diff --git a/analysis/current/analysis-report-hybrid.json b/analysis/current/analysis-report-hybrid.json index 1fb814096f16fdca0fc9bc11fba8e49285f06fae..925c2e9001e7e5bce8960063c5471ed0a6499efd 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": "20260614T180027Z", - "generated_at": "2026-06-14T18:09:50Z", + "snapshot_id": "20260615T000032Z", + "generated_at": "2026-06-15T00:08:22Z", "evidence_quality": "full", "llm_enrichment": true, "meta_bugs": [ diff --git a/analysis/current/analysis-report-hybrid.llm-reviews.json b/analysis/current/analysis-report-hybrid.llm-reviews.json index d1033b9d3e962ca4c768d089c8ff368cfddaa492..95ab97685a0a1f42d7cc57c20f9d5ebeb7285fc2 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": "20260614T180027Z", - "generated_at": "2026-06-14T18:09:50Z", + "snapshot_id": "20260615T000032Z", + "generated_at": "2026-06-15T00:08:22Z", "model": "gpt-5.4-mini?service_tier=flex", "reviews": [ { @@ -132,340 +132,298 @@ "issue:43866|issue:46402", "issue:45800|issue:46304" ], - "prepared_review_unit_hash": "7ae37cfff2afcee5ca9c1a3146002f8780827863f91586ad0a0479233064ebac", + "prepared_review_unit_hash": "8b3d64b39d31cbc89f08552441d602d75d3928dc045fec6631eb1f5fabc0f5ae", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 63, - "serialized_chars": 46528, - "estimated_input_tokens": 11632, - "estimated_eval_tokens": 23520 + "serialized_chars": 46524, + "estimated_input_tokens": 11631, + "estimated_eval_tokens": 23518 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose bundle of unrelated model/runtime/docs regressions. Only one soft edge looks like a true duplicate; most others share broad buzzwords like ZeRO-3, dtype mismatch, or version incompatibility but not the same concrete bug.", - "confidence": 0.81, - "canonical_issue_reason": "issue:45237 is the best umbrella representative for this noisy cluster: it is one of the most connected runtime/hardware failure reports and captures the broad device/backend incompatibility theme better than the more niche items.", + "summary": "This cluster is mostly heterogeneous. Only two soft-edge groups look like true duplicates: the DeepSpeed ZeRO-3/BERT index-out-of-bounds reports, and the Torch 2.4.1 compatibility reports. The rest are different model-specific bugs, docs issues, or unrelated error paths.", + "confidence": 0.74, + "canonical_issue_reason": "issue:43638 is the broadest and most representative of the only clear duplicate subcluster: the DeepSpeed ZeRO-3 / BERT index-0-out-of-bounds failure. It subsumes the narrower init-specific report in issue:43596.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45237 is the strongest global issue candidate because it is central among the linked failures and has the broadest user-impacting symptom set, even though the cluster is heterogeneous overall.", + "best_issue_reason": "issue:43638 is the best issue representative for this cluster because it matches a true duplicate pair and describes the underlying bug more generally than the companion report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44322", "right": "issue:45464", "accept": false, - "reason": "Different failures: missing config attribute vs streaming chat/completions API error." + "reason": "Different Qwen3.5 failures: missing config attribute vs streaming chat/completions bug." }, { "left": "issue:45237", "right": "issue:45685", "accept": false, - "reason": "Different backends and bugs: GPT-OSS on AMD GPUs vs MPS histogram on Int in MoE." + "reason": "Unrelated hardware/model issues: GPT-OSS on AMD vs MPS histogram on Int." }, { "left": "issue:43716", "right": "issue:45237", "accept": false, - "reason": "Different model paths: image preprocessor/model dtype mismatch vs GPT-OSS AMD GPU failure." + "reason": "Different model paths and symptoms: image preprocessor dtype mismatch vs AMD GPU runtime failure." }, { "left": "issue:43828", "right": "issue:45237", "accept": false, - "reason": "Different models and code paths: Phi-tiny-MoE autocast dtype issue vs GPT-OSS on AMD GPUs." + "reason": "Different bugs and model families; both are dtype/device-related but not the same failure." }, { "left": "issue:45237", "right": "issue:46468", "accept": false, - "reason": "Different problems: GPT-OSS AMD failure vs broken Nemotron example." + "reason": "Different models and code paths; a broken official example is not the same as the AMD GPU issue." }, { "left": "issue:45237", "right": "issue:45507", "accept": false, - "reason": "Different model-specific bug: GPT-OSS AMD GPU issue vs GraniteMoEHybrid invalid method call." + "reason": "Different MoE/model bug: AMD GPU runtime vs invalid method call." }, { "left": "issue:30277", "right": "issue:35545", "accept": false, - "reason": "Different subsystems: DeepSpeed collective mismatch vs ONNX export failure." + "reason": "DeepSpeed collective mismatch and ONNX export error are unrelated." }, { "left": "issue:35545", "right": "issue:42915", "accept": false, - "reason": "Different bugs: ModernBERT ONNX export vs Qwen3Moe FP8 failure." + "reason": "Different issues: ONNX export for ModernBERT vs FineGrainedFP8 failure for Qwen3Moe." }, { "left": "issue:35545", "right": "issue:41720", "accept": false, - "reason": "Different failures: ONNX export vs Qwen3 auto device mapping cuda assert." + "reason": "Different failure modes and models; export bug vs device-mapping CUDA assert." }, { "left": "issue:45901", "right": "issue:46015", "accept": false, - "reason": "Different subjects: table QA crash vs RAG documentation cleanup." + "reason": "One is a task crash, the other is outdated RAG documentation." }, { "left": "issue:44509", "right": "issue:44512", "accept": false, - "reason": "Both are stale docs after v5, but they refer to different removed references and doc sections." + "reason": "Both are docs cleanup for v5 removals, but they target different removed references and are not the same concrete change." }, { "left": "issue:35545", "right": "issue:36010", "accept": false, - "reason": "Different issues: export-time ONNX error vs GenerationMixin import error." + "reason": "ONNX export bug and GenerationMixin import error are unrelated." }, { "left": "issue:44918", "right": "issue:46165", "accept": false, - "reason": "Different training problems: TRL SFT embedding unpacking vs resume-training/trust-remote-code issue." + "reason": "Different training/resume bugs with no shared code-path evidence." }, { "left": "issue:44805", "right": "issue:45823", "accept": false, - "reason": "Different failure modes: mask/tensor shape mismatch vs Gemma4 device placement mismatch." + "reason": "Both mention mismatch-like behavior, but one is a mask/index shape error and the other is a Gemma4 device-map issue." }, { "left": "issue:43638", "right": "issue:44805", "accept": false, - "reason": "Different errors: ZeRO-3 Bert index error vs mask shape mismatch at indexing." + "reason": "Different underlying problems: ZeRO-3 BERT index error vs generic mask/tensor shape mismatch." }, { "left": "issue:43638", "right": "issue:46165", "accept": false, - "reason": "Different contexts: ZeRO-3 Bert initialization bug vs training resume with local model." + "reason": "ZeRO-3 BERT index error is unrelated to resume-training with trust_remote_code." }, { "left": "issue:43596", "right": "issue:44805", "accept": false, - "reason": "Different failures: ZeRO-3 Bert index error vs unrelated mask/tensor shape mismatch." + "reason": "Both are index/shape errors, but the model/setup and failure path differ." }, { "left": "issue:44805", "right": "issue:46294", "accept": false, - "reason": "Different modalities and bugs: mask indexing error vs Qwen3.5 video token/features mismatch." + "reason": "Different modalities and symptoms: mask shape mismatch vs Qwen3.5 video token/features mismatch." }, { "left": "issue:44805", "right": "issue:45161", "accept": false, - "reason": "Different model/backend issues: mask shape error vs GPT-OSS tensor-parallel failure." + "reason": "Generic tensor-shape error vs GPT-OSS tensor-parallel bug; not the same issue." }, { "left": "issue:46165", "right": "issue:46468", "accept": false, - "reason": "Different bugs: resume/trust-remote-code issue vs broken official Nemotron example." + "reason": "Different problems: training resume with remote code vs official Nemotron example failure." }, { "left": "issue:44560", "right": "issue:46165", "accept": false, - "reason": "Different failures: video StopIteration vs training resume/trust-remote-code." + "reason": "Video chat-template StopIteration and training-resume failure are unrelated." }, { "left": "issue:43596", "right": "issue:46165", "accept": false, - "reason": "Different bugs: ZeRO-3 Bert index error vs resume-training issue." + "reason": "DeepSpeed ZeRO-3 BERT index error is not the same as the resume-training issue." }, { "left": "issue:34634", "right": "issue:35545", "accept": false, - "reason": "Different domains: BarkProcessor voice preset bug vs ModernBERT ONNX export." + "reason": "BarkProcessor voice preset bug and ModernBERT ONNX export are unrelated." }, { "left": "issue:30277", "right": "issue:43638", "accept": false, - "reason": "Both involve DeepSpeed, but one is collective mismatch and the other is Bert zero-size index error; not the same bug." + "reason": "Both involve DeepSpeed, but one is a collective mismatch and the other is an index-out-of-bounds BERT issue." }, { "left": "issue:43596", "right": "issue:43638", "accept": true, - "reason": "Same underlying ZeRO-3 Bert loading/training bug with the same 'index 0 is out of bounds for dimension 0 with size 0' failure." + "reason": "Same DeepSpeed ZeRO-3 / BERT index-out-of-bounds bug, just described from init vs training/non-pretrained variants." }, { "left": "issue:43039", "right": "issue:45161", "accept": false, - "reason": "Different problems: Liger Kernel cross-entropy path vs GPT-OSS tensor-parallel failure." + "reason": "Liger Kernel cross-entropy call and GPT-OSS TP failure are different code paths." }, { "left": "issue:43975", "right": "issue:44457", "accept": false, - "reason": "Different bugs: detokenization regression vs LoRA merge/save/load mismatch." + "reason": "Different tokenizer/decoding bug vs LoRA merge/save inconsistency." }, { "left": "issue:44292", "right": "issue:44457", "accept": false, - "reason": "Different model/runtime issue vs LoRA weight merge inconsistency." + "reason": "Different models and symptoms; NVFP4 runtime error is not the LoRA reload discrepancy." }, { "left": "issue:34928", "right": "issue:44805", "accept": false, - "reason": "Different training stack failures: FSDP/checkpointing recompute mismatch vs generic mask indexing error." + "reason": "Activation-checkpointing/FSDP size mismatch is distinct from the mask-shape error." }, { "left": "issue:24643", "right": "issue:30277", "accept": false, - "reason": "Different DeepSpeed symptoms and code paths: weight shape error vs collective mismatch." + "reason": "Both are DeepSpeed-related, but the concrete failures are different: weight dimensionality vs collective mismatch." }, { "left": "issue:45859", "right": "issue:45902", "accept": false, - "reason": "Same model family, but different concrete bugs: CPU offload incompatibility vs uninitialized memory after materialization." + "reason": "Same model component, but different failures: CPU offload incompatibility vs uninitialized memory after materialization." }, { - "left": "issue:39290", - "right": "issue:45081", + "left": "issue:43866", + "right": "issue:46402", "accept": false, - "reason": "Different model/tooling bugs: Gemma3 vLLM config attribute error vs Mistral tokenizer patch crash." + "reason": "Corrupted checkpoint and missing text encoder weights are not the same bug." }, { "left": "issue:45800", "right": "issue:46515", - "accept": false, - "reason": "Both mention torch compatibility, but they are separate version constraints and failure reports, not one bug." + "accept": true, + "reason": "Both report the same Torch 2.4.1 incompatibility; 46515 is the broader restatement of 45800." }, { "left": "issue:45823", "right": "issue:45874", "accept": false, - "reason": "Different Gemma4 issues: device mismatch in forward vs expensive reverse embedding lookup." + "reason": "Different Gemma4 bugs: device mismatch in forward vs expensive reverse embedding lookup for inputs_embeds." }, { "left": "issue:43425", "right": "issue:44292", "accept": false, - "reason": "Generic torch incompatibility report vs Qwen3-8B NVFP4 runtime error." - }, - { - "left": "issue:43827", - "right": "issue:44512", - "accept": false, - "reason": "Both are docs regressions, but one is about removed pipeline tasks and the other about the removed transformers run command." + "reason": "Generic Torch 2.10 incompatibility and Qwen3-8B-NVFP4 runtime error are unrelated." }, { "left": "issue:43828", "right": "issue:45685", "accept": false, - "reason": "Different bugs: autocast dtype mismatch vs MPS histogram not implemented for Int." + "reason": "Different issues: autocast dtype mismatch vs unimplemented MPS histogram op." }, { "left": "issue:44918", "right": "issue:45685", "accept": false, - "reason": "Different subjects: TRL SFT embedding unpacking vs MPS MoE histogram failure." + "reason": "Qwen3.5 SFT-trainer embedding issue and MPS histogram error are unrelated." }, { "left": "issue:43039", - "right": "issue:43828", - "accept": false, - "reason": "Different failures: Liger Kernel cross-entropy invocation vs autocast dtype mismatch." - }, - { - "left": "issue:30277", - "right": "issue:36331", + "right": "issue:44918", "accept": false, - "reason": "Different training-related bugs: DeepSpeed collective mismatch vs custom trainer compute_loss kwarg change." + "reason": "Liger Kernel cross-entropy and Qwen3.5 SFT-trainer embedding failures are different bugs." }, { "left": "issue:34928", "right": "issue:36010", "accept": false, - "reason": "Different paths: activation checkpointing/FSDP size mismatch vs GenerationMixin import error." + "reason": "FSDP activation-checkpointing mismatch and GenerationMixin import failure are unrelated." }, { "left": "issue:43039", "right": "issue:44805", "accept": false, - "reason": "Different code paths: Liger Kernel cross entropy vs mask/tensor shape mismatch." + "reason": "Different code paths: Liger Kernel cross-entropy vs tensor mask shape mismatch." }, { "left": "issue:45237", "right": "issue:46015", "accept": false, - "reason": "Different areas: GPT-OSS AMD GPU issue vs outdated RAG examples." + "reason": "GPU/model runtime bug and outdated RAG docs are unrelated." }, { - "left": "issue:43866", + "left": "issue:45866", "right": "issue:46402", "accept": false, - "reason": "Different bugs: corrupted checkpoint vs Sam3Model text encoder weights not loaded." + "reason": "No pair provided." }, { "left": "issue:45800", "right": "issue:46304", "accept": false, - "reason": "Both are compatibility reports, but they concern different torch/distributed constraints and different failure modes." - }, - { - "left": "issue:43716", - "right": "issue:45507", - "accept": false, - "reason": "Different bugs: Mistral dtype mismatch vs GraniteMoEHybrid invalid method." - }, - { - "left": "issue:43716", - "right": "issue:45685", - "accept": false, - "reason": "Different backend/model failures: dtype mismatch in Mistral vs MPS histogram Int error." - }, - { - "left": "issue:43541", - "right": "issue:43716", - "accept": false, - "reason": "Different failures: torch dynamo grouped_mm tracing error vs Mistral image preprocessor/model dtype mismatch." - }, - { - "left": "issue:43716", - "right": "issue:46468", - "accept": false, - "reason": "Different issues: Mistral dtype mismatch vs broken Nemotron example." - }, - { - "left": "issue:43716", - "right": "issue:45237", - "accept": false, - "reason": "Different model/backend bugs: Mistral dtype mismatch vs GPT-OSS AMD GPU failure." + "reason": "Related theme of version compatibility, but different concrete breakages and code paths." } ] }, "evaluator_result": { "accept": true, - "feedback": "The cluster summary is grounded in the packet: it correctly characterizes the set as mostly heterogeneous and only treats the 43596/43638 pair as a duplicate. The soft-edge verdicts are generally conservative and do not overclaim merges where the titles and pairings point to distinct bugs or subsystems." + "feedback": "Accepted. The canonical issue choice and cluster summary are grounded in the packet, and the two accepted soft edges are conservative duplicates rather than overclaims. The remaining soft-edge rejections are appropriately cautious and do not overstate equivalence." }, "evaluator_used": true, "retried": false, @@ -579,292 +537,292 @@ "issue:43828|issue:45464", "issue:43366|issue:45464" ], - "prepared_review_unit_hash": "888b668f50eada30e0cf19b72f63dbf982c738c301a373b002c77544e17794ea", + "prepared_review_unit_hash": "20012b59bf88ee980571b99a721133dfd50a8e54fd95865d646b639498b304db", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 42, - "serialized_chars": 40030, - "estimated_input_tokens": 10008, - "estimated_eval_tokens": 20272 + "serialized_chars": 40024, + "estimated_input_tokens": 10006, + "estimated_eval_tokens": 20268 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly heterogeneous issue set. Only one soft pair looks like a true duplicate: the v5 AutoTokenizer wrong-token-ID regression across model families. The rest share broad subsystems or similar error shapes but do not appear to be the same underlying bug.", - "confidence": 0.61, - "canonical_issue_reason": "Issue 45920 is the best representative because it is the broadest, most actionable tokenizer-regression report, covering multiple model families and carrying inbound references.", + "summary": "This cluster is mostly a loose set of Transformers regression/model-loading issues, with a few near-duplicate tokenizer reports but many unrelated model- or environment-specific bugs. The strongest duplicate-like pair is the AutoTokenizer v5 regression reports; most other soft pairs should stay separate.", + "confidence": 0.86, + "canonical_issue_reason": "issue:45920 is the most representative umbrella report: it describes the broadest concrete regression (AutoTokenizer returning wrong token IDs across multiple model families in v5), which also subsumes the narrower Granite-only report.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45920 is the strongest global issue choice: broad scope, open, and likely the central tracker for the v5 AutoTokenizer token-ID regression cluster.", + "best_issue_reason": "issue:45920 is the best global issue candidate because it is the broadest, most user-impacting, and most clearly root-cause-aligned issue in the set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43296", "right": "issue:46304", "accept": false, - "reason": "Different failures: PaddleOCR-VL loading in vLLM vs a PyTorch USE_DISTRIBUTED=0 compatibility break." + "reason": "Both are loading/runtime failures, but one is PaddleOCR-VL in vLLM and the other is USE_DISTRIBUTED=0 support in v5.9.0; different code paths and failure modes." }, { "left": "issue:41762", "right": "issue:43596", "accept": false, - "reason": "Both are ZeRO-3 index errors, but the models and likely code paths differ; not enough evidence of one underlying bug." + "reason": "Both mention ZeRO-3 index errors, but the affected models and load paths differ (Gemma3 conditional generation vs BertModel)." }, { "left": "issue:39290", "right": "issue:41720", "accept": false, - "reason": "Gemma3 sliding_window_pattern attribute error and Qwen3 auto-device-map cuda assert are unrelated." + "reason": "Gemma3 config attribute missing in vLLM and Qwen3 auto device-map CUDA assert are unrelated bugs." }, { "left": "issue:44589", "right": "issue:46304", "accept": false, - "reason": "Float8 storage lookup failure is unrelated to the distributed build regression." + "reason": "Float8 storage lookup failure and USE_DISTRIBUTED=0 breakage are unrelated compatibility issues." }, { "left": "issue:43901", "right": "issue:44512", "accept": false, - "reason": "Both are docs issues, but they concern different outdated references and different behavior changes." + "reason": "Both are docs-related, but they cover different API/docs regressions and do not look like the same underlying issue." }, { "left": "issue:45237", "right": "issue:45901", "accept": false, - "reason": "GPT-OSS AMD GPU failure and table-question-answering crash are unrelated." + "reason": "GPT-OSS on AMD GPUs and a table-question-answering crash are unrelated." }, { "left": "issue:45823", "right": "issue:46294", "accept": false, - "reason": "Different model families and different failure modes: device mismatch vs apply_chat_template video-token mismatch." + "reason": "Different model families and different bugs: Gemma4 device placement vs Qwen3.5 video chat-template token mismatch." }, { "left": "issue:43039", "right": "issue:44560", "accept": false, - "reason": "Liger Kernel cross-entropy dispatch and Qwen3-VL video StopIteration are distinct problems." + "reason": "Liger Kernel cross-entropy behavior and Qwen3.5 video StopIteration are unrelated." }, { "left": "issue:44841", "right": "issue:45823", "accept": false, - "reason": "Voxtral processor failure and Gemma4 PLE device mismatch are not the same bug." + "reason": "Processor failure for Voxtral and Gemma4 device mismatch are separate model-specific bugs." }, { "left": "issue:43975", "right": "issue:46015", "accept": false, - "reason": "DeepSeek detokenization regression and outdated RAG docs are unrelated." + "reason": "Incorrect detokenization and outdated RAG docs are unrelated." }, { "left": "issue:44863", "right": "issue:46015", "accept": false, - "reason": "NemotronH checkpoint loading failure is unrelated to documentation examples." + "reason": "NemotronH checkpoint loading and docs updates do not share the same underlying bug." }, { "left": "issue:45020", "right": "issue:46304", "accept": false, - "reason": "Remote_code compatibility break and USE_DISTRIBUTED=0 incompatibility are different regressions." + "reason": "One is a broad remote_code regression report, the other is a specific USE_DISTRIBUTED=0 breakage; too different to merge." }, { "left": "issue:45812", "right": "issue:45920", "accept": true, - "reason": "Same underlying v5 tokenizer regression: wrong token IDs across model families, with 45812 appearing to be a Granite-specific instance of the broader issue in 45920." + "reason": "Both report the same concrete AutoTokenizer v5 regression: wrong token IDs across model families. One is Granite-specific, the other is the broader multi-model manifestation." }, { "left": "issue:43866", "right": "issue:44863", "accept": false, - "reason": "Corrupted Ovis2 checkpoint and NemotronH checkpoint-loading implementation bug are different." + "reason": "Corrupted Ovis2 checkpoint and NemotronH checkpoint loading are unrelated." }, { "left": "issue:43872", "right": "issue:46304", "accept": false, - "reason": "bitsandbytes Int8Params kwarg incompatibility is unrelated to the distributed build regression." + "reason": "bitsandbytes constructor incompatibility and distributed-build support regression are different problems." }, { "left": "issue:44292", "right": "issue:46015", "accept": false, - "reason": "Qwen-3-8B-NVFP4 runtime error and outdated RAG docs are unrelated." + "reason": "Qwen-3-8B-NVFP4 runtime error and RAG docs issues are unrelated." }, { "left": "issue:39290", "right": "issue:43531", "accept": false, - "reason": "Both mention sliding window, but one is a missing config attribute and the other is a Qwen3-MoE model-specific issue." + "reason": "Both mention sliding_window, but they involve different models/configs and different failure modes." }, { "left": "issue:41093", "right": "issue:43638", "accept": false, - "reason": "Both are IndexErrors, but one is a mask/tensor shape mismatch and the other is a ZeRO-3 initialization failure." + "reason": "Both are index errors, but one is a mask/tensor shape mismatch and the other is a ZeRO-3 BertModel initialization failure." }, { "left": "issue:43792", "right": "issue:45901", "accept": false, - "reason": "Whisper model runtime failure and table-question-answering crash are unrelated." + "reason": "Whisper loading failure and table-question-answering crash are unrelated." }, { "left": "issue:43854", "right": "issue:43866", "accept": false, - "reason": "GLM-4.7-Flash load failure and Ovis2 checkpoint corruption are distinct model-specific problems." + "reason": "GLM-4.7-Flash unit-test loading issue and Ovis2 checkpoint corruption are different model-specific failures." }, { "left": "issue:43425", "right": "issue:45070", "accept": false, - "reason": "Torch version incompatibility and a pydantic PretrainedConfig regression are separate compatibility issues." + "reason": "Torch 2.10 incompatibility and a PretrainedConfig pydantic break are separate compatibility regressions." }, { "left": "issue:43596", "right": "issue:45800", "accept": false, - "reason": "ZeRO-3 BertModel load failure and torch 2.4.1/transformers 5.8.0 incompatibility are different issues." + "reason": "ZeRO-3 BertModel index error and torch/transformers version incompatibility are unrelated." }, { "left": "issue:43975", "right": "issue:45901", "accept": false, - "reason": "Detokenization regression and table-question-answering crash are unrelated." + "reason": "Detokenization regression and table-QA crash do not share the same code path." }, { "left": "issue:43792", "right": "issue:45070", "accept": false, - "reason": "Whisper runtime failure and pydantic config regression are unrelated." + "reason": "Whisper cannot-run report and pydantic config break are unrelated." }, { "left": "issue:45362", "right": "issue:45464", "accept": false, - "reason": "Both involve Qwen3.5, but one is a transformers-chat crash and the other is a streaming chat/completions API failure; not clearly the same bug." + "reason": "Both are Qwen3.5-related, but one is a chat crash and the other is a streaming chat/completions API failure; not clearly the same bug." }, { "left": "issue:44292", "right": "issue:45901", "accept": false, - "reason": "Qwen-3-8B-NVFP4 runtime error and table-question-answering crash are unrelated." + "reason": "Qwen-3-8B-NVFP4 runtime error and table-QA crash are unrelated." }, { "left": "issue:39290", "right": "issue:45823", "accept": false, - "reason": "Gemma3 sliding_window_pattern and Gemma4 PLE device mismatch are different regressions." + "reason": "Different models and different failures: Gemma3 config attribute vs Gemma4 device mismatch." }, { "left": "issue:45464", "right": "issue:45685", "accept": false, - "reason": "Qwen3.5 streaming inference failure and MPS MoE histogram error are unrelated." + "reason": "Streaming inference failure and MPS histogram_mps int error are unrelated." }, { "left": "issue:43541", "right": "issue:45685", "accept": false, - "reason": "Different MoE problems: grouped_mm tracing failure vs MPS histogram unsupported for Int." + "reason": "Mixtral grouped_mm tracing and MPS histogram_mps failures are different bugs." }, { "left": "issue:43541", "right": "issue:43828", "accept": false, - "reason": "Both are MoE-related runtime errors, but the traced grouped_mm failure and autocast dtype mismatch are different concrete bugs." + "reason": "Both are MoE-related, but grouped_mm tracing and autocast dtype mismatch are distinct code paths." }, { "left": "issue:36010", "right": "issue:39290", "accept": false, - "reason": "ImportError for GenerationMixin and Gemma3 sliding_window_pattern are unrelated." + "reason": "Importing GenerationMixin and a Gemma3 config attribute error are unrelated." }, { "left": "issue:44841", "right": "issue:45084", "accept": false, - "reason": "Voxtral processor failure and non-template-node compile error are unrelated." + "reason": "Voxtral processor failure and compilation of non-template nodes are unrelated." }, { "left": "issue:43404", "right": "issue:43866", "accept": false, - "reason": "Tied lm_head weights in Mistral3 and a corrupted Ovis2 checkpoint are unrelated." + "reason": "Mistral3 tied-weights bug and Ovis2 checkpoint corruption are unrelated." }, { "left": "issue:41093", "right": "issue:41762", "accept": false, - "reason": "Both are IndexErrors, but they stem from different symptoms and model-loading contexts." + "reason": "Both are index errors, but they arise in different contexts and model-loading paths; not the same underlying bug." }, { "left": "issue:43296", "right": "issue:43366", "accept": false, - "reason": "PaddleOCR-VL loading in vLLM and GGUF gpt-oss support are different features/issues." + "reason": "PaddleOCR-VL load failure and GGUF gpt-oss support are different architecture/runtime issues." }, { "left": "issue:45800", "right": "issue:45823", "accept": false, - "reason": "Torch/transformers version incompatibility and Gemma4 device mismatch are unrelated." + "reason": "Torch/Transformers version incompatibility and Gemma4 device mismatch are different regressions." }, { "left": "issue:43572", "right": "issue:45070", "accept": false, - "reason": "StableLmConfig pad_token_idx regression and pydantic PretrainedConfig breakage are different config issues." + "reason": "Missing pad_token_idx in StableLmConfig and PretrainedConfig field breakage are different config-schema problems." }, { "left": "issue:44805", "right": "issue:44918", "accept": false, - "reason": "Mask-shape IndexError and Qwen3.5 embedding unpacking failure do not appear to share the same cause." + "reason": "Mask shape mismatch during generation and TRL embedding-unpacking failure are different code paths." }, { "left": "issue:44387", "right": "issue:45005", "accept": false, - "reason": "Int4 OOM/memory regression and tied-weights translation-model regression are different problems." + "reason": "Int4 quantization OOM and tied-weights translation-model issues are unrelated." }, { "left": "issue:43541", "right": "issue:45161", "accept": false, - "reason": "Grouped_mm tracing failure and GPT-OSS MoE tensor-parallel failure are different MoE bugs." + "reason": "Both involve MoE/parallelism, but grouped_mm tracing and TP support on GPT-OSS are distinct issues." }, { "left": "issue:43828", "right": "issue:45464", "accept": false, - "reason": "Autocast dtype mismatch in Phi-tiny-MoE and Qwen3.5 streaming inference failure are unrelated." + "reason": "Phi-tiny-MoE autocast dtype mismatch and Qwen3.5 streaming API failure are unrelated." }, { "left": "issue:43366", "right": "issue:45464", "accept": false, - "reason": "GGUF gpt-oss support and Qwen3.5 streaming failure are unrelated." + "reason": "GGUF gpt-oss support and Qwen3.5 streaming API failure are unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet, and the only accepted soft pair (issue:45812 \u2194 issue:45920) is a conservative, title-based duplicate call: both describe the same v5 AutoTokenizer wrong-token-ID regression, with 45920 clearly broader. The other soft-edge rejections stay cautious and do not overclaim a shared root cause." + "feedback": "The summary is grounded in the packet and stays conservative: it correctly treats most pairs as unrelated, and the one accepted pair (45812/45920) is well supported by both titles describing the same AutoTokenizer v5 wrong-token-ID regression, with 45920 clearly the broader umbrella report. No obvious overstatement beyond the available evidence." }, "evaluator_used": true, "retried": false, @@ -967,20 +925,20 @@ "issue:36331|issue:41762", "issue:43828|issue:44928" ], - "prepared_review_unit_hash": "3833d1ab332f1cd976fc2b744b1c6edf9ba78f1e6548289ea9499610781068e8", + "prepared_review_unit_hash": "85d364253373430f4945048ef27df94b1481e5932545a4d01b726c3712da5765", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37033, + "serialized_chars": 37035, "estimated_input_tokens": 9259, "estimated_eval_tokens": 18774 }, @@ -989,207 +947,207 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The non-persistent-buffer regression pair (43950/44534) remains the only clearly grounded duplicate. The T5/XOR-validation pair is too weakly evidenced from titles alone and should be treated as a non-duplicate for now.", - "confidence": 0.84, - "canonical_issue_reason": "43950 is the clearest, most specific representative of the shared non-persistent-buffer corruption regression.", + "summary": "Mostly a loose similarity cluster of unrelated Transformers regressions. The only clear duplicate subgroups are the non-persistent buffer corruption reports, with a couple of plausible same-bug pairs in SigLIP2/video-input/validation error-message issues.", + "confidence": 0.59, + "canonical_issue_reason": "issue:43644 is the clearest and earliest representative of the repeated non-persistent-buffer regression; its wording matches the later rephrasings in 43950 and 44534 most directly.", "canonical_pr_reason": null, - "best_issue_reason": "43950 is the best overall anchor issue because it is precise and directly describes the underlying regression.", + "best_issue_reason": "Among these mostly heterogeneous issues, 43644 is the best representative because it states a concrete, recurring bug plainly and is the broadest/earliest formulation of that sub-bug.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44291", "right": "issue:45070", "accept": false, - "reason": "Different concrete failures: unexpected init argument vs pydantic/config field break." + "reason": "Different regressions: unexpected `_is_hf_initialized` arg during empty-weights loading vs a Pydantic/PretrainedConfig field breakage." }, { "left": "issue:43531", "right": "issue:45070", "accept": false, - "reason": "Unrelated bugs in different code paths; sliding_window behavior vs PretrainedConfig/pydantic." + "reason": "`sliding_window` handling in Qwen3-MoE is a different code path from the config/Pydantic regression in 45070." }, { "left": "issue:43716", "right": "issue:46294", "accept": false, - "reason": "Both are multimodal/input issues, but the symptoms and likely code paths differ." + "reason": "Both involve multimodal models, but one is an image preprocessor/dtype mismatch and the other is a video chat-template token/feature mismatch." }, { "left": "issue:45020", "right": "issue:45800", "accept": false, - "reason": "Different root causes: remote_code breakage vs torch/transformers version incompatibility." + "reason": "Remote-code compatibility breakage and torch-version incompatibility are broad but distinct problems." }, { "left": "issue:43541", "right": "issue:44589", "accept": false, - "reason": "Distinct runtime errors in unrelated code paths." + "reason": "Grouped-mm tracing failure in Mixtral is unrelated to the missing Float8 storage lookup error." }, { "left": "issue:44479", "right": "issue:46294", - "accept": false, - "reason": "Both mention video input, but one is a broad regression and the other is a specific template/token mismatch." + "accept": true, + "reason": "Both describe the same Qwen video-input regression family: video features/tokens or template processing are mismatched." }, { "left": "issue:41093", "right": "issue:43596", "accept": false, - "reason": "Both are shape/index errors, but they are different failures with different triggers." + "reason": "Both are IndexErrors under different setups, but the mask-shape mismatch and ZeRO-3 empty-tensor indexing are not the same bug." }, { "left": "issue:43054", "right": "issue:43994", - "accept": false, - "reason": "Same model family, but the reported symptoms are not clearly the same bug." + "accept": true, + "reason": "Both report SigLIP2 producing bad/nonsensical outputs, pointing to the same underlying model/processing regression." }, { "left": "issue:45759", "right": "issue:45902", "accept": false, - "reason": "Different bugs in composite-config loading vs uninitialized rotary embedding memory." + "reason": "Different failures: composite-config unwrapping vs uninitialized rotary frequency after meta-to-empty materialization." }, { "left": "issue:39186", "right": "issue:44918", "accept": false, - "reason": "Different failures: FSDP weight dimensionality vs TRL SFT embedding unpacking." + "reason": "FSDP weight-shape failure and TRL SFT embedding unpacking failure are unrelated." }, { "left": "issue:43716", "right": "issue:44918", "accept": false, - "reason": "Shared multimodal/training context only; not the same underlying bug." + "reason": "Image dtype mismatch and TRL unpacking failure do not share the same underlying bug." }, { "left": "issue:40990", "right": "issue:44960", "accept": false, - "reason": "Completely different models and symptoms." + "reason": "High perplexity on GPT-OSS/WikiText-2 and a GLM5 issue are unrelated model-specific reports." }, { "left": "issue:43645", "right": "issue:45399", "accept": false, - "reason": "Custom-model initialization in notebooks is unrelated to flash-attn fallback gating." + "reason": "Custom-model initialization in Jupyter and flash-attn2 fallback gating are different issues." }, { "left": "issue:45632", "right": "issue:45698", "accept": false, - "reason": "Both involve custom code loading, but the failure modes are different." + "reason": "Cache-path collision for trust_remote_code and wrong custom-module loading after save_pretrained are related but distinct bugs." }, { "left": "issue:43404", "right": "issue:45005", "accept": false, - "reason": "Both mention tied weights, but affect different model families and likely different paths." + "reason": "lm_head weight tying in Mistral3 is a different model-specific bug from the broader translation-model tied-weights regression." }, { "left": "issue:42915", "right": "issue:45399", "accept": false, - "reason": "FineGrainedFP8Config failure is unrelated to flash-attn fallback checks." + "reason": "Qwen3Moe FP8 failure is unrelated to the flash-attn2 fallback checks." }, { "left": "issue:44530", "right": "issue:45005", "accept": false, - "reason": "PagedAttentionCache/linear_attention crash is unrelated to tied-weights translation behavior." + "reason": "PagedAttentionCache invalid group type and tied-weights handling are different code paths." }, { "left": "issue:43873", "right": "issue:45005", "accept": false, - "reason": "Quantization offloading behavior is a different issue from tied-weight handling." + "reason": "Quantization/offloading behavior is not the same as the tied-weights regression." }, { "left": "issue:44360", "right": "issue:44512", "accept": false, - "reason": "A DSA indexer note is unrelated to a docs typo." + "reason": "A DSA indexer math bug and a docs typo about the removed run command are unrelated." }, { "left": "issue:44863", "right": "issue:45005", "accept": false, - "reason": "NemotronH checkpoint-loading failure is not the same as tied-weight regressions." + "reason": "NemotronH checkpoint loading is unrelated to the translation tied-weights issue." }, { "left": "issue:45183", "right": "issue:46463", - "accept": false, - "reason": "Title similarity alone is not enough; 46463 looks like a T5-specific validation/error-message issue rather than a confirmed duplicate of the broader XOR-check bug." + "accept": true, + "reason": "46463 looks like a specific T5 manifestation of the broader input/decoder validation error-message bug described in 45183." }, { "left": "issue:44492", "right": "issue:44512", "accept": false, - "reason": "A cache-strategy typo is unrelated to docs mentioning a removed command." + "reason": "Cache-strategy typo and outdated CLI docs mention are unrelated documentation issues." }, { "left": "issue:44918", "right": "issue:45507", "accept": false, - "reason": "Different bugs: TRL embedding unpacking vs an invalid-method call." + "reason": "TRL SFT embedding unpacking and GraniteMoEHybrid invalid-method calls are different failures." }, { "left": "issue:46489", "right": "issue:46491", "accept": false, - "reason": "Both are tokenizer regressions, but for different model families and symptoms." + "reason": "Both are tokenizer regressions, but the models and symptoms differ enough to look like separate bugs." }, { "left": "issue:43975", "right": "issue:44292", "accept": false, - "reason": "Different tokenizer/model-loading problems." + "reason": "Incorrect detokenization and NVFP4 runtime failure are unrelated." }, { "left": "issue:44292", "right": "issue:45507", "accept": false, - "reason": "Completely different failures." + "reason": "NVFP4 execution failure and GraniteMoEHybrid invalid method are different bugs." }, { "left": "issue:43950", "right": "issue:44534", "accept": true, - "reason": "Same regression: non-persistent buffers are being corrupted/filled with junk in Transformers v5." + "reason": "Both describe the same v5 regression where non-persistent buffers get corrupted/junk values during loading." }, { "left": "issue:39692", "right": "issue:43054", "accept": false, - "reason": "Both mention SigLIP2, but one is a docs-example mismatch and the other is a quality/results complaint." + "reason": "SigLIP2 doc-example errors/quantization issues are not the same as the text-embedding quality regression." }, { "left": "issue:43644", "right": "issue:43950", - "accept": false, - "reason": "Related v5 regressions, but not the same bug." + "accept": true, + "reason": "Same underlying non-persistent-buffer corruption regression, phrased slightly differently." }, { "left": "issue:36331", "right": "issue:41762", "accept": false, - "reason": "Different error classes and contexts." + "reason": "Unexpected compute_loss keyword and ZeRO-3 empty-index loading error are unrelated." }, { "left": "issue:43828", "right": "issue:44928", "accept": false, - "reason": "Both are dtype/autocast-related, but the concrete failures differ." + "reason": "Autocast dtype mismatch and RLHF NaN/SDPA fallback issues are different failures in different setups." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The buffer-regression pairing (43950/44534) is well grounded, but the other accepted edge (45183/46463) is too assertive for the evidence shown. The packet only gives title-level similarity, and 46463 could be a related T5-specific validation/error-message issue rather than a confirmed duplicate of the broader XOR-check bug. Please downgrade that verdict to something more tentative unless you have body-level evidence tying them to the same fix." + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative: it correctly frames the set as mostly unrelated regressions, and the accepted soft edges are all reasonably supported by the issue titles/descriptions. The non-persistent-buffer cluster is clearly real, and the Qwen video, SigLIP2, and XOR validation pairs are plausible same-bug groupings without overclaiming. The rejected pairs also look appropriately cautious." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -1291,14 +1249,14 @@ "issue:45715|issue:46315", "issue:45488|issue:46489" ], - "prepared_review_unit_hash": "0f56ef49c40e4c59650856f09f26bd990bb26568404711e1a27e532ddf92886c", + "prepared_review_unit_hash": "d3a92463d9430023cbf942323fbaa3ef74b7ed1f2b04c18edcd2ab4ef8268c33", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, @@ -1313,219 +1271,219 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly heterogeneous cluster of model-loading, tokenizer, and runtime regressions. The strongest duplicate candidates are the SigLIP2 discrepancy pair and the Mistral regex/tokenizer pair; the DeepSeek/remote_code issues are related but, on this packet, not concrete enough to treat as duplicates.", - "confidence": 0.69, - "canonical_issue_reason": "issue:45920 is the best umbrella issue because it is open and explicitly covers a broad v5 tokenizer-regression class affecting multiple model families, making it the most representative issue in the cluster even though not every DeepSeek-related report is a duplicate.", + "summary": "This cluster is mostly a mix of unrelated v5 model-loading/tokenization regression reports; only a couple of soft edges look like true duplicates, mainly around SigLIP2 output mismatch and remote_code config loading.", + "confidence": 0.71, + "canonical_issue_reason": "issue:45920 is the most representative umbrella here: it is open, concrete, and captures the recurring multi-model tokenizer-regression pattern that shows up across the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 is the strongest global issue candidate because it is open, broad, and explicitly framed as a multi-model v5 tokenizer regression; that makes it the best cluster anchor without over-collapsing distinct model-specific bugs.", + "best_issue_reason": "issue:45920 is the strongest single issue to keep as the cluster representative because it is broad enough to cover the dominant tokenizer symptom while still being specific and actionable.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43872", "right": "issue:44589", "accept": false, - "reason": "Different failure modes: bitsandbytes constructor-argument incompatibility vs missing Float8 storage support." + "reason": "Both are load-time TypeErrors, but they come from different subsystems and different root causes." }, { "left": "issue:45920", "right": "issue:46489", "accept": false, - "reason": "Both are tokenizer regressions, but the packet only shows broad thematic overlap for DeepSeek/R1 vs DeepSeek-Coder; not enough evidence of one concrete duplicate bug or one mergeable fix." + "reason": "Both involve tokenizer regressions, but they affect different model families and appear to come from different code paths." }, { "left": "issue:43054", "right": "issue:45701", "accept": false, - "reason": "Both mention tokenization, but one is a SigLIP2 embedding-quality discrepancy and the other is a general tokenizer behavior change." + "reason": "One is a SigLIP2 embedding-quality issue; the other is a generic tokenizer-version regression, so the underlying bug is not the same." }, { "left": "issue:43866", "right": "issue:44403", "accept": false, - "reason": "Checkpoint corruption during loading and generic loading noise are different bugs." + "reason": "Checkpoint corruption and noisy load logging are unrelated problems." }, { "left": "issue:43950", "right": "issue:45902", "accept": false, - "reason": "A generic non-persistent buffer corruption regression and a specific Qwen3 MoE inv_freq materialization issue are not the same code-path failure." + "reason": "Both mention loading/materialization, but one is a generic non-persistent-buffer regression and the other is a model-specific uninitialized-memory bug." }, { "left": "issue:38175", "right": "issue:43054", "accept": false, - "reason": "Same model family, but the titles point to different symptoms without enough evidence of a single underlying defect." + "reason": "Same model family, but zero probabilities and degraded text embeddings are not clearly the same defect." }, { "left": "issue:41553", "right": "issue:43883", "accept": false, - "reason": "Both are model-loading failures, but for different models and different mechanisms." + "reason": "Bad AutoTokenizer error messaging and a missing all_tied_weights_keys attribute are unrelated." }, { "left": "issue:44910", "right": "issue:45859", "accept": false, - "reason": "Flash-attention packed-sequence misinterpretation and CPU-offload incompatibility are distinct failures." + "reason": "Both are Qwen3.5-related, but one is a flash-attn packed-sequence crash and the other is a CPU-offload rotary-embedding incompatibility." }, { "left": "issue:42947", "right": "issue:44387", "accept": false, - "reason": "Gradient checkpointing ineffectiveness and int4 quantization OOM are memory-related but not the same bug." + "reason": "These are different memory regressions with different mechanisms and code paths." }, { "left": "issue:45663", "right": "issue:45902", "accept": false, - "reason": "FSDP2 shared_kv_states KeyError and MoE inv_freq materialization corruption are unrelated internals." + "reason": "Different models and different failure sites; these are not the same bug." }, { "left": "issue:43054", "right": "issue:43493", "accept": true, - "reason": "Strong SigLIP2 match: both describe HF-vs-reference implementation discrepancies causing degraded or incorrect outputs." + "reason": "Both point to the same SigLIP2 HF-vs-JAX implementation mismatch, with degraded outputs as the shared symptom." }, { "left": "issue:41553", "right": "issue:43881", "accept": false, - "reason": "Both are loading errors, but for different models and different root causes." + "reason": "Different model-loading failures with no clear shared root cause." }, { "left": "issue:44403", "right": "issue:44451", "accept": false, - "reason": "Generic loading noise is not the same as a hard model-loading failure for a specific checkpoint." + "reason": "Generic load-time noise and a specific model-load failure are not the same issue." }, { "left": "issue:24643", "right": "issue:28530", "accept": false, - "reason": "DeepSpeed weight-shape training error and early-stopping metric handling are unrelated." + "reason": "Training DeepSpeed weight-shape failure and early-stopping metric detection are unrelated." }, { "left": "issue:43975", "right": "issue:45520", "accept": false, - "reason": "Tokenizer detokenization regression and Python 3.13 flash_attn import lookup are unrelated." + "reason": "Tokenizer detokenization corruption and a Python 3.13 flash_attn import KeyError are unrelated." }, { "left": "issue:44403", "right": "issue:44488", "accept": false, - "reason": "Noise during loading vs a specific model loading failure are not the same underlying bug." + "reason": "These are separate model-load complaints; no shared concrete defect is evident." }, { "left": "issue:42947", "right": "issue:43856", "accept": false, - "reason": "Both are MoE/memory themed, but one is a PEFT checkpointing issue and the other is training memory inefficiency." + "reason": "Gradient-checkpointing inefficiency and Qwen3 MoE memory usage are different performance problems." }, { "left": "issue:43643", "right": "issue:45020", - "accept": false, - "reason": "Related around remote-code/config loading, but the packet does not show a single concrete underlying bug; safer to keep them as related-but-distinct regressions." + "accept": true, + "reason": "Both are remote_code/AutoConfig loading regressions; 43643 looks like a specific incomplete-config symptom of the broader breakage." }, { "left": "issue:41762", "right": "issue:42915", "accept": false, - "reason": "Different model-loading failures triggered by different features (DeepSpeed ZeRO-3 vs FineGrainedFP8Config)." + "reason": "Different model families and different DeepSpeed/load failures." }, { "left": "issue:45685", "right": "issue:45823", "accept": false, - "reason": "MPS histogram unsupported for Int and Gemma4 device mismatch are different device-related bugs." + "reason": "An MPS histogram kernel limitation and a Gemma4 device-mismatch forward bug are unrelated." }, { "left": "issue:44991", "right": "issue:45399", "accept": false, - "reason": "Tokenizer loading regression and flash-attn fallback gating are unrelated." + "reason": "Tokenizer loading failure and flash-attn fallback blocking are separate issues." }, { "left": "issue:43824", "right": "issue:43872", "accept": false, - "reason": "Missing model-class import and bitsandbytes parameter-construction issues are unrelated." + "reason": "ImportError for a model class and bitsandbytes parameter incompatibility are unrelated." }, { "left": "issue:43782", "right": "issue:44960", "accept": false, - "reason": "Qwen3VL weight_only loading error and GLM5 are different model-specific issues." + "reason": "Qwen3VL weight_only loading and GLM5 are different problems." }, { "left": "issue:43881", "right": "issue:43883", "accept": false, - "reason": "Different model-loading failures with no shared concrete code path." + "reason": "Generic glm-4v-9b load failure and a Molmo tied-weights attribute error are different defects." }, { "left": "issue:45020", "right": "issue:45823", "accept": false, - "reason": "Remote-code breakage and Gemma4 PLE device mismatch are unrelated." + "reason": "Remote_code loading regressions and Gemma4 device placement are unrelated." }, { "left": "issue:44661", "right": "issue:45020", "accept": false, - "reason": "Tokenizer-mapping registration failure and remote-code loading regressions are different issues." + "reason": "Tokenizer-mapping model generation and remote_code config loading are different code paths." }, { "left": "issue:45081", - "right": "issue:45356", - "accept": true, - "reason": "Both hit the fix_mistral_regex / tokenizer-regex path; the crash and warning/codec regression look like the same underlying tokenizer bug." + "right": "issue:45399", + "accept": false, + "reason": "Mistral regex patching and flash-attn2 fallback gating are unrelated." }, { "left": "issue:43828", "right": "issue:45823", "accept": false, - "reason": "Autocast dtype mismatch in Phi-tiny-MoE and Gemma4 device placement are different bugs." + "reason": "Autocast dtype mismatch and Gemma4 PLE device mismatch are different bugs." }, { "left": "issue:42491", "right": "issue:44368", "accept": false, - "reason": "LoRA compatibility across versions and a tie_word_embeddings warning are not the same bug." + "reason": "LoRA compatibility across HF versions and a tie_word_embeddings warning are not the same issue." }, { "left": "issue:43742", "right": "issue:45399", "accept": false, - "reason": "MobileLLM loading key error and flash-attn fallback checks are unrelated." + "reason": "MobileLLM key error and flash-attn fallback checks are unrelated." }, { "left": "issue:45020", "right": "issue:45356", "accept": false, - "reason": "Remote-code breakage and Kimi-K2.5 tokenizer regex regression do not share the same code path." + "reason": "Remote_code loading breakage and Kimi-K2.5 tokenizer regression are different symptoms and likely different causes." }, { "left": "issue:45715", "right": "issue:46315", "accept": false, - "reason": "Two different tokenizer-conversion performance regressions in opposite directions; not one bug." + "reason": "Both are tokenizer performance regressions, but they hit different APIs and code paths." }, { "left": "issue:45488", "right": "issue:46489", "accept": false, - "reason": "Both are DeepSeek tokenizer regressions, but the packet only supports broad thematic overlap, not a clear shared concrete bug or single fixable code path." + "reason": "Both are DeepSeek tokenizer regressions, but one is a pre-tokenizer override bug and the other is a separate gap in a prior fix." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The summary is mostly grounded, but the soft-edge judgments are not consistently conservative enough. In particular, the accepted pairs `issue:43643` \u2194 `issue:45020` and `issue:45488` \u2194 `issue:46489` are supported mainly by broad thematic overlap (`remote_code`, DeepSeek/tokenizer regressions) rather than clear evidence of the same underlying bug or a single mergeable fix. `issue:45920` \u2194 `issue:46489` is also a bit broad for a duplicate call. Given the packet only provides titles/metadata, the safer stance would be to keep these as related-but-not-duplicate unless there is stronger concrete alignment." + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The representative issue choice of issue:45920 is defensible given its broad tokenizer-regression scope, and the two accepted soft edges (43054\u219443493 and 43643\u219445020) are plausible duplicates without overclaiming. The rejected pairs are phrased cautiously enough, and there are no PR-pair mergeability problems here." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -1623,20 +1581,20 @@ "issue:43950|issue:45834", "issue:43824|issue:43957" ], - "prepared_review_unit_hash": "a9e434636e321d2e27ea1b0825038be979bfe3ef09ffe3b1514a54cbbb9e5a8b", + "prepared_review_unit_hash": "ebfb2a263ad19b24fda8a3de2a37f4f00965df8fcaaf6b6061b3b82fe24d3024", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36727, + "serialized_chars": 36728, "estimated_input_tokens": 9182, "estimated_eval_tokens": 18620 }, @@ -1645,174 +1603,198 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a mix of distinct model-loading, training, and regression reports across different model families. Only a few soft pairs look like true duplicates or near-duplicates; most should stay separate.", - "confidence": 0.57, - "canonical_issue_reason": "Issue 43957 is the broadest representative of the cluster: a transformers 5.x model-loading regression around meta-device materialization, which matches the general loading-failure theme better than the more model-specific reports.", + "summary": "Mostly heterogeneous issue cluster with only a couple of strong duplicate-like pairs. I\u2019d keep the docs-v5 pipeline cleanup issue as the canonical issue; no PRs are present.", + "confidence": 0.47, + "canonical_issue_reason": "issue:43827 is the oldest and clearest representative of the only exact docs-dup pair, with a concise problem statement about v5 pipeline removals.", "canonical_pr_reason": null, - "best_issue_reason": "43957 is the best single issue anchor for triage because it captures the widest, most reusable failure mode in the set rather than a one-off model-specific symptom.", + "best_issue_reason": "issue:43827 is the best issue anchor here: it is well-scoped, clearly reproducible as a docs mismatch, and matches issue:44509 closely enough to serve as the canonical duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43827", "right": "issue:44509", "accept": true, - "reason": "Same docs cleanup: both report stale pipeline task references removed in v5." + "reason": "Both are v5 docs leftovers referencing removed text-generation/summarization/translation pipelines; same documentation cleanup." }, { "left": "issue:43404", "right": "issue:43950", "accept": false, - "reason": "Different bugs: tied lm_head in Mistral3 vs non-persistent buffer corruption in from_pretrained()." + "reason": "Mistral3 lm_head tying and non-persistent buffer corruption are different loading/state bugs." }, { "left": "issue:43872", "right": "issue:44291", "accept": true, - "reason": "Same `_is_hf_initialized` TypeError during model initialization/loading; 44291 is the broader reproduction of the same incompatibility." + "reason": "Both describe the same unexpected _is_hf_initialized argument compatibility break during model initialization/loading." }, { "left": "issue:42915", "right": "issue:43872", "accept": false, - "reason": "Different failures: FineGrainedFP8Config on Qwen3Moe vs bitsandbytes Int8Params keyword mismatch." + "reason": "FineGrainedFP8Config on Qwen3Moe and the _is_hf_initialized/bitsandbytes issue are different failures." }, { "left": "issue:43792", "right": "issue:44220", "accept": false, - "reason": "Both involve Whisper/audio, but the reported failures are not clearly the same underlying bug." + "reason": "Whisper load/run failure and _torch_extract_fbank_features are related to audio, but not the same concrete bug." }, { "left": "issue:43257", "right": "issue:45305", "accept": false, - "reason": "Different DeepSpeed problems: missing Qwen3 MoE weight conversion vs gradient averaging with GAS." + "reason": "Both involve DeepSpeed, but one is weight loading and the other is gradient averaging; different code paths." }, { "left": "issue:36331", "right": "issue:41093", "accept": false, - "reason": "Different errors and code paths: trainer loss signature mismatch vs mask/tensor shape IndexError." + "reason": "CustomTrainer compute_loss signature mismatch is unrelated to the mask/tensor shape IndexError." }, { "left": "issue:42947", "right": "issue:45923", "accept": false, - "reason": "Both mention gradient checkpointing, but one is a PEFT/LoRA effectiveness issue and the other is a missing model flag." + "reason": "Different gradient-checkpointing issues: LoRA ineffectiveness vs a missing supports_gradient_checkpointing flag." }, { "left": "issue:41553", "right": "issue:44991", "accept": false, - "reason": "Different tokenizer-loading bugs affecting different models and symptoms." + "reason": "Different tokenizer-loading regressions for different models; no shared concrete bug is evident." }, { "left": "issue:43940", "right": "issue:45313", "accept": false, - "reason": "Related DeepSpeed weight-loading failures, but the affected models and failure descriptions are not specific enough to treat as the same bug." + "reason": "Both are ZeRO-3 load failures, but for different Qwen generations and likely different model-specific paths." }, { "left": "issue:43854", "right": "issue:44863", "accept": false, - "reason": "Different model-loading failures for unrelated architectures." + "reason": "GLM-4.7-Flash loading in tests and NemotronH checkpoint loading are unrelated model-specific bugs." }, { "left": "issue:44928", "right": "issue:45588", "accept": false, - "reason": "Different bugs: RLHF NaNs from position_ids/SDPA fallback vs flash_attention.py crash on s_aux=None." + "reason": "Qwen3.5 RLHF NaNs from 3D position_ids and flash_attention s_aux=None crashes are different failures." }, { "left": "issue:41093", "right": "issue:44805", "accept": true, - "reason": "Same mask-shape mismatch IndexError pattern; 44805 looks like another report of the same underlying indexing bug." + "reason": "Same shape-mismatch IndexError pattern on mask vs indexed tensor; likely the same off-by-one bug." }, { "left": "issue:44479", "right": "issue:44560", - "accept": true, - "reason": "Same v5.3.0 video-input regression in the Qwen/VL family; 44560 is a specific StopIteration manifestation of the broader issue." + "accept": false, + "reason": "Both are Qwen video issues, but one is a broad regression report and the other is a specific StopIteration failure; not clearly the same bug." }, { "left": "issue:41553", "right": "issue:44488", "accept": false, - "reason": "Different tokenizer/model-loading complaints for different checkpoints." + "reason": "Voxtral tokenizer error-message bug and EMBEDDIA/est-roberta tokenizer loading failure are different issues." }, { "left": "issue:44863", "right": "issue:46402", "accept": false, - "reason": "Unrelated model-load failures for different architectures." + "reason": "NemotronH checkpoint loading and Sam3Model text encoder weights not loading are different model/load paths." }, { "left": "issue:44530", "right": "issue:44863", "accept": false, - "reason": "Different Qwen3.5 cache/attention crash vs NemotronH checkpoint loading." + "reason": "PagedAttentionCache linear_attention crash and NemotronH checkpoint loading are unrelated." }, { "left": "issue:43596", "right": "issue:46515", "accept": false, - "reason": "Different issues: ZeRO-3/BertModel index error vs a generic torch 2.4.1 incompatibility." + "reason": "ZeRO-3/BertModel index error is not the same as a general torch 2.4.1 incompatibility." }, { "left": "issue:43366", "right": "issue:43828", "accept": false, - "reason": "Different bugs: GGUF gpt-oss support vs Phi-tiny-MoE autocast dtype mismatch." + "reason": "GGUF gpt-oss support request and Phi-tiny-MoE autocast dtype mismatch are unrelated." }, { "left": "issue:45325", "right": "issue:45902", "accept": false, - "reason": "Both are rotary/position-related, but the model families and failure modes differ." + "reason": "Qwen2.5-VL rope scaling bug and Qwen3.5-MoE inv_freq materialization bug are different code paths." }, { "left": "issue:43638", "right": "issue:46515", "accept": false, - "reason": "Different bugs: BertModel ZeRO-3 index error vs torch version incompatibility." + "reason": "Non-pretrained Bert zero3 index error and torch-version incompatibility are different problems." }, { "left": "issue:43366", "right": "issue:43531", "accept": false, - "reason": "Different Qwen/GGUF feature requests or bugs with no shared concrete failure." + "reason": "GGUF gpt-oss support and Qwen3-MoE sliding_window behavior are unrelated." }, { "left": "issue:45663", "right": "issue:45823", "accept": false, - "reason": "Same model family, but training-time FSDP2 KeyError and device_map auto-placement mismatch are different bugs." + "reason": "Gemma4 FSDP2 shared_kv_states KeyError and Gemma4 PLE device mismatch are distinct bugs." }, { "left": "issue:43716", "right": "issue:44560", "accept": false, - "reason": "Different regressions: Mistral-3 image-preprocessor dtype mismatch vs Qwen3-VL video StopIteration." + "reason": "Mistral-3 image dtype mismatch and Qwen video StopIteration are different model/input paths." }, { "left": "issue:44155", "right": "issue:45584", "accept": false, - "reason": "Different audio-model bugs with different symptoms and code paths." + "reason": "AudioFlamingo3 track leakage and Whisper empty-transcription failure are unrelated audio bugs." + }, + { + "left": "issue:43366", + "right": "issue:44589", + "accept": false, + "reason": "GGUF architecture support and missing Float8 storage type are different compatibility issues." }, { "left": "issue:43824", "right": "issue:43957", "accept": false, - "reason": "Different failures: missing exported class vs meta-device loading regression." + "reason": "Missing Qwen2_5_VL export and meta-device loading regressions are different failures." + }, + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "Qwen3.5 PagedAttentionCache crash and NemotronH checkpoint loading are unrelated." + }, + { + "left": "issue:43854", + "right": "issue:45923", + "accept": false, + "reason": "GLM-4.7-Flash test loading and missing gradient checkpointing support are unrelated." + }, + { + "left": "issue:43950", + "right": "issue:45834", + "accept": false, + "reason": "Non-persistent buffer corruption in from_pretrained and Kosmos2.5 long OCR index errors are unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays appropriately conservative. The canonical anchor choice for issue:43957 is reasonable as the broadest loading-regression report, and the accepted soft pairs are all plausible duplicates/near-duplicates based on the issue titles. The rejections also avoid overstating similarity for merely related model-family or symptom overlaps." + "feedback": "The summary is grounded in the packet: it correctly identifies the cluster as mostly heterogeneous, notes there are no PRs, and keeps the docs v5 pipeline pair as the canonical issue. The soft-edge accepts are conservative enough and the false pairs are mostly well-justified." }, "evaluator_used": true, "retried": false, @@ -1886,8 +1868,8 @@ "issue:42915|issue:43957", "issue:44683|issue:46515", "issue:43645|issue:43646", - "issue:44560|issue:44918", "issue:43550|issue:44910", + "issue:44560|issue:44918", "issue:42617|issue:43366", "issue:36010|issue:42915", "issue:45812|issue:46489", @@ -1902,8 +1884,8 @@ "issue:44661|issue:44805", "issue:43317|issue:44756", "issue:38175|issue:39692", - "issue:42491|issue:43720", "issue:43644|issue:44534", + "issue:42491|issue:43720", "issue:41720|issue:42915", "issue:42915|issue:44756", "issue:42491|issue:45276", @@ -1912,237 +1894,238 @@ "issue:44960|issue:44977", "issue:45313|issue:45357", "issue:45507|issue:46468", - "issue:44488|issue:44991" + "issue:44488|issue:44991", + "issue:43644|issue:43645" ], - "prepared_review_unit_hash": "e1ec07567411592f8085f8af1b1e7155a5f828b28db843ed0f9078e29be13db2", + "prepared_review_unit_hash": "30bf4b7061086c6432fa081174d0cc1ff8208008a3414237efe4686c4ac147b6", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 31, - "serialized_chars": 36162, - "estimated_input_tokens": 9041, - "estimated_eval_tokens": 18338 + "soft_pair_count": 32, + "serialized_chars": 36460, + "estimated_input_tokens": 9115, + "estimated_eval_tokens": 18486 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a heterogeneous grab-bag of Transformers v5 regressions, with only a couple of clear duplicate pairs (custom model initialization; non-persistent buffers). The rest are model- or feature-specific failures that do not collapse into one shared bug theme.", - "confidence": 0.57, - "canonical_issue_reason": null, + "summary": "Mostly a noisy cluster of transformer regression reports across unrelated models/code paths. I only found a few credible duplicates: the non-persistent-buffer junk regression, the custom-model initialization regression, and the SigLIP2 bad-output reports.", + "confidence": 0.81, + "canonical_issue_reason": "issue:43644 is the clearest anchor: it is the earlier, concrete report for the non-persistent-buffer corruption regression, and issue:44534 reads like the same bug with nearly identical symptom wording.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43646 is the best available anchor only in a narrow sense: it is the broadest of the custom-model-init reports, but the cluster is too mixed for it to serve as a strong true canonical representative.", + "best_issue_reason": "43644 is the strongest representative issue because it is specific, broad enough to describe the underlying regression, and has the cleanest duplicate counterpart.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44560", "right": "issue:45685", "accept": false, - "reason": "Different bugs and model paths: Qwen3-VL video StopIteration vs MoE MPS histogram failure." + "reason": "Different models and failure modes; same broad ML area only." }, { "left": "issue:42915", "right": "issue:43957", "accept": false, - "reason": "Both are loading regressions, but they target different mechanisms and models; not the same underlying bug." + "reason": "Both involve loading/runtime failures, but the causes and affected paths are unrelated." }, { "left": "issue:44683", "right": "issue:46515", "accept": false, - "reason": "Separate torch-version incompatibilities affecting different code paths, with no evidence of a shared root cause." + "reason": "Different torch-version compatibility problems on different code paths." }, { "left": "issue:43645", "right": "issue:43646", "accept": true, - "reason": "Same Transformers v5 custom model initialization regression; 43645 looks like a narrower notebook-specific report of the broader 43646 failure." + "reason": "Both describe the same Transformers v5 custom-model initialization regression; 43645 is the notebook-specific manifestation of 43646's broader bug." }, { - "left": "issue:44560", - "right": "issue:44918", + "left": "issue:43550", + "right": "issue:44910", "accept": false, - "reason": "Different model families and different failure modes; not a duplicate." + "reason": "Same backend family, but different models and distinct failure descriptions; not enough to treat as one bug." }, { - "left": "issue:43550", - "right": "issue:44910", + "left": "issue:44560", + "right": "issue:44918", "accept": false, - "reason": "Different concrete bugs: torch.compile/SDPA on Bamba vs flash-attention position_ids handling on Qwen3.5." + "reason": "Different Qwen3-related failures with different symptoms and code paths." }, { "left": "issue:42617", "right": "issue:43366", "accept": false, - "reason": "Unrelated topics: 3d_parallel.py runtime failure vs GGUF support for gpt-oss." + "reason": "Unrelated model/run failures; no shared underlying bug." }, { "left": "issue:36010", "right": "issue:42915", "accept": false, - "reason": "GenerationMixin import error is unrelated to the Qwen3-MoE FP8 loading failure." + "reason": "ImportError for GenerationMixin versus Qwen3Moe FP8 loading failure are unrelated." }, { "left": "issue:45812", "right": "issue:46489", "accept": false, - "reason": "Both concern tokenization, but on different models with no clear evidence of the same code-path bug." + "reason": "Both concern tokenization regressions, but they affect different model families and different failure modes." }, { "left": "issue:45823", "right": "issue:46515", "accept": false, - "reason": "Gemma4 device mismatch and a generic torch 2.4.1 incompatibility are not the same issue." + "reason": "Device-map forward mismatch and torch 2.4.1 incompatibility are not the same bug." }, { "left": "issue:44661", "right": "issue:45823", "accept": false, - "reason": "Tokenizer-mapping registration failure vs Gemma4 forward device-mismatch are different problems." + "reason": "Tokenizer-mapping bug versus Gemma4 device mismatch are unrelated." }, { "left": "issue:44560", "right": "issue:44805", "accept": false, - "reason": "Different error signatures and unrelated model paths." + "reason": "Different Qwen3-vl / masking failures with no shared root cause." }, { "left": "issue:38175", "right": "issue:43994", - "accept": false, - "reason": "Same checkpoint family and similar bad-output symptoms, but the packet does not give enough evidence to confidently treat them as the same underlying bug." + "accept": true, + "reason": "Same specific SigLIP2 checkpoint and same bad inference behavior; 43994 is the broader report of the same underlying model-output regression." }, { "left": "issue:43632", "right": "issue:45834", "accept": false, - "reason": "Different regressions: _is_hf_initialized flag handling vs long-input OCR index error." + "reason": "Different regressions: HF init flag handling versus long-OCR index error." }, { "left": "issue:44877", "right": "issue:45030", "accept": false, - "reason": "Both are config-related loading failures, but they are model-specific and not clearly the same bug." + "reason": "Both are config-validation issues, but they concern different models and likely different validation paths." }, { "left": "issue:43638", "right": "issue:44661", "accept": false, - "reason": "ZeRO-3 Bert init failure and tokenizer-mapping registration failure are different code paths." + "reason": "DeepSpeed ZeRO-3/Bert initialization issue versus tokenizer-mapping failure are unrelated." }, { "left": "issue:43596", "right": "issue:44661", "accept": false, - "reason": "Different bug classes: ZeRO-3 initialization vs tokenizer mapping logic." + "reason": "Different Bert/ZeRO-3 crash versus add-new-model-like tokenizer mapping bug." }, { "left": "issue:44661", "right": "issue:44805", "accept": false, - "reason": "Tokenizer registration failure and tensor mask shape IndexError are unrelated." + "reason": "Tokenizer-mapping logic and mask-shape indexing are unrelated." }, { "left": "issue:43317", "right": "issue:44756", "accept": false, - "reason": "Different memory/offload problems with different triggers and symptoms." + "reason": "Both are load/memory related, but one is device_map offload and the other is Strix Halo mmap OOM." }, { "left": "issue:38175", "right": "issue:39692", "accept": false, - "reason": "Same model family, but one is output quality and the other is a docs/example mismatch plus quantization failure; not enough to merge." - }, - { - "left": "issue:42491", - "right": "issue:43720", - "accept": false, - "reason": "Different issues: Qwen3-MoE LoRA compatibility vs packed-weight loading for BitNet." + "reason": "Same SigLIP2 area, but one is incorrect outputs while the other is a doc example / processor mismatch issue." }, { "left": "issue:43644", "right": "issue:44534", "accept": true, - "reason": "Same non-persistent-buffer junk-fill regression, just described across different versions." + "reason": "Near-verbatim reports of the same non-persistent-buffer junk regression in Transformers v5." + }, + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "Different model families and different loading semantics; not one bug." }, { "left": "issue:41720", "right": "issue:42915", "accept": false, - "reason": "Different Qwen3-related failures with distinct triggers and stack traces." + "reason": "Qwen3 auto device-map crash is distinct from the Qwen3Moe FP8 issue." }, { "left": "issue:42915", "right": "issue:44756", "accept": false, - "reason": "FP8 loading failure and Strix Halo mmap/OOM are not the same bug." + "reason": "No shared code path; one is FP8/Qwen3Moe, the other is mmap/OOM on a specific platform." }, { "left": "issue:42491", "right": "issue:45276", "accept": false, - "reason": "LoRA compatibility regression is unrelated to Gemma4 resize_token_embeddings behavior." + "reason": "LoRA HF4/HF5 compatibility is unrelated to Gemma4 embedding resize behavior." }, { "left": "issue:46459", "right": "issue:46512", "accept": false, - "reason": "Both involve dtype/state handling, but one is ignored dtype on loading and the other is stale config.dtype after from_config; related but distinct." + "reason": "Both involve dtype handling, but one is AutoModel loading of composite checkpoints and the other is from_config model construction; different bugs." }, { "left": "issue:44451", "right": "issue:44488", "accept": false, - "reason": "Both are model-loading failures, but for different checkpoints and without evidence of a shared root cause." + "reason": "Both are model-loading failures, but the affected models and reported symptoms are not specific enough to conclude a single underlying bug." }, { "left": "issue:44960", "right": "issue:44977", "accept": false, - "reason": "Different models and different flash-attention behavior; not a duplicate." + "reason": "Different model families and different flash-attention symptoms." }, { "left": "issue:45313", "right": "issue:45357", "accept": false, - "reason": "Different Qwen3.5 regressions: ZeRO-3 weight loading vs save_pretrained key handling." + "reason": "One is ZeRO-3 weight loading, the other is save_pretrained key serialization; different directions and likely different fixes." }, { "left": "issue:45507", "right": "issue:46468", "accept": false, - "reason": "GraniteMoEHybrid invalid method call and broken Nemotron example are unrelated." + "reason": "GraniteMoEHybrid invalid method call versus Nemotron example breakage are unrelated." }, { - "left": "issue:45812", - "right": "issue:46489", + "left": "issue:44488", + "right": "issue:44991", "accept": false, - "reason": "Tokenizer regressions on different models are too broad to call the same bug." + "reason": "Both are tokenizer/model loading regressions, but they involve different repos and no clear shared root cause." }, { - "left": "issue:44488", - "right": "issue:44991", + "left": "issue:43644", + "right": "issue:43645", "accept": false, - "reason": "Different model/tokenizer load failures; same general symptom only." + "reason": "Non-persistent-buffer corruption and custom-model initialization are separate regressions." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The overall takeaway is mostly grounded, but the canonicalization is too strong for a clearly heterogeneous cluster: the packet mixes unrelated regressions (custom init, non-persistent buffers, tokenizer loading, device-mapping, flash-attention, etc.), so 43646 is not well-supported as the representative of a single main shared theme. Also, the 38175\u219443994 acceptance overstates the evidence a bit; the titles show the same model family and similar bad-output symptoms, but not enough to confidently call them the same underlying bug from the packet alone. The other soft-edge calls look more conservative." + "accept": true, + "feedback": "Grounded and appropriately conservative overall. The three accepted duplicates are well supported by the packet (near-identical non-persistent-buffer reports, the custom-model initialization reports, and the SigLIP2 output regressions), and the non-duplicate verdicts stay cautious without overclaiming." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -2172,8 +2155,6 @@ "issue:43404", "issue:43425", "issue:43454", - "issue:43644", - "issue:43645", "issue:43708", "issue:43854", "issue:43856", @@ -2190,6 +2171,7 @@ "issue:44402", "issue:44488", "issue:44756", + "issue:44863", "issue:44910", "issue:44912", "issue:44960", @@ -2201,16 +2183,17 @@ "issue:45488", "issue:45588", "issue:45663", + "issue:45685", "issue:45704", "issue:45759", "issue:45812", "issue:45823", "issue:45859", - "issue:46050" + "issue:46050", + "issue:46468" ], "soft_pairs": [ "issue:43708|issue:44075", - "issue:43644|issue:43645", "issue:43856|issue:43873", "issue:39692|issue:43994", "issue:43856|issue:44387", @@ -2228,8 +2211,8 @@ "issue:44402|issue:44488", "issue:43873|issue:44387", "issue:42947|issue:45704", - "issue:41720|issue:44155", "issue:43404|issue:43454", + "issue:41720|issue:44155", "issue:39692|issue:44960", "issue:45663|issue:45759", "issue:44291|issue:44756", @@ -2241,241 +2224,252 @@ "issue:43957|issue:44756", "issue:43335|issue:45663", "issue:42947|issue:43381", - "issue:45276|issue:45292" + "issue:45276|issue:45292", + "issue:44863|issue:46468", + "issue:45685|issue:46468" ], - "prepared_review_unit_hash": "5c2a4fc9c4ac0ae3698fbc94b56fff92f371f9308ec021bac4f0fdca62c62e18", + "prepared_review_unit_hash": "913b32883bb03f6258268a51095035fe923b72d05ec168a2a9e04218a8f2e4a2", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 33, - "serialized_chars": 37134, - "estimated_input_tokens": 9284, - "estimated_eval_tokens": 18824 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37952, + "estimated_input_tokens": 9488, + "estimated_eval_tokens": 19232 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is a loose collection of Transformers issues spanning tokenization, model loading, quantization, training, and generation. A few pairs are superficially similar, but none look like the same concrete bug/change closely enough to merge as duplicates.", - "confidence": 0.59, - "canonical_issue_reason": "If forced to choose one representative, issue 45292 is the most generic and central among the closest related reports (the resize_token_embeddings family), while the others are more model-specific variants.", + "summary": "This cluster is mostly heterogeneous model-specific bugs; only the resize_token_embeddings/output_embeddings pair looks like a true duplicate. The rest are related only at a broad subsystem level, not the same underlying bug.", + "confidence": 0.84, + "canonical_issue_reason": "issue:45292 is the cleanest generic statement of the shared resize_token_embeddings/output_embeddings regression; issue:45276 appears to be the Gemma4-specific manifestation of the same bug.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45292 is the best single anchor because it states the core resize_token_embeddings symptom in the broadest way and appears closest to multiple related reports.", + "best_issue_reason": "issue:45292 is the best representative because it captures the core API regression succinctly and broadly, making it the strongest dedupe anchor.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43708", "right": "issue:44075", "accept": false, - "reason": "Different bugs: checkpoint resume step counting vs optimizer argument handling." - }, - { - "left": "issue:43644", - "right": "issue:43645", - "accept": false, - "reason": "Different failures: buffer initialization junk vs Jupyter custom-model initialization regression." + "reason": "Different bugs: checkpoint-resume step calculation vs SGD arguments not being applied." }, { "left": "issue:43856", "right": "issue:43873", "accept": false, - "reason": "Both touch Qwen3/MoE/quantization, but one is training memory inefficiency and the other is offloading not working; not the same bug." + "reason": "Both involve Qwen3/moe or memory, but one is training-memory inefficiency and the other is quantization/offloading behavior; not the same bug." }, { "left": "issue:39692", "right": "issue:43994", "accept": false, - "reason": "Both involve SigLIP2, but one is a broken documentation example and the other is incorrect model/pipeline outputs." + "reason": "Both mention SigLIP2, but one is a broken documentation example while the other is incorrect model/pipeline outputs." }, { "left": "issue:43856", "right": "issue:44387", "accept": false, - "reason": "Both are memory-related, but the underlying problems differ: MoE training memory use vs int4 quantization OOM regression." + "reason": "Different failure modes: Qwen3 MoE training memory usage vs generic int4 quantization CUDA reserved-memory/OOM regression." }, { "left": "issue:43854", "right": "issue:43931", "accept": false, - "reason": "Different model families and failure modes; one is a unit-test loading issue, the other is a Qwen3-VL weight-shape mismatch." + "reason": "Both are load failures, but they affect different models and appear to have different root causes." }, { "left": "issue:43257", "right": "issue:43854", "accept": false, - "reason": "Distinct model/load paths: Qwen3 MoE deepspeed conversion vs GLM-4.7 unit-test loading." + "reason": "Different models and mechanisms: Qwen3 MoE conversion with accelerate/deepspeed vs GLM-4.7-Flash loading in tests." }, { "left": "issue:42491", "right": "issue:45292", "accept": false, - "reason": "LoRA/qwen3_moe compatibility regression is unrelated to resize_token_embeddings behavior." + "reason": "LoRA/qwen3_moe hf4.x vs resize_token_embeddings on output embeddings are unrelated bugs." }, { "left": "issue:42491", "right": "issue:43931", "accept": false, - "reason": "Different model and bug type: LoRA compatibility vs model weight-shape mismatch." + "reason": "Different issues in different model families and code paths." }, { "left": "issue:44960", "right": "issue:46050", "accept": false, - "reason": "Different serialization problems: GLM5 loading vs sinq quantized save/load corruption." + "reason": "Both concern quantized-model persistence, but they involve different quantization flows/models and not the same underlying bug." }, { "left": "issue:45488", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different model families and appear to stem from different code paths." + "reason": "Both are tokenizer regressions, but they affect different model families and different failure modes (pre-tokenizer override vs wrong token IDs)." }, { "left": "issue:44164", "right": "issue:44291", "accept": false, - "reason": "Different loading/save mechanisms: extra_state handling vs init_empty_weights argument compatibility." + "reason": "Different save/load bugs: extra_state handling vs init_empty_weights/_is_hf_initialized argument handling." }, { "left": "issue:45362", "right": "issue:45406", "accept": false, - "reason": "Different runtime paths and models: Transformers chat on Qwen3.5 vs transformers serve on Gemma4Processor." + "reason": "Different components and failures: chat crash for Qwen3.5-35B vs serve crash due to missing Gemma4Processor._tokenizer." }, { "left": "issue:45276", "right": "issue:45335", "accept": false, - "reason": "Same API surface, but different architectures and symptoms; not enough evidence of a single shared bug." + "reason": "Same broad API area, but the titles indicate different model classes and different embedding paths; not enough evidence of one shared duplicate bug." }, { "left": "issue:44910", "right": "issue:45588", "accept": false, - "reason": "Different Flash Attention failures: packed-sequence position_ids misinterpretation vs s_aux=None AttributeError." + "reason": "Both are Flash Attention-related, but one is a packed-sequence/position_ids illegal access and the other is an AttributeError on s_aux=None." }, { "left": "issue:43425", "right": "issue:43957", "accept": false, - "reason": "One is Torch 2.10 compatibility generally, the other is meta-device model loading on specific models." + "reason": "Unrelated load/runtime issues: torch version compatibility vs meta-device loading breakage." }, { "left": "issue:44402", "right": "issue:44488", "accept": false, - "reason": "Both are tokenization/model-loading complaints, but they concern different models and specific symptoms." + "reason": "Different model-specific vocabulary/loading problems; not the same bug." }, { "left": "issue:43873", "right": "issue:44387", "accept": false, - "reason": "Both mention quantization and memory, but one is offloading behavior and the other is a CUDA reserved-memory regression." + "reason": "Different memory problems caused by different features (offloading/quantization vs int4 reserved memory/OOM)." }, { "left": "issue:42947", "right": "issue:45704", "accept": false, - "reason": "Gradient checkpointing/LoRA bug is unrelated to the T5 apex RMSNorm memory leak." + "reason": "Different problems: LoRA gradient checkpointing vs T5 apex RMSNorm memory leak." }, { - "left": "issue:41720", - "right": "issue:44155", + "left": "issue:43404", + "right": "issue:43454", "accept": false, - "reason": "Unrelated failures: Qwen3 auto device-map CUDA assert vs AudioFlamingo3 batched inference leakage." + "reason": "Both are missing lm_head tying, but they are model-specific implementations with different affected code paths; too broad to treat as the same duplicate bug." }, { - "left": "issue:43404", - "right": "issue:43454", + "left": "issue:41720", + "right": "issue:44155", "accept": false, - "reason": "Both mention lm_head tying, but they are model-specific reports on different architectures and not clearly the same defect." + "reason": "Different models and failure modes: Qwen3 auto device mapping CUDA assert vs AudioFlamingo3 batched inference leakage." }, { "left": "issue:39692", "right": "issue:44960", "accept": false, - "reason": "Different categories entirely: SigLIP2 docs/model example vs GLM5 model loading." + "reason": "Documentation/example errors vs quantized save/load behavior are unrelated." }, { "left": "issue:45663", "right": "issue:45759", "accept": false, - "reason": "Different model families and code paths: Gemma4 FSDP2 attention bug vs Qwen multimodal config unwrapping." + "reason": "Different model families and symptoms: Gemma4 FSDP2 attention bug vs Qwen multimodal config unwrap issue." }, { "left": "issue:44291", "right": "issue:44756", "accept": false, - "reason": "Unexpected init argument during empty-weights loading is unrelated to disabling mmap for Strix Halo OOM." + "reason": "Different loading problems: unexpected init argument vs mmap/OOM on Strix Halo." }, { "left": "issue:36010", "right": "issue:39404", "accept": false, - "reason": "Import error for GenerationMixin and Whisper return_language regression are separate issues." + "reason": "ImportError for GenerationMixin is unrelated to Whisper return_language pipeline behavior." }, { "left": "issue:45663", "right": "issue:45859", "accept": false, - "reason": "Gemma4 FSDP2 attention state handling is unrelated to Qwen3.5 MoE rotary embedding CPU offload." + "reason": "Gemma4 FSDP2 attention state bug vs Qwen3.5-MoE rotary embedding CPU offload incompatibility." }, { "left": "issue:44387", "right": "issue:44756", "accept": false, - "reason": "Int4 quantization memory regression and mmap OOM avoidance are different problems." + "reason": "Different resource issues and different triggers: int4 quantization memory growth vs mmap/OOM on Strix Halo." }, { "left": "issue:44292", "right": "issue:44912", "accept": false, - "reason": "Different quantization issues: init_empty_weights loading bug vs MXFP4 fallback to bf16." + "reason": "Different quantized-model load failures: sinq save/load breakage vs MXFP4 fallback to bf16." }, { "left": "issue:44155", "right": "issue:45823", "accept": false, - "reason": "AudioFlamingo3 batch inference leakage is unrelated to Gemma4 PLE device-mismatch forward errors." + "reason": "AudioFlamingo3 cross-track leakage vs Gemma4 PLE device mismatch are unrelated." }, { "left": "issue:43957", "right": "issue:44756", "accept": false, - "reason": "Meta-device model loading failures and Strix Halo mmap OOM are not the same underlying bug." + "reason": "Meta-device loading breakage is unrelated to mmap/OOM behavior." }, { "left": "issue:43335", "right": "issue:45663", "accept": false, - "reason": "SwitchTransformers sparse-layer construction bug is unrelated to Gemma4 FSDP2 attention state rebuilding." + "reason": "SwitchTransformers sparse-layer construction bug vs Gemma4 FSDP2 attention bug are unrelated." }, { "left": "issue:42947", "right": "issue:43381", "accept": false, - "reason": "Both concern gradient checkpointing, but one is ineffective LoRA checkpointing and the other is an eval-mode restriction." + "reason": "Gradient checkpointing ineffective in PEFT LoRA is unrelated to eval-mode restriction." }, { "left": "issue:45276", "right": "issue:45292", + "accept": true, + "reason": "Both describe the same resize_token_embeddings regression where updated token embeddings are not propagated to output embeddings; one is the Gemma4-specific manifestation, the other the generic report." + }, + { + "left": "issue:44863", + "right": "issue:46468", + "accept": false, + "reason": "NemotronH checkpoint-loading failure vs broken official example are different issues." + }, + { + "left": "issue:45685", + "right": "issue:46468", "accept": false, - "reason": "Closely related resize_token_embeddings reports, but the affected model scopes differ and the titles indicate distinct manifestations rather than one exact duplicate." + "reason": "MPS histogram not implemented for Int in MoE is unrelated to the Nemotron example breakage." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The clustering is grounded in the packet and the soft-edge calls are conservative. The only accepted duplicate pair, issue:45276 with issue:45292, is well-supported by the titles and fits a shared resize_token_embeddings/output_embeddings regression. The remaining pairs are reasonably treated as different bugs or at most broad subsystem neighbors." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -2494,6 +2488,7 @@ "nodes": [ "issue:39404", "issue:42796", + "issue:43299", "issue:43381", "issue:43425", "issue:43611", @@ -2505,7 +2500,10 @@ "issue:43854", "issue:43901", "issue:43906", + "issue:43931", "issue:43957", + "issue:43994", + "issue:44079", "issue:44164", "issue:44265", "issue:44402", @@ -2517,7 +2515,6 @@ "issue:44683", "issue:44756", "issue:44805", - "issue:44863", "issue:44877", "issue:44898", "issue:44912", @@ -2530,19 +2527,15 @@ "issue:45588", "issue:45663", "issue:45684", - "issue:45685", "issue:45698", "issue:45701", "issue:45812", "issue:45823", "issue:46315", "issue:46326", - "issue:46468", "issue:46491" ], "soft_pairs": [ - "issue:44863|issue:46468", - "issue:45685|issue:46468", "issue:44479|issue:44877", "issue:44164|issue:44756", "issue:43381|issue:44928", @@ -2565,234 +2558,236 @@ "issue:43425|issue:45823", "issue:43425|issue:43716", "issue:44530|issue:45084", - "issue:44479|issue:44805", "issue:44589|issue:45084", + "issue:44479|issue:44805", "issue:43906|issue:44530", "issue:44265|issue:44898", "issue:43901|issue:44509", "issue:43425|issue:44918", - "issue:39404|issue:44912" + "issue:39404|issue:44912", + "issue:43994|issue:44079", + "issue:43299|issue:43931" ], - "prepared_review_unit_hash": "5fae6cb6382c74d8e7c359faae8794f77237a2375b09814d910d0c44f8cb46d0", + "prepared_review_unit_hash": "a2d5c8a2b16ecc555cbd2ea90423c7f6d575e86adf105b9496443bf06c7862b9", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36463, - "estimated_input_tokens": 9116, - "estimated_eval_tokens": 18488 + "serialized_chars": 37084, + "estimated_input_tokens": 9271, + "estimated_eval_tokens": 18798 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly heterogeneous cluster; only one soft pair looks like the same underlying bug (resize_token_embeddings not updating the corresponding output embeddings). The rest are adjacent-but-distinct regressions or unrelated issues.", - "confidence": 0.92, - "canonical_issue_reason": "issue:45292 is the broader, model-agnostic report of the resize_token_embeddings bug; issue:45335 looks like a specific t5gemma manifestation of the same problem.", + "summary": "This is mostly a heterogeneous grab bag of unrelated Transformers regressions, not a single tight duplicate cluster. The only clear duplicate-looking pair is the resize_token_embeddings / output_embeddings bug, and the Qwen3-VL loader pair is the next-closest family match.", + "confidence": 0.67, + "canonical_issue_reason": "Issue 45292 is the best canonical anchor: it states the generic embedding-resize regression, and 45335 is the model-specific manifestation of the same bug. The rest of the set is too mixed to collapse into one issue.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45292 is the best representative issue in the only true duplicate pair because it states the underlying regression more generally.", + "best_issue_reason": "Issue 45292 is the strongest standalone issue in the set because it is broad, concrete, and directly subsumes its model-specific duplicate. It is the cleanest representative for duplicate triage here.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44863", - "right": "issue:46468", - "accept": false, - "reason": "Different problems: NemotronH checkpoint loading vs an example/script breakage for Nemotron." - }, - { - "left": "issue:45685", - "right": "issue:46468", - "accept": false, - "reason": "MPS histogram limitation in moe code is unrelated to the Nemotron example failure." - }, { "left": "issue:44479", "right": "issue:44877", "accept": false, - "reason": "Video-input regression in Qwen VL models is not the same as strict Granite Speech config loading." + "reason": "Both are model regressions, but one is a Qwen VL video-input issue and the other is Granite Speech config strictness; different bugs." }, { "left": "issue:44164", "right": "issue:44756", "accept": false, - "reason": "save/from_pretrained extra_state handling and Strix Halo mmap OOM are unrelated." + "reason": "Unrelated save/from_pretrained extra_state handling vs mmap/OOM behavior on Strix Halo." }, { "left": "issue:43381", "right": "issue:44928", "accept": false, - "reason": "Gradient checkpointing in eval mode is a different code path from Qwen3.5 RLHF NaN/SDPA fallback." + "reason": "Gradient checkpointing eval-mode behavior and Qwen3.5 RLHF NaNs are different failures with different code paths." }, { "left": "issue:43854", "right": "issue:45216", "accept": false, - "reason": "GLM-4.7 unit-test loading issue and Qwen3.5 save_pretrained regression are distinct model-specific bugs." + "reason": "Both are loading regressions, but they affect different models and different underlying problems." }, { "left": "issue:45684", "right": "issue:45698", "accept": false, - "reason": "Permissions propagation on saved custom-model files is not the same as loading the wrong custom module." + "reason": "Both involve custom save/load flows, but one is permissions propagation and the other is wrong module resolution after save_pretrained." }, { "left": "issue:45588", "right": "issue:45663", "accept": false, - "reason": "Flash-attention aux-state crash and Gemma4 FSDP2 shared_kv_states KeyError are different failures." + "reason": "Flash-attention sink-less crash vs Gemma-4 FSDP2 shared_kv_states rebuild; unrelated code paths." }, { "left": "issue:42796", "right": "issue:46326", "accept": false, - "reason": "Both involve tokenizers, but one is BPE-marker handling and the other is continue_final_message behavior." + "reason": "Both are tokenizer issues, but BPE-marker handling and continue_final_message are distinct bugs." }, { "left": "issue:43646", "right": "issue:43708", "accept": false, - "reason": "Custom model initialization regressions and Trainer resume_from_checkpoint max_steps are unrelated." + "reason": "Custom model initialization in notebooks is unrelated to resume_from_checkpoint max_steps math." }, { "left": "issue:45292", "right": "issue:45335", "accept": true, - "reason": "Same underlying bug: resize_token_embeddings does not propagate to the output/decoder embeddings." + "reason": "Same resize_token_embeddings regression; 45335 is the T5Gemma-specific decoder.embed_tokens manifestation of the general bug." }, { "left": "issue:43828", "right": "issue:43957", "accept": false, - "reason": "Both mention dtype issues, but autocast mismatch and meta-device loading are different concrete bugs." + "reason": "Autocast dtype mismatch on Phi-tiny-MoE and meta-device loading failures are different issues." }, { "left": "issue:43645", "right": "issue:44164", "accept": false, - "reason": "Notebook custom-model initialization and extra_state save/load handling are separate issues." + "reason": "Notebook-defined custom model init and extra_state serialization are separate custom-model problems." }, { "left": "issue:44402", "right": "issue:44451", "accept": false, - "reason": "Tokenizer vocab-size mismatch and a model-loading failure for ScandiBERT are not the same bug." + "reason": "Tokenizer vocab-size mismatch and ScandiBERT loading failure are not the same underlying bug." }, { "left": "issue:43425", "right": "issue:43828", "accept": false, - "reason": "Torch 2.10 incompatibility is broader than the specific autocast dtype mismatch here." + "reason": "Torch 2.10 incompatibility is broader/version-specific, while Phi-tiny-MoE dtype mismatch is model-specific." }, { "left": "issue:44589", "right": "issue:44756", "accept": false, - "reason": "Float8 storage lookup failure and Strix Halo mmap/OOM are unrelated runtime issues." + "reason": "Float8 storage lookup failure and Strix Halo mmap OOM are unrelated." }, { "left": "issue:45812", "right": "issue:46315", "accept": false, - "reason": "Granite token-ID regression and slow convert_tokens_to_ids are different tokenizer regressions." + "reason": "Wrong Granite token IDs and slow convert_tokens_to_ids are both tokenizer regressions, but not the same bug." }, { "left": "issue:46315", "right": "issue:46491", "accept": false, - "reason": "Tokenizer performance regression and CodeLlama leading-space round-trip bug are unrelated." + "reason": "Performance regression in token lookup vs CodeLlama space round-trip correctness are different problems." }, { "left": "issue:44683", "right": "issue:45823", "accept": false, - "reason": "Compiled flex_attention failure on torch>=2.9 is unrelated to Gemma4 PLE device mismatch." + "reason": "Compiled flex_attention failure and Gemma4 device-map mismatch do not share a concrete code-path bug." }, { "left": "issue:39404", "right": "issue:45701", "accept": false, - "reason": "Whisper return_language pipeline regression is unrelated to tokenization changes in v5." + "reason": "Whisper pipeline return_language and generic tokenization changes are unrelated." }, { "left": "issue:43611", "right": "issue:43645", "accept": false, - "reason": "Both concern custom-model regressions, but loading base_model_prefix models and Jupyter initialization are distinct code paths." + "reason": "base_model_prefix loading breakage and notebook custom-model initialization are different custom-model regressions." }, { "left": "issue:43425", "right": "issue:45823", "accept": false, - "reason": "Torch 2.10 incompatibility and Gemma4 PLE device mismatch are not the same underlying issue." + "reason": "Torch compatibility and Gemma4 PLE device mismatch are not the same issue." }, { "left": "issue:43425", "right": "issue:43716", "accept": false, - "reason": "Torch version incompatibility is different from the Mistral-3 image preprocessor dtype mismatch." + "reason": "Torch version incompatibility and Mistral-3 dtype mismatch are distinct failures." }, { "left": "issue:44530", "right": "issue:45084", "accept": false, - "reason": "PagedAttentionCache linear_attention crash and non-template-node compile error are unrelated." + "reason": "Qwen3.5 PagedAttentionCache crash and non-template-node compile TypeError are unrelated." }, { - "left": "issue:44479", - "right": "issue:44805", + "left": "issue:44589", + "right": "issue:45084", "accept": false, - "reason": "Qwen VL video regression and a mask shape IndexError are different failures." + "reason": "Float8 storage lookup and compile-time template-node errors do not match." }, { - "left": "issue:44589", - "right": "issue:45084", + "left": "issue:44479", + "right": "issue:44805", "accept": false, - "reason": "Float8 storage object lookup and compile error on non-template nodes are unrelated." + "reason": "Video-input regression and mask-shape IndexError are different bugs." }, { "left": "issue:43906", "right": "issue:44530", "accept": false, - "reason": "An isolated reproduction of another issue is not evidence of the same Qwen3.5 PagedAttentionCache bug." + "reason": "Isolated reproduction of another issue and Qwen3.5 cache crash are not clearly the same underlying defect." }, { "left": "issue:44265", "right": "issue:44898", "accept": false, - "reason": "torch.export/exportable-check failures and Perceiver interpolation failures are different issues." + "reason": "torch.export export failures with torch_compilable_check are unrelated to Perceiver interpolation failures." }, { "left": "issue:43901", "right": "issue:44509", "accept": false, - "reason": "Both are docs regressions, but they cover different pipeline APIs and different removed/changed behaviors." + "reason": "Both are docs issues, but they concern different pages and different deprecated/changed pipeline text." }, { "left": "issue:43425", "right": "issue:44918", "accept": false, - "reason": "Torch 2.10 incompatibility is unrelated to Qwen3.5 embedding unpacking in TRL SFT." + "reason": "Torch incompatibility and TRL SFT input-embedding unpacking are different bugs." }, { "left": "issue:39404", "right": "issue:44912", "accept": false, - "reason": "Whisper pipeline return_language regression is unrelated to MXFP4 quantization falling back to bf16." + "reason": "Whisper return_language and MXFP4 quantization fallback are unrelated." + }, + { + "left": "issue:43994", + "right": "issue:44079", + "accept": false, + "reason": "Siglip2 bad outputs and ModelOutput key assignment semantics are different code paths." + }, + { + "left": "issue:43299", + "right": "issue:43931", + "accept": true, + "reason": "Same Qwen3-VL loading regression family; one reports the dev breakage, the other the resulting weight-shape mismatch on a specific checkpoint." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet. The accepted pair (issue:45292 / issue:45335) is strongly supported by the titles and similarity score, and the remaining soft-edge rejections stay conservative rather than overstating equivalence." + "feedback": "The summary is grounded in the packet and stays conservative. The two accepted pairs are reasonably supported by the issue titles and appear to be same-bug family matches, while the rest of the verdicts are appropriately cautious." }, "evaluator_used": true, "retried": false, @@ -2819,9 +2814,9 @@ "issue:39404", "issue:40444", "issue:40990", + "issue:42175", "issue:42796", "issue:42915", - "issue:43299", "issue:43404", "issue:43421", "issue:43452", @@ -2832,12 +2827,10 @@ "issue:43632", "issue:43643", "issue:43644", + "issue:43646", "issue:43716", "issue:43856", "issue:43906", - "issue:43931", - "issue:43994", - "issue:44079", "issue:44164", "issue:44451", "issue:44462", @@ -2849,21 +2842,22 @@ "issue:44683", "issue:44737", "issue:44756", + "issue:44811", "issue:44987", "issue:44991", "issue:45081", "issue:45084", + "issue:45584", "issue:45759", "issue:45859", "issue:46050", + "issue:46178", "issue:46285", "issue:46326", "issue:46402", "issue:46489" ], "soft_pairs": [ - "issue:43994|issue:44079", - "issue:43299|issue:43931", "issue:44671|issue:44991", "issue:36032|issue:43452", "issue:46050|issue:46326", @@ -2878,8 +2872,8 @@ "issue:44462|issue:46489", "issue:42796|issue:43404", "issue:43611|issue:43644", - "issue:39404|issue:44671", "issue:44661|issue:44683", + "issue:39404|issue:44671", "issue:44737|issue:45859", "issue:32090|issue:39404", "issue:43716|issue:45084", @@ -2891,216 +2885,225 @@ "issue:44451|issue:44987", "issue:43541|issue:45084", "issue:42915|issue:44530", - "issue:34567|issue:40444" + "issue:34567|issue:40444", + "issue:42175|issue:46178", + "issue:43644|issue:43646", + "issue:44811|issue:45584" ], - "prepared_review_unit_hash": "296f91d095a077159e313796b6f84f456a4b8e0e75a79b96f09f2e30d4da4d8e", + "prepared_review_unit_hash": "fd2463489eefb34f84e3b5cbf0d53a8d0f54094548b36df76283bcdd87453332", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 30, - "serialized_chars": 36186, - "estimated_input_tokens": 9047, - "estimated_eval_tokens": 18350 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37046, + "estimated_input_tokens": 9262, + "estimated_eval_tokens": 18780 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is broad and mostly heterogeneous: most pairs share only a loose theme (tokenizer/model loading or v5 regressions) but do not appear to be the same underlying bug or change. I found no soft-edge pair strong enough to merge. If one issue must represent the cluster, 43611 is the broadest loading-regression report.", - "confidence": 0.57, - "canonical_issue_reason": "Issue 43611 is the broadest, most representative v5 loading-regression report in the set, and it sits closest to the other loading-related items without being overly model-specific.", + "summary": "This cluster is a loose collection of Transformers v5 regressions, mostly around model/tokenizer loading and initialization, but the soft pairs generally describe distinct bugs rather than one shared fix path.", + "confidence": 0.43, + "canonical_issue_reason": "issue:43646 is the broadest and most representative v5 regression in the set, covering model initialization rather than a narrow model-specific symptom.", "canonical_pr_reason": null, - "best_issue_reason": "43611 is the best single issue to keep as a representative anchor: it is user-facing, broadly scoped, and more central than the narrowly model-specific reports.", + "best_issue_reason": "issue:43646 is the best umbrella issue here because it is the most general of the v5 init/load regressions and most likely to represent the cluster\u2019s theme.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43994", - "right": "issue:44079", + "left": "issue:44671", + "right": "issue:44991", "accept": false, - "reason": "Different failures: SigLIP2 inference quality vs ModelOutput key assignment behavior." + "reason": "Both are tokenizer-load regressions, but they affect different models with different failure modes; not the same underlying bug." }, { - "left": "issue:43299", - "right": "issue:43931", + "left": "issue:36032", + "right": "issue:43452", "accept": false, - "reason": "Both involve Qwen3-VL loading, but one is a version-loading break and the other is a specific weight-shape mismatch; not the same bug." - }, - { - "left": "issue:44671", - "right": "issue:44991", - "accept": false, - "reason": "CamemBERT MLM predictions and EMBEDDIA tokenizer loading are unrelated issues." - }, - { - "left": "issue:36032", - "right": "issue:43452", - "accept": false, - "reason": "Tokenizer method conflict vs gguf_file loading failure are different code paths and symptoms." + "reason": "One is a T5 tokenizer method-name conflict, the other is a gguf/from_pretrained loading issue; unrelated code paths." }, { "left": "issue:46050", "right": "issue:46326", "accept": false, - "reason": "Quantized save/load corruption and mistral_common continue_final_message handling are unrelated." + "reason": "Quantized save/load corruption and Whisper/mistral_common decoding behavior are different bugs." }, { "left": "issue:44164", "right": "issue:45759", "accept": false, - "reason": "extra_state serialization and Qwen multimodal config unwrapping are different problems." + "reason": "extra_state save/load handling and AutoModel text_config unwrapping are separate model-loading problems." }, { "left": "issue:43906", "right": "issue:46326", "accept": false, - "reason": "One is a reproduction of an older issue, the other is a tokenizer final-message bug; no shared underlying defect." + "reason": "A reproduction of an older issue and a tokenizer continue_final_message bug are not the same change or failure." }, { "left": "issue:43596", "right": "issue:44683", "accept": false, - "reason": "Deepspeed zero3/BertModel indexing and torch>=2.9 flex_attention compilation are unrelated." + "reason": "Deepspeed ZeRO3/BertModel initialization and torch 2.9 flex_attention compilation are unrelated." }, { "left": "issue:44991", "right": "issue:45081", "accept": false, - "reason": "Both are tokenizer loading regressions, but they affect different models and different tokenizer code paths." + "reason": "Both involve tokenizer loading, but the specific models and root causes differ." }, { "left": "issue:43856", "right": "issue:44756", "accept": false, - "reason": "Qwen3 MoE memory usage and Apple Silicon mmap OOM are distinct performance/memory issues." + "reason": "MoE memory usage and Strix Halo mmap/OOM handling are different performance issues." }, { "left": "issue:44164", "right": "issue:44479", "accept": false, - "reason": "Serialization of extra_state and video-input regression in Qwen VL models are not the same bug." + "reason": "save/from_pretrained extra_state handling and a video-input regression are unrelated." }, { "left": "issue:44488", "right": "issue:45081", "accept": false, - "reason": "Different tokenizer/model families and different failure modes." + "reason": "Tokenizer load failure and Mistral regex patch crash are different concrete bugs." }, { "left": "issue:44488", "right": "issue:44987", "accept": false, - "reason": "Both are load failures, but the underlying causes are model-specific and not clearly shared." + "reason": "Both are load failures for different repos/models, but there is no evidence of one shared defect." }, { "left": "issue:44462", "right": "issue:46489", "accept": false, - "reason": "Ignoring tokenizer.json and DeepSeek-Coder tokenizer output regression are related only at a high level." + "reason": "Ignoring tokenizer.json and DeepSeek-Coder output regression are different tokenizer problems." }, { "left": "issue:42796", "right": "issue:43404", "accept": false, - "reason": "Tokenizer BPE-marker handling and lm_head weight tying are unrelated." + "reason": "Tokenizer BPE-marker handling and lm_head weight tying are unrelated issues." }, { "left": "issue:43611", "right": "issue:43644", "accept": false, - "reason": "Both are v5 load regressions, but one is base_model_prefix handling and the other is non-persistent buffer initialization; different bugs." + "reason": "Both are v5 regressions, but one is base_model_prefix loading and the other is non-persistent buffer initialization; distinct bugs." }, { - "left": "issue:39404", - "right": "issue:44671", + "left": "issue:44661", + "right": "issue:44683", "accept": false, - "reason": "Whisper pipeline language return and CamemBERT masked-LM output are unrelated." + "reason": "add-new-model-like mapping logic and flex_attention compilation are unrelated." }, { - "left": "issue:44661", - "right": "issue:44683", + "left": "issue:39404", + "right": "issue:44671", "accept": false, - "reason": "Tokenizer mapping registration and flex_attention compilation are different subsystems." + "reason": "Whisper pipeline return_language and CamemBERT masked-LM predictions are different model-specific bugs." }, { "left": "issue:44737", "right": "issue:45859", "accept": false, - "reason": "Both mention device/offload concerns, but they affect different models and separate code paths." + "reason": "XLNet CPU arange placement and Qwen3_5Moe CPU-offload rotary embedding are not the same issue." }, { "left": "issue:32090", "right": "issue:39404", "accept": false, - "reason": "Trainer GPU broadcast TypeError and Whisper pipeline regression are unrelated." + "reason": "Trainer GPU broadcast NoneType error and Whisper return_language regression are unrelated." }, { "left": "issue:43716", "right": "issue:45084", "accept": false, - "reason": "Mistral image dtype mismatch and template compilation error have no common underlying bug." + "reason": "Mistral-3 dtype mismatch and compile non-template nodes are different failures." }, { "left": "issue:46285", "right": "issue:46402", "accept": false, - "reason": "CLIP structure inconsistency and Sam3 text-encoder weight loading are distinct model-specific regressions." + "reason": "CLIP structure inconsistency and Sam3Model weight loading are separate model-architecture regressions." }, { "left": "issue:36032", "right": "issue:42915", "accept": false, - "reason": "T5 tokenizer loading conflict and Qwen3Moe FP8 config failure are unrelated." + "reason": "Tokenizer attribute conflict and Qwen3Moe FP8 configuration failure are unrelated." }, { "left": "issue:43582", "right": "issue:43632", "accept": false, - "reason": "Apple Silicon warmup TypeError and _is_hf_initialized flag breakage are different issues." + "reason": "Apple Silicon warmup TypeError and _is_hf_initialized flag behavior are different problems." }, { "left": "issue:43643", "right": "issue:44530", "accept": false, - "reason": "trust_remote_code config field loss and PagedAttentionCache linear_attention crash are unrelated." + "reason": "trust_remote_code field loss and PagedAttentionCache group-type crash are unrelated." }, { "left": "issue:40990", "right": "issue:43421", "accept": false, - "reason": "High perplexity on gpt-oss and runtime post-processor updates are different problems." + "reason": "High perplexity on a model and tokenizer backend post-processor updates are not the same bug." }, { "left": "issue:44451", "right": "issue:44987", "accept": false, - "reason": "Model load failure for ScandiBERT and load failure for physical-intelligence/fast are different model-specific reports." + "reason": "Both are model load failures, but they concern different repositories and no shared concrete failure is shown." }, { "left": "issue:43541", "right": "issue:45084", "accept": false, - "reason": "Dynamo tracing grouped_mm error and template-node compilation error are unrelated." + "reason": "Torch dynamo grouped_mm tracing and compile non-template nodes are different compiler errors." }, { "left": "issue:42915", "right": "issue:44530", "accept": false, - "reason": "Qwen3Moe FP8 training failure and Qwen3.5 PagedAttentionCache crash do not share the same defect." + "reason": "Qwen3Moe FP8 training and Qwen3.5 PagedAttentionCache crashes are different code paths." }, { "left": "issue:34567", "right": "issue:40444", "accept": false, - "reason": "Trainer token-count tracking and Qwen2.5-VL iterable dataset failure are unrelated." + "reason": "TrainerState token counting and iterable-dataset multi-image finetuning are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:46178", + "accept": false, + "reason": "Missing tensorflow backend in extras and an unsupported tensorflow_text backend mapping are similar packaging areas but not the same bug." + }, + { + "left": "issue:43644", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 init regressions, but non-persistent buffer junk and custom model initialization are distinct failure modes." + }, + { + "left": "issue:44811", + "right": "issue:45584", + "accept": false, + "reason": "Both are Whisper decode-related, but batch_decode skipping special tokens and empty-transcription handling after align_special_tokens are different issues." } ] }, @@ -3128,7 +3131,6 @@ "issue:34567", "issue:36032", "issue:40444", - "issue:42175", "issue:42915", "issue:43089", "issue:43262", @@ -3139,7 +3141,6 @@ "issue:43452", "issue:43482", "issue:43540", - "issue:43644", "issue:43645", "issue:43646", "issue:43742", @@ -3156,47 +3157,45 @@ "issue:44451", "issue:44464", "issue:44466", + "issue:44479", "issue:44488", "issue:44534", "issue:44589", "issue:44610", - "issue:44811", + "issue:44683", "issue:44849", "issue:44987", "issue:44991", "issue:45081", "issue:45084", + "issue:45216", "issue:45325", + "issue:45357", "issue:45507", - "issue:45584", "issue:45987", - "issue:46178", "issue:46326" ], "soft_pairs": [ - "issue:42175|issue:46178", - "issue:43644|issue:43646", - "issue:44811|issue:45584", "issue:44206|issue:44220", "issue:43883|issue:44488", "issue:40444|issue:46326", "issue:36032|issue:43645", "issue:43452|issue:43482", - "issue:43317|issue:43482", "issue:45084|issue:45507", + "issue:43317|issue:43482", "issue:44987|issue:44991", "issue:43262|issue:43377", - "issue:43746|issue:46326", "issue:43646|issue:44534", - "issue:44403|issue:45081", + "issue:43746|issue:46326", "issue:44464|issue:45987", + "issue:44403|issue:45081", "issue:42915|issue:45084", "issue:43645|issue:44534", "issue:43883|issue:44991", "issue:43404|issue:44466", "issue:43994|issue:44610", - "issue:44849|issue:45325", "issue:43377|issue:43819", + "issue:44849|issue:45325", "issue:28530|issue:34567", "issue:44451|issue:45081", "issue:43322|issue:43540", @@ -3205,244 +3204,259 @@ "issue:44265|issue:44610", "issue:36032|issue:43742", "issue:43089|issue:45987", + "issue:42915|issue:44589", "issue:44062|issue:44987", - "issue:42915|issue:44589" + "issue:44220|issue:44683", + "issue:45216|issue:45357", + "issue:44479|issue:45325" ], - "prepared_review_unit_hash": "14af153ea2af38e1277be984768d21980680e93a6235c8720f401a24c079d022", + "prepared_review_unit_hash": "22aae035c4d531ffd45a94e9e6775c8cf75f41819c5064d66310252248af9360", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 33, - "serialized_chars": 37677, - "estimated_input_tokens": 9420, - "estimated_eval_tokens": 19096 + "serialized_chars": 37146, + "estimated_input_tokens": 9287, + "estimated_eval_tokens": 18830 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a noisy mixed cluster of unrelated Transformers issues. There is one clear accepted duplicate pair here: issue:43404 \u2194 issue:44466 around tied-weights / lm_head serialization. A non-persistent-buffer similarity is also plausible, but in this packet it remains only a near-match rather than a confirmed duplicate; the rest of the soft edges are conservative rejects.", - "confidence": 0.81, - "canonical_issue_reason": "issue:43404 is the earlier and clearest anchor for the tied-weights / lm_head regression shared with issue:44466.", + "summary": "Most pairs are only loosely related by subsystem, not by the same underlying bug. The only clear duplicate is the Qwen3.5 `save_pretrained` checkpoint regression pair.", + "confidence": 0.76, + "canonical_issue_reason": "issue:45216 is the best canonical issue: it describes the regression broadly, includes the version boundary, and captures the `save_pretrained` checkpoint corruption that 45357 appears to continue/specialize.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43404 is the strongest representative of the only clearly accepted duplicate pair in this packet.", + "best_issue_reason": "issue:45216 is the strongest representative issue in this set because it states the shared regression most generally and is the cleanest umbrella for the later follow-up report.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:42175", - "right": "issue:46178", - "accept": false, - "reason": "Both are backend/import failures, but they involve different backends and different error modes." - }, - { - "left": "issue:43644", - "right": "issue:43646", - "accept": false, - "reason": "Both are v5 regressions, but the concrete bugs differ: junk non-persistent buffers versus custom model initialization breakage." - }, - { - "left": "issue:44811", - "right": "issue:45584", - "accept": false, - "reason": "Shared Whisper/token handling theme, but one is batch_decode skip_special_tokens behavior and the other is empty-transcription generation after align_special_tokens." - }, { "left": "issue:44206", "right": "issue:44220", "accept": false, - "reason": "Both touch audio feature extraction, but they point to different code paths and triggers." + "reason": "Related audio feature-extraction area, but the titles point to different failures (`center` arg vs `_torch_extract_fbank_features()`), so the root cause is not clearly the same." }, { "left": "issue:43883", "right": "issue:44488", "accept": false, - "reason": "Different model-loading failures for different checkpoints; not the same underlying bug." + "reason": "Both are model-loading complaints, but they target different models and different failure modes; not enough to treat as the same bug." }, { "left": "issue:40444", "right": "issue:46326", "accept": false, - "reason": "Both are multimodal/tokenizer-related at a high level, but the affected paths are unrelated." + "reason": "Different code paths: multimodal iterable-dataset image handling vs tokenizer `continue_final_message` behavior." }, { "left": "issue:36032", "right": "issue:43645", "accept": false, - "reason": "Tokenizer attribute-name collision and custom model initialization failure are separate regressions." + "reason": "Unrelated regressions: tokenizer class method conflict vs custom model initialization in notebooks." }, { "left": "issue:43452", "right": "issue:43482", "accept": false, - "reason": "Both are GGUF-related, but they affect different APIs/models and do not look like one bug." + "reason": "Both involve GGUF/loading, but one is a generic `gguf_file` API break and the other is a specific Qwen2.5-GGUF failure; the common root cause is not clear enough." }, { - "left": "issue:43317", - "right": "issue:43482", + "left": "issue:45084", + "right": "issue:45507", "accept": false, - "reason": "device_map/offload dequantization failure is a different path from the Qwen2.5-GGUF loading failure." + "reason": "Different bugs in different code paths; no strong evidence they are the same underlying issue." }, { - "left": "issue:45084", - "right": "issue:45507", + "left": "issue:43317", + "right": "issue:43482", "accept": false, - "reason": "Different problems: template compilation error versus GraniteMoEHybrid invalid method call." + "reason": "Different failures: device-map/offload dequantized loading vs GGUF loading regression." }, { "left": "issue:44987", "right": "issue:44991", "accept": false, - "reason": "Both are v5 loading regressions, but they affect different models and likely different assumptions." + "reason": "Same version family, but different model/tokenizer load failures and no clear shared concrete bug." }, { "left": "issue:43262", "right": "issue:43377", "accept": false, - "reason": "Audio chat-template sample-rate handling and MIMI batching/padding-mask behavior are unrelated." - }, - { - "left": "issue:43746", - "right": "issue:46326", - "accept": false, - "reason": "PEFT local checkpoint loading and continue_final_message behavior are different bugs." + "reason": "Both audio-related, but one is a sampling-rate default issue and the other is a padding-mask batch consistency bug." }, { "left": "issue:43646", "right": "issue:44534", "accept": false, - "reason": "Both are in the custom-model / v5 regression neighborhood, but they are not the same concrete failure." + "reason": "Different regressions around custom model handling vs non-persistent buffer initialization." }, { - "left": "issue:44403", - "right": "issue:45081", + "left": "issue:43746", + "right": "issue:46326", "accept": false, - "reason": "Unnecessary loading noise and a tokenizer regex patch crash are unrelated." + "reason": "PEFT/local-checkpoint loading for GraniteSpeech is unrelated to the mistral_common tokenizer message-handling issue." }, { "left": "issue:44464", "right": "issue:45987", "accept": false, - "reason": "Chunked-generation inconsistency and StaticCache.get_seq_length() returning a tensor are different runtime issues." + "reason": "Compiled chunked generation inconsistency is not the same as the `StaticCache.get_seq_length()` return-type contract bug." + }, + { + "left": "issue:44403", + "right": "issue:45081", + "accept": false, + "reason": "Different problems: generic load-time noise vs Mistral regex patch crash." }, { "left": "issue:42915", "right": "issue:45084", "accept": false, - "reason": "FineGrainedFP8Config failure and non-template-node compilation are separate implementation errors." + "reason": "Qwen3Moe FP8 config failure and template-node compile error are unrelated." }, { "left": "issue:43645", "right": "issue:44534", "accept": false, - "reason": "Custom model initialization in notebooks versus non-persistent-buffer serialization are similar only at a broad v5 level." + "reason": "Custom model init in Jupyter and junk-filled non-persistent buffers are separate issues." }, { "left": "issue:43883", "right": "issue:44991", "accept": false, - "reason": "Missing all_tied_weights_keys and tokenizer loading failure for a specific model are different issues." + "reason": "Different model-load/tokenizer-load failures with no concrete shared bug." }, { "left": "issue:43404", "right": "issue:44466", - "accept": true, - "reason": "Same tied-weights / lm_head regression: one report says lm_head is untied, the other reports device-dependent lm_head.weight serialization." + "accept": false, + "reason": "Both mention tied weights, but one is a specific model not tying `lm_head` and the other is device-dependent serialization; related theme, not the same bug." }, { "left": "issue:43994", "right": "issue:44610", "accept": false, - "reason": "Different vision/model-input bugs with different symptoms and model paths." + "reason": "Different vision issues: nonsensical outputs vs processor/model input-size mismatch." }, { - "left": "issue:44849", - "right": "issue:45325", + "left": "issue:43377", + "right": "issue:43819", "accept": false, - "reason": "Different Qwen-family regressions affecting different outputs and components." + "reason": "Both are model-output mismatches, but the models and mechanics differ (`padding mask` vs `STE`)." }, { - "left": "issue:43377", - "right": "issue:43819", + "left": "issue:44849", + "right": "issue:45325", "accept": false, - "reason": "MIMI batching/padding-mask mismatch and DAC.from_latents vs forward mismatch are separate correctness issues." + "reason": "Different Qwen regressions affecting different features; not the same code-path problem." }, { "left": "issue:28530", "right": "issue:34567", "accept": false, - "reason": "Early-stopping metric selection and num_input_tokens_seen tracking are unrelated trainer problems." + "reason": "Trainer early-stopping behavior and `num_input_tokens_seen` updates are unrelated." }, { "left": "issue:44451", "right": "issue:45081", "accept": false, - "reason": "Model load failure for a specific checkpoint and a tokenizer regex patch crash are different failure modes." + "reason": "Different failures: loading a specific model vs Mistral regex patch crash." }, { "left": "issue:43322", "right": "issue:43540", "accept": false, - "reason": "Unrelated multimodal crashes in different models and code paths." + "reason": "Different multimodal loading/processing failures; no clear shared underlying bug." }, { "left": "issue:44987", "right": "issue:45081", "accept": false, - "reason": "Model loading regression versus tokenizer regex patch crash; no shared concrete bug path." + "reason": "Tokenizer/model loading regression vs Mistral tokenizer regex crash are distinct issues." }, { "left": "issue:42915", "right": "issue:44291", "accept": false, - "reason": "FineGrainedFP8Config failure and unexpected _is_hf_initialized during init_empty_weights are separate regressions." + "reason": "FP8 configuration failure and `init_empty_weights` argument mismatch are separate loading issues." }, { "left": "issue:44265", "right": "issue:44610", "accept": false, - "reason": "torch.export/exportability failure and processor image-size mismatch are unrelated." + "reason": "torch.export/compilation failure is unrelated to the OmDet-Turbo processor image-size regression." }, { "left": "issue:36032", "right": "issue:43742", "accept": false, - "reason": "Tokenizer method-name collision and loading a specific model key error are different issues." + "reason": "Different model/tokenizer loading failures with no direct evidence of the same defect." }, { "left": "issue:43089", "right": "issue:45987", "accept": false, - "reason": "Generation performance overhead and StaticCache.get_seq_length() returning a tensor are not the same bug." + "reason": "Generation performance overhead and cache-length return-type mismatch are different problems." + }, + { + "left": "issue:42915", + "right": "issue:44589", + "accept": false, + "reason": "Both mention low-level tensor/storage issues, but the affected paths and root causes are different." }, { "left": "issue:44062", "right": "issue:44987", "accept": false, - "reason": "AddedToken keyword conflict and a model-loading regression for a specific checkpoint are unrelated tokenizer/model bugs." + "reason": "AddedToken keyword collision is not the same as the model-loading failure in 44987." }, { - "left": "issue:42915", - "right": "issue:44589", + "left": "issue:44220", + "right": "issue:44683", + "accept": false, + "reason": "Different subsystems: audio fbank extraction vs compiled flex_attention on newer torch." + }, + { + "left": "issue:45216", + "right": "issue:45357", + "accept": true, + "reason": "Same Qwen3.5 `save_pretrained` regression; 45357 looks like a later, more specific report of the same incorrect checkpoint/visual-encoder-key bug." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": false, + "reason": "Related Qwen video/rope-index area, but 44479 is a broad video-input regression across models while 45325 is a specific still-image temporal-position bug; not clearly the same underlying defect." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": false, + "reason": "Related Qwen video/rope-index area, but 44479 is a broad video-input regression across models while 45325 is a specific still-image temporal-position bug; not clearly the same underlying defect." + }, + { + "left": "issue:44466", + "right": "issue:45325", "accept": false, - "reason": "Both are v5-related regressions, but the failure causes are different." + "reason": "Different Qwen regressions: tied-weight serialization vs temporal position scaling." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The lone accepted pair (issue:43404 \u2194 issue:44466) is well grounded, and most rejections are conservative. However, the overall summary overstates the evidence by claiming there are \"only two clear duplicate pairs\": in the provided packet, only one pair is explicitly accepted, and the non-persistent-buffer duplicate is plausible but not directly supported by a listed soft pair or explicit linkage. Please soften that cluster-level conclusion to match the packet." + "accept": true, + "feedback": "The judgment is mostly well grounded in the packet, and the soft-edge calls are conservative overall. The only accepted pair, issue:45216 vs issue:45357, is supported by the titles as the same Qwen3.5 save_pretrained regression family. The summary does not overclaim beyond the evidence provided." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -3485,21 +3499,20 @@ "issue:43818", "issue:43856", "issue:44079", - "issue:44220", "issue:44291", "issue:44403", - "issue:44479", + "issue:44462", "issue:44534", "issue:44568", "issue:44589", - "issue:44683", "issue:44756", "issue:44811", + "issue:44829", "issue:44877", "issue:44991", - "issue:45216", + "issue:45081", "issue:45325", - "issue:45357", + "issue:45479", "issue:45538", "issue:45542", "issue:45663", @@ -3510,17 +3523,14 @@ "issue:46491" ], "soft_pairs": [ - "issue:44220|issue:44683", - "issue:45216|issue:45357", - "issue:44479|issue:45325", "issue:43531|issue:44877", - "issue:44991|issue:46315", "issue:36032|issue:41720", + "issue:44991|issue:46315", "issue:43425|issue:44403", "issue:43611|issue:44534", "issue:43296|issue:43531", - "issue:40444|issue:45325", "issue:43632|issue:44756", + "issue:40444|issue:45325", "issue:45812|issue:46491", "issue:43792|issue:44589", "issue:44403|issue:44991", @@ -3539,233 +3549,233 @@ "issue:42175|issue:45542", "issue:41553|issue:43582", "issue:43531|issue:44291", + "issue:44079|issue:44534", "issue:43818|issue:46402", - "issue:44079|issue:44534" + "issue:44829|issue:45479", + "issue:41720|issue:45081", + "issue:44462|issue:45812" ], - "prepared_review_unit_hash": "ac39e5672b2ea73e8a1dee0c46c39ccc09236788c44f8c48be4f292a2073be47", + "prepared_review_unit_hash": "32d8bed34e6b37a7e0196e159a460adb5065cde8094b11fb135de975e0ab117e", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 31, - "serialized_chars": 36914, - "estimated_input_tokens": 9229, - "estimated_eval_tokens": 18714 + "serialized_chars": 36566, + "estimated_input_tokens": 9142, + "estimated_eval_tokens": 18540 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a heterogeneous issue set with a couple of real duplicate pairs around Qwen/VL save_pretrained and rope-index regressions; the rest are adjacent but distinct model/tokenizer/runtime bugs.", - "confidence": 0.61, - "canonical_issue_reason": "issue:45216 is the clearest duplicate anchor among the accepted overlaps: it describes the base Qwen3.5 save_pretrained regression, and 45357 reads like the same serialization bug resurfacing with incorrect visual-encoder keys.", + "summary": "This cluster is heterogeneous: most items are tokenizer/model-loading regressions or unrelated training/runtime bugs, and the soft-similarity pairs do not look like true duplicates.", + "confidence": 0.34, + "canonical_issue_reason": "issue:44462 is the strongest representative issue in the set: it describes a broad AutoTokenizer regression with repository tokenizer handling, which matches the most central tokenizer-related theme in the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45812 has the broadest impact and strongest regression framing, affecting all Granite models and already drawing inbound references, so it is the best overall representative issue in this set.", + "best_issue_reason": "issue:44462 is the best standalone issue here because it is broad, user-visible, and tied to tokenizer loading behavior with external references, making it the most useful anchor for this mixed cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44220", - "right": "issue:44683", - "accept": false, - "reason": "Different code paths: fbank feature extraction vs compiled flex_attention on newer torch." - }, - { - "left": "issue:45216", - "right": "issue:45357", - "accept": true, - "reason": "Same underlying Qwen3.5 save_pretrained serialization regression; 45357 looks like the same bug with a later symptom around visual encoder keys." - }, - { - "left": "issue:44479", - "right": "issue:45325", - "accept": true, - "reason": "Both point to the same Qwen VL position/rope indexing regression in 5.3.0, with one phrased as the broader video-input failure and the other as the specific still-image temporal-position scaling bug." - }, { "left": "issue:43531", "right": "issue:44877", "accept": false, - "reason": "Both are config/load problems, but one is a sliding_window issue and the other is strict config blocking granite_speech; not the same bug." + "reason": "Different models and failure modes: Qwen3-MoE sliding_window vs Granite speech config loading." }, { - "left": "issue:44991", - "right": "issue:46315", + "left": "issue:36032", + "right": "issue:41720", "accept": false, - "reason": "Tokenizer loading failure vs a performance regression in convert_tokens_to_ids; related subsystem, different issue." + "reason": "Unrelated bugs: T5 tokenizer method conflict vs Qwen3 device-mapping CUDA assert." }, { - "left": "issue:36032", - "right": "issue:41720", + "left": "issue:44991", + "right": "issue:46315", "accept": false, - "reason": "Unrelated: T5 tokenizer method conflict vs Qwen3 auto device mapping CUDA assert." + "reason": "Both are tokenizer regressions, but one is loading failure and the other is a convert_tokens_to_ids performance issue." }, { "left": "issue:43425", "right": "issue:44403", "accept": false, - "reason": "Torch-version incompatibility is not the same as noisy loading output." + "reason": "Torch compatibility is unrelated to noisy model-loading logs." }, { "left": "issue:43611", "right": "issue:44534", "accept": false, - "reason": "Base-model-prefix loading failure and junk-filled non-persistent buffers are different v5 regressions." + "reason": "Different code paths: base_model_prefix loading vs non-persistent buffer initialization." }, { "left": "issue:43296", "right": "issue:43531", "accept": false, - "reason": "PaddleOCR-VL load failure and Qwen3 sliding_window behavior are unrelated." + "reason": "Different model families and different symptoms: PaddleOCR-VL load failure vs Qwen3-MoE sliding_window." }, { - "left": "issue:40444", - "right": "issue:45325", + "left": "issue:43632", + "right": "issue:44756", "accept": false, - "reason": "IterableDataset multi-image finetuning failure is distinct from the Qwen2.5-VL temporal position scaling bug." + "reason": "v5 init flag regression is unrelated to Apple Silicon mmap/OOM behavior." }, { - "left": "issue:43632", - "right": "issue:44756", + "left": "issue:40444", + "right": "issue:45325", "accept": false, - "reason": "_is_hf_initialized regression and Strix Halo mmap OOM are unrelated." + "reason": "Same model family, but one is multi-image iterable training and the other is still-image temporal position scaling." }, { "left": "issue:45812", "right": "issue:46491", "accept": false, - "reason": "Both are tokenizer regressions, but Granite wrong IDs and CodeLlama leading-space round-trip are different bugs and models." + "reason": "Both involve tokenization, but the concrete bugs differ: Granite token IDs vs CodeLlama leading-space round-trip." }, { "left": "issue:43792", "right": "issue:44589", "accept": false, - "reason": "Whisper model runtime failure and Float8 storage lookup error do not share the same code path." + "reason": "Whisper runtime failure is unrelated to missing Float8 storage support." }, { "left": "issue:44403", "right": "issue:44991", "accept": false, - "reason": "Generic loading noise is not the same as a specific tokenizer load failure." + "reason": "Generic loading noise is not the same bug as tokenizer loading for est-roberta." }, { "left": "issue:42796", "right": "issue:43746", "accept": false, - "reason": "Different model families and different failure modes: BPE marker handling vs PEFT checkpoint loading." + "reason": "T5 tokenizer special-token conflict and GraniteSpeech PEFT checkpoint loading are different issues." }, { "left": "issue:33666", "right": "issue:40444", "accept": false, - "reason": "Both mention training problems for Qwen VL, but multi-GPU training and multi-image iterable dataset failures are distinct." + "reason": "Both are vision-model training issues, but the reported failures and code paths differ." }, { "left": "issue:43381", "right": "issue:45923", "accept": false, - "reason": "Gradient checkpointing in eval mode and missing supports_gradient_checkpointing flag are different issues." + "reason": "Eval-mode gradient-checkpointing behavior is distinct from a missing supports_gradient_checkpointing flag." }, { "left": "issue:44291", "right": "issue:44589", "accept": false, - "reason": "init_empty_weights/_is_hf_initialized argument handling is unrelated to missing Float8 storage." + "reason": "Unexpected _is_hf_initialized argument and missing Float8 storage are unrelated regressions." }, { "left": "issue:44079", "right": "issue:45663", "accept": false, - "reason": "ModelOutput key assignment bug and Gemma4 shared_kv_states/FSDP2 bug are different failure modes." + "reason": "ModelOutput key assignment and Gemma4 FSDP shared_kv_states are different internals." }, { "left": "issue:43632", "right": "issue:44534", "accept": false, - "reason": "Different regressions: initialization flag handling vs non-persistent buffer corruption." + "reason": "These are different v5 regressions: initialization flag handling vs buffer contents." }, { "left": "issue:40990", "right": "issue:44811", "accept": false, - "reason": "High perplexity on a language model is unrelated to Whisper batch_decode skip_special_tokens handling." + "reason": "Perplexity on WikiText and Whisper batch_decode special-token handling are unrelated." }, { "left": "issue:43482", "right": "issue:43525", "accept": false, - "reason": "Qwen2.5-GGUF loading failure and missing pad_token_id on Llama4Config are separate model-loading bugs." + "reason": "Different models and different missing pieces: Qwen2.5-GGUF loading vs Llama4Config pad_token_id." }, { "left": "issue:43482", "right": "issue:45538", "accept": false, - "reason": "GGUF loading failure and CLIPTokenizer max-length behavior are unrelated." + "reason": "GGUF loading failure is unrelated to CLIPTokenizer max_length behavior." }, { "left": "issue:35707", "right": "issue:40444", "accept": false, - "reason": "Progressive generation with inputs_embeds/past_key_values is a different code path from Qwen2.5-VL multi-image iterable dataset finetuning." + "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to multi-image finetuning failures." }, { "left": "issue:43421", "right": "issue:44568", "accept": false, - "reason": "Runtime post-processor update issue is not the same as add_special_tokens failing to add BOS/EOS." + "reason": "Runtime post-processor updates and missing BOS/EOS insertion are related only broadly, not the same bug." }, { "left": "issue:43317", "right": "issue:43856", "accept": false, - "reason": "Device-map/offload loading failure and Qwen3 MoE memory inefficiency are different problems." + "reason": "Device-map offload loading and Qwen3 MoE memory efficiency are different problems." }, { "left": "issue:42175", "right": "issue:45542", "accept": false, - "reason": "Packaging omission of TensorFlow in the torch extra is not the same as the runtime backend error when only tensorboard is installed." + "reason": "Both touch TensorFlow packaging, but one is missing backend inclusion while the other is a tensorboard-without-tf runtime error." }, { "left": "issue:41553", "right": "issue:43582", "accept": false, - "reason": "Bad AutoTokenizer error message for Voxtral is unrelated to the AppleSilicon caching_allocator_warmup TypeError." + "reason": "AutoTokenizer error messaging and Apple Silicon caching allocator TypeError are unrelated." }, { "left": "issue:43531", "right": "issue:44291", "accept": false, - "reason": "sliding_window behavior and init_empty_weights argument compatibility are unrelated." + "reason": "Qwen3 sliding_window behavior is not the same as init_empty_weights argument handling." + }, + { + "left": "issue:44079", + "right": "issue:44534", + "accept": false, + "reason": "ModelOutput key assignment and junk-filled buffers are different regressions." }, { "left": "issue:43818", "right": "issue:46402", "accept": false, - "reason": "Video-LLaVA tower weight/temporal-attention issue is unrelated to Sam3 text encoder weight loading." + "reason": "Different multimodal model issues: missing temporal attention vs text encoder weights not loading." }, { - "left": "issue:44079", - "right": "issue:44534", + "left": "issue:44829", + "right": "issue:45479", + "accept": false, + "reason": "Both are classification-training regressions, but the triggers and root causes are different." + }, + { + "left": "issue:41720", + "right": "issue:45081", + "accept": false, + "reason": "Qwen3 device-map CUDA assert is unrelated to Mistral tokenizer regex patching." + }, + { + "left": "issue:44462", + "right": "issue:45812", "accept": false, - "reason": "ModelOutput key assignment and non-persistent buffer corruption are separate bugs." + "reason": "Both are tokenizer regressions, but one is repo tokenizer.json selection and the other is wrong token IDs on Granite models." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet, and the best/canonical issue picks are reasonable. The soft-edge false verdicts are conservative, and the two accepted pairs are at least plausible from the titles alone. If anything, the wording about 'real duplicate pairs' could be softened to 'likely related duplicate candidates' for extra caution." - }, - "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 }, @@ -3784,6 +3794,7 @@ "nodes": [ "issue:34634", "issue:35707", + "issue:39692", "issue:40444", "issue:40990", "issue:41720", @@ -3795,6 +3806,7 @@ "issue:43377", "issue:43454", "issue:43482", + "issue:43550", "issue:43606", "issue:43611", "issue:43632", @@ -3805,36 +3817,31 @@ "issue:43749", "issue:43792", "issue:43818", + "issue:43819", "issue:43873", + "issue:43881", "issue:43957", "issue:43994", "issue:44079", "issue:44265", "issue:44291", - "issue:44462", "issue:44466", "issue:44479", - "issue:44829", + "issue:44488", "issue:44877", "issue:44898", "issue:45072", - "issue:45081", "issue:45325", "issue:45357", "issue:45381", - "issue:45479", "issue:45491", "issue:45584", "issue:45701", "issue:45800", - "issue:45812", "issue:45823", "issue:45834" ], "soft_pairs": [ - "issue:44829|issue:45479", - "issue:44462|issue:45812", - "issue:41720|issue:45081", "issue:43377|issue:45491", "issue:43818|issue:45357", "issue:43644|issue:44079", @@ -3842,8 +3849,8 @@ "issue:44265|issue:45072", "issue:43749|issue:44079", "issue:42175|issue:43296", - "issue:40444|issue:45381", "issue:45325|issue:45381", + "issue:40444|issue:45381", "issue:43792|issue:43957", "issue:43994|issue:44898", "issue:43454|issue:44466", @@ -3854,216 +3861,226 @@ "issue:43701|issue:45325", "issue:44291|issue:44877", "issue:43873|issue:45701", - "issue:43296|issue:45800", "issue:44479|issue:45823", + "issue:43296|issue:45800", "issue:43606|issue:44265", "issue:43262|issue:43746", "issue:40990|issue:45584", "issue:34634|issue:35707", - "issue:35707|issue:41720" + "issue:35707|issue:41720", + "issue:43454|issue:43550", + "issue:39692|issue:40444", + "issue:43606|issue:43819", + "issue:43881|issue:44488" ], - "prepared_review_unit_hash": "51759287311950a78518f0f6b8f6798d1df2989c089095808badcd5263968bd3", + "prepared_review_unit_hash": "b12ef12c749a5ba390cc91710caea5dd6f1ca7684a77911e6a619753112482f3", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36681, - "estimated_input_tokens": 9171, - "estimated_eval_tokens": 18598 + "soft_pair_count": 30, + "serialized_chars": 36672, + "estimated_input_tokens": 9168, + "estimated_eval_tokens": 18592 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a heterogeneous mix of Transformers v5 regression reports, tokenizer/model-loading bugs, and a few unrelated model-specific failures. The soft-similarity pairs mostly share broad subsystem terms but not the same concrete root cause, so I rejected them all.", - "confidence": 0.34, - "canonical_issue_reason": "issue:43611 is the broadest, clearest representative of the dominant theme here: Transformers v5 loading/init regressions. It is more general than the more specialized model- or tokenizer-specific reports.", + "summary": "This set is mostly a grab bag of unrelated model-specific bugs. A few Qwen2.5-VL video/position-id reports are superficially close, but none are strong enough to merge as the same concrete bug.", + "confidence": 0.86, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:43611 is the best single issue to anchor this cluster because it describes a core v5 compatibility break in model loading, which is the most reusable/representative problem among the set.", + "best_issue_reason": "issue:44479 is the broadest umbrella report among the video-regression items, but the overall set is not a single duplicate cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44829", - "right": "issue:45479", + "left": "issue:43377", + "right": "issue:45491", "accept": false, - "reason": "Both concern degenerate training, but one is a flash-attention-3 behavior and the other is a classification-loss edge case for num_labels=1; different bugs." + "reason": "Both involve batching/padding-related behavior, but they are different models and different failure modes." }, { - "left": "issue:44462", - "right": "issue:45812", + "left": "issue:43818", + "right": "issue:45357", "accept": false, - "reason": "Both are tokenizer regressions, but one is AutoTokenizer ignoring tokenizer.json while the other is Granite token ID corruption; different failure modes." + "reason": "One is a Video-LLaVA architecture/weight-sharing bug; the other is a Qwen3.5 save_pretrained key-serialization issue." }, { - "left": "issue:41720", - "right": "issue:45081", + "left": "issue:43644", + "right": "issue:44079", "accept": false, - "reason": "Different models and code paths: one is device-mapping/cuda assert on Qwen3, the other is a Mistral tokenizer regex patch crash." - }, - { - "left": "issue:43377", - "right": "issue:45491", - "accept": false, - "reason": "Both are batching-related, but one is missing padding-mask support in MIMI and the other is Gemma3 NaNs from sliding-window attention; not the same bug." - }, - { - "left": "issue:43818", - "right": "issue:45357", - "accept": false, - "reason": "Video-LLaVA temporal-attention/weight-sharing corruption is unrelated to Qwen3.5 save_pretrained writing wrong visual encoder keys." - }, - { - "left": "issue:43644", - "right": "issue:44079", - "accept": false, - "reason": "Non-persistent buffer junk fill and ModelOutput key assignment are separate internal behaviors with different symptoms and fixes." + "reason": "Non-persistent buffer junk vs ModelOutput key assignment are unrelated code paths." }, { "left": "issue:43611", "right": "issue:43646", "accept": false, - "reason": "Both are v5 loading regressions, but one is about base_model_prefix handling and the other about custom model initialization; too different to merge as one bug." + "reason": "Both are Transformers v5 regressions, but one is base_model_prefix loading and the other is custom model initialization; not the same bug." }, { "left": "issue:44265", "right": "issue:45072", "accept": false, - "reason": "torch.export failures with torch_compilable_check are unrelated to bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel." + "reason": "torch.export with torch_compilable_check and bf16 dtype mismatches are different problems." }, { "left": "issue:43749", "right": "issue:44079", "accept": false, - "reason": "FSDP CPU RAM efficient loading is a model-loading path issue; ModelOutput key assignment is a separate data-structure bug." + "reason": "FSDP CPU RAM efficient loading and ModelOutput key handling are unrelated." }, { "left": "issue:42175", "right": "issue:43296", "accept": false, - "reason": "Packaging/backend installation missing TensorFlow is unrelated to a vLLM model-loading failure for PaddleOCR-VL." + "reason": "Packaging/backends issue vs a specific model load failure; no shared underlying bug." }, { - "left": "issue:40444", + "left": "issue:45325", "right": "issue:45381", "accept": false, - "reason": "Both involve Qwen2.5-VL inputs, but one is iterable finetuning with multiple images and the other is a video position-id regression." + "reason": "Both are Qwen2.5-VL position-id regressions, but one is still-image temporal scaling and the other is video vision_position_ids; not clearly the same defect." }, { - "left": "issue:45325", + "left": "issue:40444", "right": "issue:45381", "accept": false, - "reason": "Both touch Qwen2.5-VL position ids, but one is still-image temporal scaling and the other is video input position ids; related area, different concrete bug." + "reason": "Multiple-images finetuning failure and video position-id corruption are different multimodal paths." }, { "left": "issue:43792", "right": "issue:43957", "accept": false, - "reason": "Whisper runtime failure and meta-device loading breakage are different issues with different causes." + "reason": "A specific model run failure and a meta-device loading regression are not the same bug." }, { "left": "issue:43994", "right": "issue:44898", "accept": false, - "reason": "Different model families and symptoms: SigLIP2 produces nonsense outputs, while Perceiver fails on non-default image resolution." + "reason": "Different models and different symptoms; only a broad image-classification similarity." }, { "left": "issue:43454", "right": "issue:44466", "accept": false, - "reason": "One is failing to tie lm_head weights in AyaVision; the other is inconsistent serialization of tied weights depending on device." + "reason": "Both mention lm_head/tied weights, but one is a missing tie causing garbage generation and the other is serialization inconsistency." }, { "left": "issue:45325", "right": "issue:45834", "accept": false, - "reason": "Qwen2.5-VL RoPE/position-id scaling is unrelated to Kosmos2.5 OCR index errors on long input." + "reason": "Both are model-specific vision/OCR issues, but the affected models and failure modes differ." }, { "left": "issue:42915", "right": "issue:43482", "accept": false, - "reason": "FineGrainedFP8Config failure and GGUF loading failure are separate model-format/runtime compatibility issues." + "reason": "Different models and unrelated failures: FP8 config vs GGUF loading under v5." }, { "left": "issue:43632", "right": "issue:45072", "accept": false, - "reason": "_is_hf_initialized flag breakage is a model-initialization regression, not a bfloat16 dtype mismatch problem." + "reason": "_is_hf_initialized handling and bf16 dtype mismatches are separate regressions." }, { "left": "issue:42796", "right": "issue:45381", "accept": false, - "reason": "Tokenizer BPE-marker handling for Ministral is unrelated to Qwen2.5-VL video position ids." + "reason": "Tokenizer BPE-marker handling is unrelated to Qwen2.5-VL video position ids." }, { "left": "issue:43701", "right": "issue:45325", "accept": false, - "reason": "resume_from_checkpoint key mismatch and Qwen2.5-VL position-id scaling are different code paths and failure modes." + "reason": "Checkpoint key mismatch and Qwen2.5-VL rope-index scaling are unrelated bugs." }, { "left": "issue:44291", "right": "issue:44877", "accept": false, - "reason": "Unexpected _is_hf_initialized argument from init_empty_weights is not the same as strict config rejection for granite_speech." + "reason": "init_empty_weights argument incompatibility and strict config loading are different failures." }, { "left": "issue:43873", "right": "issue:45701", "accept": false, - "reason": "Quantization/offloading behavior is unrelated to tokenizer-version-dependent tokenization changes." + "reason": "Quantization/offloading behavior is unrelated to tokenization changes." }, { - "left": "issue:43296", - "right": "issue:45800", + "left": "issue:44479", + "right": "issue:45823", "accept": false, - "reason": "PaddleOCR-VL loading in vLLM and torch/transformers incompatibility are different cross-version issues." + "reason": "A broad Qwen video regression and a Gemma4 device-mismatch bug are not the same code-path problem." }, { - "left": "issue:44479", - "right": "issue:45823", + "left": "issue:43296", + "right": "issue:45800", "accept": false, - "reason": "Video input regression across Qwen variants is unrelated to Gemma4 PLE device-mismatch during forward." + "reason": "PaddleOCR-VL loading failure and torch/transformers version incompatibility are unrelated." }, { "left": "issue:43606", "right": "issue:44265", "accept": false, - "reason": "CPU offload device mismatch for bark-small is not the same as torch.export failing on torch_compilable_check." + "reason": "CPU offload device mismatch and torch.export compilation failures are different bugs." }, { "left": "issue:43262", "right": "issue:43746", "accept": false, - "reason": "Audio chat-template sample-rate defaulting and GraniteSpeech PEFT checkpoint loading are unrelated." + "reason": "Audio chat-template sampling-rate defaulting and GraniteSpeech checkpoint loading are unrelated." }, { "left": "issue:40990", "right": "issue:45584", "accept": false, - "reason": "High perplexity on gpt-oss-20b and Whisper empty-transcription failure are distinct model/task bugs." + "reason": "High perplexity on one model and Whisper empty-transcription generation failure are unrelated." }, { "left": "issue:34634", "right": "issue:35707", "accept": false, - "reason": "BarkProcessor voice_preset and inputs_embeds/past_key_values generation are unrelated." + "reason": "Bark voice_preset and inputs_embeds/past_key_values generation are different issues." }, { "left": "issue:35707", "right": "issue:41720", "accept": false, - "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to Qwen3 auto device mapping cuda asserts." + "reason": "Progressive generation and Qwen3 auto device-map asserts are unrelated." + }, + { + "left": "issue:43454", + "right": "issue:43550", + "accept": false, + "reason": "lm_head tying/serialization vs torch.compile SDPA failure are not the same defect." + }, + { + "left": "issue:39692", + "right": "issue:40444", + "accept": false, + "reason": "SigLIP2 doc-example errors and Qwen2.5-VL multi-image finetuning failures are different." + }, + { + "left": "issue:43606", + "right": "issue:43819", + "accept": false, + "reason": "Bark CPU-offload device mismatch and DAC from_latents/forward mismatch are unrelated." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "Two separate model-loading failures with no shared underlying code path." } ] }, @@ -4090,26 +4107,23 @@ "issue:26413", "issue:28530", "issue:36331", - "issue:39692", - "issue:40444", "issue:40990", "issue:41720", "issue:42617", "issue:42796", "issue:42915", "issue:42947", + "issue:43065", "issue:43257", "issue:43317", "issue:43322", "issue:43377", + "issue:43450", "issue:43454", "issue:43526", - "issue:43550", "issue:43572", - "issue:43606", "issue:43720", "issue:43749", - "issue:43819", "issue:43873", "issue:43881", "issue:43883", @@ -4117,15 +4131,17 @@ "issue:44265", "issue:44387", "issue:44466", - "issue:44488", "issue:44534", "issue:44568", "issue:44610", + "issue:44756", + "issue:44877", "issue:44898", "issue:44977", "issue:44991", "issue:45072", "issue:45357", + "issue:45701", "issue:45704", "issue:45759", "issue:45823", @@ -4134,21 +4150,18 @@ "issue:46050", "issue:46209", "issue:46305", - "issue:46315" + "issue:46315", + "issue:46402" ], "soft_pairs": [ - "issue:43454|issue:43550", - "issue:39692|issue:40444", - "issue:43606|issue:43819", - "issue:43881|issue:44488", "issue:44186|issue:44265", "issue:41720|issue:45823", "issue:46305|issue:46315", "issue:43377|issue:44610", "issue:42796|issue:46209", "issue:43454|issue:45072", - "issue:44568|issue:45920", "issue:43257|issue:45923", + "issue:44568|issue:45920", "issue:43377|issue:44898", "issue:43317|issue:43873", "issue:42796|issue:44977", @@ -4161,221 +4174,246 @@ "issue:41720|issue:43322", "issue:26413|issue:28530", "issue:43526|issue:43883", - "issue:44466|issue:45357", "issue:44387|issue:45704", "issue:41720|issue:44977", + "issue:44466|issue:45357", "issue:36331|issue:42617", "issue:43881|issue:44991", - "issue:36331|issue:42915" + "issue:36331|issue:42915", + "issue:44877|issue:44991", + "issue:43450|issue:45701", + "issue:40990|issue:45701", + "issue:44186|issue:45072", + "issue:43065|issue:46402", + "issue:43873|issue:44756", + "issue:41720|issue:42617" ], - "prepared_review_unit_hash": "d4c2fee4315307e99164417b07e14941b2b17675a0d9e116d5572b3ff5efa34e", + "prepared_review_unit_hash": "32c27b3385013b417195f4b9a921e3a62342429979834001700c57630332c8ab", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36880, - "estimated_input_tokens": 9220, - "estimated_eval_tokens": 18696 + "soft_pair_count": 33, + "serialized_chars": 37626, + "estimated_input_tokens": 9407, + "estimated_eval_tokens": 19070 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly unrelated Transformers regressions grouped by loose semantic similarity; none of the soft-edge pairs look like the same underlying bug, so the cluster is not a true duplicate set.", - "confidence": 0.74, - "canonical_issue_reason": "Issue 43873 is the broadest and most central-looking report in the offloading/quantization subtheme, which is the closest thing to a cluster anchor here, though the overall set is too heterogeneous for a clean canonical duplicate.", + "summary": "These issues are a heterogeneous mix of unrelated regressions across tokenizers, loading, quantization, device mapping, and training utilities. None of the soft pairs look like the same underlying bug/change, so there is no safe duplicate cluster anchor here.", + "confidence": 0.95, + "canonical_issue_reason": "No single canonical issue: the items span different subsystems and failure modes, so there is no safe duplicate representative for the whole set.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43873 is the best single representative because it is open, broadly phrased, and closest to several offloading/device-map/quantization complaints, but it still does not cover the full heterogeneous cluster.", + "best_issue_reason": "No strong global best issue; if forced, 45920 is the broadest tokenizer regression, but it does not represent the rest of the set.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43454", - "right": "issue:43550", - "accept": false, - "reason": "Different concrete bugs: lm_head weight tying/garbage generation vs torch.compile SDPA failure." - }, - { - "left": "issue:39692", - "right": "issue:40444", - "accept": false, - "reason": "One is a documentation example/model-processor mismatch; the other is an IterableDataset multi-image finetuning failure." - }, - { - "left": "issue:43606", - "right": "issue:43819", - "accept": false, - "reason": "CPU offload device mismatch and DAC latent/forward mismatch are unrelated code paths." - }, - { - "left": "issue:43881", - "right": "issue:44488", - "accept": false, - "reason": "Model loading failure for glm-4v-9b is not the same as tokenizer loading regression for cjvt/sleng-bert." - }, { "left": "issue:44186", "right": "issue:44265", "accept": false, - "reason": "Tokenizer crashes on NER/padding are different from torch.export failures around torch_compilable_check." + "reason": "Tokenizer NER/padding crash vs torch.export + torch_compilable_check export failure are unrelated code paths." }, { "left": "issue:41720", "right": "issue:45823", "accept": false, - "reason": "Both are device-related, but one is a Qwen3 CUDA assert under auto mapping and the other is a Gemma4 PLE forward mismatch." + "reason": "Both involve device placement, but one is a CUDA assert on Qwen3 auto mapping and the other is a Gemma4 PLE forward device mismatch; different bugs." }, { "left": "issue:46305", "right": "issue:46315", "accept": false, - "reason": "Off-by-one in a streaming mask is unrelated to tokenizer conversion performance regression." + "reason": "Sliding-window mask off-by-one and slow convert_tokens_to_ids are different classes of bug." }, { "left": "issue:43377", "right": "issue:44610", "accept": false, - "reason": "MIMI padding-mask batching bug and OmDet processor/model resolution mismatch affect different subsystems." + "reason": "Missing padding-mask support in MIMI encoder vs OmDet-Turbo processor shape mismatch are unrelated." }, { "left": "issue:42796", "right": "issue:46209", "accept": false, - "reason": "Tokenizer BPE-marker handling is unrelated to FP8 inference failure on B200." + "reason": "Tokenizer BPE-marker handling vs FP8 inference failure on B200 are not the same bug." }, { "left": "issue:43454", "right": "issue:45072", "accept": false, - "reason": "lm_head tying/serialization issue and bfloat16 dtype mismatch are different failure modes." + "reason": "Weight-tying/garbage text issue vs bf16 dtype mismatch are different failure modes." }, { - "left": "issue:44568", - "right": "issue:45920", + "left": "issue:43257", + "right": "issue:45923", "accept": false, - "reason": "Missing BOS/EOS insertion for one tokenizer is not the same as wrong token IDs across many models." + "reason": "MoE weight conversion under accelerate+deepspeed is unrelated to a missing gradient_checkpointing flag." }, { - "left": "issue:43257", - "right": "issue:45923", + "left": "issue:44568", + "right": "issue:45920", "accept": false, - "reason": "Qwen3 MoE loading with accelerate+deepspeed and a missing gradient-checkpointing flag are unrelated." + "reason": "Both are tokenizer regressions, but one is missing BOS/EOS for a specific tokenizer while the other is wrong token IDs across many models; not the same bug." }, { "left": "issue:43377", "right": "issue:44898", "accept": false, - "reason": "Missing padding-mask support in MIMI is not the same bug as Perceiver non-default-resolution handling." + "reason": "MIMI batch/padding behavior and Perceiver image-size handling are unrelated model-specific issues." }, { "left": "issue:43317", "right": "issue:43873", "accept": false, - "reason": "Device-map loading of a dequantized model and general offloading-with-quantization regression are related only broadly, not the same concrete path." + "reason": "Dequantized model loading with gpu+cpu offload and generic quantization offloading problems are too broad and not the same concrete defect." }, { "left": "issue:42796", "right": "issue:44977", "accept": false, - "reason": "Tokenizer marker handling is unrelated to flash-attention generation correctness." + "reason": "Tokenizer BPE marker handling vs flash-attention generation failure are unrelated." }, { "left": "issue:43454", "right": "issue:43720", "accept": false, - "reason": "Weight tying/garbage text and packed-weight unpacking during accelerate loading are distinct bugs." + "reason": "AyaVision lm_head tying and BitNet packed-weight unpacking are different model-loading bugs." }, { "left": "issue:40990", "right": "issue:42947", "accept": false, - "reason": "High perplexity on gpt-oss and ineffective gradient checkpointing with LoRA are unrelated." + "reason": "High perplexity on a model benchmark vs ineffective gradient checkpointing with PEFT LoRA are unrelated." }, { "left": "issue:43572", "right": "issue:45759", "accept": false, - "reason": "Missing pad_token_idx in one config is unrelated to unwrapping text_config for composite multimodal configs." + "reason": "Missing pad_token_idx in StableLmConfig and unwrapping text_config for composite Qwen configs are different config-loading problems." }, { "left": "issue:43322", "right": "issue:46050", "accept": false, - "reason": "A Llava Next load-time segfault and sinq save/load quantization breakage are not the same issue." + "reason": "Llava Next segfault on load vs sinq quantized save/load breakage are not the same underlying bug." }, { "left": "issue:42617", "right": "issue:42915", "accept": false, - "reason": "Running 3d_parallel.py and Qwen3 MoE FineGrainedFP8Config failure do not share a concrete bug." + "reason": "Failure to run 3d_parallel.py and Qwen3Moe FP8 config failure are unrelated." }, { "left": "issue:43749", "right": "issue:44534", "accept": false, - "reason": "FSDP CPU RAM efficient loading and non-persistent buffer junk are different implementation bugs." + "reason": "FSDP CPU RAM efficient loading and non-persistent buffer corruption are different issues, even if both touch loading/memory." }, { "left": "issue:41720", "right": "issue:43322", "accept": false, - "reason": "Qwen3 auto device-mapping CUDA asserts and Llava Next loading segfaults are separate failures." + "reason": "Qwen3 auto device mapping CUDA assert and Llava Next load segfault are unrelated model/device failures." }, { "left": "issue:26413", "right": "issue:28530", "accept": false, - "reason": "Resume-from-checkpoint epoch_iterator failure and early-stopping metric selection are unrelated training-loop issues." + "reason": "Checkpoint resume failing on empty epoch_iterator and early stopping metric lookup are different training-loop bugs." }, { "left": "issue:43526", "right": "issue:43883", "accept": false, - "reason": "BeitImageProcessorFast label reduction and missing all_tied_weights_keys in Molmo are unrelated." - }, - { - "left": "issue:44466", - "right": "issue:45357", - "accept": false, - "reason": "lm_head serialization inconsistency and incorrect visual-encoder key saving are different save/load bugs." + "reason": "BeitImageProcessorFast label reduction bug vs missing all_tied_weights_keys on Molmo are unrelated." }, { "left": "issue:44387", "right": "issue:45704", "accept": false, - "reason": "int4 reserved-memory OOM and apex FusedRMSNorm memory leak are different resource issues." + "reason": "Int4 quantization OOM from reserved CUDA memory and an apex RMSNorm memory leak are different memory problems." }, { "left": "issue:41720", "right": "issue:44977", "accept": false, - "reason": "CUDA assert under auto device mapping is not the same as flash-attention generation breakage." + "reason": "CUDA assert under auto device mapping and flash-attention generation issues are unrelated." + }, + { + "left": "issue:44466", + "right": "issue:45357", + "accept": false, + "reason": "lm_head serialization inconsistency and incorrect visual encoder keys are different save_pretrained bugs." }, { "left": "issue:36331", "right": "issue:42617", "accept": false, - "reason": "CustomTrainer.compute_loss API mismatch and inability to run 3d_parallel.py are unrelated." + "reason": "Unexpected compute_loss kwarg and 3d_parallel.py failure are unrelated." }, { "left": "issue:43881", "right": "issue:44991", "accept": false, - "reason": "A model loading failure for glm-4v-9b is not the same as tokenizer loading regression for EMBEDDIA/est-roberta." + "reason": "Model loading failure for glm-4v-9b vs tokenizer loading regression for est-roberta are unrelated." }, { "left": "issue:36331", "right": "issue:42915", "accept": false, - "reason": "Unexpected compute_loss keyword argument and Qwen3 MoE FP8 failure are different code paths." + "reason": "Unexpected compute_loss kwarg and Qwen3Moe FP8 failure are unrelated." + }, + { + "left": "issue:44877", + "right": "issue:44991", + "accept": false, + "reason": "Strict config loading failure and tokenizer loading regression are different bugs." + }, + { + "left": "issue:43450", + "right": "issue:45701", + "accept": false, + "reason": "Batched video processor shape bug and version-dependent tokenization changes are unrelated." + }, + { + "left": "issue:40990", + "right": "issue:45701", + "accept": false, + "reason": "Perplexity regression and tokenization changes are different symptoms and code paths." + }, + { + "left": "issue:44186", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer crash with NER/padding and bf16 dtype mismatch are unrelated." + }, + { + "left": "issue:43065", + "right": "issue:46402", + "accept": false, + "reason": "Sam3 pixel decoder dummy Conv2d and Sam3 text-encoder weights not loaded are different components and different bugs." + }, + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "Quantization/offloading behavior and Strix Halo mmap OOM are not the same underlying issue." + }, + { + "left": "issue:41720", + "right": "issue:42617", + "accept": false, + "reason": "Qwen3 device-map CUDA assert and inability to run 3d_parallel.py are unrelated." } ] }, @@ -4401,28 +4439,31 @@ "nodes": [ "issue:28530", "issue:30064", + "issue:33357", "issue:33666", + "issue:34634", + "issue:35707", "issue:36010", "issue:36331", - "issue:40990", "issue:41720", "issue:42617", "issue:42915", - "issue:43065", "issue:43295", "issue:43316", "issue:43317", "issue:43322", "issue:43404", - "issue:43450", "issue:43526", "issue:43531", + "issue:43540", "issue:43550", "issue:43572", "issue:43577", "issue:43606", + "issue:43653", "issue:43708", "issue:43720", + "issue:43782", "issue:43873", "issue:43881", "issue:44008", @@ -4432,30 +4473,19 @@ "issue:44610", "issue:44671", "issue:44743", - "issue:44756", - "issue:44877", "issue:44898", "issue:44913", - "issue:44991", - "issue:45072", "issue:45161", "issue:45357", "issue:45406", "issue:45632", "issue:45684", "issue:45701", + "issue:45704", "issue:45823", - "issue:46291", - "issue:46402" + "issue:46291" ], "soft_pairs": [ - "issue:44877|issue:44991", - "issue:43450|issue:45701", - "issue:40990|issue:45701", - "issue:44186|issue:45072", - "issue:43065|issue:46402", - "issue:43873|issue:44756", - "issue:41720|issue:42617", "issue:43550|issue:43720", "issue:43577|issue:43708", "issue:43404|issue:45357", @@ -4464,8 +4494,8 @@ "issue:43550|issue:43606", "issue:44186|issue:44610", "issue:44610|issue:44898", - "issue:42915|issue:43317", "issue:44423|issue:45406", + "issue:42915|issue:43317", "issue:45632|issue:45684", "issue:36010|issue:42617", "issue:41720|issue:45161", @@ -4478,242 +4508,240 @@ "issue:43526|issue:43881", "issue:42915|issue:43322", "issue:43316|issue:44913", - "issue:28530|issue:30064" + "issue:28530|issue:30064", + "issue:43540|issue:43782", + "issue:33357|issue:35707", + "issue:43873|issue:45704", + "issue:30064|issue:34634", + "issue:43550|issue:43653", + "issue:43606|issue:43720" ], - "prepared_review_unit_hash": "221b0c78a23b4f71ac0b10c6ee91bae4bcc573f4a0aadd8de6f6d542a1d1130c", + "prepared_review_unit_hash": "e2f3b02be3a69c72caf9632ff1e39acc335a1cd52bc2df600282827d5fea1ec1", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36635, - "estimated_input_tokens": 9159, - "estimated_eval_tokens": 18574 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35831, + "estimated_input_tokens": 8958, + "estimated_eval_tokens": 18172 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is a loose semantic cluster of mostly unrelated Transformers bugs (training, loading, multimodal processors, device/offload, tokenization). The soft pairs are not true duplicates; they only share broad subsystem vocabulary.", - "confidence": 0.93, - "canonical_issue_reason": "issue:43873 is the broadest and most central runtime/loading regression in the set, with the most discussion and references; it is the closest thing to a cluster hub, though the cluster is not a tight duplicate set.", + "summary": "These are mostly unrelated issues that only share broad subsystem keywords (offloading, multimodal processors, tokenizers, training). The overlaps are not tight enough to treat any soft pair as the same underlying bug, so all soft edges should be rejected.", + "confidence": 0.79, + "canonical_issue_reason": "issue:43873 is the best umbrella representative: it is broad, current, and centered on offloading/quantization behavior, which is the closest common theme among several nearby issues. It is still not a duplicate of the others.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 is the best representative issue for the cluster because it captures the most general and widely referenced failure mode (offloading/quantization loading), even though most other items are separate bugs.", + "best_issue_reason": "issue:43873 has the clearest general scope and the strongest chance of serving as a cluster anchor, even though the surrounding issues are distinct bug reports.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44877", - "right": "issue:44991", - "accept": false, - "reason": "Both are loading regressions in v5, but one is strict config loading for granite_speech and the other is tokenizer loading for est-roberta; different failure paths and fixes." - }, - { - "left": "issue:43450", - "right": "issue:45701", - "accept": false, - "reason": "Video processor batched-shape bug vs tokenizer behavior change; they affect different code paths and are not the same defect." - }, - { - "left": "issue:40990", - "right": "issue:45701", - "accept": false, - "reason": "High perplexity on one model is a downstream quality symptom, not the same issue as a version-dependent tokenization change." - }, - { - "left": "issue:44186", - "right": "issue:45072", - "accept": false, - "reason": "Tokenizer crash on NER/padding vs bfloat16 dtype mismatch; unrelated subsystems and root causes." - }, - { - "left": "issue:43065", - "right": "issue:46402", - "accept": false, - "reason": "Same model family, but different bugs: dummy Conv2d in the pixel decoder versus missing text encoder weights." - }, - { - "left": "issue:43873", - "right": "issue:44756", - "accept": false, - "reason": "Both involve memory/offload concerns, but one is quantization/offloading behavior and the other is mmap OOM avoidance; not the same concrete bug." - }, - { - "left": "issue:41720", - "right": "issue:42617", - "accept": false, - "reason": "Qwen3 device-map CUDA assert and inability to run 3d_parallel.py are different runtime failures with no shared fix." - }, { "left": "issue:43550", "right": "issue:43720", "accept": false, - "reason": "torch.compile/SDPA failure is unrelated to packed-weight unpacking during accelerate loading." + "reason": "Different failure modes: torch.compile with SDPA vs packed weights not unpacked during accelerate loading." }, { "left": "issue:43577", "right": "issue:43708", "accept": false, - "reason": "One is dtype propagation on load; the other is checkpoint resume max_steps computation. Different code paths entirely." + "reason": "Unrelated bugs: BLIP-2 dtype handling vs Trainer resume/max_steps calculation." }, { "left": "issue:43404", "right": "issue:45357", "accept": false, - "reason": "Weight tying in generation differs from incorrect visual encoder keys on save_pretrained; both concern model wiring, but not the same bug." + "reason": "Both touch multimodal model internals, but one is lm_head tying and the other is save_pretrained visual key serialization." }, { "left": "issue:44008", "right": "issue:45823", "accept": false, - "reason": "Both are Gemma-family issues, but one is a forward variable-collision AttributeError and the other is a device-map mismatch." + "reason": "Different root causes: Gemma3n variable-name collision vs Gemma4 PLE device mismatch." }, { "left": "issue:43295", "right": "issue:45701", "accept": false, - "reason": "Processor/tokenizer API regression vs a tokenizer change across versions; broad thematic overlap only, not a duplicate." + "reason": "Custom processor/tokenizer API regression vs version-dependent tokenization changes are not the same bug." }, { "left": "issue:43550", "right": "issue:43606", "accept": false, - "reason": "Both mention device mismatches, but one is compile+SDPA and the other is CPU offload; distinct bugs and fixes." + "reason": "Both mention device/offload contexts, but one is torch.compile+SDPA and the other is Bark CPU offload mismatch." }, { "left": "issue:44186", "right": "issue:44610", "accept": false, - "reason": "Tokenizer crash on NER/padding vs processor output image-size mismatch; different components and symptoms." + "reason": "Tokenizer padding/NER crash vs processor image-size mismatch are different code paths." }, { "left": "issue:44610", "right": "issue:44898", "accept": false, - "reason": "Both are vision preprocessing issues, but one is a wrong fixed output size and the other is a non-default-resolution interpolation failure; not the same concrete defect." + "reason": "Both are vision preprocessing issues, but one is fixed-size processor output and the other is Perceiver interpolation failure." }, { - "left": "issue:42915", - "right": "issue:43317", + "left": "issue:44423", + "right": "issue:45406", "accept": false, - "reason": "Both are loading/offload problems, but one is FineGrainedFP8Config failure and the other is dequantized model loading with gpu+cpu offload." + "reason": "Both affect serve/multimodal processor paths, but the concrete crashes and root causes differ." }, { - "left": "issue:44423", - "right": "issue:45406", + "left": "issue:42915", + "right": "issue:43317", "accept": false, - "reason": "Both are serve-time multimodal crashes, but the attribute errors and triggering paths differ." + "reason": "Quantization-related, but Qwen3Moe FineGrainedFP8 and dequantized model offload are different problems." }, { "left": "issue:45632", "right": "issue:45684", "accept": false, - "reason": "Cache-path collision and read-only file-permission propagation are unrelated save/load issues." + "reason": "Cache-path collision and file-permissions propagation are unrelated save/load behaviors." }, { "left": "issue:36010", "right": "issue:42617", "accept": false, - "reason": "ImportError for GenerationMixin versus a 3d_parallel execution failure; no shared underlying bug." + "reason": "ImportError for GenerationMixin is unrelated to not being able to run 3d_parallel.py." }, { "left": "issue:41720", "right": "issue:45161", "accept": false, - "reason": "Both involve Qwen/GPT MoE and parallelism, but one is an auto device-map CUDA assert and the other is tensor-parallel support failure; too different to merge." + "reason": "Both concern model parallelism/device mapping, but the reported failures and model stacks are different." }, { "left": "issue:42915", "right": "issue:46291", "accept": false, - "reason": "Model quantization/loading failure versus pipeline import dependency/version requirement; different layers of the stack." + "reason": "FineGrainedFP8 loading bug vs pipelines import dependency error are unrelated." }, { "left": "issue:43531", "right": "issue:43572", "accept": false, - "reason": "Qwen3-MoE sliding-window behavior and missing pad_token_idx in StableLmConfig are unrelated config/model bugs." + "reason": "Qwen3-MoE sliding window behavior and StableLmConfig pad_token_idx regression are different issues." }, { "left": "issue:28530", "right": "issue:33666", "accept": false, - "reason": "Trainer early-stopping metric selection and Qwen2-VL multi-GPU training are unrelated." + "reason": "Early stopping metric selection vs Qwen2-VL multi-GPU training are not the same bug." }, { "left": "issue:44671", "right": "issue:44743", "accept": false, - "reason": "CamemBERT masked-LM regression and qwen3_5 recurrent-state reset are different model-specific bugs." + "reason": "CamemBERT masked-LM prediction regression vs Qwen3.5 recurrent state reset are unrelated model bugs." }, { "left": "issue:36331", "right": "issue:41720", "accept": false, - "reason": "Trainer compute_loss signature mismatch versus Qwen3 device-map CUDA assert; no common code path." + "reason": "Trainer compute_loss signature breakage is unrelated to Qwen3 auto device mapping cuda assert." }, { "left": "issue:44186", "right": "issue:44361", "accept": false, - "reason": "Both are tokenizer-related AttributeErrors, but the affected tokenizers, tasks, and failure modes are different enough to keep separate." + "reason": "Different tokenizer/model families and different AttributeErrors; not one underlying defect." }, { "left": "issue:43526", "right": "issue:43881", "accept": false, - "reason": "BEiT fast label reduction bug and glm-4v-9b loading failure are unrelated." + "reason": "BeitImageProcessorFast label reduction bug vs glm-4v-9b loading failure are unrelated." }, { "left": "issue:42915", "right": "issue:43322", "accept": false, - "reason": "FineGrainedFP8Config failure and segmentation fault while loading Llava Next are different failures with different likely fixes." + "reason": "One is a quantization config failure, the other is a Llava Next segfault during loading." }, { "left": "issue:43316", "right": "issue:44913", "accept": false, - "reason": "API discrepancy in Gemma3TextConfig and GPTNeoX rotary_pct reload regression are not the same issue." + "reason": "Gemma3TextConfig API discrepancy and GPTNeoX rotary_pct reload regression are separate config issues." }, { "left": "issue:28530", "right": "issue:30064", "accept": false, - "reason": "Trainer early stopping logic and image processor void-segmentation handling are unrelated bugs." - } - ] - }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, - "accepted_nontrivial_soft_edge": false, - "error_kind": null, - "error_message": null - }, - { - "label": "issue", - "component_index": 1, - "component_count": 21, - "review_unit_index": 15, - "review_unit_count": 56, - "status": "reviewed", - "reason": null, - "source": "llm", - "cache_hit": false, - "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-469", - "nodes": [ - "issue:29159", - "issue:30064", - "issue:33357", + "reason": "Early stopping/eval metric handling and void segmentation map processing are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:43782", + "accept": false, + "reason": "Video input processing in Qwen3OmniMoe is not the same as Qwen3VL from_pretrained weight_only failure." + }, + { + "left": "issue:33357", + "right": "issue:35707", + "accept": false, + "reason": "MacOS CLIP bus error and progressive generation with inputs_embeds/past_key_values are different bugs." + }, + { + "left": "issue:43873", + "right": "issue:45704", + "accept": false, + "reason": "Quantization/offload behavior is unrelated to T5's apex.FusedRMSNorm memory leak." + }, + { + "left": "issue:30064", + "right": "issue:34634", + "accept": false, + "reason": "Segmentation-map processing bug vs Bark voice_preset issue are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:43653", + "accept": false, + "reason": "torch.compile/SDPA failure is not the same as tokenizer special-token registration." + }, + { + "left": "issue:43606", + "right": "issue:43720", + "accept": false, + "reason": "CPU offload device mismatch and packed-weight loading are 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": 21, + "review_unit_index": 15, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:29159", "issue:34567", "issue:34634", "issue:35707", @@ -4722,16 +4750,15 @@ "issue:41720", "issue:42490", "issue:43329", + "issue:43381", "issue:43388", + "issue:43421", "issue:43531", - "issue:43540", - "issue:43550", "issue:43606", "issue:43653", - "issue:43720", "issue:43746", - "issue:43782", - "issue:43873", + "issue:43844", + "issue:43927", "issue:44062", "issue:44077", "issue:44368", @@ -4739,10 +4766,14 @@ "issue:44451", "issue:44492", "issue:44509", + "issue:44514", + "issue:44545", "issue:44589", "issue:44610", "issue:44625", "issue:44898", + "issue:44918", + "issue:44936", "issue:44987", "issue:44991", "issue:45042", @@ -4754,18 +4785,11 @@ "issue:45479", "issue:45588", "issue:45701", - "issue:45704", "issue:45759", "issue:45859", "issue:46402" ], "soft_pairs": [ - "issue:43540|issue:43782", - "issue:33357|issue:35707", - "issue:43873|issue:45704", - "issue:30064|issue:34634", - "issue:43550|issue:43653", - "issue:43606|issue:43720", "issue:29159|issue:36032", "issue:44387|issue:45701", "issue:43388|issue:45059", @@ -4774,225 +4798,226 @@ "issue:43531|issue:45759", "issue:42490|issue:43329", "issue:43606|issue:43653", - "issue:45357|issue:46402", "issue:45406|issue:45464", + "issue:45357|issue:46402", "issue:34634|issue:41720", "issue:34567|issue:35707", - "issue:44451|issue:44991", "issue:43531|issue:44589", + "issue:44451|issue:44991", "issue:45759|issue:45859", "issue:44077|issue:44625", "issue:44625|issue:45479", - "issue:45588|issue:45759", "issue:44898|issue:45072", + "issue:45588|issue:45759", "issue:44062|issue:44451", "issue:44987|issue:45042", "issue:44368|issue:45357", - "issue:39692|issue:44610" + "issue:39692|issue:44610", + "issue:44514|issue:44545", + "issue:43381|issue:43844", + "issue:44918|issue:44936", + "issue:43421|issue:43927" ], - "prepared_review_unit_hash": "0a75b82f8d92723d352d1961785a8be56f5b995efac52fb68b2224920dfa4e83", + "prepared_review_unit_hash": "a59eac9bd1462f8c7950f821074af1911e82a0b0d8022c512213938dde15a75a", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 29, - "serialized_chars": 36106, - "estimated_input_tokens": 9027, - "estimated_eval_tokens": 18310 + "soft_pair_count": 27, + "serialized_chars": 35582, + "estimated_input_tokens": 8896, + "estimated_eval_tokens": 18048 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose collection of mostly unrelated Transformers issues spanning tokenizers, multimodal processors, quantization/offload, generation, and docs. None of the soft pairs look like the same underlying bug or change, so they should stay separate.", - "confidence": 0.81, - "canonical_issue_reason": "issue:45701 is the broadest and most representative standalone issue here: it describes a version-induced tokenizer regression, has multiple inbound references, and sits near the center of the cluster\u2019s general tokenizer/loading theme.", + "summary": "Only one soft duplicate looks convincing: the two Qwen2_5_VLProcessor batched-input crash reports are the same underlying bug. The rest are title-level similarities across different components or failure modes, so they should stay separate.", + "confidence": 0.94, + "canonical_issue_reason": "issue:44514 and issue:44545 describe the same crash in `Qwen2_5_VLProcessor.apply_chat_template` with batched input and `padding=False`. I\u2019d keep 44514 as canonical because it was filed first and appears to have the fuller report.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45701 is the best single issue to anchor this cluster because it is clearly actionable, broadly scoped, and more representative than the very model-specific reports.", + "best_issue_reason": "issue:44514 is the best representative for the duplicated Qwen2_5_VLProcessor bug: it is specific, directly actionable, and matches the exact failure mode in the paired duplicate.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43540", - "right": "issue:43782", - "accept": false, - "reason": "Different failures on different Qwen multimodal models: one is a video-input processing error, the other is a from_pretrained weight_only loading error." - }, - { - "left": "issue:33357", - "right": "issue:35707", - "accept": false, - "reason": "Unrelated domains: MacOS bus error in CLIP model loading versus progressive generation with inputs_embeds and past_key_values." - }, - { - "left": "issue:43873", - "right": "issue:45704", - "accept": false, - "reason": "Both are runtime regressions, but one is quantization/offload behavior and the other is a T5 RMSNorm memory leak; different code paths and fixes." - }, - { - "left": "issue:30064", - "right": "issue:34634", - "accept": false, - "reason": "Image processor segmentation-map handling and BarkProcessor voice_preset support are unrelated bugs." - }, - { - "left": "issue:43550", - "right": "issue:43653", - "accept": false, - "reason": "Different subsystems and causes: SDPA/torch.compile failure in Bamba versus tokenizer special-token registration in BigBirdTokenizer." - }, - { - "left": "issue:43606", - "right": "issue:43720", - "accept": false, - "reason": "Both involve loading/offload, but one is a CPU offload device mismatch and the other is packed-weight unpacking during accelerate loading; not the same bug." - }, { "left": "issue:29159", "right": "issue:36032", "accept": false, - "reason": "General slow-vs-fast tokenizer inconsistency is not the same as the T5 add_special_tokens name conflict on load." + "reason": "Both are tokenizer-related, but one is slow/fast inconsistency while the other is a T5 tokenizer load-time `add_special_tokens` conflict; different bugs." }, { "left": "issue:44387", "right": "issue:45701", "accept": false, - "reason": "CUDA memory/OOM under int4 quantization is unrelated to version-dependent tokenization changes." + "reason": "Different problems: int4 quantization memory regression vs tokenization changes across versions. Same broad area, not the same bug." }, { "left": "issue:43388", "right": "issue:45059", "accept": false, - "reason": "Metric gathering/tuple truncation and SAM3 text+box behavior are different training/inference issues." + "reason": "One is a distributed metrics tuple-truncation bug; the other is SAM3 PCS behavior with text and boxes. No concrete overlap." }, { "left": "issue:44492", "right": "issue:44509", "accept": false, - "reason": "A typo in cache-strategy docs is not the same as outdated pipeline-task documentation." + "reason": "A typo in cache strategy docs and removed pipeline task docs are both documentation issues, but they are unrelated changes." }, { "left": "issue:41720", "right": "issue:43746", "accept": false, - "reason": "Qwen3 auto device mapping CUDA assert and GraniteSpeech PEFT local checkpoint loading are distinct model-loading failures." + "reason": "Qwen3 auto device mapping CUDA assert and GraniteSpeech PEFT local checkpoint loading are different model-loading failures." }, { "left": "issue:43531", "right": "issue:45759", "accept": false, - "reason": "Sliding-window behavior in Qwen3-MoE and unwrapping text_config in composite multimodal configs are separate Qwen issues with different fixes." + "reason": "Both mention Qwen3.x config behavior, but one is a sliding_window issue and the other is `from_config` not unwrapping `text_config`; different code paths." }, { "left": "issue:42490", "right": "issue:43329", "accept": false, - "reason": "Trailing-slash import warning and undefined video-token-count helpers in multimodal processing are unrelated." + "reason": "Trailing-slash local model import warning and a multimodal token counting video-branch NameError are unrelated." }, { "left": "issue:43606", "right": "issue:43653", "accept": false, - "reason": "Offload/device mismatch in Bark-small and BigBird tokenizer special-token registration are different bugs." + "reason": "Both are Bark-related CI issues, but one is CPU offload device mismatch and the other is a BigBirdTokenizer special-token decode bug." }, { - "left": "issue:45357", - "right": "issue:46402", + "left": "issue:45406", + "right": "issue:45464", "accept": false, - "reason": "Both are Qwen/Sam3 multimodal load/save regressions, but one is incorrect visual encoder keys on save and the other is missing text encoder weights on load." + "reason": "Gemma4Processor missing `_tokenizer` in serve and Qwen3.5 streaming chat/completions failures are different subsystems and failure modes." }, { - "left": "issue:45406", - "right": "issue:45464", + "left": "issue:45357", + "right": "issue:46402", "accept": false, - "reason": "Gemma4Processor missing _tokenizer in serve and Qwen3.5 streaming chat/completions failure are different API paths and failures." + "reason": "Both concern model saving/loading in multimodal models, but one is Qwen3.5 visual encoder keys on save and the other is Sam3Model text encoder weights not loading." }, { "left": "issue:34634", - "right": "issue:35707", + "right": "issue:41720", "accept": false, - "reason": "TrainerState token counting and progressive generation with inputs_embeds/past_key_values are unrelated." + "reason": "BarkProcessor voice_preset failure and Qwen3 auto device mapping CUDA assert are unrelated." }, { - "left": "issue:44451", - "right": "issue:44991", + "left": "issue:34567", + "right": "issue:35707", "accept": false, - "reason": "Loading failures for different models (ScandiBERT vs est-roberta) do not indicate the same root cause." + "reason": "TrainerState token counter not updating and progressive generation with inputs_embeds/past_key_values are separate trainer/runtime issues." }, { "left": "issue:43531", "right": "issue:44589", "accept": false, - "reason": "Qwen3-MoE sliding-window behavior and missing Float8 storage registration are different failures." + "reason": "Sliding window handling for Qwen3-MoE and a Float8 storage lookup TypeError are unrelated." + }, + { + "left": "issue:44451", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer/load failures for different models, but one is a generic latest-version load regression and the other is an EMBEDDIA tokenizer issue; not the same bug." }, { "left": "issue:45759", "right": "issue:45859", "accept": false, - "reason": "Config unwrapping for multimodal Qwen models and CPU-offload incompatibility in rotary embedding are separate code paths." + "reason": "Both involve Qwen 3.5 config/model mechanics, but one is `from_config` text_config unwrapping and the other is CPU-offload rotary embedding compatibility." }, { "left": "issue:44077", "right": "issue:44625", "accept": false, - "reason": "patchtsmixer post_init validation and Qwen3.5 num_labels propagation are unrelated configuration issues." + "reason": "Optional `post_init` in patchtsmixer and Qwen3.5 `num_labels` propagation are different configuration bugs." }, { "left": "issue:44625", "right": "issue:45479", "accept": false, - "reason": "Num_labels propagation in Qwen3.5 configs and degenerate zero-loss behavior in sequence classification are different bugs." + "reason": "One is config propagation for Qwen3.5; the other is zero-loss behavior for single-label classification across models. Different underlying issues." }, { - "left": "issue:45588", - "right": "issue:45759", + "left": "issue:44898", + "right": "issue:45072", "accept": false, - "reason": "Flash-attention sinkless-model crash and Qwen multimodal config unwrapping are unrelated." + "reason": "Perceiver image classification resolution failure and dtype mismatches in bfloat16 inference are separate model-processing bugs." }, { - "left": "issue:44898", - "right": "issue:45072", + "left": "issue:45588", + "right": "issue:45759", "accept": false, - "reason": "Perceiver non-default-resolution preprocessing and bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel are distinct issues." + "reason": "Flash attention `s_aux=None` crash and Qwen multimodal config unwrapping are unrelated code paths." }, { "left": "issue:44062", "right": "issue:44451", "accept": false, - "reason": "AddedToken keyword collision and ScandiBERT loading failure are not the same root cause." + "reason": "AddedToken special-argument duplication and a model load failure for ScandiBERT are different tokenizer/model issues." }, { "left": "issue:44987", "right": "issue:45042", "accept": false, - "reason": "A model-specific loading failure and PIL image processor torchvision dependency regression are unrelated." + "reason": "A loading regression for a specific model and a PIL backend torchvision dependency regression are unrelated." }, { "left": "issue:44368", "right": "issue:45357", "accept": false, - "reason": "A tie_word_embeddings warning during LoRA fine-tuning is not the same as incorrect visual-encoder keys being saved." + "reason": "A tie_word_embeddings warning during LoRA fine-tuning and incorrect visual encoder keys on save are not the same failure." }, { "left": "issue:39692", "right": "issue:44610", "accept": false, - "reason": "Both mention examples/processors, but one is a SigLIP2 docs example with multiple errors and the other is an OmDet-Turbo processor size mismatch." + "reason": "SigLIP2 docs example errors and OmDet-Turbo processor/model resolution mismatch are both vision-related but describe different problems." + }, + { + "left": "issue:44514", + "right": "issue:44545", + "accept": true, + "reason": "Exact same title and same failure mode: `Qwen2_5_VLProcessor.apply_chat_template` crashes on batched input when `padding=False`. These are duplicates." + }, + { + "left": "issue:43381", + "right": "issue:43844", + "accept": false, + "reason": "Gradient checkpointing in eval mode and gradient blow-up with DeepSpeed ZeRO-3 are distinct training issues." + }, + { + "left": "issue:44918", + "right": "issue:44936", + "accept": false, + "reason": "Qwen3.5 input-embedding unpacking with TRL SFT trainer and trainer.evaluate() failing after train() are different trainer bugs." + }, + { + "left": "issue:43421", + "right": "issue:43927", + "accept": false, + "reason": "Runtime post-processor updates for TokenizersBackend and DiaConfig losing custom token IDs after save/load are related to tokens but not the same bug." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The evaluation is grounded in the packet and the soft-edge verdicts are conservative. The only accepted duplicate, issue:44514 vs issue:44545, is directly supported by identical titles and matching failure mode. The remaining pairs are sensibly rejected as different bugs or only loosely related." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -5015,13 +5040,13 @@ "issue:41720", "issue:42886", "issue:43295", - "issue:43381", "issue:43421", "issue:43531", "issue:43540", "issue:43550", "issue:43596", "issue:43618", + "issue:43643", "issue:43645", "issue:43653", "issue:43697", @@ -5029,19 +5054,19 @@ "issue:43746", "issue:43782", "issue:43819", - "issue:43844", "issue:43873", "issue:43927", "issue:43994", + "issue:44008", + "issue:44062", "issue:44186", "issue:44220", - "issue:44514", - "issue:44545", + "issue:44488", "issue:44661", "issue:44671", "issue:44843", - "issue:44918", - "issue:44936", + "issue:44877", + "issue:44898", "issue:45072", "issue:45081", "issue:45356", @@ -5058,17 +5083,13 @@ "issue:46463" ], "soft_pairs": [ - "issue:44514|issue:44545", - "issue:43381|issue:43844", - "issue:44918|issue:44936", - "issue:43421|issue:43927", "issue:45704|issue:45923", "issue:43746|issue:46050", "issue:41720|issue:45704", "issue:45704|issue:45859", "issue:43531|issue:43645", - "issue:42886|issue:44843", "issue:46032|issue:46463", + "issue:42886|issue:44843", "issue:43782|issue:45381", "issue:43421|issue:43653", "issue:45663|issue:45923", @@ -5080,243 +5101,258 @@ "issue:45081|issue:45356", "issue:43720|issue:43819", "issue:43596|issue:44220", - "issue:45356|issue:45907", "issue:43653|issue:43720", + "issue:45356|issue:45907", "issue:43540|issue:46294", "issue:43295|issue:45834", "issue:43618|issue:43697", "issue:44220|issue:44661", "issue:43550|issue:45072", - "issue:43550|issue:43927" + "issue:43550|issue:43927", + "issue:44062|issue:44488", + "issue:44186|issue:44898", + "issue:43531|issue:43643", + "issue:43653|issue:43819", + "issue:43643|issue:44877", + "issue:43540|issue:44008" ], - "prepared_review_unit_hash": "5ef9d8266cb624c27e10ecd31847f3e10715b05a11b2d6b3364b82e7bb218791", + "prepared_review_unit_hash": "2129d5cdb17ed9259c412d778c02f1cacbf0edbe777aabdcfe43c3d82d63a22a", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 30, - "serialized_chars": 36388, - "estimated_input_tokens": 9097, - "estimated_eval_tokens": 18450 + "soft_pair_count": 32, + "serialized_chars": 37088, + "estimated_input_tokens": 9272, + "estimated_eval_tokens": 18800 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly heterogeneous, with two likely duplicate subpairs: one exact Qwen2_5_VL batched-padding crash and one offline tokenizer regression around the Mistral regex patch. The remaining candidates describe different bugs despite superficial subsystem overlap.", - "confidence": 0.84, - "canonical_issue_reason": "No single canonical issue for the whole cluster: most items are unrelated, and only small duplicate subpairs appear.", + "summary": "This cluster is heterogeneous: the items span unrelated bugs in tokenizers, multimodal inputs, quantization/offloading, config loading, and model-specific inference paths. The soft pairs are mostly only superficially similar, so none should be merged as duplicates.", + "confidence": 0.89, + "canonical_issue_reason": "No clear canonical duplicate emerges because the cluster does not center on one underlying bug; the issues are spread across distinct code paths and model families.", "canonical_pr_reason": null, - "best_issue_reason": "No global best issue across this mixed cluster; any representative would only cover a narrow subpair, not the whole set.", + "best_issue_reason": "If one issue must be used as a representative anchor, issue 43873 is the broadest and most discussion-heavy open report, but it still does not subsume the rest of the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44514", - "right": "issue:44545", - "accept": true, - "reason": "Same crash in Qwen2_5_VLProcessor.apply_chat_template on batched input with padding=False; titles differ only in wording." - }, - { - "left": "issue:43381", - "right": "issue:43844", - "accept": false, - "reason": "Different problems: eval-mode gradient checkpointing misuse vs abnormal gradients with ZeRO-3 training." - }, - { - "left": "issue:44918", - "right": "issue:44936", - "accept": false, - "reason": "Unrelated failures: Qwen3.5 embedding unpacking with TRL SFT vs trainer.evaluate() after trainer.train()." - }, - { - "left": "issue:43421", - "right": "issue:43927", - "accept": false, - "reason": "Different tokenizer/config bugs: runtime special-token post-processor update vs save/load losing custom token IDs." - }, { "left": "issue:45704", "right": "issue:45923", "accept": false, - "reason": "Different subsystems and symptoms: T5 RMSNorm memory leak vs missing supports_gradient_checkpointing flag." + "reason": "Different bugs: one is an apex RMSNorm memory leak in T5, the other is a missing supports_gradient_checkpointing flag in a Nemotron trust_remote_code variant." }, { "left": "issue:43746", "right": "issue:46050", "accept": false, - "reason": "Unrelated: GraniteSpeech PEFT checkpoint loading vs sinq quantized model save/load regression." + "reason": "Unrelated failures: PEFT/local checkpoint loading for GraniteSpeech vs broken save/load behavior for sinq-quantized models." }, { "left": "issue:41720", "right": "issue:45704", "accept": false, - "reason": "Qwen3 auto device mapping CUDA assert is unrelated to the T5 apex memory leak." + "reason": "Different code paths and symptoms: Qwen3 auto device mapping cuda assert vs a T5 RMSNorm memory leak." }, { "left": "issue:45704", "right": "issue:45859", "accept": false, - "reason": "Different issues: T5 apex RMSNorm leak vs Qwen3_5Moe CPU offload incompatibility." + "reason": "Both mention runtime issues, but one is an apex leak and the other is CPU-offload incompatibility in Qwen3_5Moe rotary embeddings." }, { "left": "issue:43531", "right": "issue:43645", "accept": false, - "reason": "Different bugs: Qwen3-MoE sliding_window behavior vs Jupyter custom-model initialization regression." - }, - { - "left": "issue:42886", - "right": "issue:44843", - "accept": true, - "reason": "Both describe HF_HUB_OFFLINE cache-loading failures caused by tokenizer code that unconditionally reaches model_info() in the Mistral regex patch path." + "reason": "Separate problems: Qwen3-MoE sliding_window behavior vs notebook custom-model initialization breakage in Transformers 5.0." }, { "left": "issue:46032", "right": "issue:46463", "accept": false, - "reason": "Different model paths and symptoms: Mamba2 use_cache correctness bug vs T5 decoder-input error-message regression." + "reason": "Different bug classes: incorrect caching results in Mamba2Mixer vs an error-message/validation issue in T5 decoder inputs." + }, + { + "left": "issue:42886", + "right": "issue:44843", + "accept": false, + "reason": "Both involve offline mode, but one is a general tokenizer cache-load regression and the other is a specific Mistral regex patch making an unconditional network call." }, { "left": "issue:43782", "right": "issue:45381", "accept": false, - "reason": "Different video/weight-loading bugs in different Qwen variants; not the same concrete failure." + "reason": "Different models and failures: Qwen3VL weight_only loading error vs Qwen2.5-VL video position-id mismatch." }, { "left": "issue:43421", "right": "issue:43653", "accept": false, - "reason": "Different tokenizer issues: runtime post-processor syncing vs BigBird mask token registration/decoding." + "reason": "Related tokenization area, but not the same bug: runtime special-token/post-processor updates vs BigBirdTokenizer mask-token registration/empty decode." }, { "left": "issue:45663", "right": "issue:45923", "accept": false, - "reason": "Gemma4 FSDP2 shared_kv_states bug is unrelated to the Nemotron gradient-checkpointing flag issue." + "reason": "Unrelated: Gemma4 FSDP2 shared_kv_states bug vs a missing gradient-checkpointing capability flag in a Nemotron model." }, { "left": "issue:39692", "right": "issue:43873", "accept": false, - "reason": "Both mention quantization/offloading broadly, but they are different bugs with different models and failure modes." + "reason": "One is a SigLIP2 documentation/model-processor mismatch with quantization issues; the other is a general offloading-with-quantization runtime bug." }, { "left": "issue:43994", "right": "issue:44671", "accept": false, - "reason": "Different model-specific inference/training bugs: SigLIP2 AutoModel/pipeline output issue vs CamemBERT MLM regression." + "reason": "Different concerns: SigLIP2 inference quality/pipeline behavior vs add-new-model-like failing when the model is in TOKENIZER_MAPPING_NAMES." }, { "left": "issue:43746", "right": "issue:43873", "accept": false, - "reason": "Different areas: PEFT checkpoint loading vs quantization/offloading behavior." + "reason": "Both touch loading/quantization-adjacent behavior, but they describe different problems: PEFT checkpoint loading vs offloading with quantization." }, { "left": "issue:43421", "right": "issue:44186", "accept": false, - "reason": "Tokenizer post-processor syncing and LayoutLMv2 NER padding/truncation crashes are different problems." + "reason": "Both are tokenizer bugs, but the concrete failures differ: runtime special-token updates vs LayoutLMv2 NER/batched padding-truncation crashes." }, { "left": "issue:30064", "right": "issue:36010", "accept": false, - "reason": "Completely unrelated: image processor segmentation-map handling vs GenerationMixin import error." + "reason": "Completely unrelated: image processor void segmentation map handling vs an ImportError for GenerationMixin." }, { "left": "issue:45081", "right": "issue:45356", "accept": false, - "reason": "Both involve Mistral/fix_mistral_regex, but one is an AttributeError crash and the other is a codec-handling regression with a warning issue." + "reason": "Both involve Mistral/Kimi tokenizer regex handling, but one is a backend_tokenizer AttributeError and the other is a codec-handling regression with a warning issue." }, { "left": "issue:43720", "right": "issue:43819", "accept": false, - "reason": "Different model internals: BitNet packed-weight loading vs DAC latent reconstruction/STE mismatch." + "reason": "Different model internals and symptoms: BitNet packed-weight loading vs DAC latent reconstruction missing STE." }, { "left": "issue:43596", "right": "issue:44220", "accept": false, - "reason": "Unrelated: DeepSpeed ZeRO-3 initialization IndexError in BertModel vs audio feature extraction issue." + "reason": "No shared underlying bug: DeepSpeed ZeRO3/BertModel init failure vs a feature-extraction helper issue." }, { - "left": "issue:45356", - "right": "issue:45907", + "left": "issue:43653", + "right": "issue:43720", "accept": false, - "reason": "Different bugs: tokenizer regression in Kimi-K2.5 vs list_repo_files revision forwarding in a CI script." + "reason": "Distinct tokenizer-special-token bug vs BitNet accelerate-loading bug; only broad infrastructure similarity." }, { - "left": "issue:43653", - "right": "issue:43720", + "left": "issue:45356", + "right": "issue:45907", "accept": false, - "reason": "BigBird tokenizer special-token registration is unrelated to BitNet accelerate loading." + "reason": "Unrelated: tokenizer codec/regex regression vs list_repo_files failing to forward revision." }, { "left": "issue:43540", "right": "issue:46294", "accept": false, - "reason": "Both are Qwen video-input issues, but they target different models and different failure points (processor ValueError vs chat-template feature/token mismatch)." + "reason": "Both concern video inputs, but they are different failures in different Qwen models and code paths." }, { "left": "issue:43295", "right": "issue:45834", "accept": false, - "reason": "Different multimodal bugs: processor.tokenizer regression vs Kosmos2.5 long-OCR index error." + "reason": "Different regressions: processor/tokenizer API breakage vs an OCR long-input index error in Kosmos2.5." }, { "left": "issue:43618", "right": "issue:43697", "accept": false, - "reason": "CLIPOutput attentions regression is unrelated to RTDetrV2 output differences." + "reason": "Different issues: CLIP attentions field assignment vs RTDetrV2 output divergence in v5." }, { "left": "issue:44220", "right": "issue:44661", "accept": false, - "reason": "Audio feature extraction bug and add-new-model-like tokenizer mapping bug are unrelated." + "reason": "No shared defect: audio fbank extraction helper vs add-new-model-like failure on TOKENIZER_MAPPING_NAMES." }, { "left": "issue:43550", "right": "issue:45072", "accept": false, - "reason": "Different model/runtime issues: Bamba SDPA torch.compile failure vs dtype mismatches in other models." + "reason": "Different runtime problems: torch.compile/SDPA failure in Bamba vs bfloat16 dtype mismatches in SwitchTransformers and TimmWrapperModel." }, { "left": "issue:43550", "right": "issue:43927", "accept": false, - "reason": "No shared underlying bug: torch.compile/SDPA failure vs DiaConfig custom token ID persistence." - } - ] - }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative: it identifies only the two clearly supported duplicate subpairs (44514/44545 and 42886/44843) and correctly treats the rest as heterogeneous. The soft-edge acceptances are justified by near-identical titles or a direct shared offline/Mistral regression path, and the rejected pairs are framed narrowly without overclaiming." - }, - "evaluator_used": true, - "retried": false, - "accepted_nontrivial_soft_edge": true, - "error_kind": null, - "error_message": null - }, - { - "label": "issue", - "component_index": 1, - "component_count": 21, - "review_unit_index": 17, - "review_unit_count": 56, - "status": "reviewed", + "reason": "Unrelated: torch.compile+SDPA failure vs DiaConfig losing custom token IDs after save/load." + }, + { + "left": "issue:44062", + "right": "issue:44488", + "accept": false, + "reason": "Only loosely tokenizer-related; the AddedToken special-argument error is not the same underlying bug as failing to load cjvt/sleng-bert." + }, + { + "left": "issue:44186", + "right": "issue:44898", + "accept": false, + "reason": "Different model classes and failure modes: LayoutLMv2 tokenizer crashes vs Perceiver image classification failing on non-default resolution." + }, + { + "left": "issue:43531", + "right": "issue:43643", + "accept": false, + "reason": "Different areas: Qwen3-MoE sliding_window behavior vs AutoConfig trust_remote_code returning missing fields." + }, + { + "left": "issue:43653", + "right": "issue:43819", + "accept": false, + "reason": "Different bugs: BigBirdTokenizer special-token registration vs DAC.from_latents mismatch with missing STE." + }, + { + "left": "issue:43643", + "right": "issue:44877", + "accept": false, + "reason": "Both involve config loading, but one is a trust_remote_code field-loss bug and the other is strict config blocking granite_speech loading." + }, + { + "left": "issue:43540", + "right": "issue:44008", + "accept": false, + "reason": "Different multimodal model bugs: Qwen3OmniMoe video processing ValueError vs Gemma 3n variable-name collision causing audio_mel_mask AttributeError." + } + ] + }, + "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": 21, + "review_unit_index": 17, + "review_unit_count": 56, + "status": "reviewed", "reason": null, "source": "llm", "cache_hit": false, @@ -5332,38 +5368,38 @@ "issue:43322", "issue:43335", "issue:43388", - "issue:43531", + "issue:43421", "issue:43540", "issue:43550", + "issue:43577", "issue:43582", "issue:43606", - "issue:43643", "issue:43644", - "issue:43653", "issue:43716", "issue:43749", "issue:43756", + "issue:43818", "issue:43819", "issue:43950", "issue:44008", - "issue:44062", "issue:44186", "issue:44442", "issue:44448", "issue:44479", "issue:44485", - "issue:44488", "issue:44509", "issue:44671", + "issue:44805", + "issue:44811", "issue:44821", "issue:44843", - "issue:44877", - "issue:44898", + "issue:44936", "issue:44991", "issue:45072", "issue:45081", "issue:45084", "issue:45183", + "issue:45381", "issue:45479", "issue:45584", "issue:45685", @@ -5373,12 +5409,6 @@ "issue:46512" ], "soft_pairs": [ - "issue:44062|issue:44488", - "issue:44186|issue:44898", - "issue:43531|issue:43643", - "issue:43653|issue:43819", - "issue:43540|issue:44008", - "issue:43643|issue:44877", "issue:43335|issue:43756", "issue:43550|issue:43819", "issue:30064|issue:36331", @@ -5388,209 +5418,201 @@ "issue:43322|issue:45685", "issue:39692|issue:45834", "issue:43388|issue:46032", - "issue:45183|issue:45479", "issue:44186|issue:44442", + "issue:45183|issue:45479", "issue:45584|issue:45834", - "issue:44448|issue:44671", - "issue:43716|issue:44008", "issue:43950|issue:45987", "issue:43644|issue:43749", + "issue:44448|issue:44671", "issue:43582|issue:46512", + "issue:43716|issue:44008", "issue:44485|issue:44509", "issue:43606|issue:45072", "issue:34634|issue:36010", - "issue:44479|issue:45834" + "issue:44479|issue:45834", + "issue:43577|issue:43818", + "issue:44805|issue:44936", + "issue:43540|issue:45381", + "issue:43421|issue:44811" ], - "prepared_review_unit_hash": "2dc8199a4dd55f14b6120ad3c61d41ebca9c9fc307942aa0a6f5556beff1a52e", + "prepared_review_unit_hash": "48c1674c00a884c64e1eea96802acdb4d4f293d90d02e9b835872194bb0b80ff", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 27, - "serialized_chars": 36061, - "estimated_input_tokens": 9016, - "estimated_eval_tokens": 18288 + "soft_pair_count": 25, + "serialized_chars": 35594, + "estimated_input_tokens": 8899, + "estimated_eval_tokens": 18054 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab bag of unrelated Transformers bug reports. A couple of pairs are superficially similar, especially the Mistral regex/loading issues, but they appear to be different failure modes rather than the same underlying bug. No soft edge should be merged.", - "confidence": 0.88, - "canonical_issue_reason": "issue:43950 is the strongest cross-cutting regression here: a `from_pretrained()` v5 buffer-corruption bug with broad impact on model loading and state handling, making it the best anchor for this mixed cluster.", + "summary": "All soft-similarity links look like false positives: the issues span unrelated models, subsystems, and failure modes, so none should be merged as duplicates. The cluster is noisy and heterogeneous, with no true duplicate pair identified.", + "confidence": 0.91, + "canonical_issue_reason": "issue:45834 is the most connected item in this noisy set and is a concrete, reproducible bug report, but the cluster overall is too heterogeneous to indicate a real canonical duplicate target.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43950 is the most generally useful representative because it describes a clear, high-impact core loading regression rather than a model-specific edge case.", + "best_issue_reason": "issue:45834 is the best representative by connectivity and specificity, though it is still only a local hub in an otherwise unrelated cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44062", - "right": "issue:44488", - "accept": false, - "reason": "Tokenizer/AddedToken construction error vs a model loading failure for a specific checkpoint; different code paths and root causes." - }, - { - "left": "issue:44186", - "right": "issue:44898", - "accept": false, - "reason": "Different domains entirely: LayoutLMv2 tokenizer/padding behavior vs Perceiver image classification resolution handling." - }, - { - "left": "issue:43531", - "right": "issue:43643", - "accept": false, - "reason": "Qwen3 sliding-window/moe behavior is unrelated to `trust_remote_code` config field loss." - }, - { - "left": "issue:43653", - "right": "issue:43819", - "accept": false, - "reason": "BigBirdTokenizer special-token registration issue vs DAC latent reconstruction mismatch; no shared bug." - }, - { - "left": "issue:43540", - "right": "issue:44008", - "accept": false, - "reason": "Video input processing in Qwen3OmniMoe vs a Gemma3n variable-name collision in forward; unrelated failures." - }, - { - "left": "issue:43643", - "right": "issue:44877", - "accept": false, - "reason": "Both involve config loading, but one is missing fields with `trust_remote_code=True` and the other is strict config validation blocking `granite_speech`; distinct bugs." - }, { "left": "issue:43335", "right": "issue:43756", "accept": false, - "reason": "Sparse-layer creation in SwitchTransformers vs Smollm3 RoPE layer count mismatch; different model internals." + "reason": "Different model/config bugs: sparse-layer construction vs RoPE-layer dropping." }, { "left": "issue:43550", "right": "issue:43819", "accept": false, - "reason": "torch.compile/SDPA failure in Bamba vs DAC `from_latents` mismatch; unrelated code paths." + "reason": "Unrelated failures in different code paths: torch.compile/SDPA vs DAC latent decoding." }, { "left": "issue:30064", "right": "issue:36331", "accept": false, - "reason": "Image processor void segmentation map handling vs trainer loss-signature incompatibility; no common underlying bug." + "reason": "Completely different subsystems: image segmentation preprocessing vs Trainer loss signature." }, { "left": "issue:41720", "right": "issue:44991", "accept": false, - "reason": "Qwen3 device-mapping CUDA assert vs tokenizer load failure for a different model; unrelated." + "reason": "Different components and symptoms: Qwen3 device-map CUDA assert vs tokenizer loading failure." }, { "left": "issue:44843", "right": "issue:45081", "accept": false, - "reason": "Both touch `_patch_mistral_regex`, but one is an offline `model_info()` call bug and the other is a tokenizer attribute-access crash; different concrete failures." + "reason": "Same helper function, but distinct bugs: offline hub call vs missing backend_tokenizer attribute." }, { "left": "issue:44821", "right": "issue:45084", "accept": false, - "reason": "AutoImageProcessor URL loading vs template compilation error; different subsystems." + "reason": "No shared underlying bug; image-processor URL loading and template compilation are unrelated." }, { "left": "issue:43322", "right": "issue:45685", "accept": false, - "reason": "Llava Next load segfault vs MPS histogram op unsupported for Int in moe code; unrelated." + "reason": "Different model/runtime failures: Llava Next segfault vs MPS histogram on Int." }, { "left": "issue:39692", "right": "issue:45834", "accept": false, - "reason": "SigLIP2 docs/model-processor mismatch vs Kosmos2.5 long OCR input indexing; different bug classes." + "reason": "Docs/example issues for SigLIP2 are unrelated to Kosmos2.5 OCR indexing errors." }, { "left": "issue:43388", "right": "issue:46032", "accept": false, - "reason": "Metric-gathering label truncation vs Mamba2 cache correctness; distinct runtime paths." - }, - { - "left": "issue:45183", - "right": "issue:45479", - "accept": false, - "reason": "Input validation error-message bug vs degenerate sequence-classification loss; not the same issue." + "reason": "Different execution paths: metric gathering in evaluation vs Mamba2 use_cache correctness." }, { "left": "issue:44186", "right": "issue:44442", "accept": false, - "reason": "LayoutLMv2 tokenizer padding/truncation crash vs AutoTokenizer failing on a custom tokenizer class; unrelated." - }, - { - "left": "issue:45584", - "right": "issue:45834", - "accept": false, - "reason": "Whisper empty-transcription generation bug vs Kosmos2.5 OCR input length bug; different models and failures." + "reason": "Tokenizer/padding crash vs AutoTokenizer loading failure; no same bug." }, { - "left": "issue:44448", - "right": "issue:44671", + "left": "issue:45183", + "right": "issue:45479", "accept": false, - "reason": "Pegasus v4/v5 output drift vs CamemBERT masked-LM prediction regression; separate model-specific issues." + "reason": "Validation error messaging and classification loss behavior are different problems." }, { - "left": "issue:43716", - "right": "issue:44008", + "left": "issue:45584", + "right": "issue:45834", "accept": false, - "reason": "Mistral-3 image preprocessor dtype mismatch vs Gemma3n audio mask attribute collision; unrelated." + "reason": "Different models and bugs: Whisper empty-transcription generation vs Kosmos2.5 OCR indexing." }, { "left": "issue:43950", "right": "issue:45987", "accept": false, - "reason": "Non-persistent buffer corruption on `from_pretrained` vs `StaticCache.get_seq_length()` returning wrong type; different API bugs." + "reason": "Both touch loading/state, but the bugs are different: buffer corruption vs cache-length return type." }, { "left": "issue:43644", "right": "issue:43749", "accept": false, - "reason": "Non-persistent buffer junk in v5 vs FSDP CPU RAM efficient loading breakage; both loading-related but not the same defect." + "reason": "Generic loading regression vs FSDP CPU RAM-efficient loading; not the same concrete defect." + }, + { + "left": "issue:44448", + "right": "issue:44671", + "accept": false, + "reason": "Different model outputs in v5, but no evidence of a shared root cause." }, { "left": "issue:43582", "right": "issue:46512", "accept": false, - "reason": "Apple Silicon warmup `TypeError` vs stale `config.dtype` after `from_config(dtype=...)`; distinct dtype-related issues." + "reason": "Apple Silicon warmup TypeError and stale config dtype are unrelated dtype/runtime issues." + }, + { + "left": "issue:43716", + "right": "issue:44008", + "accept": false, + "reason": "Image-preprocessor dtype mismatch and Gemma3n variable collision are unrelated bugs." }, { "left": "issue:44485", "right": "issue:44509", "accept": false, - "reason": "GLM-5 RoPE implementation discussion vs docs still mentioning removed pipeline tasks; unrelated." + "reason": "RoPE implementation discussion is unrelated to stale documentation pipeline-task references." }, { "left": "issue:43606", "right": "issue:45072", "accept": false, - "reason": "CPU offload device mismatch for bark-small vs bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel; different bugs." + "reason": "Bark CPU-offload device mismatch and SwitchTransformers/TimmWrapper dtype mismatch are different bugs." }, { "left": "issue:34634", "right": "issue:36010", "accept": false, - "reason": "Bark voice preset failure vs GenerationMixin import error; unrelated." + "reason": "BarkProcessor voice preset handling and GenerationMixin import regression are unrelated." }, { "left": "issue:44479", "right": "issue:45834", "accept": false, - "reason": "Qwen video-input regression vs Kosmos2.5 OCR long-input indexing; different multimodal bugs." + "reason": "Both are video/vision adjacent at most, but one is a Qwen video regression and the other is an OCR index error." + }, + { + "left": "issue:43577", + "right": "issue:43818", + "accept": false, + "reason": "Blip2 dtype loading bug and Video-LLaVA temporal-attention/weight-sharing issue are unrelated." + }, + { + "left": "issue:44805", + "right": "issue:44936", + "accept": false, + "reason": "Mask-shape IndexError and evaluate-after-train failure are not the same concrete problem." + }, + { + "left": "issue:43540", + "right": "issue:45381", + "accept": false, + "reason": "Different Qwen video bugs in different model families and code paths." + }, + { + "left": "issue:43421", + "right": "issue:44811", + "accept": false, + "reason": "Runtime post-processor updates and Whisper batch_decode special-token handling are unrelated." } ] }, @@ -5620,9 +5642,8 @@ "issue:36331", "issue:42175", "issue:43262", - "issue:43421", + "issue:43329", "issue:43526", - "issue:43540", "issue:43577", "issue:43582", "issue:43653", @@ -5631,43 +5652,41 @@ "issue:43742", "issue:43749", "issue:43782", - "issue:43818", "issue:43819", "issue:43825", "issue:43827", "issue:43994", + "issue:44060", "issue:44291", + "issue:44368", "issue:44561", + "issue:44589", "issue:44610", "issue:44671", - "issue:44805", - "issue:44811", "issue:44857", "issue:44912", - "issue:44936", "issue:44977", "issue:44991", "issue:45003", "issue:45020", + "issue:45042", "issue:45072", "issue:45081", "issue:45325", - "issue:45381", + "issue:45372", "issue:45406", + "issue:45600", "issue:45701", "issue:45987", "issue:46192", "issue:46285", "issue:46291", "issue:46294", + "issue:46326", "issue:46463" ], "soft_pairs": [ - "issue:43577|issue:43818", - "issue:44805|issue:44936", - "issue:43540|issue:45381", "issue:43825|issue:43994", - "issue:43421|issue:44811", "issue:43782|issue:44977", "issue:44991|issue:46291", "issue:46294|issue:46463", @@ -5685,212 +5704,213 @@ "issue:34634|issue:36331", "issue:43782|issue:45325", "issue:45003|issue:46285", - "issue:43653|issue:44912", "issue:43262|issue:43819", + "issue:43653|issue:44912", "issue:43526|issue:43697", "issue:43742|issue:45020", - "issue:45081|issue:45406" + "issue:45081|issue:45406", + "issue:44589|issue:45042", + "issue:45372|issue:46326", + "issue:44060|issue:44368", + "issue:43329|issue:45600" ], - "prepared_review_unit_hash": "a6547ece4b831c2ed30ea5df2c0adc078afebeb0771d18e1aa4f25d24b2820ba", + "prepared_review_unit_hash": "5be4629bf838046f6696ed46ab701d02fdde0de3ddd55380165f38d7233e33c4", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 35559, - "estimated_input_tokens": 8890, - "estimated_eval_tokens": 18036 + "serialized_chars": 36258, + "estimated_input_tokens": 9065, + "estimated_eval_tokens": 18386 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is mostly heterogeneous; only one soft pair looks like a plausible duplicate, while the rest describe distinct bugs, model families, or unrelated regressions.", - "confidence": 0.71, - "canonical_issue_reason": "issue:45701 is the broadest, most representative report here: it frames a general tokenizer regression and has the highest inbound references, making it the best cluster anchor.", + "summary": "This is a loose topical cluster of mostly unrelated Transformers bugs/regressions. None of the soft pairs look like the same concrete underlying issue, so I would not merge any of them.", + "confidence": 0.28, + "canonical_issue_reason": "issue:45020 is the broadest umbrella report in the set, covering a general class of recent `remote_code` breakages rather than a narrow model-specific symptom.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45701 is the best overall representative because it is broad, well-referenced, and closest to an umbrella regression report.", + "best_issue_reason": "issue:45020 is the best cluster anchor because it is the most general and could plausibly subsume several version-related loading regressions.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43577", - "right": "issue:43818", - "accept": false, - "reason": "Different model families and different failure modes: dtype loading vs missing temporal attention/weight sharing." - }, - { - "left": "issue:44805", - "right": "issue:44936", - "accept": false, - "reason": "One is a mask-shape IndexError; the other is a train/evaluate lifecycle failure." - }, - { - "left": "issue:43540", - "right": "issue:45381", - "accept": false, - "reason": "Both involve vision/video inputs, but they target different Qwen model families and different bugs." - }, { "left": "issue:43825", "right": "issue:43994", "accept": false, - "reason": "One is an error-message regression around pipeline removal; the other is an output-quality issue in a specific model pipeline." - }, - { - "left": "issue:43421", - "right": "issue:44811", - "accept": false, - "reason": "Tokenizer post-processor updates and Whisper batch_decode skipping special tokens are different code paths." + "reason": "Both involve pipeline/model behavior, but one is an error-message regression and the other is a model output quality issue; different bugs." }, { "left": "issue:43782", "right": "issue:44977", "accept": false, - "reason": "Loading with weight_only=True and flash-attention generation issues are unrelated." + "reason": "Different models, different failure modes (`weight_only` load error vs flash-attention generation issue)." }, { "left": "issue:44991", "right": "issue:46291", "accept": false, - "reason": "Tokenizer loading regression and kernels import/version requirements are separate breakages." + "reason": "Tokenizer loading failure and a pipelines import dependency error are unrelated code paths." }, { "left": "issue:46294", "right": "issue:46463", "accept": false, - "reason": "Qwen video chat-template mismatch and T5 decoder-input validation are unrelated bugs." + "reason": "Both touch chat-template/tokenizer logic, but they affect different models and distinct behaviors." }, { "left": "issue:44561", "right": "issue:45020", "accept": false, - "reason": "45020 is a broad remote_code regression report; 44561 is a specific removed-helper breakage, not the same concrete bug." + "reason": "Related broad theme, but one is a specific removed helper causing remote-code failures while the other is a generic recent regression report; not the same concrete bug." }, { "left": "issue:44610", "right": "issue:45072", "accept": false, - "reason": "Processor image-size mismatch and bfloat16 dtype mismatch are different problems." + "reason": "Processor image-size bug vs dtype mismatch; different subsystems and symptoms." }, { "left": "issue:43720", "right": "issue:45072", "accept": false, - "reason": "Packed-weight loading failure is unrelated to dtype mismatches in inference." + "reason": "Packed-weight loading failure is unrelated to bfloat16 dtype mismatches." }, { "left": "issue:43577", "right": "issue:43582", "accept": false, - "reason": "Blip2 dtype persistence and Apple Silicon warmup TypeError are unrelated." + "reason": "Blip2 dtype persistence and Apple Silicon allocator warmup TypeError are separate issues." }, { "left": "issue:45987", "right": "issue:46192", "accept": false, - "reason": "Cache-length return type and return_tensors precedence are different API bugs." + "reason": "Cache sequence-length return type bug and `return_tensors` kwargs handling are different API regressions." }, { "left": "issue:43825", "right": "issue:43827", "accept": false, - "reason": "Both mention v5 pipeline removals, but one is runtime messaging and the other is documentation drift." + "reason": "One is an incorrect runtime message, the other is stale documentation; related area but not the same bug." }, { "left": "issue:28218", "right": "issue:29159", - "accept": true, - "reason": "Both point to tokenizer added-token handling inconsistencies, including unwanted spacing and slow/fast tokenizer divergence." + "accept": false, + "reason": "Both are tokenizer complaints, but one is about added-token spacing and the other is a broad slow/fast inconsistency report." }, { "left": "issue:43749", "right": "issue:44291", "accept": false, - "reason": "FSDP CPU RAM-efficient loading and init_empty_weights/_is_hf_initialized are separate loading regressions." + "reason": "Different loading paths and different exceptions (`FSDP` RAM-efficient loading vs `_is_hf_initialized` argument)." }, { "left": "issue:44671", "right": "issue:45701", "accept": false, - "reason": "CamemBERT masked-LM prediction regression is not the same bug as a broad tokenizer-version change report." + "reason": "Masked LM prediction quality regression is not the same as a tokenizer behavior change report." }, { "left": "issue:42175", "right": "issue:45081", "accept": false, - "reason": "Backend-install dependency packaging and Mistral tokenizer patch crashes are unrelated." + "reason": "Backend packaging/dependency issue vs tokenizer regex patch crash; unrelated." }, { "left": "issue:43582", "right": "issue:44857", "accept": false, - "reason": "Apple Silicon allocator warmup TypeError and CUDA AMP loss crash are different platform-specific issues." + "reason": "Apple Silicon warmup TypeError and CUDA AMP crash are different platform-specific bugs." }, { "left": "issue:34634", "right": "issue:36331", "accept": false, - "reason": "BarkProcessor voice_preset and CustomTrainer.compute_loss signature mismatch are unrelated." + "reason": "Processor voice preset bug and trainer loss-signature mismatch are unrelated." }, { "left": "issue:43782", "right": "issue:45325", "accept": false, - "reason": "Qwen3VL weight-only loading and Qwen2.5-VL rope-index scaling are different bugs." + "reason": "Different Qwen VL bugs: load-time `weight_only` failure vs rope-index scaling bug." }, { "left": "issue:45003", "right": "issue:46285", "accept": false, - "reason": "Unsafe sys.modules access and CLIP text-model structure inconsistency are unrelated." + "reason": "Unsafe `sys.modules` access and CLIP class structure inconsistency are different code-path problems." }, { - "left": "issue:43653", - "right": "issue:44912", + "left": "issue:43262", + "right": "issue:43819", "accept": false, - "reason": "BigBird special-token registration and MXFP4 quantization loading are different issues." + "reason": "Audio sampling-rate default bug vs DAC latent reconstruction mismatch; not the same defect." }, { - "left": "issue:43262", - "right": "issue:43819", + "left": "issue:43653", + "right": "issue:44912", "accept": false, - "reason": "Audio sampling-rate defaults and DAC from_latents/forward mismatch are unrelated." + "reason": "Tokenizer special-token registration and MXFP4 quantization fallback are unrelated." }, { "left": "issue:43526", "right": "issue:43697", "accept": false, - "reason": "BEiT label reduction and RTDetrV2 output differences are different model bugs." + "reason": "Beit label reduction bug and RT-DETR output mismatch are different model/processor regressions." }, { "left": "issue:43742", "right": "issue:45020", "accept": false, - "reason": "A specific MobileLLM loading key error is not the same as a broad remote_code compatibility report." + "reason": "A model-specific key error is not the same as the broader remote-code regression report." }, { "left": "issue:45081", "right": "issue:45406", "accept": false, - "reason": "Mistral regex-patching failure and Gemma4Processor missing _tokenizer are separate AttributeError paths." + "reason": "Both are AttributeErrors, but they occur in different components (`mistral` tokenizer patch vs Gemma4 processor)." + }, + { + "left": "issue:44589", + "right": "issue:45042", + "accept": false, + "reason": "Float8 storage lookup failure and PIL/torchvision dependency regression are unrelated." + }, + { + "left": "issue:45372", + "right": "issue:46326", + "accept": false, + "reason": "Different tokenizer/model-loading failures with different external dependency and API surfaces." + }, + { + "left": "issue:44060", + "right": "issue:44368", + "accept": false, + "reason": "Both mention tied-weights warnings, but one is a mistaken tie across parameters and the other is a training-time warning; not the same bug." + }, + { + "left": "issue:43329", + "right": "issue:45600", + "accept": false, + "reason": "Multimodal token counting bug and auto-mapping config reference cleanup are unrelated." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet: issue:45701 does have the highest inbound references and is a reasonable umbrella tokenizer-regression anchor. The soft-edge verdicts are generally conservative, and the single accepted pair (issue:28218 / issue:29159) is plausibly the same tokenizer-added-token behavior class. No mergeability problems are apparent from the provided data." - }, - "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 }, @@ -5916,6 +5936,7 @@ "issue:43257", "issue:43262", "issue:43329", + "issue:43450", "issue:43452", "issue:43526", "issue:43531", @@ -5923,20 +5944,22 @@ "issue:43638", "issue:43643", "issue:43645", + "issue:43650", + "issue:43696", "issue:43720", "issue:43746", + "issue:43749", "issue:43819", "issue:43854", "issue:43906", "issue:43927", - "issue:44060", - "issue:44368", + "issue:43931", "issue:44387", "issue:44479", "issue:44554", - "issue:44589", "issue:44610", "issue:44617", + "issue:44756", "issue:44811", "issue:44912", "issue:44936", @@ -5944,24 +5967,17 @@ "issue:44991", "issue:45042", "issue:45084", - "issue:45372", "issue:45459", "issue:45538", - "issue:45600", "issue:45759", "issue:45803", "issue:46209", "issue:46291", - "issue:46326", "issue:46421", "issue:46424" ], "soft_pairs": [ - "issue:44589|issue:45042", - "issue:45372|issue:46326", - "issue:44060|issue:44368", "issue:45084|issue:45538", - "issue:43329|issue:45600", "issue:43906|issue:46209", "issue:43526|issue:44554", "issue:43550|issue:44912", @@ -5975,8 +5991,8 @@ "issue:43531|issue:45042", "issue:43720|issue:44610", "issue:38175|issue:44610", - "issue:43746|issue:44811", "issue:44387|issue:44617", + "issue:43746|issue:44811", "issue:39692|issue:44912", "issue:42490|issue:43257", "issue:43854|issue:44977", @@ -5984,22 +6000,27 @@ "issue:43262|issue:44811", "issue:45459|issue:45803", "issue:43329|issue:44479", - "issue:46421|issue:46424" + "issue:46421|issue:46424", + "issue:43257|issue:43696", + "issue:43650|issue:43906", + "issue:43329|issue:43450", + "issue:43257|issue:43931", + "issue:43749|issue:44756" ], - "prepared_review_unit_hash": "79ec33102c85573f6945331aff455ec99fb7fef7abea87109e3ba4058988a84d", + "prepared_review_unit_hash": "04c94ca0608427461d2c13c92fb5e6bbb25e8197671bb564217ed2a158525696", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 28, - "serialized_chars": 36471, + "soft_pair_count": 29, + "serialized_chars": 36472, "estimated_input_tokens": 9118, "estimated_eval_tokens": 18492 }, @@ -6008,186 +6029,192 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is mostly a grab bag of unrelated Transformers bugs. I found one plausible duplicate pair around the Qwen video-input regression; the rest are not the same underlying issue and should stay separate.", - "confidence": 0.69, - "canonical_issue_reason": "issue:43329 is the best canonical issue for the Qwen video regression because it identifies the concrete broken code path in `_get_num_multimodal_tokens` (undefined video helpers / merge sizing) rather than only describing the downstream user-visible failure.", + "summary": "This cluster is mostly a grab-bag of unrelated Transformer issues. I found one likely true duplicate pair around the Qwen video regression; the rest look distinct and should stay separate.", + "confidence": 0.89, + "canonical_issue_reason": "Issue 44479 is the broadest and most representative report of the Qwen video-input regression across multiple model variants; it captures the user-facing failure more completely than the narrower root-cause-style report in 43329.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43329 is the strongest global representative: it gives the clearest root cause and is the most actionable merge target for the broader v5.3.0 Qwen video-input regression reported in issue:44479.", + "best_issue_reason": "Issue 44479 is the best single issue to represent the only plausible duplicate set here because it is the most general, clearly scoped, and highest-impact description of the underlying video regression.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44589", - "right": "issue:45042", - "accept": false, - "reason": "Different failures in different subsystems: FP8 storage lookup vs PIL image processor torchvision dependency." - }, - { - "left": "issue:45372", - "right": "issue:46326", - "accept": false, - "reason": "Unrelated tokenizer/processor import failure vs mistral_common continue_final_message behavior." - }, - { - "left": "issue:44060", - "right": "issue:44368", - "accept": false, - "reason": "Both mention Qwen warnings, but one is a tied-weights bug and the other is a LoRA fine-tuning warning; not the same concrete bug." - }, { "left": "issue:45084", "right": "issue:45538", "accept": false, - "reason": "Compilation error and tokenizer max_length bug are unrelated." - }, - { - "left": "issue:43329", - "right": "issue:45600", - "accept": false, - "reason": "Video-token counting bug vs stale auto_mappings entry; different code paths and failures." + "reason": "Template-compilation failure and tokenizer max_length default are unrelated bugs." }, { "left": "issue:43906", "right": "issue:46209", "accept": false, - "reason": "A reproduction of an older issue is not the same as an FP8 inference failure on B200." + "reason": "One is a reproduction of an older issue; the other is an FP8 inference/OOM problem on B200. Different failure modes." }, { "left": "issue:43526", "right": "issue:44554", "accept": false, - "reason": "BEiT label reduction bug and MPS attention correctness issue are unrelated." + "reason": "BEiT label reduction and MPS attention correctness are separate code paths and symptoms." }, { "left": "issue:43550", "right": "issue:44912", "accept": false, - "reason": "Different model-specific failures: torch.compile/SDPA vs MXFP4 load fallback." + "reason": "torch.compile+SDPA failure in Bamba and MXFP4 quantization fallback in git-oss are different model/runtime issues." }, { "left": "issue:43638", "right": "issue:44936", "accept": false, - "reason": "Both involve training/eval, but one is a ZeRO-3 index error and the other is trainer lifecycle behavior after training." + "reason": "ZeRO3/non-pretrained Bert indexing error is not the same as train-then-evaluate failure." }, { "left": "issue:43452", "right": "issue:44991", "accept": false, - "reason": "Both concern loading, but one is gguf_file behavior across tokenizer/model loading and the other is a specific tokenizer load failure." + "reason": "gguf_file integration breakage and tokenizer loading regression for a specific model are different problems." }, { "left": "issue:43257", "right": "issue:43329", "accept": false, - "reason": "Accelerate/DeepSpeed weight conversion bug is unrelated to the multimodal video-token helper bug." + "reason": "Qwen3 MoE weight conversion/loading and multimodal video token counting are different bugs." }, { "left": "issue:43819", "right": "issue:43927", "accept": false, - "reason": "DAC STE mismatch and DiaConfig save/load token ID loss are different model/config bugs." + "reason": "DAC latent reconstruction mismatch and DiaConfig custom token loss are unrelated." }, { "left": "issue:36010", "right": "issue:43645", "accept": false, - "reason": "Importing GenerationMixin and Jupyter custom model initialization are separate breakages." + "reason": "GenerationMixin import failure and Jupyter custom-model initialization breakage are distinct issues." }, { "left": "issue:36010", "right": "issue:36331", "accept": false, - "reason": "Import error for GenerationMixin and unexpected compute_loss kwarg are unrelated." + "reason": "ImportError for GenerationMixin and compute_loss signature regression are not the same bug." }, { "left": "issue:42175", "right": "issue:46291", "accept": false, - "reason": "Packaging backend dependency issue vs kernels package version/revision requirement are different problems." + "reason": "Packaging/backends selection issue is unrelated to kernels dependency/version validation on pipelines import." }, { "left": "issue:43531", "right": "issue:45042", "accept": false, - "reason": "Qwen sliding-window behavior and PIL torchvision dependency are unrelated." + "reason": "Sliding-window behavior in Qwen3-MoE and unintended torchvision requirement in PIL backend processors do not share the same root cause." }, { "left": "issue:43720", "right": "issue:44610", "accept": false, - "reason": "BitNet accelerate-loading unpacking bug and OmDet-Turbo image-size mismatch are different code paths." + "reason": "Packed-weight loading in BitNet and processor/model input-size mismatch in OmDet-Turbo are different failure classes." }, { "left": "issue:38175", "right": "issue:44610", "accept": false, - "reason": "SigLIP2 zero probabilities and OmDet-Turbo processor/model size mismatch are unrelated." + "reason": "Unexpected zero probabilities in SigLIP2 is unrelated to OmDet-Turbo\u2019s 640px/224px preprocessing mismatch." }, { - "left": "issue:43746", - "right": "issue:44811", + "left": "issue:44387", + "right": "issue:44617", "accept": false, - "reason": "PEFT local checkpoint loading and Whisper batch_decode skip_special_tokens are unrelated bugs." + "reason": "Int4 quantization memory regression and Sam3Video OOM are different issues with different triggers." }, { - "left": "issue:44387", - "right": "issue:44617", + "left": "issue:43746", + "right": "issue:44811", "accept": false, - "reason": "General int4 memory regression and a model-specific CUDA OOM report are not the same issue." + "reason": "PEFT local checkpoint loading and Whisper batch_decode special-token handling are unrelated." }, { "left": "issue:39692", "right": "issue:44912", "accept": false, - "reason": "Doc/example errors plus quantization failure vs a load/quantization fallback bug in a different model." + "reason": "Documentation/example errors for SigLIP2 are not the same as MXFP4 quantization loading behavior." }, { "left": "issue:42490", "right": "issue:43257", "accept": false, - "reason": "Local-path import warning and Qwen MoE accelerate/deepspeed conversion are unrelated." + "reason": "Slash-path import warning and Qwen3 MoE accelerate/deepspeed loading are unrelated." }, { "left": "issue:43854", "right": "issue:44977", "accept": false, - "reason": "GLM-4.7-Flash test loading failure and Qwen3.5 flash-attention generation regression are separate model issues." + "reason": "Model loading test failures for GLM-4.7-Flash and flash-attention generation problems in Qwen3.5 are distinct." }, { "left": "issue:43643", "right": "issue:45759", "accept": false, - "reason": "Missing fields with trust_remote_code and text_config unwrapping for composite configs are different config-loading bugs." + "reason": "Missing fields in AutoConfig with trust_remote_code and not unwrapping text_config in AutoModelForCausalLM.from_config are related to config handling but not the same concrete bug." }, { "left": "issue:43262", "right": "issue:44811", "accept": false, - "reason": "Audio chat-template sample-rate default and Whisper batch_decode token skipping are unrelated." + "reason": "Audio chat-template sample-rate default and Whisper decode token filtering are different processor bugs." }, { "left": "issue:45459", "right": "issue:45803", "accept": false, - "reason": "Both involve error handling, but they affect different code paths and different classes of hidden errors." + "reason": "Both involve error handling, but they affect different functions and different swallowed exceptions." }, { "left": "issue:43329", "right": "issue:44479", "accept": true, - "reason": "Both describe the same Qwen-family video-input regression; 43329 exposes the broken `_get_num_multimodal_tokens` video branch, and 44479 reports the resulting failure across qwen2_5_vl/qwen3_vl/qwen3_5 variants." + "reason": "Both describe the same Qwen video-input regression in the multimodal/video code path; 43329 points to the undefined helper bug, while 44479 reports the broader regression across affected model variants." }, { "left": "issue:46421", "right": "issue:46424", "accept": false, - "reason": "Both mention cache behavior, but one is lazy init causing recompilation and the other is max_cache_length scoping for quantized cache only; not the same bug." + "reason": "Lazy cache initialization causing recompilation and max_cache_length handling for quantized cache are separate cache-related bugs." + }, + { + "left": "issue:43257", + "right": "issue:43696", + "accept": false, + "reason": "Qwen3 MoE loading/conversion and GPT-oss-20b OOM are unrelated." + }, + { + "left": "issue:43650", + "right": "issue:43906", + "accept": false, + "reason": "The placeholder issue title and an isolated reproduction of another issue do not indicate the same underlying bug." + }, + { + "left": "issue:43329", + "right": "issue:43450", + "accept": false, + "reason": "Undefined video token helpers and batched video shape mismatch are both video-related, but they are not the same concrete defect." + }, + { + "left": "issue:43257", + "right": "issue:43931", + "accept": false, + "reason": "MoE weight conversion/loading and Qwen3-VL shape mismatch are different model-loading failures." + }, + { + "left": "issue:43749", + "right": "issue:44756", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and mmap-related OOM on Strix Halo are different memory/loading issues." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft pair, issue:43329 vs issue:44479, looks like the same Qwen-family video-input regression and is a reasonable duplicate/merge target. The rest of the pairwise rejections are appropriately cautious and do not overstate evidence." + "feedback": "The summary is grounded in the packet and stays conservative overall. The one accepted duplicate pair (43329/44479) is plausibly the same Qwen video regression, with 44479 as the broader report. The other soft-edge rejections are mostly well-justified and do not overstate duplication. No mergeability issues apply since these are issue pairs, not PR pairs." }, "evaluator_used": true, "retried": false, @@ -6208,6 +6235,7 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-469", "nodes": [ + "issue:24643", "issue:28282", "issue:29127", "issue:29159", @@ -6215,41 +6243,40 @@ "issue:34567", "issue:34689", "issue:40990", + "issue:41720", "issue:42175", "issue:42915", - "issue:43257", "issue:43317", - "issue:43329", "issue:43404", "issue:43425", - "issue:43450", "issue:43452", "issue:43454", "issue:43525", + "issue:43540", "issue:43577", "issue:43582", "issue:43606", - "issue:43650", - "issue:43696", "issue:43697", "issue:43742", "issue:43749", "issue:43824", - "issue:43906", "issue:43927", - "issue:43931", + "issue:44062", "issue:44291", + "issue:44479", + "issue:44560", "issue:44568", - "issue:44756", "issue:44821", "issue:44849", "issue:44912", "issue:44987", + "issue:44991", "issue:45042", "issue:45127", "issue:45137", "issue:45356", "issue:45357", + "issue:45381", "issue:45405", "issue:45538", "issue:45701", @@ -6258,17 +6285,12 @@ "issue:46512" ], "soft_pairs": [ - "issue:43257|issue:43696", - "issue:43650|issue:43906", - "issue:43329|issue:43450", - "issue:43257|issue:43931", - "issue:43749|issue:44756", "issue:42915|issue:43742", "issue:43606|issue:45859", "issue:43454|issue:43577", "issue:44291|issue:45042", - "issue:42915|issue:44821", "issue:34567|issue:40990", + "issue:42915|issue:44821", "issue:43525|issue:46291", "issue:43927|issue:44912", "issue:29159|issue:34567", @@ -6287,223 +6309,245 @@ "issue:43317|issue:45538", "issue:43742|issue:44291", "issue:44849|issue:45701", - "issue:28282|issue:29127" + "issue:28282|issue:29127", + "issue:44062|issue:44991", + "issue:44479|issue:45381", + "issue:41720|issue:43540", + "issue:43540|issue:44479", + "issue:43540|issue:44560", + "issue:42915|issue:43749", + "issue:24643|issue:29159" ], - "prepared_review_unit_hash": "dd1df68bbcb64b881014299482efb5eeace63f429fcd3a566bfe1a7f9beae5ef", + "prepared_review_unit_hash": "49e0176932011d0cc61f5e5e0f5e6faeeade3aaa803f3dad32cb91ac66ded46e", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36704, - "estimated_input_tokens": 9176, - "estimated_eval_tokens": 18608 + "soft_pair_count": 32, + "serialized_chars": 37342, + "estimated_input_tokens": 9336, + "estimated_eval_tokens": 18928 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is highly heterogeneous: most pairs only share broad themes like loading, tokenization, dtype, or offload, but not the same underlying bug. I rejected all soft edges and treated this as a mixed issue set rather than a true duplicate cluster.", - "confidence": 0.64, - "canonical_issue_reason": "Issue:46512 is the most general, core API-level regression in the set (dtype state after from_config), so it is the best single representative even though the cluster is otherwise mixed.", + "summary": "This cluster is mostly heterogeneous: model loading, tokenizer, dtype, offload, and video-processing bugs are mixed together. Only the Qwen VL video-input regression pair looks like a plausible duplicate; the rest are separate issues.", + "confidence": 0.46, + "canonical_issue_reason": "issue:44479 is the best representative issue because it describes the broad 5.3.0 Qwen video-input regression across multiple affected Qwen VL variants, which is the closest thing in this cluster to a shared underlying bug.", "canonical_pr_reason": null, - "best_issue_reason": "Broad, user-facing loading/config bug with a clear current regression and the least model-specific scope among the candidates.", + "best_issue_reason": "issue:44479 is the strongest cluster anchor: it is broad, symptom-focused, and explicitly frames the regression in the same Qwen video-input code path that the closest soft match also reports.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43257", - "right": "issue:43696", - "accept": false, - "reason": "Different models and symptoms: MOE weight conversion vs CUDA OOM." - }, - { - "left": "issue:43650", - "right": "issue:43906", - "accept": false, - "reason": "No shared bug; a placeholder-style issue is unrelated to an isolated reproduction report." - }, - { - "left": "issue:43329", - "right": "issue:43450", - "accept": false, - "reason": "Both are video-related, but one is token counting and the other is processor output shape." - }, - { - "left": "issue:43257", - "right": "issue:43931", - "accept": false, - "reason": "Different Qwen variants and different loading failures; not the same bug." - }, - { - "left": "issue:43749", - "right": "issue:44756", - "accept": false, - "reason": "Both involve memory/loading, but one is FSDP RAM-efficient loading and the other is Strix Halo mmap OOM." - }, { "left": "issue:42915", "right": "issue:43742", "accept": false, - "reason": "Different model families and failures; FP8 config bug vs key error on load." + "reason": "Both are loading/quantization failures, but they affect different models and report different failures (FineGrainedFP8Config vs. key error on MobileLLM). No clear same bug." }, { "left": "issue:43606", "right": "issue:45859", "accept": false, - "reason": "Both touch CPU offload, but they hit different models and different code paths." + "reason": "Both mention CPU offload, but one is a Bark-small device mismatch and the other is a Qwen3.5 MoE rotary embedding compatibility issue. Different code paths and models." }, { "left": "issue:43454", "right": "issue:43577", "accept": false, - "reason": "Both are model-loading issues, but one is untied lm_head weights and the other is dtype propagation." + "reason": "One is about tying lm_head weights in AyaVision/Mistral3; the other is about BLIP2 dtype propagation. Similar theme, but not the same underlying bug." }, { "left": "issue:44291", "right": "issue:45042", "accept": false, - "reason": "Unexpected init argument vs torchvision dependency regression are unrelated bugs." + "reason": "Init-empty-weights argument handling and PIL image-processor torchvision requirements are unrelated." }, { - "left": "issue:42915", - "right": "issue:44821", + "left": "issue:34567", + "right": "issue:40990", "accept": false, - "reason": "Different areas: MOE FP8 loading vs image processor URL loading." + "reason": "TrainerState token accounting and perplexity on gpt-oss-20b are unrelated." }, { - "left": "issue:34567", - "right": "issue:40990", + "left": "issue:42915", + "right": "issue:44821", "accept": false, - "reason": "Trainer token counting is unrelated to perplexity quality on a specific model." + "reason": "Different features and failures: FP8 config loading vs AutoImageProcessor-from-URL behavior." }, { "left": "issue:43525", "right": "issue:46291", "accept": false, - "reason": "Missing pad_token_id on a config is unrelated to kernels package import requirements." + "reason": "Llama4 pad_token_id access and kernels package import requirements are unrelated." }, { "left": "issue:43927", "right": "issue:44912", "accept": false, - "reason": "Config token persistence bug vs quantization fallback bug are different issues." + "reason": "DiaConfig token-ID persistence and MXFP4 quantization fallback are separate problems." }, { "left": "issue:29159", "right": "issue:34567", "accept": false, - "reason": "Tokenizer inconsistency and trainer token accounting are separate problems." + "reason": "Tokenizer slow/fast inconsistency is not the same bug as TrainerState num_input_tokens_seen not updating." }, { "left": "issue:44568", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different tokenizers and different behavior." + "reason": "Different tokenizer regressions with different models and failure modes." }, { "left": "issue:44987", "right": "issue:45538", "accept": false, - "reason": "Model loading regression vs tokenizer max-length constant are not the same bug." + "reason": "Loading a specific model and CLIPTokenizer max-length handling are unrelated." }, { "left": "issue:43697", "right": "issue:45357", "accept": false, - "reason": "RTDetr output mismatch and Qwen3.5 save_pretrained key errors are unrelated." + "reason": "RTDetrV2 output drift in v5 and Qwen3.5 save_pretrained key mapping are different issues." }, { "left": "issue:43404", "right": "issue:45127", "accept": false, - "reason": "Both mention tied embeddings, but one is a missing lm_head tie and the other is LoRA merge collapse." + "reason": "Both involve tied embeddings, but one is a missing weight-tying bug in a model class and the other is LoRA merge behavior with extended vocab. Not the same concrete bug." }, { "left": "issue:42915", "right": "issue:45538", "accept": false, - "reason": "MOE FP8 loading and CLIPTokenizer max-length are unrelated." + "reason": "FP8 loading and CLIP tokenizer model_max_length are unrelated." }, { "left": "issue:43425", "right": "issue:46291", "accept": false, - "reason": "Torch compatibility and kernels import dependency issues are different root causes." + "reason": "Torch version compatibility and kernels import requirements are unrelated." }, { "left": "issue:45137", "right": "issue:45405", "accept": false, - "reason": "DeepSpeed deque failure and a PEFT version bump are unrelated." + "reason": "DeepSpeed ZeRO3 deque error and a PEFT minimum-version bump are unrelated." }, { "left": "issue:28282", "right": "issue:34689", "accept": false, - "reason": "Missing PyTorch dependency vs a specific model-loading regression are not the same issue." + "reason": "PyTorch missing from environment and Llama 3.2 90B Vision model loading regression are unrelated." }, { "left": "issue:43577", "right": "issue:46512", "accept": false, - "reason": "Both are dtype-related, but one is incorrect loaded tensor dtypes and the other is stale config metadata." + "reason": "Both concern dtype state, but one is BLIP2 loading behavior and the other is from_config leaving config.dtype stale. Different API paths and fixes." }, { "left": "issue:42175", "right": "issue:43824", "accept": false, - "reason": "Backend packaging omission and a missing model import are unrelated." + "reason": "TensorFlow extra packaging and missing Qwen2.5-VL class import are separate packaging/import issues." }, { "left": "issue:33357", "right": "issue:34567", "accept": false, - "reason": "MacOS bus error on CLIP loading is unrelated to trainer token counters." + "reason": "MacOS bus error on CLIP model loading and TrainerState token count updates are unrelated." }, { "left": "issue:43452", "right": "issue:43582", "accept": false, - "reason": "GGUF loading breakage and AppleSilicon allocator TypeError are different problems." + "reason": "GGUF file loading and Apple Silicon caching allocator warmup are unrelated." }, { "left": "issue:43317", "right": "issue:45538", "accept": false, - "reason": "Device-map offload failure is unrelated to tokenizer max-length behavior." + "reason": "Device-map/offload dequantized model loading and CLIP tokenizer max-length are unrelated." }, { "left": "issue:43742", "right": "issue:44291", "accept": false, - "reason": "Key error loading a model and an init_empty_weights signature change are different bugs." + "reason": "Key error on model loading and init_empty_weights argument handling are distinct problems." }, { "left": "issue:44849", "right": "issue:45701", "accept": false, - "reason": "Qwen3.5 hidden-state behavior and version-dependent tokenization changes do not match." + "reason": "Qwen3.5 output_hidden_states behavior and tokenizer changes across versions are different issues." }, { "left": "issue:28282", "right": "issue:29127", "accept": false, - "reason": "Missing PyTorch dependency and unclear LayoutLMv3 box-data error are unrelated." + "reason": "Missing PyTorch import and LayoutLMv3 box-content error are unrelated." + }, + { + "left": "issue:44062", + "right": "issue:44991", + "accept": false, + "reason": "AddedToken special-argument handling and est-roberta tokenizer loading are different tokenizer failures." + }, + { + "left": "issue:44479", + "right": "issue:45381", + "accept": true, + "reason": "Both report the same Qwen VL video-input regression in 5.3.0, with 45381 describing a specific symptom (vision_position_ids) of the broader broken video path in 44479." + }, + { + "left": "issue:41720", + "right": "issue:43540", + "accept": false, + "reason": "Qwen3 auto device mapping cudaErrorAssert and Qwen3OmniMoe video ValueError are different failure modes." + }, + { + "left": "issue:43540", + "right": "issue:44479", + "accept": false, + "reason": "Both are video-related Qwen issues, but they affect different model families and symptoms; not enough evidence they are the same bug." + }, + { + "left": "issue:43540", + "right": "issue:44560", + "accept": false, + "reason": "Qwen3OmniMoe video ValueError and Qwen3-vl-embedding StopIteration are distinct video bugs." + }, + { + "left": "issue:42915", + "right": "issue:43749", + "accept": false, + "reason": "FP8 config loading and FSDP CPU RAM efficient loading are unrelated training/loading issues." + }, + { + "left": "issue:24643", + "right": "issue:29159", + "accept": false, + "reason": "DeepSpeed weight dimensionality error and tokenizer slow/fast inconsistency are unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall. The cluster summary is conservative and supported by the packet, and the only accepted soft edge (44479/45381) is plausibly the same Qwen VL 5.3.0 video-input regression. The rejected pairs are mostly clearly distinct by model, API path, or failure mode, without overstating linkage." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -6520,31 +6564,26 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-469", "nodes": [ - "issue:24643", "issue:27957", "issue:28282", - "issue:29159", "issue:34567", "issue:36331", "issue:38175", - "issue:41720", "issue:42175", - "issue:42915", "issue:43257", + "issue:43404", + "issue:43452", "issue:43493", "issue:43526", - "issue:43540", + "issue:43531", "issue:43577", "issue:43742", - "issue:43749", "issue:43756", "issue:43906", "issue:43994", "issue:44062", "issue:44291", - "issue:44479", "issue:44488", - "issue:44560", "issue:44568", "issue:44589", "issue:44737", @@ -6553,36 +6592,33 @@ "issue:44898", "issue:44964", "issue:44987", - "issue:44991", "issue:45042", "issue:45072", "issue:45081", "issue:45325", + "issue:45357", "issue:45381", + "issue:45405", "issue:45479", "issue:45488", "issue:45588", + "issue:45600", "issue:45823", "issue:45907", + "issue:45920", + "issue:45987", "issue:46399", "issue:46463", "issue:46468", "issue:46491" ], "soft_pairs": [ - "issue:44062|issue:44991", - "issue:44479|issue:45381", - "issue:41720|issue:43540", - "issue:43540|issue:44560", - "issue:43540|issue:44479", - "issue:42915|issue:43749", - "issue:24643|issue:29159", "issue:42175|issue:44488", "issue:38175|issue:44898", "issue:43493|issue:43994", + "issue:27957|issue:28282", "issue:44987|issue:45907", "issue:44568|issue:45488", - "issue:27957|issue:28282", "issue:44291|issue:44821", "issue:34567|issue:36331", "issue:44737|issue:46463", @@ -6590,224 +6626,230 @@ "issue:43577|issue:45072", "issue:44792|issue:44964", "issue:44062|issue:45081", - "issue:45042|issue:45081", "issue:43742|issue:44589", + "issue:45042|issue:45081", + "issue:44737|issue:45588", "issue:44568|issue:46491", "issue:43906|issue:46468", "issue:43756|issue:46399", - "issue:44737|issue:45588", "issue:43756|issue:45325", "issue:43526|issue:45479", - "issue:45823|issue:46463" + "issue:43404|issue:45987", + "issue:45823|issue:46463", + "issue:43531|issue:44821", + "issue:45405|issue:45600", + "issue:43452|issue:44488", + "issue:43906|issue:45381", + "issue:45357|issue:45920" ], - "prepared_review_unit_hash": "1ccabac06da892fad0169fbe138f17b054ad84a716fd3c437e87871381f8b12b", + "prepared_review_unit_hash": "6db46b3076b74b6d63af50fa7eff74ddab4c13bf557f7e23ca1c374f5f4dd3fe", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36647, - "estimated_input_tokens": 9162, - "estimated_eval_tokens": 18580 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35951, + "estimated_input_tokens": 8988, + "estimated_eval_tokens": 18232 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated transformer issues, with one clear duplicate-like Qwen video-input regression pair and one plausible SigLIP2 duplicate pair.", - "confidence": 0.69, - "canonical_issue_reason": "Issue 44479 is the broadest, clearest regression report in the cluster and best captures the shared Qwen video-input bug family; 45381 reads like a narrower instance of the same problem.", + "summary": "Most proposed soft links are false positives: they share a library area or model family, but describe different failures and code paths. The one strong duplicate candidate is the pair of SigLIP2 reports, which both point to the same model producing invalid outputs. For a representative issue, the broad multi-model tokenizer regression is the best anchor.", + "confidence": 0.76, + "canonical_issue_reason": "issue:45920 is the broadest, most user-impacting regression in the set: it describes a v5 tokenizer bug affecting multiple model families, so it is the best representative issue for triage and duplicate collection.", "canonical_pr_reason": null, - "best_issue_reason": "44479 is the strongest duplicate-triage anchor because it describes the underlying regression at a higher level and matches the broader symptom scope seen in the soft duplicate candidate.", + "best_issue_reason": "issue:45920 has the widest scope and clearest cross-model impact, making it the strongest single issue to keep visible as the cluster anchor.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44062", - "right": "issue:44991", + "left": "issue:42175", + "right": "issue:44488", "accept": false, - "reason": "Different tokenizer/model-loading failures; one is an AddedToken keyword collision, the other is an AutoTokenizer load failure for est-roberta." + "reason": "Different problems: packaging/backend availability versus a specific model load failure." }, { - "left": "issue:44479", - "right": "issue:45381", + "left": "issue:38175", + "right": "issue:44898", + "accept": false, + "reason": "Both are vision-model bugs, but they concern different models and different failure modes." + }, + { + "left": "issue:43493", + "right": "issue:43994", "accept": true, - "reason": "Both point to the same Qwen2.5-VL/Qwen video-input regression around incorrect position IDs in transformers 5.3.0." + "reason": "Both describe the same SigLIP2 model producing incorrect/nonsensical outputs, likely the same underlying implementation bug." }, { - "left": "issue:41720", - "right": "issue:43540", + "left": "issue:27957", + "right": "issue:28282", "accept": false, - "reason": "Different bugs: auto device-mapping CUDA assert versus Qwen3OmniMoe video-input ValueError." + "reason": "Unrelated: Flash Attention support for XLM-Roberta versus a missing PyTorch installation error." }, { - "left": "issue:43540", - "right": "issue:44560", + "left": "issue:44987", + "right": "issue:45907", "accept": false, - "reason": "Both are video-related, but they affect different Qwen model families and have different failure modes." + "reason": "One is a model-loading regression; the other is a repo-file listing API forwarding bug in CI." }, { - "left": "issue:43540", - "right": "issue:44479", + "left": "issue:44568", + "right": "issue:45488", "accept": false, - "reason": "Different Qwen families and different symptoms; no clear shared code-path bug." + "reason": "Both are tokenizer regressions, but they affect different tokenizer implementations and different behaviors." }, { - "left": "issue:42915", - "right": "issue:43749", + "left": "issue:44291", + "right": "issue:44821", "accept": false, - "reason": "Qwen3Moe FP8 config failure and FSDP CPU RAM efficient loading are distinct loading/runtime issues." + "reason": "Different subsystems: init_empty_weights loading error versus loading an image processor from URL." }, { - "left": "issue:24643", - "right": "issue:29159", + "left": "issue:34567", + "right": "issue:36331", "accept": false, - "reason": "DeepSpeed weight shape error and tokenizer slow/fast inconsistency are unrelated." + "reason": "Both touch trainer internals, but one is token-count tracking and the other is a breaking compute_loss signature change." }, { - "left": "issue:42175", - "right": "issue:44488", + "left": "issue:44737", + "right": "issue:46463", "accept": false, - "reason": "Packaging/backend dependency issue versus a specific model load failure; not the same bug." + "reason": "Completely different code paths: XLNet positional encoding on CPU versus a T5 error-message bug." }, { - "left": "issue:38175", - "right": "issue:44898", + "left": "issue:43257", + "right": "issue:43756", "accept": false, - "reason": "Different model families and different failure modes, despite both being vision-related." + "reason": "Different model-specific bugs: MoE weight conversion versus RoPE layer dropping." }, { - "left": "issue:43493", - "right": "issue:43994", - "accept": true, - "reason": "Both describe the same SigLIP2 inference discrepancy / wrong outputs for the same model family, with 43994 looking like a user-facing reproduction of 43493." - }, - { - "left": "issue:44987", - "right": "issue:45907", + "left": "issue:43577", + "right": "issue:45072", "accept": false, - "reason": "A model load regression and a repo-file revision forwarding bug are unrelated." + "reason": "Both are dtype-related, but they affect different models and likely different forward paths." }, { - "left": "issue:44568", - "right": "issue:45488", + "left": "issue:44792", + "right": "issue:44964", "accept": false, - "reason": "Different tokenizer regressions: missing BOS/EOS insertion versus Metaspace/ByteLevel override in LlamaTokenizer." + "reason": "Different multimodal model failures with no clear shared code path." }, { - "left": "issue:27957", - "right": "issue:28282", + "left": "issue:44062", + "right": "issue:45081", "accept": false, - "reason": "Flash Attention 2 behavior for XLM-Roberta is unrelated to a missing PyTorch import error." + "reason": "Tokenizer construction crash versus Mistral regex patching crash; both are tokenizer-adjacent but distinct." }, { - "left": "issue:44291", - "right": "issue:44821", + "left": "issue:43742", + "right": "issue:44589", "accept": false, - "reason": "Init-empty-weights argument handling and AutoImageProcessor-from-URL loading are different problems." + "reason": "Different loading failures: a key error for one model versus a Float8 storage type lookup error." }, { - "left": "issue:34567", - "right": "issue:36331", + "left": "issue:45042", + "right": "issue:45081", "accept": false, - "reason": "Trainer token-count bookkeeping and a custom compute_loss signature change are not the same underlying issue." + "reason": "Image-processor dependency issue versus tokenizer regex patch crash." }, { "left": "issue:44737", - "right": "issue:46463", + "right": "issue:45588", "accept": false, - "reason": "XLNet CPU device placement bug versus T5 decoder-input validation messaging; unrelated." + "reason": "Different runtime failures in unrelated modules." }, { - "left": "issue:43257", - "right": "issue:43756", + "left": "issue:44568", + "right": "issue:46491", "accept": false, - "reason": "Qwen3 MOE weight conversion and Smollm3 RoPE-layer handling are separate model-specific bugs." + "reason": "Both involve tokenizers, but one is special-token insertion and the other is encode/decode whitespace regression." }, { - "left": "issue:43577", - "right": "issue:45072", + "left": "issue:43906", + "right": "issue:46468", "accept": false, - "reason": "Different dtype-mismatch issues in different model stacks; not one shared fix." + "reason": "Too little evidence they are the same bug; one is an isolated repro and the other is a broken example." }, { - "left": "issue:44792", - "right": "issue:44964", + "left": "issue:43756", + "right": "issue:46399", "accept": false, - "reason": "Janus test failure and Phi-4 multimodal load failure are separate issues." + "reason": "Different model families and different failure modes." }, { - "left": "issue:44062", - "right": "issue:45081", + "left": "issue:43756", + "right": "issue:45325", "accept": false, - "reason": "AddedToken duplicate keyword handling is unrelated to the Mistral tokenizer backend-tokenizer crash." + "reason": "Both are Qwen-family, but one is a RoPE-layer configuration issue and the other is a position-id scaling bug." }, { - "left": "issue:45042", - "right": "issue:45081", + "left": "issue:43526", + "right": "issue:45479", "accept": false, - "reason": "Image-processor torchvision dependency regression and tokenizer attribute crash are unrelated." + "reason": "Image-label reduction bug versus a sequence-classification loss edge case." }, { - "left": "issue:43742", - "right": "issue:44589", + "left": "issue:43404", + "right": "issue:45987", "accept": false, - "reason": "MobileLLM key error and Float8 storage lookup error are different load-path failures." + "reason": "Different core issues: weight tying in Mistral3 versus cache sequence-length return type." }, { - "left": "issue:44568", - "right": "issue:46491", + "left": "issue:45823", + "right": "issue:46463", "accept": false, - "reason": "Different tokenizer regressions: special-token insertion versus encode/decode whitespace preservation." + "reason": "Device placement bug versus T5 validation-message bug." }, { - "left": "issue:43906", - "right": "issue:46468", + "left": "issue:43531", + "right": "issue:44821", "accept": false, - "reason": "One is an isolated reproduction of an earlier issue; the other is a broken official example, but the titles do not indicate the same bug." + "reason": "Qwen3-MoE sliding-window behavior and image-processor loading are unrelated." }, { - "left": "issue:43756", - "right": "issue:46399", + "left": "issue:45405", + "right": "issue:45600", "accept": false, - "reason": "Smollm3 RoPE-layer dropping and Qwen3.5 rotary registration crash are different code paths." + "reason": "One is a dependency version pinning problem; the other is stale auto-mapping references." }, { - "left": "issue:44737", - "right": "issue:45588", + "left": "issue:43452", + "right": "issue:44488", "accept": false, - "reason": "XLNet CPU arange placement and flash_attention s_aux handling are unrelated." + "reason": "Both involve model loading, but the reported failures and likely root causes are different." }, { - "left": "issue:43756", - "right": "issue:45325", + "left": "issue:43906", + "right": "issue:45381", "accept": false, - "reason": "Smollm3 RoPE-layer count issue and Qwen2.5-VL temporal position-ID scaling are distinct bugs." + "reason": "Both mention Qwen2.5-VL-like behavior only indirectly; the described bugs are not the same code-path problem." }, { - "left": "issue:43526", - "right": "issue:45479", + "left": "issue:45357", + "right": "issue:45920", "accept": false, - "reason": "BEiTImageProcessorFast label reduction and sequence-classification zero-loss behavior are different issues." + "reason": "Different areas: save_pretrained visual-encoder key serialization versus AutoTokenizer token-id regression." }, { - "left": "issue:45823", - "right": "issue:46463", + "left": "issue:46399", + "right": "issue:45920", "accept": false, - "reason": "Gemma4 device-mapping mismatch and T5 decoder-input validation messaging are unrelated." + "reason": "Completely different regressions: kernel registration failure versus tokenizer token-id corruption." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The two accepted soft edges are defensible from the titles alone: 44479/45381 are clearly in the same Qwen video-input regression family, and 43493/43994 are plausibly the same SigLIP2 issue. No overstatement is evident, and the many rejections are appropriately cautious." + "feedback": "Grounded overall. The canonical anchor at issue:45920 is supported by its broad multi-model tokenizer regression scope, and the soft-edge verdicts stay mostly conservative. The SigLIP2 pair is the only accepted duplicate candidate, and that linkage is reasonable from the titles without overstating evidence. " }, "evaluator_used": true, "retried": false, @@ -6835,26 +6877,26 @@ "issue:35141", "issue:38175", "issue:41762", + "issue:42898", "issue:43295", - "issue:43404", "issue:43452", "issue:43531", + "issue:43577", "issue:43643", "issue:43653", "issue:43688", "issue:43697", "issue:43720", + "issue:43742", "issue:43749", "issue:43819", - "issue:43906", "issue:44361", "issue:44442", "issue:44466", - "issue:44488", "issue:44493", + "issue:44514", "issue:44625", "issue:44811", - "issue:44821", "issue:44843", "issue:44877", "issue:44898", @@ -6863,32 +6905,25 @@ "issue:45005", "issue:45200", "issue:45216", - "issue:45357", + "issue:45356", "issue:45372", "issue:45375", "issue:45381", - "issue:45405", - "issue:45600", + "issue:45542", "issue:45823", - "issue:45920", - "issue:45987", "issue:46050", "issue:46121", "issue:46170", - "issue:46463" + "issue:46178", + "issue:46463", + "issue:46491" ], "soft_pairs": [ - "issue:43404|issue:45987", - "issue:43531|issue:44821", - "issue:45405|issue:45600", - "issue:43452|issue:44488", - "issue:43906|issue:45381", - "issue:45357|issue:45920", "issue:44625|issue:44877", "issue:44913|issue:46121", "issue:43720|issue:44898", - "issue:43697|issue:45216", "issue:43643|issue:45375", + "issue:43697|issue:45216", "issue:24643|issue:29127", "issue:28282|issue:32090", "issue:41762|issue:44936", @@ -6898,185 +6933,203 @@ "issue:44466|issue:45005", "issue:43819|issue:44811", "issue:43452|issue:44843", - "issue:43688|issue:43749", "issue:44361|issue:44442", + "issue:43688|issue:43749", "issue:38175|issue:43295", "issue:44936|issue:45823", "issue:43653|issue:44361", - "issue:45372|issue:46170" + "issue:45372|issue:46170", + "issue:45542|issue:46178", + "issue:43452|issue:43577", + "issue:42898|issue:46491", + "issue:44514|issue:45381", + "issue:43531|issue:43742", + "issue:43295|issue:45356" ], - "prepared_review_unit_hash": "eb2161c06308db4cd182c95abc4aa11090f607d03aa07102daaf22dfc858b72a", + "prepared_review_unit_hash": "e9d91b57b93055788946e68e20acc8e35079039326d70e659168744ebb0e909c", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 26, - "serialized_chars": 35860, - "estimated_input_tokens": 8965, - "estimated_eval_tokens": 18186 + "serialized_chars": 35351, + "estimated_input_tokens": 8838, + "estimated_eval_tokens": 17932 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: most items are unrelated regression reports across different models, configs, and loading paths. I do not see any soft pair that is clearly the same underlying bug/change, so no duplicate merge is warranted.", - "confidence": 0.92, - "canonical_issue_reason": null, + "summary": "The proposed soft edges are mostly superficial theme overlaps (configs, tokenizers, tied weights, loading paths) rather than the same concrete bug. I would reject all of them; this set is too heterogeneous to form a single duplicate cluster.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue fits: the items span unrelated subsystems and failure modes, so there isn\u2019t one underlying bug to anchor the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44466 is the most concrete and actionable report in the loose tied-weights subtheme, with a specific regression and a clear serialization failure mode; however, the cluster as a whole is not coherent enough to nominate a true canonical issue.", + "best_issue_reason": "No global best issue: the cluster is heterogeneous, so any one issue would be a misleading representative of the others.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43404", - "right": "issue:45987", - "accept": false, - "reason": "Different bugs: tied lm_head weights in Mistral vs StaticCache seq-length return type." - }, - { - "left": "issue:43531", - "right": "issue:44821", - "accept": false, - "reason": "Unrelated: Qwen3-MoE sliding_window behavior vs AutoImageProcessor URL loading." - }, - { - "left": "issue:45405", - "right": "issue:45600", - "accept": false, - "reason": "Dependency version bump issue vs removed auto_mapping config references; not the same bug." - }, - { - "left": "issue:43452", - "right": "issue:44488", - "accept": false, - "reason": "Both are load failures, but one is gguf_file handling and the other is a model-specific load regression." - }, - { - "left": "issue:43906", - "right": "issue:45381", - "accept": false, - "reason": "One is a reproduction of an older issue; the other is a Qwen2.5-VL video position_ids bug." - }, - { - "left": "issue:45357", - "right": "issue:45920", - "accept": false, - "reason": "Different layers: save_pretrained key serialization vs AutoTokenizer producing wrong token IDs." - }, { "left": "issue:44625", "right": "issue:44877", "accept": false, - "reason": "Config propagation bug vs strict config rejection when loading granite_speech." + "reason": "Both are config-related, but one is Qwen3.5 num_labels propagation and the other is strict loading for granite_speech; different models and code paths." }, { "left": "issue:44913", "right": "issue:46121", "accept": false, - "reason": "Both involve RoPE config, but one is reload-default drift and the other is a TypeError on JSON-loaded lists." + "reason": "Both mention rope/config handling, but one is a persisted config field regression and the other is a TypeError in rope validation for JSON-loaded lists." }, { "left": "issue:43720", "right": "issue:44898", "accept": false, - "reason": "Packed-weight loading bug vs image classification interpolation issue; no shared code path." + "reason": "Different subsystems entirely: BitNet accelerate loading vs Perceiver image classification resolution handling." }, { - "left": "issue:43697", - "right": "issue:45216", + "left": "issue:43643", + "right": "issue:45375", "accept": false, - "reason": "RTDetrV2 output drift and Qwen3.5 checkpoint serialization are different regressions." + "reason": "Both concern missing config fields, but one is generic AutoConfig with trust_remote_code and the other is a Qwen3_5MoeVision strict-field issue; not the same bug." }, { - "left": "issue:43643", - "right": "issue:45375", + "left": "issue:43697", + "right": "issue:45216", "accept": false, - "reason": "Both mention missing fields, but one is trust_remote_code config loading and the other is @strict dropping a field." + "reason": "Different regressions: RTDetrV2 output changes vs Qwen3.5 checkpoint save_pretrained corruption." }, { "left": "issue:24643", "right": "issue:29127", "accept": false, - "reason": "DeepSpeed training weight-dimension error is unrelated to a LayoutLMv3 error-message clarity issue." + "reason": "DeepSpeed training runtime error vs LayoutLMv3 error-message clarity; unrelated failures." }, { "left": "issue:28282", "right": "issue:32090", "accept": false, - "reason": "Missing PyTorch import error vs trainer _gpu_broadcast_one NoneType failure." + "reason": "PyTorch import detection vs Trainer broadcast TypeError; unrelated." }, { "left": "issue:41762", "right": "issue:44936", "accept": false, - "reason": "ZeRO-3 loading IndexError is not the same as trainer.evaluate() failing after train()." + "reason": "Gemma3 loading with ZeRO-3 vs trainer.evaluate after trainer.train; different code paths." }, { "left": "issue:45200", "right": "issue:46463", "accept": false, - "reason": "Required mm_token_type_ids defaulting vs T5 decoder input validation error message; distinct bugs." + "reason": "Gemma 4 multimodal token defaults vs T5 decoder input validation messaging; not the same underlying bug." }, { "left": "issue:28282", "right": "issue:35141", "accept": false, - "reason": "Missing torch import issue vs post_init reinitializing resized embeddings; unrelated." + "reason": "ImportError for missing PyTorch vs tied-weights reinitialization on resize; unrelated." }, { "left": "issue:44493", "right": "issue:46050", "accept": false, - "reason": "Unexpected position-id key regression vs quantized save/load breakage." + "reason": "Unexpected position-id keys vs sinq quantization save/load breakage; different bug classes." }, { "left": "issue:44466", "right": "issue:45005", "accept": false, - "reason": "Related tied-weights theme, but one is device-dependent lm_head serialization and the other is a broader translation-model regression; not clearly the same concrete bug." + "reason": "Both involve tied weights, but one is device-dependent lm_head serialization and the other is translation-model tied-weight issues in v5; too broad to merge." }, { "left": "issue:43819", "right": "issue:44811", "accept": false, - "reason": "DAC.from_latents mismatch is unrelated to Whisper batch_decode skipping special tokens." + "reason": "DAC latent conversion mismatch vs Whisper batch_decode skip_special_tokens behavior; unrelated." }, { - "left": "issue:43688", - "right": "issue:43749", + "left": "issue:43452", + "right": "issue:44843", "accept": false, - "reason": "Auxiliary-loss normalization and FSDP CPU RAM efficient loading are different problems." + "reason": "Both involve AutoTokenizer loading, but one is gguf_file support breaking tokenizer/model loads and the other is an offline model_info call in mistral regex patching." }, { "left": "issue:44361", "right": "issue:44442", "accept": false, - "reason": "MLukeTokenizer task AttributeError is unrelated to FastSpeech2ConformerTokenizer load failure." + "reason": "Different tokenizer failures on different tokenizers; not the same concrete bug." + }, + { + "left": "issue:43688", + "right": "issue:43749", + "accept": false, + "reason": "Auxiliary loss normalization in MoE models vs FSDP CPU RAM efficient loading breakage." }, { "left": "issue:38175", "right": "issue:43295", "accept": false, - "reason": "Unexpected zero probabilities in SigLIP2 vs processor.tokenizer/images regression in v4.57.5." + "reason": "Unexpected zero probabilities in SigLIP2 vs processor.tokenizer/image-passing regression; unrelated." + }, + { + "left": "issue:44936", + "right": "issue:45823", + "accept": false, + "reason": "Trainer lifecycle failure vs Gemma4 PLE device mismatch during forward; different issues." }, { "left": "issue:43653", "right": "issue:44361", "accept": false, - "reason": "BigBird special-token decode bug and MLuke tokenizer task AttributeError are distinct." + "reason": "BigBird special-token decode bug vs MLukeTokenizer AttributeError; unrelated tokenizers and failure modes." }, { "left": "issue:45372", "right": "issue:46170", "accept": false, - "reason": "Gemma 4 processor import failure is unrelated to the Windows ROCm torch.distributed dependency issue." + "reason": "Gemma 4 processor import dependency vs Windows ROCm torch.distributed dependency problem; unrelated." + }, + { + "left": "issue:45542", + "right": "issue:46178", + "accept": false, + "reason": "Backend registration/availability issues, but one is tensorflow backend undefined and the other is tensorflow_text not present in BACKENDS_MAPPING; different concrete bugs." + }, + { + "left": "issue:43452", + "right": "issue:43577", + "accept": false, + "reason": "Tokenizer/model loading with gguf_file vs BLIP2 dtype propagation; unrelated." + }, + { + "left": "issue:42898", + "right": "issue:46491", + "accept": false, + "reason": "General v5 cleanup-space behavior change vs CodeLlama leading-space round-trip regression; related area but not the same underlying bug." + }, + { + "left": "issue:44514", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen2.5-VL issues, but one is batched chat-template padding crash and the other is incorrect video vision_position_ids; different code paths." + }, + { + "left": "issue:43531", + "right": "issue:43742", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior vs MobileLLM-125M loading key error; unrelated." + }, + { + "left": "issue:43295", + "right": "issue:45356", + "accept": false, + "reason": "Both are regressions in tokenizer/processor behavior, but one is processor.tokenizer/image handling and the other is Kimi-K2.5 codec handling plus a warning; not the same bug." } ] }, @@ -7105,17 +7158,17 @@ "issue:33357", "issue:33666", "issue:36010", + "issue:36331", "issue:41720", - "issue:42898", - "issue:43295", - "issue:43452", - "issue:43531", + "issue:43504", "issue:43550", "issue:43577", + "issue:43643", "issue:43646", "issue:43653", "issue:43688", "issue:43697", + "issue:43720", "issue:43742", "issue:43749", "issue:43756", @@ -7126,7 +7179,10 @@ "issue:44479", "issue:44484", "issue:44514", + "issue:44554", + "issue:44561", "issue:44568", + "issue:44610", "issue:44625", "issue:44704", "issue:44734", @@ -7135,246 +7191,250 @@ "issue:44843", "issue:44977", "issue:44991", - "issue:45356", - "issue:45381", + "issue:45290", "issue:45440", "issue:45491", - "issue:45542", "issue:45701", "issue:45704", "issue:46002", "issue:46050", - "issue:46178", "issue:46402", - "issue:46491", "issue:46612" ], "soft_pairs": [ - "issue:45542|issue:46178", - "issue:42898|issue:46491", - "issue:43452|issue:43577", - "issue:44514|issue:45381", - "issue:43295|issue:45356", - "issue:43531|issue:43742", "issue:44484|issue:44704", "issue:43577|issue:43697", - "issue:43653|issue:44568", "issue:43646|issue:43749", - "issue:44843|issue:44991", + "issue:43653|issue:44568", "issue:44464|issue:44743", + "issue:44843|issue:44991", "issue:43688|issue:45704", "issue:29159|issue:33357", "issue:43653|issue:43927", "issue:43756|issue:46402", "issue:43688|issue:45440", - "issue:43688|issue:45701", "issue:44479|issue:44625", + "issue:43688|issue:45701", "issue:43550|issue:44361", - "issue:45491|issue:46002", "issue:44734|issue:46612", + "issue:45491|issue:46002", "issue:33666|issue:36010", "issue:44464|issue:44977", "issue:32090|issue:33666", "issue:43742|issue:44821", "issue:41720|issue:44464", - "issue:44466|issue:46050" + "issue:44466|issue:46050", + "issue:43504|issue:43720", + "issue:43504|issue:43550", + "issue:43643|issue:44561", + "issue:44514|issue:45290", + "issue:43577|issue:44554", + "issue:43504|issue:44610", + "issue:43720|issue:43927", + "issue:33666|issue:36331" ], - "prepared_review_unit_hash": "b0e049a5e71a34ecb904d8ac16186ad3b57abd5d44a92d0d10667556ced0bdfb", + "prepared_review_unit_hash": "3eaf698472fa23c64a85e6a6cdda0083d8b5063ddbe338db1cfdf85440e1b027", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 28, - "serialized_chars": 36193, - "estimated_input_tokens": 9049, - "estimated_eval_tokens": 18354 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36228, + "estimated_input_tokens": 9057, + "estimated_eval_tokens": 18370 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous. A small subgroup concerns tokenizer regressions/behavior changes in v5, but most pairings are different bugs, models, or code paths and should not be deduplicated.", - "confidence": 0.37, - "canonical_issue_reason": "issue:42898 is the broadest tokenizer-regression report in the set and is the closest fit for the tokenization-behavior subgroup; the rest of the cluster is too mixed for a single true canonical.", + "summary": "This cluster is highly heterogeneous: the items span unrelated tokenizer, loading, generation, training, serialization, and model-specific regression reports. Every soft pair is only lexically similar and should not be merged as duplicates.", + "confidence": 0.88, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:42898 is the best issue to anchor the cluster because it generalizes to tokenizer behavior changes in v5, while the other issues are narrower or unrelated.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:45542", - "right": "issue:46178", + "left": "issue:44484", + "right": "issue:44704", "accept": false, - "reason": "Both are backend-related errors, but they point to different optional-backend problems (undefined tf backend vs missing BACKENDS_MAPPING entry)." + "reason": "max_shard_size/save_pretrained behavior is unrelated to AutoProcessor kwargs forwarding to cached_file." }, { - "left": "issue:42898", - "right": "issue:46491", + "left": "issue:43577", + "right": "issue:43697", "accept": false, - "reason": "Both involve tokenization regressions, but one is about cleanup-space behavior in v5 and the other is a CodeLlama leading-space round-trip bug." + "reason": "Blip2 dtype loading bug and RTDetr output divergence affect different model paths and failure modes." }, { - "left": "issue:43452", - "right": "issue:43577", - "accept": false, - "reason": "Unrelated failures: gguf loading broke in one, while the other is a Blip2 dtype propagation issue." - }, - { - "left": "issue:44514", - "right": "issue:45381", - "accept": false, - "reason": "Same model family, but different code paths: chat template batching vs video vision position ids." - }, - { - "left": "issue:43295", - "right": "issue:45356", - "accept": false, - "reason": "Different regressions affecting different models and mechanisms; not the same underlying bug." - }, - { - "left": "issue:43531", - "right": "issue:43742", - "accept": false, - "reason": "Different issues: sliding-window behavior in Qwen3-MoE versus a load-time key error for MobileLLM-125M." - }, - { - "left": "issue:44484", - "right": "issue:44704", - "accept": false, - "reason": "One is a save_pretrained sharding question, the other is missing kwargs propagation in AutoProcessor loading." - }, - { - "left": "issue:43577", - "right": "issue:43697", + "left": "issue:43646", + "right": "issue:43749", "accept": false, - "reason": "Different bug classes: dtype handling in Blip2 versus divergent outputs in RTDetrV2." + "reason": "trust_remote_code config object field loss is a different issue from FSDP CPU RAM efficient loading." }, { "left": "issue:43653", "right": "issue:44568", "accept": false, - "reason": "Both concern special tokens, but they are different tokenizers and different failure modes." + "reason": "Both involve tokenizers, but one is special-token registration and the other is BOS/EOS insertion with add_special_tokens." }, { - "left": "issue:43646", - "right": "issue:43749", + "left": "issue:44464", + "right": "issue:44743", "accept": false, - "reason": "Custom model initialization breakage is unrelated to the FSDP CPU RAM-efficient loading regression." + "reason": "Compiled chunked generation inconsistency is not the same as recurrent-state reset with cache in modular_qwen3_5.py." }, { "left": "issue:44843", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer-loading regressions, but one is an offline model_info call bug and the other is a separate tokenizer load failure." - }, - { - "left": "issue:44464", - "right": "issue:44743", - "accept": false, - "reason": "Both involve generation/cache state, but they are different concrete bugs in different code paths." + "reason": "Offline model_info patching in AutoTokenizer is a different failure than tokenizer loading for est-roberta." }, { "left": "issue:43688", "right": "issue:45704", "accept": false, - "reason": "Completely different failures: auxiliary-loss normalization versus an RMSNorm memory leak." + "reason": "Auxiliary loss normalization is unrelated to the apex FusedRMSNorm memory leak." }, { "left": "issue:29159", "right": "issue:33357", "accept": false, - "reason": "Tokenization inconsistency versus a MacOS bus error; no shared underlying defect." + "reason": "Tokenizer slow/fast inconsistency is not the same as a MacOS bus error loading a CLIP model." }, { "left": "issue:43653", "right": "issue:43927", "accept": false, - "reason": "Special-token registration bug versus config losing custom token IDs on save/load." + "reason": "Special-token registration in BigBirdTokenizer and custom token ID persistence in DiaConfig are distinct bugs." }, { "left": "issue:43756", "right": "issue:46402", "accept": false, - "reason": "RoPE-layer dropout in Smollm3 is unrelated to missing text-encoder weight loading in Sam3Model." + "reason": "Smollm3 RoPE layer mismatch and Sam3 text-encoder weight loading are unrelated model-specific problems." }, { "left": "issue:43688", "right": "issue:45440", "accept": false, - "reason": "Different model implementations and different numerical issues; not the same bug." + "reason": "Aux loss normalization does not match the DeepseekV3MoE divergence from remote implementation." }, { - "left": "issue:43688", - "right": "issue:45701", + "left": "issue:44479", + "right": "issue:44625", "accept": false, - "reason": "Auxiliary-loss normalization is unrelated to a broad tokenization-version-change report." + "reason": "Video-input regression in Qwen VL models is unrelated to num_labels propagation in Qwen3.5 config." }, { - "left": "issue:44479", - "right": "issue:44625", + "left": "issue:43688", + "right": "issue:45701", "accept": false, - "reason": "Video-input regression for VL models versus num_labels propagation in Qwen3.5 config." + "reason": "Incorrect auxiliary-loss normalization and tokenization-version differences are different bugs." }, { "left": "issue:43550", "right": "issue:44361", "accept": false, - "reason": "Torch.compile/SDPA failure in Bamba is unrelated to an AttributeError in MLukeTokenizer." + "reason": "torch.compile/SDPA failure in Bamba is not the same as MLukeTokenizer AttributeError on tasks." }, { - "left": "issue:45491", - "right": "issue:46002", + "left": "issue:44734", + "right": "issue:46612", "accept": false, - "reason": "Both mention NaNs, but the causes and affected components are different." + "reason": "Serve KV-cache continuation indexing and beam-search cache reorder omission are different generation code paths." }, { - "left": "issue:44734", - "right": "issue:46612", + "left": "issue:45491", + "right": "issue:46002", "accept": false, - "reason": "Both touch generation caches, but one is server KV-continuation indexing and the other is beam-search reorder being skipped for some models." + "reason": "NaNs from sliding-window attention with mixed-length batches are not the same as NaNs in classification heads after checkout/init." }, { "left": "issue:33666", "right": "issue:36010", "accept": false, - "reason": "Multi-GPU training issue versus an import error for GenerationMixin; different failure modes." + "reason": "Qwen2-VL multi-GPU training and the GenerationMixin import error are unrelated failures." }, { "left": "issue:44464", "right": "issue:44977", "accept": false, - "reason": "Different generation regressions: compiled-forward chunking versus flash-attention behavior in Qwen3.5." + "reason": "Compiled-forward chunked generation inconsistency is not the same as flash-attention regression in Qwen3.5." }, { "left": "issue:32090", "right": "issue:33666", "accept": false, - "reason": "Trainer broadcast error and Qwen2-VL multi-GPU training are not the same bug." + "reason": "NoneType GPU broadcast in Trainer is a different problem from Qwen2-VL multi-GPU training." }, { "left": "issue:43742", "right": "issue:44821", "accept": false, - "reason": "A model-loading key error and inability to load AutoImageProcessor from URL are unrelated." + "reason": "Model loading key error and AutoImageProcessor-from-URL failure involve different loaders." }, { "left": "issue:41720", "right": "issue:44464", "accept": false, - "reason": "Auto device mapping cuda assert on A800 is unrelated to compiled-forward chunked generation inconsistency." + "reason": "Qwen3 auto device mapping cuda assert is unrelated to compiled-forward chunked generation output differences." }, { "left": "issue:44466", "right": "issue:46050", "accept": false, - "reason": "Different persistence bugs: device-dependent lm_head serialization versus quantized-model save/load breakage." + "reason": "Inconsistent lm_head.weight serialization is not the same as sinq quantization breaking save/load." + }, + { + "left": "issue:43504", + "right": "issue:43720", + "accept": false, + "reason": "Beit legacy-field loading and BitNet packed-weight unpacking are distinct model-loading defects." + }, + { + "left": "issue:43504", + "right": "issue:43550", + "accept": false, + "reason": "A Beit preset-loading regression is not the same as Bamba torch.compile with SDPA." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "Missing config fields with trust_remote_code and removal of is_torch_fx_available are separate v5 compatibility breaks." + }, + { + "left": "issue:44514", + "right": "issue:45290", + "accept": false, + "reason": "Batched apply_chat_template padding behavior and assistant tool-call/no-content crashes are close but not the same bug." + }, + { + "left": "issue:43504", + "right": "issue:44610", + "accept": false, + "reason": "Legacy-field model loading and processor/model image-size mismatch are unrelated." + }, + { + "left": "issue:43720", + "right": "issue:43927", + "accept": false, + "reason": "Packed-weight accelerate loading and DiaConfig custom token ID persistence are different issues." + }, + { + "left": "issue:33666", + "right": "issue:36331", + "accept": false, + "reason": "Qwen2-VL multi-GPU training and CustomTrainer.compute_loss signature mismatch are unrelated trainer problems." } ] }, @@ -7401,22 +7461,23 @@ "issue:28218", "issue:33357", "issue:33666", - "issue:36331", "issue:39692", + "issue:43257", "issue:43262", "issue:43278", "issue:43295", + "issue:43299", + "issue:43334", "issue:43452", "issue:43493", - "issue:43504", "issue:43525", - "issue:43550", + "issue:43540", "issue:43572", - "issue:43577", + "issue:43575", "issue:43606", - "issue:43643", "issue:43646", "issue:43688", + "issue:43701", "issue:43720", "issue:43825", "issue:43844", @@ -7426,17 +7487,15 @@ "issue:44451", "issue:44466", "issue:44488", - "issue:44514", "issue:44554", - "issue:44561", + "issue:44568", "issue:44589", - "issue:44610", "issue:44625", + "issue:44792", "issue:44843", "issue:44849", "issue:44964", "issue:44977", - "issue:45290", "issue:45357", "issue:45405", "issue:45520", @@ -7445,27 +7504,20 @@ "issue:46050", "issue:46192", "issue:46209", - "issue:46365" + "issue:46365", + "issue:46489" ], "soft_pairs": [ - "issue:43504|issue:43550", - "issue:43504|issue:43720", - "issue:43643|issue:44561", - "issue:44514|issue:45290", - "issue:43577|issue:44554", - "issue:43504|issue:44610", - "issue:43720|issue:43927", - "issue:33666|issue:36331", "issue:43262|issue:43295", "issue:43606|issue:43927", "issue:44554|issue:46365", + "issue:44977|issue:46209", "issue:43525|issue:43572", "issue:43278|issue:43688", - "issue:44977|issue:46209", "issue:43720|issue:44361", - "issue:43452|issue:44451", "issue:44964|issue:45405", "issue:39692|issue:43493", + "issue:43452|issue:44451", "issue:43493|issue:43825", "issue:43994|issue:46192", "issue:45823|issue:46050", @@ -7479,228 +7531,236 @@ "issue:33357|issue:33666", "issue:44488|issue:44843", "issue:44849|issue:45357", - "issue:44589|issue:45520" + "issue:44589|issue:45520", + "issue:43278|issue:43701", + "issue:43540|issue:44977", + "issue:44568|issue:46489", + "issue:43257|issue:43299", + "issue:43575|issue:45405", + "issue:43927|issue:44361", + "issue:43334|issue:43572", + "issue:44792|issue:45405" ], - "prepared_review_unit_hash": "484cdf78378ad3d340d565fb6231ead3fbb327506b4e7f7c5fb740f0c7aca3f4", + "prepared_review_unit_hash": "71e258f99a75fccbe373e919d918b44ea02edaf99f725c9505d1c26c61ef5d64", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37405, - "estimated_input_tokens": 9352, - "estimated_eval_tokens": 18960 + "serialized_chars": 37286, + "estimated_input_tokens": 9322, + "estimated_eval_tokens": 18900 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The soft links are mostly false positives: they connect unrelated bug reports that merely share broad themes like loading, tokenizers, configs, or model-specific regressions. None of the candidate pairs looks like the same underlying issue.", - "confidence": 0.94, - "canonical_issue_reason": "issue:43504 is a clean, self-contained bug report with a concrete failure mode and a likely narrow fix path, making it a reasonable representative issue.", + "summary": "The cluster is dominated by unrelated transformers regressions, tokenizer/config bugs, and model-specific loading issues. None of the soft pairs look like the same underlying bug or change, so all soft edges should be rejected. The most representative issue is the broad v5 custom model initialization regression.", + "confidence": 0.39, + "canonical_issue_reason": "issue:43646 is the broadest, most central regression report in the set: it targets Transformers 5.0.0 breaking custom model initialization, which is a wider umbrella than the more specialized tokenizer/config/model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43504 is the most actionable representative here: specific scope, clear symptom, and a plausible single root cause. It is stronger as a triage anchor than the broader or more speculative reports.", + "best_issue_reason": "issue:43646 is the best general representative because it captures a high-level v5 regression affecting custom model init rather than a narrow model-specific symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43504", - "right": "issue:43550", - "accept": false, - "reason": "Both are model-related failures, but one is a Beit preset loading bug and the other is a Bamba torch.compile/SDPA issue; different code paths." - }, - { - "left": "issue:43504", - "right": "issue:43720", - "accept": false, - "reason": "Legacy-field model loading in Beit is unrelated to BitNet packed-weight unpacking during accelerate loading." - }, - { - "left": "issue:43643", - "right": "issue:44561", - "accept": false, - "reason": "trust_remote_code field loss in AutoConfig is a different compatibility bug from removing `is_torch_fx_available` for remote-code models." - }, - { - "left": "issue:44514", - "right": "issue:45290", - "accept": false, - "reason": "These are both chat-template/processor issues, but one is batched Qwen2.5-VL padding behavior and the other is tool-call messages with no content." - }, - { - "left": "issue:43577", - "right": "issue:44554", - "accept": false, - "reason": "One is a dtype propagation problem in BLIP2 loading; the other is an MPS attention correctness bug when value and query head dims differ." - }, - { - "left": "issue:43504", - "right": "issue:44610", - "accept": false, - "reason": "Beit pretrained preset loading and OmDet-Turbo image-size mismatch are unrelated vision-model bugs." - }, - { - "left": "issue:43720", - "right": "issue:43927", - "accept": false, - "reason": "BitNet weight unpacking during load is unrelated to DiaConfig losing custom token IDs on save/load." - }, - { - "left": "issue:33666", - "right": "issue:36331", - "accept": false, - "reason": "Multi-GPU training for Qwen2-VL is unrelated to a Trainer `compute_loss` signature regression." - }, { "left": "issue:43262", "right": "issue:43295", "accept": false, - "reason": "Audio processor sampling-rate defaults and processor.tokenizer/image-handling regressions are different processor API bugs." + "reason": "Audio processor chat-template sampling-rate bug and processor/tokenizer regression are different code paths and symptoms." }, { "left": "issue:43606", "right": "issue:43927", "accept": false, - "reason": "CPU offload device mismatch in Bark is unrelated to DiaConfig token-ID persistence." + "reason": "CPU offload device mismatch for bark-small is unrelated to DiaConfig losing custom token IDs on save/load." }, { "left": "issue:44554", "right": "issue:46365", "accept": false, - "reason": "MPS attention math correctness is unrelated to an import failure for Trainer caused by datasets installation." + "reason": "MPS attention correctness issue is unrelated to importing Trainer breaking because datasets is installed." + }, + { + "left": "issue:44977", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug and FP8 inference-on-B200 issue are distinct backend/hardware problems." }, { "left": "issue:43525", "right": "issue:43572", "accept": false, - "reason": "Missing `pad_token_id` on Llama4Config and missing `pad_token_idx` on StableLmConfig are separate config-field regressions." + "reason": "Llama4Config pad_token_id and StableLmConfig pad_token_idx are similar attribute-missing reports but for different configs and fixes." }, { "left": "issue:43278", "right": "issue:43688", "accept": false, - "reason": "BLIP2 dtype mismatch between train/eval and auxiliary-loss normalization in OLMoE/GPT Oss are different numerical issues." - }, - { - "left": "issue:44977", - "right": "issue:46209", - "accept": false, - "reason": "Qwen3.5 flash-attention generation and FP8 inference on B200 are different model/hardware failures." + "reason": "BF16/FP32 embedding dtype drift in evaluation is unrelated to auxiliary-loss normalization in MoE models." }, { "left": "issue:43720", "right": "issue:44361", "accept": false, - "reason": "BitNet accelerate-loading unpacking and MLukeTokenizer task AttributeError are unrelated failures." - }, - { - "left": "issue:43452", - "right": "issue:44451", - "accept": false, - "reason": "gguf_file breaks specific AutoTokenizer/AutoModel loading, while the ScandiBERT report is a separate model-loading regression." + "reason": "BitNet packed-weight unpacking during accelerate loading is a different failure from MLukeTokenizer task AttributeError." }, { "left": "issue:44964", "right": "issue:45405", "accept": false, - "reason": "Phi-4 multimodal loading and the PEFT minimum-version bump are unrelated issues." + "reason": "Phi-4 multimodal loading and PEFT minimum-version bump are unrelated; one is a model compatibility issue, the other a dependency constraint issue." }, { "left": "issue:39692", "right": "issue:43493", "accept": false, - "reason": "Both mention SigLIP2, but one is a broken documentation example and the other is a core HF-vs-JAX implementation discrepancy." + "reason": "SigLIP2 doc/example errors and SigLIP2 implementation discrepancy are both about the same model family but describe different problems and would not be merged as one fix." + }, + { + "left": "issue:43452", + "right": "issue:44451", + "accept": false, + "reason": "gguf_file breaks tokenizer/model loading is unrelated to inability to load a specific model checkpoint." }, { "left": "issue:43493", "right": "issue:43825", "accept": false, - "reason": "SigLIP2 implementation mismatch and pipeline translation-task messaging are unrelated bugs." + "reason": "SigLIP2 implementation discrepancy and pipeline translation-task error message are unrelated subsystems." }, { "left": "issue:43994", "right": "issue:46192", "accept": false, - "reason": "A SigLIP2 model/pipeline correctness issue is unrelated to `return_tensors` being ignored when `text_kwargs` is passed." + "reason": "SigLIP2 nonsensical outputs and return_tensors being ignored are different behavioral bugs with different APIs." }, { "left": "issue:45823", "right": "issue:46050", "accept": false, - "reason": "Gemma4 device-mapping mismatch is unrelated to sinq quantized model save/load breakage." + "reason": "Gemma4 PLE device mismatch and sinq quantization save/load breakage are different model/runtime issues." }, { "left": "issue:44625", "right": "issue:45584", "accept": false, - "reason": "Qwen3.5 `num_labels` propagation and Whisper empty-transcription generation are different model-specific bugs." + "reason": "Qwen3.5 num_labels propagation and Whisper empty-transcription failure are unrelated model-specific bugs." }, { "left": "issue:28218", "right": "issue:33357", "accept": false, - "reason": "Tokenizer added-space behavior and a MacOS bus error in a CLIP model are entirely different failures." + "reason": "Tokenizer extra-space behavior and MacOS bus error with a CLIP model are completely unrelated." }, { "left": "issue:44849", "right": "issue:46209", "accept": false, - "reason": "Qwen3.5 hidden-states output bug and FP8 inference failure on B200 are not the same code-path problem." + "reason": "Qwen3.5 hidden-state bug and FP8 B200 inference bug are different execution-path and hardware issues." }, { "left": "issue:39692", "right": "issue:43295", "accept": false, - "reason": "A SigLIP2 docs example error and a regression around processor.tokenizer/image passing are unrelated." + "reason": "SigLIP2 doc/example failures and custom model/tokenizer regression are different models and different bug classes." }, { "left": "issue:43278", "right": "issue:43844", "accept": false, - "reason": "BF16-vs-FP32 dtype drift in evaluation and ZeRO-3 gradient growth are separate training stability issues." + "reason": "Embedding dtype mismatch and ZeRO-3 gradient growth are unrelated training behaviors." }, { "left": "issue:43493", "right": "issue:43844", "accept": false, - "reason": "SigLIP2 implementation mismatch is unrelated to a DeepSpeed ZeRO-3 gradient increase bug." + "reason": "SigLIP2 implementation discrepancy and ZeRO-3 gradient inflation do not share the same underlying bug." }, { "left": "issue:43646", "right": "issue:44466", "accept": false, - "reason": "Custom model initialization breakage and device-dependent lm_head serialization are different v5 regressions." + "reason": "Custom model initialization regression and lm_head serialization inconsistency are different issues in different parts of the stack." }, { "left": "issue:33357", "right": "issue:33666", "accept": false, - "reason": "A MacOS bus error in CLIP and Qwen2-VL multi-GPU training are unrelated reports." + "reason": "MacOS bus error in a pretrained CLIP model and Qwen2-VL multi-GPU training are unrelated." }, { "left": "issue:44488", "right": "issue:44843", "accept": false, - "reason": "Model loading failure for a specific BERT model and an offline-mode tokenizer regex patch bug have different causes." + "reason": "Model loading failure for a specific checkpoint and unconditional hub model_info() calls in offline mode are different bugs." }, { "left": "issue:44849", "right": "issue:45357", "accept": false, - "reason": "Qwen3.5 hidden-state output and incorrect visual encoder keys on save_pretrained are separate regressions." + "reason": "Qwen3.5 hidden-state output bug and incorrect visual encoder key serialization are separate issues." }, { "left": "issue:44589", "right": "issue:45520", "accept": false, - "reason": "Float8 storage deserialization and a Python 3.13 flash_attn import KeyError are unrelated runtime compatibility issues." + "reason": "Float8 storage lookup failure and flash_attn import KeyError on Python 3.13 are different environment/runtime compatibility problems." + }, + { + "left": "issue:43278", + "right": "issue:43701", + "accept": false, + "reason": "Embedding dtype change and resume_from_checkpoint key mismatch are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44977", + "accept": false, + "reason": "Video-processing ValueError in Qwen3OmniMoe and Qwen3.5 flash-attention generation bug are different modalities and failure modes." + }, + { + "left": "issue:44568", + "right": "issue:46489", + "accept": false, + "reason": "mdeberta-v3 add_special_tokens behavior and DeepSeek-Coder tokenizer output regression are tokenizer bugs, but not the same concrete code-path problem." + }, + { + "left": "issue:43257", + "right": "issue:43299", + "accept": false, + "reason": "Qwen3 MOE accelerate+deepspeed weight conversion and Qwen3VL MOE loading breakage are different loading failures with different root causes." + }, + { + "left": "issue:43575", + "right": "issue:45405", + "accept": false, + "reason": "Tensor-parallel OOM on Qwen2-57B and PEFT version pinning are unrelated." + }, + { + "left": "issue:43927", + "right": "issue:44361", + "accept": false, + "reason": "DiaConfig token-ID persistence bug and MLukeTokenizer task AttributeError are unrelated config/tokenizer failures." + }, + { + "left": "issue:43334", + "right": "issue:43572", + "accept": false, + "reason": "Qwen3-VL pad_token_id missing and StableLmConfig pad_token_idx missing are similar at a high level, but they affect different config classes and likely require different fixes." + }, + { + "left": "issue:44792", + "right": "issue:45405", + "accept": false, + "reason": "Janus image-generation test failure and PEFT minimum-version bump are unrelated." } ] }, @@ -7724,20 +7784,20 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-469", "nodes": [ + "issue:15354", + "issue:28218", "issue:29159", "issue:29942", "issue:30333", "issue:32090", + "issue:34567", + "issue:38175", "issue:42490", "issue:43065", - "issue:43257", - "issue:43278", - "issue:43299", "issue:43334", + "issue:43493", "issue:43504", - "issue:43540", "issue:43572", - "issue:43575", "issue:43606", "issue:43653", "issue:43696", @@ -7745,22 +7805,22 @@ "issue:43824", "issue:43844", "issue:43873", - "issue:43927", "issue:43994", "issue:44112", "issue:44291", - "issue:44361", + "issue:44423", "issue:44464", "issue:44493", + "issue:44554", "issue:44568", "issue:44589", - "issue:44792", + "issue:44734", "issue:44964", - "issue:44977", "issue:45084", "issue:45200", "issue:45229", - "issue:45405", + "issue:45406", + "issue:45446", "issue:45520", "issue:45584", "issue:45588", @@ -7768,30 +7828,22 @@ "issue:45812", "issue:45823", "issue:46050", + "issue:46121", "issue:46285", "issue:46369", - "issue:46424", - "issue:46489" + "issue:46424" ], "soft_pairs": [ - "issue:43278|issue:43701", - "issue:44568|issue:46489", - "issue:43540|issue:44977", - "issue:43257|issue:43299", - "issue:43927|issue:44361", - "issue:43575|issue:45405", - "issue:43334|issue:43572", - "issue:44568|issue:45812", - "issue:44792|issue:45405", "issue:43504|issue:43606", + "issue:44568|issue:45812", "issue:43334|issue:45759", "issue:43824|issue:44589", "issue:43606|issue:44112", "issue:43572|issue:44291", "issue:44464|issue:46424", "issue:45200|issue:45229", - "issue:44493|issue:46285", "issue:43504|issue:43653", + "issue:44493|issue:46285", "issue:44464|issue:45584", "issue:45229|issue:45823", "issue:43994|issue:44493", @@ -7802,210 +7854,204 @@ "issue:29159|issue:32090", "issue:42490|issue:43696", "issue:43065|issue:46369", - "issue:45084|issue:45520" + "issue:45084|issue:45520", + "issue:44423|issue:44734", + "issue:44554|issue:45446", + "issue:28218|issue:34567", + "issue:15354|issue:38175", + "issue:45406|issue:46121", + "issue:43493|issue:43701" ], - "prepared_review_unit_hash": "2cf30f97c2edb93d6cb130915ee2e348ec13d7d40912962a7fe9d865870600ed", + "prepared_review_unit_hash": "4f0a869316be7f304e589f0ca809cffe1b2497d649025f73cd93becb06b62ab0", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36362, - "estimated_input_tokens": 9091, - "estimated_eval_tokens": 18438 + "soft_pair_count": 27, + "serialized_chars": 35991, + "estimated_input_tokens": 8998, + "estimated_eval_tokens": 18252 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "None of the proposed soft pairs look like true duplicates; they mostly share only a broad subsystem or model family, not the same concrete bug. I would keep them separate.", - "confidence": 0.94, - "canonical_issue_reason": "No single canonical issue stands out because the items span many unrelated regressions (tokenizers, model loading, quantization, CI, memory, and config issues).", + "summary": "These are mostly unrelated issue reports. None of the soft pairs look like the same underlying bug/change, though a few are in adjacent tokenizer, multimodal, or quantization areas.", + "confidence": 0.92, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "No global best issue for deduplication purposes; the set is too heterogeneous to use one issue as a representative canonical cluster.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43278", - "right": "issue:43701", - "accept": false, - "reason": "Different failures: dtype drift during train/eval vs a resume_from_checkpoint key mismatch." - }, - { - "left": "issue:44568", - "right": "issue:46489", - "accept": false, - "reason": "Both are tokenizer regressions, but for different models and likely different code paths; not enough to treat as the same bug." - }, - { - "left": "issue:43540", - "right": "issue:44977", - "accept": false, - "reason": "Qwen3OmniMoe video-input processing and Qwen3.5 flash-attention generation are unrelated failures." - }, - { - "left": "issue:43257", - "right": "issue:43299", - "accept": false, - "reason": "Both involve Qwen3 MOE loading, but one is weight conversion with accelerate/deepspeed and the other is a dev-version load regression; distinct enough to keep separate." - }, - { - "left": "issue:43927", - "right": "issue:44361", - "accept": false, - "reason": "Different models and symptoms: saved DiaConfig token IDs vs MLukeTokenizer AttributeError." - }, - { - "left": "issue:43575", - "right": "issue:45405", - "accept": false, - "reason": "OOM on a large Qwen model is unrelated to an unreleased PEFT version bump." - }, - { - "left": "issue:43334", - "right": "issue:43572", + "left": "issue:43504", + "right": "issue:43606", "accept": false, - "reason": "Both are missing-token-field regressions, but they hit different configs and attribute names; not the same underlying bug." + "reason": "Different bugs: a legacy-field model loading failure versus a CPU offload device-mismatch on a specific model." }, { "left": "issue:44568", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but one is mdeberta special-token handling and the other is Granite token IDs; too model-specific to merge." - }, - { - "left": "issue:44792", - "right": "issue:45405", - "accept": false, - "reason": "Janus image-generation test failure is unrelated to the PEFT version release issue." - }, - { - "left": "issue:43504", - "right": "issue:43606", - "accept": false, - "reason": "Legacy field loading bug in BEiT is unrelated to a CPU-offload device mismatch in bark-small." + "reason": "Both are tokenizer regressions, but one is mDeBERTa special-token behavior and the other is Granite token-ID handling; not the same code path." }, { "left": "issue:43334", "right": "issue:45759", "accept": false, - "reason": "Both concern Qwen multimodal configs, but one is a missing pad_token_id and the other is text_config unwrapping in from_config." + "reason": "Related Qwen multimodal area, but one is a missing pad_token_id on load and the other is AutoModelForCausalLM config unwrapping; different fixes." }, { "left": "issue:43824", "right": "issue:44589", "accept": false, - "reason": "ImportError for a Qwen2.5 VL class and a Float8 storage lookup failure are unrelated." + "reason": "Completely different failures: an import-name error for Qwen2.5-VL versus a missing Float8 storage type." }, { "left": "issue:43606", "right": "issue:44112", "accept": false, - "reason": "Different models and different failure modes: CPU offload device mismatch vs a stale GraniteSpeech CI test." + "reason": "One is a model/offload runtime mismatch; the other is a stale CI test in GraniteSpeech." }, { "left": "issue:43572", "right": "issue:44291", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig is not the same as an init_empty_weights unexpected-argument regression." + "reason": "Both involve model loading regressions after v5.0, but the symptoms and code paths differ: StableLm config field vs init_empty_weights argument handling." }, { "left": "issue:44464", "right": "issue:46424", "accept": false, - "reason": "Compiled-forward chunked generation inconsistency and cache max length behavior are separate generation/cache bugs." + "reason": "Both mention caching/generation, but one is compiled-forward chunked generation mismatch and the other is max_cache_length handling for quantized cache only." }, { "left": "issue:45200", "right": "issue:45229", "accept": false, - "reason": "Same model family, but one is a missing default token-type tensor and the other is multi-GPU CUDA OOM." + "reason": "Different Gemma4 problems: missing default token-type IDs for text-only fine-tuning versus multi-GPU CUDA OOM." }, { - "left": "issue:44493", - "right": "issue:46285", + "left": "issue:43504", + "right": "issue:43653", "accept": false, - "reason": "Both are structural regressions, but they affect different models and different APIs." + "reason": "One is a pretrained-model load bug; the other is a tokenizer special-token registration bug with empty decode output." }, { - "left": "issue:43824", - "right": "issue:44589", + "left": "issue:44493", + "right": "issue:46285", "accept": false, - "reason": "Importing a model class and missing Float8 storage support are unrelated issues." + "reason": "Different regressions: a position-id key mismatch across many models versus a CLIP text-model structure inconsistency." }, { "left": "issue:44464", "right": "issue:45584", "accept": false, - "reason": "Chunked generation with compiled forward and Whisper empty-transcription handling are different bugs." + "reason": "Different generation issues: compiled-forward inconsistency versus Whisper empty-transcription failure after align_special_tokens." }, { "left": "issue:45229", "right": "issue:45823", "accept": false, - "reason": "Both are Gemma4 inference issues, but OOM and device mismatch are separate root causes." + "reason": "One reports OOM, the other a device mismatch during forward; not the same underlying bug." }, { "left": "issue:43994", "right": "issue:44493", "accept": false, - "reason": "Siglip2 nonsensical outputs and unexpected position-id keys are not the same defect." + "reason": "SigLIP2 wrong outputs and a position-id state-dict/key issue are not the same concrete failure." }, { "left": "issue:43873", "right": "issue:46050", "accept": false, - "reason": "Quantization/offloading behavior and save/load corruption after sinq quantization are related only broadly, not the same code-path bug." + "reason": "Both touch quantization/offloading, but one is offloading behavior and the other is broken save/load for sinq-quantized models." }, { "left": "issue:43844", "right": "issue:44964", "accept": false, - "reason": "ZeRO-3 gradient growth and Phi-4 multimodal loading failure are unrelated." + "reason": "Different problems: ZeRO-3 gradient growth versus a multimodal model loading failure." }, { "left": "issue:45588", "right": "issue:46050", "accept": false, - "reason": "Flash-attention integration crash with sink-less models is unrelated to sinq save/load breakage." + "reason": "Different subsystems: flash_attention.py crashing on s_aux=None versus quantized model persistence after save_pretrained/from_pretrained." }, { "left": "issue:29942", "right": "issue:30333", "accept": false, - "reason": "Flash Attention 2 test failures and MLflow reporting after training failure are completely separate." + "reason": "A Flash Attention 2 test failure and an MLFlow reporting issue are unrelated." }, { "left": "issue:29159", "right": "issue:32090", "accept": false, - "reason": "Tokenizer inconsistency and a GPU broadcast NoneType error are unrelated." + "reason": "Tokenizer slow/fast inconsistency is unrelated to a Trainer GPU broadcast TypeError." }, { "left": "issue:42490", "right": "issue:43696", "accept": false, - "reason": "Local-path import warning and GPT-oss OOM are unrelated." + "reason": "A local-path import warning is unrelated to a GPT-oss-20b CUDA out-of-memory error." }, { "left": "issue:43065", "right": "issue:46369", "accept": false, - "reason": "A dummy Conv2d implementation and memory overflow in sam3-litetext are different problems." + "reason": "Same broad SAM3 area, but one is a dummy Conv2d implementation detail and the other is a memory overflow report; not the same bug." }, { "left": "issue:45084", "right": "issue:45520", "accept": false, - "reason": "Template-node compilation error and missing flash_attn import key on Python 3.13 are unrelated." + "reason": "Different failures: template-node compilation error versus Python 3.13 flash_attn import KeyError." + }, + { + "left": "issue:44423", + "right": "issue:44734", + "accept": false, + "reason": "Both are serve crashes, but one is a multimodal .to() failure and the other is a KV-cache continuation indexing bug." + }, + { + "left": "issue:44554", + "right": "issue:45446", + "accept": false, + "reason": "Both mention attention/flex_attention, but one is an MPS correctness issue and the other is a PyTorch version-check import bug." + }, + { + "left": "issue:28218", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer spacing behavior and TrainerState token counting are unrelated." + }, + { + "left": "issue:15354", + "right": "issue:38175", + "accept": false, + "reason": "Different model issues: TorchScript export failure for ViT versus zero probabilities on SigLIP2." + }, + { + "left": "issue:45406", + "right": "issue:46121", + "accept": false, + "reason": "A serve-time processor attribute error and a rope-parameter JSON list TypeError are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:43701", + "accept": false, + "reason": "SigLIP2 implementation mismatch and resume_from_checkpoint key mismatch are different bugs." } ] }, @@ -8029,35 +8075,35 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-469", "nodes": [ - "issue:15354", - "issue:28218", - "issue:34567", "issue:37428", "issue:38175", "issue:39692", "issue:42490", "issue:43232", + "issue:43278", "issue:43295", + "issue:43334", "issue:43450", + "issue:43452", "issue:43493", + "issue:43525", "issue:43540", "issue:43550", "issue:43572", "issue:43575", "issue:43653", - "issue:43701", "issue:43824", "issue:43844", + "issue:43856", "issue:43873", + "issue:43950", "issue:43957", "issue:44112", "issue:44186", - "issue:44423", "issue:44466", "issue:44521", "issue:44554", "issue:44610", - "issue:44734", "issue:44849", "issue:44898", "issue:44987", @@ -8068,27 +8114,20 @@ "issue:45290", "issue:45356", "issue:45381", - "issue:45406", "issue:45412", - "issue:45446", "issue:45520", "issue:45823", - "issue:46121", "issue:46178", "issue:46192", - "issue:46209" + "issue:46209", + "issue:46285", + "issue:46512" ], "soft_pairs": [ - "issue:44423|issue:44734", - "issue:44554|issue:45446", - "issue:28218|issue:34567", - "issue:15354|issue:38175", - "issue:45406|issue:46121", - "issue:43493|issue:43701", "issue:45072|issue:45198", + "issue:43295|issue:43450", "issue:43540|issue:45200", "issue:37428|issue:43957", - "issue:43295|issue:43450", "issue:44186|issue:46192", "issue:45200|issue:45823", "issue:45823|issue:46192", @@ -8109,232 +8148,221 @@ "issue:44466|issue:45127", "issue:38175|issue:43493", "issue:43653|issue:44112", - "issue:44898|issue:45412" + "issue:44898|issue:45412", + "issue:43950|issue:46285", + "issue:43334|issue:43525", + "issue:43278|issue:43856", + "issue:43452|issue:46512" ], - "prepared_review_unit_hash": "54abbd07c31c6d50bb8b290b740b278fbee5c211379ed92324182fdb80af267b", + "prepared_review_unit_hash": "3116d6b5b501261d4a402f31dd3de27dc37c47fecba7ee577d946b1bb9665489", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 31, - "serialized_chars": 37295, - "estimated_input_tokens": 9324, - "estimated_eval_tokens": 18904 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 36157, + "estimated_input_tokens": 9040, + "estimated_eval_tokens": 18336 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly heterogeneous issue cluster with one likely duplicate pair in the SigLIP2 reports. The rest of the items look like separate tokenizer, import, loading, generation, processor, or CI bugs based on titles alone.", - "confidence": 0.5, - "canonical_issue_reason": "issue:43493 is the best representative of the SigLIP2 correctness subgroup and is the broader title-level match for issue:38175.", + "summary": "This cluster is mostly heterogeneous issue reports with only loose theme overlap around model loading, tokenization, and v5 regressions. I found no soft-edge pair that is clearly the same underlying bug/change, so all candidate edges are rejected.", + "confidence": 0.69, + "canonical_issue_reason": "Issue 43950 is the strongest representative of the cluster: it is a detailed, core loading regression with clear reproduction and broad impact on `from_pretrained()`, whereas the other issues are narrower and mostly unrelated one-off failures.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43493 is the strongest overall issue to keep as the representative SigLIP2 report; it best covers the only clear duplicate-looking pair in this packet.", + "best_issue_reason": "Issue 43950 is the best single issue to anchor the cluster because it targets a central library path, has a well-defined failure mode, and is more broadly reusable as a representative bug than the model-specific edge cases here.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44423", - "right": "issue:44734", - "accept": false, - "reason": "Both are serve crashes, but the titles point to different problems: continuous batching multimodal failure vs KV-cache continuation indexing." - }, - { - "left": "issue:44554", - "right": "issue:45446", - "accept": false, - "reason": "MPS attention correctness and a flex_attention import/version check are different bugs." - }, - { - "left": "issue:28218", - "right": "issue:34567", - "accept": false, - "reason": "Tokenizer spacing after an added token is unrelated to TrainerState token accounting." - }, - { - "left": "issue:15354", - "right": "issue:38175", - "accept": false, - "reason": "TorchScript GeneratorExp export support and SigLIP2 zero probabilities do not describe the same bug." - }, - { - "left": "issue:45406", - "right": "issue:46121", - "accept": false, - "reason": "A Gemma4Processor attribute error and a rope-validation list TypeError are different failure modes." - }, - { - "left": "issue:43493", - "right": "issue:43701", + "left": "issue:45072", + "right": "issue:45198", "accept": false, - "reason": "A SigLIP2 implementation discrepancy and a resume_from_checkpoint key mismatch are unrelated by title." + "reason": "Different failures: SDPA/bfloat16 dtype mismatch in SwitchTransformers/TimmWrapperModel vs Wav2Vec2 save/tokenization failure." }, { - "left": "issue:45072", - "right": "issue:45198", + "left": "issue:43295", + "right": "issue:43450", "accept": false, - "reason": "These are different model/runtime failures: CI dtype mismatch versus Wav2Vec2 save/tokenization problems." + "reason": "Both involve processors, but one is a regression in `processor.tokenizer` / image passing and the other is batched video-shape output; not the same bug." }, { "left": "issue:43540", "right": "issue:45200", "accept": false, - "reason": "A video-input ValueError in Qwen3OmniMoe and a Gemma4 token-type default issue are not the same bug." + "reason": "Qwen3OmniMoe video-processing ValueError is unrelated to Gemma 4 text-only fine-tuning needing `mm_token_type_ids`." }, { "left": "issue:37428", "right": "issue:43957", "accept": false, - "reason": "Missing a flash-attention helper import and meta-device loading breakage are distinct issues." - }, - { - "left": "issue:43295", - "right": "issue:43450", - "accept": false, - "reason": "Processor.tokenizer regression and batched video shape errors describe different processor problems." + "reason": "ImportError for a missing flash-attn helper is not the same as meta-device loading breaking some models." }, { "left": "issue:44186", "right": "issue:46192", "accept": false, - "reason": "LayoutLMv2 tokenizer crashes on NER/padding are unrelated to return_tensors being ignored with text_kwargs." + "reason": "LayoutLMv2 tokenizer crashes on NER/batched padding, while the other issue is about `return_tensors` being ignored when `text_kwargs` is passed." }, { "left": "issue:45200", "right": "issue:45823", "accept": false, - "reason": "Gemma4 token-type defaults and a device_map forward mismatch are different bugs." + "reason": "Different code paths: Gemma 4 token-type defaults vs Gemma 4 PLE device placement with `device_map=\"auto\"`." }, { "left": "issue:45823", "right": "issue:46192", "accept": false, - "reason": "Device placement mismatch and return_tensors handling are unrelated." + "reason": "Device-mapping forward mismatch is unrelated to `return_tensors` argument handling." }, { "left": "issue:43295", "right": "issue:44610", "accept": false, - "reason": "Processor API regression and OmDet-Turbo image-size mismatch are different issues." + "reason": "Processor/tokenizer regression is not the same as OmDet-Turbo\u2019s image-size mismatch." }, { "left": "issue:43295", "right": "issue:43873", "accept": false, - "reason": "Processor regression and offloading/quantization behavior are not the same underlying problem." + "reason": "Custom processor/tokenizer regression and quantization offloading behavior are different bugs." }, { "left": "issue:43232", "right": "issue:44849", "accept": false, - "reason": "Generation kwargs after sync_gpus and Qwen3.5 hidden-state output issues are separate generation bugs." + "reason": "Generation kwargs update after `sync_gpus` is unrelated to `output_hidden_states=True` in Qwen3.5." }, { "left": "issue:43450", "right": "issue:43575", "accept": false, - "reason": "Batched video shape handling and tensor-parallel OOM during load are unrelated." + "reason": "Batched video shape bug and tensor-parallel OOM on Qwen2-57B are distinct issues." }, { "left": "issue:43572", "right": "issue:44987", "accept": false, - "reason": "A missing pad_token_idx and a load failure for a specific model/package are different problems." + "reason": "Missing `pad_token_idx` in StableLmConfig is not the same as loading `physical-intelligence/fast` failing on transformers>=5.1.0." }, { "left": "issue:45520", "right": "issue:46178", "accept": false, - "reason": "Both are import/backend errors, but they mention different backends and different failure mechanisms." + "reason": "Different import-registry problems: `flash_attn` on Python 3.13 vs missing `tensorflow_text` backend mapping." }, { "left": "issue:43550", "right": "issue:43844", "accept": false, - "reason": "torch.compile SDPA failure and ZeRO-3 gradient growth are different training/runtime bugs." + "reason": "One is a torch.compile + SDPA model failure; the other is abnormal gradient growth under ZeRO-3 during training." }, { "left": "issue:44521", "right": "issue:45290", "accept": false, - "reason": "Assistant-mask generation and tool-call tokenization crashes are both chat-template related but not the same bug." + "reason": "Both touch chat templates, but one is zero assistant masks for multimodal inputs and the other is a crash on tool-call assistant messages with no content." }, { "left": "issue:39692", "right": "issue:46192", "accept": false, - "reason": "SigLIP2 documentation example errors and return_tensors handling are unrelated." + "reason": "SigLIP2 doc-example errors are unrelated to `return_tensors` being ignored with explicit `text_kwargs`." }, { "left": "issue:43572", "right": "issue:43957", "accept": false, - "reason": "Missing pad_token_idx and meta-device loading breakage are distinct model-loading issues." + "reason": "A missing pad-token config field and meta-device loading failures are different regression classes." }, { "left": "issue:44554", "right": "issue:45356", "accept": false, - "reason": "MPS attention correctness and a tokenizer regression for Kimi-K2.5 are unrelated." + "reason": "MPS attention correctness with mismatched head dims is unrelated to the Kimi-K2.5 tokenizer/codec regression." }, { "left": "issue:37428", "right": "issue:43824", "accept": false, - "reason": "These are both import errors, but they concern different missing symbols/modules." + "reason": "Both are import errors, but they concern different missing symbols and different models." }, { "left": "issue:45381", "right": "issue:46209", "accept": false, - "reason": "Wrong Qwen2.5-VL video position ids and FP8 inference failure on B200 are unrelated." + "reason": "Wrong `vision_position_ids` for Qwen2.5-VL video input is unrelated to FP8 inference failure on B200." }, { "left": "issue:42490", "right": "issue:43493", "accept": false, - "reason": "A trailing-slash local-path import warning and a SigLIP2 implementation discrepancy are different problems." + "reason": "Trailing-slash local-path warning and SigLIP2 JAX/HF discrepancy are not the same underlying bug." }, { "left": "issue:44466", "right": "issue:45127", "accept": false, - "reason": "Serialization of tied lm_head weights and LoRA merge collapse with extended vocab are only loosely related." + "reason": "Serialization of tied weights by device is unrelated to LoRA merge collapse with extended vocabulary." }, { "left": "issue:38175", "right": "issue:43493", - "accept": true, - "reason": "Both are SigLIP2 correctness reports at the title level, and 38175 fits as a narrower symptom under the broader discrepancy described by 43493." + "accept": false, + "reason": "Both involve SigLIP2, but one reports zero probabilities and the other a broader HF-vs-JAX implementation discrepancy; not enough to treat as the same bug." }, { "left": "issue:43653", "right": "issue:44112", "accept": false, - "reason": "BigBirdTokenizer special-token decode issues and a GraniteSpeech CI stale-device test are unrelated." + "reason": "BigBirdTokenizer special-token registration is unrelated to a GraniteSpeech CI stale test failure." }, { "left": "issue:44898", "right": "issue:45412", "accept": false, - "reason": "Perceiver resolution handling and RT-DETR memory release are different issues entirely." + "reason": "Perceiver image-size interpolation failure and RT-DETR memory-release issues are unrelated." + }, + { + "left": "issue:43950", + "right": "issue:46285", + "accept": false, + "reason": "Non-persistent buffer corruption in `from_pretrained()` is a loading regression; CLIP text-model structural inconsistency is a different API change." + }, + { + "left": "issue:43334", + "right": "issue:43525", + "accept": false, + "reason": "Both mention missing `pad_token_id`, but they affect different model configs and loading paths; no clear shared bug." + }, + { + "left": "issue:43278", + "right": "issue:43856", + "accept": false, + "reason": "Embedding dtype drift between train/eval is unrelated to Qwen3 MoE memory inefficiency." + }, + { + "left": "issue:43452", + "right": "issue:46512", + "accept": false, + "reason": "GGUF handling in `from_pretrained()` is unrelated to stale `config.dtype` after `from_config(dtype=...)`." } ] }, - "evaluator_result": { - "accept": false, - "feedback": "The high-level conclusion is mostly reasonable, but the response is not conservative enough for this packet. Several soft-edge reasons assert detailed failure modes and code paths that are not supported by the packet data (which only provides titles/metadata), e.g. the serve/KV-cache distinctions and other specific implementation claims. The 38175\u219443493 accept is plausible from the titles, but the writeup should stay at the title-level commonality. Also, saying 43493 is the 'only credible duplicate subgroup' slightly overstates the evidence given that other SigLIP2-related items exist, even if they look distinct." - }, - "evaluator_used": true, - "retried": true, - "accepted_nontrivial_soft_edge": true, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -8352,29 +8380,28 @@ "cluster_id": "cluster-10000-469", "nodes": [ "issue:28218", + "issue:29942", + "issue:32090", "issue:35141", "issue:37428", "issue:38175", "issue:41720", "issue:42915", "issue:43262", - "issue:43278", - "issue:43334", "issue:43335", "issue:43388", "issue:43425", "issue:43450", - "issue:43452", - "issue:43525", "issue:43526", "issue:43550", "issue:43606", + "issue:43611", + "issue:43653", "issue:43819", "issue:43824", "issue:43827", - "issue:43856", "issue:43901", - "issue:43950", + "issue:43931", "issue:43975", "issue:44112", "issue:44206", @@ -8401,17 +8428,13 @@ "issue:46512" ], "soft_pairs": [ - "issue:43950|issue:46285", - "issue:43334|issue:43525", "issue:28218|issue:35141", - "issue:43278|issue:43856", - "issue:43452|issue:46512", "issue:42915|issue:43824", "issue:43262|issue:43450", "issue:43526|issue:46192", "issue:43824|issue:43975", - "issue:45200|issue:45325", "issue:43388|issue:43526", + "issue:45200|issue:45325", "issue:44442|issue:45198", "issue:44792|issue:45563", "issue:43819|issue:44112", @@ -8420,212 +8443,216 @@ "issue:43827|issue:43901", "issue:44561|issue:46234", "issue:41720|issue:46209", - "issue:45759|issue:46285", "issue:44206|issue:45584", + "issue:45759|issue:46285", "issue:43335|issue:43550", - "issue:43819|issue:45083", "issue:44857|issue:45593", + "issue:43819|issue:45083", "issue:45520|issue:46170", - "issue:43606|issue:45083", "issue:45759|issue:46512", - "issue:37428|issue:43425" + "issue:43606|issue:45083", + "issue:37428|issue:43425", + "issue:29942|issue:32090", + "issue:43335|issue:43653", + "issue:43550|issue:44112", + "issue:43611|issue:43931" ], - "prepared_review_unit_hash": "95f4ee0ad6746cc6d0f93d1f20b22a8b0f257c18509360f891e70f63eb099e47", + "prepared_review_unit_hash": "f9cce3f063192b6b2f467ebbb0510067a889060e87bcb306a6e7b7a9a48477d9", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 36335, - "estimated_input_tokens": 9084, - "estimated_eval_tokens": 18424 + "serialized_chars": 35695, + "estimated_input_tokens": 8924, + "estimated_eval_tokens": 18104 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose collection of unrelated Transformers issues and regressions across tokenization, multimodal configs, loaders, processors, and hardware/runtime compatibility. None of the soft pairs look like true duplicates of the same underlying bug/change, so I would not merge them.", - "confidence": 0.89, - "canonical_issue_reason": "issue:43950 is the best representative of the cluster because it is a broad core-loading regression with clear impact, while the other items are mostly isolated model-specific, processor-specific, or environment-specific bugs.", + "summary": "These soft links are mostly title-level keyword matches across unrelated Transformer bugs. I found no true duplicate pairs, so I rejected all candidate edges.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue stands out: the items span many unrelated bugs, regressions, and docs issues rather than one underlying problem.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43950 is the strongest single issue to anchor the cluster: concrete, widely relevant, and closer to a central library regression than the more niche reports.", + "best_issue_reason": "None of the issues is a strong cluster representative because the proposed links do not form a coherent duplicate set.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43950", - "right": "issue:46285", - "accept": false, - "reason": "Both concern model loading/structure, but one is about non-persistent buffers in from_pretrained and the other about CLIP text model attribute structure; different bugs and code paths." - }, - { - "left": "issue:43334", - "right": "issue:43525", - "accept": false, - "reason": "Both mention missing pad_token_id on config objects, but they affect different model families and likely different config classes/pathways." - }, { "left": "issue:28218", "right": "issue:35141", "accept": false, - "reason": "Tokenizer spacing after added tokens is unrelated to embedding reinitialization during resize_token_embeddings." - }, - { - "left": "issue:43278", - "right": "issue:43856", - "accept": false, - "reason": "dtype drift between training/evaluate is a different issue from Qwen3 MoE memory usage during training." - }, - { - "left": "issue:43452", - "right": "issue:46512", - "accept": false, - "reason": "gguf_file loading failures and stale config.dtype after from_config are separate loading/configuration bugs." + "reason": "Both mention token embeddings, but one is a tokenizer spacing bug and the other is output embedding reinitialization after resize; different code paths." }, { "left": "issue:42915", "right": "issue:43824", "accept": false, - "reason": "Qwen3Moe FP8 failure and Qwen2.5-VL import error are unrelated problems in different components." + "reason": "FP8Config failure in Qwen3MoE is unrelated to a missing Qwen2.5-VL import class." }, { "left": "issue:43262", "right": "issue:43450", "accept": false, - "reason": "Audio chat-template sampling-rate defaults and batched video processor shape issues are unrelated processor bugs." + "reason": "Audio chat-template sampling rate mismatch and batched video shape handling are separate processor bugs." }, { "left": "issue:43526", "right": "issue:46192", "accept": false, - "reason": "Label reduction in BeitImageProcessorFast and return_tensors being ignored with text_kwargs are different API bugs." + "reason": "One is label reduction dropping values; the other is return_tensors being ignored when text_kwargs is passed." }, { "left": "issue:43824", "right": "issue:43975", "accept": false, - "reason": "Importing Qwen2.5-VL and detokenization behavior for deepseek-coder are unrelated." + "reason": "An import error for a model class is not the same as a detokenization regression in v5." }, { - "left": "issue:45200", - "right": "issue:45325", + "left": "issue:43388", + "right": "issue:43526", "accept": false, - "reason": "Gemma mm_token_type_ids defaults and Qwen2.5-VL rope index scaling are different multimodal issues." + "reason": "Both touch labels, but one is gather_for_metrics truncation and the other is BeitImageProcessorFast reduce_labels behavior." }, { - "left": "issue:43388", - "right": "issue:43526", + "left": "issue:45200", + "right": "issue:45325", "accept": false, - "reason": "Both involve labels, but gather_for_metrics batch slicing and BeitImageProcessorFast reduce_labels are distinct code paths and bugs." + "reason": "Gemma 4 mm_token_type_ids defaults and Qwen2.5-VL RoPE temporal scaling are different multimodal bugs." }, { "left": "issue:44442", "right": "issue:45198", "accept": false, - "reason": "FastSpeech2ConformerTokenizer loading failure and Wav2Vec2 save/tokenization failure are not the same underlying defect." + "reason": "FastSpeech2ConformerTokenizer loading failure is unrelated to Wav2Vec2 save_pretrained/tokenization breakage." }, { "left": "issue:44792", "right": "issue:45563", "accept": false, - "reason": "A failing janus image-generation test and a stale generate warning are different generate-path issues." + "reason": "A Janus image-generation test failure and a stale paged-generate warning are not the same problem." }, { "left": "issue:43819", "right": "issue:44112", "accept": false, - "reason": "DAC latent decoding mismatch and a GraniteSpeech CI test issue are unrelated." + "reason": "DAC latent/forward mismatch and GraniteSpeech CI device-override flakiness are unrelated." }, { "left": "issue:44610", "right": "issue:44857", "accept": false, - "reason": "OmDet-Turbo input sizing and LwDetrImageLoss AMP crash are separate vision bugs." + "reason": "Processor output size mismatch and LwDetrImageLoss AMP crash are different detection issues." }, { "left": "issue:38175", "right": "issue:45412", "accept": false, - "reason": "SigLIP2 zero probabilities and RT-DETR memory release are unrelated symptoms and subsystems." + "reason": "Unexpected zero probabilities in SigLIP2 is unrelated to RT-DETR memory-release behavior." }, { "left": "issue:43827", "right": "issue:43901", "accept": false, - "reason": "Both are documentation mismatches, but they concern different pipeline APIs and different docs cleanup work." + "reason": "Both are docs-related, but one is pipeline removal fallout and the other is a stale return_all_scores doc note." }, { "left": "issue:44561", "right": "issue:46234", "accept": false, - "reason": "Missing utility APIs in import_utils is broad, but these are different removed/absent helpers affecting different downstream code." + "reason": "Removed is_torch_fx_available breaking trust_remote_code is not the same as a missing is_falcon_mamba_ssm_available helper." }, { "left": "issue:41720", "right": "issue:46209", "accept": false, - "reason": "Both are GPU/runtime failures, but one is a device-mapping assert on A800 and the other is an FP8 inference incompatibility on B200; different bugs." + "reason": "Auto device mapping cuda assert on A800 and FP8 inference failure on B200 are separate hardware/runtime issues." }, { - "left": "issue:45759", - "right": "issue:46285", + "left": "issue:44206", + "right": "issue:45584", "accept": false, - "reason": "from_config not unwrapping text_config and CLIP text model structural inconsistency are distinct configuration issues." + "reason": "Unsupported center arg in LasrFeatureExtractor and Whisper empty-transcription failure after align_special_tokens are unrelated." }, { - "left": "issue:44206", - "right": "issue:45584", + "left": "issue:45759", + "right": "issue:46285", "accept": false, - "reason": "Unsupported feature-extractor arg and Whisper empty-transcription failure are unrelated audio bugs." + "reason": "Composite Qwen multimodal config unwrapping and CLIP text-model structure inconsistency are different model/config bugs." }, { "left": "issue:43335", "right": "issue:43550", "accept": false, - "reason": "SwitchTransformers sparse-layer construction and Bamba torch.compile SDPA failure do not share the same underlying code-path problem." + "reason": "SwitchTransformers zero sparse-layer config and Bamba torch.compile SDPA failure do not share the same bug." }, { - "left": "issue:43819", - "right": "issue:45083", + "left": "issue:44857", + "right": "issue:45593", "accept": false, - "reason": "DAC from_latents mismatch and qwen3_omni_moe feature-length helper behavior are unrelated." + "reason": "Float16 AMP crash in LwDetrImageLoss is unrelated to D-FINE skipping auxiliary losses when denoising is off." }, { - "left": "issue:44857", - "right": "issue:45593", + "left": "issue:43819", + "right": "issue:45083", "accept": false, - "reason": "A CUDA AMP crash in loss computation and missing auxiliary losses when denoising is off are different model-training bugs." + "reason": "DAC.from_latents mismatch and qwen3_omni_moe feat-extract length behavior are different helper/model issues." }, { "left": "issue:45520", "right": "issue:46170", "accept": false, - "reason": "Python 3.13 flash_attn import handling and Windows ROCm torch.distributed dependency issues are separate environment compatibility problems." + "reason": "Python 3.13 flash_attn import KeyError and Windows ROCm torch.distributed dependency problems are unrelated." }, { - "left": "issue:43606", - "right": "issue:45083", + "left": "issue:45759", + "right": "issue:46512", "accept": false, - "reason": "CPU offload device mismatch in bark-small and qwen3_omni_moe output-length helper behavior are unrelated." + "reason": "Both involve from_config, but one is about unwrapping text_config and the other is stale config.dtype after dtype construction." }, { - "left": "issue:45759", - "right": "issue:46512", + "left": "issue:43606", + "right": "issue:45083", "accept": false, - "reason": "Both are from_config-related, but one is about unwrapping composite multimodal configs and the other about syncing config.dtype after dtype-aware construction." + "reason": "CPU offload device mismatch for bark-small is unrelated to qwen3_omni_moe feature-length helper behavior." }, { "left": "issue:37428", "right": "issue:43425", "accept": false, - "reason": "A missing flash-attention utility import and Torch 2.10 incompatibility are different compatibility problems." + "reason": "Missing _flash_supports_window_size is a specific import break, not the same as broad Torch 2.10 incompatibility." + }, + { + "left": "issue:29942", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention 2 test failures and a Trainer _gpu_broadcast_one NoneType error are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "SwitchTransformers sparse-layer config bug and BigBirdTokenizer special-token registration issue are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:44112", + "accept": false, + "reason": "A torch.compile/SDPA model failure is not the same as a stale CI device-override test." + }, + { + "left": "issue:43611", + "right": "issue:43931", + "accept": false, + "reason": "base_model_prefix loading regression and Qwen3-VL weight-shape mismatch are different model-loading failures." } ] }, @@ -8652,7 +8679,6 @@ "issue:26413", "issue:28282", "issue:29127", - "issue:29942", "issue:31501", "issue:32090", "issue:33357", @@ -8660,18 +8686,16 @@ "issue:35141", "issue:37428", "issue:39692", + "issue:42490", "issue:43278", "issue:43335", "issue:43504", "issue:43526", - "issue:43550", "issue:43606", - "issue:43611", - "issue:43653", "issue:43720", "issue:43819", "issue:43825", - "issue:43931", + "issue:43854", "issue:44112", "issue:44206", "issue:44351", @@ -8680,28 +8704,28 @@ "issue:44514", "issue:44521", "issue:44743", + "issue:44779", "issue:44855", "issue:44857", + "issue:45072", "issue:45083", "issue:45127", "issue:45200", "issue:45245", "issue:45290", "issue:45310", + "issue:45459", "issue:45478", "issue:45588", "issue:45593", "issue:45823", "issue:45859", + "issue:46234", "issue:46326", "issue:46431", "issue:46566" ], "soft_pairs": [ - "issue:29942|issue:32090", - "issue:43335|issue:43653", - "issue:43550|issue:44112", - "issue:43611|issue:43931", "issue:28282|issue:33357", "issue:44514|issue:44521", "issue:44448|issue:44743", @@ -8725,213 +8749,224 @@ "issue:26413|issue:31501", "issue:44206|issue:45823", "issue:45310|issue:45478", - "issue:43606|issue:44857" + "issue:43606|issue:44857", + "issue:42490|issue:43854", + "issue:39692|issue:44493", + "issue:45459|issue:46234", + "issue:44448|issue:44779", + "issue:44857|issue:45072" ], - "prepared_review_unit_hash": "8522dcba99b3add4b3c238025bd33a094152e12d215c88e4a561f7dfc5d64e49", + "prepared_review_unit_hash": "f8b3a01a5252b506ac09545d37cc44586053b7c4af8ac6ef56e3e9eaba7fbe6c", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 28, - "serialized_chars": 35797, - "estimated_input_tokens": 8950, - "estimated_eval_tokens": 18156 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36746, + "estimated_input_tokens": 9187, + "estimated_eval_tokens": 18630 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly unrelated bug reports that only share broad surface similarity. The one strong duplicate pair is the identical Qwen3.5 Moe `from_pretrained` issue (45310/45478); the rest should stay separate.", - "confidence": 0.93, - "canonical_issue_reason": "Issue 45310 is the best canonical anchor for the duplicate pair because it is the earlier report and has the same title and error description as 45478.", + "summary": "This cluster is mostly false-positive similarity matches across unrelated bugs. The only clear duplicate pair is the Qwen3.5 MoE `from_pretrained` regression (`45310` / `45478`).", + "confidence": 0.95, + "canonical_issue_reason": "Issue `45310` is the clearest canonical representative because it is a concrete, reproducible bug report and the matching duplicate (`45478`) is the same Qwen3.5 MoE `from_pretrained` failure.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45310 is the best representative of the duplicate pair since it is the earliest copy of the same Qwen3.5 Moe `from_pretrained` failure.", + "best_issue_reason": "`45310` is the best issue to keep as the cluster representative: it has the strongest duplicate evidence and the most direct problem statement.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:29942", - "right": "issue:32090", - "accept": false, - "reason": "Different failures and code paths: Flash Attention 2 test breakage vs a Trainer `_gpu_broadcast_one` NoneType error." - }, - { - "left": "issue:43335", - "right": "issue:43653", - "accept": false, - "reason": "Unrelated bugs in different subsystems: SwitchTransformers sparse-layer config vs BigBirdTokenizer special-token decode behavior." - }, - { - "left": "issue:43550", - "right": "issue:44112", - "accept": false, - "reason": "Different models and symptoms: Bamba/SDPA `torch.compile` failure vs GraniteSpeech CI stale device override test." - }, - { - "left": "issue:43611", - "right": "issue:43931", - "accept": false, - "reason": "Both are loading errors, but `base_model_prefix` handling and Qwen3-VL weight-shape mismatch are distinct problems." - }, { "left": "issue:28282", "right": "issue:33357", "accept": false, - "reason": "A missing-PyTorch ImportError is not the same as a MacOS bus error in a community CLIP model." + "reason": "Different failure modes: missing PyTorch vs MacOS bus error on CLIP loading; same library, not the same bug." }, { "left": "issue:44514", "right": "issue:44521", "accept": false, - "reason": "Both involve `apply_chat_template`, but the underlying defects differ: batched padding=False crash vs all-zero assistant masks for multimodal inputs." + "reason": "Both involve chat templates, but one is a batched padding crash and the other is incorrect assistant masks for multimodal inputs." }, { "left": "issue:44448", "right": "issue:44743", "accept": false, - "reason": "Different models and code paths: Pegasus v4/v5 output drift vs Qwen3.5 modular recurrent-state reset." + "reason": "Unrelated model behaviors: Pegasus output drift across versions vs Qwen3 recurrent state reset with cache." }, { "left": "issue:43335", "right": "issue:43504", "accept": false, - "reason": "SwitchTransformers sparse-layer creation and BEiT pretrained-preset loading are unrelated bugs." + "reason": "Different models and code paths: SwitchTransformers sparse-layer config bug vs BEiT legacy pretrained-field loading issue." }, { "left": "issue:44855", "right": "issue:45083", "accept": false, - "reason": "Python 3.13 JIT parsing/IndentationError in DebertaV2 is unrelated to qwen3_omni_moe output-length behavior." + "reason": "Python import/parse failure in DebertaV2 is unrelated to the Qwen3 Omni MoE feature-length helper behavior." }, { "left": "issue:45290", "right": "issue:46326", "accept": false, - "reason": "Both touch chat-template logic, but tool-call/no-content crashes and `continue_final_message` behavior are different underlying issues." + "reason": "Both are tokenizer/chat-template bugs, but one is tool-call content handling and the other is `continue_final_message` for mistral_common." }, { "left": "issue:46431", "right": "issue:46566", "accept": false, - "reason": "A missing chat template error is unrelated to DiffusionGemma weights staying on `meta` with `device_map=\"auto\"`." + "reason": "A missing chat template error is unrelated to DiffusionGemma weights \u10d3\u10d0\u10e0\u10e9ing on `meta` with `device_map='auto'`." }, { "left": "issue:37428", "right": "issue:44351", "accept": false, - "reason": "Different import failures for different symbols; no evidence of the same bug or code path." + "reason": "Different missing-symbol import errors (`_flash_supports_window_size` vs `HybridCache`), not the same underlying breakage." }, { "left": "issue:29127", "right": "issue:32090", "accept": false, - "reason": "LayoutLMv3 box-information error and Trainer GPU broadcast NoneType error are unrelated." + "reason": "Clarity of an error message in LayoutLMv3 is unrelated to a Trainer `_gpu_broadcast_one` NoneType crash." }, { "left": "issue:43278", "right": "issue:43825", "accept": false, - "reason": "Embedding dtype drift during evaluate and a pipeline translation-task error message are different bugs." + "reason": "Embedding dtype drift during eval is unrelated to an incorrect pipeline translation-task warning." }, { "left": "issue:43335", "right": "issue:45083", "accept": false, - "reason": "SwitchTransformers sparse-layer misconfiguration and helper output-length calculation are unrelated." + "reason": "Different component bugs: SwitchTransformers config construction vs Qwen3 Omni MoE feature-length helper behavior." }, { "left": "issue:43720", "right": "issue:44112", "accept": false, - "reason": "BitNet packed-weight loading and GraniteSpeech device override CI failure are different problems." + "reason": "Packed-weight loading in BitNet is unrelated to a stale GraniteSpeech CI test." }, { "left": "issue:45588", "right": "issue:45859", "accept": false, - "reason": "Both involve attention/offload areas, but `s_aux=None` flash-attention crashes and CPU offload incompatibility are distinct failures." + "reason": "Flash-attention sink handling and CPU-offload rotary embedding issues are different code paths and symptoms." }, { "left": "issue:45200", "right": "issue:45588", "accept": false, - "reason": "Gemma 4 `mm_token_type_ids` defaults and flash_attention `s_aux=None` crashes are unrelated." + "reason": "Gemma 4 multimodal token-type defaults are unrelated to the flash_attention `s_aux=None` crash." }, { "left": "issue:43335", "right": "issue:44112", "accept": false, - "reason": "Sparse-layer configuration in SwitchTransformers and GraniteSpeech CI device override are unrelated." + "reason": "No shared bug: one is a SwitchTransformers config edge case, the other is a GraniteSpeech CI failure." }, { "left": "issue:45127", "right": "issue:45245", "accept": false, - "reason": "LoRA merge collapse with tied embeddings and the 2^24 category limit are completely different issues." + "reason": "LoRA/extended-vocab collapse is unrelated to the category-cardinality runtime limit." }, { "left": "issue:43335", "right": "issue:45593", "accept": false, - "reason": "SwitchTransformers sparse-layer creation and D-FINE auxiliary-loss behavior do not match." + "reason": "SwitchTransformers sparse-layer creation and D-FINE auxiliary-loss handling are different issues." }, { "left": "issue:39692", "right": "issue:43526", "accept": false, - "reason": "SigLIP2 docs example problems are unrelated to BeitImageProcessorFast `reduce_labels` behavior." + "reason": "Docs/example errors for SigLIP2 are unrelated to BeitImageProcessorFast's `reduce_labels` bug." }, { "left": "issue:43819", "right": "issue:44493", "accept": false, - "reason": "DAC `from_latents`/STE mismatch and unexpected position-id keys are different model behaviors." + "reason": "DAC latent-forward mismatch is unrelated to unexpected position-id keys across many models." }, { "left": "issue:34689", "right": "issue:35141", "accept": false, - "reason": "Llama 3.2 Vision loading breakage is not the same as output-embedding reinitialization after resizing embeddings." + "reason": "Llama 3.2 Vision model loading regression is not the same as token-embedding resize reinitialization." }, { "left": "issue:26413", "right": "issue:31501", "accept": false, - "reason": "Checkpoint resume epoch-iterator failure and FineWeb training not starting are both training-related but not the same bug." + "reason": "Training resume iterator failure and FineWeb training not starting are both training-related, but not the same underlying bug." }, { "left": "issue:44206", "right": "issue:45823", "accept": false, - "reason": "LasrFeatureExtractor regression and Gemma4 PLE device mismatch are unrelated." + "reason": "Unsupported feature-extractor args and Gemma4 device-map mismatch are unrelated failures." }, { "left": "issue:45310", "right": "issue:45478", "accept": true, - "reason": "Same underlying bug: identical title and the same Qwen3.5 Moe `from_pretrained` error report, so these are duplicates." + "reason": "Exact duplicate: both report the same `transformers>=5.4.0` Qwen3.5 MoE `from_pretrained` error." }, { "left": "issue:43606", "right": "issue:44857", "accept": false, - "reason": "CPU offload device mismatch in bark-small and float16 AMP crashes in LwDetrImageLoss are different failures." + "reason": "CPU-offload device mismatch in Bark is unrelated to float16 AMP crashes in LwDetrImageLoss." + }, + { + "left": "issue:42490", + "right": "issue:43854", + "accept": false, + "reason": "A local-path import warning is unrelated to the GLM-4.7-Flash unit-test loading failure." + }, + { + "left": "issue:39692", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 documentation issues are not the same as broad position-id key warnings." + }, + { + "left": "issue:45459", + "right": "issue:46234", + "accept": false, + "reason": "Tokenizer protobuf error masking is unrelated to a missing `is_falcon_mamba_ssm_available` import utility symbol." + }, + { + "left": "issue:44448", + "right": "issue:44779", + "accept": false, + "reason": "Pegasus v4/v5 output differences are unrelated to Deepseek tokenizer regressions." + }, + { + "left": "issue:44857", + "right": "issue:45072", + "accept": false, + "reason": "These are distinct numerical/runtime issues: AMP crash in LwDetrImageLoss vs dtype mismatch in SwitchTransformers/TimmWrapperModel inference." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The duplicate judgment for 45310/45478 is plausible from the identical titles and very similar report text, but the rationale overstates what the packet proves. In particular, the claim that 45310 is earlier and has more discussion/references is only partly true (45478 has more discussion activity), and wording like \"clearest and most complete report\" is not supported by the metadata alone. The rest of the soft-edge rejections are conservative, but the overall summary should be tightened to avoid unsupported certainty." + "accept": true, + "feedback": "The summary is well grounded in the packet: the only clearly supported duplicate is 45310/45478, and the other soft-edge rejections stay conservative by citing distinct titles/failure modes rather than overstating certainty. No mergeability issues apply here since these are issue pairs, not PR pairs." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -8951,6 +8986,7 @@ "nodes": [ "issue:10000", "issue:15354", + "issue:25251", "issue:27957", "issue:28218", "issue:29942", @@ -8961,8 +8997,8 @@ "issue:35141", "issue:37428", "issue:38175", - "issue:39692", "issue:40444", + "issue:41720", "issue:42490", "issue:43278", "issue:43329", @@ -8972,11 +9008,12 @@ "issue:43575", "issue:43701", "issue:43746", - "issue:43854", "issue:43873", "issue:43927", + "issue:43976", "issue:44360", "issue:44448", + "issue:44466", "issue:44493", "issue:44561", "issue:44779", @@ -8984,26 +9021,19 @@ "issue:44857", "issue:44977", "issue:44987", - "issue:45072", "issue:45083", "issue:45229", "issue:45290", "issue:45381", - "issue:45459", + "issue:45440", "issue:45588", "issue:45698", "issue:45920", "issue:46192", "issue:46209", - "issue:46234", "issue:46399" ], "soft_pairs": [ - "issue:45459|issue:46234", - "issue:39692|issue:44493", - "issue:42490|issue:43854", - "issue:44448|issue:44779", - "issue:44857|issue:45072", "issue:28218|issue:32090", "issue:44360|issue:44561", "issue:43701|issue:45083", @@ -9016,8 +9046,8 @@ "issue:43550|issue:45083", "issue:45920|issue:46192", "issue:44448|issue:44977", - "issue:44855|issue:46399", "issue:27957|issue:34689", + "issue:44855|issue:46399", "issue:44977|issue:45229", "issue:43278|issue:43329", "issue:10000|issue:15354", @@ -9028,210 +9058,214 @@ "issue:37428|issue:44987", "issue:44561|issue:44779", "issue:43873|issue:45229", - "issue:32090|issue:33290" + "issue:32090|issue:33290", + "issue:25251|issue:32090", + "issue:41720|issue:45229", + "issue:44448|issue:44466", + "issue:43976|issue:45440" ], - "prepared_review_unit_hash": "cfbe84d1202a39c843d8627d19e64c515ff4e7d1c63cd9257639db3e929e54b5", + "prepared_review_unit_hash": "dd23d6346c22253fce35606f2bb5b4a98eecf21e975528d312e4e92eab9c56a2", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36766, - "estimated_input_tokens": 9192, - "estimated_eval_tokens": 18640 + "soft_pair_count": 29, + "serialized_chars": 36429, + "estimated_input_tokens": 9108, + "estimated_eval_tokens": 18472 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster looks like a loose text-similarity group of mostly unrelated bugs across tokenization, flash attention, model loading, dtype, and training issues. None of the soft pairs look like the same concrete bug/change, so I would not merge any of them as duplicates. The broadest representative issue is the v5 tokenizer regression affecting multiple models.", - "confidence": 0.88, - "canonical_issue_reason": "issue:45920 is the broadest and most central report in the set: it describes a v5 tokenizer regression impacting multiple model families, which makes it the best umbrella issue for the cluster.", + "summary": "These issues are broadly heterogeneous; the soft links mostly reflect superficial similarity in error types or subsystem names, not the same underlying bug. No soft pair looks safe to merge as a duplicate.", + "confidence": 0.93, + "canonical_issue_reason": "If one issue must anchor the cluster, issue 45920 is the best representative: it is a broad, clearly described v5 tokenizer regression affecting multiple models and has the widest apparent user impact.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 is the best issue candidate because it is open, broadly scoped, and explicitly aggregates multiple model failures rather than a single model-specific edge case.", + "best_issue_reason": "Issue 45920 is the strongest triage target because it is open, concrete, and summarizes a multi-model regression in a way that can guide follow-up work.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:45459", - "right": "issue:46234", - "accept": false, - "reason": "Both are import-related, but they concern different missing-symbol/tokenizer-error paths and different modules; not the same bug." - }, - { - "left": "issue:39692", - "right": "issue:44493", - "accept": false, - "reason": "One is a SigLIP2 docs/example problem, the other is a position-id regression; no shared underlying failure." - }, - { - "left": "issue:42490", - "right": "issue:43854", - "accept": false, - "reason": "Local-path trailing-slash warning and GLM-4.7-Flash loading failure are different load-time issues." - }, - { - "left": "issue:44448", - "right": "issue:44779", - "accept": false, - "reason": "Both mention v5 regressions, but they affect different models/tokenizers and different outputs." - }, - { - "left": "issue:44857", - "right": "issue:45072", - "accept": false, - "reason": "Both are dtype/AMP-related, but they hit different models and code paths; not mergeable as one fix." - }, { "left": "issue:28218", "right": "issue:32090", "accept": false, - "reason": "Tokenizer spacing bug vs distributed Trainer broadcast TypeError; unrelated failures." + "reason": "Added-token spacing in tokenization and a Trainer _gpu_broadcast_one NoneType error are unrelated failures." }, { "left": "issue:44360", "right": "issue:44561", "accept": false, - "reason": "A missing ReLU in DSA indexer is unrelated to the v5 API-removal breakage in remote-code models." + "reason": "DSA indexer ReLU discussion and a v5 trust_remote_code compatibility break are different code paths and symptoms." }, { "left": "issue:43701", "right": "issue:45083", "accept": false, - "reason": "Checkpoint key mismatch and feature-extraction length helper behavior are separate bugs." + "reason": "Checkpoint key mismatches and a qwen3_omni_moe helper behavior issue are not the same bug." }, { "left": "issue:38175", "right": "issue:44493", "accept": false, - "reason": "One is a SigLIP2 probability issue, the other a position-id warning regression; different root causes." + "reason": "Zero probabilities in SigLIP2 and unexpected position-id keys are distinct model-output problems." }, { "left": "issue:45290", "right": "issue:45381", "accept": false, - "reason": "Chat-template tool-call crash and Qwen2.5-VL video position ids are distinct multimodal bugs." + "reason": "Chat template crashes on tool-call messages and Qwen2.5-VL video position-id issues are separate failures." }, { "left": "issue:32090", "right": "issue:33357", "accept": false, - "reason": "Trainer broadcast NoneType error and MacOS bus error with CLIP are unrelated." + "reason": "Distributed broadcast NoneType support and a MacOS bus error with CLIP are unrelated." }, { "left": "issue:40444", "right": "issue:43526", "accept": false, - "reason": "Multi-image Qwen2.5-VL finetuning failure and BeitImageProcessorFast label reduction bug are not the same issue." + "reason": "Multi-image iterable dataset finetuning and BeitImageProcessorFast reduce_labels behavior are different bugs." }, { "left": "issue:42490", "right": "issue:43575", "accept": false, - "reason": "Trailing-slash import warning and tensor-parallel OOM are different load-time problems." + "reason": "Local-path import warning and tensor-parallel OOM are not the same issue." }, { "left": "issue:29942", "right": "issue:35141", "accept": false, - "reason": "Flash Attention test failures and output-embedding reinitialization are unrelated." + "reason": "Flash Attention 2 test failures and post_init reinitializing embeddings are unrelated." }, { "left": "issue:43550", "right": "issue:45083", "accept": false, - "reason": "torch.compile/SDPA failure in Bamba is not the same bug as qwen3_omni_moe feature-length behavior." + "reason": "torch.compile/SDPA failure in Bamba and qwen3_omni_moe helper behavior are distinct." }, { "left": "issue:45920", "right": "issue:46192", "accept": false, - "reason": "The first is a tokenizer-ID regression across models; the second is return_tensors handling in text kwargs." + "reason": "Wrong tokenizer IDs across models and silently ignored return_tensors are different tokenizer regressions." }, { "left": "issue:44448", "right": "issue:44977", "accept": false, - "reason": "Different model families and different flash-attention/generation failures; not one concrete bug." + "reason": "Pegasus v4/v5 output drift and Qwen3.5 flash-attention generation failure are unrelated." }, { - "left": "issue:44855", - "right": "issue:46399", + "left": "issue:27957", + "right": "issue:34689", "accept": false, - "reason": "Python parsing/import issue in one model vs missing-decorator registration error in another; superficially decorator-related only." + "reason": "XLMRoberta Flash Attention 2 and Llama 3.2 Vision loading are separate model-specific problems." }, { - "left": "issue:27957", - "right": "issue:34689", + "left": "issue:44855", + "right": "issue:46399", "accept": false, - "reason": "XLMRoberta Flash Attention 2 support and Llama 3.2 Vision model loading are distinct issues." + "reason": "Python 3.13 decorator/comment parsing error is not the same as a missing apply_rotary_pos_emb decorator registration bug." }, { "left": "issue:44977", "right": "issue:45229", "accept": false, - "reason": "Flash-attention generation bug and multi-GPU CUDA OOM are different failure modes." + "reason": "Flash-attention generation issues and a multi-GPU CUDA OOM are different failure modes." }, { "left": "issue:43278", "right": "issue:43329", "accept": false, - "reason": "BF16/FP32 dtype drift in eval is unrelated to undefined video-branch helpers in multimodal token counting." + "reason": "Embedding dtype drift and a multimodal-token counting video-branch bug are unrelated." }, { "left": "issue:10000", "right": "issue:15354", "accept": false, - "reason": "German DistilBertModel issue and torch.jit GeneratorExp export failure are unrelated legacy bugs." + "reason": "German DistilBertModel and GeneratorExp torch.jit.script export are different historical issues." }, { "left": "issue:44857", "right": "issue:45588", "accept": false, - "reason": "AMP/CUDA crash in loss computation and flash_attention AttributeError on s_aux=None are different code paths." + "reason": "LwDetrImageLoss AMP crash and flash_attention.py s_aux=None AttributeError are not the same code-path bug." }, { "left": "issue:43927", "right": "issue:45698", "accept": false, - "reason": "One is custom token IDs being lost on save/load; the other is the wrong custom module being loaded after save_pretrained." + "reason": "Custom token IDs lost on save/load and wrong custom module loading after save_pretrained are related only broadly to serialization, not the same defect." }, { "left": "issue:43278", "right": "issue:43381", "accept": false, - "reason": "Embedding dtype mismatch and gradient checkpointing in eval mode are unrelated behaviors." + "reason": "BF16-to-FP32 embedding dtype change and gradient checkpointing in eval mode are unrelated." }, { "left": "issue:43746", "right": "issue:46209", "accept": false, - "reason": "PEFT checkpoint loading and FP8 inference on B200 are separate issues with no shared failure path." + "reason": "PEFT adapter checkpoint loading and FP8 inference on B200 are separate issues." }, { "left": "issue:37428", "right": "issue:44987", "accept": false, - "reason": "Importing a missing flash-attention helper and loading physical-intelligence/fast on v5 are not shown to be the same bug." + "reason": "Missing _flash_supports_window_size import and a model-loading failure for physical-intelligence/fast are too vague to treat as the same bug." }, { "left": "issue:44561", "right": "issue:44779", "accept": false, - "reason": "Global API removal breaking trust_remote_code is unrelated to DeepSeek tokenizer regression." + "reason": "Removed helper breaking trust_remote_code and DeepSeek tokenizer regression are different failures." }, { "left": "issue:43873", "right": "issue:45229", "accept": false, - "reason": "Quantization offloading behavior and Gemma4 multi-GPU OOM are different resource-management problems." + "reason": "Quantization offloading behavior and Gemma4 multi-GPU OOM are not the same underlying problem." + }, + { + "left": "issue:32090", + "right": "issue:33290", + "accept": false, + "reason": "Trainer broadcast TypeError and Adafactor/DeepSpeed OOM are unrelated." + }, + { + "left": "issue:25251", + "right": "issue:32090", + "accept": false, + "reason": "Pipeline top_k nesting behavior and a distributed broadcast TypeError do not share a code path." + }, + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "Qwen3 auto device mapping cudaErrorAssert and Gemma4 OOM are distinct memory/runtime issues." + }, + { + "left": "issue:44448", + "right": "issue:44466", + "accept": false, + "reason": "Pegasus output differences and lm_head serialization inconsistency are different regressions." + }, + { + "left": "issue:43976", + "right": "issue:45440", + "accept": false, + "reason": "Python version compatibility in Transformers 5.1.0 and DeepseekV3MoE divergence are unrelated." } ] }, @@ -9255,19 +9289,18 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-469", "nodes": [ - "issue:25251", + "issue:27957", "issue:29127", "issue:29942", "issue:30064", - "issue:32090", "issue:33357", "issue:34567", "issue:39692", - "issue:41720", "issue:42490", "issue:42915", "issue:42994", "issue:43065", + "issue:43066", "issue:43278", "issue:43381", "issue:43412", @@ -9277,20 +9310,20 @@ "issue:43701", "issue:43756", "issue:43873", - "issue:43976", "issue:43994", "issue:44186", "issue:44206", + "issue:44265", "issue:44361", - "issue:44448", - "issue:44466", + "issue:44442", + "issue:44462", "issue:44479", "issue:44568", "issue:44743", "issue:44871", "issue:44977", "issue:45198", - "issue:45229", + "issue:45245", "issue:45372", "issue:45406", "issue:45412", @@ -9302,29 +9335,26 @@ "issue:45712", "issue:45803", "issue:45823", - "issue:46121" + "issue:46121", + "issue:46369" ], "soft_pairs": [ - "issue:25251|issue:32090", - "issue:41720|issue:45229", - "issue:44448|issue:44466", - "issue:43976|issue:45440", "issue:45588|issue:45823", - "issue:44568|issue:45491", "issue:42490|issue:43873", + "issue:44568|issue:45491", "issue:45198|issue:45803", - "issue:45372|issue:45588", "issue:43412|issue:45588", + "issue:45372|issue:45588", "issue:44977|issue:45563", - "issue:33357|issue:39692", "issue:43278|issue:43575", + "issue:33357|issue:39692", "issue:29127|issue:34567", "issue:43653|issue:44743", "issue:43381|issue:43701", "issue:43611|issue:46121", "issue:44871|issue:44977", - "issue:45372|issue:45406", "issue:44361|issue:45198", + "issue:45372|issue:45406", "issue:44186|issue:45198", "issue:42915|issue:42994", "issue:44206|issue:44479", @@ -9332,204 +9362,243 @@ "issue:29942|issue:30064", "issue:43065|issue:43994", "issue:45459|issue:45712", - "issue:43756|issue:45440" + "issue:43756|issue:45440", + "issue:45412|issue:46369", + "issue:43653|issue:44442", + "issue:45245|issue:45563", + "issue:45372|issue:45459", + "issue:43994|issue:45412", + "issue:44265|issue:45198", + "issue:29127|issue:33357", + "issue:43066|issue:44462", + "issue:27957|issue:29942" ], - "prepared_review_unit_hash": "90716c4a1d729f0a576b2ac5a6ceacb71d1ef6f61aec574bef5602a9b84e583d", + "prepared_review_unit_hash": "35ab63a6113f29f36ac92e6467a30125430d064be194df7f65606b5a23c9e870", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 28, - "serialized_chars": 36231, - "estimated_input_tokens": 9058, - "estimated_eval_tokens": 18372 + "soft_pair_count": 33, + "serialized_chars": 37588, + "estimated_input_tokens": 9397, + "estimated_eval_tokens": 19050 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is a heterogeneous set of unrelated issues; the soft-similarity pairs are topical overlaps only and should not be collapsed as duplicates.", - "confidence": 0.95, - "canonical_issue_reason": "No single canonical issue: the items span separate bugs across different models, tokenizers, trainers, and regressions, and none of the soft pairs describe the same underlying defect.", + "summary": "This is not a duplicate cluster: the items span many unrelated bugs/regressions across different models, tokenizers, export paths, and runtime behaviors. The soft-similarity pairs mostly share broad subsystem keywords, not the same underlying defect.", + "confidence": 0.98, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "No issue in this set cleanly serves as a duplicate representative; the strongest matches are still different code paths or different failure modes.", + "best_issue_reason": "issue:43873 is the best anchor only because it is the most active open item with broad impact and discussion; it is not a duplicate representative of the other issues.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:25251", - "right": "issue:32090", + "left": "issue:45588", + "right": "issue:45823", "accept": false, - "reason": "Top_k nesting output bug vs Trainer _gpu_broadcast_one NoneType error; unrelated components and symptoms." + "reason": "Different bugs: flash_attention.py crashes on sink-less models, while Gemma4 PLE has a device-mismatch during forward." }, { - "left": "issue:41720", - "right": "issue:45229", + "left": "issue:42490", + "right": "issue:43873", "accept": false, - "reason": "Qwen3 device-mapping cuda assert vs Gemma4 multi-GPU OOM; both are GPU failures but not the same bug." - }, - { - "left": "issue:44448", - "right": "issue:44466", - "accept": false, - "reason": "Generation/output discrepancy vs lm_head.weight serialization/device behavior; different code paths and bug shapes." - }, - { - "left": "issue:43976", - "right": "issue:45440", - "accept": false, - "reason": "Python version compatibility regression vs DeepseekV3MoE implementation divergence; unrelated problems." - }, - { - "left": "issue:45588", - "right": "issue:45823", - "accept": false, - "reason": "flash_attention integration s_aux=None crash vs Gemma4 PLE device mismatch; distinct failures." + "reason": "Unrelated: trailing-slash local model import warning vs quantization/offloading behavior." }, { "left": "issue:44568", "right": "issue:45491", "accept": false, - "reason": "Tokenizer special-token insertion bug vs NaN embeddings from sliding-window attention; not the same defect." - }, - { - "left": "issue:42490", - "right": "issue:43873", - "accept": false, - "reason": "Local path trailing-slash import warning vs quantization offloading behavior; different subsystems and outcomes." + "reason": "Different tokenizer special-token regression vs Gemma3 NaN embeddings with mixed-length sliding-window batching." }, { "left": "issue:45198", "right": "issue:45803", "accept": false, - "reason": "Wav2Vec2 save/tokenization failure vs bare except swallowing errors in Fuyu batch conversion; unrelated." + "reason": "Wav2Vec2 save/tokenization failure is unrelated to Fuyu batch tensor conversion swallowing errors." }, { - "left": "issue:45372", + "left": "issue:43412", "right": "issue:45588", "accept": false, - "reason": "Gemma4 processor import dependency issue vs flash_attention AttributeError; different root causes." + "reason": "Gemma3n export failure and flash_attention sink-less crash are different code paths and symptoms." }, { - "left": "issue:43412", + "left": "issue:45372", "right": "issue:45588", "accept": false, - "reason": "ExecuTorch export guard/erfinv support issue vs sink-less flash_attention crash; not mergeable as one fix." + "reason": "Gemma 4 processor import breakage is unrelated to the flash_attention.py AttributeError." }, { "left": "issue:44977", "right": "issue:45563", "accept": false, - "reason": "Qwen3.5 flash-attention generation bug vs stale num_return_sequences warning; unrelated behavior." + "reason": "Qwen3.5 flash-attention generation bug vs stale paged-generate warning; not the same defect." }, { - "left": "issue:33357", - "right": "issue:39692", + "left": "issue:43278", + "right": "issue:43575", "accept": false, - "reason": "MacOS CLIP bus error vs SigLIP2 docs example errors; different models and failure modes." + "reason": "Different runtime issues: BF16/FP32 dtype drift vs tensor-parallel OOM on a specific Qwen model." }, { - "left": "issue:43278", - "right": "issue:43575", + "left": "issue:33357", + "right": "issue:39692", "accept": false, - "reason": "Embedding dtype regression vs tensor-parallel OOM on Qwen2-57B; different issues entirely." + "reason": "MacOS bus error loading a CLIP community model is unrelated to SigLIP2 doc example errors." }, { "left": "issue:29127", "right": "issue:34567", "accept": false, - "reason": "LayoutLMv3 error-message clarity vs TrainerState token counter not updating; not the same bug." + "reason": "LayoutLMv3 box-clarity error message issue vs TrainerState token-count update bug." }, { "left": "issue:43653", "right": "issue:44743", "accept": false, - "reason": "BigBirdTokenizer special-token registration/decoding issue vs Qwen recurrent-state reset; unrelated." + "reason": "BigBirdTokenizer special-token registration bug is unrelated to Qwen3.5 recurrent-state reset behavior." }, { "left": "issue:43381", "right": "issue:43701", "accept": false, - "reason": "Gradient checkpointing in eval mode vs resume_from_checkpoint key mismatch; separate training flows." + "reason": "Gradient checkpointing in eval mode and resume_from_checkpoint key mismatch are different training bugs." }, { "left": "issue:43611", "right": "issue:46121", "accept": false, - "reason": "base_model_prefix loading regression vs rope-param validation TypeError; no shared code-path bug." + "reason": "base_model_prefix loading breakage is unrelated to rope-validation JSON list handling." }, { "left": "issue:44871", "right": "issue:44977", "accept": false, - "reason": "Gemma3 eos_token_id mismatch vs Qwen3.5 flash-attention generation issue; different model families and symptoms." + "reason": "Gemma-3 eos_token_id mismatch is not the same as Qwen3.5 flash-attention generation failure." }, { - "left": "issue:45372", - "right": "issue:45406", + "left": "issue:44361", + "right": "issue:45198", "accept": false, - "reason": "Gemma4 processor import error vs transformers serve missing _tokenizer attribute; related model area but different failure points." + "reason": "MLukeTokenizer task AttributeError differs from Wav2Vec2 save_pretrained/tokenization failure." }, { - "left": "issue:44361", - "right": "issue:45198", + "left": "issue:45372", + "right": "issue:45406", "accept": false, - "reason": "MLukeTokenizer AttributeError vs Wav2Vec2 save/tokenization failure; unrelated tokenizer/model bugs." + "reason": "Both involve Gemma 4, but one is a missing ReasoningEffort import and the other is a serve-side _tokenizer AttributeError." }, { "left": "issue:44186", "right": "issue:45198", "accept": false, - "reason": "LayoutLMv2 NER padding/truncation crash vs Wav2Vec2 save/tokenization failure; different tokenizer stacks." + "reason": "LayoutLMv2Tokenizer NER/padding crash is unrelated to Wav2Vec2 save/tokenization failure." }, { "left": "issue:42915", "right": "issue:42994", "accept": false, - "reason": "Qwen3Moe FP8 config failure vs quantized model saving failure; quantization-themed but distinct bugs." + "reason": "Qwen3Moe FineGrainedFP8Config failure and quantized model saving failure are distinct quantization-related bugs." }, { "left": "issue:44206", "right": "issue:44479", "accept": false, - "reason": "LasrFeatureExtractor unsupported center arg vs Qwen video-input regression; different components and fixes." + "reason": "Feature extractor unsupported-arg crash is different from the video-input regression in Qwen VL models." }, { "left": "issue:43065", "right": "issue:45412", "accept": false, - "reason": "Dummy Conv2d placeholder in Sam3PixelDecoder vs RT-DETR memory leak; no common underlying defect." + "reason": "Dummy Conv2d in Sam3PixelDecoder and RT-DETR memory not released are unrelated model implementation bugs." }, { "left": "issue:29942", "right": "issue:30064", "accept": false, - "reason": "Flash Attention 2 test failures vs void segmentation map processing; different test failure classes." + "reason": "Failing Flash Attention 2 tests are not the same as void segmentation map processing failure." }, { "left": "issue:43065", "right": "issue:43994", "accept": false, - "reason": "Dummy placeholder class cleanup vs SigLIP2 nonsensical outputs; unrelated." + "reason": "Sam3PixelDecoder dummy layer issue is unrelated to SigLIP2 nonsensical AutoModel/pipeline results." }, { "left": "issue:45459", "right": "issue:45712", "accept": false, - "reason": "Tokenizer error handling around protobuf vs leftover dummy classes in repo checks; separate maintenance issues." + "reason": "Tokenizer protobuf error handling is unrelated to leftover dummy classes in dummy_pt_objects.py." }, { "left": "issue:43756", "right": "issue:45440", "accept": false, - "reason": "Smollm3 RoPE-layer config mismatch vs DeepseekV3MoE divergence from remote implementation; different models and bugs." + "reason": "Smollm3 RoPE-layer dropout bug is unrelated to DeepseekV3MoE divergence from remote implementation." + }, + { + "left": "issue:45412", + "right": "issue:46369", + "accept": false, + "reason": "RT-DETR memory leak on deletion is different from Sam3-LiteText memory overflow during use." + }, + { + "left": "issue:43653", + "right": "issue:44442", + "accept": false, + "reason": "BigBirdTokenizer special-token registration bug is unrelated to AutoTokenizer failing to load FastSpeech2ConformerTokenizer." + }, + { + "left": "issue:45245", + "right": "issue:45563", + "accept": false, + "reason": "Category-count RuntimeError and stale num_return_sequences warning are unrelated issues." + }, + { + "left": "issue:45372", + "right": "issue:45459", + "accept": false, + "reason": "Gemma 4 processor import failure is unrelated to protobuf exception masking tokenizer errors." + }, + { + "left": "issue:43994", + "right": "issue:45412", + "accept": false, + "reason": "SigLIP2 model/pipeline mismatch is unrelated to RT-DETR memory not being released." + }, + { + "left": "issue:44265", + "right": "issue:45198", + "accept": false, + "reason": "torch.export failure with torch_compilable_check is unrelated to Wav2Vec2 save/tokenization failure." + }, + { + "left": "issue:29127", + "right": "issue:33357", + "accept": false, + "reason": "LayoutLMv3 error-message clarity and MacOS CLIP bus error are unrelated bugs." + }, + { + "left": "issue:43066", + "right": "issue:44462", + "accept": false, + "reason": "Wrong tokenizer decoder type in v5 and AutoTokenizer ignoring tokenizer.json are different loading/tokenizer issues." + }, + { + "left": "issue:27957", + "right": "issue:29942", + "accept": false, + "reason": "XLMRoberta Flash Attention 2 support and failing Flash Attention 2 tests are not the same underlying bug." } ] }, @@ -9555,43 +9624,43 @@ "nodes": [ "issue:27957", "issue:29127", - "issue:29942", "issue:33290", - "issue:33357", + "issue:34567", "issue:34689", "issue:35141", "issue:42490", "issue:42915", - "issue:43066", + "issue:43262", "issue:43334", "issue:43381", "issue:43421", "issue:43526", "issue:43531", "issue:43575", - "issue:43653", + "issue:43646", "issue:43701", "issue:43761", "issue:43824", + "issue:43873", "issue:43931", "issue:43994", + "issue:44077", "issue:44186", - "issue:44265", + "issue:44206", "issue:44279", "issue:44387", "issue:44442", - "issue:44462", + "issue:44464", + "issue:44743", + "issue:44811", "issue:44857", + "issue:44898", + "issue:44977", "issue:45072", - "issue:45198", "issue:45245", "issue:45250", - "issue:45372", - "issue:45412", "issue:45440", - "issue:45459", "issue:45491", - "issue:45563", "issue:45588", "issue:45593", "issue:45600", @@ -9603,15 +9672,6 @@ "issue:46640" ], "soft_pairs": [ - "issue:45412|issue:46369", - "issue:45245|issue:45563", - "issue:43653|issue:44442", - "issue:45372|issue:45459", - "issue:43994|issue:45412", - "issue:44265|issue:45198", - "issue:29127|issue:33357", - "issue:43066|issue:44462", - "issue:27957|issue:29942", "issue:45440|issue:45593", "issue:43575|issue:43701", "issue:43421|issue:45491", @@ -9627,238 +9687,244 @@ "issue:43824|issue:44279", "issue:43526|issue:44857", "issue:46121|issue:46399", - "issue:43334|issue:43531", "issue:27957|issue:29127", + "issue:43334|issue:43531", "issue:27957|issue:33290", "issue:43761|issue:45859", "issue:33290|issue:34689", "issue:45588|issue:45600", - "issue:33290|issue:35141" + "issue:33290|issue:35141", + "issue:43262|issue:44206", + "issue:44977|issue:45245", + "issue:33290|issue:34567", + "issue:43381|issue:43873", + "issue:44464|issue:45491", + "issue:43646|issue:44077", + "issue:44743|issue:44811", + "issue:44442|issue:44898", + "issue:42490|issue:44743" ], - "prepared_review_unit_hash": "b8ee085b3bb3104772b1dafb9b0fc89cf76deb6ecc488ea0eef65350ef0655e1", + "prepared_review_unit_hash": "1eb57403c1e5dd71b92394a7fb339f5b18dfd732044552946de5e0f92ee97d35", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36815, - "estimated_input_tokens": 9204, - "estimated_eval_tokens": 18664 + "serialized_chars": 36765, + "estimated_input_tokens": 9192, + "estimated_eval_tokens": 18640 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most pairs are false positives driven by broad title similarity. The only plausible duplicate is the two generic Flash Attention 2 issue reports; the rest describe different models, subsystems, or failure modes.", - "confidence": 0.87, - "canonical_issue_reason": "issue:45250 is the broadest and most central Flash Attention 2 report in this set, making it the best duplicate target for the one plausible FA2 duplicate pair.", + "summary": "This cluster is highly heterogeneous: the soft-similarity pairs are mostly broad title overlaps, not the same underlying bug or change. No pull requests are present, and none of the soft edges should be merged.", + "confidence": 0.64, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:45250 is the best issue to anchor the cluster because it is the most generic, topic-level Flash Attention 2 report and can subsume the nearby FA2 duplicate.", + "best_issue_reason": "issue:43873 is the strongest representative issue in the set: it has the most discussion/inbound references and a concrete, well-scoped bug report.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:45412", - "right": "issue:46369", - "accept": false, - "reason": "Both mention memory problems, but they are different models and different failure modes; not the same bug." - }, - { - "left": "issue:45245", - "right": "issue:45563", - "accept": false, - "reason": "Unrelated issues: category-limit runtime error vs a stale generate warning." - }, - { - "left": "issue:43653", - "right": "issue:44442", - "accept": false, - "reason": "Both concern tokenizers, but one is BigBird special-token registration and the other is AutoTokenizer loading a specific tokenizer; different bugs." - }, - { - "left": "issue:45372", - "right": "issue:45459", - "accept": false, - "reason": "Both involve loading/import-related failures, but the concrete root causes and code paths are different." - }, - { - "left": "issue:43994", - "right": "issue:45412", - "accept": false, - "reason": "Nonsensical model outputs and RT-DETR memory retention are unrelated." - }, - { - "left": "issue:44265", - "right": "issue:45198", - "accept": false, - "reason": "torch.export/torch_compilable_check is a different subsystem from Wav2Vec2 save_pretrained/tokenization failures." - }, - { - "left": "issue:29127", - "right": "issue:33357", - "accept": false, - "reason": "LayoutLMv3 error messaging and a MacOS bus error in CLIP are unrelated." - }, - { - "left": "issue:43066", - "right": "issue:44462", - "accept": false, - "reason": "Different tokenizer bugs: decoder type mismatch in v5 versus ignoring tokenizer.json from a repo." - }, - { - "left": "issue:27957", - "right": "issue:29942", - "accept": false, - "reason": "Both mention Flash Attention 2, but one is XLMRoberta-specific while the other is a broader test failure; no clear same bug." - }, { "left": "issue:45440", "right": "issue:45593", "accept": false, - "reason": "DeepSeekV3MoE divergence and D-FINE auxiliary-loss behavior are unrelated model-specific issues." + "reason": "Different models/features and different failure modes; only a broad ML bug theme is shared." }, { "left": "issue:43575", "right": "issue:43701", "accept": false, - "reason": "TP loading OOM and resume-from-checkpoint key mismatch are different problems." + "reason": "One is an OOM during tensor-parallel model load, the other is a checkpoint resume key mismatch." }, { "left": "issue:43421", "right": "issue:45491", "accept": false, - "reason": "Tokenizer post-processor updates and Gemma3 NaN embeddings stem from different code paths." + "reason": "Tokenizer post-processor runtime behavior vs GPU NaNs in Gemma3; unrelated code paths." }, { "left": "issue:29127", "right": "issue:35141", "accept": false, - "reason": "Error-message clarity in LayoutLMv3 is unrelated to embedding reinitialization after resize." + "reason": "LayoutLMV3 error messaging and embedding reinit after resize are different bugs." }, { "left": "issue:43526", "right": "issue:45834", "accept": false, - "reason": "BeitImageProcessorFast label reduction and Kosmos2.5 OCR long-input indexing are different bugs." + "reason": "BeitImageProcessorFast label reduction bug vs Kosmos2.5 OCR index error; not the same issue." }, { "left": "issue:42490", "right": "issue:43381", "accept": false, - "reason": "A local-path import warning and gradient checkpointing in eval mode are unrelated." + "reason": "Local-model import warning with trailing slash vs gradient checkpointing eval-mode restriction." }, { "left": "issue:45250", "right": "issue:46640", - "accept": true, - "reason": "Both are generic Flash Attention 2 issue reports and appear to target the same underlying FA2 problem/request." + "accept": false, + "reason": "Both mention Flash Attention 2, but the titles are too generic to prove the same concrete bug." }, { "left": "issue:43575", "right": "issue:44387", "accept": false, - "reason": "TP load OOM and increased CUDA reserved memory under int4 quantization are different OOM causes." + "reason": "Both mention OOM, but one is TP model loading and the other is int4 quantization memory growth." }, { "left": "issue:43994", "right": "issue:46369", "accept": false, - "reason": "Different models and different memory-related symptoms; not the same underlying issue." + "reason": "SigLIP2 bad outputs vs SAM3-LiteText memory overflow; different models and symptoms." }, { "left": "issue:43526", "right": "issue:44186", "accept": false, - "reason": "Different tokenizer/processor bugs in different model families." + "reason": "Image processor label handling and LayoutLMv2 tokenizer crashes are unrelated." }, { "left": "issue:44442", "right": "issue:45072", "accept": false, - "reason": "Tokenizer loading and bfloat16 dtype mismatches are unrelated." + "reason": "Tokenizer loading failure vs dtype mismatch in inference; not the same code-path problem." }, { "left": "issue:42915", "right": "issue:43931", "accept": false, - "reason": "FP8 config failure in Qwen3Moe and weight-shape mismatch in Qwen3-VL are distinct issues." + "reason": "FineGrainedFP8Config failure for Qwen3MoE vs weight-shape mismatch loading Qwen3-VL." }, { "left": "issue:43824", "right": "issue:44279", "accept": false, - "reason": "An import error for a specific class and a generic dependency issue are not the same bug." + "reason": "Missing Qwen2.5-VL import and a generic dependency issue are not the same bug." }, { "left": "issue:43526", "right": "issue:44857", "accept": false, - "reason": "Image label reduction and LwDetrImageLoss AMP/CUDA crashes are unrelated." + "reason": "Label-reduction behavior in an image processor vs AMP crash in LwDetrImageLoss." }, { "left": "issue:46121", "right": "issue:46399", "accept": false, - "reason": "ROPE-parameter JSON list handling and Qwen3.5 kernel registration are different failures." + "reason": "RoPE validation TypeError vs missing decorator for apply_rotary_pos_emb registration." }, { - "left": "issue:43334", - "right": "issue:43531", + "left": "issue:27957", + "right": "issue:29127", "accept": false, - "reason": "Missing pad_token_id on Qwen3-VL and sliding_window issues on Qwen3-MoE are separate bugs." + "reason": "Flash Attention support for XLM-Roberta vs LayoutLMv3 error-message clarity; unrelated." }, { - "left": "issue:27957", - "right": "issue:29127", + "left": "issue:43334", + "right": "issue:43531", "accept": false, - "reason": "Flash Attention 2 support and LayoutLMv3 error messaging are unrelated." + "reason": "Qwen3-VL pad_token_id attribute error vs Qwen3-MoE sliding-window behavior." }, { "left": "issue:27957", "right": "issue:33290", "accept": false, - "reason": "Flash Attention 2 support and Deepspeed Adafactor OOM are unrelated." + "reason": "Flash Attention support vs AdaFactor/DeepSpeed OOM; different problems." }, { "left": "issue:43761", "right": "issue:45859", "accept": false, - "reason": "CLIP hidden_states regression and Qwen3_5Moe CPU-offload incompatibility are different model/code paths." + "reason": "CLIP hidden_states regression vs Qwen3_5Moe CPU offload incompatibility." }, { "left": "issue:33290", "right": "issue:34689", "accept": false, - "reason": "Optimizer-related OOM and a model-loading regression for Llama 3.2 Vision are unrelated." + "reason": "DeepSpeed/AdaFactor OOM vs model-loading regression for Llama 3.2 Vision." }, { "left": "issue:45588", "right": "issue:45600", "accept": false, - "reason": "Flash-attention integration crash and removed auto_mappings references are different CI/runtime problems." + "reason": "Flash-attention sink-less crash vs stale auto_mappings references; different subsystems." }, { "left": "issue:33290", "right": "issue:35141", "accept": false, - "reason": "OOM during optimizer use and embedding reinitialization after resize are unrelated." + "reason": "Optimizer OOM vs output-embedding reinitialization after resizing token embeddings." + }, + { + "left": "issue:43262", + "right": "issue:44206", + "accept": false, + "reason": "Audio chat-template sampling-rate default vs unsupported center arg in feature extractor." + }, + { + "left": "issue:44977", + "right": "issue:45245", + "accept": false, + "reason": "Qwen3.5 flash-attention generation issue vs category-count limit runtime error." + }, + { + "left": "issue:33290", + "right": "issue:34567", + "accept": false, + "reason": "OOM in DeepSpeed/AdaFactor vs TrainerState token counter not updating." + }, + { + "left": "issue:43381", + "right": "issue:43873", + "accept": false, + "reason": "Gradient checkpointing eval-mode restriction and quantization offloading are distinct behaviors." + }, + { + "left": "issue:44464", + "right": "issue:45491", + "accept": false, + "reason": "Compiled-forward chunked generation inconsistency vs Gemma3 sliding-window NaNs." + }, + { + "left": "issue:43646", + "right": "issue:44077", + "accept": false, + "reason": "Custom model initialization breakage in v5 vs patchtsmixer post_init API restriction." + }, + { + "left": "issue:44743", + "right": "issue:44811", + "accept": false, + "reason": "Qwen3 recurrent-state reset vs Whisper batch_decode skip_special_tokens bug." + }, + { + "left": "issue:44442", + "right": "issue:44898", + "accept": false, + "reason": "Tokenizer loading failure vs Perceiver image-size/interpolation failure; unrelated." + }, + { + "left": "issue:42490", + "right": "issue:44743", + "accept": false, + "reason": "Local path import warning and Qwen3 recurrent-state reset are unrelated issues." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The assessment is grounded in the packet and stays conservative on the soft edges. The only accepted pair is the clearly generic Flash Attention 2 title match (issue:45250 with issue:46640), while the other pairwise rejections are reasoned from distinct models, subsystems, or failure modes rather than overstating duplicate evidence." - }, - "evaluator_used": true, + "evaluator_result": null, + "evaluator_used": false, "retried": false, - "accepted_nontrivial_soft_edge": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -9875,20 +9941,21 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-469", "nodes": [ - "issue:33290", - "issue:34567", + "issue:25251", + "issue:28282", "issue:34689", "issue:42175", - "issue:42490", + "issue:42617", + "issue:42907", "issue:43232", - "issue:43262", "issue:43316", - "issue:43381", "issue:43388", "issue:43493", "issue:43525", + "issue:43575", "issue:43611", - "issue:43646", + "issue:43643", + "issue:43701", "issue:43720", "issue:43723", "issue:43749", @@ -9900,40 +9967,30 @@ "issue:44077", "issue:44079", "issue:44188", - "issue:44206", + "issue:44190", "issue:44442", "issue:44464", "issue:44479", - "issue:44743", - "issue:44811", + "issue:44496", "issue:44855", "issue:44871", - "issue:44898", - "issue:44977", + "issue:44913", "issue:44987", "issue:45003", "issue:45137", "issue:45216", - "issue:45245", "issue:45440", "issue:45479", - "issue:45491", + "issue:45698", "issue:45759", "issue:45803", "issue:45907", "issue:46178", - "issue:46234" + "issue:46234", + "issue:46421", + "issue:46506" ], "soft_pairs": [ - "issue:43262|issue:44206", - "issue:44977|issue:45245", - "issue:33290|issue:34567", - "issue:43381|issue:43873", - "issue:44464|issue:45491", - "issue:43646|issue:44077", - "issue:44743|issue:44811", - "issue:44442|issue:44898", - "issue:42490|issue:44743", "issue:34689|issue:43611", "issue:45003|issue:45440", "issue:43720|issue:44442", @@ -9946,210 +10003,205 @@ "issue:43493|issue:43827", "issue:43761|issue:44188", "issue:43873|issue:44871", - "issue:44987|issue:46178", "issue:43232|issue:43388", "issue:43749|issue:43931", + "issue:44987|issue:46178", "issue:43931|issue:45216", - "issue:43316|issue:45759", "issue:44442|issue:45803", - "issue:44855|issue:45137" + "issue:43316|issue:45759", + "issue:44855|issue:45137", + "issue:43643|issue:44913", + "issue:42907|issue:45698", + "issue:43701|issue:46506", + "issue:44190|issue:44496", + "issue:44464|issue:46421", + "issue:42617|issue:43575", + "issue:25251|issue:28282" ], - "prepared_review_unit_hash": "d903613c8a8f9f4cc1f045b465a209a9b1f50040eb863fee7e6c98d9fcceba30", + "prepared_review_unit_hash": "dcc1efa9a0ff8ec757c1ad2fe3e28f3b7f196f2f2f175a0b114312bb3a409401", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 28, - "serialized_chars": 36320, - "estimated_input_tokens": 9080, - "estimated_eval_tokens": 18416 + "soft_pair_count": 26, + "serialized_chars": 35636, + "estimated_input_tokens": 8909, + "estimated_eval_tokens": 18074 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose collection of unrelated Transformers issues with only superficial title similarity. I would reject every soft edge as a duplicate and treat one broad v5 loading-regression issue as the cluster representative.", - "confidence": 0.93, - "canonical_issue_reason": "issue:43611 is the broadest, most representative issue in the set: a core Transformers 5.0.0 model-loading regression with a clear, general failure mode.", + "summary": "These issues span many distinct bugs across loading, configs, generation, datasets, and docs. Despite superficial subsystem overlap, none of the soft pairs look like the same underlying bug or change, so all should be rejected as duplicates.", + "confidence": 0.92, + "canonical_issue_reason": "Issue 43873 is the broadest and most actively discussed open regression in the set, so it is the least-specific representative if a canonical must be chosen. That said, the cluster is heterogeneous and has no strong true canonical.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43611 is the best overall issue to stand in for the cluster because it captures a high-level, reusable regression pattern rather than a narrow model-specific symptom.", + "best_issue_reason": "43873 is the best-fit issue only in a weak, administrative sense: it is open, high-activity, and broadly impactful. There is no stronger cluster-wide representative among these unrelated reports.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43262", - "right": "issue:44206", - "accept": false, - "reason": "Audio chat-template sample-rate handling and a LasrFeatureExtractor unsupported-arg crash are different code paths and different failure modes." - }, - { - "left": "issue:44977", - "right": "issue:45245", - "accept": false, - "reason": "Flash-attention generation failure for Qwen3.5 is unrelated to the 2^24 category-limit error." - }, - { - "left": "issue:33290", - "right": "issue:34567", - "accept": false, - "reason": "DeepSpeed Adafactor OOM and TrainerState token-count tracking are unrelated training issues." - }, - { - "left": "issue:43381", - "right": "issue:43873", - "accept": false, - "reason": "Gradient checkpointing in eval mode and quantization offloading are different subsystems and bugs." - }, - { - "left": "issue:44464", - "right": "issue:45491", - "accept": false, - "reason": "Compiled chunked generation inconsistency is not the same bug as Gemma3 NaNs from mixed-length sliding-window batching." - }, - { - "left": "issue:43646", - "right": "issue:44077", - "accept": false, - "reason": "Custom model initialization breakage is broader and different from the specific patchtsmixer post_init policy issue." - }, - { - "left": "issue:44743", - "right": "issue:44811", - "accept": false, - "reason": "Recurrent-state reset with cache is unrelated to Whisper batch_decode ignoring skip_special_tokens." - }, - { - "left": "issue:44442", - "right": "issue:44898", - "accept": false, - "reason": "Tokenizer loading failure and Perceiver interpolation failure affect different components and code paths." - }, - { - "left": "issue:42490", - "right": "issue:44743", - "accept": false, - "reason": "Local-path import warning has nothing to do with Qwen3.5 recurrent-state reset behavior." - }, { "left": "issue:34689", "right": "issue:43611", "accept": false, - "reason": "These are separate model-loading regressions in different versions with different root causes and symptoms." + "reason": "Both are model-loading regressions, but they describe different failures, models, and likely different code paths." }, { "left": "issue:45003", "right": "issue:45440", "accept": false, - "reason": "Unsafe sys.modules access is a generic import bug, not the same as DeepSeekV3MoE divergence from remote implementation." + "reason": "`sys.modules` access safety and DeepSeekV3 parity are unrelated bugs with different root causes." }, { "left": "issue:43720", "right": "issue:44442", "accept": false, - "reason": "BitNet packed-weight loading and FastSpeech2 tokenizer loading are unrelated failures." + "reason": "One is BitNet accelerate-weight loading, the other is tokenizer loading; these are separate components." }, { "left": "issue:43316", "right": "issue:43525", "accept": false, - "reason": "Gemma3TextConfig API mismatch and Llama4Config missing pad_token_id are separate config-schema issues." + "reason": "Both are config API discrepancies, but for different models and different missing fields, not the same bug." }, { "left": "issue:42175", "right": "issue:46234", "accept": false, - "reason": "Missing TensorFlow in extras installation is unrelated to a missing import_utils availability helper." + "reason": "Package extra/backend inclusion and a missing import_utils availability helper are different issues." }, { "left": "issue:43723", "right": "issue:45479", "accept": false, - "reason": "Tokenizer loading in v5 and zero-loss sequence-classification behavior are different bugs with different affected paths." + "reason": "Tokenizer loading in v5 and sequence-classification loss behavior are unrelated." }, { "left": "issue:44062", "right": "issue:45907", "accept": false, - "reason": "AddedToken special-argument duplication is unrelated to list_repo_files not forwarding revision." + "reason": "AddedToken keyword duplication and list_repo_files revision forwarding are unrelated code paths." }, { "left": "issue:43761", "right": "issue:44079", "accept": false, - "reason": "CLIPVisionModel hidden_states regression is not clearly the same underlying bug as ModelOutput key reassignment." + "reason": "One is CLIPVision hidden_states output, the other is generic ModelOutput key assignment; too different to merge." }, { "left": "issue:44077", "right": "issue:44479", "accept": false, - "reason": "patchtsmixer post_init validation and Qwen video-input regression are distinct model-specific issues." + "reason": "PatchTSMixer post_init validation and Qwen video-input regression are separate model-specific bugs." }, { "left": "issue:43493", "right": "issue:43827", "accept": false, - "reason": "SigLIP2 implementation drift and docs still referencing pipeline() are unrelated." + "reason": "Model implementation parity and docs still referencing pipeline() are unrelated." }, { "left": "issue:43761", "right": "issue:44188", "accept": false, - "reason": "A CLIPVisionModel output bug is unrelated to divergent attention kernels under torch.compile." + "reason": "Hidden-state output regression and attention-kernel divergence under torch.compile are different problems." }, { "left": "issue:43873", "right": "issue:44871", "accept": false, - "reason": "Quantization/offloading behavior and Gemma3 eos_token_id config mismatch do not share the same bug." - }, - { - "left": "issue:44987", - "right": "issue:46178", - "accept": false, - "reason": "A model-loading regression for physical-intelligence/fast is unrelated to missing tensorflow_text backend mapping." + "reason": "Quantization/offloading behavior and Gemma eos_token_id configuration mismatch are unrelated." }, { "left": "issue:43232", "right": "issue:43388", "accept": false, - "reason": "Generation kwargs update after sync_gpus and gather_for_metrics label truncation are different execution paths." + "reason": "Generation kwargs handling and gather_for_metrics label dropping are different trainer/generation bugs." }, { "left": "issue:43749", "right": "issue:43931", "accept": false, - "reason": "FSDP CPU RAM efficient loading and Qwen3-VL weight-shape mismatch are not the same failure." + "reason": "FSDP CPU RAM-efficient loading and Qwen3-VL shape mismatch are both loading-related but have different failure modes and causes." }, { - "left": "issue:43931", - "right": "issue:45216", + "left": "issue:44987", + "right": "issue:46178", "accept": false, - "reason": "Model weight-shape mismatch on load is unrelated to a broken save_pretrained checkpoint regression." + "reason": "A model-loading regression and a backend mapping error for tensorflow_text are not the same issue." }, { - "left": "issue:43316", - "right": "issue:45759", + "left": "issue:43931", + "right": "issue:45216", "accept": false, - "reason": "A config API discrepancy is not the same as AutoModelForCausalLM.from_config failing to unwrap text_config." + "reason": "Both involve Qwen model persistence/loading, but one is a weight-shape mismatch and the other is a save_pretrained regression." }, { "left": "issue:44442", "right": "issue:45803", "accept": false, - "reason": "Tokenizer loading failure and Fuyu batch-feature exception handling are unrelated bugs." + "reason": "Tokenizer loading vs Fuyu batch feature tensor conversion error; different functionality and symptoms." + }, + { + "left": "issue:43316", + "right": "issue:45759", + "accept": false, + "reason": "Both touch config handling, but one is a field discrepancy and the other is unwrapping text_config for composite multimodal configs." }, { "left": "issue:44855", "right": "issue:45137", "accept": false, - "reason": "Python 3.13 IndentationError in torch.jit.script parsing and a DeepSpeed ZeRO3 deque underflow are unrelated." + "reason": "Python 3.13 torch.jit parsing and DeepSpeed ZeRO3 deque underflow are unrelated." + }, + { + "left": "issue:43643", + "right": "issue:44913", + "accept": false, + "reason": "Remote-code config field loss and GPTNeoX rotary_pct not persisting are both config-related but not the same concrete bug." + }, + { + "left": "issue:42907", + "right": "issue:45698", + "accept": false, + "reason": "Saving dequantized models and loading the wrong custom module after save_pretrained are different save/load failures." + }, + { + "left": "issue:43701", + "right": "issue:46506", + "accept": false, + "reason": "Checkpoint key mismatch and CPU/CUDA output differences are unrelated." + }, + { + "left": "issue:44190", + "right": "issue:44496", + "accept": false, + "reason": "Dataset loading in a training script and missing model_type in config.json are unrelated." + }, + { + "left": "issue:44464", + "right": "issue:46421", + "accept": false, + "reason": "Both mention chunked generation/prefill, but one is output inconsistency and the other is cache lazy-init recompilation; not the same bug." + }, + { + "left": "issue:42617", + "right": "issue:43575", + "accept": false, + "reason": "A parallel script failure and a tensor-parallel OOM on a specific model are different distributed-runtime issues." + }, + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Pipeline top_k nesting behavior and a PyTorch ImportError are completely unrelated." } ] }, @@ -10175,32 +10227,33 @@ "nodes": [ "issue:16998", "issue:22355", - "issue:25251", "issue:27957", - "issue:28282", "issue:32090", "issue:42175", - "issue:42617", "issue:42796", - "issue:42907", "issue:43065", "issue:43295", "issue:43316", + "issue:43352", + "issue:43404", + "issue:43441", "issue:43452", + "issue:43540", "issue:43550", "issue:43575", "issue:43606", - "issue:43643", "issue:43646", + "issue:43650", "issue:43701", "issue:43716", "issue:43742", + "issue:43761", "issue:43825", "issue:44077", "issue:44162", - "issue:44190", - "issue:44464", + "issue:44479", "issue:44496", + "issue:44514", "issue:44534", "issue:44610", "issue:44617", @@ -10210,28 +10263,20 @@ "issue:45290", "issue:45584", "issue:45632", - "issue:45698", "issue:45803", "issue:46050", "issue:46170", + "issue:46202", "issue:46234", "issue:46291", "issue:46365", "issue:46369", - "issue:46421", "issue:46424", - "issue:46506" + "issue:46640" ], "soft_pairs": [ - "issue:43643|issue:44913", - "issue:42907|issue:45698", - "issue:43701|issue:46506", - "issue:44190|issue:44496", - "issue:44464|issue:46421", - "issue:42617|issue:43575", - "issue:25251|issue:28282", - "issue:43550|issue:43701", "issue:27957|issue:32090", + "issue:43550|issue:43701", "issue:46234|issue:46291", "issue:43575|issue:43742", "issue:43606|issue:43701", @@ -10239,8 +10284,8 @@ "issue:42796|issue:45290", "issue:16998|issue:22355", "issue:43295|issue:44871", - "issue:43316|issue:43716", "issue:44871|issue:45584", + "issue:43316|issue:43716", "issue:42175|issue:43316", "issue:45803|issue:46365", "issue:44610|issue:46369", @@ -10251,210 +10296,203 @@ "issue:45803|issue:46170", "issue:44162|issue:44792", "issue:43452|issue:44496", - "issue:43065|issue:44617" + "issue:43065|issue:44617", + "issue:43650|issue:46202", + "issue:43404|issue:43441", + "issue:43352|issue:46640", + "issue:43540|issue:44514", + "issue:43761|issue:44479" ], - "prepared_review_unit_hash": "ff0f29d3112b3fc8f7fa711a9516314f581c1e8317afccc3e6bac75ec448cd45", + "prepared_review_unit_hash": "f0a2ce758fce7b43043e439006f3c7343f599fd7c28c4a553794e310d79422bc", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36059, - "estimated_input_tokens": 9015, - "estimated_eval_tokens": 18286 + "soft_pair_count": 27, + "serialized_chars": 35439, + "estimated_input_tokens": 8860, + "estimated_eval_tokens": 17976 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose collection of unrelated Transformers bug reports; the only weak commonality is around loading/saving/config regressions. No soft pair looks like the same underlying issue, so all are rejected.", - "confidence": 0.39, - "canonical_issue_reason": "Issue #45698 is the best representative because it is the most actively discussed, broadly user-facing save/load regression in the set: custom modules are mis-selected after save_pretrained/from_pretrained. It is also open, so it fits ongoing triage better than the narrower closed reports.", + "summary": "The cluster is a loose mix of mostly unrelated Transformers issues that only share superficial subsystem or keyword overlap; none of the soft pairs look like true duplicates, so no pair should be merged.", + "confidence": 0.92, + "canonical_issue_reason": "issue:43825 is the clearest self-contained, user-facing regression in the set, with explicit expected behavior and inbound references, so it is the best single representative if one must be chosen.", "canonical_pr_reason": null, - "best_issue_reason": "#45698 is the strongest single issue to anchor the cluster: it has the highest apparent user impact and discussion, and its problem statement is concrete enough to serve as the canonical save/load custom-module regression.", + "best_issue_reason": "issue:43825 is the strongest global issue candidate because it is concrete, well-scoped, and clearly describes a reproducible API regression.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43643", - "right": "issue:44913", - "accept": false, - "reason": "Both are config-related, but one is a trust_remote_code field-loss bug and the other is a GPTNeoX rotary_pct reload bug; different models and different failure modes." - }, - { - "left": "issue:42907", - "right": "issue:45698", - "accept": false, - "reason": "Both involve save/load flows, but one is about failing to save dequantized models and the other is about loading the wrong custom module after save_pretrained; not the same bug." - }, - { - "left": "issue:43701", - "right": "issue:46506", - "accept": false, - "reason": "Unrelated issues: checkpoint key mismatch versus CPU/CUDA output differences in PPDocLayoutV3." - }, - { - "left": "issue:44190", - "right": "issue:44496", - "accept": false, - "reason": "Different subsystems and symptoms: local dataset loading in an example script versus model type recognition during from_pretrained." - }, - { - "left": "issue:44464", - "right": "issue:46421", - "accept": false, - "reason": "Both mention chunking/cache, but one is inconsistent generation under compiled forward and the other is spurious recompilation from lazy cache initialization; not the same code-path problem." - }, - { - "left": "issue:42617", - "right": "issue:43575", - "accept": false, - "reason": "Both are parallelism-related, but one is a script failure and the other is tensor-parallel OOM when loading a specific model." - }, - { - "left": "issue:25251", - "right": "issue:28282", + "left": "issue:27957", + "right": "issue:32090", "accept": false, - "reason": "Completely different problems: pipeline output nesting versus missing PyTorch dependency for AutoModel." + "reason": "Different failures: Flash Attention support in XLM-Roberta vs Trainer GPU broadcast NoneType error." }, { "left": "issue:43550", "right": "issue:43701", "accept": false, - "reason": "Different failures: torch.compile/SDPA on Bamba versus resume_from_checkpoint key mismatch." - }, - { - "left": "issue:27957", - "right": "issue:32090", - "accept": false, - "reason": "Flash Attention 2 support on XLMRoberta is unrelated to a Trainer broadcast NoneType error." + "reason": "Unrelated bugs: torch.compile/SDPA failure vs resume_from_checkpoint key mismatch." }, { "left": "issue:46234", "right": "issue:46291", "accept": false, - "reason": "Both are import/dependency issues, but one is a missing helper in import_utils and the other is kernels v0.15 requiring version/revision; distinct root causes." + "reason": "Both involve imports, but one is a missing utility symbol and the other is a kernels package version requirement." }, { "left": "issue:43575", "right": "issue:43742", "accept": false, - "reason": "Loading OOM for Qwen2 with tensor parallelism is not the same as a key error loading MobileLLM-125M." + "reason": "Different loading problems: tensor-parallel OOM vs key error loading a small model." }, { "left": "issue:43606", "right": "issue:43701", "accept": false, - "reason": "CPU offload device mismatch for bark-small is unrelated to checkpoint key mismatches." + "reason": "CPU offload device mismatch is unrelated to checkpoint resume key mapping." }, { "left": "issue:43295", "right": "issue:43825", "accept": false, - "reason": "A regression in processor/tokenizer behavior is not the same as an incorrect v5 pipeline error message about translation tasks." + "reason": "Different regressions: processor.tokenizer/images handling vs a pipeline error message about translation tasks." }, { "left": "issue:42796", "right": "issue:45290", "accept": false, - "reason": "Both involve tokenization, but one is a tokenizer BPE-marker handling bug and the other is chat template crashes on assistant tool-call messages." + "reason": "Tokenizer BPE-marker handling and chat-template crashes are distinct code paths and symptoms." }, { "left": "issue:16998", "right": "issue:22355", "accept": false, - "reason": "A DeBERTa model_max_length question is unrelated to a transformers.onnx import error." + "reason": "Model max length question is unrelated to the missing transformers.onnx module import error." }, { "left": "issue:43295", "right": "issue:44871", "accept": false, - "reason": "Both are regressions around processor/model config, but one is about processor.tokenizer and image passing, while the other is eos_token_id inconsistency for Gemma-3." + "reason": "One is a processor/tokenizer regression; the other is an eos_token_id config inconsistency." }, { - "left": "issue:43316", - "right": "issue:43716", + "left": "issue:44871", + "right": "issue:45584", "accept": false, - "reason": "API discrepancy in Gemma3TextConfig is unrelated to a dtype mismatch between image preprocessor and model." + "reason": "Gemma eos_token_id mismatch and Whisper empty-transcription failure are different bugs." }, { - "left": "issue:44871", - "right": "issue:45584", + "left": "issue:43316", + "right": "issue:43716", "accept": false, - "reason": "Different models and bugs: Gemma-3 eos_token_id mismatch versus Whisper empty-transcription generation failure." + "reason": "API discrepancy in config classes is unrelated to image preprocessor dtype mismatch." }, { "left": "issue:42175", "right": "issue:43316", "accept": false, - "reason": "Packaging/backend dependency mismatch is unrelated to a config API discrepancy." + "reason": "Packaging/backend dependency issue does not match a config API discrepancy." }, { "left": "issue:45803", "right": "issue:46365", "accept": false, - "reason": "Bare except swallowing errors in FuyuBatchFeature.convert_to_tensors is unrelated to Trainer import issues caused by datasets." + "reason": "Bare except swallowing errors in Fuyu conversion is unrelated to Trainer import failure with datasets." }, { "left": "issue:44610", "right": "issue:46369", "accept": false, - "reason": "Processor input size mismatch for OmDet-Turbo is not the same as Sam3-LiteText memory overflow." + "reason": "Processor image-size mismatch and Sam3LiteText memory overflow are different model bugs." }, { "left": "issue:44077", "right": "issue:44534", "accept": false, - "reason": "Optional post_init policy and non-persistent buffer corruption are distinct model-initialization versus state-buffer bugs." + "reason": "Optional post_init enforcement and junk non-persistent buffers are unrelated v5 issues." }, { "left": "issue:43646", "right": "issue:44913", "accept": false, - "reason": "Both are regressions, but custom model initialization breakage is not the same as rotary_pct resetting on reload." + "reason": "Generic custom model initialization breakage is not the same as GPTNeoX rotary_pct reload regression." }, { "left": "issue:44871", "right": "issue:46050", "accept": false, - "reason": "Both touch save/load or config persistence, but the Gemma-3 eos token mismatch is different from quantized sinq models breaking after save_pretrained/from_pretrained." + "reason": "Gemma eos_token_id inconsistency and sinq quantization roundtrip failure are separate serialization/loading problems." }, { "left": "issue:45632", "right": "issue:46424", "accept": false, - "reason": "Trust-remote-code cache path collisions are unrelated to max_cache_length being ignored outside Quantized Cache." + "reason": "Cache-path collisions and max_cache_length behavior are different cache-related issues." }, { "left": "issue:45803", "right": "issue:46170", "accept": false, - "reason": "Swallowed exceptions in FuyuBatchFeature conversion are unrelated to Windows/ROCm import failures involving torch.distributed." + "reason": "Fuyu tensor conversion exception handling is unrelated to Windows/ROCm torch.distributed import errors." }, { "left": "issue:44162", "right": "issue:44792", "accept": false, - "reason": "ESM2 breakage and janus image-generation test failures are unrelated model-specific issues." + "reason": "ESM2 model breakage and janus image-generation test failure are unrelated." }, { "left": "issue:43452", "right": "issue:44496", "accept": false, - "reason": "Both concern loading models, but gguf_file handling and missing model_type are different loading paths with different failure modes." + "reason": "gguf_file from_pretrained breakage is a different loading path from a missing model_type config error." }, { "left": "issue:43065", "right": "issue:44617", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder and Sam3Video CUDA OOM are not the same bug; one is architecture stub logic, the other is memory usage." + "reason": "Dummy Conv2d in Sam3PixelDecoder and Sam3Video OOM are not the same underlying problem." + }, + { + "left": "issue:43650", + "right": "issue:46202", + "accept": false, + "reason": "Placeholder issue title with no technical detail is unrelated to Gemma pipeline roundtrip behavior." + }, + { + "left": "issue:43404", + "right": "issue:43441", + "accept": false, + "reason": "Weight tying in Mistral3 is unrelated to FlashAttention support in Ministral-3." + }, + { + "left": "issue:43352", + "right": "issue:46640", + "accept": false, + "reason": "A specific model lacking Flash Attention 2 support is not enough to match a vague generic flash-attention issue." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Video-input ValueError in Qwen3OmniMoe is unrelated to batched chat-template padding crashes." + }, + { + "left": "issue:43761", + "right": "issue:44479", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and Qwen video-input regression affect different models and code paths." } ] }, @@ -10481,27 +10519,26 @@ "issue:35141", "issue:41950", "issue:42490", - "issue:43352", + "issue:43065", "issue:43388", - "issue:43404", - "issue:43441", "issue:43475", "issue:43504", "issue:43525", "issue:43531", - "issue:43540", "issue:43572", - "issue:43650", + "issue:43644", "issue:43673", "issue:43688", "issue:43761", + "issue:43827", "issue:43931", + "issue:43994", "issue:44060", "issue:44190", "issue:44242", "issue:44479", - "issue:44514", "issue:44534", + "issue:44610", "issue:44617", "issue:44734", "issue:44743", @@ -10515,24 +10552,19 @@ "issue:45381", "issue:45406", "issue:45542", + "issue:45563", "issue:45584", "issue:45588", "issue:45600", "issue:45632", "issue:45698", "issue:46178", - "issue:46202", "issue:46234", "issue:46369", - "issue:46617", - "issue:46640" + "issue:46421", + "issue:46617" ], "soft_pairs": [ - "issue:43650|issue:46202", - "issue:43352|issue:46640", - "issue:43404|issue:43441", - "issue:43540|issue:44514", - "issue:43761|issue:44479", "issue:41950|issue:43475", "issue:43388|issue:43688", "issue:43572|issue:46178", @@ -10540,198 +10572,210 @@ "issue:45542|issue:46234", "issue:43504|issue:45698", "issue:43673|issue:44743", - "issue:44792|issue:44977", "issue:44792|issue:45600", + "issue:44792|issue:44977", "issue:43525|issue:43531", "issue:44534|issue:44913", - "issue:44734|issue:45406", "issue:44617|issue:45584", + "issue:44734|issue:45406", "issue:44190|issue:44964", "issue:35141|issue:42490", "issue:45632|issue:46617", "issue:43931|issue:44060", "issue:45381|issue:46369", "issue:44855|issue:45588", - "issue:43761|issue:44242" + "issue:43761|issue:44242", + "issue:43673|issue:46421", + "issue:43827|issue:43994", + "issue:45563|issue:45584", + "issue:44479|issue:44734", + "issue:43644|issue:44913", + "issue:43065|issue:44610" ], - "prepared_review_unit_hash": "32d604d2f91a86af2c016ae0e30cc2c798fe7fd8894b14d5d44d0857c5b807e4", + "prepared_review_unit_hash": "a645066301604423111e5871037bf312c94ef8f96cf08724f3db4bf646c73c4f", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 25, - "serialized_chars": 35239, - "estimated_input_tokens": 8810, - "estimated_eval_tokens": 17876 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35043, + "estimated_input_tokens": 8761, + "estimated_eval_tokens": 17778 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly unrelated issues that only share broad subsystem words (cache, flash attention, video, loading, etc.). I would not collapse any of the proposed soft pairs; the closest cluster theme is trust_remote_code/cache loading, with 46617 the best single anchor if one must be chosen.", - "confidence": 0.94, - "canonical_issue_reason": "No single true duplicate group emerges, but issue 46617 is the broadest, most current cache-loading failure and the best representative anchor among the list.", + "summary": "This cluster is mostly a grab-bag of unrelated Transformers issues; the soft pairs share broad vocabulary (cache, loading, vision, v5 regressions) but not the same underlying bug, so I rejected all proposed duplicates. I used a broad, user-facing loading regression as the representative issue.", + "confidence": 0.38, + "canonical_issue_reason": "issue:45698 is the broadest representative here: it is a high-discussion, user-facing loading regression around from_pretrained/save_pretrained and custom modules, which makes it a better cluster anchor than the many narrow model-specific bugs.", "canonical_pr_reason": null, - "best_issue_reason": "46617 is the most general user-facing failure mode here and is broad enough to serve as the cluster anchor; the others are mostly model-specific or unrelated edge cases.", + "best_issue_reason": "issue:45698 is the most suitable single issue to represent the cluster because it is open, broadly impactful, and has the strongest evidence of recurring user impact (discussion/inbound refs) compared with the mostly isolated model- or version-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43650", - "right": "issue:46202", - "accept": false, - "reason": "Unrelated: one is a nonspecific placeholder issue, the other is a Gemma pipeline roundtrip/Unicode token bug." - }, - { - "left": "issue:43352", - "right": "issue:46640", - "accept": false, - "reason": "Too vague to merge safely: both mention Flash Attention 2, but one is a specific unsupported-model error and the other is an unspecific generic report." - }, - { - "left": "issue:43404", - "right": "issue:43441", - "accept": false, - "reason": "Different failures: tied lm_head weights in Mistral3 vs FlashAttention support failure in Ministral-3." - }, - { - "left": "issue:43540", - "right": "issue:44514", - "accept": false, - "reason": "Different code paths: Qwen3OmniMoe video-input processing vs Qwen2_5_VL batched chat template padding behavior." - }, - { - "left": "issue:43761", - "right": "issue:44479", - "accept": false, - "reason": "Unrelated regressions: CLIPVision hidden states vs a Qwen video-input regression across several models." - }, { "left": "issue:41950", "right": "issue:43475", "accept": false, - "reason": "Different subsystems: video-classification processor selection vs SAM3 video encoder attribute access." + "reason": "Both are vision/video-adjacent, but one is a pipeline processor selection bug and the other is a missing attribute in Sam3 video outputs; different code paths." }, { "left": "issue:43388", "right": "issue:43688", "accept": false, - "reason": "Different bugs: last-batch metric gathering drops tuple labels vs auxiliary-loss normalization in MoE models." + "reason": "Metric gathering drops labels in the last batch; the other is auxiliary-loss normalization. Same training context, different bug." }, { "left": "issue:43572", "right": "issue:46178", "accept": false, - "reason": "Only loosely related via config/backend naming; one is a StableLm config field regression, the other is a backend-mapping error for tensorflow_text." + "reason": "StableLm config missing pad_token_idx is unrelated to a missing tensorflow_text backend mapping." }, { "left": "issue:45127", "right": "issue:45200", "accept": false, - "reason": "Not the same bug: LoRA/extended-vocab collapse with tied embeddings vs Gemma4 token-type ID defaults." + "reason": "LoRA merge / tied embeddings corruption is a different problem from Gemma 4 mm_token_type_ids defaulting." }, { "left": "issue:45542", "right": "issue:46234", "accept": false, - "reason": "Different issues: TensorBoard-only TensorFlow backend handling vs a missing import_utils availability helper." + "reason": "Undefined tf backend due to tensorboard-only install is not the same as a missing import_utils availability helper." }, { "left": "issue:43504", "right": "issue:45698", "accept": false, - "reason": "Both are loading-related, but one is a legacy-field preset load failure and the other is a wrong custom-module reload after save_pretrained." + "reason": "Legacy BEiT preset field failure is a model-config loading issue, but not the same as loading the wrong custom module after save_pretrained." }, { "left": "issue:43673", "right": "issue:44743", "accept": false, - "reason": "Both mention cache/state, but the concrete bugs differ: missing GenerationMixin cache in chunked prefill vs recurrent states resetting with seq_len>1." + "reason": "Both mention cache, but one is chunked_prefill cache initialization and the other is recurrent state reset in a specific modular model." }, { "left": "issue:44792", - "right": "issue:44977", + "right": "issue:45600", "accept": false, - "reason": "Unrelated model behaviors: a failed janus image-generation test vs Qwen3.5 flash-attention generation issues." + "reason": "Janus image-generation test failure and stale auto_mappings entries are unrelated." }, { "left": "issue:44792", - "right": "issue:45600", + "right": "issue:44977", "accept": false, - "reason": "Different problems: janus generation test failure vs broken auto_mappings references to removed Sam3LiteText configs." + "reason": "Janus test failure is not the same as Qwen3.5 flash-attention generation misbehavior." }, { "left": "issue:43525", "right": "issue:43531", "accept": false, - "reason": "Different config/model bugs: missing pad_token_id in Llama4Config vs sliding_window behavior in Qwen3-MoE." + "reason": "Missing pad_token_id in Llama4Config and Qwen3-MoE sliding_window issues are different config bugs." }, { "left": "issue:44534", "right": "issue:44913", "accept": false, - "reason": "Different persistence bugs: non-persistent buffer junk in v5 vs GPTNeoXConfig rotary_pct not persisting on reload." + "reason": "Non-persistent buffer junk in v5 is unrelated to GPTNeoX rotary_pct not persisting on reload." }, { - "left": "issue:44734", - "right": "issue:45406", + "left": "issue:44617", + "right": "issue:45584", "accept": false, - "reason": "Different serve failures: KV-cache continuation indexing bug vs Gemma4Processor missing _tokenizer." + "reason": "Sam3Video OOM and Whisper empty-transcription failure are different model/runtime failures." }, { - "left": "issue:44617", - "right": "issue:45584", + "left": "issue:44734", + "right": "issue:45406", "accept": false, - "reason": "Completely different domains: SAM3 video OOM vs Whisper empty-transcription generation after token alignment." + "reason": "KV-cache tensor indexing in transformers serve is unrelated to Gemma4Processor missing _tokenizer." }, { "left": "issue:44190", "right": "issue:44964", "accept": false, - "reason": "Different loading paths: local dataset loading in an example script vs loading a Phi-4 multimodal model in transformers." + "reason": "Local dataset loading in a training script is unrelated to loading a multimodal checkpoint." }, { "left": "issue:35141", "right": "issue:42490", "accept": false, - "reason": "Different model initialization issues: output embedding reinit after resize vs import warning for local model paths ending with a slash." + "reason": "Token embedding resize reinit is unrelated to path-with-trailing-slash import warnings." }, { "left": "issue:45632", "right": "issue:46617", "accept": false, - "reason": "Related only at a high level: both involve trust_remote_code/cache behavior, but one is a leaf-directory cache collision and the other is a FileNotFoundError from cache loading." + "reason": "Both involve trust_remote_code cache usage, but one is cache-path collision and the other is FileNotFoundError from cache loading; not the same bug." }, { "left": "issue:43931", "right": "issue:44060", "accept": false, - "reason": "Different Qwen-family issues: weight-shape mismatch on load vs an incorrect tied-weights warning involving dt_bias." + "reason": "Qwen3-VL weight-shape mismatch and Qwen3-Next tied-weights warning point to different model/config problems." }, { "left": "issue:45381", "right": "issue:46369", "accept": false, - "reason": "Different problems: wrong video position IDs in qwen2.5-vl vs memory overflow in sam3-litetext." + "reason": "Wrong video position ids in Qwen2.5-VL is unrelated to Sam3-LiteText memory overflow." }, { "left": "issue:44855", "right": "issue:45588", "accept": false, - "reason": "Different failure modes: Python 3.13 import/IndentationError vs flash_attention.py crashing on s_aux=None." + "reason": "Python 3.13 import parsing failure in DebertaV2Model is unrelated to a flash_attention s_aux=None crash." }, { "left": "issue:43761", "right": "issue:44242", "accept": false, - "reason": "Different model behaviors: CLIPVision hidden-state regression vs load-balancing loss omission when output_router_logits=False." + "reason": "CLIP hidden_states regression and MoE load-balancing loss omission are different functional bugs." + }, + { + "left": "issue:43673", + "right": "issue:46421", + "accept": false, + "reason": "Both mention chunked prefill/cache, but one is a missing GenerationMixin cache and the other is spurious recompilation from lazy cache init." + }, + { + "left": "issue:43827", + "right": "issue:43994", + "accept": false, + "reason": "Docs still referencing pipeline() is unrelated to SigLIP2 producing bad outputs through AutoModel/pipeline." + }, + { + "left": "issue:45563", + "right": "issue:45584", + "accept": false, + "reason": "Paged generate warning about num_return_sequences is unrelated to Whisper empty transcription failure." + }, + { + "left": "issue:44479", + "right": "issue:44734", + "accept": false, + "reason": "Video-input regression across Qwen VL models is unrelated to KV-cache continuation tensor indexing in transformers serve." + }, + { + "left": "issue:43644", + "right": "issue:44913", + "accept": false, + "reason": "Non-persistent buffer junk in v5 and rotary_pct reset on reload are different serialization/regression issues." + }, + { + "left": "issue:43065", + "right": "issue:44610", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder is not the same as OmDet-Turbo processor/model resolution mismatch." } ] }, @@ -10757,35 +10801,34 @@ "nodes": [ "issue:25251", "issue:27957", + "issue:36246", + "issue:41628", "issue:41720", "issue:42796", - "issue:43065", "issue:43116", "issue:43295", "issue:43335", "issue:43540", "issue:43550", - "issue:43644", - "issue:43673", - "issue:43827", - "issue:43994", + "issue:43643", "issue:44077", "issue:44112", "issue:44190", + "issue:44246", "issue:44462", "issue:44464", - "issue:44479", + "issue:44496", "issue:44521", - "issue:44610", "issue:44617", "issue:44734", "issue:44737", "issue:44779", "issue:44855", "issue:44877", - "issue:44913", "issue:44933", "issue:44938", + "issue:44964", + "issue:45003", "issue:45127", "issue:45245", "issue:45276", @@ -10799,20 +10842,15 @@ "issue:45632", "issue:45803", "issue:45907", + "issue:46082", + "issue:46121", "issue:46234", "issue:46369", - "issue:46421", "issue:46489" ], "soft_pairs": [ - "issue:43673|issue:46421", - "issue:43827|issue:43994", - "issue:45563|issue:45584", - "issue:44479|issue:44734", - "issue:43065|issue:44610", - "issue:43644|issue:44913", - "issue:44933|issue:45362", "issue:25251|issue:27957", + "issue:44933|issue:45362", "issue:41720|issue:45245", "issue:43335|issue:44855", "issue:43550|issue:44190", @@ -10832,211 +10870,220 @@ "issue:45276|issue:45584", "issue:45632|issue:46234", "issue:44462|issue:45907", - "issue:42796|issue:44521" + "issue:42796|issue:44521", + "issue:44246|issue:45003", + "issue:36246|issue:41628", + "issue:46082|issue:46121", + "issue:44964|issue:45600", + "issue:45245|issue:45584", + "issue:43643|issue:44496" ], - "prepared_review_unit_hash": "0608074ee7c04647186c227dadaedfcfeb39c3d5504d3e0905a1476d60404a1c", + "prepared_review_unit_hash": "b00f853ac23f83dea1c1b24c12ea9fe08effb8b2729b5866819cefb1ec75f75d", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36004, - "estimated_input_tokens": 9001, - "estimated_eval_tokens": 18258 + "serialized_chars": 36203, + "estimated_input_tokens": 9051, + "estimated_eval_tokens": 18358 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is a grab bag of unrelated Transformers bug reports and regressions. The soft pairs mostly share broad subsystems (tokenizers, vision, generation, config, serving) but not the same concrete bug or fix path, so they should all stay separate.", - "confidence": 0.95, - "canonical_issue_reason": "issue:44521 is the strongest standalone issue in the set: it is open, has the most inbound references, and describes a clearly scoped multimodal chat-template bug with broad user impact.", + "summary": "This cluster is mostly heterogeneous transformer bug reports with only one clear duplicate-like pair. The only strong match is the DeepSeek tokenizer regression pair; the rest differ by model, subsystem, or failure mode.", + "confidence": 0.91, + "canonical_issue_reason": "issue:44779 is the broader DeepSeek tokenizer regression report, while issue:46489 appears to be a narrower follow-up about the same v5 tokenizer breakage and explicitly references a gap in PR #44801's fix.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 is the best representative issue because it has active discussion and a concrete, reusable failure description; the rest are mostly isolated model- or subsystem-specific reports.", + "best_issue_reason": "issue:44779 is the best representative because it states the underlying regression generically and covers the wider DeepSeek tokenizer failure surface.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43673", - "right": "issue:46421", - "accept": false, - "reason": "Both mention cache behavior during chunked prefill, but one is about missing GenerationMixin cache in v5 and the other about lazy cache init causing recompilation; different bugs and likely different fixes." - }, - { - "left": "issue:43827", - "right": "issue:43994", - "accept": false, - "reason": "Docs referencing removed pipeline() APIs is unrelated to SigLIP2 producing bad outputs; one is documentation cleanup, the other is model/pipeline behavior." - }, - { - "left": "issue:45563", - "right": "issue:45584", - "accept": false, - "reason": "Paged generate warning on num_return_sequences and Whisper empty-transcription failure are different generation edge cases with different triggers and code paths." - }, - { - "left": "issue:44479", - "right": "issue:44734", - "accept": false, - "reason": "Both involve video or serving regressions, but the affected paths differ: Qwen VL video vision_position_ids vs KV-cache continuation indexing in transformers serve." - }, - { - "left": "issue:43065", - "right": "issue:44610", - "accept": false, - "reason": "A dummy Conv2d placeholder in Sam3PixelDecoder and an OmDet-Turbo processor/model size mismatch are unrelated architecture and preprocessing issues." - }, - { - "left": "issue:43644", - "right": "issue:44913", + "left": "issue:25251", + "right": "issue:27957", "accept": false, - "reason": "Non-persistent buffer initialization in v5 and GPTNeoXConfig rotary_pct not persisting on reload are separate config/state bugs." + "reason": "Different problems: pipeline top_k output shape vs XLM-RoBERTa Flash Attention 2 support." }, { "left": "issue:44933", "right": "issue:45362", "accept": false, - "reason": "Missing import from image_utils and Qwen3.5-35B crashing in chat are not the same issue; one is API surface regression, the other is model runtime failure." - }, - { - "left": "issue:25251", - "right": "issue:27957", - "accept": false, - "reason": "Pipeline top_k changing output shape and XLMRoberta Flash Attention 2 support are unrelated; one is postprocessing shape semantics, the other is backend/model support." + "reason": "Unrelated: missing image_utils import vs Qwen3.5 chat crash." }, { "left": "issue:41720", "right": "issue:45245", "accept": false, - "reason": "Qwen3 auto device mapping cudaErrorAssert and category-count limit 2^24 are completely different failure modes and subsystems." + "reason": "Different failure modes and subsystems: Qwen3 device-mapping CUDA assert vs category-limit RuntimeError." }, { "left": "issue:43335", "right": "issue:44855", "accept": false, - "reason": "SwitchTransformers sparse-layer creation and DebertaV2 import indentation failure are unrelated config vs Python parsing problems." + "reason": "Different code paths: SwitchTransformers sparse-layer config bug vs DebertaV2 Python 3.13 import parsing issue." }, { "left": "issue:43550", "right": "issue:44190", "accept": false, - "reason": "Bamba torch.compile/SDPA failure and local dataset loading in the no-trainer image classification script are different bugs with no shared fix path." + "reason": "Unrelated: torch.compile/SDPA model failure vs example script dataset loading bug." }, { "left": "issue:44077", "right": "issue:44877", "accept": false, - "reason": "PatchTST config post_init policy and strict config blocking granite_speech loading are both config-related, but they address different model constraints and different checks." + "reason": "Both touch config handling, but the concrete bugs are different models and different validation failures." }, { "left": "issue:44464", "right": "issue:45563", "accept": false, - "reason": "Chunked generation inconsistency with compiled forward and a stale warning for paged generate are distinct generation behaviors; one is output correctness, the other is warning logic." + "reason": "Different generation issues: compiled forward inconsistency vs stale warning in paged generate." }, { "left": "issue:44617", "right": "issue:46369", "accept": false, - "reason": "Sam3Video OOM and sam3-litetext memory overflow are both memory issues, but they concern different models and likely different causes." + "reason": "Same broad model family, but different products and symptoms; not clearly the same underlying bug." }, { "left": "issue:43335", "right": "issue:44190", "accept": false, - "reason": "SwitchTransformers sparse-layer construction and local dataset loading are unrelated." + "reason": "Different subsystems: model config construction vs example-script dataset loading." }, { "left": "issue:45127", "right": "issue:45357", "accept": false, - "reason": "LoRA merge collapse with tied embeddings and Qwen3.5 visual encoder keys saved incorrectly are different model-state bugs, not one underlying problem." + "reason": "Different regression surfaces: LoRA vocab merging collapse vs incorrect visual encoder keys on save_pretrained." }, { "left": "issue:44521", "right": "issue:45381", "accept": false, - "reason": "All-zero assistant masks in apply_chat_template and wrong vision_position_ids for qwen2.5-vl video input are both multimodal, but they affect different outputs and code paths." + "reason": "Both multimodal-related, but assistant-mask generation and vision_position_ids are distinct bugs." }, { "left": "issue:44112", "right": "issue:45600", "accept": false, - "reason": "GraniteSpeech stale device override CI failure and auto_mappings.py referencing removed Sam3LiteText configs are separate CI/config maintenance issues." + "reason": "Both are CI/test maintenance issues, but they affect different model mappings and different failures." }, { "left": "issue:43116", "right": "issue:44190", "accept": false, - "reason": "Empty multi-label classification results in an example script and local dataset loading in another example are unrelated example-script problems." + "reason": "Different example scripts and different behaviors: multi-label outputs vs image-classification dataset loading." }, { "left": "issue:44779", "right": "issue:46489", - "accept": false, - "reason": "DeepSeek tokenizer regression in v5 and DeepSeek-Coder v1 tokenizer gap relative to PR #44801 are both tokenizer issues, but they concern different model families and fixes." + "accept": true, + "reason": "Same underlying DeepSeek tokenizer regression in v5; 46489 is a narrower report that points to the same breakage and references the prior fix gap." }, { "left": "issue:43540", "right": "issue:44734", "accept": false, - "reason": "Qwen3OmniMoe video-processing ValueError and serve KV-cache continuation indexing are different multimodal vs serving bugs." + "reason": "Different runtime paths: multimodal video input processing vs serve KV-cache continuation indexing." }, { "left": "issue:43295", "right": "issue:45276", "accept": false, - "reason": "Custom processor.tokenizer/images regression and gemma4 resize_token_embeddings not updating per-layer embeddings are separate API regressions with different affected objects." + "reason": "Both are regressions, but one is processor/tokenizer API breakage and the other is embedding resize propagation." }, { "left": "issue:44737", "right": "issue:45803", "accept": false, - "reason": "XLNet CPU arange bug and bare except in FuyuBatchFeature.convert_to_tensors are unrelated runtime and error-handling issues." + "reason": "Different bugs: XLNet device placement inefficiency vs Fuyu exception handling." }, { "left": "issue:44938", "right": "issue:45542", "accept": false, - "reason": "Python 3.14 load failure and tensorboard-installed-without-TensorFlow backend error are different environment/package issues." + "reason": "Python 3.14 import failure is unrelated to the TensorFlow/tensorboard backend mismatch." }, { "left": "issue:45276", "right": "issue:45584", "accept": false, - "reason": "Gemma4 embedding resize propagation and Whisper empty transcription failure are separate model-state vs generation-behavior bugs." + "reason": "Unrelated: token embedding resize propagation vs Whisper empty-transcription generation failure." }, { "left": "issue:45632", "right": "issue:46234", "accept": false, - "reason": "Local trust_remote_code cache collision and missing is_falcon_mamba_ssm_available import are unrelated cache-path vs utility-export issues." + "reason": "Different areas: trust_remote_code cache collision vs missing import_utils availability helper." }, { "left": "issue:44462", "right": "issue:45907", "accept": false, - "reason": "AutoTokenizer ignoring tokenizer.json and list_repo_files not forwarding revision are both repository-loading related, but they affect different APIs and failure modes." + "reason": "Tokenizer-file resolution bug vs list_repo_files revision forwarding are not the same underlying issue." }, { "left": "issue:42796", "right": "issue:44521", "accept": false, - "reason": "Tokenizer BPE-marker handling in Ministral and multimodal assistant masks are unrelated tokenizer vs chat-template bugs." + "reason": "Different tokenization/template problems affecting different modalities and APIs." + }, + { + "left": "issue:44246", + "right": "issue:45003", + "accept": false, + "reason": "Both relate to import-time behavior, but one is slowness and the other is unsafe sys.modules access." + }, + { + "left": "issue:36246", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but for different symbols and different model areas." + }, + { + "left": "issue:46082", + "right": "issue:46121", + "accept": false, + "reason": "Different config-validation bugs: Llama head_dim constraints vs rope validation list handling." + }, + { + "left": "issue:44964", + "right": "issue:45600", + "accept": false, + "reason": "Different failures: model loading for Phi-4 multimodal vs removed Sam3LiteText references in auto_mappings CI." + }, + { + "left": "issue:45245", + "right": "issue:45584", + "accept": false, + "reason": "Category-count runtime limit vs empty-transcription generation are unrelated." + }, + { + "left": "issue:43643", + "right": "issue:44496", + "accept": false, + "reason": "Related to config loading, but one is trust_remote_code field loss and the other is unrecognized model/config metadata." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The clustering looks grounded in the packet, and the duplicate call between issue:44779 and issue:46489 is conservative and well-supported by the titles. The other soft-edge rejections stay cautious and do not overclaim cross-issue equivalence." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -11058,59 +11105,54 @@ "issue:35141", "issue:36246", "issue:38617", - "issue:41628", "issue:41950", "issue:42175", "issue:42907", "issue:43295", + "issue:43381", "issue:43404", "issue:43441", "issue:43479", "issue:43526", "issue:43531", - "issue:43643", "issue:43650", "issue:43673", "issue:43854", "issue:43901", "issue:43976", "issue:44016", + "issue:44062", "issue:44112", "issue:44186", "issue:44220", - "issue:44246", "issue:44279", "issue:44295", + "issue:44336", "issue:44496", "issue:44683", "issue:44734", + "issue:44792", "issue:44964", "issue:44987", - "issue:45003", "issue:45042", "issue:45083", "issue:45216", - "issue:45245", "issue:45335", "issue:45375", "issue:45446", - "issue:45584", + "issue:45588", + "issue:45593", "issue:45600", - "issue:46082", - "issue:46121", + "issue:45759", + "issue:46123", + "issue:46369", "issue:46424", "issue:46506" ], "soft_pairs": [ - "issue:44246|issue:45003", - "issue:46082|issue:46121", - "issue:36246|issue:41628", - "issue:45245|issue:45584", - "issue:44964|issue:45600", - "issue:43643|issue:44496", "issue:42175|issue:44279", - "issue:44112|issue:44964", "issue:36246|issue:38617", + "issue:44112|issue:44964", "issue:43854|issue:44964", "issue:44683|issue:45446", "issue:27957|issue:35141", @@ -11129,204 +11171,231 @@ "issue:43976|issue:44987", "issue:43526|issue:43901", "issue:44496|issue:45042", - "issue:42907|issue:43441" + "issue:42907|issue:43441", + "issue:43901|issue:44496", + "issue:43381|issue:45593", + "issue:43673|issue:44336", + "issue:43976|issue:44279", + "issue:43441|issue:45588", + "issue:44016|issue:44062", + "issue:46123|issue:46369", + "issue:44792|issue:46424", + "issue:44496|issue:45759" ], - "prepared_review_unit_hash": "4f0d25ac9523d0f0fb3a0524d2a49e19852859291ea8bb53a4eca32dfb589151", + "prepared_review_unit_hash": "8f7103a18a80e7f95256b9b8f151542fefbfa825ea79066e96d4a8dd5486d512", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 28, - "serialized_chars": 35503, - "estimated_input_tokens": 8876, - "estimated_eval_tokens": 18008 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37044, + "estimated_input_tokens": 9261, + "estimated_eval_tokens": 18778 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The soft pairs mostly share only broad subsystem vocabulary, not the same concrete bug or fix. I rejected all of them; this set does not form a coherent duplicate cluster, and there are no PRs to nominate.", - "confidence": 0.96, - "canonical_issue_reason": "No single canonical issue fits here: the listed issues span many unrelated bugs, so picking one as the cluster representative would be misleading.", + "summary": "Loose similarity cluster with many unrelated issues; none of the soft edges look like true duplicates. The set spans different models, configs, pipelines, and runtime failures rather than one underlying bug.", + "confidence": 0.9, + "canonical_issue_reason": "No single canonical duplicate stands out because the cluster is heterogeneous. If a representative is needed, issue 44964 is the closest fit: it is a concrete, well-scoped model-loading regression with a detailed repro.", "canonical_pr_reason": null, - "best_issue_reason": "No best issue for deduplication: none of these issues is a clear umbrella report for the others, and the candidate pairs are not true duplicates.", + "best_issue_reason": "Issue 44964 is the best manual anchor in this set because it is specific, reproducible, and well described, but it does not unify the rest of the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44246", - "right": "issue:45003", + "left": "issue:42175", + "right": "issue:44279", "accept": false, - "reason": "Both concern import behavior, but one is about slow import and the other about unsafe sys.modules access; different bugs." + "reason": "TensorFlow backend packaging vs a generic dependency issue; different failure modes and likely different fixes." }, { - "left": "issue:46082", - "right": "issue:46121", + "left": "issue:36246", + "right": "issue:38617", "accept": false, - "reason": "These are different config-validation problems: Llama head_dim handling vs rope-parameter list conversion." + "reason": "Different missing imports from different modules (`qwen2_5_vl` vs `configuration_utils`); unrelated code paths." }, { - "left": "issue:36246", - "right": "issue:41628", + "left": "issue:44112", + "right": "issue:44964", "accept": false, - "reason": "Both are ImportErrors, but they involve different missing symbols and different modules/code paths." + "reason": "CI test flake in GraniteSpeech vs model loading failure for Phi-4 multimodal; no shared root bug." }, { - "left": "issue:45245", - "right": "issue:45584", + "left": "issue:43854", + "right": "issue:44964", "accept": false, - "reason": "Completely different failures: category-count limit vs Whisper empty-transcription generation." + "reason": "Both are model-loading reports, but for different models and different likely causes; not the same bug." }, { - "left": "issue:44964", - "right": "issue:45600", + "left": "issue:44683", + "right": "issue:45446", "accept": false, - "reason": "One is a model-loading failure; the other is stale auto_mappings references breaking CI. Same area, not the same bug." + "reason": "Torch version / compiled flex attention issue vs an incorrect version check for a different import path." }, { - "left": "issue:43643", - "right": "issue:44496", + "left": "issue:27957", + "right": "issue:35141", "accept": false, - "reason": "Different config-loading bugs: missing fields from trust_remote_code vs unrecognized model without model_type." - }, - { - "left": "issue:42175", - "right": "issue:44279", - "accept": false, - "reason": "Generic dependency complaints, but not the same concrete packaging or backend issue." - }, - { - "left": "issue:44112", - "right": "issue:44964", - "accept": false, - "reason": "A CI test flake in GraniteSpeech is unrelated to a Phi-4 multimodal model-loading failure." - }, - { - "left": "issue:36246", - "right": "issue:38617", - "accept": false, - "reason": "Both are import errors, but the missing names and affected modules are different." - }, - { - "left": "issue:43854", - "right": "issue:44964", - "accept": false, - "reason": "Different model-loading failures for different models; no shared code-path bug is evident." - }, - { - "left": "issue:44683", - "right": "issue:45446", - "accept": false, - "reason": "Both touch flex_attention, but one is a torch>=2.9 compiled path failure and the other is a version-check/import gate issue." - }, - { - "left": "issue:27957", - "right": "issue:35141", - "accept": false, - "reason": "Flash Attention support for XLMRoberta is unrelated to token-embedding resizing reinitialization." + "reason": "Flash Attention support for XLM-Roberta is unrelated to token embedding reinitialization after resize." }, { "left": "issue:43404", "right": "issue:45216", "accept": false, - "reason": "Different models and different weight-saving/tie-weight problems; they are not the same fix." + "reason": "Weight tying bug in Mistral3 vs checkpoint save regression in Qwen3.5; different model families and code paths." }, { "left": "issue:44295", "right": "issue:46506", "accept": false, - "reason": "Position_ids buffer access and CPU/CUDA output divergence are unrelated bugs." + "reason": "Buffer/position_ids access bug vs CPU/CUDA numerical divergence; unrelated symptoms." }, { "left": "issue:44186", "right": "issue:44734", "accept": false, - "reason": "Tokenizer padding/truncation crashes and serve KV-cache indexing crashes are different code paths." + "reason": "Tokenizer crash on NER padding/truncation vs serve KV-cache indexing bug; different layers and fixes." }, { "left": "issue:41950", "right": "issue:44016", "accept": false, - "reason": "A video-classification pipeline bug is unrelated to a notebook syntax error." + "reason": "Pipeline image-processor lookup bug vs a notebook syntax error; not the same underlying issue." }, { "left": "issue:43650", "right": "issue:44220", "accept": false, - "reason": "The title placeholder issue is unrelated to the audio feature-extraction bug." + "reason": "Vague request for data vs fbank feature extraction bug; no common failure." }, { "left": "issue:43673", "right": "issue:46424", "accept": false, - "reason": "Both involve cache, but one is missing GenerationMixin cache and the other is max_cache_length behavior for quantized cache; different defects." + "reason": "Chunked prefill cache regression vs max_cache_length only applying to Quantized Cache; related area but distinct behavior." }, { "left": "issue:32090", "right": "issue:35141", "accept": false, - "reason": "Trainer broadcast of NoneType is unrelated to post_init reinitializing embeddings." + "reason": "Trainer broadcast NoneType error vs output embedding reinitialization after resize; different subsystems." }, { "left": "issue:43479", "right": "issue:45375", "accept": false, - "reason": "Both are config-shape problems, but they affect different model families and different fields." + "reason": "Different multimodal config validation bugs on different fields/models; not mergeable as one fix." }, { "left": "issue:44016", "right": "issue:44220", "accept": false, - "reason": "Notebook syntax error and feature-extraction helper bug are unrelated." + "reason": "Notebook syntax error and feature extraction bug are unrelated." }, { "left": "issue:43531", "right": "issue:45600", "accept": false, - "reason": "A Qwen3-MoE sliding-window issue is unrelated to broken auto_mappings references." + "reason": "Qwen3-MoE sliding-window behavior vs stale auto_mappings references to removed configs; unrelated." }, { "left": "issue:45083", "right": "issue:46506", "accept": false, - "reason": "Different models and different symptoms: helper length calculation vs CPU/CUDA output mismatch." + "reason": "Helper length calculation in qwen3_omni_moe vs CPU/CUDA output mismatch; different bugs." }, { "left": "issue:43295", "right": "issue:45335", "accept": false, - "reason": "Both mention embedding/tokenizer behavior, but one is a processor/tokenizer regression and the other is decoder.embed_tokens not updating on resize." + "reason": "Processor/tokenizer regression vs `resize_token_embeddings` not affecting decoder embeddings; different code paths." }, { "left": "issue:43976", "right": "issue:44987", "accept": false, - "reason": "A Python version compatibility problem is unrelated to a specific model-loading regression." + "reason": "Python version compatibility issue vs loading a specific model failing on transformers>=5.1.0; not the same bug." }, { "left": "issue:43526", "right": "issue:43901", "accept": false, - "reason": "ImageProcessorFast label reduction and a documentation/behavior mismatch in text classification are not the same bug." + "reason": "BeitImageProcessorFast label handling vs docs mentioning `return_all_scores`; one is runtime behavior, the other documentation drift." }, { "left": "issue:44496", "right": "issue:45042", "accept": false, - "reason": "An unrecognized-model config error is unrelated to the PIL backend's torchvision dependency requirement." + "reason": "Unrecognized model config vs PIL image processors requiring torchvision; unrelated model-loading and dependency issues." }, { "left": "issue:42907", "right": "issue:43441", "accept": false, - "reason": "Same model family, but one report is about save_pretrained on dequantized models and the other is a FlashAttention failure; different fixes." + "reason": "Saving dequantized Ministrals/Devstrals vs FlashAttention failure in Ministral-3 RC; same family name but different concrete failures." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "Documentation mismatch vs unrecognized model config; no shared bug." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Gradient checkpointing in eval mode vs missing auxiliary losses when denoising is off; unrelated behavior." + }, + { + "left": "issue:43673", + "right": "issue:44336", + "accept": false, + "reason": "Cache handling bug vs ANSI codes emitted in loading report; different components entirely." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Python support regression vs generic dependency issue; no common fix." + }, + { + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "Both mention flash attention, but one is a model-specific RC failure and the other is an AttributeError on `s_aux=None`; not enough to be the same bug." + }, + { + "left": "issue:44016", + "right": "issue:44062", + "accept": false, + "reason": "Notebook syntax error vs tokenizer `AddedToken` argument conflict; unrelated." + }, + { + "left": "issue:46123", + "right": "issue:46369", + "accept": false, + "reason": "Partial-batch mask generation bug vs memory overflow in sam3-litetext; different symptoms and fixes." + }, + { + "left": "issue:44792", + "right": "issue:46424", + "accept": false, + "reason": "Janus image generation test failure vs cache max length semantics; unrelated." + }, + { + "left": "issue:44496", + "right": "issue:45759", + "accept": false, + "reason": "Unrecognized model type vs `from_config` not unwrapping `text_config`; both loading-related but distinct model/config bugs." } ] }, @@ -11356,59 +11425,49 @@ "issue:42796", "issue:43295", "issue:43299", - "issue:43381", - "issue:43441", + "issue:43479", + "issue:43525", "issue:43577", "issue:43650", - "issue:43673", + "issue:43653", "issue:43824", + "issue:43854", "issue:43901", "issue:43976", - "issue:44016", - "issue:44062", "issue:44075", "issue:44242", "issue:44279", - "issue:44336", "issue:44351", "issue:44368", - "issue:44496", + "issue:44466", + "issue:44521", "issue:44617", + "issue:44683", "issue:44779", - "issue:44792", "issue:44933", "issue:45020", "issue:45103", "issue:45200", "issue:45216", + "issue:45276", "issue:45278", + "issue:45325", "issue:45335", "issue:45357", "issue:45372", "issue:45405", + "issue:45542", "issue:45584", - "issue:45588", "issue:45593", - "issue:45759", + "issue:45812", "issue:45823", "issue:46082", - "issue:46123", - "issue:46369", + "issue:46209", "issue:46399", - "issue:46424", "issue:46512", "issue:46515" ], "soft_pairs": [ - "issue:43381|issue:45593", - "issue:43901|issue:44496", - "issue:43673|issue:44336", - "issue:43976|issue:44279", - "issue:43441|issue:45588", - "issue:44016|issue:44062", - "issue:46123|issue:46369", - "issue:44792|issue:46424", - "issue:44496|issue:45759", "issue:43824|issue:44933", "issue:43299|issue:45216", "issue:44279|issue:46515", @@ -11418,226 +11477,222 @@ "issue:44279|issue:45405", "issue:42796|issue:43650", "issue:43577|issue:45216", - "issue:45103|issue:45372", "issue:39692|issue:43901", "issue:43295|issue:44779", "issue:41950|issue:45020", + "issue:45103|issue:45372", "issue:44351|issue:45278", "issue:43299|issue:45357", "issue:43976|issue:45405", "issue:44242|issue:45200", "issue:38175|issue:44617", "issue:46082|issue:46512", - "issue:41950|issue:45823" + "issue:41950|issue:45823", + "issue:44779|issue:45812", + "issue:44683|issue:45542", + "issue:43299|issue:43854", + "issue:43479|issue:43525", + "issue:43650|issue:46209", + "issue:44466|issue:45276", + "issue:44521|issue:45325", + "issue:43653|issue:45593" ], - "prepared_review_unit_hash": "abbf1b3da85c34214a38c28eb1770104c9b3d3600dd2fd1ca28959ede87c543b", + "prepared_review_unit_hash": "acad1401fae057a8be4bf6f9303eb0c260c698c4623abec2a2271a58d683bb57", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36375, - "estimated_input_tokens": 9094, - "estimated_eval_tokens": 18444 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35558, + "estimated_input_tokens": 8890, + "estimated_eval_tokens": 18036 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is very loose: it mixes unrelated Transformers issues across model loading, tokenizer behavior, pipelines, docs, and environment/version regressions. I don\u2019t see any soft pair that looks like the same underlying defect.", - "confidence": 0.95, - "canonical_issue_reason": "No strong canonical duplicate emerges; issue 45278 is only the closest umbrella because it is the broadest post-upgrade import-regression report, but it does not actually subsume most of the cluster.", + "summary": "This cluster is heterogeneous: it mixes unrelated tokenizer, config, model-loading, pipeline, documentation, and dependency/version regression reports. None of the soft edges look like the same underlying bug, so no duplicate merge is justified.", + "confidence": 0.97, + "canonical_issue_reason": "44521 ('apply_chat_template returns all-zero assistant_masks for multimodal inputs') is the closest thing to a canonical issue because it is the most concrete, reproducible, and currently open bug in the set.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45278 is the most usable anchor for triage because it is the broadest and most general regression report in the set.", + "best_issue_reason": "44521 is the best issue to keep as the representative problem: narrow scope, clear failure mode, and actionable reproduction details.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43381", - "right": "issue:45593", - "accept": false, - "reason": "Different bugs: eval-mode gradient checkpointing vs missing auxiliary loss when denoising is off." - }, - { - "left": "issue:43901", - "right": "issue:44496", - "accept": false, - "reason": "Docs mismatch in a pipeline API vs a model config/import error; no shared code-path." - }, - { - "left": "issue:43673", - "right": "issue:44336", - "accept": false, - "reason": "Generation cache regression vs ANSI code output in loading_report; unrelated subsystems." - }, - { - "left": "issue:43976", - "right": "issue:44279", - "accept": false, - "reason": "Python version compatibility issue vs a generic dependency complaint; not the same bug." - }, - { - "left": "issue:43441", - "right": "issue:45588", - "accept": false, - "reason": "Both mention flash attention, but they describe different failures and different code paths." - }, - { - "left": "issue:44016", - "right": "issue:44062", - "accept": false, - "reason": "Notebook syntax error vs tokenizer AddedToken kwarg conflict; unrelated." - }, - { - "left": "issue:46123", - "right": "issue:46369", - "accept": false, - "reason": "Partial-batch handling bug vs SAM3-LiteText memory overflow; different symptoms and causes." - }, - { - "left": "issue:44792", - "right": "issue:46424", - "accept": false, - "reason": "Janus image-generation test failure vs cache max-length semantics; not the same issue." - }, - { - "left": "issue:44496", - "right": "issue:45759", - "accept": false, - "reason": "Different model-loading failures for different model families and different root causes." - }, { "left": "issue:43824", "right": "issue:44933", "accept": false, - "reason": "Distinct import problems for different symbols/modules; not the same missing import." + "reason": "Both are import-related, but they fail on different symbols and different code paths: Qwen2.5-VL class import vs a bad image_utils import." }, { "left": "issue:43299", "right": "issue:45216", "accept": false, - "reason": "Qwen3VL loading regression vs Qwen3.5 save_pretrained checkpoint corruption; separate defects." + "reason": "Both affect Qwen models, but one is a model-loading failure for Qwen3VL MoE and the other is an incorrect save_pretrained checkpoint for Qwen3.5." }, { "left": "issue:44279", "right": "issue:46515", "accept": false, - "reason": "Generic dependency issue vs torch 2.4.1 incompatibility; too broad and unrelated." + "reason": "These are unrelated dependency/version complaints: a generic transformers dependency issue vs torch 2.4.1 incompatibility." }, { "left": "issue:45335", "right": "issue:45584", "accept": false, - "reason": "Embedding resize bug vs Whisper empty-transcription generation bug; different models and paths." + "reason": "Different models and different bugs: token embedding resize not propagating in t5gemma vs Whisper generation failure after align_special_tokens." }, { "left": "issue:44075", "right": "issue:44368", "accept": false, - "reason": "Optimizer argument handling vs a config warning about tied embeddings; not duplicates." + "reason": "Optimizer SGD args being ignored is unrelated to a Qwen3.5 LoRA warning about tie_word_embeddings." }, { "left": "issue:46082", "right": "issue:46399", "accept": false, - "reason": "Config validation for Llama head_dim vs Qwen3.5 kernel registration failure; unrelated." + "reason": "LlamaConfig head_dim validation and Qwen3.5 rotary-kernel registration are separate config/code-path bugs." }, { - "left": "issue:45103", - "right": "issue:45372", + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "A generic dependency issue is not the same as bumping MIN_PEFT_VERSION to an unreleased release." + }, + { + "left": "issue:42796", + "right": "issue:43650", + "accept": false, + "reason": "Tokenizer BPE-marker handling and an extremely vague 'ADD THE DATA' report do not describe the same bug." + }, + { + "left": "issue:43577", + "right": "issue:45216", "accept": false, - "reason": "Docstring parser crash vs missing ReasoningEffort import breaking Gemma 4 processor loading." + "reason": "Blip2 dtype propagation during loading and Qwen3.5 checkpoint serialization are different regression points." }, { "left": "issue:39692", "right": "issue:43901", "accept": false, - "reason": "Different documentation problems: SigLIP2 example errors vs return_all_scores docs mismatch." + "reason": "Both mention docs, but one is a SigLIP2 example with multiple runtime errors while the other is a TextClassificationPipeline docs mismatch." }, { "left": "issue:43295", "right": "issue:44779", "accept": false, - "reason": "Both involve tokenizers, but the affected model paths and failure modes are different." + "reason": "Both are tokenizer-related regressions, but they involve different model families and distinct failure modes." }, { "left": "issue:41950", "right": "issue:45020", "accept": false, - "reason": "Video-classification processor lookup bug vs remote_code regression; not the same underlying change." + "reason": "A video-classification pipeline processor lookup bug is unrelated to remote_code model breakage." + }, + { + "left": "issue:45103", + "right": "issue:45372", + "accept": false, + "reason": "auto_docstring crashing on future annotations and Gemma 4 processor loading failing on a mistral_common import are separate issues." }, { "left": "issue:44351", "right": "issue:45278", "accept": false, - "reason": "45278 is a broad umbrella for many import errors, but 44351 is just one specific missing symbol; not the same concrete bug." + "reason": "A missing HybridCache import and broad import failures after upgrade are related only at a very high level, not the same concrete bug." }, { "left": "issue:43299", "right": "issue:45357", "accept": false, - "reason": "Both are Qwen3.5-related, but one is a load regression and the other is a save_pretrained key bug." + "reason": "Both are Qwen3.5/Qwen3VL regressions, but one is model loading and the other is save_pretrained writing wrong visual encoder keys." }, { "left": "issue:43976", "right": "issue:45405", "accept": false, - "reason": "Python support regression vs PEFT version bump; unrelated." + "reason": "Python version compatibility and an unreleased PEFT version bump are unrelated release/dependency problems." }, { "left": "issue:44242", "right": "issue:45200", "accept": false, - "reason": "Load-balancing loss omission vs mm_token_type_ids defaulting issue; different model logic." + "reason": "Missing load-balancing loss when output_router_logits=False is not the same as mm_token_type_ids defaulting to zeros." }, { "left": "issue:38175", "right": "issue:44617", "accept": false, - "reason": "SigLIP probability anomaly vs SAM3Video CUDA OOM; unrelated model failures." + "reason": "SigLIP2 zero probabilities and Sam3Video CUDA OOM are completely different model/runtime problems." }, { "left": "issue:46082", "right": "issue:46512", "accept": false, - "reason": "Config validation error vs stale model.config.dtype after from_config; different code paths." + "reason": "Config head_dim validation and stale model.config.dtype after from_config(dtype=...) are separate config issues." }, { "left": "issue:41950", "right": "issue:45823", "accept": false, - "reason": "Video pipeline processor lookup bug vs Gemma4 device-map mismatch; not duplicates." + "reason": "A pipeline processor lookup bug and Gemma4 PLE device mismatch are not the same underlying defect." }, { - "left": "issue:42796", - "right": "issue:43650", + "left": "issue:44779", + "right": "issue:45812", "accept": false, - "reason": "Specific tokenizer BPE-marker bug vs an unrelated low-information issue title." + "reason": "Both are tokenizer regressions, but they affect different model families and do not clearly share the same concrete code-path bug." }, { - "left": "issue:43577", - "right": "issue:45216", + "left": "issue:44683", + "right": "issue:45542", "accept": false, - "reason": "Blip2 dtype propagation bug vs Qwen3.5 save_pretrained regression; different models and behavior." + "reason": "Compiled flex_attention on torch>=2.9 and a missing TensorFlow backend check are unrelated." }, { - "left": "issue:44279", - "right": "issue:45405", + "left": "issue:43299", + "right": "issue:43854", "accept": false, - "reason": "Generic dependency issue vs an explicit unreleased PEFT version bump; no shared defect." + "reason": "Both mention Qwen VL models, but one is a load failure and the other is an import-name failure for a different class." }, { - "left": "issue:44279", - "right": "issue:46515", + "left": "issue:43479", + "right": "issue:43525", + "accept": false, + "reason": "Phi4MultimodalConfig default-init behavior and Llama4Config missing pad_token_id are different configuration bugs." + }, + { + "left": "issue:43650", + "right": "issue:46209", + "accept": false, + "reason": "The first issue is too vague to match, and the second is a specific FP8 inference failure on B200." + }, + { + "left": "issue:44466", + "right": "issue:45276", + "accept": false, + "reason": "Both involve weights/embeddings, but one is tied-weight serialization varying by device and the other is resize_token_embeddings not updating Gemma4 embeddings." + }, + { + "left": "issue:44521", + "right": "issue:45325", + "accept": false, + "reason": "Both are multimodal, but assistant mask generation and rope-index temporal scaling are distinct preprocessing bugs." + }, + { + "left": "issue:43653", + "right": "issue:45593", "accept": false, - "reason": "Dependency complaint vs torch incompatibility; too generic to be a duplicate." + "reason": "BigBirdTokenizer special-token registration and D-FINE auxiliary-loss gating are unrelated." } ] }, @@ -11664,39 +11719,39 @@ "issue:33290", "issue:42175", "issue:42994", - "issue:43299", "issue:43317", "issue:43441", - "issue:43479", "issue:43525", - "issue:43650", - "issue:43653", + "issue:43643", "issue:43673", "issue:43723", "issue:43761", - "issue:43854", + "issue:43784", + "issue:43824", "issue:43881", "issue:43901", "issue:43976", "issue:43994", "issue:44016", "issue:44162", + "issue:44188", "issue:44246", "issue:44261", "issue:44263", "issue:44291", + "issue:44297", + "issue:44336", "issue:44361", + "issue:44373", "issue:44462", - "issue:44466", - "issue:44521", + "issue:44589", "issue:44623", "issue:44671", "issue:44683", "issue:44749", "issue:44779", "issue:45081", - "issue:45276", - "issue:45325", + "issue:45362", "issue:45397", "issue:45542", "issue:45593", @@ -11704,29 +11759,21 @@ "issue:45656", "issue:45712", "issue:45800", - "issue:45812", "issue:45823", "issue:45834", "issue:45907", - "issue:46209" + "issue:46291", + "issue:46399" ], "soft_pairs": [ - "issue:44779|issue:45812", - "issue:44683|issue:45542", - "issue:43299|issue:43854", - "issue:43650|issue:46209", - "issue:43479|issue:43525", - "issue:44466|issue:45276", - "issue:44521|issue:45325", - "issue:43653|issue:45593", "issue:44016|issue:44291", "issue:44462|issue:45081", "issue:43441|issue:45834", "issue:42994|issue:43317", "issue:33290|issue:45656", "issue:44361|issue:45593", - "issue:44261|issue:44263", "issue:44623|issue:44749", + "issue:44261|issue:44263", "issue:45600|issue:45712", "issue:45081|issue:45907", "issue:44623|issue:45397", @@ -11737,212 +11784,220 @@ "issue:44016|issue:45823", "issue:43441|issue:44683", "issue:43723|issue:44162", - "issue:43673|issue:43761", "issue:43901|issue:44246", - "issue:43525|issue:43881" + "issue:43673|issue:43761", + "issue:43525|issue:43881", + "issue:46291|issue:46399", + "issue:43824|issue:43976", + "issue:43525|issue:43643", + "issue:44188|issue:44336", + "issue:44589|issue:45362", + "issue:44297|issue:45397", + "issue:43784|issue:45542", + "issue:44336|issue:44373" ], - "prepared_review_unit_hash": "e91dda31c618d150bfce020a03550ee6cbe12d21627fcc80607b0d2084bf9075", + "prepared_review_unit_hash": "cebb9a70916ea6c317f7bfbd6857f4df7ede0c0c082ceb6663d8057068353c59", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36251, - "estimated_input_tokens": 9063, - "estimated_eval_tokens": 18382 + "serialized_chars": 36254, + "estimated_input_tokens": 9064, + "estimated_eval_tokens": 18384 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is over-broad and mixes many unrelated v5 regressions: tokenizer/output bugs, model-loading failures, config defaults, attention-kernel issues, packaging/backend problems, docs, and CI cleanup. None of the soft pairs look like the same underlying bug/change.", - "confidence": 0.94, - "canonical_issue_reason": "No single canonical issue: the items are too heterogeneous to represent one underlying bug.", + "summary": "This cluster is a loose collection of mostly unrelated Transformers issues across loading, tokenization, attention, packaging, docs, and model-specific regressions. None of the soft pairs appears to be the same underlying bug/change, so no duplicate merges are justified.", + "confidence": 0.91, + "canonical_issue_reason": "No single issue is a clear duplicate hub for this set; if one representative must be chosen, issue 45656 is the clearest current, actionable bug report.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45812 is the broadest representative if one must be chosen: it describes a cross-model tokenizer regression with clear v4\u2192v5 impact and multiple references.", + "best_issue_reason": "Issue 45656 is the best representative because it is a specific, reproducible, currently open deepspeed bug, while most other items are orthogonal or less central.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44779", - "right": "issue:45812", - "accept": false, - "reason": "Both are tokenizer regressions, but they affect different model families and likely different code paths." - }, - { - "left": "issue:44683", - "right": "issue:45542", - "accept": false, - "reason": "Unrelated subsystems: flex_attention/torch kernel failure versus TensorFlow backend packaging error." - }, - { - "left": "issue:43299", - "right": "issue:43854", - "accept": false, - "reason": "Both are model-loading failures, but for different models and symptoms; no clear shared root cause." - }, - { - "left": "issue:43650", - "right": "issue:46209", - "accept": false, - "reason": "Completely unrelated: a vague data request versus an FP8 inference failure on B200." - }, - { - "left": "issue:43479", - "right": "issue:43525", - "accept": false, - "reason": "Different models and different config bugs: default nested configs versus missing pad_token_id." - }, - { - "left": "issue:44466", - "right": "issue:45276", - "accept": false, - "reason": "Different failures: serialization of tied lm_head weights versus resize_token_embeddings not propagating." - }, - { - "left": "issue:44521", - "right": "issue:45325", - "accept": false, - "reason": "Both involve multimodal/VL logic, but one is chat-template masks and the other is rope/position-id scaling." - }, - { - "left": "issue:43653", - "right": "issue:45593", - "accept": false, - "reason": "Tokenizer special-token registration bug versus auxiliary-loss logic in D-FINE; unrelated." - }, { "left": "issue:44016", "right": "issue:44291", "accept": false, - "reason": "Notebook syntax error is unrelated to init_empty_weights/_is_hf_initialized loading failure." + "reason": "Notebook syntax error vs init_empty_weights TypeError: different code paths and symptoms." }, { "left": "issue:44462", "right": "issue:45081", "accept": false, - "reason": "Both concern tokenizer loading, but one ignores tokenizer.json and the other crashes in a Mistral regex patch." + "reason": "AutoTokenizer ignoring tokenizer.json is unrelated to Mistral regex patching crash." }, { "left": "issue:43441", "right": "issue:45834", "accept": false, - "reason": "Different models and different bugs: FlashAttention failure versus long-ocr index error." + "reason": "Different models and failures: FlashAttention regression vs OCR long-input index error." }, { "left": "issue:42994", "right": "issue:43317", "accept": false, - "reason": "Quantized save failure and dequantized load/offload failure are distinct code paths and symptoms." + "reason": "Both involve quantization, but save failure and dequantized-load/offload failure are different bugs." }, { "left": "issue:33290", "right": "issue:45656", "accept": false, - "reason": "Both mention deepspeed/optimizer behavior, but one is OOM and the other is double optimizer-step invocation." + "reason": "OOM with Adafactor under deepspeed is not the same as optimizer.step being called twice." }, { "left": "issue:44361", "right": "issue:45593", "accept": false, - "reason": "Tokenizer AttributeError on tasks versus D-FINE auxiliary-loss behavior; unrelated." + "reason": "Tokenizer AttributeError vs missing auxiliary losses in a vision model." }, { - "left": "issue:44261", - "right": "issue:44263", + "left": "issue:44623", + "right": "issue:44749", "accept": false, - "reason": "Different MLAs/GLM issues: missing rms_norm_eps precision drift versus torch.split return-value handling." + "reason": "Processor save_pretrained missing files is unrelated to data filtering slowdown." }, { - "left": "issue:44623", - "right": "issue:44749", + "left": "issue:44261", + "right": "issue:44263", "accept": false, - "reason": "Saving missing files is unrelated to a 10x preprocessing slowdown regression." + "reason": "Different GLM-related internals: layernorm config precision issue vs torch.split return values." }, { "left": "issue:45600", "right": "issue:45712", "accept": false, - "reason": "Both are CI/repo hygiene issues, but one is stale auto_mappings references and the other is leftover dummy classes." + "reason": "Both are repo-maintenance issues, but they affect different files and failure modes." }, { "left": "issue:45081", "right": "issue:45907", "accept": false, - "reason": "Tokenizer regex crash versus repo file listing revision forwarding; unrelated." + "reason": "Mistral tokenizer regex crash is unrelated to list_repo_files revision forwarding." }, { "left": "issue:44623", "right": "issue:45397", "accept": false, - "reason": "Processor save_pretrained missing files is unrelated to Gemma-4 zero3 from_pretrained loading." + "reason": "Generic processor save bug does not match gemma-4 zero3 loading failure." }, { "left": "issue:42175", "right": "issue:43976", "accept": false, - "reason": "Packaging/backend dependency issue versus Python version compatibility; different problems." + "reason": "Packaging missing TensorFlow backend vs Python version incompatibility are distinct environment issues." }, { "left": "issue:45542", "right": "issue:45800", "accept": false, - "reason": "TensorBoard-without-TensorFlow backend error is unrelated to torch/transformers version incompatibility." + "reason": "Missing tf backend from partial install is not the same as torch/transformers version incompatibility." }, { "left": "issue:43901", "right": "issue:43994", "accept": false, - "reason": "Docs mismatch for return_all_scores versus a model/pipeline inference regression." + "reason": "Documentation mismatch for return_all_scores is unrelated to incorrect SigLIP2 predictions." }, { "left": "issue:44671", "right": "issue:44779", "accept": false, - "reason": "Both are v5 regressions in model outputs, but they concern different models and likely different fixes." + "reason": "CamemBERT masked-LM wrong outputs and DeepSeek tokenizer regression are different subsystems." }, { "left": "issue:44016", "right": "issue:45823", "accept": false, - "reason": "Notebook syntax error is unrelated to Gemma4 device_map forward device mismatch." + "reason": "Notebook syntax error and Gemma4 PLE device mismatch are unrelated." }, { "left": "issue:43441", "right": "issue:44683", "accept": false, - "reason": "Both are attention-related, but one is FlashAttention model support and the other is compiled flex_attention on new torch." + "reason": "FlashAttention/model loading issue is not the same as compiled flex_attention failing on newer torch." }, { "left": "issue:43723", "right": "issue:44162", "accept": false, - "reason": "Tokenizer loading regression versus ESM2 model breakage; too different to be the same bug." + "reason": "Tokenizer loading regression and ESM2 model breakage are separate bugs." + }, + { + "left": "issue:43901", + "right": "issue:44246", + "accept": false, + "reason": "Docs issue and slow import issue do not share the same root cause." }, { "left": "issue:43673", "right": "issue:43761", "accept": false, - "reason": "Generation cache missing in chunked prefill versus CLIPVision hidden_states behavior; unrelated code paths." + "reason": "Chunked prefill cache regression and CLIP hidden_states bug affect different paths." }, { - "left": "issue:43901", - "right": "issue:44246", + "left": "issue:43525", + "right": "issue:43881", + "accept": false, + "reason": "Different model loading failures; one is pad_token_id missing, the other is a GLM4V load problem." + }, + { + "left": "issue:46291", + "right": "issue:46399", "accept": false, - "reason": "Docs mismatch versus import-time slowness; no shared underlying issue." + "reason": "Kernels revision/version requirement failure is not the same as missing registration decorator in apply_rotary_pos_emb." + }, + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "Qwen2.5-VL import error and Python version incompatibility are unrelated." }, { "left": "issue:43525", - "right": "issue:43881", + "right": "issue:43643", + "accept": false, + "reason": "Llama4Config pad_token_id error and trust_remote_code missing fields are different API bugs." + }, + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "Torch.compile attention-kernel divergence is unrelated to ANSI codes in loading_report." + }, + { + "left": "issue:44589", + "right": "issue:45362", + "accept": false, + "reason": "Missing Float8 storage support and Qwen3.5 chat crash are different failure modes." + }, + { + "left": "issue:44297", + "right": "issue:45397", + "accept": false, + "reason": "Tokenizer save metadata mismatch is unrelated to gemma-4 zero3 loading." + }, + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "sentence-transformers import NameError and TensorFlow backend install issue are not the same bug." + }, + { + "left": "issue:44336", + "right": "issue:44373", "accept": false, - "reason": "Different model configs and different failure modes: missing pad_token_id versus GLM-4V loading failure." + "reason": "ANSI color output bug and wrong docstring are unrelated." } ] }, @@ -11972,58 +12027,50 @@ "issue:42491", "issue:42757", "issue:42886", + "issue:43257", + "issue:43262", "issue:43299", "issue:43475", - "issue:43525", "issue:43526", "issue:43531", "issue:43550", - "issue:43643", - "issue:43784", - "issue:43824", + "issue:43596", + "issue:43673", "issue:43901", "issue:43976", - "issue:44188", + "issue:44162", "issue:44220", "issue:44242", "issue:44246", "issue:44295", - "issue:44297", "issue:44336", "issue:44351", - "issue:44373", "issue:44485", "issue:44496", - "issue:44589", + "issue:44561", "issue:44734", + "issue:44749", "issue:44779", + "issue:44857", "issue:44871", "issue:45003", "issue:45103", + "issue:45137", + "issue:45200", "issue:45292", "issue:45362", "issue:45397", - "issue:45542", "issue:45561", "issue:45563", "issue:45593", "issue:45859", "issue:46132", - "issue:46291", "issue:46369", "issue:46399", "issue:46424", "issue:46506" ], "soft_pairs": [ - "issue:46291|issue:46399", - "issue:43824|issue:43976", - "issue:43525|issue:43643", - "issue:44188|issue:44336", - "issue:44589|issue:45362", - "issue:44297|issue:45397", - "issue:43784|issue:45542", - "issue:44336|issue:44373", "issue:41628|issue:43976", "issue:44246|issue:44496", "issue:45561|issue:45563", @@ -12038,237 +12085,248 @@ "issue:41950|issue:44220", "issue:43550|issue:46506", "issue:36246|issue:42757", + "issue:36246|issue:44351", "issue:44336|issue:46424", "issue:42886|issue:46132", - "issue:36246|issue:44351", "issue:44779|issue:45292", "issue:44734|issue:46424", - "issue:45003|issue:45561", "issue:45859|issue:46399", + "issue:45003|issue:45561", "issue:43901|issue:45003", - "issue:43475|issue:46369" + "issue:43475|issue:46369", + "issue:43257|issue:45137", + "issue:44242|issue:44857", + "issue:36246|issue:43596", + "issue:44295|issue:44749", + "issue:44496|issue:45003", + "issue:43673|issue:44561", + "issue:44162|issue:45200", + "issue:43262|issue:44246" ], - "prepared_review_unit_hash": "e05a676017856ffda366c0622548b0ed4501894c1f4579c72a26e76150ac1a49", + "prepared_review_unit_hash": "588b074d9afce8e533688590c90d8856d29a09ec989fa72619039350270de465", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36899, - "estimated_input_tokens": 9225, - "estimated_eval_tokens": 18706 + "serialized_chars": 36905, + "estimated_input_tokens": 9227, + "estimated_eval_tokens": 18710 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: most items are unrelated import/export regressions, model-specific runtime bugs, cache/offline issues, docs, or infrastructure problems. I do not see a single canonical issue or any convincing duplicate PR pair among the soft candidates.", - "confidence": 0.96, - "canonical_issue_reason": null, + "summary": "This cluster is mostly heterogeneous Transformers bugs, import errors, and model-specific regressions. I only see one plausible duplicate pocket: the Qwen3-MoE loading breakage around the hf4.x\u2192hf5.x / v5 transition.", + "confidence": 0.81, + "canonical_issue_reason": "No single canonical issue fits the whole cluster; 42491 is the closest anchor because it is the broadest Qwen3-MoE loading regression report and best matches the only likely duplicate pocket.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "42491 is the best representative issue for the cluster's one near-duplicate area (Qwen3-MoE loading incompatibility on the new branch/version).", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:46291", - "right": "issue:46399", - "accept": false, - "reason": "Both mention kernels, but one is a pipelines import/dependency failure and the other is a Qwen3.5 model-specific registration crash; different code paths and fixes." - }, - { - "left": "issue:43824", - "right": "issue:43976", - "accept": false, - "reason": "One is a missing Qwen2.5-VL export, the other is a Python-version compatibility problem; not the same bug." - }, - { - "left": "issue:43525", - "right": "issue:43643", - "accept": false, - "reason": "Both involve missing fields, but one is Llama4Config pad_token_id and the other is trust_remote_code returning incomplete AutoConfig data; distinct issues." - }, - { - "left": "issue:44188", - "right": "issue:44336", - "accept": false, - "reason": "Attention kernel divergence and ANSI codes in loading_report are unrelated subsystems and failures." - }, - { - "left": "issue:44589", - "right": "issue:45362", - "accept": false, - "reason": "Storage-object lookup failure and Qwen3.5 chat crash are unrelated runtime problems." - }, - { - "left": "issue:44297", - "right": "issue:45397", - "accept": false, - "reason": "Tokenizer save metadata mismatch vs gemma-4 zero3 from_pretrained are different behaviors and code paths." - }, - { - "left": "issue:43784", - "right": "issue:45542", - "accept": false, - "reason": "Sentence-transformers import NameError and undefined tf backend due tensorboard-only install are unrelated import/environment bugs." - }, - { - "left": "issue:44336", - "right": "issue:44373", - "accept": false, - "reason": "Loading-report ANSI output and a wrong docstring for position_ids are unrelated." - }, { "left": "issue:41628", "right": "issue:43976", "accept": false, - "reason": "Missing AutoImageProcessor export and a Python 3.9+/3.10+ support issue are different problems." + "reason": "Both are import-related, but one is a missing AutoImageProcessor import and the other is a Python-version/package-compatibility failure; not the same bug." }, { - "left": "issue:44242", + "left": "issue:44246", "right": "issue:44496", "accept": false, - "reason": "Missing load-balancing loss in MoE training and an unrecognized model/config error are unrelated." + "reason": "Import slowdown and unrecognized-model/config parsing are unrelated failure modes." }, { "left": "issue:45561", "right": "issue:45563", "accept": false, - "reason": "A pytest-xdist file race and a stale warning in paged generate() are different test/runtime issues." + "reason": "A pytest-xdist file race and a stale generation warning are unrelated issues." }, { "left": "issue:45103", "right": "issue:46399", "accept": false, - "reason": "Auto-docstring crash with future annotations and a Qwen3.5 kernel-registration error do not share the same failure path." + "reason": "Auto-docstring annotation handling and Qwen3.5 kernel registration are different code paths and symptoms." }, { "left": "issue:44242", "right": "issue:45593", "accept": false, - "reason": "Both mention missing auxiliary losses, but they are on different models and mechanisms; not the same concrete bug." + "reason": "Both mention missing losses, but they are different model families and different auxiliary-loss mechanisms." }, { "left": "issue:42491", "right": "issue:43299", - "accept": false, - "reason": "Both are Qwen3 MoE-related, but one is LoRA compatibility across branches and the other is a Qwen3VL MoE loading regression; too different to merge." + "accept": true, + "reason": "Both report Qwen3-MoE loading incompatibility after the hf4.x/main \u2192 hf5.x/v5 transition; they look like the same regression." }, { "left": "issue:44295", "right": "issue:45397", "accept": false, - "reason": "Position_ids buffer access error and gemma-4 zero3 from_pretrained failure are unrelated." + "reason": "A position_ids buffer error and a Gemma-4 ZeRO3 from_pretrained failure are not the same concrete bug." }, { "left": "issue:43526", "right": "issue:45003", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug and unsafe sys.modules access in modeling_utils are unrelated." + "reason": "A BeitImageProcessorFast label reduction bug and unsafe sys.modules access are unrelated." }, { "left": "issue:43531", "right": "issue:45362", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior and Qwen3.5 chat crash are different model-path bugs." + "reason": "Both are Qwen3-related, but one is a sliding_window issue and the other is a chat loading crash; too different to merge." }, { "left": "issue:44246", "right": "issue:44871", "accept": false, - "reason": "Import-time slowness and Gemma-3 eos_token_id inconsistency are unrelated." + "reason": "Import performance and Gemma eos_token configuration are unrelated." }, { "left": "issue:44336", "right": "issue:44485", "accept": false, - "reason": "ANSI output in loading_report and GLM-5 RoPE implementation are unrelated topics." + "reason": "ANSI color handling in loading_report and GLM-5 RoPE implementation are unrelated." }, { "left": "issue:41950", "right": "issue:44220", "accept": false, - "reason": "Video-classification image-processor lookup and _torch_extract_fbank_features() are different preprocessing bugs." + "reason": "Video-classification image-processor lookup and audio fbank extraction are different preprocessing bugs." }, { "left": "issue:43550", "right": "issue:46506", "accept": false, - "reason": "torch.compile SDPA failure on Bamba and CPU/CUDA result mismatch on PPDocLayoutV3 are different code paths." + "reason": "torch.compile+SDPA breakage and CPU/CUDA result divergence are distinct model/runtime problems." }, { "left": "issue:36246", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but they concern different symbols and packages; not the same underlying defect." + "reason": "Both are import errors, but they concern different missing symbols and likely different packaging regressions." + }, + { + "left": "issue:36246", + "right": "issue:44351", + "accept": false, + "reason": "Different missing exports (Qwen2.5 VL image processor vs HybridCache); not the same underlying issue." }, { "left": "issue:44336", "right": "issue:46424", "accept": false, - "reason": "Loading-report ANSI codes and cache max length handling are unrelated." + "reason": "Terminal ANSI output and cache-length semantics are unrelated." }, { "left": "issue:42886", "right": "issue:46132", "accept": false, - "reason": "HF_HUB_OFFLINE cache loading and AttentionInterface.register behavior are unrelated." - }, - { - "left": "issue:36246", - "right": "issue:44351", - "accept": false, - "reason": "Different missing exports in transformers; same symptom class, but not the same bug or fix." + "reason": "Offline cache loading and AttentionInterface.register behavior are unrelated." }, { "left": "issue:44779", "right": "issue:45292", "accept": false, - "reason": "Deepseek tokenizer regression and resize_token_embeddings not updating output embeddings are different model/tokenizer issues." + "reason": "A DeepSeek tokenizer regression and resize_token_embeddings not updating output embeddings are different bugs." }, { "left": "issue:44734", "right": "issue:46424", "accept": false, - "reason": "KV-cache continuation crash in serve and cache max length being ignored for non-quantized cache are different cache problems." + "reason": "KV-cache continuation indexing and max_cache_length honoring are different cache issues." }, { - "left": "issue:45003", - "right": "issue:45561", + "left": "issue:45859", + "right": "issue:46399", "accept": false, - "reason": "Unsafe sys.modules access and a testing-utils file race are unrelated." + "reason": "Both are Qwen3.5/MoE-related, but one is CPU-offload incompatibility and the other is a registration/decorator crash; not the same bug." }, { - "left": "issue:45859", - "right": "issue:46399", + "left": "issue:45003", + "right": "issue:45561", "accept": false, - "reason": "Both involve Qwen3.5 MoE, but one is CPU offload compatibility and the other is kernel registration missing a decorator; different fixes." + "reason": "Unsafe sys.modules access and an xdist file race are unrelated." }, { "left": "issue:43901", "right": "issue:45003", "accept": false, - "reason": "A docs/behavior mismatch in TextClassificationPipeline and a sys.modules access bug are unrelated." + "reason": "Documentation drift about return_all_scores is unrelated to sys.modules access." }, { "left": "issue:43475", "right": "issue:46369", "accept": false, - "reason": "SAM3 video encoder attribute error and SAM3 lite-text memory overflow are different model variants and failure modes." + "reason": "A missing attribute in Sam3VisionEncoderOutput and a SAM3 LiteText memory overflow are different failures." + }, + { + "left": "issue:43257", + "right": "issue:45137", + "accept": false, + "reason": "Qwen3 MOE weight conversion and DeepSpeed ZeRO3 deque/index errors are different problems." + }, + { + "left": "issue:44242", + "right": "issue:44857", + "accept": false, + "reason": "Missing load-balancing loss and an AMP CUDA crash are unrelated." + }, + { + "left": "issue:36246", + "right": "issue:43596", + "accept": false, + "reason": "A missing import symbol and a ZeRO3 BertModel index error are unrelated." + }, + { + "left": "issue:44295", + "right": "issue:44749", + "accept": false, + "reason": "A position_ids buffer access bug and a slowdown after upgrade are not the same issue." + }, + { + "left": "issue:44496", + "right": "issue:45003", + "accept": false, + "reason": "An unrecognized model/config error and unsafe sys.modules access are unrelated." + }, + { + "left": "issue:43673", + "right": "issue:44561", + "accept": false, + "reason": "Chunked-prefill cache missing and removal of is_torch_fx_available affect different parts of generation/model loading." + }, + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "ESM2 breakage and Gemma-4 mm_token_type_ids defaults are unrelated model bugs." + }, + { + "left": "issue:43262", + "right": "issue:44246", + "accept": false, + "reason": "Audio chat-template sampling-rate handling and import slowdown are unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft edge, 42491 \u2194 43299, is plausibly the same Qwen3-MoE loading regression around the hf4.x/hf5.x transition. The remaining pairs are rejected with appropriately narrow reasoning and no overstatement." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -12285,35 +12343,29 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-469", "nodes": [ - "issue:36246", "issue:42886", - "issue:43257", - "issue:43262", + "issue:43065", "issue:43352", + "issue:43408", "issue:43475", "issue:43540", - "issue:43596", "issue:43653", - "issue:43673", + "issue:43906", + "issue:43931", "issue:44038", "issue:44060", "issue:44077", "issue:44162", - "issue:44242", - "issue:44246", - "issue:44295", + "issue:44188", "issue:44368", "issue:44466", "issue:44479", - "issue:44496", + "issue:44483", "issue:44561", - "issue:44749", - "issue:44857", "issue:44877", "issue:44977", "issue:44987", "issue:44995", - "issue:45003", "issue:45137", "issue:45200", "issue:45229", @@ -12324,31 +12376,28 @@ "issue:45397", "issue:45412", "issue:45459", + "issue:45563", "issue:45584", + "issue:45588", "issue:45600", "issue:45656", + "issue:45685", "issue:45701", "issue:45759", "issue:46123", "issue:46170", + "issue:46202", "issue:46399", + "issue:46431", "issue:46506" ], "soft_pairs": [ - "issue:43257|issue:45137", - "issue:44242|issue:44857", - "issue:36246|issue:43596", - "issue:44295|issue:44749", - "issue:44496|issue:45003", - "issue:43673|issue:44561", - "issue:44162|issue:45200", - "issue:43262|issue:44246", "issue:45375|issue:45759", "issue:44877|issue:45375", "issue:45292|issue:45584", - "issue:45292|issue:45701", - "issue:44038|issue:46506", "issue:45137|issue:45600", + "issue:44038|issue:46506", + "issue:45292|issue:45701", "issue:44561|issue:46399", "issue:43352|issue:45250", "issue:44466|issue:45335", @@ -12361,198 +12410,213 @@ "issue:45397|issue:45656", "issue:44368|issue:44977", "issue:43653|issue:44162", - "issue:42886|issue:45229" + "issue:42886|issue:45229", + "issue:44995|issue:45588", + "issue:44877|issue:46170", + "issue:43065|issue:43408", + "issue:44368|issue:45563", + "issue:43906|issue:46202", + "issue:43931|issue:44368", + "issue:45685|issue:46431", + "issue:44188|issue:46202", + "issue:44483|issue:46202" ], - "prepared_review_unit_hash": "bcbdbfcf15ae5472c9a81264772ee380841e930dba880f41f5bc09446538395d", + "prepared_review_unit_hash": "10644e12b8a955241c3200f2a09591a5867207d27a2ae090c989fd8b5c8112d4", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 27, - "serialized_chars": 35863, - "estimated_input_tokens": 8966, - "estimated_eval_tokens": 18188 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35414, + "estimated_input_tokens": 8854, + "estimated_eval_tokens": 17964 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is very noisy: most pairs are only loosely similar by subsystem or version, not true duplicates. The closest shared theme is Transformers v5/Qwen multimodal regressions, especially video-input handling, but the artifacts still describe different bugs and code paths.", - "confidence": 0.56, - "canonical_issue_reason": "issue:44479 is the broadest, most representative concrete regression in the cluster: it describes a v5.3.0 video-input failure across multiple Qwen multimodal variants, which best matches the cluster\u2019s dominant Qwen/v5 regression theme.", + "summary": "This cluster is heterogeneous: the items span unrelated model bugs, tokenizer/config issues, serving regressions, and backend-specific failures. I found no pair that clearly looks like the same underlying bug/change, so all soft edges should be rejected.", + "confidence": 0.84, + "canonical_issue_reason": "issue:45229 is the strongest representative issue in the set: it is concrete, high-impact, and well-documented with substantial discussion. It is more substantial than the very broad or highly model-specific reports nearby.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44479 is the strongest standalone issue to anchor the cluster because it is specific, user-facing, and spans several related model variants without being as narrow as the single-model reports.", + "best_issue_reason": "issue:45229 is the best single issue to anchor the cluster because it describes a reproducible, high-severity regression with clear scope and community attention.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43257", - "right": "issue:45137", - "accept": false, - "reason": "Both involve DeepSpeed, but one is a Qwen3 MoE weight-conversion bug and the other is a ZeRO3 deque/indexing failure; different failure modes and code paths." - }, - { - "left": "issue:44242", - "right": "issue:44857", - "accept": false, - "reason": "Unrelated models and subsystems: MoE load-balancing loss vs RT-DETR AMP/CUDA crash." - }, - { - "left": "issue:36246", - "right": "issue:43596", - "accept": false, - "reason": "A Qwen2.5-VL import error and a BertModel ZeRO3 init IndexError are unrelated." - }, - { - "left": "issue:44295", - "right": "issue:44749", - "accept": false, - "reason": "Position-IDs buffer access bug vs a general performance regression report; not the same issue." - }, - { - "left": "issue:44496", - "right": "issue:45003", - "accept": false, - "reason": "One is an unrecognized model/config loading failure; the other is unsafe sys.modules access in modeling_utils. Related area, different bug." - }, - { - "left": "issue:43673", - "right": "issue:44561", - "accept": false, - "reason": "Both are v5 compatibility regressions, but one is cache/chunked prefill and the other is removal of a utility breaking trust_remote_code models." - }, - { - "left": "issue:44162", - "right": "issue:45200", - "accept": false, - "reason": "ESM2 broken vs Gemma 4 mm_token_type_ids defaults; different model families and symptoms." - }, - { - "left": "issue:43262", - "right": "issue:44246", - "accept": false, - "reason": "Audio processor sampling-rate default vs import latency; no shared underlying bug." - }, { "left": "issue:45375", "right": "issue:45759", "accept": false, - "reason": "Both involve Qwen multimodal configs, but one drops a missing field under strict config while the other fails to unwrap text_config in from_config; distinct root causes." + "reason": "Both are about Qwen multimodal configs, but one is a missing config field and the other is AutoModel CausalLM config unwrapping; different failure modes and fixes." }, { "left": "issue:44877", "right": "issue:45375", "accept": false, - "reason": "Different models and different config-validation bugs." + "reason": "Both mention strict config behavior, but they affect different model families and different config fields; not the same bug." }, { "left": "issue:45292", "right": "issue:45584", "accept": false, - "reason": "Resize-token-embeddings behavior is unrelated to Whisper empty-transcription generation after align_special_tokens." + "reason": "One is about resize_token_embeddings not updating output embeddings, the other is Whisper failing on empty transcription after token alignment; unrelated code paths." }, { - "left": "issue:45292", - "right": "issue:45701", + "left": "issue:45137", + "right": "issue:45600", "accept": false, - "reason": "A specific embeddings-update bug and a broad tokenization-regression report are not the same issue." + "reason": "DeepSpeed ZeRO3 deque pop error and Sam3LiteText auto-mapping references are unrelated issues." }, { "left": "issue:44038", "right": "issue:46506", "accept": false, - "reason": "Both are multimodal/model-specific problems, but Qwen3-VL-Moe video loading and PPDocLayoutV3 CPU/CUDA divergence are unrelated." + "reason": "Qwen3-VL-Moe loading/regression and PPDocLayoutV3 CPU/CUDA divergence are different model/backend problems." }, { - "left": "issue:45137", - "right": "issue:45600", + "left": "issue:45292", + "right": "issue:45701", "accept": false, - "reason": "DeepSpeed deque underflow vs auto_mappings CI breakage; different code paths." + "reason": "Both involve tokenization/embedding behavior, but one is about embedding resizing while the other is version-dependent tokenization output; not the same bug." }, { "left": "issue:44561", "right": "issue:46399", "accept": false, - "reason": "A removed helper breaking trust_remote_code is unrelated to Qwen3.5 kernel registration for apply_rotary_pos_emb." + "reason": "Removal of a utility in v5 breaking trust_remote_code models is unrelated to Qwen3.5 kernel registration for apply_rotary_pos_emb." }, { "left": "issue:43352", "right": "issue:45250", "accept": false, - "reason": "One is an unsupported-model error for Flash Attention 2.0; the other is a generic feature request/topic, not the same concrete bug." + "reason": "One is a specific Nemotron Flash Attention 2 incompatibility; the other is a generic Flash Attention 2 topic with no matching concrete bug." }, { "left": "issue:44466", "right": "issue:45335", "accept": false, - "reason": "Different tied-weights / resize-token-embeddings serialization behaviors in different models." + "reason": "Both touch tied embeddings, but one is device-dependent serialization and the other is resize propagation to decoder embeddings; different defects." }, { "left": "issue:45412", "right": "issue:46123", "accept": false, - "reason": "Memory leak in RT-DETR is unrelated to MaskGenerationPipeline partial-batch result dropping." + "reason": "RT-DETR memory release and MaskGenerationPipeline partial-batch dropping are unrelated." }, { "left": "issue:44987", "right": "issue:45459", "accept": false, - "reason": "Both are loading-related, but one is a specific transformers>=5.1.0 model load failure and the other is tokenizer error masking when protobuf is absent." + "reason": "Model loading failure for physical-intelligence/fast and tokenizer decode errors when protobuf is missing are unrelated." }, { "left": "issue:44479", "right": "issue:46170", "accept": false, - "reason": "Qwen multimodal video regression vs Windows/ROCm torch.distributed import issue; unrelated." + "reason": "A video-input regression for several Qwen models is unrelated to Windows/ROCm torch.distributed import issues." }, { "left": "issue:44060", "right": "issue:44077", "accept": false, - "reason": "Incorrect tied-weights warning vs patchtsmixer post_init policy; not the same underlying bug." + "reason": "Incorrect tied-weights warning and patchtsmixer post_init restriction are different configuration/model-authoring issues." }, { "left": "issue:43475", "right": "issue:43540", "accept": false, - "reason": "SAM 3 Video output attribute error and Qwen3OmniMoe video ValueError are different model-specific failures." + "reason": "SAM3 vision encoder attribute error and Qwen3OmniMoe video-processing ValueError are different model-specific bugs." }, { "left": "issue:44995", "right": "issue:45200", "accept": false, - "reason": "GlmMoeDsa stale cache on second forward pass is unrelated to Gemma 4 mm_token_type_ids defaults." + "reason": "A stale cache crash on second forward pass is unrelated to Gemma4 needing mm_token_type_ids defaults." }, { "left": "issue:45397", "right": "issue:45656", "accept": false, - "reason": "Gemma-4 ZeRO3 from_pretrained and double optimizer-step are different DeepSpeed-related bugs." + "reason": "Gemma4 zero3 loading and optimizer step being called twice under deepspeed are different DeepSpeed-related problems." }, { "left": "issue:44368", "right": "issue:44977", "accept": false, - "reason": "A tied-word-embeddings warning and a flash-attention generation bug are distinct Qwen3.5 issues." + "reason": "A tied-word-embeddings warning for Qwen3.5 fine-tuning is not the same as a flash-attention generation failure." }, { "left": "issue:43653", "right": "issue:44162", "accept": false, - "reason": "BigBirdTokenizer special-token registration bug is unrelated to ESM2 being broken." + "reason": "BigBirdTokenizer special-token registration and ESM2 breakage are different tokenizer/model failures." }, { "left": "issue:42886", "right": "issue:45229", "accept": false, - "reason": "Offline tokenizer cache loading and Gemma 4 multi-GPU OOM are unrelated." + "reason": "Offline tokenizer cache loading and Gemma4 multi-GPU CUDA OOM are entirely unrelated." + }, + { + "left": "issue:44995", + "right": "issue:45588", + "accept": false, + "reason": "Stale indexer cache crash and flash_attention.py crashing on s_aux=None are unrelated failures." + }, + { + "left": "issue:44877", + "right": "issue:46170", + "accept": false, + "reason": "Strict granite_speech config loading and Windows/ROCm distributed import errors do not share the same bug." + }, + { + "left": "issue:43065", + "right": "issue:43408", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder and sam3_video vs sam3_tracker model-type mismatch are different SAM3 issues." + }, + { + "left": "issue:44368", + "right": "issue:45563", + "accept": false, + "reason": "Both concern generation warnings, but one is tied embeddings and the other is a stale num_return_sequences warning in paged generate()." + }, + { + "left": "issue:43906", + "right": "issue:46202", + "accept": false, + "reason": "An isolated reproduction of an earlier issue and a Gemma pipeline roundtrip failure on \u2581 are not the same underlying bug." + }, + { + "left": "issue:43931", + "right": "issue:44368", + "accept": false, + "reason": "Qwen3-VL-30B-A3B-Instruct weight shape mismatch is unrelated to the tied-embeddings warning in Qwen3.5 fine-tuning." + }, + { + "left": "issue:45685", + "right": "issue:46431", + "accept": false, + "reason": "MPS histogram not implemented for Int and missing chat template in a processor are unrelated." + }, + { + "left": "issue:44188", + "right": "issue:46202", + "accept": false, + "reason": "torch.compile branching causing diverging attention kernels is unrelated to the Gemma \u2581 roundtrip failure." + }, + { + "left": "issue:44483", + "right": "issue:46202", + "accept": false, + "reason": "A v5.3 chat/completions request rejection and a Gemma tokenization roundtrip failure are different API/model issues." } ] }, @@ -12579,65 +12643,55 @@ "issue:28282", "issue:30990", "issue:41628", + "issue:42371", "issue:42994", "issue:43065", "issue:43232", - "issue:43408", "issue:43475", "issue:43482", + "issue:43525", "issue:43643", "issue:43646", "issue:43673", - "issue:43906", - "issue:43931", + "issue:43723", + "issue:43784", "issue:44038", "issue:44162", "issue:44188", "issue:44297", "issue:44361", - "issue:44368", - "issue:44483", "issue:44623", "issue:44749", "issue:44869", "issue:44871", - "issue:44877", "issue:44933", - "issue:44995", + "issue:44964", + "issue:45092", + "issue:45137", "issue:45230", "issue:45310", "issue:45356", "issue:45375", "issue:45381", "issue:45397", + "issue:45406", "issue:45478", - "issue:45563", "issue:45584", - "issue:45588", "issue:45656", - "issue:45685", "issue:45987", "issue:46082", "issue:46123", "issue:46132", "issue:46170", "issue:46202", - "issue:46421", - "issue:46431" + "issue:46326", + "issue:46399", + "issue:46421" ], "soft_pairs": [ - "issue:44995|issue:45588", - "issue:43065|issue:43408", - "issue:44877|issue:46170", - "issue:44368|issue:45563", - "issue:43906|issue:46202", - "issue:43931|issue:44368", - "issue:45685|issue:46431", - "issue:44188|issue:46202", - "issue:44483|issue:46202", "issue:45987|issue:46082", - "issue:44188|issue:46421", "issue:45584|issue:46170", + "issue:44188|issue:46421", "issue:43065|issue:46123", "issue:44297|issue:44623", "issue:44361|issue:44749", @@ -12655,216 +12709,218 @@ "issue:44038|issue:44623", "issue:45375|issue:45478", "issue:41628|issue:44933", - "issue:43475|issue:45381" + "issue:43475|issue:45381", + "issue:42371|issue:45092", + "issue:45310|issue:45375", + "issue:42994|issue:43525", + "issue:44964|issue:45137", + "issue:43784|issue:46399", + "issue:43723|issue:46326", + "issue:45406|issue:46202", + "issue:43232|issue:44188" ], - "prepared_review_unit_hash": "a03f2ed1c79f147fe78e09fc3a1671756dbead6f1112b24ca68b40c053e2bd27", + "prepared_review_unit_hash": "93ab6ad855f604707df2b88f44d2fdf7f68eb07e46aa50e8855a417489e560d9", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36557, - "estimated_input_tokens": 9140, - "estimated_eval_tokens": 18536 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35623, + "estimated_input_tokens": 8906, + "estimated_eval_tokens": 18068 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a false-positive similarity group: the issues span unrelated import/load/save/generation/cache/regression bugs across different models and APIs. None of the soft pairs convincingly describe the same underlying bug or mergeable fix.", - "confidence": 0.82, - "canonical_issue_reason": "No canonical issue: the items are heterogeneous and do not form one duplicate family, even when titles share model names or API terms.", + "summary": "This cluster is heterogeneous: it mixes unrelated tokenizer, model-loading, generation, save/load, and performance regressions. None of the soft pairs look like the same concrete bug, so I reject all of them. As a representative issue, I\u2019d anchor the cluster on the broad v5 custom-model-init regression.", + "confidence": 0.64, + "canonical_issue_reason": "issue:43646 is the broadest, most umbrella-like regression report in the set, so it\u2019s the best canonical issue to represent this otherwise mixed cluster.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a strong cluster representative; 45375 is the closest to a repeat of the Qwen3.5-Moe regression, but the rest of the cluster is too unrelated for it to be a true canonical.", + "best_issue_reason": "issue:43646 has the clearest high-level framing of a general Transformers v5 breakage, making it the most useful issue to summarize the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44995", - "right": "issue:45588", - "accept": false, - "reason": "Both are crashes, but they hit different code paths: GlmMoeDsa second-forward cache state vs flash_attention s_aux=None handling." - }, - { - "left": "issue:43065", - "right": "issue:43408", + "left": "issue:45987", + "right": "issue:46082", "accept": false, - "reason": "Same SAM3 family, but one is a dummy Conv2d in the pixel decoder and the other is a model-type mismatch warning." + "reason": "Return-type contract bug in StaticCache vs LlamaConfig validation error; different code paths and failure modes." }, { - "left": "issue:44877", + "left": "issue:45584", "right": "issue:46170", "accept": false, - "reason": "Different failures: strict config loading for granite_speech vs Windows/ROCm import dependency issues." - }, - { - "left": "issue:44368", - "right": "issue:45563", - "accept": false, - "reason": "Both mention warnings, but they are unrelated: tie_word_embeddings vs stale num_return_sequences in paged generate()." - }, - { - "left": "issue:43906", - "right": "issue:46202", - "accept": false, - "reason": "An isolated reproduction of an older issue is not the same bug as Gemma pipeline roundtrip failure on the \u2581 token." - }, - { - "left": "issue:43931", - "right": "issue:44368", - "accept": false, - "reason": "Qwen3-VL weight-shape mismatch is unrelated to a Qwen3.5 fine-tuning warning about tie_word_embeddings." - }, - { - "left": "issue:45685", - "right": "issue:46431", - "accept": false, - "reason": "MPS histogram support in moe and missing chat template in a processor are unrelated problems." - }, - { - "left": "issue:44188", - "right": "issue:46202", - "accept": false, - "reason": "torch.compile attention-kernel divergence is a different bug from Gemma token roundtrip behavior." - }, - { - "left": "issue:44483", - "right": "issue:46202", - "accept": false, - "reason": "API request rejection in /v1/chat/completions is unrelated to Gemma pipeline token roundtrip failure." - }, - { - "left": "issue:45987", - "right": "issue:46082", - "accept": false, - "reason": "StaticCache return-type regression and LlamaConfig head_dim validation are distinct issues." + "reason": "Whisper empty-transcription generation bug vs Windows/ROCm torch.distributed import dependency issue." }, { "left": "issue:44188", "right": "issue:46421", "accept": false, - "reason": "Both touch generation/cache/compile, but one is kernel divergence under torch.compile and the other is spurious recompilation from lazy cache init." - }, - { - "left": "issue:45584", - "right": "issue:46170", - "accept": false, - "reason": "Whisper empty-transcription generation failure is unrelated to the Windows ROCm torch.distributed import problem." + "reason": "Both mention generation/cache behavior, but one is attention-kernel divergence under torch.compile and the other is cache lazy-init recompilation; not the same bug." }, { "left": "issue:43065", "right": "issue:46123", "accept": false, - "reason": "SAM3 pixel-decoder initialization and MaskGenerationPipeline final partial-batch handling are different subsystems and bugs." + "reason": "SAM3 pixel decoder dummy layer issue vs partial-batch masking pipeline dropping results; unrelated." }, { "left": "issue:44297", "right": "issue:44623", "accept": false, - "reason": "Both are save_pretrained regressions, but tokenizer_class mismatch and missing processor files are different serialization bugs." + "reason": "Tokenizer save metadata mismatch vs processor save missing files; both are save_pretrained-related but not the same artifact or bug." }, { "left": "issue:44361", "right": "issue:44749", "accept": false, - "reason": "MLukeTokenizer AttributeError and a data-filtering slowdown after upgrade are not the same underlying issue." + "reason": "MLukeTokenizer AttributeError vs a general throughput regression; different problems." }, { "left": "issue:44749", "right": "issue:45656", "accept": false, - "reason": "Performance regression in filtering is unrelated to deepspeed causing optimizer.step() to run twice." + "reason": "Performance slowdown regression vs DeepSpeed optimizer.step being called twice; not the same underlying issue." }, { "left": "issue:43646", "right": "issue:44162", "accept": false, - "reason": "A broad Transformers 5.0 custom-model init break and an ESM2-specific regression are not the same bug." + "reason": "Generic custom model init breakage vs ESM2-specific breakage; different models and failure paths." }, { "left": "issue:44623", "right": "issue:45656", "accept": false, - "reason": "Processor.save_pretrained missing files and duplicate optimizer steps are unrelated code paths." + "reason": "Processor save_pretrained file emission bug vs DeepSpeed optimizer-step duplication; unrelated." }, { "left": "issue:44749", "right": "issue:45310", "accept": false, - "reason": "Upgrade-related performance slowdown is unrelated to Qwen3.5 Moe from_pretrained failures." + "reason": "General performance regression vs Qwen3.5 MoE from_pretrained error; different symptoms and fixes." }, { "left": "issue:44871", "right": "issue:46132", "accept": false, - "reason": "Gemma eos_token_id inconsistency and AttentionInterface.register behavior changes are separate issues." + "reason": "Gemma eos_token_id inconsistency vs AttentionInterface.register behavior change; not the same bug." }, { "left": "issue:43232", "right": "issue:43673", "accept": false, - "reason": "Both involve generation/cache, but one is kwargs updating after sync_gpus and the other is missing cache during chunked_prefill." + "reason": "Generation kwargs update after sync_gpus vs missing GenerationMixin cache in chunked prefill; both generation-related but distinct." }, { "left": "issue:42994", "right": "issue:43482", "accept": false, - "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are opposite directions and not the same bug." + "reason": "Quantized model saving failure vs Qwen2.5-GGUF loading failure; opposite directions and different subsystems." }, { "left": "issue:45230", "right": "issue:45397", "accept": false, - "reason": "A placeholder bug report is too underspecified to match a specific gemma-4 zero3 from_pretrained error." + "reason": "Generic placeholder bug report vs Gemma-4 zero3 from_pretrained failure; no concrete overlap." }, { "left": "issue:43475", "right": "issue:43643", "accept": false, - "reason": "Missing fpn_position_embeddings in a SAM3 output object and missing fields from trust_remote_code AutoConfig are different missing-field bugs." + "reason": "Missing Sam3 vision encoder field vs trust_remote_code AutoConfig field-loss issue; both are missing-field reports but not the same one." }, { "left": "issue:44869", "right": "issue:45356", "accept": false, - "reason": "Whisper word-timestamp crash on trailing replacement characters is unrelated to the Kimi-K2.5 tokenizer codec regression." + "reason": "Whisper timestamp decode crash vs Kimi tokenizer codec/regex regression; different models and code paths." }, { "left": "issue:28282", "right": "issue:30990", "accept": false, - "reason": "PyTorch-not-found import error and a Sentence Transformers load hang are different failure modes." + "reason": "PyTorch-missing ImportError vs Sentence-Transformers loading hang; unrelated import vs runtime stall." }, { "left": "issue:44038", "right": "issue:44623", "accept": false, - "reason": "Qwen3-VL-Moe from_pretrained problems and processor.save_pretrained missing files are unrelated." + "reason": "Qwen3-VL-Moe/v5 bug vs processor.save_pretrained missing files; different artifacts and failures." }, { "left": "issue:45375", "right": "issue:45478", "accept": false, - "reason": "These are the closest pair, but one is a specific missing config field dropped by @strict and the other is only a generic Qwen3.5-Moe from_pretrained error; not enough to confirm the same root cause." + "reason": "Missing config field dropped by @strict vs generic Qwen3.5 MoE from_pretrained error; related model family but not the same concrete bug." }, { "left": "issue:41628", "right": "issue:44933", "accept": false, - "reason": "Both are import-related, but they refer to different symbols/modules and likely different broken imports." + "reason": "Both are import errors, but they concern different missing symbols from different paths; too broad to merge." }, { "left": "issue:43475", "right": "issue:45381", "accept": false, - "reason": "SAM3 vision encoder missing a field and qwen2.5-vl video position IDs being wrong are unrelated model-specific bugs." + "reason": "Sam3VisionEncoderOutput missing attribute vs qwen2.5-vl video position-id error; unrelated model families and failures." + }, + { + "left": "issue:42371", + "right": "issue:45092", + "accept": false, + "reason": "TF32 behavior/API settings issue vs InternVL2 meta-initialization incompatibility; different concerns." + }, + { + "left": "issue:45310", + "right": "issue:45375", + "accept": false, + "reason": "One is a Qwen3.5 MoE from_pretrained failure, the other a missing config field being dropped; not enough evidence they are the same bug." + }, + { + "left": "issue:42994", + "right": "issue:43525", + "accept": false, + "reason": "Quantized save failure vs LlamaConfig pad_token_id validation error; unrelated." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Phi-4 multimodal load failure vs DeepSpeed ZeRO3 deque IndexError; different components and symptoms." + }, + { + "left": "issue:43784", + "right": "issue:46399", + "accept": false, + "reason": "sentence-transformers import NameError vs Qwen3.5 kernel-registration ValueError; unrelated." + }, + { + "left": "issue:43723", + "right": "issue:46326", + "accept": false, + "reason": "AutoTokenizer loading regression in v5 vs continue_final_message bug for mistral_common tokenizers; different tokenizer paths." + }, + { + "left": "issue:45406", + "right": "issue:46202", + "accept": false, + "reason": "transformers serve crash on missing _tokenizer vs Gemma roundtrip failure on a specific character; unrelated." + }, + { + "left": "issue:43232", + "right": "issue:44188", + "accept": false, + "reason": "Generation kwargs update bug vs attention-kernel divergence under torch.compile; both touch generation/attention but are different concrete problems." } ] }, @@ -12895,62 +12951,54 @@ "issue:36010", "issue:41628", "issue:42175", - "issue:42371", + "issue:42222", "issue:42673", "issue:42831", "issue:42898", "issue:42994", "issue:43232", "issue:43278", + "issue:43377", + "issue:43479", "issue:43504", - "issue:43525", "issue:43704", - "issue:43723", "issue:43784", "issue:43824", "issue:43873", "issue:43976", + "issue:44016", "issue:44038", - "issue:44188", + "issue:44079", "issue:44222", + "issue:44242", "issue:44351", "issue:44496", "issue:44568", + "issue:44623", "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:45406", "issue:45561", "issue:45600", "issue:45712", "issue:45753", + "issue:45812", "issue:46077", "issue:46123", - "issue:46202", - "issue:46326", - "issue:46399" + "issue:46231" ], "soft_pairs": [ - "issue:42371|issue:45092", - "issue:45310|issue:45375", - "issue:42994|issue:43525", - "issue:44964|issue:45137", - "issue:43784|issue:46399", - "issue:43723|issue:46326", - "issue:45406|issue:46202", - "issue:43232|issue:44188", "issue:43873|issue:44945", "issue:41628|issue:44351", "issue:44945|issue:45753", - "issue:42673|issue:43704", "issue:45081|issue:45103", + "issue:42673|issue:43704", "issue:42994|issue:43824", "issue:44222|issue:45310", "issue:44038|issue:45310", @@ -12961,228 +13009,236 @@ "issue:42175|issue:44933", "issue:44496|issue:45561", "issue:43504|issue:45561", - "issue:45561|issue:45600", "issue:42175|issue:45712", + "issue:45561|issue:45600", "issue:43976|issue:44933", "issue:30333|issue:33290", "issue:46077|issue:46123", "issue:30990|issue:35141", - "issue:42898|issue:44568" + "issue:42898|issue:44568", + "issue:43377|issue:44242", + "issue:44623|issue:45310", + "issue:43479|issue:44869", + "issue:42222|issue:45200", + "issue:43479|issue:44079", + "issue:44016|issue:46231", + "issue:42898|issue:45812", + "issue:43784|issue:43824" ], - "prepared_review_unit_hash": "c33d55ac6729f2dcf5d2e6d558a267762f9b8fe4ee27f902a6d737cfa8878037", + "prepared_review_unit_hash": "571db8331624a0872e0d7387beff84fceaa5883c22572b4c31ed99e25ae89a43", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36605, - "estimated_input_tokens": 9152, - "estimated_eval_tokens": 18560 + "serialized_chars": 36639, + "estimated_input_tokens": 9160, + "estimated_eval_tokens": 18576 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly heterogeneous transformer-related issues, with one clear duplicate pair: Qwen3ForCausalLM VRAM leak (42673/43704). The other soft matches are title-similarity only and do not look like the same concrete bug.", - "confidence": 0.96, - "canonical_issue_reason": "issue:42673 is the older, fuller report for the only clear duplicate pair; issue:43704 repeats the same Qwen3ForCausalLM VRAM leak symptom in multiple dataloader threads.", + "summary": "Mostly unrelated issues. The only strong duplicate-looking pair is the two Qwen3ForCausalLM VRAM leak reports, which share the same title and very similar symptom description; the rest appear to cover different bugs or features.", + "confidence": 0.93, + "canonical_issue_reason": "issue:42673 is the best canonical issue because it is the fuller Qwen3ForCausalLM VRAM-leak report, while issue:43704 appears to repeat the same title and symptom.", "canonical_pr_reason": null, - "best_issue_reason": "issue:42673 is the best cluster anchor because it has the strongest duplicate evidence and matches issue:43704 exactly.", + "best_issue_reason": "issue:42673 is the best representative of the duplicate pair because it has the more detailed description of the same VRAM leak behavior reported again in issue:43704.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:42371", - "right": "issue:45092", - "accept": false, - "reason": "TF32 behavior guidance and an InternVL2 meta-init checkpoint incompatibility are different problems." - }, - { - "left": "issue:45310", - "right": "issue:45375", - "accept": false, - "reason": "Both mention Qwen3.5 MoE, but one is a from_pretrained error and the other is a missing config field being dropped; not the same bug." - }, - { - "left": "issue:42994", - "right": "issue:43525", - "accept": false, - "reason": "Quantized model saving failure is unrelated to a missing pad_token_id attribute on Llama4Config." - }, - { - "left": "issue:44964", - "right": "issue:45137", - "accept": false, - "reason": "Model loading with latest transformers and a DeepSpeed ZeRO3 deque error are different code paths." - }, - { - "left": "issue:43784", - "right": "issue:46399", - "accept": false, - "reason": "A sentence-transformers import NameError and a Qwen3.5 rotary registration crash are unrelated." - }, - { - "left": "issue:43723", - "right": "issue:46326", - "accept": false, - "reason": "Tokenizer loading in v5 and continue_final_message for mistral_common are different tokenizer issues." - }, - { - "left": "issue:45406", - "right": "issue:46202", - "accept": false, - "reason": "A processor _tokenizer attribute error and a Gemma tokenizer roundtrip failure are different bugs." - }, - { - "left": "issue:43232", - "right": "issue:44188", - "accept": false, - "reason": "Generation kwargs sync after sync_gpus and attention-kernel branching under torch.compile are distinct issues." - }, { "left": "issue:43873", "right": "issue:44945", "accept": false, - "reason": "Quantization/offloading behavior is not the same as incorrect output under pipeline parallelism." + "reason": "Quantization/offloading behavior is a different bug from incorrect outputs under pipeline parallelism." }, { "left": "issue:41628", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for different symbols and different failure paths." + "reason": "Both are import errors, but they concern different missing symbols and different code paths." }, { "left": "issue:44945", "right": "issue:45753", "accept": false, - "reason": "Incorrect pipeline-parallel output is different from an infinite loop on a specific image input." - }, - { - "left": "issue:42673", - "right": "issue:43704", - "accept": true, - "reason": "Identical title and same VRAM leak symptom in multiple dataloader threads; clear duplicate." + "reason": "Pipeline-parallelism output corruption is unrelated to a model-specific infinite loop on one image." }, { "left": "issue:45081", "right": "issue:45103", "accept": false, - "reason": "A Mistral regex patch crash and an auto_docstring future-annotations crash are unrelated." + "reason": "Different AttributeError sites: tokenizer regex patching vs auto-docstring keyword parsing." + }, + { + "left": "issue:42673", + "right": "issue:43704", + "accept": true, + "reason": "Same title and same underlying symptom: Qwen3ForCausalLM leaking VRAM across multiple dataloader threads." }, { "left": "issue:42994", "right": "issue:43824", "accept": false, - "reason": "Model saving failure and an import error for Qwen2_5_VLForConditionalGeneration are different bugs." + "reason": "Quantized model saving failure is not the same bug as an import error for a model class." }, { "left": "issue:44222", "right": "issue:45310", "accept": false, - "reason": "FP8 save_pretrained on MoE models is different from a Qwen3.5 from_pretrained error." + "reason": "FP8 save_pretrained failure and Qwen3.5 MoE from_pretrained failure are different operations with different failure modes." }, { "left": "issue:44038", "right": "issue:45310", "accept": false, - "reason": "Different model variants and different failure points; not the same underlying issue." + "reason": "They both mention Qwen3/MoE, but the reported symptoms and entry points are different enough to treat as separate issues." }, { "left": "issue:36010", "right": "issue:41628", "accept": false, - "reason": "Separate missing-import regressions for different objects." + "reason": "Different missing imports from different modules; no sign of the same broken export." }, { "left": "issue:42831", "right": "issue:44222", "accept": false, - "reason": "FP8 accuracy drift is not the same as FP8 model saving failing." + "reason": "One is an FP8 accuracy regression, the other is an FP8 save_pretrained problem." }, { "left": "issue:43278", "right": "issue:45137", "accept": false, - "reason": "Embedding dtype drift during eval is unrelated to a ZeRO3 deque IndexError." + "reason": "Embedding dtype drift in eval is unrelated to a DeepSpeed ZeRO3 deque IndexError." }, { "left": "issue:43232", "right": "issue:44792", "accept": false, - "reason": "Generation-kwarg handling and a janus image-generation test failure are not the same bug." + "reason": "Generation kwargs handling after sync_gpus is unrelated to a Janus image-generation test failure." }, { "left": "issue:42175", "right": "issue:44933", "accept": false, - "reason": "Installing torch-only extras and a removed image_utils import are different problems." + "reason": "Backend dependency wiring is unrelated to a missing image_utils import." }, { "left": "issue:44496", "right": "issue:45561", "accept": false, - "reason": "Config loading rejection and a pytest-xdist race in testing utils are unrelated." + "reason": "Unrecognized model/config loading is unrelated to the pytest-xdist captured_info.txt race." }, { "left": "issue:43504", "right": "issue:45561", "accept": false, - "reason": "Legacy config-field loading failure is different from a captured_info.txt race in tests." + "reason": "Legacy-field model loading and the test-file race are different failures." }, { - "left": "issue:45561", - "right": "issue:45600", + "left": "issue:42175", + "right": "issue:45712", "accept": false, - "reason": "A test race and stale auto_mappings references are separate issues." + "reason": "Torch extra backend dependency issues are not the same as leftover dummy classes leaking into dir(transformers)." }, { - "left": "issue:42175", - "right": "issue:45712", + "left": "issue:45561", + "right": "issue:45600", "accept": false, - "reason": "Package-install extras and dummy class cleanup are not the same code-path problem." + "reason": "A test race and stale auto_mappings references are different CI/test problems." }, { "left": "issue:43976", "right": "issue:44933", "accept": false, - "reason": "Python version compatibility breakage is unrelated to a missing import from image_utils." + "reason": "Python version compatibility failure is unrelated to a missing image_utils import." }, { "left": "issue:30333", "right": "issue:33290", "accept": false, - "reason": "MLflow reporting after training failure is unrelated to an Adafactor/DeepSpeed OOM." + "reason": "MLflow job-state reporting and DeepSpeed Adafactor OOM are unrelated." }, { "left": "issue:46077", "right": "issue:46123", "accept": false, - "reason": "Bidirectional attention masking and final-partial-batch result dropping are different bugs." + "reason": "Attention-mask construction for Gemma4 is unrelated to final-batch tracking in MaskGenerationPipeline." }, { "left": "issue:30990", "right": "issue:35141", "accept": false, - "reason": "A loading hang is not the same as output embeddings being reinitialized after resizing." + "reason": "A loading hang for Sentence Transformers is unrelated to embedding reinitialization after resize." }, { "left": "issue:42898", "right": "issue:44568", "accept": false, - "reason": "Tokenizer cleanup behavior changes and missing BOS/EOS insertion are distinct regressions." + "reason": "Tokenization cleanup behavior changes and missing BOS/EOS insertion are distinct tokenizer regressions." + }, + { + "left": "issue:43377", + "right": "issue:44242", + "accept": false, + "reason": "Missing padding-mask support in MIMI is unrelated to omitted load-balancing loss." + }, + { + "left": "issue:44623", + "right": "issue:45310", + "accept": false, + "reason": "Missing files on processor save_pretrained is unrelated to the Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:43479", + "right": "issue:44869", + "accept": false, + "reason": "Default multimodal config initialization is unrelated to Whisper timestamp decode crashing on a trailing replacement character." + }, + { + "left": "issue:42222", + "right": "issue:45200", + "accept": false, + "reason": "Broken vitpose model loading is unrelated to Gemma 4 mm_token_type_ids defaults." + }, + { + "left": "issue:43479", + "right": "issue:44079", + "accept": false, + "reason": "Default config initialization is unrelated to ModelOutput key assignment for previously-None fields." + }, + { + "left": "issue:44016", + "right": "issue:46231", + "accept": false, + "reason": "A notebook syntax error and a docstring parameter-name mismatch are different documentation issues." + }, + { + "left": "issue:42898", + "right": "issue:45812", + "accept": false, + "reason": "clean_up_tokenization_spaces behavior and Granite token-ID regression are separate tokenizer issues." + }, + { + "left": "issue:43784", + "right": "issue:43824", + "accept": false, + "reason": "Different import failures with different missing symbols; no evidence of a shared root cause." } ] }, "evaluator_result": { "accept": false, - "feedback": "The cluster summary is mostly grounded, and the Qwen3ForCausalLM pair (42673/43704) is a solid duplicate. However, at least one soft-edge verdict is not conservative enough: the 42371/45092 rationale introduces an unsupported \u2018packaging/backend extras\u2019 explanation that is not grounded in the packet\u2019s visible titles/metadata. Because the review should only accept when every soft-edge verdict stays conservative, I have to reject this output." + "feedback": "The soft-edge verdicts are generally conservative, but the issue-level reasoning overstates the evidence: the packet only shows two issues with identical titles, not a confirmed duplicate link or externally confirmed duplicate topic. Phrases like \"best representative of the cluster\u2019s only real duplicate topic\" and \"matches the confirmed duplicate topic exactly\" go beyond the packet. Keep the conclusion limited to \"these two reports appear to be duplicates based on identical titles and similar descriptions.\"" }, "evaluator_used": true, "retried": true, @@ -13210,68 +13266,60 @@ "issue:34689", "issue:35141", "issue:36331", - "issue:42222", + "issue:41628", + "issue:42757", "issue:42796", "issue:42831", "issue:42898", "issue:43295", - "issue:43377", + "issue:43475", "issue:43479", "issue:43644", "issue:43723", - "issue:43784", - "issue:43824", "issue:43874", - "issue:44016", "issue:44060", "issue:44075", - "issue:44079", "issue:44222", - "issue:44242", + "issue:44263", "issue:44295", "issue:44297", + "issue:44373", "issue:44448", + "issue:44485", "issue:44568", "issue:44623", "issue:44734", "issue:44779", - "issue:44869", "issue:44933", + "issue:44977", + "issue:45071", "issue:45092", "issue:45103", "issue:45125", - "issue:45200", "issue:45245", "issue:45310", "issue:45356", "issue:45405", "issue:45656", "issue:45712", - "issue:45812", + "issue:46082", "issue:46132", - "issue:46231" + "issue:46231", + "issue:46305" ], "soft_pairs": [ - "issue:43377|issue:44242", - "issue:44623|issue:45310", - "issue:43479|issue:44869", - "issue:42222|issue:45200", - "issue:43479|issue:44079", - "issue:44016|issue:46231", "issue:43874|issue:45092", - "issue:42898|issue:45812", - "issue:43784|issue:43824", "issue:43723|issue:44568", "issue:42898|issue:44779", "issue:44734|issue:45103", "issue:44075|issue:46231", "issue:44295|issue:45656", "issue:42898|issue:45356", - "issue:42898|issue:43295", "issue:27957|issue:30333", + "issue:42898|issue:43295", "issue:30333|issue:34689", - "issue:45405|issue:45712", "issue:44060|issue:45125", + "issue:45405|issue:45712", "issue:44222|issue:44297", "issue:44933|issue:45712", "issue:30333|issue:36331", @@ -13283,228 +13331,229 @@ "issue:44222|issue:44623", "issue:44448|issue:46132", "issue:42796|issue:43723", - "issue:45245|issue:46132" + "issue:45245|issue:46132", + "issue:41628|issue:42757", + "issue:44297|issue:45310", + "issue:43475|issue:46305", + "issue:45071|issue:46082", + "issue:44373|issue:44485", + "issue:44263|issue:45092", + "issue:43723|issue:44977" ], - "prepared_review_unit_hash": "a1ccc7ddd4406e7130b8a47af6c8aeafed55bc2991de22cf75b19cfaa00b814c", + "prepared_review_unit_hash": "d523dab0ac982152e94ea198f25006b82b89a6f7f19a057af939bca55e8a67be", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 32, - "serialized_chars": 36879, - "estimated_input_tokens": 9220, - "estimated_eval_tokens": 18696 + "soft_pair_count": 31, + "serialized_chars": 36367, + "estimated_input_tokens": 9092, + "estimated_eval_tokens": 18440 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, - "split": true, - "analyst_result": { - "summary": "The cluster is mostly a set of unrelated Transformers bug reports spanning tokenizers, multimodal models, training, and documentation. None of the soft pairs look like true duplicates or mergeable PR equivalents under a strict same-bug standard.", - "confidence": 0.81, - "canonical_issue_reason": "issue:42898 is the most central item in the cluster and a concrete, user-facing v5 regression report with several related tokenizer-regression links, making it the best representative issue.", - "canonical_pr_reason": null, - "best_issue_reason": "issue:42898 is the strongest standalone representative of the cluster: specific, reproducible, and broadly relevant to a common regression theme.", - "best_pr_reason": null, - "soft_edge_verdicts": [ - { - "left": "issue:43377", - "right": "issue:44242", - "accept": false, - "reason": "Different bugs: MIMI batch-vs-single padding-mask mismatch vs MoE load-balancing loss omission." - }, - { - "left": "issue:44623", - "right": "issue:45310", - "accept": false, - "reason": "Different code paths: processor.save_pretrained file omission vs Qwen3.5 MoE from_pretrained failure." - }, - { - "left": "issue:43479", - "right": "issue:44869", - "accept": false, - "reason": "Unrelated multimodal config initialization bug vs Whisper timestamp decode crash." - }, - { - "left": "issue:42222", - "right": "issue:45200", - "accept": false, - "reason": "Different model families and failures: VitPose breakage vs Gemma 4 mm_token_type_ids defaulting." - }, - { - "left": "issue:43479", - "right": "issue:44079", - "accept": false, - "reason": "Config-default initialization issue vs ModelOutput key assignment bug." - }, - { - "left": "issue:44016", - "right": "issue:46231", - "accept": false, - "reason": "Notebook syntax error is unrelated to a docs parameter-name mismatch." - }, - { - "left": "issue:43874", - "right": "issue:45092", - "accept": false, - "reason": "Different failures: missing image-patch API vs remote-code/meta-init incompatibility." - }, - { - "left": "issue:42898", - "right": "issue:45812", - "accept": false, - "reason": "Both are tokenizer regressions, but they affect different behaviors and model families." - }, + "split": true, + "analyst_result": { + "summary": "This cluster is very noisy: most items are only loosely related by broad Transformers subsystem or v5-regression wording, not the same underlying bug. I found no convincing duplicate pairs, so all soft edges are rejected.", + "confidence": 0.24, + "canonical_issue_reason": "Selected `issue:42898` as the most broadly representative tokenizer regression report in v5; it is generic enough to stand in for the tokenizer-regression subset.", + "canonical_pr_reason": null, + "best_issue_reason": "`issue:42898` is the best overall representative of the cluster\u2019s recurring tokenizer/v5 regression theme, but the set is too heterogeneous for true duplicate consolidation.", + "best_pr_reason": null, + "soft_edge_verdicts": [ { - "left": "issue:43784", - "right": "issue:43824", + "left": "issue:43874", + "right": "issue:45092", "accept": false, - "reason": "Different import failures: missing nn in sentence-transformers integration vs missing Qwen2.5-VL class export." + "reason": "Different failures: missing image-processor method vs remote-code/meta-init incompatibility." }, { "left": "issue:43723", "right": "issue:44568", "accept": false, - "reason": "Tokenizer-loading regression vs special-tokens/BOS-EOS behavior regression; not the same bug." + "reason": "Tokenizer loading failure is distinct from special-token insertion behavior for a specific model." }, { "left": "issue:42898", "right": "issue:44779", "accept": false, - "reason": "Both are v5 tokenizer regressions, but they concern different models and likely different root causes." + "reason": "Both are tokenizer regressions in v5, but they describe different symptoms and likely different code paths." }, { "left": "issue:44734", "right": "issue:45103", "accept": false, - "reason": "Serving KV-cache indexing crash is unrelated to auto_docstring annotation handling." + "reason": "Serving KV-cache indexing bug is unrelated to auto-docstring crash under future annotations." }, { "left": "issue:44075", "right": "issue:46231", "accept": false, - "reason": "Optimizer argument usage bug vs documentation typo; unrelated." + "reason": "Optimizer argument handling is unrelated to a documentation parameter-name mismatch." }, { "left": "issue:44295", "right": "issue:45656", "accept": false, - "reason": "Position_ids buffer read error is unrelated to Deepspeed double-step behavior." + "reason": "Buffer read error and duplicate optimizer step under deepspeed are different runtime problems." }, { "left": "issue:42898", "right": "issue:45356", "accept": false, - "reason": "Both tokenizer regressions, but they affect different models and different tokenizer logic." + "reason": "Both concern tokenizer regressions, but one is clean-up spacing behavior and the other is codec/warning handling for Kimi-K2.5." }, { - "left": "issue:42898", - "right": "issue:43295", + "left": "issue:27957", + "right": "issue:30333", "accept": false, - "reason": "Shared tokenizer/processor area, but one is clean_up_tokenization_spaces behavior and the other is processor.tokenizer/images access regression." + "reason": "Flash Attention 2 model support and MLflow reporting on failed training jobs are unrelated." }, { - "left": "issue:27957", - "right": "issue:30333", + "left": "issue:42898", + "right": "issue:43295", "accept": false, - "reason": "Flash Attention model issue and MLflow reporting failure are unrelated." + "reason": "Tokenizer clean-up behavior and processor/tokenizer API regression are distinct breakages." }, { "left": "issue:30333", "right": "issue:34689", "accept": false, - "reason": "MLflow job-state reporting bug vs Llama 3.2 Vision model-loading regression." + "reason": "Training-job reporting bug is unrelated to Llama 3.2 Vision model-loading regression." }, { - "left": "issue:45405", - "right": "issue:45712", + "left": "issue:44060", + "right": "issue:45125", "accept": false, - "reason": "PEFT version bump problem is unrelated to dummy class cleanup/check_repo failures." + "reason": "Incorrect tied-weights warning is not the same as missing tensor-parallel plan metadata." }, { - "left": "issue:44060", - "right": "issue:45125", + "left": "issue:45405", + "right": "issue:45712", "accept": false, - "reason": "Different Qwen3 family issues: tied-weights warning vs missing tensor-parallel plan." + "reason": "PEFT version pinning issue is unrelated to dummy class cleanup and repo checks." }, { "left": "issue:44222", "right": "issue:44297", "accept": false, - "reason": "Different save_pretrained bugs in different components: FP8 MoE vs tokenizer metadata." + "reason": "Both mention save_pretrained, but one is FP8 MoE serialization and the other is tokenizer_class metadata mismatch." }, { "left": "issue:44933", "right": "issue:45712", "accept": false, - "reason": "Missing import from image_utils is unrelated to leftover dummy classes." + "reason": "Nonexistent import from image_utils is unrelated to leftover dummy PT classes." }, { "left": "issue:30333", "right": "issue:36331", "accept": false, - "reason": "MLflow reporting failure vs Trainer.compute_loss signature breakage." + "reason": "MLflow reporting on failure is unrelated to CustomTrainer.compute_loss signature changes." }, { "left": "issue:30333", "right": "issue:35141", "accept": false, - "reason": "Training-job reporting bug vs embedding reinitialization during post_init." + "reason": "Failed-job reporting and embedding reinitialization are separate training behaviors." }, { "left": "issue:44297", "right": "issue:45656", "accept": false, - "reason": "Tokenizer save metadata bug is unrelated to Deepspeed optimizer stepping." + "reason": "Tokenizer save metadata bug does not match optimizer stepping twice under deepspeed." }, { "left": "issue:22355", "right": "issue:28282", "accept": false, - "reason": "Both are import errors, but one is missing transformers.onnx and the other is missing PyTorch." + "reason": "Missing transformers.onnx module and missing PyTorch dependency are different import errors." }, { "left": "issue:42831", "right": "issue:44623", "accept": false, - "reason": "FineGrainedFP8 accuracy issue is unrelated to processor.save_pretrained missing files." + "reason": "FP8 accuracy issue and processor save_pretrained missing files are not the same bug." }, { "left": "issue:43479", "right": "issue:43644", "accept": false, - "reason": "Multimodal config defaults are unrelated to non-persistent buffer initialization junk." + "reason": "Default multimodal config initialization and non-persistent buffer junk are unrelated initialization bugs." }, { "left": "issue:44222", "right": "issue:44623", "accept": false, - "reason": "Both mention save_pretrained, but they concern different artifacts and failures." + "reason": "Different save_pretrained failures: FP8 MoE serialization vs processor artifact omission." }, { "left": "issue:44448", "right": "issue:46132", "accept": false, - "reason": "Pegasus v4/v5 output regression is unrelated to AttentionInterface.register behavior changes." + "reason": "PEGASUS v4/v5 output drift is unrelated to AttentionInterface.register behavior changes." }, { "left": "issue:42796", "right": "issue:43723", "accept": false, - "reason": "Different tokenizer problems: BPE marker handling vs tokenizer loading in v5." + "reason": "Tokenizer marker handling and tokenizer loading/import behavior are different problems." }, { "left": "issue:45245", "right": "issue:46132", "accept": false, - "reason": "Category-count runtime error is unrelated to AttentionInterface.register behavior." + "reason": "Category-cardinality runtime error is unrelated to attention registry behavior." + }, + { + "left": "issue:41628", + "right": "issue:42757", + "accept": false, + "reason": "Missing AutoImageProcessor export and missing is_offline_mode import come from different modules and causes." + }, + { + "left": "issue:44297", + "right": "issue:45310", + "accept": false, + "reason": "Tokenizer save metadata mismatch and Qwen3.5 MoE from_pretrained error are separate issues." + }, + { + "left": "issue:43475", + "right": "issue:46305", + "accept": false, + "reason": "SAM 3 Video attribute error and MoonshineStreaming off-by-one mask are unrelated model bugs." + }, + { + "left": "issue:45071", + "right": "issue:46082", + "accept": false, + "reason": "Type-checking regression in PretrainedConfig is different from LlamaConfig head_dim validation." + }, + { + "left": "issue:44373", + "right": "issue:44485", + "accept": false, + "reason": "Docstring typo and GLM-5 RoPE implementation discussion are not the same bug." + }, + { + "left": "issue:44263", + "right": "issue:45092", + "accept": false, + "reason": "torch.split return handling in GLM MoE code is unrelated to remote-code meta-initialization compatibility." + }, + { + "left": "issue:43723", + "right": "issue:44977", + "accept": false, + "reason": "Both are tokenizer issues in v5, but one is loading failure and the other is incorrect Deepseek tokenization output." } ] }, @@ -13528,17 +13577,20 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-469", "nodes": [ + "issue:22355", + "issue:27957", "issue:30990", "issue:38175", "issue:41628", - "issue:42757", "issue:42831", "issue:42907", "issue:43122", "issue:43232", - "issue:43475", + "issue:43408", + "issue:43479", "issue:43644", "issue:43723", + "issue:43824", "issue:44038", "issue:44190", "issue:44206", @@ -13547,18 +13599,15 @@ "issue:44263", "issue:44279", "issue:44295", - "issue:44297", "issue:44360", "issue:44373", "issue:44483", "issue:44484", - "issue:44485", "issue:44623", "issue:44734", "issue:44829", "issue:44871", - "issue:44977", - "issue:45071", + "issue:44913", "issue:45092", "issue:45103", "issue:45310", @@ -13572,250 +13621,230 @@ "issue:46209", "issue:46305", "issue:46328", + "issue:46402", "issue:46421", "issue:46424", "issue:46431" ], "soft_pairs": [ - "issue:41628|issue:42757", - "issue:44297|issue:45310", - "issue:43475|issue:46305", - "issue:45071|issue:46082", - "issue:44373|issue:44485", - "issue:44263|issue:45092", - "issue:45381|issue:46305", - "issue:43723|issue:44977", "issue:43122|issue:46209", + "issue:45381|issue:46305", "issue:44222|issue:44484", - "issue:44206|issue:45103", "issue:46209|issue:46431", + "issue:44206|issue:45103", "issue:43232|issue:46424", "issue:44734|issue:46421", "issue:44190|issue:46082", "issue:44295|issue:44623", "issue:43232|issue:45834", - "issue:44263|issue:44360", "issue:43723|issue:44871", + "issue:44263|issue:44360", "issue:44279|issue:46328", - "issue:44261|issue:45092", - "issue:30990|issue:38175", "issue:43644|issue:46305", + "issue:30990|issue:38175", + "issue:44261|issue:45092", "issue:45092|issue:45357", "issue:42907|issue:46050", "issue:42831|issue:45310", "issue:44263|issue:44373", "issue:44829|issue:46002", "issue:44483|issue:45406", - "issue:42831|issue:44038" + "issue:42831|issue:44038", + "issue:22355|issue:27957", + "issue:41628|issue:43824", + "issue:43479|issue:44913", + "issue:43408|issue:46402" ], - "prepared_review_unit_hash": "10c21a28bbf87936ef85a007efde76faba12af182070760883801b981db7cfd8", + "prepared_review_unit_hash": "34b581b9fee06bf65fbfcac4ffc098574c543037226dfb4d1ff47974bf207617", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 30, - "serialized_chars": 35833, - "estimated_input_tokens": 8959, - "estimated_eval_tokens": 18174 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35534, + "estimated_input_tokens": 8884, + "estimated_eval_tokens": 18024 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab-bag of Transformers v5 regressions and model-specific bugs; no pull requests are present. None of the soft pairs look like the same underlying fix, so all are rejected.", - "confidence": 0.39, - "canonical_issue_reason": "issue:43723 is the broadest user-facing regression here: a generic tokenizer loading failure via AutoTokenizer.from_pretrained in v5, which makes it the best representative of the cluster\u2019s load/save/API breakage theme.", + "summary": "This cluster is a grab bag of mostly unrelated Transformers issues spanning tokenization, cache/generation, multimodal configs, serving, quantization, and model-specific regressions. None of the proposed soft pairs look like true duplicates of the same bug or change.", + "confidence": 0.79, + "canonical_issue_reason": "There is no strong canonical issue because the set is heterogeneous; if one anchor is needed, issue 46050 is the most concrete open bug with a reproducible save/load regression.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43723 is the most suitable anchor for triage because it is broadly scoped, clearly actionable, and closer to the common v5 compatibility problems than the more model-specific reports.", + "best_issue_reason": "Issue 46050 is the best overall triage anchor because it is open, specific, and describes a concrete model quantization save/load failure.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:41628", - "right": "issue:42757", - "accept": false, - "reason": "Both are import errors after v5 changes, but they involve different symbols and different dependency modules, so they do not appear to be the same bug." - }, - { - "left": "issue:44297", - "right": "issue:45310", - "accept": false, - "reason": "Both concern tokenizer/model loading, but one is a save_pretrained metadata mismatch while the other is a Qwen3.5 MoE from_pretrained failure; different code paths and symptoms." - }, - { - "left": "issue:43475", - "right": "issue:46305", - "accept": false, - "reason": "These are unrelated model bugs: one is a missing attribute in SAM 3 Video, the other is an off-by-one in MoonshineStreaming masking." - }, - { - "left": "issue:45071", - "right": "issue:46082", - "accept": false, - "reason": "Both touch config validation, but one is a general PreTrainedConfig typing regression and the other is a LlamaConfig constraint on head_dim; not the same fix." - }, - { - "left": "issue:44373", - "right": "issue:44485", - "accept": false, - "reason": "One is a docstring issue for position_ids, the other is a GLM-5 RoPE implementation discussion; no shared underlying bug is evident." - }, - { - "left": "issue:44263", - "right": "issue:45092", + "left": "issue:43122", + "right": "issue:46209", "accept": false, - "reason": "These affect different areas: a GlmMoeDsaIndexer split behavior versus remote-code checkpoint compatibility with meta initialization." + "reason": "Tokenizer-version mismatch and FP8 inference on B200 are unrelated bugs." }, { "left": "issue:45381", "right": "issue:46305", "accept": false, - "reason": "Both involve vision/video position-related behavior, but the reported failures are different and not clearly the same concrete code path." - }, - { - "left": "issue:43723", - "right": "issue:44977", - "accept": false, - "reason": "Tokenizer loading failure and flash-attention generation failure are distinct regressions affecting different subsystems." - }, - { - "left": "issue:43122", - "right": "issue:46209", - "accept": false, - "reason": "Tokenization differences across versions and FP8 inference on B200 are unrelated bugs." + "reason": "Different model paths and failures: Qwen2.5-VL vision positions vs MoonshineStreaming mask off-by-one." }, { "left": "issue:44222", "right": "issue:44484", "accept": false, - "reason": "FP8 save_pretrained/moe behavior and a question about max_shard_size default are not the same issue." + "reason": "FP8 save_pretrained/moe serialization is unrelated to max_shard_size default behavior." }, { - "left": "issue:44206", - "right": "issue:45103", + "left": "issue:46209", + "right": "issue:46431", "accept": false, - "reason": "A feature-extractor crash from an unsupported center argument is unrelated to the auto_docstring annotations crash." + "reason": "FP8 inference on B200 and missing chat template are different serving/model issues." }, { - "left": "issue:46209", - "right": "issue:46431", + "left": "issue:44206", + "right": "issue:45103", "accept": false, - "reason": "FP8 inference on B200 and a chat template absence error are different runtime failures." + "reason": "Feature extractor argument crash and auto_docstring annotations crash are separate code paths." }, { "left": "issue:43232", "right": "issue:46424", "accept": false, - "reason": "Generation state syncing and cache max length handling are different cache/generation behaviors, not one bug." + "reason": "Generation kwargs handling after sync_gpus is distinct from cache max_length configuration." }, { "left": "issue:44734", "right": "issue:46421", "accept": false, - "reason": "Wrong tensor indexing in serve KV-cache continuation is unrelated to lazy cache initialization causing recompilation." + "reason": "KV-cache continuation tensor indexing and lazy cache init recompilation are different bugs." }, { "left": "issue:44190", "right": "issue:46082", "accept": false, - "reason": "Local dataset loading in an example script and a LlamaConfig validation rule are unrelated." + "reason": "Local dataset loading failure is unrelated to LlamaConfig head_dim validation." }, { "left": "issue:44295", "right": "issue:44623", "accept": false, - "reason": "Reading a registered buffer and processor.save_pretrained missing files are separate save/load issues with different failure modes." + "reason": "position_ids buffer access and processor.save_pretrained missing files do not share the same cause." }, { "left": "issue:43232", "right": "issue:45834", "accept": false, - "reason": "Generation keyword/state handling and a long OCR input index error are not the same underlying defect." + "reason": "Generation sync_gpus behavior and Kosmos2.5 OCR index errors are unrelated." }, { - "left": "issue:44263", - "right": "issue:44360", + "left": "issue:43723", + "right": "issue:44871", "accept": false, - "reason": "Both mention the DSA indexer, but one is about torch.split return values and the other about a missing ReLU; distinct implementation bugs." + "reason": "AutoTokenizer load failure and Gemma-3 eos_token_id mismatch are separate issues." }, { - "left": "issue:43723", - "right": "issue:44871", + "left": "issue:44263", + "right": "issue:44360", "accept": false, - "reason": "Tokenizer loading regression and inconsistent eos_token_id config are different model/config issues." + "reason": "Both mention the DSA indexer, but one is split return handling and the other is a missing ReLU; different defects." }, { "left": "issue:44279", "right": "issue:46328", "accept": false, - "reason": "A generic dependency issue is not the same as a padding performance regression." + "reason": "A dependency problem and a padding performance regression are not the same bug." }, { - "left": "issue:44261", - "right": "issue:45092", + "left": "issue:43644", + "right": "issue:46305", "accept": false, - "reason": "MLA layernorm precision from missing rms_norm_eps is unrelated to meta-init incompatibility in old InternVL2 checkpoints." + "reason": "Non-persistent buffer junk and sliding-window off-by-one are unrelated." }, { "left": "issue:30990", "right": "issue:38175", "accept": false, - "reason": "Sentence Transformers loading hang and zero probabilities in SigLIP2 are unrelated model-specific issues." + "reason": "Loading stalls and zero probabilities are different failure modes." }, { - "left": "issue:43644", - "right": "issue:46305", + "left": "issue:44261", + "right": "issue:45092", "accept": false, - "reason": "Non-persistent buffer junk in v5 and a MoonshineStreaming mask off-by-one are different bug classes." + "reason": "Missing rms_norm_eps precision issue is unrelated to old InternVL2 meta-init incompatibility." }, { "left": "issue:45092", "right": "issue:45357", "accept": false, - "reason": "Remote-code meta-init incompatibility and incorrect visual encoder keys on save_pretrained are different serialization/loading problems." + "reason": "Both are v5 multimodal regressions, but they affect different model families and code paths." }, { "left": "issue:42907", "right": "issue:46050", "accept": false, - "reason": "Dequantized Ministrals/Devstrals save failures and sinq quantized model save/load breakage are both quantization-related but not the same concrete bug." + "reason": "Both involve save/load of quantized models, but dequantized Ministrals/Devstrals and sinq round-trip failures are not the same bug." }, { "left": "issue:42831", "right": "issue:45310", "accept": false, - "reason": "An accuracy issue in FineGrainedFP8 and a Qwen3.5 MoE from_pretrained error are different failure modes." + "reason": "FineGrainedFP8 accuracy and Qwen3.5-Moe from_pretrained failure are unrelated." }, { "left": "issue:44263", "right": "issue:44373", "accept": false, - "reason": "The DSA indexer split bug and the position_ids docstring issue do not describe the same underlying defect." + "reason": "A DSA indexer tensor bug and a docstring issue for position_ids are not duplicates." }, { "left": "issue:44829", "right": "issue:46002", "accept": false, - "reason": "Degenerate training with flash_attention_3 and NaNs after checkout/init are both training-path issues but not the same root cause." + "reason": "Flash-attention training degeneration and NaNs after checkout/init are different instability reports." }, { "left": "issue:44483", "right": "issue:45406", "accept": false, - "reason": "An API request rejection in transformers serve and a Gemma4Processor missing _tokenizer attribute are separate serve/processor bugs." + "reason": "Chat/completions request rejection and Gemma4Processor missing _tokenizer are separate serving bugs." }, { "left": "issue:42831", "right": "issue:44038", "accept": false, - "reason": "FineGrainedFP8 accuracy concerns and a Qwen3-VL-Moe v5 bug are unrelated." + "reason": "FP8 accuracy regression and Qwen3-VL-Moe bug are different issues." + }, + { + "left": "issue:22355", + "right": "issue:27957", + "accept": false, + "reason": "Missing transformers.onnx import and XLMRoberta Flash Attention 2 are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but they are missing different exported symbols and likely separate fixes." + }, + { + "left": "issue:43479", + "right": "issue:44913", + "accept": false, + "reason": "Default multimodal config initialization and GPTNeoX rotary_pct reload behavior are unrelated." + }, + { + "left": "issue:43408", + "right": "issue:46402", + "accept": false, + "reason": "sam3_video/sam3_tracker warning and missing Sam3Model text encoder weights are different SAM3-family problems." } ] }, @@ -13857,13 +13886,12 @@ "issue:43122", "issue:43232", "issue:43408", - "issue:43479", "issue:43550", "issue:43576", "issue:43646", "issue:43756", "issue:43784", - "issue:43824", + "issue:43937", "issue:43976", "issue:44060", "issue:44117", @@ -13874,10 +13902,11 @@ "issue:44297", "issue:44368", "issue:44373", + "issue:44488", "issue:44568", "issue:44704", - "issue:44913", "issue:44945", + "issue:44964", "issue:45278", "issue:45593", "issue:45600", @@ -13885,14 +13914,9 @@ "issue:46002", "issue:46132", "issue:46305", - "issue:46369", - "issue:46402" + "issue:46369" ], "soft_pairs": [ - "issue:22355|issue:27957", - "issue:41628|issue:43824", - "issue:43479|issue:44913", - "issue:43408|issue:46402", "issue:22355|issue:28218", "issue:42831|issue:45753", "issue:42757|issue:45278", @@ -13903,13 +13927,13 @@ "issue:43232|issue:45593", "issue:43408|issue:46369", "issue:42886|issue:42994", - "issue:44188|issue:44373", "issue:42796|issue:43122", + "issue:44188|issue:44373", "issue:43576|issue:43784", "issue:43646|issue:46002", "issue:42371|issue:44263", - "issue:44261|issue:44373", "issue:43550|issue:46132", + "issue:44261|issue:44373", "issue:44295|issue:44297", "issue:42907|issue:44945", "issue:44117|issue:44704", @@ -13917,210 +13941,207 @@ "issue:43756|issue:44060", "issue:41628|issue:45278", "issue:43784|issue:45600", - "issue:43976|issue:44945" + "issue:43976|issue:44945", + "issue:39401|issue:44488", + "issue:42907|issue:43937", + "issue:42907|issue:44964" ], - "prepared_review_unit_hash": "ee080a2167e4b3e2ec5ecf1ba6ca0e958f853ff70e491b6ef94025a3254d1b8d", + "prepared_review_unit_hash": "3e72d286063f1dfa75f3aa5eccc4a35c0adf7c1e4a2932f5f5860157acc2213f", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 35966, - "estimated_input_tokens": 8992, - "estimated_eval_tokens": 18240 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35025, + "estimated_input_tokens": 8757, + "estimated_eval_tokens": 17770 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly heterogeneous; the soft pairs look like loosely related Transformer regressions, but not the same underlying bug/change. No PRs are present.", - "confidence": 0.63, - "canonical_issue_reason": "issue:45278 is the broadest umbrella report here, covering multiple import failures after a version upgrade, so it is the best representative of the small import-regression subtheme in this cluster.", + "summary": "This cluster is heterogeneous: it mixes unrelated tokenizer, import, generation, attention, and model-specific bugs. No soft edge looks like the same underlying bug/change, so I rejected all candidate merges.", + "confidence": 0.36, + "canonical_issue_reason": "issue:45278 is the broadest representative here: a general import-regression report after upgrading Transformers, which overlaps the most with the other import-related items. It is still not a duplicate of the others, just the best anchor for this noisy cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the most generally useful issue to anchor triage because it describes a wide import-breakage regression rather than a single model- or tokenizer-specific symptom.", + "best_issue_reason": "issue:45278 is the best single issue to represent the cluster because it is the most generic and highest-level failure mode among the listed items. It is the closest thing to a cluster hub, even though the overall set is not a true duplicate group.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:22355", - "right": "issue:27957", - "accept": false, - "reason": "Different problems: missing transformers.onnx module vs XLMRoberta Flash Attention 2 behavior." - }, - { - "left": "issue:41628", - "right": "issue:43824", - "accept": false, - "reason": "Both are import errors, but for different symbols and likely different code paths." - }, - { - "left": "issue:43479", - "right": "issue:44913", - "accept": false, - "reason": "Both involve config defaults, but one is multimodal config init and the other is GPTNeoX rotary_pct reload semantics." - }, - { - "left": "issue:43408", - "right": "issue:46402", - "accept": false, - "reason": "Same SAM3 family, but one is a model-type warning and the other is missing text encoder weights." - }, { "left": "issue:22355", "right": "issue:28218", "accept": false, - "reason": "Unrelated: module import failure vs tokenizer spacing after added token." + "reason": "Different failure classes: missing module import vs tokenizer spacing behavior." }, { "left": "issue:42831", "right": "issue:45753", "accept": false, - "reason": "Different bugs: FineGrainedFP8 accuracy vs an image-triggered infinite loop in Qwen3_5." + "reason": "Unrelated problems: FP8 accuracy regression vs image-specific infinite loop." }, { "left": "issue:42757", "right": "issue:45278", "accept": false, - "reason": "Both mention imports, but one is a specific huggingface_hub symbol and the other is a broad upgrade regression." + "reason": "Both are import-breakage reports, but one is a specific missing hub symbol and the other is a broad upgrade-wide import failure; not the same concrete bug." }, { "left": "issue:39401", "right": "issue:44568", "accept": false, - "reason": "Both are tokenizer issues, but they affect different models and different tokenization behaviors." + "reason": "Both concern tokenization, but they affect different models and different symptoms (offset mapping vs special-token insertion)." }, { "left": "issue:27957", "right": "issue:30990", "accept": false, - "reason": "Flash Attention on XLMRoberta is not the same as Sentence Transformers getting stuck while loading." + "reason": "Different subsystems and symptoms: Flash Attention 2 on XLM-Roberta vs loading hangs in Sentence Transformers." }, { "left": "issue:44368", "right": "issue:46305", "accept": false, - "reason": "Different subsystems: tied-weights warning in Qwen3.5 fine-tuning vs a MoonshineStreaming mask off-by-one." + "reason": "Unrelated: a tied-embeddings warning in fine-tuning vs a streaming sliding-window off-by-one bug." }, { "left": "issue:42886", "right": "issue:43066", "accept": false, - "reason": "Offline cache loading failure is distinct from the v5 tokenizer decoder-type mismatch." + "reason": "Both are tokenizer-related, but cache/offline loading and decoder-type mismatch are distinct code paths." }, { "left": "issue:43232", "right": "issue:45593", "accept": false, - "reason": "Generation kwargs after sync_gpus is unrelated to D-FINE auxiliary-loss behavior." + "reason": "Generation kwargs handling and D-FINE auxiliary-loss behavior are unrelated." }, { "left": "issue:43408", "right": "issue:46369", "accept": false, - "reason": "Both concern SAM3/SAM3-lite text models, but one is a model-type warning and the other is memory overflow." + "reason": "Same broad model family (sam3) is not enough; one is a type-mismatch warning, the other is a memory overflow." }, { "left": "issue:42886", "right": "issue:42994", "accept": false, - "reason": "Both involve model/tokenizer saving or loading, but one is offline cache loading and the other is quantized model saving." + "reason": "Offline tokenizer cache loading and quantized model saving are different bugs." }, { - "left": "issue:44188", - "right": "issue:44373", + "left": "issue:42796", + "right": "issue:43122", "accept": false, - "reason": "Attention-kernel divergence under torch.compile is unrelated to a docstring typo." + "reason": "The former is a tokenizer BPE-marker bug for one model; the latter is a general tokenization-difference regression across versions." }, { - "left": "issue:42796", - "right": "issue:43122", + "left": "issue:44188", + "right": "issue:44373", "accept": false, - "reason": "Both are tokenizer regressions, but one is BPE-marker handling and the other is a version-to-version tokenization difference." + "reason": "Attention-kernel divergence under torch.compile is unrelated to a docstring issue." }, { "left": "issue:43576", "right": "issue:43784", "accept": false, - "reason": "Broken CLI env command is not the same as a NameError importing sentence-transformers." + "reason": "A broken CLI/env command is not the same bug as an import-time NameError in sentence-transformers." }, { "left": "issue:43646", "right": "issue:46002", "accept": false, - "reason": "Custom model initialization breakage is unrelated to NaNs in classification heads." + "reason": "Custom model initialization breakage and NaNs in classification heads are different failure modes." }, { "left": "issue:42371", "right": "issue:44263", "accept": false, - "reason": "TF32 settings guidance is unrelated to torch.split return-value handling in GlmMoeDsaIndexer." + "reason": "TF32 API settings guidance and torch.split behavior in GlmMoeDsaIndexer are unrelated." }, { - "left": "issue:44261", - "right": "issue:44373", + "left": "issue:43550", + "right": "issue:46132", "accept": false, - "reason": "Precision error from missing rms_norm_eps is unrelated to the position_ids docstring." + "reason": "Both involve attention/compile concepts, but they do not describe the same code-path problem." }, { - "left": "issue:43550", - "right": "issue:46132", + "left": "issue:44261", + "right": "issue:44373", "accept": false, - "reason": "Both touch attention-related code, but one is a torch.compile SDPA failure and the other is a register API behavior change." + "reason": "A precision/config bug in MLA layernorm is unrelated to a docstring correction." }, { "left": "issue:44295", "right": "issue:44297", "accept": false, - "reason": "Reading a buffer-registered position_ids tensor is different from tokenizer_class mismatch on save." + "reason": "Serialization/buffer handling and tokenizer_class metadata mismatch are separate issues." }, { "left": "issue:42907", "right": "issue:44945", "accept": false, - "reason": "Quantized model saving failure is unrelated to incorrect pipeline-parallel LLM output." + "reason": "Quantized-model save failures and pipeline-parallel output errors are unrelated." }, { "left": "issue:44117", "right": "issue:44704", "accept": false, - "reason": "Tokenizer mapping None handling and AutoProcessor kwargs forwarding are different from_pretrained bugs." + "reason": "Both are loading-path issues, but one is tokenizer mapping assumptions and the other is AutoProcessor kwargs propagation." }, { "left": "issue:30990", "right": "issue:33357", "accept": false, - "reason": "Stuck loading and MacOS bus error are different failure modes with different triggers." + "reason": "Both mention loading, but they are different models and different failure types (stuck load vs bus error)." }, { "left": "issue:43756", "right": "issue:44060", "accept": false, - "reason": "Smollm3 RoPE-layer configuration is not the same as Qwen3-Next tied-weights warning." + "reason": "RoPE-layer misconfiguration and a tied-weights warning are not the same bug." }, { "left": "issue:41628", "right": "issue:45278", "accept": false, - "reason": "Both are import regressions, but 45278 is broad and version-wide while 41628 is a single missing symbol." + "reason": "Missing AutoImageProcessor export is a specific API regression; the other is a broad import-error report." + }, + { + "left": "issue:43784", + "right": "issue:45600", + "accept": false, + "reason": "Import-time NameError in sentence-transformers is unrelated to auto_mappings CI breakage." + }, + { + "left": "issue:43976", + "right": "issue:44945", + "accept": false, + "reason": "Python version compatibility and pipeline-parallel output correctness are separate issues." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "Both touch tokenization/model loading, but the reported bugs are different and model-specific." }, { - "left": "issue:43784", - "right": "issue:45600", + "left": "issue:42907", + "right": "issue:43937", "accept": false, - "reason": "Import NameError in sentence-transformers is unrelated to stale auto_mappings references for Sam3LiteText." + "reason": "Quantized save failure and invalid GenerationConfig are unrelated." }, { - "left": "issue:43976", - "right": "issue:44945", + "left": "issue:42907", + "right": "issue:44964", "accept": false, - "reason": "Python version compatibility breakage is unrelated to pipeline parallelism output corruption." + "reason": "Saving a dequantized model and failing to load a multimodal model are different code paths." } ] }, @@ -14163,6 +14184,7 @@ "issue:43452", "issue:43475", "issue:43493", + "issue:43611", "issue:43824", "issue:43937", "issue:44038", @@ -14173,7 +14195,6 @@ "issue:44351", "issue:44360", "issue:44485", - "issue:44488", "issue:44521", "issue:44625", "issue:44704", @@ -14181,7 +14202,6 @@ "issue:44829", "issue:44938", "issue:44945", - "issue:44964", "issue:45072", "issue:45200", "issue:45216", @@ -14189,21 +14209,19 @@ "issue:45290", "issue:45478", "issue:45753", + "issue:46002", "issue:46326", "issue:46328", "issue:46614" ], "soft_pairs": [ - "issue:39401|issue:44488", - "issue:42907|issue:43937", - "issue:42907|issue:44964", "issue:44360|issue:44485", "issue:44704|issue:45290", - "issue:39401|issue:44521", "issue:44625|issue:44829", + "issue:39401|issue:44521", "issue:43122|issue:46326", - "issue:43122|issue:44521", "issue:44038|issue:44297", + "issue:43122|issue:44521", "issue:43116|issue:45200", "issue:43408|issue:43475", "issue:39401|issue:46614", @@ -14221,232 +14239,228 @@ "issue:22355|issue:29127", "issue:43493|issue:45753", "issue:43937|issue:44485", - "issue:44188|issue:46328", "issue:38617|issue:45278", + "issue:44188|issue:46328", "issue:42898|issue:46614", "issue:42673|issue:45072", - "issue:43066|issue:44779" + "issue:43066|issue:44779", + "issue:43611|issue:46002", + "issue:43116|issue:43493" ], - "prepared_review_unit_hash": "2771944a4e06d4cd5d142327b9ce1aca3bd6e15b0eaa29ea90a3c43899d4f4dd", + "prepared_review_unit_hash": "13870aa0146acb441fa079205f9701b9c4e9310ea4b74668a9db7de1f3f0aae6", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 32, - "serialized_chars": 36975, - "estimated_input_tokens": 9244, - "estimated_eval_tokens": 18744 + "soft_pair_count": 31, + "serialized_chars": 36663, + "estimated_input_tokens": 9166, + "estimated_eval_tokens": 18588 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated Transformers issues are grouped together here. A loose tokenizer/v5-regression theme appears in a few titles, but none of the soft pairs look like the same concrete bug or change, so I would reject all soft edges.", - "confidence": 0.87, - "canonical_issue_reason": "Issue 43122 is the best representative only in a very loose sense: it is a broad tokenizer regression report with downstream references, and it sits near the few tokenizer-related titles in this set. That said, the cluster is too heterogeneous to have a true canonical duplicate issue.", + "summary": "This cluster is very heterogeneous: import-surface regressions, tokenizer mismatches, multimodal chat-template bugs, model-family-specific issues, and training/runtime problems. I do not see any soft pair that is clearly the same underlying bug/change, so all soft edges should be rejected. If forced to pick an umbrella issue, the broad 4.57\u21925.5 import-regression report is the most representative, but only for a small subset of the set.", + "confidence": 0.84, + "canonical_issue_reason": "Issue 45278 is the broadest regression report here and is the closest thing to an umbrella for the import-error subset, though the overall cluster is still too mixed to make it a true duplicate hub.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43122 is the strongest global issue candidate because it is the broadest regression report among the mostly tokenizer-related subset and has the most chance of serving as a central reference point. It is still not a strong duplicate-canonical choice for the full cluster.", + "best_issue_reason": "Issue 45278 is the most general, regression-oriented issue among the set and is the best single representative for triage, even if it does not cover the full cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:39401", - "right": "issue:44488", - "accept": false, - "reason": "Different bugs: Qwen3 tokenizer offset mapping vs loading a specific model. Same broad area, not the same underlying issue." - }, - { - "left": "issue:42907", - "right": "issue:43937", - "accept": false, - "reason": "Unrelated model-save bug vs GenerationConfig validation error." - }, - { - "left": "issue:42907", - "right": "issue:44964", - "accept": false, - "reason": "Different code paths: dequantized save_pretrained vs loading Phi-4 multimodal." - }, { "left": "issue:44360", "right": "issue:44485", "accept": false, - "reason": "Completely different problems: DSA indexer discussion vs GLM-5 RoPE implementation." + "reason": "Different bugs in different areas: DSA indexer/ReLU discussion vs GLM-5 RoPE implementation." }, { "left": "issue:44704", "right": "issue:45290", "accept": false, - "reason": "Both touch processor/chat-template APIs, but the failures are different and not the same concrete bug." - }, - { - "left": "issue:39401", - "right": "issue:44521", - "accept": false, - "reason": "Tokenizer offset mapping bug vs multimodal assistant mask generation bug." + "reason": "Different code paths: cached_file kwargs handling vs apply_chat_template crash on tool-call messages." }, { "left": "issue:44625", "right": "issue:44829", "accept": false, - "reason": "Qwen3.5 num_labels propagation and flash_attention_3 training degeneration are different issues." + "reason": "Different model behaviors: num_labels propagation vs flash_attention_3 training degeneration." }, { - "left": "issue:43122", - "right": "issue:46326", + "left": "issue:39401", + "right": "issue:44521", "accept": false, - "reason": "Both are tokenizer-related, but one is a v5 tokenization regression and the other is continue_final_message behavior for mistral_common." + "reason": "Both involve tokenization, but one is offset_mapping and the other is assistant_masks for multimodal chat templates." }, { "left": "issue:43122", - "right": "issue:44521", + "right": "issue:46326", "accept": false, - "reason": "Different symptoms and code paths: tokenization drift vs multimodal assistant masks." + "reason": "Both are tokenizer regressions, but the symptoms and affected APIs are different (tokenization output vs continue_final_message)." }, { "left": "issue:44038", "right": "issue:44297", "accept": false, - "reason": "Unrelated: Qwen3-VL-Moe bug vs tokenizer_config serialization mismatch." + "reason": "Different failures: Qwen3-VL-Moe loading vs tokenizer save_pretrained metadata mismatch." + }, + { + "left": "issue:43122", + "right": "issue:44521", + "accept": false, + "reason": "Different problems: text tokenization drift vs multimodal assistant mask generation." }, { "left": "issue:43116", "right": "issue:45200", "accept": false, - "reason": "Different model/task bugs: multi-label example script output vs Gemma 4 text-only fine-tuning defaults." + "reason": "Unrelated features: example-script multi-label classification vs Gemma 4 mm_token_type_ids handling." }, { "left": "issue:43408", "right": "issue:43475", "accept": false, - "reason": "Same general SAM 3 area, but a model-type warning and a missing attribute are not the same underlying bug." + "reason": "Same model family, but one is a config warning and the other is an attribute error in vision encoder outputs." }, { "left": "issue:39401", "right": "issue:46614", "accept": false, - "reason": "Different tokenizer/phonemizer bugs affecting different models and behaviors." + "reason": "Different tokenizers and symptoms: Qwen3 offset mapping vs Wav2Vec2 phonemizer backend state leakage." }, { "left": "issue:44038", "right": "issue:44295", "accept": false, - "reason": "Different failures: Qwen3-VL-Moe bug vs position_ids buffer access error." + "reason": "Different bug classes: Qwen3-VL-Moe import/loading issue vs position_ids buffer access error." }, { "left": "issue:42907", "right": "issue:43452", "accept": false, - "reason": "Different lifecycle paths: saving dequantized models vs gguf loading via AutoTokenizer/AutoModel." + "reason": "Different operations and artifacts: dequantized save_pretrained vs gguf_file loading through AutoTokenizer/AutoModel." }, { "left": "issue:44038", "right": "issue:44222", "accept": false, - "reason": "Both involve model saving/quantization, but the concrete bugs are different." + "reason": "Both mention Qwen3-VL-Moe, but one is a loading/import issue and the other is FP8 save_pretrained behavior." }, { "left": "issue:22355", "right": "issue:29159", "accept": false, - "reason": "No module import error for transformers.onnx vs slow/fast tokenizer inconsistency." + "reason": "No real overlap: ONNX module import failure vs slow/fast tokenizer inconsistency." }, { "left": "issue:45216", "right": "issue:45478", "accept": false, - "reason": "Both concern Qwen3.5 regressions, but one is save_pretrained checkpoint correctness and the other is from_pretrained loading failure." + "reason": "Same model family/version area, but save_pretrained checkpoint corruption and from_pretrained error are distinct code paths." }, { "left": "issue:43824", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but they involve different missing symbols and likely different fixes." + "reason": "Both are missing imports, but for different symbols and likely different export paths." }, { "left": "issue:43122", "right": "issue:44779", "accept": false, - "reason": "Both mention tokenizer regressions in v5, but the affected models and symptoms are different." + "reason": "Both are tokenizer regressions in v5, but the affected models and reported failures are different." }, { "left": "issue:42757", "right": "issue:44351", "accept": false, - "reason": "Shared import-error pattern only; different packages/symbols, not the same bug." + "reason": "Different missing imports from different modules; no evidence of the same underlying fix." }, { "left": "issue:36296", "right": "issue:43116", "accept": false, - "reason": "Tensor parallel training bug vs example script multi-label classification bug are unrelated." + "reason": "Completely unrelated: tensor-parallel training bug vs an example script classification issue." }, { "left": "issue:43278", "right": "issue:44945", "accept": false, - "reason": "Different parallelism/dtype issues with different symptoms and code paths." + "reason": "Different execution contexts: dtype drift in eval vs incorrect output under pipeline parallelism." }, { "left": "issue:42175", "right": "issue:44938", "accept": false, - "reason": "Backend extra-install issue vs Python 3.14 load failure are unrelated." + "reason": "Unrelated environment/setup issues: backend extras install vs Python 3.14 load failure." }, { "left": "issue:22355", "right": "issue:29127", "accept": false, - "reason": "No module named transformers.onnx vs LayoutLMv3 error-message clarity are unrelated." + "reason": "Different failures: missing transformers.onnx module vs LayoutLMv3 error-message clarity." }, { "left": "issue:43493", "right": "issue:45753", "accept": false, - "reason": "Different vision-model bugs: SigLIP2 implementation discrepancy vs Qwen3_5 image infinite loop." + "reason": "Both concern image/multimodal models, but one is a SigLIP2 implementation discrepancy and the other is an infinite-loop bug in Qwen3_5." }, { "left": "issue:43937", "right": "issue:44485", "accept": false, - "reason": "Same model family (GLM-5) but different concrete problems: GenerationConfig validation vs RoPE implementation." + "reason": "Same GLM-5 branding, but GenerationConfig validation and RoPE implementation are distinct problems." }, { - "left": "issue:44188", - "right": "issue:46328", + "left": "issue:38617", + "right": "issue:45278", "accept": false, - "reason": "Attention kernel divergence under torch.compile vs SequenceFeatureExtractor.pad inefficiency are unrelated." + "reason": "Both are import regressions, but one is a specific missing symbol and the other is a broad upgrade-wide import-error report; not enough to call them the same bug." }, { - "left": "issue:38617", - "right": "issue:45278", + "left": "issue:44188", + "right": "issue:46328", "accept": false, - "reason": "Both are import-related regressions, but the specific missing symbol and the broader upgrade failure are not clearly the same issue." + "reason": "Different subsystems: torch.compile attention-kernel branching vs SequenceFeatureExtractor.pad performance." }, { "left": "issue:42898", "right": "issue:46614", "accept": false, - "reason": "Tokenizer cleanup behavior change vs stale phonemizer backend state are different bugs." + "reason": "Tokenizer-related, but one is a v5 behavior change in cleanup and the other is stale phonemizer language state." }, { "left": "issue:42673", "right": "issue:45072", "accept": false, - "reason": "VRAM leak in multi-threaded Qwen3ForCausalLM use vs dtype mismatches in bfloat16 inference are unrelated." + "reason": "Different failure modes: VRAM leak in multi-threaded inference vs dtype mismatches in bfloat16 inference." }, { "left": "issue:43066", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer v5 regressions, but one is decoder type selection and the other is Deepseek tokenization correctness; not the same concrete bug." + "reason": "Both are tokenizer regressions in v5, but the specific behavior changes and affected models are different." + }, + { + "left": "issue:43611", + "right": "issue:46002", + "accept": false, + "reason": "Unrelated: base_model_prefix loading breakage vs NaNs during classification-head init/checkout." + }, + { + "left": "issue:43116", + "right": "issue:43493", + "accept": false, + "reason": "Different domains: text classification example behavior vs multimodal model discrepancy." } ] }, @@ -14481,8 +14495,6 @@ "issue:43116", "issue:43441", "issue:43450", - "issue:43493", - "issue:43611", "issue:43673", "issue:43825", "issue:43827", @@ -14503,6 +14515,7 @@ "issue:44704", "issue:44779", "issue:44936", + "issue:44977", "issue:45071", "issue:45083", "issue:45356", @@ -14520,13 +14533,11 @@ "issue:46640" ], "soft_pairs": [ - "issue:43611|issue:46002", - "issue:43116|issue:43493", "issue:44355|issue:46231", "issue:42371|issue:43874", "issue:43976|issue:45446", - "issue:43992|issue:44704", "issue:44038|issue:45478", + "issue:43992|issue:44704", "issue:44625|issue:46002", "issue:44075|issue:44936", "issue:28348|issue:46640", @@ -14545,220 +14556,215 @@ "issue:45071|issue:46614", "issue:43673|issue:46328", "issue:44291|issue:45071", + "issue:43450|issue:43986", "issue:44466|issue:46002", "issue:42673|issue:43906", - "issue:43450|issue:43986", "issue:39401|issue:44779", - "issue:45405|issue:45446" + "issue:45405|issue:45446", + "issue:43937|issue:44977" ], - "prepared_review_unit_hash": "1b82c631b77b2df79855ae424e24147039cbf149e6838a4a0567bd18d264e634", + "prepared_review_unit_hash": "029cb6b74f68d35ab787ade6b15dafd840cb7eb270354ea87312c3c8479cdc5e", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36567, - "estimated_input_tokens": 9142, - "estimated_eval_tokens": 18540 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35669, + "estimated_input_tokens": 8918, + "estimated_eval_tokens": 18092 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: the items span unrelated tokenizer, loading, generation, vision, optimizer, docs, and version-compatibility bugs. The soft pairs are mostly superficial subsystem matches rather than the same underlying defect, so none should be merged as duplicates.", - "confidence": 0.91, - "canonical_issue_reason": "No reliable canonical issue: the cluster spans unrelated bugs across tokenizers, generation, model loading, docs, and vision.", + "summary": "Mostly false-positive similarity links: the pairs span unrelated tokenizer, attention, processor, training, docs, and config bugs. I would not merge any soft edge. If one issue must represent the set, issue:45478 is the clearest concrete runtime regression in the Qwen/MoE loading-failure subset.", + "confidence": 0.39, + "canonical_issue_reason": "issue:45478 is the most concrete, user-visible runtime failure here, with a clearly scoped from_pretrained regression and a model-specific failure path.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45478 is the most concrete, actionable bug report in the set (Qwen3.5 MoE from_pretrained error), so it is the best anchor, though not a true duplicate representative.", + "best_issue_reason": "It is specific, reproducible, and easy to triage as a real bug; among these heterogeneous issues it is the strongest standalone representative.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43611", - "right": "issue:46002", - "accept": false, - "reason": "Base-model-prefix loading breakage and NaNs in classification heads are different failure modes and code paths." - }, - { - "left": "issue:43116", - "right": "issue:43493", - "accept": false, - "reason": "Example-script multi-label empties and SigLIP2 implementation drift are unrelated bugs." - }, { "left": "issue:44355", "right": "issue:46231", "accept": false, - "reason": "Compiled Python file errors and a doc parameter-name mismatch are not the same issue." + "reason": "Compiled-file import errors and a docs parameter-name mismatch are unrelated." }, { "left": "issue:42371", "right": "issue:43874", "accept": false, - "reason": "TF32 API guidance and a missing image-patch method in GLM46V are unrelated." + "reason": "TF32 behavior settings and a missing image-processor method affect different code paths." }, { "left": "issue:43976", "right": "issue:45446", "accept": false, - "reason": "General Python version compatibility and a flex_attention import version check are different problems." + "reason": "Python-version compatibility and a flex_attention import/version check are not the same bug." }, { - "left": "issue:43992", - "right": "issue:44704", + "left": "issue:44038", + "right": "issue:45478", "accept": false, - "reason": "Missing embed token weights in UMT5 and AutoProcessor kwargs forwarding are unrelated loading bugs." + "reason": "Both touch Qwen MoE models, but the affected models, APIs, and failure modes differ." }, { - "left": "issue:44038", - "right": "issue:45478", + "left": "issue:43992", + "right": "issue:44704", "accept": false, - "reason": "Both involve Qwen MoE models, but the model variants and underlying from_pretrained failures are not the same bug." + "reason": "Missing model weights and AutoProcessor kwarg forwarding are unrelated loading issues." }, { "left": "issue:44625", "right": "issue:46002", "accept": false, - "reason": "Qwen3.5 num_labels propagation and NaNs in classification heads are distinct issues." + "reason": "Label propagation and NaNs in classification heads are different defects." }, { "left": "issue:44075", "right": "issue:44936", "accept": false, - "reason": "Optimizer argument handling and trainer.evaluate() failing after train() are different training bugs." + "reason": "Optimizer args being ignored and evaluate-after-train failure are separate trainer bugs." }, { "left": "issue:28348", "right": "issue:46640", "accept": false, - "reason": "FlashAttention 2 support for GPT2LMHeadModel is not the same as a generic Flash Attention 2 issue." + "reason": "Feature request for GPT2 FlashAttention 2 support is not the same as a vague Flash Attention bug report." }, { "left": "issue:45357", "right": "issue:45478", "accept": false, - "reason": "save_pretrained key serialization and from_pretrained loading failures are separate paths with different symptoms." + "reason": "Both are Qwen3.5-related, but one is save_pretrained key serialization and the other is from_pretrained loading failure." }, { "left": "issue:42222", "right": "issue:43441", "accept": false, - "reason": "Vitpose model breakage and Ministral FlashAttention failures are unrelated model-specific bugs." + "reason": "VitPose breakage and a Ministral flash-attention failure are unrelated." }, { "left": "issue:45083", "right": "issue:45446", "accept": false, - "reason": "A qwen3_omni_moe helper-function bug and a flex_attention version check are unrelated." + "reason": "Feature-extractor length math and a PyTorch version guard in flex_attention are different issues." }, { "left": "issue:43116", "right": "issue:43825", "accept": false, - "reason": "Multi-label empty results and a stale pipeline() translation error message are unrelated." + "reason": "Multi-label example output failure and a pipeline translation error message are not the same bug." }, { "left": "issue:42673", "right": "issue:42796", "accept": false, - "reason": "VRAM leakage in Qwen3ForCausalLM and tokenizer BPE-marker handling are unrelated." + "reason": "A VRAM leak in dataloader threads is unrelated to tokenizer BPE-marker handling." }, { "left": "issue:43064", "right": "issue:43673", "accept": false, - "reason": "FSDP2/PEFT optimizer-state corruption and chunked_prefill cache handling are different code paths." + "reason": "FSDP/PEFT optimizer-state corruption and GenerationMixin cache handling are different code paths." }, { "left": "issue:43937", "right": "issue:45563", "accept": false, - "reason": "Invalid GenerationConfig and a stale paged-generate warning are not the same bug." + "reason": "An invalid GenerationConfig error and a stale warning for num_return_sequences are different symptoms." }, { "left": "issue:43827", "right": "issue:45446", "accept": false, - "reason": "Documentation cleanup after pipeline removals and a PyTorch import version check are unrelated." + "reason": "Documentation cleanup after pipeline removals does not match a flex_attention version-check bug." }, { "left": "issue:44380", "right": "issue:46077", "accept": false, - "reason": "GPT2 attention scaling ignored in backends and Gemma4 bidirectional masks are different attention bugs." + "reason": "GPT2 backend scaling being ignored and Gemma4 bidirectional-mask construction are distinct attention bugs." }, { "left": "issue:33453", "right": "issue:39401", "accept": false, - "reason": "Tokenizer loading regression and Qwen3 offset_mapping errors are distinct tokenizer issues." + "reason": "Both are tokenizer regressions, but loading regression and offset_mapping correctness are different defects." }, { "left": "issue:45356", "right": "issue:46614", "accept": false, - "reason": "Kimi-K2.5 codec regression and stale phonemizer language state are unrelated tokenizer/processor bugs." + "reason": "Different tokenizer/model families and different failure modes; not the same underlying bug." }, { "left": "issue:44117", "right": "issue:45588", "accept": false, - "reason": "Tokenizer mapping None handling and a flash_attention crash on s_aux=None are unrelated." + "reason": "Tokenizer mapping None and a flash_attention crash on s_aux=None are unrelated." }, { "left": "issue:45071", "right": "issue:46614", "accept": false, - "reason": "PretrainedConfig type checking and phonemizer backend caching are unrelated regressions." + "reason": "PretrainedConfig type-checking and stale phonemizer backend state are unrelated." }, { "left": "issue:43673", "right": "issue:46328", "accept": false, - "reason": "Generation cache handling and SequenceFeatureExtractor.pad performance are different concerns." + "reason": "Generation cache absence and a padding performance issue are not the same bug." }, { "left": "issue:44291", "right": "issue:45071", "accept": false, - "reason": "init_empty_weights argument handling and PretrainedConfig type checking are separate v5 compatibility issues." + "reason": "Unexpected init_empty_weights argument handling and config type-checking are different regressions." + }, + { + "left": "issue:43450", + "right": "issue:43986", + "accept": false, + "reason": "Batched video-processor shape issues and missing torchvision crash behavior are related only loosely by area." }, { "left": "issue:44466", "right": "issue:46002", "accept": false, - "reason": "lm_head serialization inconsistency and classification-head NaNs are unrelated." + "reason": "Serialization of tied lm_head weights and NaNs in classification heads are unrelated." }, { "left": "issue:42673", "right": "issue:43906", "accept": false, - "reason": "VRAM leak in Qwen3ForCausalLM and a reproduction of another issue are not duplicates." - }, - { - "left": "issue:43450", - "right": "issue:43986", - "accept": false, - "reason": "Batched video processor shape errors and missing torchvision crash handling are related to video input, but not the same concrete bug." + "reason": "An isolated reproduction of another issue is not enough to show the same VRAM-leak defect." }, { "left": "issue:39401", "right": "issue:44779", "accept": false, - "reason": "Qwen3 offset mapping corruption and Deepseek tokenizer regression are distinct model-specific tokenizer bugs." + "reason": "Both are tokenizer bugs, but Qwen3 offset mapping and DeepSeek incorrect results are different model-specific regressions." }, { "left": "issue:45405", "right": "issue:45446", "accept": false, - "reason": "A PEFT minimum-version bump and a flex_attention import check are unrelated release/compatibility issues." + "reason": "A PEFT version bump and a flex_attention import/version check are unrelated." + }, + { + "left": "issue:43937", + "right": "issue:44977", + "accept": false, + "reason": "Invalid GenerationConfig and flash-attention generation failure are not the same concrete bug." } ] }, @@ -14811,7 +14817,6 @@ "issue:44829", "issue:44877", "issue:44933", - "issue:44977", "issue:45071", "issue:45081", "issue:45216", @@ -14832,18 +14837,17 @@ "issue:46617" ], "soft_pairs": [ - "issue:43937|issue:44977", "issue:44410|issue:45310", + "issue:44829|issue:45216", "issue:45341|issue:46328", "issue:45341|issue:45561", - "issue:44829|issue:45216", "issue:45071|issue:46512", - "issue:44117|issue:45081", "issue:38617|issue:41628", + "issue:44117|issue:45081", "issue:43122|issue:46298", "issue:42617|issue:45341", - "issue:43408|issue:43742", "issue:42222|issue:46178", + "issue:43408|issue:43742", "issue:45362|issue:46328", "issue:42371|issue:44261", "issue:44877|issue:46231", @@ -14852,8 +14856,8 @@ "issue:44484|issue:45310", "issue:39401|issue:46209", "issue:43937|issue:44496", - "issue:38617|issue:44220", "issue:44933|issue:45278", + "issue:38617|issue:44220", "issue:44410|issue:45478", "issue:44075|issue:45656", "issue:43066|issue:43122", @@ -14863,214 +14867,208 @@ "issue:42175|issue:42222", "issue:43673|issue:45341" ], - "prepared_review_unit_hash": "e6def4a086582952c1779dd0f08cacea45b6ce97e8372a6b5eb22af26d02ffcc", + "prepared_review_unit_hash": "304be38acb463c53e76698a036be56eea7d21d2001466e19d8e404a611e64227", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36238, - "estimated_input_tokens": 9060, - "estimated_eval_tokens": 18376 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35382, + "estimated_input_tokens": 8846, + "estimated_eval_tokens": 17948 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose collection of unrelated Transformers regression reports (imports, tokenizer behavior, config/loading, backend issues, and model-specific bugs). None of the soft pairs look like the same concrete underlying bug/change, so I would not merge any of them as duplicates.", - "confidence": 0.67, - "canonical_issue_reason": "issue:45278 is the broadest user-facing regression report here and best fits a general duplicate target for version-upgrade import breakages.", + "summary": "This set is mostly a grab bag of unrelated bugs: tokenizer regressions, import errors, backend/config loading issues, testing-utils problems, and model-specific failures. I don\u2019t see any soft pair that clearly describes the same underlying bug or change.", + "confidence": 0.9, + "canonical_issue_reason": "Issue 45561 is the most clearly scoped and actionable open bug in the set: a concrete pytest-xdist race in patched testing utilities, with reproducible behavior and active discussion.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the most reusable issue to point to when triaging broad 5.x compatibility/import failures.", + "best_issue_reason": "Issue 45561 is the best representative issue because it has the clearest failure mode, strong reproducibility, and multiple references; it is more triage-worthy than the other loosely related reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43937", - "right": "issue:44977", + "left": "issue:44410", + "right": "issue:45310", "accept": false, - "reason": "Both mention Qwen-related failures, but one is a GenerationConfig validation error and the other is a flash-attention generation bug." + "reason": "Both involve Qwen-family models, but one is a missing projection bug in qwen3next and the other is a Qwen3.5 MoE from_pretrained failure." }, { - "left": "issue:44410", - "right": "issue:45310", + "left": "issue:44829", + "right": "issue:45216", "accept": false, - "reason": "Same model family, but one reports missing projections in layer 0 while the other is a from_pretrained loading error." + "reason": "Different problems: flash_attention_3 training degeneration vs a Qwen3.5 save_pretrained checkpoint regression." }, { "left": "issue:45341", "right": "issue:46328", "accept": false, - "reason": "One is a testing_utils bug; the other is a SequenceFeatureExtractor.pad performance issue." + "reason": "Unrelated: a small testing_utils bug vs a SequenceFeatureExtractor.pad performance issue." }, { "left": "issue:45341", "right": "issue:45561", "accept": false, - "reason": "Both touch testing utilities, but one is a generic bug report and the other is a specific xdist race on captured_info.txt." - }, - { - "left": "issue:44829", - "right": "issue:45216", - "accept": false, - "reason": "Training degeneration with flash_attention_3 is unrelated to a Qwen3.5 save_pretrained regression." + "reason": "Both touch testing utils, but one is a generic bug and the other is a specific xdist file race; not the same concrete failure." }, { "left": "issue:45071", "right": "issue:46512", "accept": false, - "reason": "Both involve config handling, but one is type checking and the other is stale config dtype after from_config()." + "reason": "Different regressions: PreTrainedConfig type checking vs stale model.config.dtype after from_config(dtype=...)." }, { - "left": "issue:44117", - "right": "issue:45081", + "left": "issue:38617", + "right": "issue:41628", "accept": false, - "reason": "Tokenizer mapping None and Mistral regex patch crash are different code paths and symptoms." + "reason": "Separate import errors for different symbols: layer_type_validation vs AutoImageProcessor." }, { - "left": "issue:38617", - "right": "issue:41628", + "left": "issue:44117", + "right": "issue:45081", "accept": false, - "reason": "Both are import errors, but they concern different missing symbols and different modules." + "reason": "Both are tokenizer-related, but one is a TOKENIZER_MAPPING_NAMES None-handling bug and the other is a Mistral regex patch crash." }, { "left": "issue:43122", "right": "issue:46298", "accept": false, - "reason": "A tokenization regression and a noisy logging message are unrelated issues." + "reason": "Tokenization regression vs a noisy logging message; no shared underlying bug." }, { "left": "issue:42617", "right": "issue:45341", "accept": false, - "reason": "A script execution failure and a testing_utils bug are not the same problem." + "reason": "A runtime failure running 3d_parallel.py is unrelated to a testing_utils bug." }, { - "left": "issue:43408", - "right": "issue:43742", + "left": "issue:42222", + "right": "issue:46178", "accept": false, - "reason": "A model-type warning for sam3_tracker is unrelated to a MobileLLM key error." + "reason": "Vitpose model breakage and a tensorflow_text backend-mapping error are different issues." }, { - "left": "issue:42222", - "right": "issue:46178", + "left": "issue:43408", + "right": "issue:43742", "accept": false, - "reason": "Broken vitpose models and a missing tensorflow_text backend definition are different failures." + "reason": "Different model-loading problems: sam3_video/sam3_tracker warning vs MobileLLM config/key error." }, { "left": "issue:45362", "right": "issue:46328", "accept": false, - "reason": "Qwen3.5 chat crashes and a padding performance issue do not share the same bug." + "reason": "Qwen3.5 chat crash and a padding performance issue do not share the same code-path." }, { "left": "issue:42371", "right": "issue:44261", "accept": false, - "reason": "TF32 API guidance and an rms_norm_eps precision issue are unrelated." + "reason": "TF32 API guidance is unrelated to the MLA rms_norm_eps precision bug." }, { "left": "issue:44877", "right": "issue:46231", "accept": false, - "reason": "Strict config loading for granite_speech is unrelated to a documentation parameter mismatch." + "reason": "A strict config loading bug and a documentation parameter-name typo are not the same bug." }, { "left": "issue:42490", "right": "issue:42673", "accept": false, - "reason": "A trailing-slash local model warning and a VRAM leak in dataloader threads are different bugs." + "reason": "Local-path import warning vs VRAM leak in multi-dataloader-thread use are unrelated." }, { "left": "issue:42371", "right": "issue:46617", "accept": false, - "reason": "TF32 behavior guidance is unrelated to a trust_remote_code cache FileNotFoundError." + "reason": "TF32 backend settings and trust_remote_code cache loading failures are unrelated." }, { "left": "issue:44484", "right": "issue:45310", "accept": false, - "reason": "A save_pretrained sharding question is unrelated to a Qwen3.5 MoE from_pretrained error." + "reason": "A save_pretrained shard-size question is unrelated to a Qwen3.5 MoE loading error." }, { "left": "issue:39401", "right": "issue:46209", "accept": false, - "reason": "Tokenizer offset mapping and FP8 inference on B200 are completely different issues." + "reason": "Qwen3 tokenizer offset_mapping bug and FP8 inference failure on B200 are different problems." }, { "left": "issue:43937", "right": "issue:44496", "accept": false, - "reason": "A GenerationConfig validation failure and an unrecognized-model config error are both config-related, but not the same bug." + "reason": "Invalid GenerationConfig and unrecognized model/config loading are distinct failures." }, { - "left": "issue:38617", - "right": "issue:44220", + "left": "issue:44933", + "right": "issue:45278", "accept": false, - "reason": "Different import/path failures in different modules; not the same underlying defect." + "reason": "The latter is a broad upgrade-related import regression report; the former is one specific import issue, not clearly the same bug." }, { - "left": "issue:44933", - "right": "issue:45278", + "left": "issue:38617", + "right": "issue:44220", "accept": false, - "reason": "Both mention imports, but one is a specific nonexistent symbol while the other is a broad upgrade regression report." + "reason": "An import failure and an _torch_extract_fbank_features issue are unrelated." }, { "left": "issue:44410", "right": "issue:45478", "accept": false, - "reason": "Different Qwen3.5/Qwen3next failures with no shared concrete code path." + "reason": "Similar model family, but the concrete bugs differ: missing projections vs a Qwen3.5 MoE from_pretrained error." }, { "left": "issue:44075", "right": "issue:45656", "accept": false, - "reason": "Optimizer args ignored and optimizer step called twice with deepspeed are different optimizer problems." + "reason": "Optimizer SGD args being ignored is different from optimizer.step being called twice under deepspeed." }, { "left": "issue:43066", "right": "issue:43122", "accept": false, - "reason": "Both concern tokenizer behavior in v5, but one is a decoder-type issue and the other is a broader tokenization-change report; not clearly the same concrete bug." + "reason": "Both mention tokenizer behavior in v5, but one is a wrong decoder-type issue and the other is changed tokenization output; not enough evidence they are the same code-path bug." }, { "left": "issue:42222", "right": "issue:45542", "accept": false, - "reason": "A vitpose model breakage and a missing TensorFlow backend after installing tensorboard are unrelated." + "reason": "Vitpose model breakage is unrelated to a missing TensorFlow backend setup." }, { "left": "issue:44263", "right": "issue:44485", "accept": false, - "reason": "A torch.split return-value issue and a GLM-5 RoPE implementation discussion are different code-path problems." + "reason": "torch.split return semantics and GLM-5 RoPE implementation are unrelated." }, { "left": "issue:43097", "right": "issue:44493", "accept": false, - "reason": "Removed tie_embeddings_and_encoder_decoder behavior and unexpected position-id keys are unrelated regressions." + "reason": "API removal around tie_embeddings_and_encoder_decoder is unrelated to unexpected position-id keys." }, { "left": "issue:42175", "right": "issue:42222", "accept": false, - "reason": "TensorFlow backend packaging and broken vitpose models are different issues." + "reason": "Backend packaging/install issue vs vitpose model breakage are different problems." }, { "left": "issue:43673", "right": "issue:45341", "accept": false, - "reason": "GenerationMixin cache handling and a testing_utils bug are unrelated." + "reason": "Generation cache behavior and a testing_utils bug do not share the same underlying issue." } ] }, @@ -15144,9 +15142,9 @@ "issue:46369" ], "soft_pairs": [ - "issue:43874|issue:44261", "issue:43012|issue:43577", "issue:41628|issue:45071", + "issue:43874|issue:44261", "issue:45071|issue:45542", "issue:43931|issue:44410", "issue:44263|issue:46172", @@ -15163,8 +15161,8 @@ "issue:44933|issue:46231", "issue:43874|issue:44263", "issue:44261|issue:44485", - "issue:44117|issue:44987", "issue:43673|issue:46123", + "issue:44117|issue:44987", "issue:43986|issue:44484", "issue:43531|issue:45341", "issue:45440|issue:46172", @@ -15175,235 +15173,235 @@ "issue:43097|issue:46050", "issue:44410|issue:45375", "issue:44038|issue:44410", - "issue:42890|issue:46298", - "issue:38617|issue:44351" + "issue:38617|issue:44351", + "issue:42890|issue:46298" ], - "prepared_review_unit_hash": "23f23be7431904d99e58e0cb273e0d81cf02257864e63d51c3e724d9d43af007", + "prepared_review_unit_hash": "b22905402ef03e3d5232fd2fa63a3696c62ecf10cbcface323d4cd6b02977540", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37377, - "estimated_input_tokens": 9345, - "estimated_eval_tokens": 18946 + "serialized_chars": 37375, + "estimated_input_tokens": 9344, + "estimated_eval_tokens": 18944 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: it mixes unrelated import errors, model-specific loading/shape bugs, flash-attention issues, training/runtime regressions, and a few docs/test nits. None of the soft pairs look like the same underlying defect or change.", - "confidence": 0.97, - "canonical_issue_reason": "No single issue is a safe canonical here. The cluster does not converge on one bug family; the strongest similarity is only superficial keyword overlap across unrelated reports.", - "canonical_pr_reason": "No PRs are present in this cluster.", - "best_issue_reason": "No global best issue is suitable. Picking any one issue would misrepresent the cluster because the pairwise links are mostly broad subsystem or wording similarity rather than duplicate bugs.", - "best_pr_reason": "No PRs are present in this cluster.", + "summary": "These items are mostly heterogeneous bug reports with only superficial similarity in a few v5/import/model-loading themes. None of the soft pairs look like the same underlying bug or change, so I would not merge any of them as duplicates.", + "confidence": 0.88, + "canonical_issue_reason": "issue:43097 is the broadest, most central v5-regression report in the set and has the strongest signal to serve as an anchor if a single canonical issue must be chosen.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43097 is the best representative of the recurring compatibility-breakage theme, but the cluster is not cohesive enough to justify treating the others as duplicates of it.", + "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43874", - "right": "issue:44261", - "accept": false, - "reason": "Different failures: missing image-patch method vs missing rms_norm_eps in MLA. Not the same code path." - }, { "left": "issue:43012", "right": "issue:43577", "accept": false, - "reason": "Compile-time float32 warning vs dtype not propagating on load. Related theme, but different bugs and fixes." + "reason": "Both involve dtype/precision behavior, but one is a compilation warning and the other is a model-loading dtype bug for a different architecture." }, { "left": "issue:41628", "right": "issue:45071", "accept": false, - "reason": "Importing AutoImageProcessor and PreTrainedConfig type-checking are unrelated issues." + "reason": "These are unrelated failures: one is an AutoImageProcessor import error, the other is a PreTrainedConfig type-checking regression." + }, + { + "left": "issue:43874", + "right": "issue:44261", + "accept": false, + "reason": "Different models and different failures: a missing image-processor method versus an MLA layernorm config precision issue." }, { "left": "issue:45071", "right": "issue:45542", "accept": false, - "reason": "Config type-checking breakage is not the same as the TensorFlow backend guard problem." + "reason": "A config type-checking issue and a TensorFlow backend detection issue are not the same bug." }, { "left": "issue:43931", "right": "issue:44410", "accept": false, - "reason": "Both are Qwen-related, but one is a weight-shape mismatch and the other is missing projections in a different model." + "reason": "Both are Qwen-related, but one is a weight-shape mismatch on model load and the other is a missing projection set in a different model variant." }, { "left": "issue:44263", "right": "issue:46172", "accept": false, - "reason": "torch.split return-value handling and dtype/transpose handling are different defects." + "reason": "These hit different code paths in different DeepSeek/GLM-style models; no evidence they share the same defect." }, { "left": "issue:44360", "right": "issue:46172", "accept": false, - "reason": "Missing ReLU in the DSA indexer is unrelated to the Deepseek dtype/transpose bug." + "reason": "A missing ReLU in one indexer and a dtype/transpose bug in another model are distinct issues." }, { "left": "issue:44483", "right": "issue:45464", "accept": false, - "reason": "General chat/completions request rejection vs streaming inference failure on a specific model are not the same bug." + "reason": "Both mention chat/completions, but one is a general request rejection in v5 and the other is a model-specific streaming failure." }, { "left": "issue:44484", "right": "issue:45478", "accept": false, - "reason": "A save_pretrained shard-size question is unrelated to a Qwen3.5 MoE from_pretrained error." + "reason": "One concerns save_pretrained sharding defaults; the other is a from_pretrained load error for Qwen3.5 MoE." }, { "left": "issue:27957", "right": "issue:43352", "accept": false, - "reason": "Both mention Flash Attention 2, but they concern different models and different support gaps." + "reason": "Both are Flash Attention 2 topics, but they are model-specific support/compatibility reports, not the same underlying bug." }, { "left": "issue:43441", "right": "issue:44829", "accept": false, - "reason": "Ministral FlashAttention failure and flash_attention_3 training degeneration are different code paths and symptoms." + "reason": "A FlashAttention compatibility failure and a FlashAttention-3 degenerate-training bug are different defects." }, { "left": "issue:43931", "right": "issue:44038", "accept": false, - "reason": "Separate Qwen3-VL family issues: one loading mismatch, one Qwen3-VL-Moe bug. Not the same change." + "reason": "These are different Qwen model families with different symptoms: weight mismatch versus MoE load failure." }, { "left": "issue:43097", "right": "issue:43452", "accept": false, - "reason": "API removal warning for tie_embeddings_and_encoder_decoder is unrelated to gguf_file loader breakage." + "reason": "A removed config field and a gguf loading regression are unrelated code paths." }, { "left": "issue:45003", "right": "issue:45341", "accept": false, - "reason": "Unsafe sys.modules access and a small testing_utils bug are unrelated." + "reason": "An unsafe sys.modules access bug is unrelated to the testing_utils issue." }, { "left": "issue:43986", "right": "issue:45341", "accept": false, - "reason": "Video AutoProcessor crash without torchvision is unrelated to the testing_utils bug." + "reason": "A processor-loading crash without torchvision and a test utility bug are not the same bug." }, { "left": "issue:45753", "right": "issue:46369", "accept": false, - "reason": "Infinite loop on a specific image and SAM memory overflow are different model/runtime failures." + "reason": "Different model families and different failure modes: infinite loop versus memory overflow." }, { "left": "issue:44933", "right": "issue:46231", "accept": false, - "reason": "Missing import from image_utils and a documentation parameter-name typo are unrelated." + "reason": "A missing import and a documentation parameter-name mismatch are unrelated." }, { "left": "issue:43874", "right": "issue:44263", "accept": false, - "reason": "Different GLM-family bugs: missing image-patch API vs torch.split handling." + "reason": "Different missing-method / tensor-splitting failures in different model components." }, { "left": "issue:44261", "right": "issue:44485", "accept": false, - "reason": "Missing rms_norm_eps precision bug is unrelated to the GLM-5 RoPE implementation discussion." + "reason": "A missing rms_norm_eps precision issue is not the same as a GLM-5 RoPE implementation bug." }, { - "left": "issue:44117", - "right": "issue:44987", + "left": "issue:43673", + "right": "issue:46123", "accept": false, - "reason": "TOKENIZER_MAPPING_NAMES None handling and a specific model-load failure are not the same defect." + "reason": "One is a GenerationMixin cache regression; the other is a mask-generation batching bug." }, { - "left": "issue:43673", - "right": "issue:46123", + "left": "issue:44117", + "right": "issue:44987", "accept": false, - "reason": "Generation cache/chunked_prefill bug is unrelated to the mask pipeline partial-batch flag issue." + "reason": "A tokenizer mapping None-handling bug and a package load failure are too different to treat as duplicates." }, { "left": "issue:43986", "right": "issue:44484", "accept": false, - "reason": "Video processor crash without torchvision is unrelated to the max_shard_size question." + "reason": "These affect different surfaces entirely: processor loading versus model sharding defaults." }, { "left": "issue:43531", "right": "issue:45341", "accept": false, - "reason": "Qwen3-MoE sliding_window bug and a testing_utils issue are unrelated." + "reason": "A sliding-window bug in Qwen3-MoE is unrelated to the testing_utils issue." }, { "left": "issue:45440", "right": "issue:46172", "accept": false, - "reason": "DeepseekV3MoE divergence from remote implementation is a different bug from the Deepseek-v4 dtype/transpose issue." + "reason": "Both involve DeepSeek-style models, but the reported problems are different code paths and symptoms." }, { "left": "issue:43064", "right": "issue:44315", "accept": false, - "reason": "FSDP2 optimizer-state corruption and Liger Kernel not applied with model_init are different trainer integration bugs." + "reason": "A distributed optimizer-state bug and a Liger Kernel application issue are not the same change." }, { "left": "issue:43986", "right": "issue:45372", "accept": false, - "reason": "Missing torchvision crash and missing ReasoningEffort import are different dependency/import issues." + "reason": "A missing torchvision crash and a mistral_common import error affect different processor-loading paths." }, { "left": "issue:38617", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but they involve different missing symbols from different modules and fixes." + "reason": "Both are import errors, but for different symbols and different dependency breakages." }, { "left": "issue:43097", "right": "issue:45588", "accept": false, - "reason": "Embedding tie removal notice is unrelated to the flash_attention.py s_aux=None crash." + "reason": "A removed config field and a flash_attention.py None-handling crash are unrelated." }, { "left": "issue:43097", "right": "issue:46050", "accept": false, - "reason": "API removal notice and quantized save/load breakage are different problems." + "reason": "A config/API removal issue and a quantization save/load breakage are distinct bugs." }, { "left": "issue:44410", "right": "issue:45375", "accept": false, - "reason": "Missing attention projections and a missing config field silently dropped by strict are different bugs." + "reason": "Both mention Qwen variants, but one is missing attention projections while the other is a config field being dropped." }, { "left": "issue:44038", "right": "issue:44410", "accept": false, - "reason": "Different Qwen model variants and different failure modes; not a shared underlying fix." + "reason": "These are different Qwen model families with different load/config defects." }, { - "left": "issue:42890", - "right": "issue:46298", + "left": "issue:38617", + "right": "issue:44351", "accept": false, - "reason": "Seed-related flaky tests and a noisy logging message are unrelated." + "reason": "Both are import errors, but they stem from different missing symbols and are not the same underlying regression." }, { - "left": "issue:38617", - "right": "issue:44351", + "left": "issue:42890", + "right": "issue:46298", "accept": false, - "reason": "Both are import errors, but they refer to different missing names and unrelated modules." + "reason": "A flaky integration test from missing set_seed calls is unrelated to a noisy logging message." } ] }, @@ -15512,238 +15510,238 @@ "issue:42490|issue:43012", "issue:42757|issue:43824" ], - "prepared_review_unit_hash": "a84208d5a6060a49ef4aad5db58797924b333dce24d25d5c1ddec8cfbb7e2b30", + "prepared_review_unit_hash": "dc47ce412cdd2d039edef94e5310f04b084add8de980e7925ae82e55b4dcc4b0", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 34, - "serialized_chars": 37717, - "estimated_input_tokens": 9430, - "estimated_eval_tokens": 19116 + "serialized_chars": 37715, + "estimated_input_tokens": 9429, + "estimated_eval_tokens": 19114 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is highly heterogeneous and does not look like one duplicate family. I would not merge any of the soft pairs; they mostly share only broad subsystem keywords (imports, saving/loading, attention, training, compilation).", - "confidence": 0.83, - "canonical_issue_reason": "No single issue cleanly represents the whole cluster, but issue 43502 is the best anchor because it is a concrete, narrowly scoped runtime bug with a clear symptom and repro.", + "summary": "This set is heterogeneous: the soft pairs mostly share broad subsystem language (tokenizers, loading, saving, compilation, processors) but describe different failure modes and code paths, so none should be deduplicated.", + "confidence": 0.96, + "canonical_issue_reason": "issue:43097 is the broadest, most central regression report in the set, with downstream references and a clear, actionable compatibility break.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43502 is the strongest standalone candidate: it has a specific user-facing failure mode and is easier to triage than the broader or more ambiguous reports.", + "best_issue_reason": "issue:43097 is the most representative user-facing regression among these otherwise unrelated issues.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44062", "right": "issue:44117", "accept": false, - "reason": "Both are tokenizer-related, but one is an AddedToken argument error and the other is a mapping lookup assumption; different code paths." + "reason": "Both are tokenizer-related, but one is an AddedToken argument conflict and the other is a TOKENIZER_MAPPING_NAMES None assumption; different bugs." }, { "left": "issue:43097", "right": "issue:43873", "accept": false, - "reason": "One is a v5 API removal complaint, the other is an offloading/quantization behavior bug; not the same underlying issue." + "reason": "A v5 API removal warning is unrelated to quantized offloading behavior; different feature areas and symptoms." }, { "left": "issue:43502", "right": "issue:43992", "accept": false, - "reason": "Local-files-only network access and missing embed token weights are unrelated loading bugs." + "reason": "One reports network/API calls despite local_files_only, the other a missing embed_tokens weight in UMT5Encoder; not the same failure." }, { "left": "issue:44936", "right": "issue:46365", "accept": false, - "reason": "Trainer.evaluate after train and Trainer import failing due to datasets are different failure modes." + "reason": "trainer.evaluate() breaking after train is unrelated to importing Trainer failing because of datasets; different code paths." }, { "left": "issue:44704", "right": "issue:45103", "accept": false, - "reason": "Passing kwargs to cached_file and auto_docstring annotation crashes touch different subsystems." + "reason": "AutoProcessor kwarg propagation to cached_file and an auto_docstring annotations crash are unrelated bugs." }, { "left": "issue:44380", "right": "issue:46421", "accept": false, - "reason": "GPT2 attention scaling under SDPA/FlashAttention is unrelated to cache lazy initialization recompilation." + "reason": "GPT2 attention scaling being ignored is separate from cache lazy initialization causing recompilation; not the same issue." }, { "left": "issue:43986", "right": "issue:44117", "accept": false, - "reason": "AutoProcessor video-model import crash without torchvision is unrelated to tokenizer mapping returning None." + "reason": "Missing torchvision during AutoProcessor load is unrelated to tokenizer mapping returning None." }, { "left": "issue:43986", "right": "issue:44016", "accept": false, - "reason": "Video-model import handling and a notebook syntax error are not the same bug." + "reason": "A runtime dependency/import crash is not the same as a notebook syntax error." }, { "left": "issue:44938", "right": "issue:46178", "accept": false, - "reason": "Python 3.14 import failure and a tensorflow_text backend mapping error are distinct dependency/import issues." + "reason": "Python 3.14 load failure and an unknown tensorflow_text backend are different compatibility problems." }, { "left": "issue:43208", "right": "issue:43550", "accept": false, - "reason": "xLSTM training bugs and Bamba SDPA torch.compile failure are different model-specific problems." + "reason": "xLSTM training bugs smaller than 7B and Bamba torch.compile/SDPA failure are distinct model/backend issues." }, { "left": "issue:42907", "right": "issue:44655", "accept": false, - "reason": "Saving dequantized models and saving pipeline objects involve different save_pretrained paths." + "reason": "Failed dequantized save behavior and inability to save Pipeline objects are different serialization paths." }, { "left": "issue:42907", "right": "issue:45753", "accept": false, - "reason": "Model save failure and Qwen3_5 infinite-loop generation are unrelated." + "reason": "Dequantized save failure is unrelated to a Qwen3.5 image loop during generation." }, { "left": "issue:43825", "right": "issue:44655", "accept": false, - "reason": "Pipeline translation messaging and pipeline save_pretrained support are separate issues." + "reason": "Pipeline translation support messaging and saving Pipeline objects are unrelated pipeline issues." }, { "left": "issue:43012", "right": "issue:43550", "accept": false, - "reason": "A bfloat16 precision warning is not the same as a compile-time SDPA failure." + "reason": "A bf16 compile warning is not the same as a Bamba SDPA compile failure." }, { "left": "issue:43992", "right": "issue:44484", "accept": false, - "reason": "Missing embed token weights and shard-size default behavior are unrelated save/load concerns." + "reason": "Missing embed_tokens weights and max_shard_size defaults are different save/load concerns." }, { "left": "issue:43012", "right": "issue:44368", "accept": false, - "reason": "PyTorch precision warning and a tie_word_embeddings config warning are different warnings with different causes." + "reason": "A PyTorch precision warning during compile is unrelated to a tie_word_embeddings config warning." }, { "left": "issue:45712", "right": "issue:46298", "accept": false, - "reason": "Leftover dummy classes leaking into dir(transformers) is unrelated to noisy logging." + "reason": "Leftover dummy classes in repo checks are not the same as a noisy logging message." }, { "left": "issue:43066", "right": "issue:43906", "accept": false, - "reason": "Wrong tokenizer decoder type and an isolated reproduction issue are not clearly the same bug." + "reason": "Wrong tokenizer decoder type in v5 and an isolated reproduction of another issue are not the same bug." }, { "left": "issue:43097", "right": "issue:46298", "accept": false, - "reason": "v5 API removal and a noisy logging message do not share a concrete code-path bug." + "reason": "A breaking API removal is unrelated to a noisy logging message." }, { "left": "issue:43352", "right": "issue:44380", "accept": false, - "reason": "Flash Attention support gating for Nemotron and GPT2 attention scaling being ignored are different attention issues." + "reason": "Nemotron Flash Attention unsupported and GPT2 attention scaling ignored are different attention/backend problems." }, { "left": "issue:43352", "right": "issue:43976", "accept": false, - "reason": "Model-specific Flash Attention support and Python version compatibility are unrelated." + "reason": "Flash Attention support for a model and Python-version compatibility are unrelated." }, { "left": "issue:44162", "right": "issue:46172", "accept": false, - "reason": "ESM2 breakage and a deepseek tensor transpose bug are different model/runtime failures." + "reason": "ESM2 breakage and a DeepSeek tensor transpose bug are separate model-specific failures." }, { "left": "issue:45584", "right": "issue:46365", "accept": false, - "reason": "Whisper empty-transcription failure and Trainer import failure are unrelated." + "reason": "Whisper empty-transcription failure is unrelated to importing Trainer being broken by datasets." }, { "left": "issue:44380", "right": "issue:46132", "accept": false, - "reason": "Ignored GPT2 attention scaling config and AttentionInterface.register behavior changes are separate attention API issues." + "reason": "Attention scaling being ignored is not the same as AttentionInterface.register changing function behavior." }, { "left": "issue:44077", "right": "issue:46365", "accept": false, - "reason": "patchtsmixer post_init validation and Trainer import failure are unrelated." + "reason": "PatchTSMixer post_init validation and Trainer import failure are unrelated." }, { "left": "issue:44222", "right": "issue:45478", "accept": false, - "reason": "Both mention MoE/save/load, but one is FP8 save_pretrained and the other is a from_pretrained error; too different to treat as the same bug." + "reason": "FP8 save_pretrained issues and a Qwen3.5 MoE from_pretrained error are similar only at a high level, but not the same concrete bug." }, { "left": "issue:44380", "right": "issue:45479", "accept": false, - "reason": "Attention backend config handling and single-label classification zero-loss behavior are unrelated." + "reason": "Attention backend config loss and degenerate zero-loss classification are unrelated." }, { "left": "issue:36683", "right": "issue:43582", "accept": false, - "reason": "Gemma3Config missing vocab_size and Apple Silicon allocator TypeError are unrelated." + "reason": "Gemma3Config missing vocab_size and an Apple Silicon allocator TypeError are unrelated runtime issues." }, { "left": "issue:43064", "right": "issue:44380", "accept": false, - "reason": "Distributed optimizer-state mismatch and GPT2 attention scaling are different parts of the stack." + "reason": "FSDP2/PEFT optimizer state corruption is a training-distributed bug, not an attention scaling regression." }, { "left": "issue:42175", "right": "issue:46365", "accept": false, - "reason": "Tensorflow backend packaging and a Trainer import conflict are different dependency/import problems." + "reason": "Backend extras packaging and Trainer import interference from datasets are different dependency problems." }, { "left": "issue:43208", "right": "issue:43526", "accept": false, - "reason": "xLSTM training blockers and BeitImageProcessorFast label reduction are unrelated." + "reason": "xLSTM training bugs and BeitImageProcessorFast label reduction are separate components and failures." }, { "left": "issue:43012", "right": "issue:44683", "accept": false, - "reason": "A bfloat16 warning and a compiled flex_attention failure are different precision/backend issues." + "reason": "A compile-time precision warning is not the same as compiled flex_attention breaking on torch>=2.9." }, { "left": "issue:42490", "right": "issue:43012", "accept": false, - "reason": "Local path slash warning and a bfloat16 precision warning are unrelated." + "reason": "Trailing-slash local model import warning and bf16 compile warning are unrelated issues." }, { "left": "issue:42757", "right": "issue:43824", "accept": false, - "reason": "Missing is_offline_mode and missing Qwen2_5_VLForConditionalGeneration are different import errors from different symbols." + "reason": "Missing is_offline_mode import and missing Qwen2_5_VLForConditionalGeneration import are both import errors, but from different modules and causes." } ] }, @@ -15825,8 +15823,8 @@ "issue:43618|issue:43761", "issue:43299|issue:43867", "issue:42994|issue:44655", - "issue:44483|issue:45290", "issue:36683|issue:43475", + "issue:44483|issue:45290", "issue:43756|issue:44485", "issue:44625|issue:46365", "issue:42898|issue:43066", @@ -15848,220 +15846,220 @@ "issue:44075|issue:46365", "issue:43010|issue:45278" ], - "prepared_review_unit_hash": "75545a37cdc18976536f546c9d40863e34a6b55b992e9a7bbca95ac3e744fcc3", + "prepared_review_unit_hash": "0fef355932228ce142a3289efaef5cf644e00c0d2b160d45b94aa1c3cf3d51cc", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 31, - "serialized_chars": 35897, - "estimated_input_tokens": 8975, - "estimated_eval_tokens": 18206 + "serialized_chars": 35893, + "estimated_input_tokens": 8974, + "estimated_eval_tokens": 18204 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is highly heterogeneous: mostly unrelated import errors, config regressions, save/load issues, tokenizer changes, and model-specific bugs. None of the soft pairs looks like a true duplicate.", - "confidence": 0.96, - "canonical_issue_reason": "No clear canonical issue: the items span unrelated import, config, model-output, save/load, and tokenizer regressions rather than one underlying bug.", + "summary": "These are mostly distinct regression reports. The similarity is often only at the subsystem level (imports, save/load, CLIP outputs, tokenizer behavior), but the concrete bugs differ enough that none of the soft edges should be merged.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 45698 is the strongest umbrella candidate: it is open, highly discussed, and centers on a broad from_pretrained/save_pretrained deserialization mismatch that is more representative of the save/load regressions here than the narrower model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "If forced to choose a representative, issue 45278 is the broadest regression report, but it still only covers the import-error slice and does not explain most of the cluster.", + "best_issue_reason": "Issue 45698 is the best representative issue in this mixed set because it has the most discussion and captures a broadly applicable save/load failure mode.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45600", "right": "issue:46298", "accept": false, - "reason": "Unrelated: one is a CI break from removed Sam3LiteText references, the other is just a noisy logging message." + "reason": "Different bugs: one is a CI break from removed Sam3LiteText mappings, the other is a noisy logging message." }, { "left": "issue:44483", "right": "issue:44514", "accept": false, - "reason": "Both mention chat/completions, but one is request validation and the other is a batched chat-template crash with tool calls." + "reason": "Both affect chat APIs, but one is request validation for /v1/chat/completions and the other is a batched apply_chat_template crash with padding=False." }, { "left": "issue:38617", "right": "issue:43502", "accept": false, - "reason": "Different failures: an import error for layer_type_validation versus local_files_only still triggering network requests." + "reason": "Unrelated import-error vs network-request behavior: layer_type_validation import failure is not the same as local_files_only still making API calls." }, { "left": "issue:44038", "right": "issue:44230", "accept": false, - "reason": "Same model family, but different bugs: general Qwen3-VL-Moe loading versus fp8 inference support." + "reason": "Same model family, but different features: Qwen3-VL-Moe loading failure vs fp8 inference support request." }, { "left": "issue:45698", "right": "issue:46172", "accept": false, - "reason": "Different code paths: save/load custom-module resolution versus a DeepSeek-V4 tensor transpose bug." + "reason": "Different code paths and symptoms: wrong custom module loading after save_pretrained is not the deepseek-v4 dtype/transpose bug." }, { "left": "issue:43618", "right": "issue:44380", "accept": false, - "reason": "Both touch attention-related behavior, but one is a CLIP output regression and the other is GPT-2 attention scaling under SDPA/FlashAttention." + "reason": "Different models and regressions: missing CLIP attentions vs GPT2 attention scaling ignored under SDPA/FlashAttention." }, { "left": "issue:43618", "right": "issue:43761", "accept": false, - "reason": "Related CLIP regressions, but they affect different missing outputs and are not clearly the same bug." + "reason": "Both are CLIP output regressions, but one is attentions missing and the other is hidden_states missing; not enough evidence of one underlying bug." }, { "left": "issue:43299", "right": "issue:43867", "accept": false, - "reason": "Both are loading errors, but they involve different models and different failure mechanisms." + "reason": "Different loading failures: Qwen3VL MoE model loading vs a generic state_dict-sorting load error." }, { "left": "issue:42994", "right": "issue:44655", "accept": false, - "reason": "Both involve saving, but one is quantized model serialization and the other is Pipeline.save_pretrained." + "reason": "Both mention saving, but quantized model saving and pipeline save_pretrained are different bugs and likely different fixes." }, { - "left": "issue:44483", - "right": "issue:45290", + "left": "issue:36683", + "right": "issue:43475", "accept": false, - "reason": "Both mention chat completion paths, but one is API request rejection and the other is a tokenizer/template crash." + "reason": "Different model configs and missing attributes: Gemma3Config vocab_size vs SAM 3 Video fpn_position_embeddings." }, { - "left": "issue:36683", - "right": "issue:43475", + "left": "issue:44483", + "right": "issue:45290", "accept": false, - "reason": "Completely different models and attributes: Gemma3Config vocab_size versus Sam3VisionEncoderOutput fpn_position_embeddings." + "reason": "Different request surfaces: chat completion validation vs assistant tool-call template crashes." }, { "left": "issue:43756", "right": "issue:44485", "accept": false, - "reason": "Both mention RoPE, but they are distinct model-specific implementation issues." + "reason": "Both mention RoPE, but the Smollm3 layer-count issue and the GLM-5 RoPE implementation issue are model-specific and not the same bug." }, { "left": "issue:44625", "right": "issue:46365", "accept": false, - "reason": "Different problem classes: Qwen3.5 num_labels propagation versus Trainer import failure tied to datasets." + "reason": "Unrelated issues: Qwen3.5 num_labels propagation vs Trainer import failure caused by datasets." }, { "left": "issue:42898", "right": "issue:43066", "accept": false, - "reason": "Both are tokenizer-related v5 changes, but they concern different behaviors and code paths." + "reason": "Both are tokenizer-related v5 reports, but clean_up_tokenization_spaces behavior and decoder type selection are different code paths." }, { "left": "issue:45276", "right": "issue:46077", "accept": false, - "reason": "Same model family, but one is embedding resizing and the other is bidirectional attention masking." + "reason": "Both are Gemma4-related, but one is embedding resize propagation and the other is bidirectional attention mask construction." }, { "left": "issue:36010", "right": "issue:42757", "accept": false, - "reason": "Unrelated import errors from different packages and symbols." + "reason": "Both are import errors, but from different packages and symbols: GenerationMixin from transformers vs is_offline_mode from huggingface_hub." }, { "left": "issue:43352", "right": "issue:43937", "accept": false, - "reason": "Different issues: Flash Attention support for Nemotron versus invalid GenerationConfig for GLM-5." + "reason": "Different failures: Flash Attention 2.0 unsupported on NemotronH vs invalid GenerationConfig for GLM-5." }, { "left": "issue:43611", "right": "issue:44655", "accept": false, - "reason": "Different subsystems: model loading with base_model_prefix versus saving Pipeline objects." + "reason": "Different bugs: base_model_prefix loading regression vs inability to save Pipeline objects." }, { "left": "issue:43901", "right": "issue:44655", "accept": false, - "reason": "Docs/behavior mismatch for TextClassificationPipeline is unrelated to Pipeline.save_pretrained." + "reason": "One is a docs mismatch about return_all_scores; the other is a runtime save_pretrained failure." }, { "left": "issue:36683", "right": "issue:44077", "accept": false, - "reason": "Different model configs and symptoms; no shared underlying bug." + "reason": "Both are model/config related, but Gemma3Config vocab_size and patchtsmixer post_init enforcement are unrelated." }, { "left": "issue:43867", "right": "issue:45216", "accept": false, - "reason": "Both involve checkpoint loading/saving, but the concrete failures are different and not mergeable into one fix." + "reason": "Generic load-model state_dict ordering error is not the same as Qwen3.5 save_pretrained checkpoint corruption." }, { "left": "issue:43976", "right": "issue:44938", "accept": false, - "reason": "Both are version-compatibility reports, but they target different Python versions and different breakages." + "reason": "Both are Python-version compatibility problems, but they target different versions and failure modes." }, { "left": "issue:44625", "right": "issue:46077", "accept": false, - "reason": "Different Gemma3.5 vs Gemma4 bugs; one is label propagation, the other attention masking." + "reason": "Different Qwen3.5 config propagation vs Gemma4 attention-mask behavior." }, { "left": "issue:46077", "right": "issue:46082", "accept": false, - "reason": "Different config issues in different models: bidirectional masks versus Llama head_dim validation." + "reason": "Unrelated model/config issues: Gemma4 bidirectional attention masks vs LlamaConfig head_dim validation." }, { "left": "issue:36683", "right": "issue:43525", "accept": false, - "reason": "Different missing config attributes in different models; not the same regression." + "reason": "Both are AttributeError reports on configs, but the missing attributes and model families are different." }, { "left": "issue:43618", "right": "issue:45216", "accept": false, - "reason": "CLIP output fields versus Qwen3.5 checkpoint saving are unrelated." + "reason": "CLIP output-field regression and Qwen3.5 checkpoint saving corruption are unrelated." }, { "left": "issue:36683", "right": "issue:44625", "accept": false, - "reason": "Both are config propagation bugs, but for different models and different fields." + "reason": "Different config propagation bugs in different models: Gemma3 vocab_size vs Qwen3.5 num_labels." }, { "left": "issue:43992", "right": "issue:45310", "accept": false, - "reason": "Both involve from_pretrained, but one misses an embedding weight and the other is a Qwen3.5 MoE loading error." + "reason": "Different from_pretrained/model-loading failures on different model families." }, { "left": "issue:43064", "right": "issue:44492", "accept": false, - "reason": "Completely unrelated: distributed optimizer state bug versus a cache-strategy typo." + "reason": "Training-state optimizer mismatch is unrelated to a typo in cache strategy naming." }, { "left": "issue:44075", "right": "issue:46365", "accept": false, - "reason": "Different areas: optimizer argument handling versus importing Trainer." + "reason": "Optimizer SGD arguments not applied is unrelated to Trainer import failure from datasets." }, { "left": "issue:43010", "right": "issue:45278", "accept": false, - "reason": "Unrelated: no_grad decoration for cache updates versus broad import errors after upgrading transformers." + "reason": "Cache update decoration/no_grad behavior is unrelated to broad import errors after upgrading Transformers." } ] }, @@ -16158,8 +16156,8 @@ "issue:43502|issue:44062", "issue:43874|issue:46617", "issue:43723|issue:44556", - "issue:44387|issue:44556", "issue:44222|issue:45125", + "issue:44387|issue:44556", "issue:44734|issue:46617", "issue:43010|issue:43065", "issue:43576|issue:46132", @@ -16168,226 +16166,226 @@ "issue:45125|issue:45478", "issue:30333|issue:33453" ], - "prepared_review_unit_hash": "2d33810df8473537e1b508517f2daae6953a191d1c6188536f5b8eee7767902c", + "prepared_review_unit_hash": "94492bbf8a27b7f4265251a5e669b22b2f4bf3a31ee5cdaedb3254f531e41df8", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 36968, - "estimated_input_tokens": 9242, - "estimated_eval_tokens": 18740 + "serialized_chars": 36969, + "estimated_input_tokens": 9243, + "estimated_eval_tokens": 18742 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly a grab bag of unrelated Transformers issues, with several broad v5-regression reports and many model-specific loading or test failures. I do not see any soft pair that clearly represents the same underlying bug/change.", - "confidence": 0.88, - "canonical_issue_reason": "Issue 45850 is the broadest umbrella-style report in the set (\u201cSince 5.0 version it breaks too many models\u201d), so it is the closest thing to a cluster anchor; however, it is too vague to be a strong duplicate target.", + "summary": "This is a heterogeneous issue-only cluster. The soft-similarity pairs mostly overlap on broad subsystems (loading, imports, tests, tensor-parallelism) but do not describe the same underlying bug, so no dedupe merges are justified.", + "confidence": 0.98, + "canonical_issue_reason": "No single canonical issue fits this cluster: it spans many unrelated regressions and test flakes rather than one shared bug.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45850 is the best global issue candidate only because it is the most general post-v5 regression complaint, but it is not precise enough to function as a high-quality canonical duplicate.", + "best_issue_reason": "No issue stands out as a representative duplicate anchor; the cluster is too mixed to pick a meaningful best issue.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44355", "right": "issue:44855", "accept": false, - "reason": "Both involve Python 3.13 import/loading failures, but one is about compiled file errors and the other is a specific torch.jit decorator parsing/IndentationError; different failure modes." + "reason": "Different failures: compiled Python file import errors vs a torch.jit decorator/comment parsing issue." }, { "left": "issue:44938", "right": "issue:45071", "accept": false, - "reason": "One is a Python 3.14 load failure, the other is a PretrainedConfig type-checking breakage in v5.4; unrelated bugs." + "reason": "One is a Python 3.14 load failure; the other is a PretrainedConfig type-checking regression." }, { "left": "issue:43867", "right": "issue:45375", "accept": false, - "reason": "State-dict sorting/load ordering issue versus a Qwen3.5 vision config field being dropped; different code paths and symptoms." + "reason": "Generic load/state_dict sorting issue vs missing Qwen3.5 vision config field." }, { "left": "issue:43867", "right": "issue:46202", "accept": false, - "reason": "A load-model/state_dict issue and a Gemma pipeline roundtrip tokenization failure are not the same underlying problem." + "reason": "Different model paths and symptoms: load failure vs Gemma pipeline roundtrip token issue." }, { "left": "issue:36683", "right": "issue:43064", "accept": false, - "reason": "Gemma3Config missing vocab_size is a model config bug; the other is an FSDP2/PEFT optimizer-state mismatch during training." + "reason": "Gemma3Config missing `vocab_size` is unrelated to the FSDP2+PEFT optimizer-state bug." }, { "left": "issue:42890", "right": "issue:45561", "accept": false, - "reason": "Both are test failures, but one is missing set_seed() coverage and the other is an xdist race on captured_info.txt; different bugs." + "reason": "Both are test-related, but one is missing `set_seed()` coverage and the other is a pytest-xdist file race." }, { "left": "issue:43010", "right": "issue:45850", "accept": false, - "reason": "One is a concrete request to add @torch.no_grad to update methods; the other is a broad complaint that many models break in v5." + "reason": "A specific `@torch.no_grad` cache/layer update fix is not the same as a vague broad regression report." }, { "left": "issue:41669", "right": "issue:44246", "accept": false, - "reason": "Both mention slow import, but 41669 names a specific root cause (import * in models) while 44246 is only a vague symptom report; not enough to confirm the same change." + "reason": "Both mention import slowness, but one is about removing `import *` and the other is an intermittent slow import report." }, { "left": "issue:36296", "right": "issue:43575", "accept": false, - "reason": "Tensor-parallel training bug versus OOM when loading a specific Qwen2-57B model; different resource/problem classes." + "reason": "Tensor-parallel training bug vs OOM when loading a specific Qwen2 model with TP." }, { "left": "issue:43867", "right": "issue:45357", "accept": false, - "reason": "Load/state_dict ordering problem versus incorrect visual encoder keys in save_pretrained; distinct regression types." + "reason": "Different save/load paths: state_dict sorting vs incorrect visual encoder keys on save_pretrained." }, { "left": "issue:43010", "right": "issue:43976", "accept": false, - "reason": "A cache update no_grad issue and a Python version compatibility problem are unrelated." + "reason": "Cache-update no-grad semantics are unrelated to Python version compatibility." }, { "left": "issue:43618", "right": "issue:44077", "accept": false, - "reason": "Missing CLIPOutput attentions and an optional post_init policy issue in patchtsmixer are unrelated." + "reason": "CLIP attentions assignment regression vs patchtsmixer `post_init` validation are different bugs." }, { "left": "issue:43992", "right": "issue:45478", "accept": false, - "reason": "Both are from_pretrained/load complaints, but they concern different models and different failure mechanisms." + "reason": "UMT5 missing `embed_tokens.weight` is unrelated to Qwen3.5 MoE from_pretrained errors." }, { "left": "issue:28282", "right": "issue:33453", "accept": false, - "reason": "PyTorch missing ImportError versus tokenizer loading regression; unrelated." + "reason": "Missing PyTorch backend import error is unrelated to tokenizer-loading regression." }, { "left": "issue:43618", "right": "issue:46285", "accept": false, - "reason": "Both touch CLIP, but one is about attentions on outputs and the other is a structural .text_model inconsistency; not the same bug." + "reason": "Same general CLIP area, but different concrete breakages: missing attentions vs inconsistent model structure." }, { "left": "issue:43618", "right": "issue:44857", "accept": false, - "reason": "CLIP output attentions versus an AMP/CUDA crash in LwDetrImageLoss; unrelated." + "reason": "CLIPOutput attentions regression is unrelated to the LwDetrImageLoss AMP crash." }, { "left": "issue:44360", "right": "issue:45440", "accept": false, - "reason": "DSA indexer missing ReLU and DeepseekV3MoE divergence are separate implementation issues." + "reason": "DSA indexer ReLU discussion and DeepseekV3MoE divergence are unrelated code paths." }, { "left": "issue:42890", "right": "issue:44964", "accept": false, - "reason": "A flaky SAM-HQ integration test and a Phi-4 multimodal model load failure are unrelated." + "reason": "A flaky integration test is not the same as a model-loading failure for Phi-4 multimodal." }, { "left": "issue:42890", "right": "issue:44112", "accept": false, - "reason": "Both are CI/test problems, but one is seed-related flakiness and the other is a stale device-override test; different root causes." + "reason": "Both are CI/test issues, but one is missing seeding and the other is a stale GraniteSpeech device override test." }, { "left": "issue:42890", "right": "issue:45600", "accept": false, - "reason": "A flaky SAM-HQ test and broken auto_mappings references are separate CI issues." + "reason": "Test flake vs auto_mappings CI break from removed Sam3LiteText configs." }, { "left": "issue:43502", "right": "issue:44062", "accept": false, - "reason": "local_files_only still making API requests versus an AddedToken keyword-argument TypeError; unrelated." + "reason": "Local-files-only network leak and `AddedToken()` keyword collision are different bugs." }, { "left": "issue:43874", "right": "issue:46617", "accept": false, - "reason": "Missing get_number_of_image_patches in a GLM46V fast processor versus a trust_remote_code cache FileNotFoundError; different code paths." + "reason": "Missing image-patch helper in GLM46V is unrelated to cache FileNotFoundError with trust_remote_code." }, { "left": "issue:43723", "right": "issue:44556", "accept": false, - "reason": "Both are loading-related regressions, but one is tokenizer loading in v5 and the other is checkpoint reload compatibility after upgrading; not the same concrete bug." - }, - { - "left": "issue:44387", - "right": "issue:44556", - "accept": false, - "reason": "int4 quantization memory growth/OOM versus checkpoint reload failure after upgrade; unrelated." + "reason": "Tokenizer loading regression and checkpoint reload-after-upgrade failure are different load paths." }, { "left": "issue:44222", "right": "issue:45125", "accept": false, - "reason": "FP8 save_pretrained for MoE versus missing _tp_plan for tensor parallelism; distinct issues." + "reason": "FP8 save_pretrained issue is unrelated to missing tensor-parallel plan metadata." }, { - "left": "issue:44734", - "right": "issue:46617", + "left": "issue:44387", + "right": "issue:44556", "accept": false, - "reason": "KV-cache continuation tensor indexing crash versus cache loading FileNotFoundError; unrelated." + "reason": "Quantization memory/OOM regression is unrelated to checkpoint reload compatibility." }, { - "left": "issue:43824", + "left": "issue:44734", "right": "issue:46617", "accept": false, - "reason": "Qwen2_5_VL import error versus trust_remote_code cache loading failure; different failures." + "reason": "KV cache continuation crash and cache FileNotFoundError are different failure modes." }, { - "left": "issue:45125", - "right": "issue:45478", + "left": "issue:43010", + "right": "issue:43065", "accept": false, - "reason": "Same broad Qwen3.5 MoE area, but one is missing tensor-parallel metadata and the other is a generic from_pretrained error; too different to merge as the same bug." + "reason": "`@torch.no_grad` on cache updates and the dummy Conv2d in Sam3PixelDecoder are unrelated changes." }, { - "left": "issue:30333", - "right": "issue:33453", + "left": "issue:43576", + "right": "issue:46132", "accept": false, - "reason": "MLFlow reporting on failed training and tokenizer loading regression are unrelated." + "reason": "Broken `transformers env` command and changed `AttentionInterface.register` behavior are separate issues." }, { "left": "issue:43502", "right": "issue:44485", "accept": false, - "reason": "Network access despite local_files_only and a GLM-5 RoPE implementation discussion are unrelated." + "reason": "Local-files-only network leakage is unrelated to GLM-5 RoPE implementation details." + }, + { + "left": "issue:43824", + "right": "issue:46617", + "accept": false, + "reason": "Qwen2.5-VL import error and trust_remote_code cache FileNotFoundError are not the same bug." }, { - "left": "issue:43010", - "right": "issue:43065", + "left": "issue:45125", + "right": "issue:45478", "accept": false, - "reason": "A no_grad decorator request for cache/layer updates and a dummy Conv2d in Sam3PixelDecoder are unrelated." + "reason": "Missing `_tp_plan` for tensor parallelism is different from the Qwen3.5 MoE from_pretrained error." }, { - "left": "issue:43576", - "right": "issue:46132", + "left": "issue:30333", + "right": "issue:33453", "accept": false, - "reason": "A broken transformers env command and a changed AttentionInterface.register behavior are different API regressions, not the same fix." + "reason": "MLflow job-status reporting and tokenizer-loading regression are unrelated." } ] }, @@ -16466,8 +16464,8 @@ "issue:45125|issue:45375", "issue:43010|issue:43232", "issue:43502|issue:44222", - "issue:43502|issue:44961", "issue:30990|issue:31795", + "issue:43502|issue:44961", "issue:43827|issue:44961", "issue:43299|issue:44556", "issue:30333|issue:36296", @@ -16496,238 +16494,232 @@ "issue:43673|issue:44492", "issue:44623|issue:45230" ], - "prepared_review_unit_hash": "03b21724466bcc31fa1bd89b0cbdacbe7824243610d093ffdea40b67c7cb7103", + "prepared_review_unit_hash": "70b8d53f40524fc89b34bb0541a07a6ec8d8dd7754549664e96914829ce7e7fc", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 34, - "serialized_chars": 37023, - "estimated_input_tokens": 9256, - "estimated_eval_tokens": 18768 + "serialized_chars": 37020, + "estimated_input_tokens": 9255, + "estimated_eval_tokens": 18766 }, "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 that only share broad v5/import/loading/regression wording; none of the soft pairs look like true duplicates.", - "confidence": 0.88, - "canonical_issue_reason": "If a representative is needed, issue 45850 is the broadest umbrella for the recurring v5 breakage theme, but the set is too heterogeneous for a true canonical duplicate.", + "summary": "This cluster is mostly a grab bag of unrelated Transformers regressions, import errors, loading issues, and docs problems. None of the soft pairs look like true duplicates; the closest ones are still too different in concrete code path or symptom. The best representative issue is the broad v5 breakage report.", + "confidence": 0.63, + "canonical_issue_reason": "issue:45850 is the broadest umbrella in the set: it captures the general 'v5 breaks many models' theme that recurs across the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "45850 is the most general and central to the visible regression theme, but it is still too vague to collapse the cluster cleanly.", + "best_issue_reason": "issue:45850 is the best global representative because it spans the widest set of user-facing v5 regressions, even though it is too generic to be a precise duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44355", "right": "issue:44749", "accept": false, - "reason": "Different bugs: compiled Python file execution errors vs a data-processing slowdown after upgrade." + "reason": "Different failures: compiled Python file errors vs slower filtering after upgrade." }, { "left": "issue:42548", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for different symbols and likely different breakages." + "reason": "Both are import errors, but for different symbols and likely different fixes." }, { "left": "issue:45125", "right": "issue:45375", "accept": false, - "reason": "Different Qwen3.5-Moe issues: missing tensor-parallel plan vs a missing config field being dropped." + "reason": "Different Qwen3.5-MoE problems: missing tensor-parallel plan vs dropped config field." }, { "left": "issue:43010", "right": "issue:43232", "accept": false, - "reason": "Cache no_grad decoration and generation sync_gpus behavior are separate code paths." + "reason": "Both touch generation internals, but the underlying bugs and code paths differ." }, { "left": "issue:43502", "right": "issue:44222", "accept": false, - "reason": "local_files_only network leakage is unrelated to FP8 save_pretrained behavior." + "reason": "Local-files-only network leakage vs FP8 save_pretrained support are unrelated." }, { - "left": "issue:43502", - "right": "issue:44961", + "left": "issue:30990", + "right": "issue:31795", "accept": false, - "reason": "A cache/network issue and an unrelated placeholder issue are not the same bug." + "reason": "Loading stall vs documentation confusion; not the same bug." }, { - "left": "issue:30990", - "right": "issue:31795", + "left": "issue:43502", + "right": "issue:44961", "accept": false, - "reason": "Loading hangs and documentation confusion are not the same underlying problem." + "reason": "API requests during load and the placeholder issue 'racoon' are unrelated." }, { "left": "issue:43827", "right": "issue:44961", "accept": false, - "reason": "Docs referencing pipeline() and an unrelated placeholder issue are clearly different." + "reason": "Docs referencing removed pipeline() vs unrelated placeholder issue." }, { "left": "issue:43299", "right": "issue:44556", "accept": false, - "reason": "Both mention v5 regressions, but one is Qwen3VL model loading and the other is checkpoint reload compatibility." + "reason": "Both are version-regression loading issues, but for different models and failure modes." }, { "left": "issue:30333", "right": "issue:36296", "accept": false, - "reason": "MLflow reporting on failed training jobs is unrelated to tensor-parallel training bugs." + "reason": "MLflow reporting on failed jobs is unrelated to tensor-parallel training." }, { "left": "issue:44355", "right": "issue:45071", "accept": false, - "reason": "Compiled-file loading errors and PretrainedConfig type-checking regressions are distinct." + "reason": "Compiled-file execution errors and PretrainedConfig type-checking are different problems." }, { "left": "issue:44355", "right": "issue:44655", "accept": false, - "reason": "Running compiled Python files is different from saving Pipeline objects." + "reason": "Compiled-file errors vs save_pretrained on Pipeline objects are unrelated." }, { "left": "issue:33453", "right": "issue:34689", "accept": false, - "reason": "Tokenizer loading regression and Llama 3.2 Vision model loading failure are not the same defect." + "reason": "Both mention loading regressions, but they affect different models and likely different loaders." }, { "left": "issue:27957", "right": "issue:33453", "accept": false, - "reason": "Flash Attention support for XLM-Roberta is unrelated to tokenizer loading regressions." + "reason": "Flash Attention 2 support for XLMRoberta is unrelated to tokenizer loading regressions." }, { "left": "issue:42757", "right": "issue:43576", "accept": false, - "reason": "Missing huggingface_hub import and a broken transformers env command are different issues." + "reason": "Missing hub import and the broken transformers env command are different failures." }, { "left": "issue:43576", "right": "issue:45071", "accept": false, - "reason": "Broken env command and config type-checking breakage do not share the same code path." + "reason": "Broken env command and type-checking breakage are not the same issue." }, { "left": "issue:44315", "right": "issue:44360", "accept": false, - "reason": "Liger Kernel application in model_init and a DSA/ReLU discussion are unrelated." + "reason": "Liger-kernel/model_init integration and DSA ReLU discussion are unrelated." }, { "left": "issue:43576", "right": "issue:43824", "accept": false, - "reason": "A CLI/env issue and a Qwen2.5-VL import error are different failures." + "reason": "Env-command breakage and Qwen2.5-VL import failure are different code paths." }, { "left": "issue:31795", "right": "issue:38617", "accept": false, - "reason": "Documentation confusion and a missing layer_type_validation import are not the same bug." + "reason": "Docs confusion in model.forward is unrelated to an import error for layer_type_validation." }, { "left": "issue:44961", "right": "issue:45468", "accept": false, - "reason": "Placeholder issue title and Gemma-4 audio positional encoding bug are unrelated." + "reason": "Placeholder issue vs Gemma-4 audio positional encoding bug are unrelated." }, { "left": "issue:44351", "right": "issue:46617", "accept": false, - "reason": "HybridCache import failure and trust_remote_code cache FileNotFoundError are different import/load problems." + "reason": "HybridCache import failure and trust_remote_code cache FileNotFoundError are different issues." }, { "left": "issue:43576", "right": "issue:44938", "accept": false, - "reason": "Broken env command and Python 3.14 load failure are not the same underlying issue." + "reason": "Broken env command in v5 vs Python 3.14 load failure are not the same underlying bug." }, { "left": "issue:44485", "right": "issue:44961", "accept": false, - "reason": "GLM-5 RoPE implementation discussion is unrelated to the placeholder issue." + "reason": "GLM-5 RoPE implementation and placeholder issue are unrelated." }, { "left": "issue:44908", "right": "issue:44961", "accept": false, - "reason": "Scheduler kwargs handling and the placeholder issue are unrelated." + "reason": "Scheduler kwargs bug vs placeholder issue are unrelated." }, { "left": "issue:44222", "right": "issue:44230", "accept": false, - "reason": "FP8 save_pretrained moe and FP8 inference support are related only by topic, not the same bug." + "reason": "FP8 save_pretrained behavior and FP8 inference support are distinct problems." }, { "left": "issue:36296", "right": "issue:43901", "accept": false, - "reason": "Tensor parallel training and pipeline docs behavior mismatch are different problems." + "reason": "Tensor-parallel training bug and pipeline docs mismatch are unrelated." }, { "left": "issue:46143", "right": "issue:46144", "accept": false, - "reason": "Missing kwargs passthrough and attention-interface usage are close but still distinct code-path issues." + "reason": "Closely related RoFormer issues, but not the same concrete bug." }, { "left": "issue:36296", "right": "issue:43526", "accept": false, - "reason": "Tensor-parallel training bug and BeitImageProcessorFast reduce_labels behavior are unrelated." + "reason": "Tensor-parallel training and BeitImageProcessorFast label reduction are unrelated." }, { "left": "issue:41628", "right": "issue:42548", "accept": false, - "reason": "Different import symbols failing to load; same broad area, not the same defect." + "reason": "Different missing exports: AutoImageProcessor vs PreTrainedModel." }, { "left": "issue:43475", "right": "issue:46617", "accept": false, - "reason": "SAM 3 Video attribute error and trust_remote_code cache loading failure are unrelated." + "reason": "SAM 3 Video missing attribute and cache FileNotFoundError are unrelated." }, { "left": "issue:44315", "right": "issue:46209", "accept": false, - "reason": "Liger Kernel application and FP8 inference on B200 are different code paths." + "reason": "Model_init Liger-kernel issue and FP8 inference on B200 are different bugs." }, { "left": "issue:45362", "right": "issue:45850", "accept": false, - "reason": "Specific Qwen3.5 chat crash and a generic 'too many models break' complaint are not the same concrete bug." - }, - { - "left": "issue:43673", - "right": "issue:44492", - "accept": false, - "reason": "Generation cache missing and a typo in cache strategies are not the same issue." + "reason": "Specific Qwen3.5 chat crash is not the same as the broad 'v5 breaks many models' complaint." }, { "left": "issue:44623", "right": "issue:45230", "accept": false, - "reason": "Missing processor save files and a vague bug report do not identify the same underlying change." + "reason": "A concrete save_pretrained file bug and an unspecified 'Bug report' are not the same underlying issue." } ] }, @@ -16803,16 +16795,16 @@ "issue:44273|issue:45542", "issue:44038|issue:45125", "issue:44273|issue:45003", - "issue:44393|issue:45753", "issue:44393|issue:45071", + "issue:44393|issue:45753", "issue:43576|issue:44556", "issue:43525|issue:45850", "issue:41669|issue:43673", "issue:44230|issue:46095", "issue:44556|issue:45092", "issue:43572|issue:45850", - "issue:44492|issue:45850", "issue:45278|issue:45850", + "issue:44492|issue:45850", "issue:42371|issue:44492", "issue:41669|issue:46218", "issue:43408|issue:44315", @@ -16821,8 +16813,8 @@ "issue:43576|issue:44908", "issue:44246|issue:44273", "issue:43976|issue:44393", - "issue:44704|issue:44908", "issue:44295|issue:45230", + "issue:44704|issue:44908", "issue:44908|issue:45071", "issue:44297|issue:45230", "issue:41669|issue:44351", @@ -16842,280 +16834,286 @@ "issue:44273|issue:44908", "issue:43576|issue:46218" ], - "prepared_review_unit_hash": "fbd266dba02bdcf600328adf02d64e885407b40e7d379306b88299eda4c217e3", + "prepared_review_unit_hash": "b09c14bbdbaf7438408ac2f81dc1ce09b26b28754e1985dce5b254fb330a3a9b", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 41, - "serialized_chars": 38848, - "estimated_input_tokens": 9712, - "estimated_eval_tokens": 19680 + "serialized_chars": 38850, + "estimated_input_tokens": 9713, + "estimated_eval_tokens": 19682 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly a grab bag of unrelated Transformer issues, with a few loose similarities around v5 regressions, imports, and lazy loading. None of the soft pairs look like the same concrete bug/change strongly enough to merge, so I would keep them separate.", - "confidence": 0.79, - "canonical_issue_reason": "Issue 45278 is the best anchor for the cluster because it is the most concrete broad regression report: many import errors after upgrading to v5.5.0. It represents the shared v5 breakage theme better than the vaguer umbrella complaints.", + "summary": "This cluster is mostly a grab bag of Transformers v5 regression reports and model-specific compatibility bugs. The titles are superficially similar, but the concrete failure modes differ enough that none of the soft edges look like true duplicates.", + "confidence": 0.39, + "canonical_issue_reason": "issue:45850 is the broadest umbrella report here, capturing the general 'v5 breaks too many models' theme that many of the other reports orbit around.", "canonical_pr_reason": null, - "best_issue_reason": "45278 is the most suitable overall issue to triage against because it is broad but still actionable, and it is more precise than the generic 'v5 breaks too many models' style reports.", + "best_issue_reason": "issue:45850 is the best cluster anchor by scope, but it is still too vague to represent the concrete bugs in the other issues.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44273", "right": "issue:45542", "accept": false, - "reason": "Different bugs: lazy loading/import behavior vs a TensorFlow backend error in a tensorboard-only install." + "reason": "Lazy loading/import performance is a different problem from the undefined TensorFlow backend error." }, { "left": "issue:44038", "right": "issue:45125", "accept": false, - "reason": "Same model family, but one is a general v5/Qwen3-VL-Moe issue and the other is a specific missing _tp_plan tensor-parallelism bug." + "reason": "Both involve Qwen3/Moe, but one is a broader model bug and the other is missing tensor-parallel metadata." }, { "left": "issue:44273", "right": "issue:45003", "accept": false, - "reason": "Both touch import/lazy-loading internals, but one is a lazy loading failure and the other is an unsafe sys.modules access bug; not the same concrete issue." + "reason": "Lazy loading not working and unsafe sys.modules access are related to imports, but not the same bug." }, { "left": "issue:44393", - "right": "issue:45753", + "right": "issue:45071", "accept": false, - "reason": "Both involve Qwen3-VL/Qwen3.5-VL style models, but the symptoms and likely code paths differ completely." + "reason": "Qwen3-VL bounding-box errors are unrelated to PretrainedConfig type-checking regressions." }, { "left": "issue:44393", - "right": "issue:45071", + "right": "issue:45753", "accept": false, - "reason": "A Qwen3-VL bounding-box output problem is unrelated to PretrainedConfig type-checking breakage." + "reason": "Both are Qwen3-VL-family issues, but one is bbox output failure 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 transformers env command and the other is checkpoint reload compatibility." + "reason": "Broken env command and checkpoint reload incompatibility are both v5 regressions, but they are distinct code paths." }, { "left": "issue:43525", "right": "issue:45850", "accept": false, - "reason": "Specific missing pad_token_id config attribute vs a vague umbrella complaint about many models breaking." + "reason": "Missing pad_token_id in Llama4Config is a specific config bug, not the broad model-breakage complaint." }, { "left": "issue:41669", "right": "issue:43673", "accept": false, - "reason": "Import performance from star-imports is not the same as a missing GenerationMixin cache during chunked prefill." + "reason": "Import-star slowdown during model import is different from missing GenerationMixin cache during chunked prefill." }, { "left": "issue:44230", "right": "issue:46095", "accept": false, - "reason": "One is fp8 inference support for Qwen3-VL/Qwen3.5-Moe; the other is a DeepSeek weight conversion question." + "reason": "FP8 support for Qwen3-VL/Moe and a DeepSeek conversion-script question are not the same issue." }, { "left": "issue:44556", "right": "issue:45092", "accept": false, - "reason": "Both are checkpoint-loading regressions after v5, but they describe different failure modes and model families." + "reason": "Both mention upgrading to v5, but one is general checkpoint reload breakage and the other is remote-code meta-init incompatibility." }, { "left": "issue:43572", "right": "issue:45850", "accept": false, - "reason": "A StableLmConfig pad_token_idx regression is not the same as a generic 'many models break' report." + "reason": "StableLmConfig pad_token_idx is a specific config migration bug, not the broad v5 breakage report." }, { - "left": "issue:44492", + "left": "issue:45278", "right": "issue:45850", "accept": false, - "reason": "A typo in cache strategies is unrelated to the broad v5 breakage complaint." + "reason": "Both are broad v5 regression complaints, but one is specifically about import errors after upgrade." }, { - "left": "issue:45278", + "left": "issue:44492", "right": "issue:45850", "accept": false, - "reason": "Both are broad v5 complaints, but one is specifically about import errors while the other is a vague meta complaint; not the same bug." + "reason": "A cache-strategy typo is a narrow docs/wording issue, not the general model-breakage problem." }, { "left": "issue:42371", "right": "issue:44492", "accept": false, - "reason": "TF32 settings guidance and a cache-strategy typo are unrelated." + "reason": "TF32 API guidance and cache-strategy typo are unrelated fixes." }, { "left": "issue:41669", "right": "issue:46218", "accept": false, - "reason": "Import-time slowdown from star imports is not the same as failing to import Qwen3_5ForCausalLM." + "reason": "Import-star performance regression is not the same as failing to import Qwen3_5ForCausalLM." }, { "left": "issue:43408", "right": "issue:44315", "accept": false, - "reason": "A model-type mismatch warning is unrelated to Liger Kernel not being applied with model_init." + "reason": "sam3_tracker config warning and Liger Kernel not applied with model_init are unrelated issues." }, { "left": "issue:42890", "right": "issue:44908", "accept": false, - "reason": "Missing set_seed in tests is unrelated to inverse_sqrt scheduler kwargs being ignored." + "reason": "Missing set_seed in tests and inverse_sqrt kwargs being ignored are unrelated test/scheduler problems." }, { "left": "issue:43352", "right": "issue:44393", "accept": false, - "reason": "Flash Attention 2 support in NemotronH is unrelated to a Qwen3-VL 2D bounding-box hallucination/output bug." + "reason": "Flash Attention 2 support for NemotronH and Qwen3-VL bbox output errors are different model-specific bugs." }, { "left": "issue:43576", "right": "issue:44908", "accept": false, - "reason": "The transformers env command and scheduler kwargs handling are different code paths and symptoms." + "reason": "The env command regression is not the same underlying bug as scheduler kwargs being ignored." }, { "left": "issue:44246", "right": "issue:44273", "accept": false, - "reason": "Both relate loosely to import/lazy loading, but one is intermittent import slowness and the other is lazy loading malfunction; too broad to merge." + "reason": "Both concern import behavior, but one is intermittent slow import and the other is lazy loading failure." }, { "left": "issue:43976", "right": "issue:44393", "accept": false, - "reason": "Python version compatibility and Qwen3-VL output errors are unrelated." + "reason": "Python-version incompatibility is unrelated to the Qwen3-VL hallucination/bbox issue." }, { - "left": "issue:44704", - "right": "issue:44908", + "left": "issue:44295", + "right": "issue:45230", "accept": false, - "reason": "AutoProcessor kwargs passing and scheduler kwargs handling are separate APIs with different bugs." + "reason": "A position_ids buffer access error is too specific to be merged with a generic bug report." }, { - "left": "issue:44295", - "right": "issue:45230", + "left": "issue:44704", + "right": "issue:44908", "accept": false, - "reason": "A position_ids buffer access error is unrelated to a generic bug report." + "reason": "AutoProcessor cached_file kwargs handling is unrelated to inverse_sqrt scheduler kwargs." }, { "left": "issue:44908", "right": "issue:45071", "accept": false, - "reason": "Scheduler kwargs and PretrainedConfig type checking are unrelated." + "reason": "Scheduler kwargs plumbing and PretrainedConfig type checking are different failures." }, { "left": "issue:44297", "right": "issue:45230", "accept": false, - "reason": "Tokenizer save_pretrained metadata mismatch is unrelated to a generic bug report." + "reason": "Tokenizer save_pretrained mismatch is a concrete bug; the other item is too generic to match." + }, + { + "left": "issue:44315", + "right": "issue:44829", + "accept": false, + "reason": "Liger Kernel application in model_init and degenerate flash_attention_3 training are different code-path problems." }, { "left": "issue:41669", "right": "issue:44351", "accept": false, - "reason": "Star-import performance and a missing HybridCache import are different import-related issues." + "reason": "Import-star slowdown is not the same as a missing HybridCache export." }, { "left": "issue:44492", "right": "issue:46328", "accept": false, - "reason": "A cache-strategy typo and SequenceFeatureExtractor.pad list conversion overhead are different problems." + "reason": "Cache-strategy typo and SequenceFeatureExtractor.pad performance are unrelated." }, { "left": "issue:41669", "right": "issue:45341", "accept": false, - "reason": "Import-star cleanup is unrelated to a small testing_utils bug." + "reason": "Model import slowdown and a testing_utils bug are unrelated." }, { "left": "issue:43976", "right": "issue:45850", "accept": false, - "reason": "Python compatibility and broad v5 breakage are not the same underlying issue." + "reason": "Python 3.9 incompatibility is a different regression class from broad model breakage." }, { "left": "issue:44393", "right": "issue:45478", "accept": false, - "reason": "Two Qwen3.5/Qwen3-VL issues, but the reported failures differ and do not point to one concrete bug." + "reason": "Qwen3-VL bbox hallucinations and Qwen3.5 Moe from_pretrained errors are different model failures." }, { "left": "issue:43976", "right": "issue:46218", "accept": false, - "reason": "Python version support and missing Qwen3_5ForCausalLM import are separate compatibility issues." + "reason": "Python-version support issues are unrelated to importing Qwen3_5ForCausalLM." }, { "left": "issue:44077", "right": "issue:44861", "accept": false, - "reason": "Optional post_init enforcement and a tied-weights AttributeError are unrelated." + "reason": "Optional post_init policy and _get_tied_weight_keys crashing are different model-initialization bugs." }, { "left": "issue:43827", "right": "issue:44908", "accept": false, - "reason": "Documentation still mentioning pipeline() and scheduler kwarg handling are unrelated." + "reason": "Docs still referencing pipeline() is unrelated to scheduler kwargs handling." }, { "left": "issue:31795", "right": "issue:33453", "accept": false, - "reason": "Documentation confusion in model.forward and tokenizer loading regression are different issues." + "reason": "Model.forward docs confusion and tokenizer loading regression are unrelated issues." }, { "left": "issue:44315", "right": "issue:44829", "accept": false, - "reason": "Liger Kernel application in model_init and flash_attention_3 degenerate training are different failures." + "reason": "Model-init integration for Liger Kernel is not the same as flash_attention_3 causing degenerate training." }, { "left": "issue:43576", "right": "issue:44273", "accept": false, - "reason": "The transformers env command and lazy loading problems are different features and code paths." + "reason": "The env command regression is different from lazy loading not working properly." }, { "left": "issue:44360", "right": "issue:45468", "accept": false, - "reason": "A DSA indexer discussion and a Gemma-4 positional-encoding bug are unrelated." + "reason": "DSA indexer activation discussion and Gemma-4 audio positional encoding bug are unrelated." }, { "left": "issue:44655", "right": "issue:44861", "accept": false, - "reason": "Saving Pipeline objects and _get_tied_weight_keys crashing are unrelated." + "reason": "Pipeline save_pretrained failure and tied-weight-key crashes are separate serialization/init bugs." }, { "left": "issue:45230", "right": "issue:45310", "accept": false, - "reason": "A generic bug report does not identify the same concrete issue as the Qwen3.5 MoE from_pretrained error." + "reason": "A generic bug report does not identify the same Qwen3.5 Moe from_pretrained error." }, { "left": "issue:44273", "right": "issue:44908", "accept": false, - "reason": "Lazy loading/import behavior and scheduler kwarg handling are unrelated." + "reason": "Lazy loading/import behavior and inverse_sqrt scheduler kwargs are unrelated." }, { "left": "issue:43576", "right": "issue:46218", "accept": false, - "reason": "The env command regression and missing Qwen3_5ForCausalLM import are different problems." + "reason": "Broken env command and missing Qwen3_5ForCausalLM import are different import-related issues." } ] }, @@ -17237,316 +17235,316 @@ "issue:43240|issue:43668", "issue:43668|issue:43813" ], - "prepared_review_unit_hash": "37006d11933b7cd413201728818989247908bcf123555c55884dae4185123355", + "prepared_review_unit_hash": "7eb27afa28ecbce1b6fb1896d7597741d032b3bd6ef31d752230824f4241bad0", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 47, - "serialized_chars": 40846, - "estimated_input_tokens": 10212, - "estimated_eval_tokens": 20680 + "serialized_chars": 40842, + "estimated_input_tokens": 10211, + "estimated_eval_tokens": 20678 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly false-positive lexical similarity: the items span unrelated bugs/features across models, tokenizer loading, configs, and API behavior, with no true duplicate pair and no PRs to merge.", - "confidence": 0.94, - "canonical_issue_reason": "No single issue is a safe canonical; the cluster splits into unrelated topics (tokenizer/load regressions, flash-attention requests, config/import bugs, and model-specific implementation issues).", + "summary": "This is a loose grab-bag of Transformers issues: model-specific regressions, tokenizer/loading bugs, performance problems, and import/API breakages. There is no strong single underlying defect tying the cluster together; the closest thing to an umbrella topic is broad v5 regression fallout.", + "confidence": 0.41, + "canonical_issue_reason": "issue:45850 is the broadest umbrella-style report in the set ('Since 5.0 version it breaks too many models'), so it best represents the cluster despite being vague.", "canonical_pr_reason": null, - "best_issue_reason": "No strong global best issue either; 43519 has the most soft links, but the linked issues are semantically mismatched, so it is not a reliable representative.", + "best_issue_reason": "issue:45850 is the most cluster-like representative because it covers multiple model breakages rather than one narrow symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43519", "right": "issue:44485", "accept": false, - "reason": "Different bugs: Qwen3VL timestamp calculation vs GLM-5 RoPE implementation." + "reason": "Different model families and different bugs: timestamp calculation vs RoPE implementation." }, { "left": "issue:43519", "right": "issue:44908", "accept": false, - "reason": "Different subsystems: processor timestamp logic vs scheduler kwargs handling." + "reason": "Processor timestamp math and scheduler kwargs are unrelated code paths." }, { "left": "issue:41669", "right": "issue:43704", "accept": false, - "reason": "Import star cleanup is unrelated to a Qwen3ForCausalLM VRAM leak." + "reason": "Import-star performance issue and VRAM leak are unrelated symptoms." }, { "left": "issue:46018", "right": "issue:46172", "accept": false, - "reason": "Same model family, but different concrete bugs: shared expert gating vs comb transpose/dtype issue." + "reason": "Both mention deepseek-v4, but one is gating logic and the other is a tensor dtype/transpose bug." }, { "left": "issue:31795", "right": "issue:34689", "accept": false, - "reason": "Documentation confusion is not the same as a model-loading regression." + "reason": "Documentation confusion for forward args vs a model-loading regression are not the same bug." }, { "left": "issue:31515", "right": "issue:31795", "accept": false, - "reason": "Slow checkpoint loading and forward-argument docs are unrelated." + "reason": "Slow from_pretrained loading and forward-doc confusion are different issues." }, { "left": "issue:43704", "right": "issue:44485", "accept": false, - "reason": "VRAM leak in threaded dataloaders is unrelated to GLM-5 RoPE code." + "reason": "VRAM leak under dataloader threads does not match RoPE implementation concerns." }, { "left": "issue:44485", "right": "issue:45468", "accept": false, - "reason": "Different model implementations: GLM-5 RoPE vs Gemma-4 audio positional encoding." + "reason": "Different models and different numerical/encoding problems." }, { "left": "issue:45850", "right": "issue:46218", "accept": false, - "reason": "Too broad vs specific import failure; not the same change or bug." + "reason": "Broad version-breakage complaint vs a specific import error; not the same failure mode." }, { "left": "issue:42371", "right": "issue:43704", "accept": false, - "reason": "TF32 settings are unrelated to a Qwen3ForCausalLM memory leak." + "reason": "TF32 configuration API guidance is unrelated to a VRAM leak." }, { "left": "issue:43502", "right": "issue:43519", "accept": false, - "reason": "Local-files-only network requests and timestamp calculation are unrelated." + "reason": "local_files_only network access and timestamp calculation are different bugs." }, { "left": "issue:43519", "right": "issue:44623", "accept": false, - "reason": "Processor timestamp bug is unrelated to save_pretrained file output." + "reason": "Processor timestamp math and save_pretrained missing files are separate processor issues." }, { "left": "issue:43519", "right": "issue:43704", "accept": false, - "reason": "Different code paths: timestamp calculation vs VRAM leak." + "reason": "Unrelated processor bug vs multi-thread VRAM leak." }, { "left": "issue:44263", "right": "issue:45468", "accept": false, - "reason": "GlmMoeDsaIndexer split behavior is unrelated to Gemma-4 audio encoding." + "reason": "Different model internals and different arithmetic/shape issues." }, { "left": "issue:43298", "right": "issue:46095", "accept": false, - "reason": "Continuous batching audio support and DeepSeek-V4 weight conversion are different requests." + "reason": "Continuous batching audio support and a DeepSeek-V4 weight-conversion question are not the same bug." }, { "left": "issue:43606", "right": "issue:45468", "accept": false, - "reason": "CPU offload device mismatch is unrelated to Gemma-4 audio positional encoding." + "reason": "CPU offload device mismatch vs Gemma-4 positional encoding are unrelated." }, { "left": "issue:44261", "right": "issue:45468", "accept": false, - "reason": "RMS norm epsilon precision issue is unrelated to Gemma-4 audio positional encoding." + "reason": "Layernorm epsilon precision bug and Gemma-4 positional encoding are different code paths." }, { "left": "issue:44351", "right": "issue:46218", "accept": false, - "reason": "Different import failures for different symbols/modules." + "reason": "Both are import errors, but they concern different missing symbols/modules." }, { "left": "issue:43352", "right": "issue:46218", "accept": false, - "reason": "Nemotron FA2 support and Qwen3_5 import failure are unrelated." + "reason": "Flash Attention support gap for Nemotron is unrelated to importing Qwen3_5ForCausalLM." }, { "left": "issue:42913", "right": "issue:43066", "accept": false, - "reason": "Both tokenizer-related, but the symptoms and likely fixes differ." + "reason": "Both are tokenizer v5-related, but one is a broad behavioral regression and the other a specific decoder-type issue." }, { "left": "issue:28348", "right": "issue:45250", "accept": false, - "reason": "GPT2-specific FA2 support request vs a generic flash-attention issue; not the same concrete change." + "reason": "GPT2LMHeadModel-specific Flash Attention support request vs a generic Flash Attention topic." }, { "left": "issue:44908", "right": "issue:44998", "accept": false, - "reason": "Scheduler kwargs bug is unrelated to the unrelated title." + "reason": "Scheduler kwargs bug and an unrelated issue titled 'Unemployment' are clearly not the same." }, { "left": "issue:44492", "right": "issue:44998", "accept": false, - "reason": "Cache strategy typo and unrelated title are not duplicates." + "reason": "A typo in cache strategy docs/code is unrelated to the other issue." }, { "left": "issue:43618", "right": "issue:44998", "accept": false, - "reason": "CLIPOutput attentions regression is unrelated to the unrelated title." + "reason": "CLIPOutput attentions regression is unrelated to the other issue." }, { "left": "issue:44998", "right": "issue:45850", "accept": false, - "reason": "No meaningful duplication; one is unrelated noise and the other is a broad complaint." + "reason": "The latter is a broad Transformers v5 regression; 'Unemployment' is unrelated noise." }, { "left": "issue:44938", "right": "issue:44998", "accept": false, - "reason": "Python 3.14 load failure is unrelated to the unrelated title." + "reason": "Python 3.14 load failure is unrelated to the other issue." }, { "left": "issue:44861", "right": "issue:45290", "accept": false, - "reason": "Tied-weights helper crash and chat-template tool-call crash are different failures." + "reason": "Tied-weights key handling and chat template tool-call crashes are different bugs." }, { "left": "issue:43525", "right": "issue:44861", "accept": false, - "reason": "Missing pad_token_id in Llama4Config is unrelated to tied-weights key handling." + "reason": "Missing pad_token_id on Llama4Config is unrelated to tied-weight key handling." }, { "left": "issue:31515", "right": "issue:38617", "accept": false, - "reason": "Slow from_pretrained loading and missing layer_type_validation import are unrelated." + "reason": "Slow checkpoint loading and an import error for layer_type_validation are unrelated." }, { "left": "issue:43572", "right": "issue:44861", "accept": false, - "reason": "Missing pad_token_idx and tied-weights list crash are different config/code-path bugs." + "reason": "StableLmConfig pad_token_idx regression is not the same as tied-weight key crashes." }, { "left": "issue:28348", "right": "issue:43352", "accept": false, - "reason": "Both mention FA2 support, but they target different models and separate implementations." + "reason": "Different models and different Flash Attention 2.0 support gaps." }, { "left": "issue:44230", "right": "issue:44623", "accept": false, - "reason": "FP8 inference support and missing processor save files are unrelated." + "reason": "FP8 support request and missing processor save files are unrelated." }, { "left": "issue:30333", "right": "issue:31515", "accept": false, - "reason": "MLflow reporting on failed jobs is unrelated to slow checkpoint loading." + "reason": "MLFlow reporting on failed jobs and slow from_pretrained loading are different problems." }, { "left": "issue:31515", "right": "issue:36296", "accept": false, - "reason": "Checkpoint loading slowness and tensor-parallel training bug are unrelated." + "reason": "Slow checkpoint loading and tensor-parallel training bug are unrelated." }, { "left": "issue:44230", "right": "issue:45310", "accept": false, - "reason": "FP8 support request and Qwen3.5 MoE from_pretrained regression are different problems." + "reason": "FP8 support request and Qwen3.5 MoE from_pretrained error are different issues." }, { "left": "issue:31515", "right": "issue:33453", "accept": false, - "reason": "from_pretrained slowness and tokenizer-loading regression are different issues." + "reason": "Load slowness and tokenizer-loading regression are not the same bug." }, { "left": "issue:43298", "right": "issue:44230", "accept": false, - "reason": "Audio continuous batching and Qwen3-VL/Qwen3.5 FP8 support are unrelated features." + "reason": "Audio continuous batching support and FP8 inference support are separate feature gaps." }, { "left": "issue:43717", "right": "issue:46144", "accept": false, - "reason": "Mamba init_weights behavior and RoFormer attention interface usage are unrelated." + "reason": "Init-weights behavior in Mamba models and RoFormer attention interface usage are unrelated." }, { "left": "issue:42936", "right": "issue:46018", "accept": false, - "reason": "Mask2Former ignore_value handling is unrelated to DeepSeek-V4 gating." + "reason": "Mask2Former ignore_value misuse and DeepSeek-V4 shared-expert gating are different model-specific bugs." }, { "left": "issue:43668", "right": "issue:44031", "accept": false, - "reason": "Incorrect type hint and tokenizer regex warning are unrelated." + "reason": "Incorrect type hint and regex warning regression are unrelated." }, { "left": "issue:42936", "right": "issue:43240", "accept": false, - "reason": "Mask2Former ignore_value and loss kwargs propagation are unrelated." + "reason": "Mask2Former ignore_value and loss kwargs propagation are different issues." }, { "left": "issue:43240", "right": "issue:46144", "accept": false, - "reason": "Loss kwargs propagation and RoFormer attention interface use are unrelated." + "reason": "Loss kwargs propagation and RoFormer attention interface usage are unrelated." }, { "left": "issue:42936", "right": "issue:44031", "accept": false, - "reason": "Mask2Former ignore_value bug and tokenizer regex warning are unrelated." + "reason": "Different subsystems: Mask2Former config/use bug vs tokenizer regex warnings." }, { "left": "issue:42936", "right": "issue:46144", "accept": false, - "reason": "Mask2Former ignore_value bug and RoFormer attention interface are unrelated." + "reason": "Different model families and unrelated bugs." }, { "left": "issue:42936", "right": "issue:43813", "accept": false, - "reason": "Mask2Former ignore_value bug and a typo in cache strategies are unrelated." + "reason": "Mask2Former ignore_value issue is unrelated to the typo in conversion code." }, { "left": "issue:43240", "right": "issue:43668", "accept": false, - "reason": "Loss kwargs propagation and ModernBERT type-hint issue are unrelated." + "reason": "Loss-kwargs forwarding and a config type-hint typo are unrelated." }, { "left": "issue:43668", "right": "issue:43813", "accept": false, - "reason": "ModernBERT type hint and cache-strategy typo are unrelated." + "reason": "Type-hint correction and conversion typo are different trivial fixes." } ] }, @@ -17578,20 +17576,20 @@ "issue:37934|issue:43668", "issue:42936|issue:43668" ], - "prepared_review_unit_hash": "ebff4a254b6e9bfeadf495d4dd750145c068c671d7f393d20fa90cf7716459a9", + "prepared_review_unit_hash": "f1ebd3f31807d351532933f488be4730514dd107d0089edacfe24b9b0a47be73", "component_budget": { "node_count": 469, "item_count": 469, "soft_pair_count": 1718, - "serialized_chars": 891815, - "estimated_input_tokens": 222954, - "estimated_eval_tokens": 446164 + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 }, "budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 2311, + "serialized_chars": 2312, "estimated_input_tokens": 578, "estimated_eval_tokens": 1412 }, @@ -17600,24 +17598,24 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No duplicate cluster here: the three issues are about different models and different defect types (norm semantics, an unused ignore_value, and a type-hint mismatch).", - "confidence": 0.97, - "canonical_issue_reason": "There is no shared underlying bug to canonicalize; if a representative must be named, issue 42936 is the most concrete code defect, but it is still unrelated to the other two.", + "summary": "These three issues are unrelated: one asks about model design, one reports an unused ignore_value in Mask2Former, and one reports an incorrect type hint in ModernBERTConfig. No pair looks like the same underlying bug or change.", + "confidence": 0.93, + "canonical_issue_reason": "issue:42936 is the most concrete and actionable bug report: it identifies a specific runtime/behavioral problem (ignore_value defined but not used) with a clear code path in Mask2Former.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 42936 is the strongest standalone report because it describes a specific implementation problem, whereas 37934 is a design question and 43668 is a type-hint issue.", + "best_issue_reason": "issue:42936 is the best representative issue because it describes an actual functional defect rather than a design question or a type annotation mismatch.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:37934", "right": "issue:43668", "accept": false, - "reason": "Different components and different problem types: Llama4TextL2Norm behavior question vs ModernBERTConfig annotation error. No same underlying bug." + "reason": "Different subjects and failure modes: Llama4TextL2Norm normalization semantics vs. ModernBERTConfig type annotation. No shared code path or concrete same-bug evidence." }, { "left": "issue:42936", "right": "issue:43668", "accept": false, - "reason": "Mask2former ignore_value usage bug and ModernBERTConfig norm_eps type hint are unrelated code paths and defects." + "reason": "One is a Mask2Former behavioral bug about ignore_value not being used; the other is a ModernBERTConfig type-hint issue. They are distinct and not mergeable as one underlying fix." } ] }, @@ -17650,52 +17648,52 @@ "issue:10519|issue:11005", "issue:11005|issue:11307" ], - "prepared_review_unit_hash": "4ff7be35793faca2be4dfc34df740db2b2b5ae5f940f703ca2494e96a848ce01", + "prepared_review_unit_hash": "96f90a9a0e55dc267a3f6fe181e4a74fa8c9261c1f0dbd3b8fa0fce7a92c0b0d", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 2877, - "estimated_input_tokens": 720, - "estimated_eval_tokens": 1696 + "serialized_chars": 2874, + "estimated_input_tokens": 719, + "estimated_eval_tokens": 1694 }, "budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 2877, - "estimated_input_tokens": 720, - "estimated_eval_tokens": 1696 + "serialized_chars": 2874, + "estimated_input_tokens": 719, + "estimated_eval_tokens": 1694 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The three issues are unrelated feature requests: tokenizer truncation direction, ReduceLROnPlateau-like scheduling, and Wav2Vec2 word boundary offsets. There is no duplicate cluster here.", + "summary": "The three issues are about unrelated feature requests: truncation direction in tokenization, ReduceLROnPlateau-like scheduling behavior, and Wav2Vec2 word time offsets. There is no credible duplicate relationship among them.", "confidence": 0.97, - "canonical_issue_reason": "No canonical issue: these items do not describe the same bug or change, just separate feature requests in different areas.", + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "If a single issue must be picked as the cluster anchor, 11307 is the most concrete and discussion-heavy, but it is still unrelated to the other two.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:10519", "right": "issue:11307", "accept": false, - "reason": "Token truncation-direction feature request vs Wav2Vec2 word time offsets; different code paths and different user problems." + "reason": "Different domains and code paths: token truncation behavior vs. Wav2Vec2 word boundary offsets." }, { "left": "issue:10519", "right": "issue:11005", "accept": false, - "reason": "Tokenizer truncation behavior is unrelated to a ReduceLROnPlateau-like learning-rate scheduler feature." + "reason": "Unrelated requests: truncation strategy in preprocessing vs. learning-rate scheduler functionality." }, { "left": "issue:11005", "right": "issue:11307", "accept": false, - "reason": "Learning-rate scheduling request and speech-model word offset extraction are not the same underlying bug or change." + "reason": "No shared underlying bug or change; scheduler feature request and Wav2Vec2 timing offsets are unrelated." } ] }, @@ -17727,7 +17725,7 @@ "issue:13244|issue:16618", "issue:15980|issue:16618" ], - "prepared_review_unit_hash": "4c3577824442a17f23eb1a10915184de2258af670fd2db2432670571d38730be", + "prepared_review_unit_hash": "b7c348bdc4f8ada873d1df621d290ccd67e5a5ba6ccc08a4efb235fd5c539a10", "component_budget": { "node_count": 3, "item_count": 3, @@ -17749,24 +17747,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The cluster is heterogeneous: 13244 is about TAPAS tokenization, while 15980 and 16618 are both about model download/load failures but with different concrete symptoms. None of the pairs look like the same underlying bug.", - "confidence": 0.82, - "canonical_issue_reason": "issue 16618 is the best representative because it is the most general and most discussed of the three, and it is the only item with any soft similarity to another issue.", + "summary": "This cluster is not cohesive: issue #13244 is about Tapas tokenization and is unrelated to the model-loading/login problems in #15980 and #16618. The title overlap between #15980 and #16618 is too weak to treat them as the same underlying bug without stronger body/diff evidence.", + "confidence": 0.94, + "canonical_issue_reason": "If forced to pick one, #16618 is the broadest and most discussion-heavy model-loading report, but it is still only a loose fit and does not represent the whole cluster well.", "canonical_pr_reason": null, - "best_issue_reason": "issue 16618 is the strongest standalone issue in the set: it is the broad model-loading failure and has the highest discussion activity, so it is the most useful canonical anchor even though the cluster is not a true duplicate set.", + "best_issue_reason": "No strong global best issue exists because the cluster mixes unrelated problems; #16618 is only the closest representative by breadth, not by true semantic match.", "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. Different subsystem and likely different fix." + "reason": "Different problem domains: Tapas tokenization vs. model loading/authentication. Same-subsystem wording is not enough to make them the same bug." }, { "left": "issue:15980", "right": "issue:16618", "accept": false, - "reason": "Both mention model download/load errors, but one is specifically a private-model auth/error-message issue and the other is a generic inability to load bert-base-uncased. Too different in concrete failure mode to treat as the same bug." + "reason": "Both involve model loading, but one is specifically about a bad error when downloading a private model without login, while the other is a generic failure to load bert-base-uncased. Not enough evidence they are the same concrete issue." } ] }, @@ -17796,12 +17794,12 @@ "soft_pairs": [ "issue:26350|issue:33260" ], - "prepared_review_unit_hash": "97f54628efe8359a4f6c4174d31b54c052b623ac5cf306a6f77baf0dec6562c2", + "prepared_review_unit_hash": "94e44515163ddc2cddbe9fa99ae3c1d7a222c982c5d8241a5f88497ce36b03d3", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1589, + "serialized_chars": 1590, "estimated_input_tokens": 398, "estimated_eval_tokens": 1052 }, @@ -17809,7 +17807,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1589, + "serialized_chars": 1590, "estimated_input_tokens": 398, "estimated_eval_tokens": 1052 }, @@ -17818,18 +17816,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates: one requests Flash Attention 2 support across more architectures, the other requests GGUF support across more architectures. They share only a generic community-contribution framing and broad architecture scope.", + "summary": "These two issues are not duplicates: one tracks Flash Attention 2 support for more architectures, the other tracks GGUF support for more architectures. They are separate feature efforts with different implementation paths.", "confidence": 0.98, - "canonical_issue_reason": null, + "canonical_issue_reason": "Issue 26350 is the older and more-discussed tracker, so it is the better representative if one issue must stand in for the cluster.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Issue 26350 has higher discussion activity and broader momentum, making it the best issue representative here.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:26350", "right": "issue:33260", "accept": false, - "reason": "Different underlying feature requests and code paths: Flash Attention 2 support vs GGUF support. Similar phrasing about adding support for more architectures is too generic to treat as the same issue." + "reason": "Different features and code paths: Flash Attention 2 support vs GGUF support. They share a broad \u2018community contribution\u2019 framing only, not the same underlying change or bug." } ] }, @@ -17888,187 +17886,187 @@ "issue:44370|issue:46226", "issue:46222|issue:46226" ], - "prepared_review_unit_hash": "78b28103d1e643f7a76d5abfc5aaf52a76b0b6590737f6c0e465af5ed04833b8", + "prepared_review_unit_hash": "395f65a190c06ba61c1a7773c94a719b488dd2a95eb86a5087ec0eced2bca8ef", "component_budget": { "node_count": 8, "item_count": 8, "soft_pair_count": 24, - "serialized_chars": 13528, - "estimated_input_tokens": 3382, - "estimated_eval_tokens": 7020 + "serialized_chars": 13529, + "estimated_input_tokens": 3383, + "estimated_eval_tokens": 7022 }, "budget": { "node_count": 8, "item_count": 8, "soft_pair_count": 24, - "serialized_chars": 13528, - "estimated_input_tokens": 3382, - "estimated_eval_tokens": 7020 + "serialized_chars": 13529, + "estimated_input_tokens": 3383, + "estimated_eval_tokens": 7022 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster splits into distinct language-specific translation requests, one near-identical placeholder-template subgroup (43834/44365/44370) that looks duplicate-like but is only supported by template similarity in the packet, and a separate exact-duplicate Thai pair (46222/46226). The cross-language items should stay separate.", + "summary": "This cluster is mostly a mix of similar-looking docs-translation requests across different languages. Only the exact placeholder-template duplicates and the identical Thai pair look like true duplicates; the concrete Turkish/Spanish/Romanian issues are not the same underlying issue.", "confidence": 0.79, - "canonical_issue_reason": "issue:43834 is the best canonical issue for the placeholder-template subgroup because it is the earliest of the near-identical generic i18n template issues and has the broadest, most generic wording.", + "canonical_issue_reason": "issue:43834 is the oldest exact-match member of the repeated placeholder-template subgroup, so it is the best canonical target for that duplicated text.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43834 is the best representative issue overall for the placeholder-template subgroup; it captures the repeated generic i18n template pattern without overclaiming exact duplication, while the language-specific requests and Thai pair are separate cases.", + "best_issue_reason": "issue:43834 is the clearest representative of the exact duplicate subgroup; there is no single best issue for the whole mixed cluster because several language-specific requests are distinct.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:27088", "right": "issue:44370", "accept": false, - "reason": "Turkish translation request vs generic placeholder template; not the same underlying issue." + "reason": "Different language-targeted docs requests; same broad theme is not enough to make them duplicates." }, { "left": "issue:27088", "right": "issue:44365", "accept": false, - "reason": "Different language-specific request vs placeholder template; similar wording is not enough." + "reason": "Turkish translation request vs. a generic placeholder-title issue; not the same concrete issue." }, { "left": "issue:27088", "right": "issue:43834", "accept": false, - "reason": "Concrete Turkish request vs generic placeholder template; not a duplicate." + "reason": "Concrete Turkish request does not match the generic placeholder-template issue." }, { "left": "issue:27088", "right": "issue:38435", "accept": false, - "reason": "Different language translation request; same docs-translation theme but not the same issue." + "reason": "Different target languages, so these are separate translation requests." }, { "left": "issue:27088", "right": "issue:28936", "accept": false, - "reason": "Turkish vs Spanish translation request; distinct underlying tasks." + "reason": "Turkish and Spanish docs translation requests are distinct issues." }, { "left": "issue:28936", "right": "issue:43834", "accept": false, - "reason": "Spanish translation request vs generic placeholder template; only broad similarity." + "reason": "Spanish request is concrete; the other item is a generic placeholder-template issue." }, { "left": "issue:28936", "right": "issue:44365", "accept": false, - "reason": "Different language request; not the same underlying issue." + "reason": "Spanish translation request vs. placeholder-template issue; not the same underlying bug/change." }, { "left": "issue:28936", "right": "issue:38435", "accept": false, - "reason": "Spanish vs Romanian translation request; distinct issues." + "reason": "Spanish and Romanian docs translation requests are different language targets." }, { "left": "issue:28936", "right": "issue:44370", "accept": false, - "reason": "Different language request; template similarity alone is not enough." + "reason": "Concrete Spanish request does not match the generic placeholder-template issue." }, { "left": "issue:38435", "right": "issue:44370", "accept": false, - "reason": "Romanian translation request vs placeholder template; not the same issue." + "reason": "Romanian request vs. generic placeholder-template issue; not enough to merge." }, { "left": "issue:38435", "right": "issue:44365", "accept": false, - "reason": "Different underlying language request; only superficially similar." + "reason": "Romanian request is concrete, while the other item is a generic placeholder-template issue." }, { "left": "issue:38435", "right": "issue:43834", "accept": false, - "reason": "Concrete Romanian request vs generic placeholder template; not a duplicate." + "reason": "Different issue contents: a specific Romanian request versus a generic placeholder title." }, { "left": "issue:38435", "right": "issue:46222", "accept": false, - "reason": "Unrelated language-specific request despite superficial docs-translation similarity." + "reason": "Romanian translation request and the Thai-text issue are unrelated." }, { "left": "issue:38435", "right": "issue:46226", "accept": false, - "reason": "Unrelated language-specific request despite superficial docs-translation similarity." + "reason": "Romanian translation request and the Thai-text issue are unrelated." }, { "left": "issue:43834", "right": "issue:44370", "accept": true, - "reason": "Near-identical generic placeholder i18n issue; looks like the same underlying template request." + "reason": "Same placeholder translation-doc title and near-identical body; looks like a duplicate issue." }, { "left": "issue:43834", "right": "issue:44365", "accept": true, - "reason": "Near-identical generic placeholder i18n issue; highly likely duplicate of the same template request." + "reason": "Same placeholder title and almost identical issue text; likely the same underlying issue." }, { "left": "issue:43834", "right": "issue:46222", "accept": false, - "reason": "Generic placeholder template vs concrete Thai text; not the same issue." + "reason": "Different title/content; high superficial similarity is not enough to call these duplicates." }, { "left": "issue:43834", "right": "issue:46226", "accept": false, - "reason": "Generic placeholder template vs concrete Thai text; not the same issue." + "reason": "Different title/content; this is not the same underlying issue." }, { "left": "issue:44365", "right": "issue:44370", "accept": true, - "reason": "Near-identical placeholder i18n issues; same underlying template request." + "reason": "Identical placeholder title and highly similar body; these appear to be duplicates." }, { "left": "issue:44365", "right": "issue:46222", "accept": false, - "reason": "Placeholder template issue vs concrete Thai request; not a duplicate." + "reason": "Placeholder translation issue vs. unrelated Thai-text issue; not a duplicate pair." }, { "left": "issue:44365", "right": "issue:46226", "accept": false, - "reason": "Placeholder template issue vs concrete Thai request; not a duplicate." + "reason": "Placeholder translation issue vs. unrelated Thai-text issue; not a duplicate pair." }, { "left": "issue:44370", "right": "issue:46222", "accept": false, - "reason": "Placeholder template issue vs concrete Thai request; not the same underlying task." + "reason": "Placeholder translation issue vs. unrelated Thai-text issue; not a duplicate pair." }, { "left": "issue:44370", "right": "issue:46226", "accept": false, - "reason": "Placeholder template issue vs concrete Thai request; not the same underlying task." + "reason": "Placeholder translation issue vs. unrelated Thai-text issue; not a duplicate pair." }, { "left": "issue:46222", "right": "issue:46226", "accept": true, - "reason": "Same Thai-titled issue with effectively identical content; true duplicate." + "reason": "Same Thai title and effectively identical content; clear duplicate pair." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The cluster summary is a bit too strong on evidence. The Thai pair (46222/46226) looks like an exact duplicate, but the placeholder i18n issues (43834/44365/44370) are only near-identical from the packet, not proven exact duplicates. Please soften that wording to match the available evidence. The rest of the duplicate/non-duplicate split is reasonably grounded." + "accept": true, + "feedback": "The summary stays grounded in the packet: it distinguishes the language-specific translation requests from the repeated placeholder-template subgroup and the identical Thai pair, without overclaiming duplicates among the mixed-language items. The soft-edge verdicts are conservative, and the canonical/best-issue choice is reasonable for the duplicated placeholder subgroup." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -18141,12 +18139,12 @@ "issue:43837|issue:45177", "issue:45322|issue:46133" ], - "prepared_review_unit_hash": "d46b08286da568f8fdcd2130cce932b6bd284f5d3acaa8de1b0d44c0f980be85", + "prepared_review_unit_hash": "49b5b513bf6c9f6fdd3acab8fe7b33d88bb694b2e0d4b77778793abe879fab77", "component_budget": { "node_count": 16, "item_count": 16, "soft_pair_count": 36, - "serialized_chars": 21658, + "serialized_chars": 21659, "estimated_input_tokens": 5415, "estimated_eval_tokens": 11086 }, @@ -18154,7 +18152,7 @@ "node_count": 16, "item_count": 16, "soft_pair_count": 36, - "serialized_chars": 21658, + "serialized_chars": 21659, "estimated_input_tokens": 5415, "estimated_eval_tokens": 11086 }, @@ -18163,156 +18161,156 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly a family of model-support request issues with similar wording, but they target different concrete models and capabilities. None of the soft pairs look like true duplicates or the same underlying change, so no soft edges are accepted.", - "confidence": 0.93, - "canonical_issue_reason": "No issue is a clear duplicate canonical for the whole cluster. If one must be chosen as the most representative, issue 43671 is a broad, request-style model-support ticket with many similarity links, but it still describes a distinct TTS model request rather than the same change as the others.", + "summary": "This cluster is a set of distinct model-support requests (different architectures and modalities), not true duplicates. The broadest umbrella issue is 42738, but it still does not subsume the rest as the same underlying request.", + "confidence": 0.91, + "canonical_issue_reason": "No single issue is a true duplicate anchor; 42738 is the closest umbrella because it is the most general model-family request, but the others are separate model additions.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43671 is the best single representative of the cluster because it is a straightforward model-support request and sits near several other audio/model-request issues, but it should not be treated as a duplicate of them.", + "best_issue_reason": "42738 is the best representative only in a loose thematic sense, since it is broader than the other titles; however, it should not be treated as a duplicate of the rest.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36817", "right": "issue:41211", "accept": false, - "reason": "Different model requests: EuroBert config vs DEIMv2. Similar template, not the same underlying addition." + "reason": "Different model requests: EuroBert vs DEIMv2 are unrelated additions." }, { "left": "issue:36817", "right": "issue:43671", "accept": false, - "reason": "BERT config request vs Qwen3-TTS support; unrelated model families and changes." + "reason": "BERT-style config support vs Qwen3-TTS audio support are different code paths and model types." }, { "left": "issue:36817", "right": "issue:36879", "accept": false, - "reason": "EuroBert config and RF-DETR are different model additions with no shared concrete change." + "reason": "EuroBert config support and RF-DETR model support are separate model additions." }, { "left": "issue:36879", "right": "issue:41211", "accept": false, - "reason": "RF-DETR and DEIMv2 are distinct vision models; not one bug/change." + "reason": "RF-DETR and DEIMv2 are different vision models, not the same change." }, { "left": "issue:36879", "right": "issue:43909", "accept": false, - "reason": "RF-DETR vs LFM2.5 Audio 1.5B are different modalities and code paths." + "reason": "RF-DETR vs LFM2.5 Audio are different modalities and implementations." }, { "left": "issue:36879", "right": "issue:44121", "accept": false, - "reason": "Different model requests: RF-DETR vs weight-sparse transformer." + "reason": "RF-DETR detection support and weight-sparse transformer support are unrelated requests." }, { "left": "issue:36879", "right": "issue:43837", "accept": false, - "reason": "RF-DETR and Qwen3-ASR are separate model-support requests." + "reason": "RF-DETR and Qwen3-ASR are different model families and tasks." }, { "left": "issue:36879", "right": "issue:43671", "accept": false, - "reason": "RF-DETR computer-vision support and Qwen3-TTS audio support are unrelated." + "reason": "RF-DETR vision support and Qwen3-TTS audio support are not the same underlying change." }, { "left": "issue:37015", "right": "issue:43837", "accept": false, - "reason": "NeoBERT vs Qwen3-ASR are different model families and implementation paths." + "reason": "NeoBERT and Qwen3-ASR target different models and code paths." }, { "left": "issue:37015", "right": "issue:44121", "accept": false, - "reason": "NeoBERT and weight-sparse transformer are not the same model/change." + "reason": "NeoBERT support is unrelated to circuit-sparsity transformer support." }, { "left": "issue:37015", "right": "issue:43909", "accept": false, - "reason": "NeoBERT text model request vs audio model request." + "reason": "NeoBERT and LFM2.5 Audio are distinct model requests." }, { "left": "issue:36879", "right": "issue:37015", "accept": false, - "reason": "RF-DETR and NeoBERT are distinct model additions." + "reason": "RF-DETR and NeoBERT are unrelated additions." }, { "left": "issue:37015", "right": "issue:42503", "accept": false, - "reason": "NeoBERT and ModernVBERT are different models; only the request pattern is similar." + "reason": "NeoBERT and ModernVBERT are different model families." }, { "left": "issue:41115", "right": "issue:43837", "accept": false, - "reason": "MiniCPM3 architecture support and Qwen3-ASR are different concrete model requests." + "reason": "MiniCPM3 architecture support and Qwen3-ASR are different tasks." }, { "left": "issue:41115", "right": "issue:44121", "accept": false, - "reason": "MiniCPM3 and weight-sparse transformer are unrelated additions." + "reason": "MiniCPM3 architecture support and weight-sparse transformer support are unrelated." }, { "left": "issue:41115", "right": "issue:42503", "accept": false, - "reason": "MiniCPM3 architecture and ModernVBERT models are distinct support requests." + "reason": "MiniCPM3 and ModernVBERT are different model additions." }, { "left": "issue:41115", "right": "issue:45027", "accept": false, - "reason": "MiniCPM3 and Voxtral TTS are different model types and code paths." + "reason": "MiniCPM3 architecture support and Voxtral TTS support are separate changes." }, { "left": "issue:41211", "right": "issue:43671", "accept": false, - "reason": "DEIMv2 and Qwen3-TTS are unrelated model support requests." + "reason": "DEIMv2 and Qwen3-TTS are different model types and tasks." }, { "left": "issue:42503", "right": "issue:43837", "accept": false, - "reason": "ModernVBERT and Qwen3-ASR are different model families." + "reason": "ModernVBERT and Qwen3-ASR are unrelated model support requests." }, { "left": "issue:42503", "right": "issue:43671", "accept": false, - "reason": "ModernVBERT and Qwen3-TTS are distinct changes." + "reason": "ModernVBERT and Qwen3-TTS are different model families." }, { "left": "issue:42503", "right": "issue:42738", "accept": false, - "reason": "ModernVBERT models and BERT-like models with RoPE are related in theme but not the same concrete request." + "reason": "ModernVBERT and BERT-like RoPE support overlap only loosely, but are not the same concrete request." }, { "left": "issue:42738", "right": "issue:45027", "accept": false, - "reason": "BERT-like RoPE support vs Voxtral TTS support are different code paths." + "reason": "BERT-like RoPE support and Voxtral TTS support are not the same underlying bug or change." }, { "left": "issue:43671", "right": "issue:43837", "accept": false, - "reason": "Qwen3-TTS and Qwen3-ASR are closely related but still separate model requests and likely separate implementations." + "reason": "Qwen3-TTS and Qwen3-ASR are related by naming only; they are separate audio tasks and fixes." }, { "left": "issue:43671", "right": "issue:45027", "accept": false, - "reason": "Qwen3-TTS and Voxtral TTS are different models; not the same underlying change." + "reason": "Qwen3-TTS and Voxtral TTS are different model integrations." }, { "left": "issue:43671", @@ -18324,67 +18322,67 @@ "left": "issue:43671", "right": "issue:44121", "accept": false, - "reason": "Audio TTS support vs weight-sparse transformer request are unrelated." + "reason": "TTS support and weight-sparse transformer support are different code paths." }, { "left": "issue:43671", "right": "issue:43909", "accept": false, - "reason": "Qwen3-TTS and LFM2.5 Audio are different audio-model support items." + "reason": "Qwen3-TTS and LFM2.5 Audio are separate audio model integrations, not duplicates." }, { "left": "issue:43837", "right": "issue:43909", "accept": false, - "reason": "Qwen3-ASR and LFM2.5 Audio are different speech model requests." + "reason": "Qwen3-ASR and LFM2.5 Audio are different audio models/tasks." }, { "left": "issue:43837", "right": "issue:45322", "accept": false, - "reason": "ASR support and EUPE perception encoder are different model additions." + "reason": "Qwen3-ASR and EUPE are unrelated model requests." }, { "left": "issue:43837", "right": "issue:44121", "accept": false, - "reason": "Qwen3-ASR and circuit-sparsity transformer are unrelated changes." + "reason": "ASR support and weight-sparse transformer support are not the same concrete change." }, { "left": "issue:43909", "right": "issue:44205", "accept": false, - "reason": "LFM2.5 Audio and SAM3-LiteText are different models and domains." + "reason": "LFM2.5 Audio and SAM3-LiteText are different model families." }, { "left": "issue:44121", "right": "issue:45322", "accept": false, - "reason": "Weight-sparse transformer request and EUPE encoder request are unrelated." + "reason": "Weight-sparse transformer support and EUPE support are unrelated." }, { "left": "issue:43837", "right": "issue:45027", "accept": false, - "reason": "Qwen3-ASR and Voxtral TTS are different speech-model additions." + "reason": "Qwen3-ASR and Voxtral TTS are different audio model support requests." }, { "left": "issue:37015", "right": "issue:45027", "accept": false, - "reason": "NeoBERT and Voxtral TTS are unrelated model support requests." + "reason": "NeoBERT and Voxtral TTS are unrelated model additions." }, { "left": "issue:43837", "right": "issue:45177", "accept": false, - "reason": "Qwen3-ASR and DeepSeek-OCR2 are different model families." + "reason": "Qwen3-ASR and DeepSeek-OCR2 are different model integrations." }, { "left": "issue:45322", "right": "issue:46133", "accept": false, - "reason": "EUPE encoder and TIPSv2 are both vision-language related, but they are different model requests with different intended architecture changes." + "reason": "EUPE and TIPSv2 are distinct vision-language model requests, not one concrete fix." } ] }, @@ -18414,12 +18412,12 @@ "soft_pairs": [ "issue:38740|issue:46443" ], - "prepared_review_unit_hash": "d569c1d4ca1aca1c481ad902a7c53b2d2fd9b59726c61b986a1245b2d22a979c", + "prepared_review_unit_hash": "cdb241b76377d1a93f968eab2d84fd884e0688596722d682adab66dc339a3d64", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1506, + "serialized_chars": 1505, "estimated_input_tokens": 377, "estimated_eval_tokens": 1010 }, @@ -18427,7 +18425,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1506, + "serialized_chars": 1505, "estimated_input_tokens": 377, "estimated_eval_tokens": 1010 }, @@ -18436,18 +18434,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are unrelated: one is a docs request to add Pruna as an optimization framework, the other is a Vision RoPE refactor.", - "confidence": 0.99, - "canonical_issue_reason": "No canonical issue for deduplication; the titles and likely scopes are unrelated.", + "summary": "The two issues are unrelated: one is a docs request to add a framework listing, the other is a technical vision RoPE refactor. They should not be clustered as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": "Issue #46443 is the stronger representative because it is the substantive technical problem in the cluster; #38740 is a narrow documentation request and does not describe the same kind of work.", "canonical_pr_reason": null, - "best_issue_reason": "If one issue must be retained as the cluster anchor, 46443 is the more concrete technical issue, but it is not a duplicate of 38740.", + "best_issue_reason": "#46443 is the best issue overall because it describes a concrete codebase refactor with clear implementation scope, unlike the docs-only request in #38740.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:38740", "right": "issue:46443", "accept": false, - "reason": "Different topics and code paths: a documentation/framework listing request versus a vision-model RoPE refactor. They do not describe the same underlying bug or change." + "reason": "Different domains and intents: one asks for documentation to add 'pruna' as an optimization framework, while the other concerns a Vision RoPE refactor. They do not appear to be the same underlying bug or change." } ] }, @@ -18477,12 +18475,12 @@ "soft_pairs": [ "issue:41084|issue:43037" ], - "prepared_review_unit_hash": "711ecfb2350dba9789d73810aac994ee78e4e16606e1da88480fa74d31cf2670", + "prepared_review_unit_hash": "0fa2efda354708653686c0d7369b3fc0e2090a19be97ae32e45f8291ede0d39f", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1480, + "serialized_chars": 1479, "estimated_input_tokens": 370, "estimated_eval_tokens": 996 }, @@ -18490,7 +18488,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1480, + "serialized_chars": 1479, "estimated_input_tokens": 370, "estimated_eval_tokens": 996 }, @@ -18499,18 +18497,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are unrelated: one asks for a generic block decoding feature, while the other requests support for a specific model version (DeepSeek v3.2).", + "summary": "The two issues appear unrelated: one is about block decoding behavior, the other is a model-specific support request for DeepSeek v3.2. No duplicate relationship is evident.", "confidence": 0.96, - "canonical_issue_reason": "issue:41084 is the more general feature request and is better suited as the broader issue, but it is not a duplicate of issue:43037.", + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:41084 is the clearest standalone issue because it describes a distinct implementation area (block decoding) without tying it to a particular model integration.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41084", "right": "issue:43037", "accept": false, - "reason": "The titles and implied scopes differ: block decoding is a core decoding capability, while DeepSeek v3.2 support is a model-specific compatibility request. They do not appear to be the same underlying bug or change." + "reason": "Titles and implied scopes differ substantially; there is no shared code path, explicit target, or description evidence that these are the same bug/request." } ] }, @@ -18540,7 +18538,7 @@ "soft_pairs": [ "issue:43598|issue:43600" ], - "prepared_review_unit_hash": "92255e2bb219f8352635275b6fa14de7642db0f3ddaa058bd27254f3ce96dedc", + "prepared_review_unit_hash": "b74e521f9aa4728e9792c3ac755c444c30eb3723a7d6fb1fff1906ec659ea0fd", "component_budget": { "node_count": 2, "item_count": 2, @@ -18562,18 +18560,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related only at a high level through Trainer customizability, but they target different concrete behaviors: one is about `remove_unused_column` handling, the other about FLOPs estimation for custom trainers. They should not be merged as duplicates.", - "confidence": 0.95, - "canonical_issue_reason": "Issue #43598 is the broader, more general Trainer customizability request and is the better representative of this small cluster, though the two issues are not duplicates.", + "summary": "The two issues are thematically related (custom trainers / Trainer extensibility), but they describe different problems: one is about `remove_unused_column` behavior, the other about FLOPS estimation for custom trainers. I would not treat them as the same underlying bug.", + "confidence": 0.91, + "canonical_issue_reason": "Issue #43598 is the broader and more directly API-facing Trainer customizability request, so it is the better representative issue for this small cluster.", "canonical_pr_reason": null, - "best_issue_reason": "#43598 is the more general and conceptually central issue title, making it the better issue to anchor the cluster even though #43600 addresses a separate problem.", + "best_issue_reason": "#43598 has the clearer, more general problem statement and is more likely to anchor discussion around Trainer customization than the narrower FLOPS-estimation concern.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43598", "right": "issue:43600", "accept": false, - "reason": "Related theme only; one concerns `remove_unused_column` behavior, the other FLOPs counting for custom trainers. Different code paths and fixes, so not the same underlying bug." + "reason": "They share the general Trainer/custom-trainer theme, but the concrete problems differ: `remove_unused_column` customization vs. FLOPS counting logic for custom trainers. No indication they are the same bug or change." } ] }, @@ -18603,12 +18601,12 @@ "soft_pairs": [ "issue:43684|issue:44405" ], - "prepared_review_unit_hash": "ddf148fbc0e198a7a581f54f2a71b888b55c7ff31f9d4d780914c21f0db95ea4", + "prepared_review_unit_hash": "537345de5da0b91be24c3f059528467f3c1f6249e67479c4ae853409052c6e3a", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1611, + "serialized_chars": 1612, "estimated_input_tokens": 403, "estimated_eval_tokens": 1062 }, @@ -18616,7 +18614,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1611, + "serialized_chars": 1612, "estimated_input_tokens": 403, "estimated_eval_tokens": 1062 }, @@ -18625,18 +18623,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues request different Hugging Face model integration features for different model/config families; they are not duplicates.", - "confidence": 0.98, - "canonical_issue_reason": null, + "summary": "These are related but not duplicates: one requests broad Qwen3-Omni registration in AutoModel/AutoModelForConditionalGeneration, while the other asks for sequence-classification support for Qwen3.5. They target different model families and different auto-class heads.", + "confidence": 0.93, + "canonical_issue_reason": "Issue #43684 is the broader, more central request about registering Qwen3-Omni with core auto classes; it better represents the cluster\u2019s general model-registration theme.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Issue #43684 is the best representative issue because it covers a wider API-registration gap rather than a single task-specific head.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43684", "right": "issue:44405", "accept": false, - "reason": "Different targets and bugs: one asks for Qwen3-Omni registration in AutoModel/AutoModelForConditionalGeneration, the other asks for AutoModelForSequenceClassification support for Qwen3.5 (Qwen3_5Config)." + "reason": "Different underlying requests: Qwen3-Omni auto-model registration vs Qwen3.5 sequence-classification support. Same broad area, but not the same bug/change." } ] }, @@ -18673,12 +18671,12 @@ "issue:43845|issue:43859", "issue:43846|issue:43859" ], - "prepared_review_unit_hash": "e7f78fbd08da588cd60d38b2efd2c10a19c5dd7b77aaf7eac697e422d36fae44", + "prepared_review_unit_hash": "751b68aadb4e8cdbebd650e3483155214fd1746936b8075bacefc52dc58bc75e", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4267, + "serialized_chars": 4265, "estimated_input_tokens": 1067, "estimated_eval_tokens": 2390 }, @@ -18686,7 +18684,7 @@ "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4267, + "serialized_chars": 4265, "estimated_input_tokens": 1067, "estimated_eval_tokens": 2390 }, @@ -18695,58 +18693,55 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Three issues share the exact title \"huggingface \" and appear to be duplicate reports; the \"Fett\" issue is unrelated. I would collapse the huggingface trio onto the earliest one.", - "confidence": 0.86, - "canonical_issue_reason": "Issue 43845 is the earliest of the near-identical \"huggingface \" reports and is a reasonable representative for the duplicate cluster.", + "summary": "One clear outlier is issue 43835 ('Fett'). The three 'huggingface ' issues look like a near-duplicate cluster by title alone, but there isn\u2019t enough content to prove they report the exact same underlying problem.", + "confidence": 0.58, + "canonical_issue_reason": "issue:43845 is the best representative of the likely duplicate 'huggingface ' set because it is the earliest of the three matching-title reports and has the most context available among them.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43845 best represents the cluster because it is earliest and matches the repeated title pattern seen in 43846 and 43859.", + "best_issue_reason": "issue:43845 is the most suitable cluster anchor for the same reason: earliest matching-title report, while 43835 appears unrelated.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43835", "right": "issue:43845", "accept": false, - "reason": "Different titles with no concrete shared details; \"Fett\" does not look like the same report as \"huggingface \"." + "reason": "Different titles ('Fett' vs 'huggingface '), no shared files/targets, and no body evidence tying them to the same issue." }, { "left": "issue:43835", "right": "issue:43846", "accept": false, - "reason": "No evidence these refer to the same underlying problem; the titles are unrelated." + "reason": "Different titles with no concrete overlap beyond generic similarity; insufficient evidence of the same underlying bug." }, { "left": "issue:43835", "right": "issue:43859", "accept": false, - "reason": "No evidence these refer to the same underlying problem; the titles are unrelated." + "reason": "Different titles and no shared targets/files or descriptive details linking them." }, { "left": "issue:43845", "right": "issue:43846", - "accept": true, - "reason": "Exact same title and very close creation times; these look like duplicate reports of the same issue." + "accept": false, + "reason": "Identical truncated titles are suggestive, but the bodies are too sparse to confirm the same underlying bug rather than separate reports." }, { "left": "issue:43845", "right": "issue:43859", - "accept": true, - "reason": "Same exact title and highly similar clustering signal; likely the same underlying report." + "accept": false, + "reason": "Same short title, but no concrete problem description, target, or code-path evidence to safely merge as duplicates." }, { "left": "issue:43846", "right": "issue:43859", - "accept": true, - "reason": "Same exact title and strong similarity signal; likely duplicate reports of the same issue." + "accept": false, + "reason": "Same short title only; the available metadata is too thin to prove these are the same issue." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet: 43845, 43846, and 43859 share the exact same title and are close enough in time/similarity to justify treating them as a likely duplicate cluster, while 43835 has a different title and no concrete overlap. The representative choice of 43845 as the earliest in that cluster is reasonable, and the soft-edge verdicts stay conservative." - }, - "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 }, @@ -18780,88 +18775,88 @@ "issue:44572|issue:44937", "issue:44573|issue:44937" ], - "prepared_review_unit_hash": "51820d8f6ad532b0dee15a575322be065468664bdcad8514862b184564f77a89", + "prepared_review_unit_hash": "ab715269cc3debbd73d875564acc153da8495549378cc777816b031c2ed859e0", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 9, - "serialized_chars": 5788, - "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": 5788, - "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": "This cluster is dominated by low-signal spam/placeholder issues and one unrelated documentation-style post; there is no clear shared underlying bug or change.", - "confidence": 0.83, - "canonical_issue_reason": "Issue 44572 is the least arbitrary anchor because it is one of the exact `` duplicates (with 44573), but it still does not represent a real substantive issue family.", + "summary": "These are not duplicates of a single real bug/change; they look like unrelated low-quality or spammy issues, with only one vaguely substantive documentation request.", + "confidence": 0.96, + "canonical_issue_reason": "No issue clearly represents a shared underlying problem. If a representative must be chosen, #44418 is the least spam-like and the only one with a meaningful topic.", "canonical_pr_reason": null, - "best_issue_reason": "44572 is the best available representative only because it matches another issue exactly in form; overall the cluster is too noisy and heterogeneous to be a strong canonical issue.", + "best_issue_reason": "#44418 is the best representative only because it is the only non-spam, semantically interpretable issue title; it is still too vague to serve as a strong canonical for the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44371", "right": "issue:44572", "accept": false, - "reason": "Both are placeholder spam titles, but there is no evidence they describe the same underlying issue." + "reason": "Both are generic spam placeholders; no evidence they describe the same bug or change." }, { "left": "issue:44371", "right": "issue:44573", "accept": false, - "reason": "Both are placeholder spam titles, but nothing ties them to one concrete duplicate bug." + "reason": "Both are generic spam placeholders; no shared underlying issue is described." }, { "left": "issue:44371", "right": "issue:44418", "accept": false, - "reason": "One is spam and the other is a generic documentation request; they are not the same issue." + "reason": "One is spam, the other is a vague documentation request; not the same underlying problem." }, { "left": "issue:44371", "right": "issue:44937", "accept": false, - "reason": "The titles are unrelated spam/marketing text with no shared underlying problem." + "reason": "Spam placeholder versus a specific-looking title about Google LLC; unrelated." }, { "left": "issue:44418", "right": "issue:44572", "accept": false, - "reason": "A documentation enhancement request is not the same as a spam placeholder issue." + "reason": "Vague documentation suggestion versus spam placeholder; no concrete overlap." }, { "left": "issue:44418", "right": "issue:44573", "accept": false, - "reason": "A documentation enhancement request is not the same as a spam placeholder issue." + "reason": "Vague documentation suggestion versus spam placeholder; not duplicative." }, { "left": "issue:44572", "right": "issue:44573", "accept": false, - "reason": "They match as generic spam placeholders, but that is not enough to show a shared underlying bug or change." + "reason": "Both are spam placeholders, but there is no evidence they refer to the same actual issue." }, { "left": "issue:44572", "right": "issue:44937", "accept": false, - "reason": "A spam placeholder and a Google LLC-related title do not indicate the same issue." + "reason": "Spam placeholder versus unrelated specific title; not the same bug/change." }, { "left": "issue:44573", "right": "issue:44937", "accept": false, - "reason": "A spam placeholder and a Google LLC-related title do not indicate the same issue." + "reason": "Spam placeholder versus unrelated specific title; not the same bug/change." } ] }, @@ -18891,46 +18886,46 @@ "soft_pairs": [ "issue:45145|issue:45151" ], - "prepared_review_unit_hash": "30d362801e8ab0da983ed6d6ab4b5a7291a9fc08162676eb5c1b84cf261fb146", + "prepared_review_unit_hash": "21ec7d7a715eafe58d5e092d7593867b6b3cc4176b2d25d79021ed0b0ddfcfc1", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1645, - "estimated_input_tokens": 412, - "estimated_eval_tokens": 1080 + "serialized_chars": 1644, + "estimated_input_tokens": 411, + "estimated_eval_tokens": 1078 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1645, - "estimated_input_tokens": 412, - "estimated_eval_tokens": 1080 + "serialized_chars": 1644, + "estimated_input_tokens": 411, + "estimated_eval_tokens": 1078 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two closed issues have the same title and very similar body lengths, so they are probably related duplicate reports, but the packet does not include the bodies or any shared links/files. That makes a hard duplicate claim too strong; treat them as likely duplicates with limited evidence.", + "summary": "The two closed issues are highly similar and likely duplicates: they have identical titles and very similar body sizes, but the evidence is only soft similarity, so I would not treat the duplicate relationship as definitive. #45145 is the better canonical representative because it has the slightly stronger linkage history.", "confidence": 0.84, - "canonical_issue_reason": "Issue #45145 is the better canonical choice because it was created first and has an inbound reference, which makes it the more central report.", + "canonical_issue_reason": "#45145 is a reasonable canonical issue because it matches #45151 exactly in title and has one inbound reference, giving it slightly better linkage history.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #45145 is the best overall issue candidate for the same reason: earliest creation time and slightly stronger evidence that it is the reference copy.", + "best_issue_reason": "#45145 is the best issue representative for the cluster because it is textually identical to #45151 and has the stronger reference signal.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45145", "right": "issue:45151", "accept": false, - "reason": "The titles match and the body lengths are close, but the packet does not expose the actual bodies or any shared files/links. That is enough for a likely-duplicate signal, not enough to confidently merge them as the same underlying issue." + "reason": "The issues look highly similar, but the evidence is only soft similarity with no explicit cross-links, shared targets, or diffs, and the pair is marked non-deterministic. That supports likely duplication, not a confident merge-level duplicate verdict." } ] }, "evaluator_result": { "accept": false, - "feedback": "The pairwise conclusion is a bit too strong for the evidence in the packet. The packet supports same title, close body lengths, and a soft similarity signal, but it does not show the actual bodies or any shared links/files, so claims like \"near-exact duplicates\" and \"no material differences\" are overstated. The canonical-choice rationale is grounded, but the soft-edge verdict should be more conservative." + "feedback": "The packet supports strong similarity: the titles are identical, body lengths are close, and issue #45145 has one inbound reference while #45151 has none. However, the soft pair is marked deterministic_accept=false, so accepting the duplicate verdict is too strong on the available evidence. The summary also overstates certainty by saying they are definitively the same underlying issue report; a more conservative phrasing would be that they appear highly similar or likely duplicates. Canonical selection of #45145 is reasonable, but the soft-edge verdict should be more cautious." }, "evaluator_used": true, "retried": true, @@ -19011,40 +19006,40 @@ "soft_pairs": [ "issue:45672|issue:45674" ], - "prepared_review_unit_hash": "01a1f1d567333290655d98d286ea63606b06fddd6140a28a85a34bfc60230c35", + "prepared_review_unit_hash": "aec920a2d4fee0abe927c9c4cdd112ca03cbf77210bc256ad8435337a16f56d7", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1673, - "estimated_input_tokens": 419, - "estimated_eval_tokens": 1094 + "serialized_chars": 1672, + "estimated_input_tokens": 418, + "estimated_eval_tokens": 1092 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1673, - "estimated_input_tokens": 419, - "estimated_eval_tokens": 1094 + "serialized_chars": 1672, + "estimated_input_tokens": 418, + "estimated_eval_tokens": 1092 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates: one is a Gemma4 audio-module dtype failure during NF4 inference, the other is a BitsAndBytesConfig regression that clears the default lm_head exclusion and triggers an AssertionError. They only overlap at a broad \u201c4-bit inference\u201d level.", - "confidence": 0.95, - "canonical_issue_reason": "No clear canonical duplicate: the two issues describe different root causes and different fixes.", + "summary": "These are two separate 4-bit inference bugs: one specific to Gemma4 audio modules with uint8 weights, and one in BitsAndBytesConfig module-exclusion handling. They are related by subsystem but not the same underlying issue.", + "confidence": 0.91, + "canonical_issue_reason": "Issue #45674 is the better cluster representative because it describes a broader, reusable configuration bug with a clear code-path in BitsAndBytesConfig, while #45672 is a model-specific failure mode.", "canonical_pr_reason": null, - "best_issue_reason": "issue 45674 is the better representative because it is a broader, config-level regression with a clearer general scope; issue 45672 is a narrower model/module-specific dtype bug.", + "best_issue_reason": "#45674 is the most generally useful issue to keep as the representative report: it has a clearer root-cause pattern, broader impact, and is less tied to one model family than #45672.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45672", "right": "issue:45674", "accept": false, - "reason": "Different underlying bugs: 45672 is about torch.finfo() being called on uint8 weights in Gemma4 audio modules, while 45674 is about llm_int8_skip_modules removing the default lm_head exclusion and causing an AssertionError. Same subsystem, not the same fix." + "reason": "Do not merge: the failures are different. #45672 is a torch.finfo TypeError triggered by Gemma4 audio modules using uint8 weights in NF4 inference; #45674 is an AssertionError caused by llm_int8_skip_modules clearing the default lm_head exclusion. Same broad area, different bug and fix path." } ] }, @@ -19074,7 +19069,7 @@ "soft_pairs": [ "issue:45762|issue:45763" ], - "prepared_review_unit_hash": "39bef1a3ff924b0e0b788d85ee9bc1367c6ded479dfc6c369831868dcba6f5e1", + "prepared_review_unit_hash": "9aa01dd876a025d5929f54c6ae227cdcdce7822630e0d0e6b37d94fff716dc52", "component_budget": { "node_count": 2, "item_count": 2, @@ -19096,18 +19091,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are just person-name/date titles with no body, links, or shared technical context; they do not appear to describe the same underlying bug or change.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "Two closed issues with unrelated person-name titles and no body/context; the soft similarity signal is not enough to treat them as duplicates.", + "confidence": 0.22, + "canonical_issue_reason": "Issue 45762 is only marginally earlier and otherwise just as sparse as 45763, so it is the least arbitrary representative among the issues.", "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a good canonical candidate because there is no evidence of a shared problem or any technical description to compare.", + "best_issue_reason": "Issue 45762 is the best issue candidate only by timestamp; neither issue provides enough context to indicate a stronger representative.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45762", "right": "issue:45763", "accept": false, - "reason": "Titles only contain different names and dates; no body or metadata suggests the same underlying issue." + "reason": "The titles refer to different people and dates, with no shared body, target, or problem description; they do not look like the same underlying issue." } ] }, @@ -19137,40 +19132,40 @@ "soft_pairs": [ "issue:45764|issue:45765" ], - "prepared_review_unit_hash": "dbefb33dda79d68f9e54324a78bcb9442ddb555f558c02f734745ab906adb5a2", + "prepared_review_unit_hash": "7cc8ac1b1f2ca7423097a338d38cdd18104a78854d877ac7802373a8fdd2c79f", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1480, - "estimated_input_tokens": 370, - "estimated_eval_tokens": 996 + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1480, - "estimated_input_tokens": 370, - "estimated_eval_tokens": 996 + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two empty closed issues with different personal names and dates; they do not describe the same bug or feature, so this cluster looks like unrelated noise rather than a duplicate set.", + "summary": "The two closed issues appear unrelated: both are standalone title-only entries with different personal names and dates, and there is no body, linkage, or shared technical context to indicate a duplicate relationship.", "confidence": 0.98, - "canonical_issue_reason": "No clear canonical issue: both issues are content-free and describe different named entities, not the same problem.", + "canonical_issue_reason": "issue:45764 is a reasonable canonical only by ordering; it has no substantive advantage over issue:45765 because both lack content and context.", "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a strong representative because neither contains actionable problem detail; if forced, either could serve only as a placeholder.", + "best_issue_reason": "Neither issue is a strong global canonical candidate since both are empty and non-technical, but issue:45764 is the slightly better placeholder due to being first.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45764", "right": "issue:45765", "accept": false, - "reason": "The titles reference different people and different dates, with no shared technical subject or actionable issue to indicate the same underlying bug." + "reason": "The titles are different names/dates with no shared technical subject, body, or links; this does not look like the same underlying issue." } ] }, @@ -19200,7 +19195,7 @@ "soft_pairs": [ "issue:45810|issue:46295" ], - "prepared_review_unit_hash": "7934498ef6730f5a7019c826260ead06a05760cfe2256614aa2d0ea0474131a6", + "prepared_review_unit_hash": "717d0d846340051464630a2075807245d2fe5de794e4bb90c19d672873c73dc0", "component_budget": { "node_count": 2, "item_count": 2, @@ -19222,18 +19217,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are about different feature areas: adding a Qwen3_5 token classification head for value modeling versus FP8 training support. They do not look like duplicates or the same underlying request.", - "confidence": 0.92, - "canonical_issue_reason": "issue:45810 is the more concrete, implementation-targeted feature request, with a specific model/class named and a clear intended use case.", + "summary": "These issues are not duplicates: one requests adding a specific token classification model class, the other asks for FP8 training support. They target different features and code paths.", + "confidence": 0.97, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:45810 is the best representative issue in this cluster because it is narrowly scoped and directly actionable, unlike the broader FP8 training request.", + "best_issue_reason": "Issue 45810 is the more concrete, narrowly scoped feature request and is easier to route or implement independently, but it is not a duplicate of 46295.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45810", "right": "issue:46295", "accept": false, - "reason": "Different underlying problems: one requests a new Qwen3_5 token-classification/value-model head, the other requests FP8 training support. No shared target code path or concrete overlap." + "reason": "Different underlying requests: model-class support for Qwen3_5 token classification vs FP8 training support. They do not describe the same bug or change." } ] }, @@ -19273,91 +19268,91 @@ "issue:45871|issue:45873", "issue:45872|issue:45873" ], - "prepared_review_unit_hash": "b268d1292f41f537f986528a6c299567512bf44dc0c57b03849a72dcd1db77be", + "prepared_review_unit_hash": "276b799f8799f5c797b95a925522dc2c78020d9ecdebd9979fa921234cfc78a6", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5725, - "estimated_input_tokens": 1432, - "estimated_eval_tokens": 3120 + "serialized_chars": 5724, + "estimated_input_tokens": 1431, + "estimated_eval_tokens": 3118 }, "budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5725, - "estimated_input_tokens": 1432, - "estimated_eval_tokens": 3120 + "serialized_chars": 5724, + "estimated_input_tokens": 1431, + "estimated_eval_tokens": 3118 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The cluster splits into two exact duplicate issue pairs: 45870/45871 and 45872/45873. Issue 45869 is thematically similar but not the same request, so it should stay separate.", - "confidence": 0.84, - "canonical_issue_reason": "45870 is an exact duplicate of 45871 and is the earliest clear instance of the DataForge financial example issue, making it a clean canonical anchor for that duplicate pair.", + "summary": "The cluster contains two likely duplicate issue pairs: 45870/45871 and 45872/45873. Issue 45869 is only thematically related to the financial-data domain and appears distinct from both request pairs.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 45870 is a good canonical representative for the first likely duplicate pair because it is the earlier of the two identical-title reports and cleanly anchors that request.", "canonical_pr_reason": null, - "best_issue_reason": "45872 is the most concrete, centrally named DataForge request in the cluster, and 45873 is an exact duplicate of it.", + "best_issue_reason": "Issue 45870 is the best single issue to keep as the cluster anchor because it is one of the clearest duplicate pair representatives and it predates 45871; the 45872/45873 pair looks like a separate request, and 45869 is unrelated.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45869", "right": "issue:45870", "accept": false, - "reason": "Related financial/DataForge theme, but different request: a dataset bundle ad versus a prediction example." + "reason": "Same broad financial domain, but one is a dataset bundle pitch and the other is an example request; not the same underlying issue." }, { "left": "issue:45869", "right": "issue:45871", "accept": false, - "reason": "Same broad topic only; not the same underlying issue or change." + "reason": "The dataset bundle and the example request are only loosely related by topic, not duplicates." }, { "left": "issue:45870", "right": "issue:45871", "accept": true, - "reason": "Exact same title and matching body length; these look like duplicate reports of the same issue." + "reason": "Same title and matching body length with high similarity; these appear to be duplicates for the same example request." }, { "left": "issue:45870", "right": "issue:45872", "accept": false, - "reason": "Both mention Financial/DataForge, but one is an example page and the other is a scraping-service request." + "reason": "They share a financial-time-series theme, but the request scope differs: example prediction versus a scraping-service ask." }, { "left": "issue:45870", "right": "issue:45873", "accept": false, - "reason": "Same broad DataForge theme, but not the same concrete issue." + "reason": "Different issue text and scope; not the same underlying request." }, { "left": "issue:45871", "right": "issue:45872", "accept": false, - "reason": "Different underlying request despite similar wording." + "reason": "The identical example request is still distinct from the scraping-service-focused request." }, { "left": "issue:45871", "right": "issue:45873", "accept": false, - "reason": "Different underlying request despite similar wording." + "reason": "Different request scope and wording; only loosely related by domain." }, { "left": "issue:45872", "right": "issue:45873", "accept": true, - "reason": "Exact same title and matching body length; these look like duplicate reports of the same issue." + "reason": "Same title and matching body length with high similarity; these appear to be duplicates for the same request." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The grouping is well grounded in the packet: 45870/45871 and 45872/45873 have identical titles, identical body lengths, and the pair scores indicate very strong similarity, while 45869 is only loosely related by theme. The soft-edge rejections are conservative, and the canonical/best anchors are defensible from the timestamps and titles." + "accept": false, + "feedback": "The clustering is mostly grounded, but the summary overstates the evidence by calling 45870/45871 and 45872/45873 \u201cexact duplicate\u201d pairs. The packet only shows identical titles, equal body lengths, and soft-similarity scores\u2014not the full bodies\u2014so the conservative wording should be \u201clikely duplicate\u201d or \u201cappears to be a duplicate.\u201d The soft-edge verdicts themselves are mostly conservative, but the duplicated/exact language is too strong for the available evidence." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -19381,40 +19376,40 @@ "soft_pairs": [ "issue:46439|issue:46441" ], - "prepared_review_unit_hash": "70598a700bfc63ba49a6279120d6a79153e817eb98a87b5d479734602b9b1110", + "prepared_review_unit_hash": "4d02d81009ed5786f80a59f9c6b1b1b34f3bf654cd8f2bfbad4a4f0140fe45b5", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1637, - "estimated_input_tokens": 410, - "estimated_eval_tokens": 1076 + "serialized_chars": 1635, + "estimated_input_tokens": 409, + "estimated_eval_tokens": 1074 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1637, - "estimated_input_tokens": 410, - "estimated_eval_tokens": 1076 + "serialized_chars": 1635, + "estimated_input_tokens": 409, + "estimated_eval_tokens": 1074 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both issues involve `cache_implementation=\"static\"` and `linear_attention`, but they describe different failure modes: one is shape corruption from `early_initialization`, the other is a `KeyError` in Qwen3.5. They look related, but not clearly the same underlying bug, so I would not deduplicate them.", - "confidence": 0.63, - "canonical_issue_reason": "Issue 46439 is the better representative issue because it states a concrete root symptom (`early_initialization` corrupts linear-attention layers and causes shape/runtime failures) rather than a model-specific crash report.", + "summary": "Two open issues both involve static cache and linear-attention in Qwen3.5, but they describe different failure modes: one is a shape corruption/RuntimeError during `early_initialization`, the other is a `KeyError: 'linear_attention'` when using `cache_implementation=\"static\"`.", + "confidence": 0.83, + "canonical_issue_reason": "Issue 46439 is the better canonical issue because it describes the more specific low-level failure in static-cache initialization and directly names the linear-attention shape corruption path.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 46439 is the strongest standalone report: it is broader, more diagnostic, and names the likely problematic code path, making it the best issue to keep as the reference if the cluster is later linked.", + "best_issue_reason": "Issue 46439 is the strongest representative of the cluster since it is more diagnostic and points to the likely underlying code-path bug, while 46441 reads more like a separate symptom or configuration lookup failure.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46439", "right": "issue:46441", "accept": false, - "reason": "Related subsystem, but not clearly the same bug. 46439 reports wrong shapes/runtime failure from `early_initialization`; 46441 reports a `KeyError: 'linear_attention'` on Qwen3.5. Different symptoms and likely different root causes, so do not merge as duplicates." + "reason": "Do not merge: both mention static cache and linear-attention, but one reports shape corruption/RuntimeError during early initialization and the other reports a KeyError for a missing `linear_attention` entry. Those are not clearly the same underlying bug." } ] }, @@ -19444,7 +19439,7 @@ "soft_pairs": [ "issue:46551|issue:46552" ], - "prepared_review_unit_hash": "eff75d89862632ecdd29afbdcdb19831d9ba16cb12699abd894f28ac78604123", + "prepared_review_unit_hash": "c8fd468e2a28e027b0fe0e8e2d146114acf9ef010e3c43f3fc3a97421385706a", "component_budget": { "node_count": 2, "item_count": 2, @@ -19466,27 +19461,27 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are near-identical duplicates about Wav2Vec2CTCTokenizer v5 incorrectly treating the word delimiter as a special token. Issue #46552 is the better canonical issue because it is the open, referenced thread; #46551 appears to be the duplicate/closed copy.", - "confidence": 0.99, - "canonical_issue_reason": "#46552 is the open issue with inbound references, so it is the better long-lived tracking thread for this exact bug report.", + "summary": "The two issues are very likely duplicates: they have identical titles, matching body lengths, and a soft-similarity score of 1.0. The open issue looks like the better canonical representative.", + "confidence": 0.94, + "canonical_issue_reason": "issue:46552 is the better canonical issue because it is open and already has inbound references, and it appears to describe the same report as issue:46551 based on the identical title and matching body length.", "canonical_pr_reason": null, - "best_issue_reason": "#46552 best represents the unique underlying issue because it matches #46551 exactly and is the active/open discussion target.", + "best_issue_reason": "issue:46552 is the strongest cluster representative since it is the active/open copy of the report and has received inbound references.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46551", "right": "issue:46552", "accept": true, - "reason": "Exact same title and bug description; these are duplicate reports of the same tokenizer special-token leakage issue." + "reason": "The pair has the same title, the same body length, and a soft-similarity score of 1.0, which strongly indicates the same underlying issue even though the body text is not shown here." } ] }, "evaluator_result": { - "accept": true, - "feedback": "Grounded in the packet: the two issues have identical titles and body lengths, and #46552 is the open item with inbound references while #46551 is closed. The duplicate verdict is appropriately conservative, and the canonical choice of #46552 is supported by the available metadata." + "accept": false, + "feedback": "Reject: the conclusion is plausible, but the response overstates evidence from the packet. The packet shows identical titles, equal body lengths, and a soft-similarity score of 1.0, but it does not actually display the bodies, so claiming they have the \"same body\" or are a \"clear duplicate pair\" is stronger than the available evidence. The canonical/open-issue reasoning is also slightly overstated as \"already being referenced\" is only supported for issue:46552 via inbound references, not for the pair as a whole." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -19499,8 +19494,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-39895-14", "nodes": [ @@ -19703,8 +19698,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-40546-2", "nodes": [ @@ -19766,8 +19761,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-40755-2", "nodes": [ @@ -19892,8 +19887,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-41797-3", "nodes": [ @@ -19963,8 +19958,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-42975-2", "nodes": [ @@ -20026,8 +20021,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-42978-6", "nodes": [ @@ -20163,8 +20158,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43291-14", "nodes": [ @@ -20392,8 +20387,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43488-2", "nodes": [ @@ -20455,8 +20450,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43612-2", "nodes": [ @@ -20518,8 +20513,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43651-2", "nodes": [ @@ -20581,8 +20576,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43656-2", "nodes": [ @@ -20710,8 +20705,8 @@ "review_unit_count": 2, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43996-28", "nodes": [ @@ -21275,8 +21270,8 @@ "review_unit_count": 2, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43996-28", "nodes": [ @@ -21370,8 +21365,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43998-13", "nodes": [ @@ -21717,8 +21712,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-44662-2", "nodes": [ @@ -21780,8 +21775,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-44979-2", "nodes": [ @@ -21843,8 +21838,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45086-2", "nodes": [ @@ -21906,8 +21901,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45115-4", "nodes": [ @@ -21985,8 +21980,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45332-2", "nodes": [ @@ -22048,8 +22043,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45513-2", "nodes": [ @@ -22111,8 +22106,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45548-2", "nodes": [ @@ -22174,8 +22169,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45639-2", "nodes": [ @@ -22237,8 +22232,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45670-3", "nodes": [ @@ -22315,8 +22310,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45720-2", "nodes": [ @@ -22444,8 +22439,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45932-2", "nodes": [ @@ -22507,8 +22502,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-46073-3", "nodes": [ @@ -22585,8 +22580,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-46126-2", "nodes": [ @@ -22648,8 +22643,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-46138-9", "nodes": [ @@ -22893,8 +22888,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-46152-2", "nodes": [ @@ -22956,8 +22951,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-46173-3", "nodes": [ @@ -23030,8 +23025,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-46236-2", "nodes": [ @@ -23093,8 +23088,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-46310-2", "nodes": [ @@ -23156,8 +23151,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-46376-2", "nodes": [ @@ -23219,8 +23214,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "llm", - "cache_hit": false, + "source": "cache", + "cache_hit": true, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-46603-2", "nodes": [ diff --git a/analysis/current/manifest.json b/analysis/current/manifest.json index 58265d7db2b21e87ea3d3f2086ecfcc3489879e7..0845269584ccc6769eacf38c0cd3b205299603f1 100644 --- a/analysis/current/manifest.json +++ b/analysis/current/manifest.json @@ -1,8 +1,8 @@ { - "analysis_id": "hybrid-model-20260614t180027z", + "analysis_id": "hybrid-model-20260615t000032z", "archived_artifacts": { - "hybrid": "snapshots/20260614T180027Z/analysis-runs/hybrid-model-20260614t180027z/analysis-report-hybrid.json", - "hybrid_reviews": "snapshots/20260614T180027Z/analysis-runs/hybrid-model-20260614t180027z/analysis-report-hybrid.llm-reviews.json" + "hybrid": "snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/analysis-report-hybrid.llm-reviews.json" }, "artifacts": { "hybrid": "analysis/current/analysis-report-hybrid.json", @@ -10,9 +10,9 @@ }, "channel": "canonical", "model": null, - "published_at": "2026-06-14T18:09:51Z", + "published_at": "2026-06-15T00:08:24Z", "repo": "huggingface/transformers", "schema_version": 1, - "snapshot_id": "20260614T180027Z", + "snapshot_id": "20260615T000032Z", "variant": "hybrid" } diff --git a/snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/analysis-report-hybrid.json b/snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/analysis-report-hybrid.json new file mode 100644 index 0000000000000000000000000000000000000000..925c2e9001e7e5bce8960063c5471ed0a6499efd --- /dev/null +++ b/snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/analysis-report-hybrid.json @@ -0,0 +1,2590 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260615T000032Z", + "generated_at": "2026-06-15T00:08:22Z", + "evidence_quality": "full", + "llm_enrichment": true, + "meta_bugs": [ + { + "cluster_id": "cluster-43979-11", + "summary": "Cluster of 1 issues and 10 PRs centered on issue #43979.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 43979, + "canonical_pr_number": 44007, + "issue_numbers": [ + 43979 + ], + "pr_numbers": [ + 43996, + 44007, + 44013, + 44044, + 44066, + 44072, + 44085, + 44129, + 44154, + 44722 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 43996, + "right_pr_number": 44007, + "code_similarity": 0.179, + "size_similarity": 0.576, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.429, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44013, + "code_similarity": 0.122, + "size_similarity": 0.318, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.392, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44044, + "code_similarity": 0.245, + "size_similarity": 0.864, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.479, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44066, + "code_similarity": 0.225, + "size_similarity": 0.818, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.408, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44072, + "code_similarity": 0.14, + "size_similarity": 0.303, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.528, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44085, + "code_similarity": 0.216, + "size_similarity": 0.783, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.398, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44129, + "code_similarity": 0.163, + "size_similarity": 0.643, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.229, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44154, + "code_similarity": 0.153, + "size_similarity": 0.535, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.31, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44722, + "code_similarity": 0.225, + "size_similarity": 0.848, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.368, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44013, + "code_similarity": 0.19, + "size_similarity": 0.553, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.531, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44044, + "code_similarity": 0.186, + "size_similarity": 0.667, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.354, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44066, + "code_similarity": 0.188, + "size_similarity": 0.704, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.315, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44072, + "code_similarity": 0.212, + "size_similarity": 0.526, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.708, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44085, + "code_similarity": 0.195, + "size_similarity": 0.735, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.318, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44129, + "code_similarity": 0.103, + "size_similarity": 0.37, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.191, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44154, + "code_similarity": 0.238, + "size_similarity": 0.93, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.344, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44722, + "code_similarity": 0.178, + "size_similarity": 0.679, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.28, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44044, + "code_similarity": 0.126, + "size_similarity": 0.368, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.351, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44066, + "code_similarity": 0.127, + "size_similarity": 0.389, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.325, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44072, + "code_similarity": 0.29, + "size_similarity": 0.952, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.667, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44085, + "code_similarity": 0.131, + "size_similarity": 0.406, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.329, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44129, + "code_similarity": 0.07, + "size_similarity": 0.205, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.192, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44154, + "code_similarity": 0.177, + "size_similarity": 0.594, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.389, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44722, + "code_similarity": 0.118, + "size_similarity": 0.375, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.287, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44066, + "code_similarity": 0.25, + "size_similarity": 0.947, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.404, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44072, + "code_similarity": 0.136, + "size_similarity": 0.351, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.442, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44085, + "code_similarity": 0.24, + "size_similarity": 0.906, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.394, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44129, + "code_similarity": 0.147, + "size_similarity": 0.555, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.243, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44154, + "code_similarity": 0.17, + "size_similarity": 0.62, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.306, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44722, + "code_similarity": 0.257, + "size_similarity": 0.982, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.402, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44072, + "code_similarity": 0.133, + "size_similarity": 0.37, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.393, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44085, + "code_similarity": 0.763, + "size_similarity": 0.957, + "file_overlap": 0.5, + "area_overlap": 0.825, + "patch_similarity": 0.887, + "shared_filenames": [ + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 41 + ], + [ + 174, + 181 + ], + [ + 250, + 257 + ], + [ + 398, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 466 + ], + [ + 471, + 488 + ], + [ + 518, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 606 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ], + "right_ranges": [ + [ + 33, + 40 + ], + [ + 173, + 180 + ], + [ + 249, + 256 + ], + [ + 397, + 405 + ], + [ + 408, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 465 + ], + [ + 470, + 489 + ], + [ + 517, + 539 + ], + [ + 553, + 559 + ], + [ + 565, + 574 + ], + [ + 579, + 597 + ], + [ + 600, + 611 + ], + [ + 633, + 639 + ], + [ + 645, + 652 + ], + [ + 657, + 673 + ], + [ + 716, + 728 + ], + [ + 737, + 743 + ], + [ + 748, + 771 + ], + [ + 789, + 794 + ] + ] + } + ] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44129, + "code_similarity": 0.145, + "size_similarity": 0.526, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.263, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44154, + "code_similarity": 0.174, + "size_similarity": 0.654, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.286, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44722, + "code_similarity": 0.907, + "size_similarity": 0.964, + "file_overlap": 1.0, + "area_overlap": 0.808, + "patch_similarity": 0.874, + "shared_filenames": [ + "src/transformers/models/codegen/modeling_codegen.py", + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/codegen/modeling_codegen.py", + "left_ranges": [ + [ + 245, + 266 + ] + ], + "right_ranges": [ + [ + 228, + 234 + ] + ] + }, + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 41 + ], + [ + 174, + 181 + ], + [ + 250, + 257 + ], + [ + 398, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 466 + ], + [ + 471, + 488 + ], + [ + 518, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 606 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ], + "right_ranges": [ + [ + 33, + 42 + ], + [ + 175, + 182 + ], + [ + 251, + 258 + ], + [ + 399, + 420 + ], + [ + 426, + 435 + ], + [ + 460, + 467 + ], + [ + 472, + 489 + ], + [ + 519, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 612 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 663 + ], + [ + 666, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ] + } + ] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44085, + "code_similarity": 0.137, + "size_similarity": 0.387, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.398, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44129, + "code_similarity": 0.074, + "size_similarity": 0.195, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.231, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44154, + "code_similarity": 0.175, + "size_similarity": 0.566, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.414, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44722, + "code_similarity": 0.124, + "size_similarity": 0.357, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.347, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44129, + "code_similarity": 0.141, + "size_similarity": 0.503, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.272, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44154, + "code_similarity": 0.18, + "size_similarity": 0.684, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.289, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44722, + "code_similarity": 0.728, + "size_similarity": 0.923, + "file_overlap": 0.5, + "area_overlap": 0.791, + "patch_similarity": 0.78, + "shared_filenames": [ + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 40 + ], + [ + 173, + 180 + ], + [ + 249, + 256 + ], + [ + 397, + 405 + ], + [ + 408, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 465 + ], + [ + 470, + 489 + ], + [ + 517, + 539 + ], + [ + 553, + 559 + ], + [ + 565, + 574 + ], + [ + 579, + 597 + ], + [ + 600, + 611 + ], + [ + 633, + 639 + ], + [ + 645, + 652 + ], + [ + 657, + 673 + ], + [ + 716, + 728 + ], + [ + 737, + 743 + ], + [ + 748, + 771 + ], + [ + 789, + 794 + ] + ], + "right_ranges": [ + [ + 33, + 42 + ], + [ + 175, + 182 + ], + [ + 251, + 258 + ], + [ + 399, + 420 + ], + [ + 426, + 435 + ], + [ + 460, + 467 + ], + [ + 472, + 489 + ], + [ + 519, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 612 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 663 + ], + [ + 666, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ] + } + ] + }, + { + "left_pr_number": 44129, + "right_pr_number": 44154, + "code_similarity": 0.099, + "size_similarity": 0.344, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.199, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44129, + "right_pr_number": 44722, + "code_similarity": 0.146, + "size_similarity": 0.545, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.247, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44154, + "right_pr_number": 44722, + "code_similarity": 0.164, + "size_similarity": 0.631, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.255, + "shared_filenames": [], + "shared_file_areas": [] + } + ] + }, + { + "cluster_id": "cluster-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": 0.797, + "size_similarity": 0.885, + "file_overlap": 0.75, + "area_overlap": 0.831, + "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": [ + [ + 730, + 737 + ] + ] + }, + { + "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": [ + [ + 720, + 726 + ], + [ + 1331, + 1337 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/__init__.py", + "left_ranges": [ + [ + 1, + 29 + ] + ], + "right_ranges": [ + [ + 1, + 29 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/configuration_minicpm3.py", + "left_ranges": [ + [ + 1, + 141 + ] + ], + "right_ranges": [ + [ + 1, + 126 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/modeling_minicpm3.py", + "left_ranges": [ + [ + 1, + 544 + ] + ], + "right_ranges": [ + [ + 1, + 522 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/modular_minicpm3.py", + "left_ranges": [ + [ + 1, + 444 + ] + ], + "right_ranges": [ + [ + 1, + 342 + ] + ] + }, + { + "filename": "tests/models/minicpm3/test_modeling_minicpm3.py", + "left_ranges": [ + [ + 1, + 109 + ] + ], + "right_ranges": [ + [ + 1, + 136 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-45647-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45647.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 45647, + "canonical_pr_number": 45738, + "issue_numbers": [ + 45647 + ], + "pr_numbers": [ + 45737, + 45738 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 45737, + "right_pr_number": 45738, + "code_similarity": 0.526, + "size_similarity": 0.11, + "file_overlap": 0.5, + "area_overlap": 1.0, + "patch_similarity": 0.024, + "shared_filenames": [ + "src/transformers/models/musicgen_melody/modeling_musicgen_melody.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/musicgen_melody/modeling_musicgen_melody.py", + "left_ranges": [ + [ + 491, + 497 + ] + ], + "right_ranges": [ + [ + 491, + 497 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-45987-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45987.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 45987, + "canonical_pr_number": 46173, + "issue_numbers": [ + 45987 + ], + "pr_numbers": [ + 46173, + 46474 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 46173, + "right_pr_number": 46474, + "code_similarity": 1.0, + "size_similarity": 1.0, + "file_overlap": 1.0, + "area_overlap": 1.0, + "patch_similarity": 1.0, + "shared_filenames": [ + "src/transformers/cache_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/cache_utils.py", + "left_ranges": [ + [ + 69, + 75 + ], + [ + 375, + 381 + ] + ], + "right_ranges": [ + [ + 69, + 75 + ], + [ + 375, + 381 + ] + ] + } + ] + } + ] + }, + { + "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": 0.494, + "size_similarity": 0.619, + "file_overlap": 0.5, + "area_overlap": 0.434, + "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": [ + [ + 3595, + 3602 + ], + [ + 3819, + 3853 + ] + ], + "right_ranges": [ + [ + 3525, + 3558 + ], + [ + 3782, + 3788 + ] + ] + }, + { + "filename": "tests/utils/test_testing_utils.py", + "left_ranges": [ + [ + 1, + 86 + ] + ], + "right_ranges": [ + [ + 1, + 114 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-46305-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #46305.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 46305, + "canonical_pr_number": 46546, + "issue_numbers": [ + 46305 + ], + "pr_numbers": [ + 46310, + 46546 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 46310, + "right_pr_number": 46546, + "code_similarity": 0.566, + "size_similarity": 0.444, + "file_overlap": 0.5, + "area_overlap": 0.867, + "patch_similarity": 0.16, + "shared_filenames": [ + "src/transformers/models/moonshine_streaming/modeling_moonshine_streaming.py", + "src/transformers/models/moonshine_streaming/modular_moonshine_streaming.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/moonshine_streaming/modeling_moonshine_streaming.py", + "left_ranges": [ + [ + 356, + 362 + ] + ], + "right_ranges": [ + [ + 355, + 361 + ] + ] + }, + { + "filename": "src/transformers/models/moonshine_streaming/modular_moonshine_streaming.py", + "left_ranges": [ + [ + 269, + 275 + ] + ], + "right_ranges": [ + [ + 269, + 275 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-46614-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #46614.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 46614, + "canonical_pr_number": 46616, + "issue_numbers": [ + 46614 + ], + "pr_numbers": [ + 46616, + 46638 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 46616, + "right_pr_number": 46638, + "code_similarity": 0.83, + "size_similarity": 0.917, + "file_overlap": 1.0, + "area_overlap": 0.828, + "patch_similarity": 0.377, + "shared_filenames": [ + "src/transformers/models/wav2vec2_phoneme/tokenization_wav2vec2_phoneme.py", + "tests/models/wav2vec2_phoneme/test_tokenization_wav2vec2_phoneme.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/wav2vec2_phoneme/tokenization_wav2vec2_phoneme.py", + "left_ranges": [ + [ + 255, + 261 + ] + ], + "right_ranges": [ + [ + 256, + 262 + ] + ] + }, + { + "filename": "tests/models/wav2vec2_phoneme/test_tokenization_wav2vec2_phoneme.py", + "left_ranges": [ + [ + 229, + 249 + ] + ], + "right_ranges": [ + [ + 229, + 246 + ] + ] + } + ] + } + ] + }, + { + "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-46612-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #46612.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 46612, + "canonical_pr_number": 46639, + "issue_numbers": [ + 46612 + ], + "pr_numbers": [ + 46615, + 46639 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 46615, + "right_pr_number": 46639, + "code_similarity": 0.204, + "size_similarity": 0.144, + "file_overlap": 0.5, + "area_overlap": 0.0, + "patch_similarity": 0.169, + "shared_filenames": [ + "src/transformers/generation/utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/generation/utils.py", + "left_ranges": [ + [ + 955, + 989 + ], + [ + 3421, + 3429 + ] + ], + "right_ranges": [ + [ + 3392, + 3404 + ] + ] + } + ] + } + ] + }, + { + "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-45415-2", + "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "status": "open", + "confidence": 0.5, + "canonical_issue_number": null, + "canonical_pr_number": 45425, + "issue_numbers": [], + "pr_numbers": [ + 45415, + 45425 + ], + "evidence_types": [ + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 45415, + "right_pr_number": 45425, + "code_similarity": 0.092, + "size_similarity": 0.171, + "file_overlap": 0.065, + "area_overlap": 0.053, + "patch_similarity": 0.13, + "shared_filenames": [ + "src/transformers/modeling_flash_attention_utils.py", + "src/transformers/modeling_utils.py", + "utils/check_types.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/modeling_flash_attention_utils.py", + "left_ranges": [ + [ + 12, + 18 + ], + [ + 148, + 156 + ] + ], + "right_ranges": [ + [ + 31, + 37 + ], + [ + 148, + 154 + ] + ] + }, + { + "filename": "src/transformers/modeling_utils.py", + "left_ranges": [ + [ + 29, + 35 + ], + [ + 175, + 181 + ], + [ + 187, + 193 + ], + [ + 295, + 301 + ], + [ + 449, + 458 + ], + [ + 840, + 851 + ], + [ + 1032, + 1038 + ], + [ + 1062, + 1068 + ], + [ + 1566, + 1572 + ], + [ + 1822, + 1828 + ], + [ + 1833, + 1841 + ], + [ + 2393, + 2399 + ], + [ + 2990, + 2996 + ], + [ + 3136, + 3143 + ], + [ + 3262, + 3268 + ], + [ + 3291, + 3299 + ], + [ + 3320, + 3326 + ], + [ + 4010, + 4016 + ], + [ + 4177, + 4183 + ], + [ + 4190, + 4196 + ], + [ + 4505, + 4511 + ], + [ + 4623, + 4629 + ], + [ + 4747, + 4753 + ], + [ + 4765, + 4771 + ] + ], + "right_ranges": [ + [ + 29, + 35 + ], + [ + 121, + 127 + ], + [ + 139, + 147 + ], + [ + 181, + 187 + ], + [ + 191, + 214 + ], + [ + 343, + 360 + ], + [ + 374, + 383 + ], + [ + 428, + 438 + ], + [ + 903, + 923 + ], + [ + 962, + 968 + ], + [ + 1010, + 1016 + ], + [ + 1079, + 1095 + ], + [ + 1112, + 1129 + ], + [ + 1626, + 1633 + ], + [ + 1853, + 1862 + ], + [ + 1873, + 1892 + ], + [ + 1895, + 1903 + ], + [ + 1924, + 1930 + ], + [ + 2448, + 2469 + ], + [ + 3059, + 3065 + ], + [ + 3205, + 3212 + ], + [ + 3328, + 3338 + ], + [ + 4095, + 4106 + ], + [ + 4293, + 4301 + ], + [ + 4635, + 4641 + ], + [ + 4757, + 4763 + ], + [ + 4836, + 4849 + ], + [ + 4889, + 4895 + ], + [ + 4907, + 4913 + ] + ] + }, + { + "filename": "utils/check_types.py", + "left_ranges": [ + [ + 4, + 10 + ], + [ + 15, + 28 + ], + [ + 32, + 54 + ] + ], + "right_ranges": [ + [ + 16, + 22 + ], + [ + 26, + 32 + ] + ] + } + ] + } + ] + }, + { + "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-45415-2", + "canonical_pr_number": 45425, + "duplicate_pr_numbers": [ + 45415 + ], + "target_issue_number": null, + "reason": "PRs in cluster-45415-2 are treated as duplicates because they share 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-45987-3", + "canonical_pr_number": 46173, + "duplicate_pr_numbers": [ + 46474 + ], + "target_issue_number": 45987, + "reason": "PRs in cluster-45987-3 are treated as duplicates because they converge on issue #45987 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-46305-3", + "canonical_pr_number": 46546, + "duplicate_pr_numbers": [ + 46310 + ], + "target_issue_number": 46305, + "reason": "PRs in cluster-46305-3 are treated as duplicates because they converge on issue #46305 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-46612-3", + "canonical_pr_number": 46639, + "duplicate_pr_numbers": [ + 46615 + ], + "target_issue_number": 46612, + "reason": "PRs in cluster-46612-3 are treated as duplicates because they converge on issue #46612 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-46614-3", + "canonical_pr_number": 46616, + "duplicate_pr_numbers": [ + 46638 + ], + "target_issue_number": 46614, + "reason": "PRs in cluster-46614-3 are treated as duplicates because they converge on issue #46614 with closing_reference, shared_issue_target evidence." + } + ], + "best_issue": { + "cluster_id": "cluster-41211-3", + "issue_number": 41211, + "reason": "Issue #41211 is the strongest global issue candidate because it is open, belongs to a cluster with 3 artifacts, and carries 6 discussion comments plus 6 inbound references.", + "score": 148.0 + }, + "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/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/analysis-report-hybrid.llm-reviews.json b/snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/analysis-report-hybrid.llm-reviews.json new file mode 100644 index 0000000000000000000000000000000000000000..95ab97685a0a1f42d7cc57c20f9d5ebeb7285fc2 --- /dev/null +++ b/snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/analysis-report-hybrid.llm-reviews.json @@ -0,0 +1,23405 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260615T000032Z", + "generated_at": "2026-06-15T00:08:22Z", + "model": "gpt-5.4-mini?service_tier=flex", + "reviews": [ + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 1, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:24643", + "issue:30277", + "issue:34634", + "issue:34928", + "issue:35545", + "issue:36010", + "issue:36331", + "issue:39290", + "issue:41720", + "issue:41762", + "issue:42915", + "issue:43039", + "issue:43425", + "issue:43541", + "issue:43596", + "issue:43638", + "issue:43716", + "issue:43827", + "issue:43828", + "issue:43866", + "issue:43975", + "issue:44292", + "issue:44322", + "issue:44457", + "issue:44509", + "issue:44512", + "issue:44560", + "issue:44805", + "issue:44918", + "issue:45081", + "issue:45161", + "issue:45237", + "issue:45464", + "issue:45507", + "issue:45685", + "issue:45800", + "issue:45823", + "issue:45859", + "issue:45874", + "issue:45901", + "issue:45902", + "issue:46015", + "issue:46165", + "issue:46294", + "issue:46304", + "issue:46402", + "issue:46468", + "issue:46515" + ], + "soft_pairs": [ + "issue:44322|issue:45464", + "issue:45237|issue:45685", + "issue:43716|issue:45237", + "issue:43828|issue:45237", + "issue:45237|issue:46468", + "issue:45237|issue:45507", + "issue:30277|issue:35545", + "issue:35545|issue:42915", + "issue:35545|issue:41720", + "issue:45901|issue:46015", + "issue:44509|issue:44512", + "issue:35545|issue:36010", + "issue:44918|issue:46165", + "issue:44805|issue:45823", + "issue:43638|issue:44805", + "issue:43638|issue:46165", + "issue:43596|issue:44805", + "issue:44805|issue:46294", + "issue:44805|issue:45161", + "issue:46165|issue:46468", + "issue:44560|issue:46165", + "issue:43596|issue:46165", + "issue:34634|issue:35545", + "issue:30277|issue:43638", + "issue:43596|issue:43638", + "issue:43039|issue:45161", + "issue:43975|issue:44457", + "issue:44292|issue:44457", + "issue:34928|issue:44805", + "issue:24643|issue:30277", + "issue:45859|issue:45902", + "issue:43638|issue:45161", + "issue:34928|issue:36331", + "issue:43596|issue:45161", + "issue:43716|issue:45507", + "issue:44292|issue:45237", + "issue:45874|issue:46294", + "issue:39290|issue:45081", + "issue:45800|issue:46515", + "issue:45823|issue:45874", + "issue:43425|issue:44292", + "issue:34928|issue:41720", + "issue:43828|issue:45161", + "issue:43828|issue:44292", + "issue:43541|issue:43716", + "issue:43716|issue:46468", + "issue:43716|issue:45685", + "issue:44918|issue:45161", + "issue:44292|issue:44918", + "issue:43716|issue:44292", + "issue:45507|issue:45685", + "issue:43039|issue:44918", + "issue:41762|issue:43638", + "issue:43827|issue:44512", + "issue:43828|issue:45685", + "issue:44918|issue:45685", + "issue:43039|issue:43828", + "issue:30277|issue:36331", + "issue:34928|issue:36010", + "issue:43039|issue:44805", + "issue:45237|issue:46015", + "issue:43866|issue:46402", + "issue:45800|issue:46304" + ], + "prepared_review_unit_hash": "8b3d64b39d31cbc89f08552441d602d75d3928dc045fec6631eb1f5fabc0f5ae", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 63, + "serialized_chars": 46524, + "estimated_input_tokens": 11631, + "estimated_eval_tokens": 23518 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous. Only two soft-edge groups look like true duplicates: the DeepSpeed ZeRO-3/BERT index-out-of-bounds reports, and the Torch 2.4.1 compatibility reports. The rest are different model-specific bugs, docs issues, or unrelated error paths.", + "confidence": 0.74, + "canonical_issue_reason": "issue:43638 is the broadest and most representative of the only clear duplicate subcluster: the DeepSpeed ZeRO-3 / BERT index-0-out-of-bounds failure. It subsumes the narrower init-specific report in issue:43596.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43638 is the best issue representative for this cluster because it matches a true duplicate pair and describes the underlying bug more generally than the companion report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44322", + "right": "issue:45464", + "accept": false, + "reason": "Different Qwen3.5 failures: missing config attribute vs streaming chat/completions bug." + }, + { + "left": "issue:45237", + "right": "issue:45685", + "accept": false, + "reason": "Unrelated hardware/model issues: GPT-OSS on AMD vs MPS histogram on Int." + }, + { + "left": "issue:43716", + "right": "issue:45237", + "accept": false, + "reason": "Different model paths and symptoms: image preprocessor dtype mismatch vs AMD GPU runtime failure." + }, + { + "left": "issue:43828", + "right": "issue:45237", + "accept": false, + "reason": "Different bugs and model families; both are dtype/device-related but not the same failure." + }, + { + "left": "issue:45237", + "right": "issue:46468", + "accept": false, + "reason": "Different models and code paths; a broken official example is not the same as the AMD GPU issue." + }, + { + "left": "issue:45237", + "right": "issue:45507", + "accept": false, + "reason": "Different MoE/model bug: AMD GPU runtime vs invalid method call." + }, + { + "left": "issue:30277", + "right": "issue:35545", + "accept": false, + "reason": "DeepSpeed collective mismatch and ONNX export error are unrelated." + }, + { + "left": "issue:35545", + "right": "issue:42915", + "accept": false, + "reason": "Different issues: ONNX export for ModernBERT vs FineGrainedFP8 failure for Qwen3Moe." + }, + { + "left": "issue:35545", + "right": "issue:41720", + "accept": false, + "reason": "Different failure modes and models; export bug vs device-mapping CUDA assert." + }, + { + "left": "issue:45901", + "right": "issue:46015", + "accept": false, + "reason": "One is a task crash, the other is outdated RAG documentation." + }, + { + "left": "issue:44509", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs cleanup for v5 removals, but they target different removed references and are not the same concrete change." + }, + { + "left": "issue:35545", + "right": "issue:36010", + "accept": false, + "reason": "ONNX export bug and GenerationMixin import error are unrelated." + }, + { + "left": "issue:44918", + "right": "issue:46165", + "accept": false, + "reason": "Different training/resume bugs with no shared code-path evidence." + }, + { + "left": "issue:44805", + "right": "issue:45823", + "accept": false, + "reason": "Both mention mismatch-like behavior, but one is a mask/index shape error and the other is a Gemma4 device-map issue." + }, + { + "left": "issue:43638", + "right": "issue:44805", + "accept": false, + "reason": "Different underlying problems: ZeRO-3 BERT index error vs generic mask/tensor shape mismatch." + }, + { + "left": "issue:43638", + "right": "issue:46165", + "accept": false, + "reason": "ZeRO-3 BERT index error is unrelated to resume-training with trust_remote_code." + }, + { + "left": "issue:43596", + "right": "issue:44805", + "accept": false, + "reason": "Both are index/shape errors, but the model/setup and failure path differ." + }, + { + "left": "issue:44805", + "right": "issue:46294", + "accept": false, + "reason": "Different modalities and symptoms: mask shape mismatch vs Qwen3.5 video token/features mismatch." + }, + { + "left": "issue:44805", + "right": "issue:45161", + "accept": false, + "reason": "Generic tensor-shape error vs GPT-OSS tensor-parallel bug; not the same issue." + }, + { + "left": "issue:46165", + "right": "issue:46468", + "accept": false, + "reason": "Different problems: training resume with remote code vs official Nemotron example failure." + }, + { + "left": "issue:44560", + "right": "issue:46165", + "accept": false, + "reason": "Video chat-template StopIteration and training-resume failure are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:46165", + "accept": false, + "reason": "DeepSpeed ZeRO-3 BERT index error is not the same as the resume-training issue." + }, + { + "left": "issue:34634", + "right": "issue:35545", + "accept": false, + "reason": "BarkProcessor voice preset bug 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 an index-out-of-bounds BERT issue." + }, + { + "left": "issue:43596", + "right": "issue:43638", + "accept": true, + "reason": "Same DeepSpeed ZeRO-3 / BERT index-out-of-bounds bug, just described from init vs training/non-pretrained variants." + }, + { + "left": "issue:43039", + "right": "issue:45161", + "accept": false, + "reason": "Liger Kernel cross-entropy call and GPT-OSS TP failure are different code paths." + }, + { + "left": "issue:43975", + "right": "issue:44457", + "accept": false, + "reason": "Different tokenizer/decoding bug vs LoRA merge/save inconsistency." + }, + { + "left": "issue:44292", + "right": "issue:44457", + "accept": false, + "reason": "Different models and symptoms; NVFP4 runtime error is not the LoRA reload discrepancy." + }, + { + "left": "issue:34928", + "right": "issue:44805", + "accept": false, + "reason": "Activation-checkpointing/FSDP size mismatch is distinct from the mask-shape error." + }, + { + "left": "issue:24643", + "right": "issue:30277", + "accept": false, + "reason": "Both are DeepSpeed-related, but the concrete failures are different: weight dimensionality vs collective mismatch." + }, + { + "left": "issue:45859", + "right": "issue:45902", + "accept": false, + "reason": "Same model component, but different failures: CPU offload incompatibility vs uninitialized memory after materialization." + }, + { + "left": "issue:43866", + "right": "issue:46402", + "accept": false, + "reason": "Corrupted checkpoint and missing text encoder weights are not the same bug." + }, + { + "left": "issue:45800", + "right": "issue:46515", + "accept": true, + "reason": "Both report the same Torch 2.4.1 incompatibility; 46515 is the broader restatement of 45800." + }, + { + "left": "issue:45823", + "right": "issue:45874", + "accept": false, + "reason": "Different Gemma4 bugs: device mismatch in forward vs expensive reverse embedding lookup for inputs_embeds." + }, + { + "left": "issue:43425", + "right": "issue:44292", + "accept": false, + "reason": "Generic Torch 2.10 incompatibility and Qwen3-8B-NVFP4 runtime error are unrelated." + }, + { + "left": "issue:43828", + "right": "issue:45685", + "accept": false, + "reason": "Different issues: autocast dtype mismatch vs unimplemented MPS histogram op." + }, + { + "left": "issue:44918", + "right": "issue:45685", + "accept": false, + "reason": "Qwen3.5 SFT-trainer embedding issue and MPS histogram error are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:44918", + "accept": false, + "reason": "Liger Kernel cross-entropy and Qwen3.5 SFT-trainer embedding failures are different bugs." + }, + { + "left": "issue:34928", + "right": "issue:36010", + "accept": false, + "reason": "FSDP activation-checkpointing mismatch and GenerationMixin import failure are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:44805", + "accept": false, + "reason": "Different code paths: Liger Kernel cross-entropy vs tensor mask shape mismatch." + }, + { + "left": "issue:45237", + "right": "issue:46015", + "accept": false, + "reason": "GPU/model runtime bug and outdated RAG docs are unrelated." + }, + { + "left": "issue:45866", + "right": "issue:46402", + "accept": false, + "reason": "No pair provided." + }, + { + "left": "issue:45800", + "right": "issue:46304", + "accept": false, + "reason": "Related theme of version compatibility, but different concrete breakages and code paths." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Accepted. The canonical issue choice and cluster summary are grounded in the packet, and the two accepted soft edges are conservative duplicates rather than overclaims. The remaining soft-edge rejections are appropriately cautious and do not overstate equivalence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 2, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:36010", + "issue:39290", + "issue:41093", + "issue:41720", + "issue:41762", + "issue:43039", + "issue:43296", + "issue:43366", + "issue:43404", + "issue:43425", + "issue:43531", + "issue:43541", + "issue:43572", + "issue:43596", + "issue:43638", + "issue:43792", + "issue:43828", + "issue:43854", + "issue:43866", + "issue:43872", + "issue:43901", + "issue:43975", + "issue:44292", + "issue:44387", + "issue:44512", + "issue:44560", + "issue:44589", + "issue:44805", + "issue:44841", + "issue:44863", + "issue:44918", + "issue:45005", + "issue:45020", + "issue:45070", + "issue:45084", + "issue:45161", + "issue:45237", + "issue:45362", + "issue:45464", + "issue:45685", + "issue:45800", + "issue:45812", + "issue:45823", + "issue:45901", + "issue:45920", + "issue:46015", + "issue:46294", + "issue:46304" + ], + "soft_pairs": [ + "issue:43296|issue:46304", + "issue:41762|issue:43596", + "issue:39290|issue:41720", + "issue:44589|issue:46304", + "issue:43901|issue:44512", + "issue:45237|issue:45901", + "issue:45823|issue:46294", + "issue:43039|issue:44560", + "issue:44841|issue:45823", + "issue:43975|issue:46015", + "issue:44863|issue:46015", + "issue:45020|issue:46304", + "issue:45812|issue:45920", + "issue:43866|issue:44863", + "issue:43872|issue:46304", + "issue:44292|issue:46015", + "issue:39290|issue:43531", + "issue:41093|issue:43638", + "issue:43792|issue:45901", + "issue:43854|issue:43866", + "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:43541|issue:45685", + "issue:43541|issue:43828", + "issue:36010|issue:39290", + "issue:44841|issue:45084", + "issue:43404|issue:43866", + "issue:41093|issue:41762", + "issue:43296|issue:43366", + "issue:45800|issue:45823", + "issue:43572|issue:45070", + "issue:44805|issue:44918", + "issue:44387|issue:45005", + "issue:43541|issue:45161", + "issue:43828|issue:45464", + "issue:43366|issue:45464" + ], + "prepared_review_unit_hash": "20012b59bf88ee980571b99a721133dfd50a8e54fd95865d646b639498b304db", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 42, + "serialized_chars": 40024, + "estimated_input_tokens": 10006, + "estimated_eval_tokens": 20268 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose set of Transformers regression/model-loading issues, with a few near-duplicate tokenizer reports but many unrelated model- or environment-specific bugs. The strongest duplicate-like pair is the AutoTokenizer v5 regression reports; most other soft pairs should stay separate.", + "confidence": 0.86, + "canonical_issue_reason": "issue:45920 is the most representative umbrella report: it describes the broadest concrete regression (AutoTokenizer returning wrong token IDs across multiple model families in v5), which also subsumes the narrower Granite-only report.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45920 is the best global issue candidate because it is the broadest, most user-impacting, and most clearly root-cause-aligned issue in the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43296", + "right": "issue:46304", + "accept": false, + "reason": "Both are loading/runtime failures, but one is PaddleOCR-VL in vLLM and the other is USE_DISTRIBUTED=0 support in v5.9.0; different code paths and failure modes." + }, + { + "left": "issue:41762", + "right": "issue:43596", + "accept": false, + "reason": "Both mention ZeRO-3 index errors, but the affected models and load paths differ (Gemma3 conditional generation vs BertModel)." + }, + { + "left": "issue:39290", + "right": "issue:41720", + "accept": false, + "reason": "Gemma3 config attribute missing in vLLM and Qwen3 auto device-map CUDA assert are unrelated bugs." + }, + { + "left": "issue:44589", + "right": "issue:46304", + "accept": false, + "reason": "Float8 storage lookup failure and USE_DISTRIBUTED=0 breakage are unrelated compatibility issues." + }, + { + "left": "issue:43901", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs-related, but they cover different API/docs regressions and do not look like the same underlying issue." + }, + { + "left": "issue:45237", + "right": "issue:45901", + "accept": false, + "reason": "GPT-OSS on AMD GPUs and a table-question-answering crash are unrelated." + }, + { + "left": "issue:45823", + "right": "issue:46294", + "accept": false, + "reason": "Different model families and different bugs: Gemma4 device placement vs Qwen3.5 video chat-template token mismatch." + }, + { + "left": "issue:43039", + "right": "issue:44560", + "accept": false, + "reason": "Liger Kernel cross-entropy behavior and Qwen3.5 video StopIteration are unrelated." + }, + { + "left": "issue:44841", + "right": "issue:45823", + "accept": false, + "reason": "Processor failure for Voxtral and Gemma4 device mismatch are separate model-specific bugs." + }, + { + "left": "issue:43975", + "right": "issue:46015", + "accept": false, + "reason": "Incorrect detokenization and outdated RAG docs are unrelated." + }, + { + "left": "issue:44863", + "right": "issue:46015", + "accept": false, + "reason": "NemotronH checkpoint loading and docs updates do not share the same underlying bug." + }, + { + "left": "issue:45020", + "right": "issue:46304", + "accept": false, + "reason": "One is a broad remote_code regression report, the other is a specific USE_DISTRIBUTED=0 breakage; too different to merge." + }, + { + "left": "issue:45812", + "right": "issue:45920", + "accept": true, + "reason": "Both report the same concrete AutoTokenizer v5 regression: wrong token IDs across model families. One is Granite-specific, the other is the broader multi-model manifestation." + }, + { + "left": "issue:43866", + "right": "issue:44863", + "accept": false, + "reason": "Corrupted Ovis2 checkpoint and NemotronH checkpoint loading are unrelated." + }, + { + "left": "issue:43872", + "right": "issue:46304", + "accept": false, + "reason": "bitsandbytes constructor incompatibility and distributed-build support regression are different problems." + }, + { + "left": "issue:44292", + "right": "issue:46015", + "accept": false, + "reason": "Qwen-3-8B-NVFP4 runtime error and RAG docs issues are unrelated." + }, + { + "left": "issue:39290", + "right": "issue:43531", + "accept": false, + "reason": "Both mention sliding_window, but they involve different models/configs and different failure modes." + }, + { + "left": "issue:41093", + "right": "issue:43638", + "accept": false, + "reason": "Both are index errors, but one is a mask/tensor shape mismatch and the other is a ZeRO-3 BertModel initialization failure." + }, + { + "left": "issue:43792", + "right": "issue:45901", + "accept": false, + "reason": "Whisper loading failure and table-question-answering crash are unrelated." + }, + { + "left": "issue:43854", + "right": "issue:43866", + "accept": false, + "reason": "GLM-4.7-Flash unit-test loading issue and Ovis2 checkpoint corruption are different model-specific failures." + }, + { + "left": "issue:43425", + "right": "issue:45070", + "accept": false, + "reason": "Torch 2.10 incompatibility and a PretrainedConfig pydantic break are separate compatibility regressions." + }, + { + "left": "issue:43596", + "right": "issue:45800", + "accept": false, + "reason": "ZeRO-3 BertModel index error and torch/transformers version incompatibility are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:45901", + "accept": false, + "reason": "Detokenization regression and table-QA crash do not share the same code path." + }, + { + "left": "issue:43792", + "right": "issue:45070", + "accept": false, + "reason": "Whisper cannot-run report and pydantic config break are unrelated." + }, + { + "left": "issue:45362", + "right": "issue:45464", + "accept": false, + "reason": "Both are Qwen3.5-related, but one is a chat crash and the other is a streaming chat/completions API failure; not clearly the same bug." + }, + { + "left": "issue:44292", + "right": "issue:45901", + "accept": false, + "reason": "Qwen-3-8B-NVFP4 runtime error and table-QA crash are unrelated." + }, + { + "left": "issue:39290", + "right": "issue:45823", + "accept": false, + "reason": "Different models and different failures: Gemma3 config attribute vs Gemma4 device mismatch." + }, + { + "left": "issue:45464", + "right": "issue:45685", + "accept": false, + "reason": "Streaming inference failure and MPS histogram_mps int error are unrelated." + }, + { + "left": "issue:43541", + "right": "issue:45685", + "accept": false, + "reason": "Mixtral grouped_mm tracing and MPS histogram_mps failures are different bugs." + }, + { + "left": "issue:43541", + "right": "issue:43828", + "accept": false, + "reason": "Both are MoE-related, but grouped_mm tracing and autocast dtype mismatch are distinct code paths." + }, + { + "left": "issue:36010", + "right": "issue:39290", + "accept": false, + "reason": "Importing GenerationMixin and a Gemma3 config attribute error are unrelated." + }, + { + "left": "issue:44841", + "right": "issue:45084", + "accept": false, + "reason": "Voxtral processor failure and compilation of non-template nodes are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:43866", + "accept": false, + "reason": "Mistral3 tied-weights bug and Ovis2 checkpoint corruption are unrelated." + }, + { + "left": "issue:41093", + "right": "issue:41762", + "accept": false, + "reason": "Both are index errors, but they arise in different contexts and model-loading paths; not the same underlying bug." + }, + { + "left": "issue:43296", + "right": "issue:43366", + "accept": false, + "reason": "PaddleOCR-VL load failure and GGUF gpt-oss support are different architecture/runtime issues." + }, + { + "left": "issue:45800", + "right": "issue:45823", + "accept": false, + "reason": "Torch/Transformers version incompatibility and Gemma4 device mismatch are different regressions." + }, + { + "left": "issue:43572", + "right": "issue:45070", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig and PretrainedConfig field breakage are different config-schema problems." + }, + { + "left": "issue:44805", + "right": "issue:44918", + "accept": false, + "reason": "Mask shape mismatch during generation and TRL embedding-unpacking failure are different code paths." + }, + { + "left": "issue:44387", + "right": "issue:45005", + "accept": false, + "reason": "Int4 quantization OOM and tied-weights translation-model issues are unrelated." + }, + { + "left": "issue:43541", + "right": "issue:45161", + "accept": false, + "reason": "Both involve MoE/parallelism, but grouped_mm tracing and TP support on GPT-OSS are distinct issues." + }, + { + "left": "issue:43828", + "right": "issue:45464", + "accept": false, + "reason": "Phi-tiny-MoE autocast dtype mismatch and Qwen3.5 streaming API failure are unrelated." + }, + { + "left": "issue:43366", + "right": "issue:45464", + "accept": false, + "reason": "GGUF gpt-oss support and Qwen3.5 streaming API failure are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative: it correctly treats most pairs as unrelated, and the one accepted pair (45812/45920) is well supported by both titles describing the same AutoTokenizer v5 wrong-token-ID regression, with 45920 clearly the broader umbrella report. No obvious overstatement beyond 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": 21, + "review_unit_index": 3, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:36331", + "issue:39186", + "issue:39692", + "issue:40990", + "issue:41093", + "issue:41762", + "issue:42915", + "issue:43054", + "issue:43404", + "issue:43531", + "issue:43541", + "issue:43596", + "issue:43644", + "issue:43645", + "issue:43716", + "issue:43828", + "issue:43873", + "issue:43950", + "issue:43975", + "issue:43994", + "issue:44291", + "issue:44292", + "issue:44360", + "issue:44479", + "issue:44492", + "issue:44512", + "issue:44530", + "issue:44534", + "issue:44589", + "issue:44863", + "issue:44918", + "issue:44928", + "issue:44960", + "issue:45005", + "issue:45020", + "issue:45070", + "issue:45183", + "issue:45399", + "issue:45507", + "issue:45632", + "issue:45698", + "issue:45759", + "issue:45800", + "issue:45902", + "issue:46294", + "issue:46463", + "issue:46489", + "issue:46491" + ], + "soft_pairs": [ + "issue:44291|issue:45070", + "issue:43531|issue:45070", + "issue:43716|issue:46294", + "issue:45020|issue:45800", + "issue:43541|issue:44589", + "issue:44479|issue:46294", + "issue:41093|issue:43596", + "issue:43054|issue:43994", + "issue:45759|issue:45902", + "issue:39186|issue:44918", + "issue:43716|issue:44918", + "issue:40990|issue:44960", + "issue:43645|issue:45399", + "issue:45632|issue:45698", + "issue:43404|issue:45005", + "issue:42915|issue:45399", + "issue:44530|issue:45005", + "issue:43873|issue:45005", + "issue:44360|issue:44512", + "issue:44863|issue:45005", + "issue:45183|issue:46463", + "issue:44492|issue:44512", + "issue:44918|issue:45507", + "issue:46489|issue:46491", + "issue:43975|issue:44292", + "issue:44292|issue:45507", + "issue:43950|issue:44534", + "issue:39692|issue:43054", + "issue:43644|issue:43950", + "issue:36331|issue:41762", + "issue:43828|issue:44928" + ], + "prepared_review_unit_hash": "85d364253373430f4945048ef27df94b1481e5932545a4d01b726c3712da5765", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37035, + "estimated_input_tokens": 9259, + "estimated_eval_tokens": 18774 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a loose similarity cluster of unrelated Transformers regressions. The only clear duplicate subgroups are the non-persistent buffer corruption reports, with a couple of plausible same-bug pairs in SigLIP2/video-input/validation error-message issues.", + "confidence": 0.59, + "canonical_issue_reason": "issue:43644 is the clearest and earliest representative of the repeated non-persistent-buffer regression; its wording matches the later rephrasings in 43950 and 44534 most directly.", + "canonical_pr_reason": null, + "best_issue_reason": "Among these mostly heterogeneous issues, 43644 is the best representative because it states a concrete, recurring bug plainly and is the broadest/earliest formulation of that sub-bug.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44291", + "right": "issue:45070", + "accept": false, + "reason": "Different regressions: unexpected `_is_hf_initialized` arg during empty-weights loading vs a Pydantic/PretrainedConfig field breakage." + }, + { + "left": "issue:43531", + "right": "issue:45070", + "accept": false, + "reason": "`sliding_window` handling in Qwen3-MoE is a different code path from the config/Pydantic regression in 45070." + }, + { + "left": "issue:43716", + "right": "issue:46294", + "accept": false, + "reason": "Both involve multimodal models, but one is an image preprocessor/dtype mismatch and the other is a video chat-template token/feature mismatch." + }, + { + "left": "issue:45020", + "right": "issue:45800", + "accept": false, + "reason": "Remote-code compatibility breakage and torch-version incompatibility are broad but distinct problems." + }, + { + "left": "issue:43541", + "right": "issue:44589", + "accept": false, + "reason": "Grouped-mm tracing failure in Mixtral is unrelated to the missing Float8 storage lookup error." + }, + { + "left": "issue:44479", + "right": "issue:46294", + "accept": true, + "reason": "Both describe the same Qwen video-input regression family: video features/tokens or template processing are mismatched." + }, + { + "left": "issue:41093", + "right": "issue:43596", + "accept": false, + "reason": "Both are IndexErrors under different setups, but the mask-shape mismatch and ZeRO-3 empty-tensor indexing are not the same bug." + }, + { + "left": "issue:43054", + "right": "issue:43994", + "accept": true, + "reason": "Both report SigLIP2 producing bad/nonsensical outputs, pointing to the same underlying model/processing regression." + }, + { + "left": "issue:45759", + "right": "issue:45902", + "accept": false, + "reason": "Different failures: composite-config unwrapping vs uninitialized rotary frequency after meta-to-empty materialization." + }, + { + "left": "issue:39186", + "right": "issue:44918", + "accept": false, + "reason": "FSDP weight-shape failure and TRL SFT embedding unpacking failure are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:44918", + "accept": false, + "reason": "Image dtype mismatch and TRL unpacking failure do not share the same underlying bug." + }, + { + "left": "issue:40990", + "right": "issue:44960", + "accept": false, + "reason": "High perplexity on GPT-OSS/WikiText-2 and a GLM5 issue are unrelated model-specific reports." + }, + { + "left": "issue:43645", + "right": "issue:45399", + "accept": false, + "reason": "Custom-model initialization in Jupyter and flash-attn2 fallback gating are different issues." + }, + { + "left": "issue:45632", + "right": "issue:45698", + "accept": false, + "reason": "Cache-path collision for trust_remote_code and wrong custom-module loading after save_pretrained are related but distinct bugs." + }, + { + "left": "issue:43404", + "right": "issue:45005", + "accept": false, + "reason": "lm_head weight tying in Mistral3 is a different model-specific bug from the broader translation-model tied-weights regression." + }, + { + "left": "issue:42915", + "right": "issue:45399", + "accept": false, + "reason": "Qwen3Moe FP8 failure is unrelated to the flash-attn2 fallback checks." + }, + { + "left": "issue:44530", + "right": "issue:45005", + "accept": false, + "reason": "PagedAttentionCache invalid group type and tied-weights handling are different code paths." + }, + { + "left": "issue:43873", + "right": "issue:45005", + "accept": false, + "reason": "Quantization/offloading behavior is not the same as the tied-weights regression." + }, + { + "left": "issue:44360", + "right": "issue:44512", + "accept": false, + "reason": "A DSA indexer math bug and a docs typo about the removed run command are unrelated." + }, + { + "left": "issue:44863", + "right": "issue:45005", + "accept": false, + "reason": "NemotronH checkpoint loading is unrelated to the translation tied-weights issue." + }, + { + "left": "issue:45183", + "right": "issue:46463", + "accept": true, + "reason": "46463 looks like a specific T5 manifestation of the broader input/decoder validation error-message bug described in 45183." + }, + { + "left": "issue:44492", + "right": "issue:44512", + "accept": false, + "reason": "Cache-strategy typo and outdated CLI docs mention are unrelated documentation issues." + }, + { + "left": "issue:44918", + "right": "issue:45507", + "accept": false, + "reason": "TRL SFT embedding unpacking and GraniteMoEHybrid invalid-method calls are different failures." + }, + { + "left": "issue:46489", + "right": "issue:46491", + "accept": false, + "reason": "Both are tokenizer regressions, but the models and symptoms differ enough to look like separate bugs." + }, + { + "left": "issue:43975", + "right": "issue:44292", + "accept": false, + "reason": "Incorrect detokenization and NVFP4 runtime failure are unrelated." + }, + { + "left": "issue:44292", + "right": "issue:45507", + "accept": false, + "reason": "NVFP4 execution failure and GraniteMoEHybrid invalid method are different bugs." + }, + { + "left": "issue:43950", + "right": "issue:44534", + "accept": true, + "reason": "Both describe the same v5 regression where non-persistent buffers get corrupted/junk values during loading." + }, + { + "left": "issue:39692", + "right": "issue:43054", + "accept": false, + "reason": "SigLIP2 doc-example errors/quantization issues are not the same as the text-embedding quality regression." + }, + { + "left": "issue:43644", + "right": "issue:43950", + "accept": true, + "reason": "Same underlying non-persistent-buffer corruption regression, phrased slightly differently." + }, + { + "left": "issue:36331", + "right": "issue:41762", + "accept": false, + "reason": "Unexpected compute_loss keyword and ZeRO-3 empty-index loading error are unrelated." + }, + { + "left": "issue:43828", + "right": "issue:44928", + "accept": false, + "reason": "Autocast dtype mismatch and RLHF NaN/SDPA fallback issues are different failures in different setups." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative: it correctly frames the set as mostly unrelated regressions, and the accepted soft edges are all reasonably supported by the issue titles/descriptions. The non-persistent-buffer cluster is clearly real, and the Qwen video, SigLIP2, and XOR validation pairs are plausible same-bug groupings without overclaiming. The rejected pairs also look appropriately cautious." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 4, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:24643", + "issue:28530", + "issue:38175", + "issue:41553", + "issue:41762", + "issue:42491", + "issue:42915", + "issue:42947", + "issue:43054", + "issue:43493", + "issue:43643", + "issue:43742", + "issue:43782", + "issue:43824", + "issue:43828", + "issue:43856", + "issue:43866", + "issue:43872", + "issue:43881", + "issue:43883", + "issue:43950", + "issue:43975", + "issue:44368", + "issue:44387", + "issue:44403", + "issue:44451", + "issue:44488", + "issue:44589", + "issue:44661", + "issue:44910", + "issue:44960", + "issue:44991", + "issue:45020", + "issue:45081", + "issue:45356", + "issue:45399", + "issue:45488", + "issue:45520", + "issue:45663", + "issue:45685", + "issue:45701", + "issue:45715", + "issue:45823", + "issue:45859", + "issue:45902", + "issue:45920", + "issue:46315", + "issue:46489" + ], + "soft_pairs": [ + "issue:43872|issue:44589", + "issue:45920|issue:46489", + "issue:43054|issue:45701", + "issue:43866|issue:44403", + "issue:43950|issue:45902", + "issue:38175|issue:43054", + "issue:41553|issue:43883", + "issue:44910|issue:45859", + "issue:42947|issue:44387", + "issue:45663|issue:45902", + "issue:43054|issue:43493", + "issue:41553|issue:43881", + "issue:44403|issue:44451", + "issue:24643|issue:28530", + "issue:43975|issue:45520", + "issue:44403|issue:44488", + "issue:42947|issue:43856", + "issue:43643|issue:45020", + "issue:41762|issue:42915", + "issue:45685|issue:45823", + "issue:44991|issue:45399", + "issue:43824|issue:43872", + "issue:43782|issue:44960", + "issue:43881|issue:43883", + "issue:45020|issue:45823", + "issue:44661|issue:45020", + "issue:45081|issue:45399", + "issue:43828|issue:45823", + "issue:42491|issue:44368", + "issue:43742|issue:45399", + "issue:45020|issue:45356", + "issue:45715|issue:46315", + "issue:45488|issue:46489" + ], + "prepared_review_unit_hash": "d3a92463d9430023cbf942323fbaa3ef74b7ed1f2b04c18edcd2ab4ef8268c33", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37917, + "estimated_input_tokens": 9480, + "estimated_eval_tokens": 19216 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a mix of unrelated v5 model-loading/tokenization regression reports; only a couple of soft edges look like true duplicates, mainly around SigLIP2 output mismatch and remote_code config loading.", + "confidence": 0.71, + "canonical_issue_reason": "issue:45920 is the most representative umbrella here: it is open, concrete, and captures the recurring multi-model tokenizer-regression pattern that shows up across the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45920 is the strongest single issue to keep as the cluster representative because it is broad enough to cover the dominant tokenizer symptom while still being specific and actionable.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43872", + "right": "issue:44589", + "accept": false, + "reason": "Both are load-time TypeErrors, but they come from different subsystems and different root causes." + }, + { + "left": "issue:45920", + "right": "issue:46489", + "accept": false, + "reason": "Both involve tokenizer regressions, but they affect different model families and appear to come from different code paths." + }, + { + "left": "issue:43054", + "right": "issue:45701", + "accept": false, + "reason": "One is a SigLIP2 embedding-quality issue; the other is a generic tokenizer-version regression, so the underlying bug is not the same." + }, + { + "left": "issue:43866", + "right": "issue:44403", + "accept": false, + "reason": "Checkpoint corruption and noisy load logging are unrelated problems." + }, + { + "left": "issue:43950", + "right": "issue:45902", + "accept": false, + "reason": "Both mention loading/materialization, but one is a generic non-persistent-buffer regression and the other is a model-specific uninitialized-memory bug." + }, + { + "left": "issue:38175", + "right": "issue:43054", + "accept": false, + "reason": "Same model family, but zero probabilities and degraded text embeddings are not clearly the same defect." + }, + { + "left": "issue:41553", + "right": "issue:43883", + "accept": false, + "reason": "Bad AutoTokenizer error messaging and a missing all_tied_weights_keys attribute are unrelated." + }, + { + "left": "issue:44910", + "right": "issue:45859", + "accept": false, + "reason": "Both are Qwen3.5-related, but one is a flash-attn packed-sequence crash and the other is a CPU-offload rotary-embedding incompatibility." + }, + { + "left": "issue:42947", + "right": "issue:44387", + "accept": false, + "reason": "These are different memory regressions with different mechanisms and code paths." + }, + { + "left": "issue:45663", + "right": "issue:45902", + "accept": false, + "reason": "Different models and different failure sites; these are not the same bug." + }, + { + "left": "issue:43054", + "right": "issue:43493", + "accept": true, + "reason": "Both point to the same SigLIP2 HF-vs-JAX implementation mismatch, with degraded outputs as the shared symptom." + }, + { + "left": "issue:41553", + "right": "issue:43881", + "accept": false, + "reason": "Different model-loading failures with no clear shared root cause." + }, + { + "left": "issue:44403", + "right": "issue:44451", + "accept": false, + "reason": "Generic load-time noise and a specific model-load failure are not the same issue." + }, + { + "left": "issue:24643", + "right": "issue:28530", + "accept": false, + "reason": "Training DeepSpeed weight-shape failure and early-stopping metric detection are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:45520", + "accept": false, + "reason": "Tokenizer detokenization corruption and a Python 3.13 flash_attn import KeyError are unrelated." + }, + { + "left": "issue:44403", + "right": "issue:44488", + "accept": false, + "reason": "These are separate model-load complaints; no shared concrete defect is evident." + }, + { + "left": "issue:42947", + "right": "issue:43856", + "accept": false, + "reason": "Gradient-checkpointing inefficiency and Qwen3 MoE memory usage are different performance problems." + }, + { + "left": "issue:43643", + "right": "issue:45020", + "accept": true, + "reason": "Both are remote_code/AutoConfig loading regressions; 43643 looks like a specific incomplete-config symptom of the broader breakage." + }, + { + "left": "issue:41762", + "right": "issue:42915", + "accept": false, + "reason": "Different model families and different DeepSpeed/load failures." + }, + { + "left": "issue:45685", + "right": "issue:45823", + "accept": false, + "reason": "An MPS histogram kernel limitation and a Gemma4 device-mismatch forward bug are unrelated." + }, + { + "left": "issue:44991", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer loading failure and flash-attn fallback blocking are separate issues." + }, + { + "left": "issue:43824", + "right": "issue:43872", + "accept": false, + "reason": "ImportError for a model class and bitsandbytes parameter incompatibility are unrelated." + }, + { + "left": "issue:43782", + "right": "issue:44960", + "accept": false, + "reason": "Qwen3VL weight_only loading and GLM5 are different problems." + }, + { + "left": "issue:43881", + "right": "issue:43883", + "accept": false, + "reason": "Generic glm-4v-9b load failure and a Molmo tied-weights attribute error are different defects." + }, + { + "left": "issue:45020", + "right": "issue:45823", + "accept": false, + "reason": "Remote_code loading regressions and Gemma4 device placement are unrelated." + }, + { + "left": "issue:44661", + "right": "issue:45020", + "accept": false, + "reason": "Tokenizer-mapping model generation and remote_code config loading are different code paths." + }, + { + "left": "issue:45081", + "right": "issue:45399", + "accept": false, + "reason": "Mistral regex patching and flash-attn2 fallback gating are unrelated." + }, + { + "left": "issue:43828", + "right": "issue:45823", + "accept": false, + "reason": "Autocast dtype mismatch and Gemma4 PLE device mismatch are different bugs." + }, + { + "left": "issue:42491", + "right": "issue:44368", + "accept": false, + "reason": "LoRA compatibility across HF versions and a tie_word_embeddings warning are not the same issue." + }, + { + "left": "issue:43742", + "right": "issue:45399", + "accept": false, + "reason": "MobileLLM key error and flash-attn fallback checks are unrelated." + }, + { + "left": "issue:45020", + "right": "issue:45356", + "accept": false, + "reason": "Remote_code loading breakage and Kimi-K2.5 tokenizer regression are different symptoms and likely different causes." + }, + { + "left": "issue:45715", + "right": "issue:46315", + "accept": false, + "reason": "Both are tokenizer performance regressions, but they hit different APIs and code paths." + }, + { + "left": "issue:45488", + "right": "issue:46489", + "accept": false, + "reason": "Both are DeepSeek tokenizer regressions, but one is a pre-tokenizer override bug and the other is a separate gap in a prior fix." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The representative issue choice of issue:45920 is defensible given its broad tokenizer-regression scope, and the two accepted soft edges (43054\u219443493 and 43643\u219445020) are plausible duplicates without overclaiming. The rejected pairs are phrased cautiously enough, and there are no PR-pair mergeability problems here." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 5, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:24643", + "issue:30064", + "issue:36331", + "issue:41093", + "issue:41553", + "issue:42915", + "issue:42947", + "issue:43257", + "issue:43366", + "issue:43404", + "issue:43531", + "issue:43596", + "issue:43638", + "issue:43716", + "issue:43792", + "issue:43824", + "issue:43827", + "issue:43828", + "issue:43854", + "issue:43872", + "issue:43940", + "issue:43950", + "issue:43957", + "issue:44155", + "issue:44220", + "issue:44291", + "issue:44479", + "issue:44488", + "issue:44509", + "issue:44530", + "issue:44560", + "issue:44589", + "issue:44805", + "issue:44863", + "issue:44928", + "issue:44991", + "issue:45305", + "issue:45313", + "issue:45325", + "issue:45584", + "issue:45588", + "issue:45663", + "issue:45823", + "issue:45834", + "issue:45902", + "issue:45923", + "issue:46402", + "issue:46515" + ], + "soft_pairs": [ + "issue:43827|issue:44509", + "issue:43404|issue:43950", + "issue:43872|issue:44291", + "issue:42915|issue:43872", + "issue:43792|issue:44220", + "issue:43257|issue:45305", + "issue:36331|issue:41093", + "issue:42947|issue:45923", + "issue:41553|issue:44991", + "issue:43940|issue:45313", + "issue:43854|issue:44863", + "issue:44928|issue:45588", + "issue:41093|issue:44805", + "issue:44479|issue:44560", + "issue:41553|issue:44488", + "issue:24643|issue:30064", + "issue:44863|issue:46402", + "issue:44530|issue:44863", + "issue:43596|issue:46515", + "issue:43366|issue:43828", + "issue:45325|issue:45902", + "issue:43638|issue:46515", + "issue:43366|issue:43531", + "issue:45663|issue:45823", + "issue:43716|issue:44560", + "issue:44155|issue:45584", + "issue:43366|issue:44589", + "issue:43950|issue:45834", + "issue:43824|issue:43957" + ], + "prepared_review_unit_hash": "ebfb2a263ad19b24fda8a3de2a37f4f00965df8fcaaf6b6061b3b82fe24d3024", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36728, + "estimated_input_tokens": 9182, + "estimated_eval_tokens": 18620 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly heterogeneous issue cluster with only a couple of strong duplicate-like pairs. I\u2019d keep the docs-v5 pipeline cleanup issue as the canonical issue; no PRs are present.", + "confidence": 0.47, + "canonical_issue_reason": "issue:43827 is the oldest and clearest representative of the only exact docs-dup pair, with a concise problem statement about v5 pipeline removals.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43827 is the best issue anchor here: it is well-scoped, clearly reproducible as a docs mismatch, and matches issue:44509 closely enough to serve as the canonical duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43827", + "right": "issue:44509", + "accept": true, + "reason": "Both are v5 docs leftovers referencing removed text-generation/summarization/translation pipelines; same documentation cleanup." + }, + { + "left": "issue:43404", + "right": "issue:43950", + "accept": false, + "reason": "Mistral3 lm_head tying and non-persistent buffer corruption are different loading/state bugs." + }, + { + "left": "issue:43872", + "right": "issue:44291", + "accept": true, + "reason": "Both describe the same unexpected _is_hf_initialized argument compatibility break during model initialization/loading." + }, + { + "left": "issue:42915", + "right": "issue:43872", + "accept": false, + "reason": "FineGrainedFP8Config on Qwen3Moe and the _is_hf_initialized/bitsandbytes issue are different failures." + }, + { + "left": "issue:43792", + "right": "issue:44220", + "accept": false, + "reason": "Whisper load/run failure and _torch_extract_fbank_features are related to audio, but not the same concrete bug." + }, + { + "left": "issue:43257", + "right": "issue:45305", + "accept": false, + "reason": "Both involve DeepSpeed, but one is weight loading and the other is gradient averaging; different code paths." + }, + { + "left": "issue:36331", + "right": "issue:41093", + "accept": false, + "reason": "CustomTrainer compute_loss signature mismatch is unrelated to the mask/tensor shape IndexError." + }, + { + "left": "issue:42947", + "right": "issue:45923", + "accept": false, + "reason": "Different gradient-checkpointing issues: LoRA ineffectiveness vs a missing supports_gradient_checkpointing flag." + }, + { + "left": "issue:41553", + "right": "issue:44991", + "accept": false, + "reason": "Different tokenizer-loading regressions for different models; no shared concrete bug is evident." + }, + { + "left": "issue:43940", + "right": "issue:45313", + "accept": false, + "reason": "Both are ZeRO-3 load failures, but for different Qwen generations and likely different model-specific paths." + }, + { + "left": "issue:43854", + "right": "issue:44863", + "accept": false, + "reason": "GLM-4.7-Flash loading in tests and NemotronH checkpoint loading are unrelated model-specific bugs." + }, + { + "left": "issue:44928", + "right": "issue:45588", + "accept": false, + "reason": "Qwen3.5 RLHF NaNs from 3D position_ids and flash_attention s_aux=None crashes are different failures." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": true, + "reason": "Same shape-mismatch IndexError pattern on mask vs indexed tensor; likely the same off-by-one bug." + }, + { + "left": "issue:44479", + "right": "issue:44560", + "accept": false, + "reason": "Both are Qwen video issues, but one is a broad regression report and the other is a specific StopIteration failure; not clearly the same bug." + }, + { + "left": "issue:41553", + "right": "issue:44488", + "accept": false, + "reason": "Voxtral tokenizer error-message bug and EMBEDDIA/est-roberta tokenizer loading failure are different issues." + }, + { + "left": "issue:44863", + "right": "issue:46402", + "accept": false, + "reason": "NemotronH checkpoint loading and Sam3Model text encoder weights not loading are different model/load paths." + }, + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash and NemotronH checkpoint loading are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:46515", + "accept": false, + "reason": "ZeRO-3/BertModel index error is not the same as a general torch 2.4.1 incompatibility." + }, + { + "left": "issue:43366", + "right": "issue:43828", + "accept": false, + "reason": "GGUF gpt-oss support request and Phi-tiny-MoE autocast dtype mismatch are unrelated." + }, + { + "left": "issue:45325", + "right": "issue:45902", + "accept": false, + "reason": "Qwen2.5-VL rope scaling bug and Qwen3.5-MoE inv_freq materialization bug are different code paths." + }, + { + "left": "issue:43638", + "right": "issue:46515", + "accept": false, + "reason": "Non-pretrained Bert zero3 index error and torch-version incompatibility are different problems." + }, + { + "left": "issue:43366", + "right": "issue:43531", + "accept": false, + "reason": "GGUF gpt-oss support and Qwen3-MoE sliding_window behavior are unrelated." + }, + { + "left": "issue:45663", + "right": "issue:45823", + "accept": false, + "reason": "Gemma4 FSDP2 shared_kv_states KeyError and Gemma4 PLE device mismatch are distinct bugs." + }, + { + "left": "issue:43716", + "right": "issue:44560", + "accept": false, + "reason": "Mistral-3 image dtype mismatch and Qwen video StopIteration are different model/input paths." + }, + { + "left": "issue:44155", + "right": "issue:45584", + "accept": false, + "reason": "AudioFlamingo3 track leakage and Whisper empty-transcription failure are unrelated audio bugs." + }, + { + "left": "issue:43366", + "right": "issue:44589", + "accept": false, + "reason": "GGUF architecture support and missing Float8 storage type are different compatibility issues." + }, + { + "left": "issue:43824", + "right": "issue:43957", + "accept": false, + "reason": "Missing Qwen2_5_VL export and meta-device loading regressions are different failures." + }, + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "Qwen3.5 PagedAttentionCache crash and NemotronH checkpoint loading are unrelated." + }, + { + "left": "issue:43854", + "right": "issue:45923", + "accept": false, + "reason": "GLM-4.7-Flash test loading and missing gradient checkpointing support are unrelated." + }, + { + "left": "issue:43950", + "right": "issue:45834", + "accept": false, + "reason": "Non-persistent buffer corruption in from_pretrained and Kosmos2.5 long OCR index errors are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: it correctly identifies the cluster as mostly heterogeneous, notes there are no PRs, and keeps the docs v5 pipeline pair as the canonical issue. The soft-edge accepts are conservative enough and the false pairs are mostly well-justified." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 6, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:36010", + "issue:38175", + "issue:39692", + "issue:41720", + "issue:42491", + "issue:42617", + "issue:42915", + "issue:43317", + "issue:43366", + "issue:43550", + "issue:43596", + "issue:43632", + "issue:43638", + "issue:43644", + "issue:43645", + "issue:43646", + "issue:43720", + "issue:43957", + "issue:43994", + "issue:44451", + "issue:44488", + "issue:44534", + "issue:44560", + "issue:44661", + "issue:44683", + "issue:44756", + "issue:44805", + "issue:44877", + "issue:44910", + "issue:44918", + "issue:44960", + "issue:44977", + "issue:44991", + "issue:45030", + "issue:45276", + "issue:45313", + "issue:45357", + "issue:45507", + "issue:45685", + "issue:45812", + "issue:45823", + "issue:45834", + "issue:46459", + "issue:46468", + "issue:46489", + "issue:46512", + "issue:46515" + ], + "soft_pairs": [ + "issue:44560|issue:45685", + "issue:42915|issue:43957", + "issue:44683|issue:46515", + "issue:43645|issue:43646", + "issue:43550|issue:44910", + "issue:44560|issue:44918", + "issue:42617|issue:43366", + "issue:36010|issue:42915", + "issue:45812|issue:46489", + "issue:45823|issue:46515", + "issue:44661|issue:45823", + "issue:44560|issue:44805", + "issue:38175|issue:43994", + "issue:43632|issue:45834", + "issue:44877|issue:45030", + "issue:43638|issue:44661", + "issue:43596|issue:44661", + "issue:44661|issue:44805", + "issue:43317|issue:44756", + "issue:38175|issue:39692", + "issue:43644|issue:44534", + "issue:42491|issue:43720", + "issue:41720|issue:42915", + "issue:42915|issue:44756", + "issue:42491|issue:45276", + "issue:46459|issue:46512", + "issue:44451|issue:44488", + "issue:44960|issue:44977", + "issue:45313|issue:45357", + "issue:45507|issue:46468", + "issue:44488|issue:44991", + "issue:43644|issue:43645" + ], + "prepared_review_unit_hash": "30bf4b7061086c6432fa081174d0cc1ff8208008a3414237efe4686c4ac147b6", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36460, + "estimated_input_tokens": 9115, + "estimated_eval_tokens": 18486 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a noisy cluster of transformer regression reports across unrelated models/code paths. I only found a few credible duplicates: the non-persistent-buffer junk regression, the custom-model initialization regression, and the SigLIP2 bad-output reports.", + "confidence": 0.81, + "canonical_issue_reason": "issue:43644 is the clearest anchor: it is the earlier, concrete report for the non-persistent-buffer corruption regression, and issue:44534 reads like the same bug with nearly identical symptom wording.", + "canonical_pr_reason": null, + "best_issue_reason": "43644 is the strongest representative issue because it is specific, broad enough to describe the underlying regression, and has the cleanest duplicate counterpart.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44560", + "right": "issue:45685", + "accept": false, + "reason": "Different models and failure modes; same broad ML area only." + }, + { + "left": "issue:42915", + "right": "issue:43957", + "accept": false, + "reason": "Both involve loading/runtime failures, but the causes and affected paths are unrelated." + }, + { + "left": "issue:44683", + "right": "issue:46515", + "accept": false, + "reason": "Different torch-version compatibility problems on different code paths." + }, + { + "left": "issue:43645", + "right": "issue:43646", + "accept": true, + "reason": "Both describe the same Transformers v5 custom-model initialization regression; 43645 is the notebook-specific manifestation of 43646's broader bug." + }, + { + "left": "issue:43550", + "right": "issue:44910", + "accept": false, + "reason": "Same backend family, but different models and distinct failure descriptions; not enough to treat as one bug." + }, + { + "left": "issue:44560", + "right": "issue:44918", + "accept": false, + "reason": "Different Qwen3-related failures with different symptoms and code paths." + }, + { + "left": "issue:42617", + "right": "issue:43366", + "accept": false, + "reason": "Unrelated model/run failures; no shared underlying bug." + }, + { + "left": "issue:36010", + "right": "issue:42915", + "accept": false, + "reason": "ImportError for GenerationMixin versus Qwen3Moe FP8 loading failure are unrelated." + }, + { + "left": "issue:45812", + "right": "issue:46489", + "accept": false, + "reason": "Both concern tokenization regressions, but they affect different model families and different failure modes." + }, + { + "left": "issue:45823", + "right": "issue:46515", + "accept": false, + "reason": "Device-map forward mismatch and torch 2.4.1 incompatibility are not the same bug." + }, + { + "left": "issue:44661", + "right": "issue:45823", + "accept": false, + "reason": "Tokenizer-mapping bug versus Gemma4 device mismatch are unrelated." + }, + { + "left": "issue:44560", + "right": "issue:44805", + "accept": false, + "reason": "Different Qwen3-vl / masking failures with no shared root cause." + }, + { + "left": "issue:38175", + "right": "issue:43994", + "accept": true, + "reason": "Same specific SigLIP2 checkpoint and same bad inference behavior; 43994 is the broader report of the same underlying model-output regression." + }, + { + "left": "issue:43632", + "right": "issue:45834", + "accept": false, + "reason": "Different regressions: HF init flag handling versus long-OCR index error." + }, + { + "left": "issue:44877", + "right": "issue:45030", + "accept": false, + "reason": "Both are config-validation issues, but they concern different models and likely different validation paths." + }, + { + "left": "issue:43638", + "right": "issue:44661", + "accept": false, + "reason": "DeepSpeed ZeRO-3/Bert initialization issue versus tokenizer-mapping failure are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:44661", + "accept": false, + "reason": "Different Bert/ZeRO-3 crash versus add-new-model-like tokenizer mapping bug." + }, + { + "left": "issue:44661", + "right": "issue:44805", + "accept": false, + "reason": "Tokenizer-mapping logic and mask-shape indexing are unrelated." + }, + { + "left": "issue:43317", + "right": "issue:44756", + "accept": false, + "reason": "Both are load/memory related, but one is device_map offload and the other is Strix Halo mmap OOM." + }, + { + "left": "issue:38175", + "right": "issue:39692", + "accept": false, + "reason": "Same SigLIP2 area, but one is incorrect outputs while the other is a doc example / processor mismatch issue." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "Near-verbatim reports of the same non-persistent-buffer junk regression in Transformers v5." + }, + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "Different model families and different loading semantics; not one bug." + }, + { + "left": "issue:41720", + "right": "issue:42915", + "accept": false, + "reason": "Qwen3 auto device-map crash is distinct from the Qwen3Moe FP8 issue." + }, + { + "left": "issue:42915", + "right": "issue:44756", + "accept": false, + "reason": "No shared code path; one is FP8/Qwen3Moe, the other is mmap/OOM on a specific platform." + }, + { + "left": "issue:42491", + "right": "issue:45276", + "accept": false, + "reason": "LoRA HF4/HF5 compatibility is unrelated to Gemma4 embedding resize behavior." + }, + { + "left": "issue:46459", + "right": "issue:46512", + "accept": false, + "reason": "Both involve dtype handling, but one is AutoModel loading of composite checkpoints and the other is from_config model construction; different bugs." + }, + { + "left": "issue:44451", + "right": "issue:44488", + "accept": false, + "reason": "Both are model-loading failures, but the affected models and reported symptoms are not specific enough to conclude a single underlying bug." + }, + { + "left": "issue:44960", + "right": "issue:44977", + "accept": false, + "reason": "Different model families and different flash-attention symptoms." + }, + { + "left": "issue:45313", + "right": "issue:45357", + "accept": false, + "reason": "One is ZeRO-3 weight loading, the other is save_pretrained key serialization; different directions and likely different fixes." + }, + { + "left": "issue:45507", + "right": "issue:46468", + "accept": false, + "reason": "GraniteMoEHybrid invalid method call versus Nemotron example breakage are unrelated." + }, + { + "left": "issue:44488", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer/model loading regressions, but they involve different repos and no clear shared root cause." + }, + { + "left": "issue:43644", + "right": "issue:43645", + "accept": false, + "reason": "Non-persistent-buffer corruption and custom-model initialization are separate regressions." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded and appropriately conservative overall. The three accepted duplicates are well supported by the packet (near-identical non-persistent-buffer reports, the custom-model initialization reports, and the SigLIP2 output regressions), and the non-duplicate verdicts stay cautious without overclaiming." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 7, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:36010", + "issue:39404", + "issue:39692", + "issue:41720", + "issue:42491", + "issue:42947", + "issue:43257", + "issue:43335", + "issue:43381", + "issue:43404", + "issue:43425", + "issue:43454", + "issue: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:44863", + "issue:44910", + "issue:44912", + "issue:44960", + "issue:45276", + "issue:45292", + "issue:45335", + "issue:45362", + "issue:45406", + "issue:45488", + "issue:45588", + "issue:45663", + "issue:45685", + "issue:45704", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:45859", + "issue:46050", + "issue:46468" + ], + "soft_pairs": [ + "issue:43708|issue:44075", + "issue:43856|issue:43873", + "issue:39692|issue:43994", + "issue:43856|issue:44387", + "issue:43854|issue:43931", + "issue:43257|issue:43854", + "issue:42491|issue:45292", + "issue:42491|issue:43931", + "issue:44960|issue:46050", + "issue:45488|issue:45812", + "issue:44164|issue:44291", + "issue:45362|issue:45406", + "issue:45276|issue:45335", + "issue:44910|issue:45588", + "issue:43425|issue:43957", + "issue:44402|issue:44488", + "issue:43873|issue:44387", + "issue:42947|issue:45704", + "issue:43404|issue:43454", + "issue:41720|issue:44155", + "issue:39692|issue:44960", + "issue:45663|issue:45759", + "issue:44291|issue:44756", + "issue:36010|issue:39404", + "issue:45663|issue:45859", + "issue:44387|issue:44756", + "issue:44292|issue:44912", + "issue:44155|issue:45823", + "issue:43957|issue:44756", + "issue:43335|issue:45663", + "issue:42947|issue:43381", + "issue:45276|issue:45292", + "issue:44863|issue:46468", + "issue:45685|issue:46468" + ], + "prepared_review_unit_hash": "913b32883bb03f6258268a51095035fe923b72d05ec168a2a9e04218a8f2e4a2", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37952, + "estimated_input_tokens": 9488, + "estimated_eval_tokens": 19232 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous model-specific bugs; only the resize_token_embeddings/output_embeddings pair looks like a true duplicate. The rest are related only at a broad subsystem level, not the same underlying bug.", + "confidence": 0.84, + "canonical_issue_reason": "issue:45292 is the cleanest generic statement of the shared resize_token_embeddings/output_embeddings regression; issue:45276 appears to be the Gemma4-specific manifestation of the same bug.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45292 is the best representative because it captures the core API regression succinctly and broadly, making it the strongest dedupe anchor.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43708", + "right": "issue:44075", + "accept": false, + "reason": "Different bugs: checkpoint-resume step calculation vs SGD arguments not being applied." + }, + { + "left": "issue:43856", + "right": "issue:43873", + "accept": false, + "reason": "Both involve Qwen3/moe or memory, but one is training-memory inefficiency and the other is quantization/offloading behavior; not the same bug." + }, + { + "left": "issue:39692", + "right": "issue:43994", + "accept": false, + "reason": "Both mention SigLIP2, but one is a broken documentation example while the other is incorrect model/pipeline outputs." + }, + { + "left": "issue:43856", + "right": "issue:44387", + "accept": false, + "reason": "Different failure modes: Qwen3 MoE training memory usage vs generic int4 quantization CUDA reserved-memory/OOM regression." + }, + { + "left": "issue:43854", + "right": "issue:43931", + "accept": false, + "reason": "Both are load failures, but they affect different models and appear to have different root causes." + }, + { + "left": "issue:43257", + "right": "issue:43854", + "accept": false, + "reason": "Different models and mechanisms: Qwen3 MoE conversion with accelerate/deepspeed vs GLM-4.7-Flash loading in tests." + }, + { + "left": "issue:42491", + "right": "issue:45292", + "accept": false, + "reason": "LoRA/qwen3_moe hf4.x vs resize_token_embeddings on output embeddings are unrelated bugs." + }, + { + "left": "issue:42491", + "right": "issue:43931", + "accept": false, + "reason": "Different issues in different model families and code paths." + }, + { + "left": "issue:44960", + "right": "issue:46050", + "accept": false, + "reason": "Both concern quantized-model persistence, but they involve different quantization flows/models and not the same underlying bug." + }, + { + "left": "issue:45488", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and different failure modes (pre-tokenizer override vs wrong token IDs)." + }, + { + "left": "issue:44164", + "right": "issue:44291", + "accept": false, + "reason": "Different save/load bugs: extra_state handling vs init_empty_weights/_is_hf_initialized argument handling." + }, + { + "left": "issue:45362", + "right": "issue:45406", + "accept": false, + "reason": "Different components and failures: chat crash for Qwen3.5-35B vs serve crash due to missing Gemma4Processor._tokenizer." + }, + { + "left": "issue:45276", + "right": "issue:45335", + "accept": false, + "reason": "Same broad API area, but the titles indicate different model classes and different embedding paths; not enough evidence of one shared duplicate bug." + }, + { + "left": "issue:44910", + "right": "issue:45588", + "accept": false, + "reason": "Both are Flash Attention-related, but one is a packed-sequence/position_ids illegal access and the other is an AttributeError on s_aux=None." + }, + { + "left": "issue:43425", + "right": "issue:43957", + "accept": false, + "reason": "Unrelated load/runtime issues: torch version compatibility vs meta-device loading breakage." + }, + { + "left": "issue:44402", + "right": "issue:44488", + "accept": false, + "reason": "Different model-specific vocabulary/loading problems; not the same bug." + }, + { + "left": "issue:43873", + "right": "issue:44387", + "accept": false, + "reason": "Different memory problems caused by different features (offloading/quantization vs int4 reserved memory/OOM)." + }, + { + "left": "issue:42947", + "right": "issue:45704", + "accept": false, + "reason": "Different problems: LoRA gradient checkpointing vs T5 apex RMSNorm memory leak." + }, + { + "left": "issue:43404", + "right": "issue:43454", + "accept": false, + "reason": "Both are missing lm_head tying, but they are model-specific implementations with different affected code paths; too broad to treat as the same duplicate bug." + }, + { + "left": "issue:41720", + "right": "issue:44155", + "accept": false, + "reason": "Different models and failure modes: Qwen3 auto device mapping CUDA assert vs AudioFlamingo3 batched inference leakage." + }, + { + "left": "issue:39692", + "right": "issue:44960", + "accept": false, + "reason": "Documentation/example errors vs quantized save/load behavior are unrelated." + }, + { + "left": "issue:45663", + "right": "issue:45759", + "accept": false, + "reason": "Different model families and symptoms: Gemma4 FSDP2 attention bug vs Qwen multimodal config unwrap issue." + }, + { + "left": "issue:44291", + "right": "issue:44756", + "accept": false, + "reason": "Different loading problems: unexpected init argument vs mmap/OOM on Strix Halo." + }, + { + "left": "issue:36010", + "right": "issue:39404", + "accept": false, + "reason": "ImportError for GenerationMixin is unrelated to Whisper return_language pipeline behavior." + }, + { + "left": "issue:45663", + "right": "issue:45859", + "accept": false, + "reason": "Gemma4 FSDP2 attention state bug vs Qwen3.5-MoE rotary embedding CPU offload incompatibility." + }, + { + "left": "issue:44387", + "right": "issue:44756", + "accept": false, + "reason": "Different resource issues and different triggers: int4 quantization memory growth vs mmap/OOM on Strix Halo." + }, + { + "left": "issue:44292", + "right": "issue:44912", + "accept": false, + "reason": "Different quantized-model load failures: sinq save/load breakage vs MXFP4 fallback to bf16." + }, + { + "left": "issue:44155", + "right": "issue:45823", + "accept": false, + "reason": "AudioFlamingo3 cross-track leakage vs Gemma4 PLE device mismatch are unrelated." + }, + { + "left": "issue:43957", + "right": "issue:44756", + "accept": false, + "reason": "Meta-device loading breakage is unrelated to mmap/OOM behavior." + }, + { + "left": "issue:43335", + "right": "issue:45663", + "accept": false, + "reason": "SwitchTransformers sparse-layer construction bug vs Gemma4 FSDP2 attention bug are unrelated." + }, + { + "left": "issue:42947", + "right": "issue:43381", + "accept": false, + "reason": "Gradient checkpointing ineffective in PEFT LoRA is unrelated to eval-mode restriction." + }, + { + "left": "issue:45276", + "right": "issue:45292", + "accept": true, + "reason": "Both describe the same resize_token_embeddings regression where updated token embeddings are not propagated to output embeddings; one is the Gemma4-specific manifestation, the other the generic report." + }, + { + "left": "issue:44863", + "right": "issue:46468", + "accept": false, + "reason": "NemotronH checkpoint-loading failure vs broken official example are different issues." + }, + { + "left": "issue:45685", + "right": "issue:46468", + "accept": false, + "reason": "MPS histogram not implemented for Int in MoE is unrelated to the Nemotron example breakage." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The clustering is grounded in the packet and the soft-edge calls are conservative. The only accepted duplicate pair, issue:45276 with issue:45292, is well-supported by the titles and fits a shared resize_token_embeddings/output_embeddings regression. The remaining pairs are reasonably treated as different bugs or at most broad subsystem neighbors." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 8, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:39404", + "issue:42796", + "issue:43299", + "issue:43381", + "issue:43425", + "issue:43611", + "issue:43645", + "issue:43646", + "issue:43708", + "issue:43716", + "issue:43828", + "issue:43854", + "issue:43901", + "issue:43906", + "issue:43931", + "issue:43957", + "issue:43994", + "issue:44079", + "issue:44164", + "issue:44265", + "issue:44402", + "issue:44451", + "issue:44479", + "issue:44509", + "issue:44530", + "issue:44589", + "issue:44683", + "issue:44756", + "issue:44805", + "issue:44877", + "issue:44898", + "issue:44912", + "issue:44918", + "issue:44928", + "issue:45084", + "issue:45216", + "issue:45292", + "issue:45335", + "issue:45588", + "issue:45663", + "issue:45684", + "issue:45698", + "issue:45701", + "issue:45812", + "issue:45823", + "issue:46315", + "issue:46326", + "issue:46491" + ], + "soft_pairs": [ + "issue:44479|issue:44877", + "issue:44164|issue:44756", + "issue:43381|issue:44928", + "issue:43854|issue:45216", + "issue:45684|issue:45698", + "issue:45588|issue:45663", + "issue:42796|issue:46326", + "issue:43646|issue:43708", + "issue:45292|issue:45335", + "issue:43828|issue:43957", + "issue:43645|issue:44164", + "issue:44402|issue:44451", + "issue:43425|issue:43828", + "issue:44589|issue:44756", + "issue:45812|issue:46315", + "issue:46315|issue:46491", + "issue:44683|issue:45823", + "issue:39404|issue:45701", + "issue:43611|issue:43645", + "issue:43425|issue:45823", + "issue:43425|issue:43716", + "issue:44530|issue:45084", + "issue:44589|issue:45084", + "issue:44479|issue:44805", + "issue:43906|issue:44530", + "issue:44265|issue:44898", + "issue:43901|issue:44509", + "issue:43425|issue:44918", + "issue:39404|issue:44912", + "issue:43994|issue:44079", + "issue:43299|issue:43931" + ], + "prepared_review_unit_hash": "a2d5c8a2b16ecc555cbd2ea90423c7f6d575e86adf105b9496443bf06c7862b9", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37084, + "estimated_input_tokens": 9271, + "estimated_eval_tokens": 18798 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is mostly a heterogeneous grab bag of unrelated Transformers regressions, not a single tight duplicate cluster. The only clear duplicate-looking pair is the resize_token_embeddings / output_embeddings bug, and the Qwen3-VL loader pair is the next-closest family match.", + "confidence": 0.67, + "canonical_issue_reason": "Issue 45292 is the best canonical anchor: it states the generic embedding-resize regression, and 45335 is the model-specific manifestation of the same bug. The rest of the set is too mixed to collapse into one issue.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45292 is the strongest standalone issue in the set because it is broad, concrete, and directly subsumes its model-specific duplicate. It is the cleanest representative for duplicate triage here.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44479", + "right": "issue:44877", + "accept": false, + "reason": "Both are model regressions, but one is a Qwen VL video-input issue and the other is Granite Speech config strictness; different bugs." + }, + { + "left": "issue:44164", + "right": "issue:44756", + "accept": false, + "reason": "Unrelated save/from_pretrained extra_state handling vs mmap/OOM behavior on Strix Halo." + }, + { + "left": "issue:43381", + "right": "issue:44928", + "accept": false, + "reason": "Gradient checkpointing eval-mode behavior and Qwen3.5 RLHF NaNs are different failures with different code paths." + }, + { + "left": "issue:43854", + "right": "issue:45216", + "accept": false, + "reason": "Both are loading regressions, but they affect different models and different underlying problems." + }, + { + "left": "issue:45684", + "right": "issue:45698", + "accept": false, + "reason": "Both involve custom save/load flows, but one is permissions propagation and the other is wrong module resolution after save_pretrained." + }, + { + "left": "issue:45588", + "right": "issue:45663", + "accept": false, + "reason": "Flash-attention sink-less crash vs Gemma-4 FSDP2 shared_kv_states rebuild; unrelated code paths." + }, + { + "left": "issue:42796", + "right": "issue:46326", + "accept": false, + "reason": "Both are tokenizer issues, but BPE-marker handling and continue_final_message are distinct bugs." + }, + { + "left": "issue:43646", + "right": "issue:43708", + "accept": false, + "reason": "Custom model initialization in notebooks is unrelated to resume_from_checkpoint max_steps math." + }, + { + "left": "issue:45292", + "right": "issue:45335", + "accept": true, + "reason": "Same resize_token_embeddings regression; 45335 is the T5Gemma-specific decoder.embed_tokens manifestation of the general bug." + }, + { + "left": "issue:43828", + "right": "issue:43957", + "accept": false, + "reason": "Autocast dtype mismatch on Phi-tiny-MoE and meta-device loading failures are different issues." + }, + { + "left": "issue:43645", + "right": "issue:44164", + "accept": false, + "reason": "Notebook-defined custom model init and extra_state serialization are separate custom-model problems." + }, + { + "left": "issue:44402", + "right": "issue:44451", + "accept": false, + "reason": "Tokenizer vocab-size mismatch and ScandiBERT loading failure are not the same underlying bug." + }, + { + "left": "issue:43425", + "right": "issue:43828", + "accept": false, + "reason": "Torch 2.10 incompatibility is broader/version-specific, while Phi-tiny-MoE dtype mismatch is model-specific." + }, + { + "left": "issue:44589", + "right": "issue:44756", + "accept": false, + "reason": "Float8 storage lookup failure and Strix Halo mmap OOM are unrelated." + }, + { + "left": "issue:45812", + "right": "issue:46315", + "accept": false, + "reason": "Wrong Granite token IDs and slow convert_tokens_to_ids are both tokenizer regressions, but not the same bug." + }, + { + "left": "issue:46315", + "right": "issue:46491", + "accept": false, + "reason": "Performance regression in token lookup vs CodeLlama space round-trip correctness are different problems." + }, + { + "left": "issue:44683", + "right": "issue:45823", + "accept": false, + "reason": "Compiled flex_attention failure and Gemma4 device-map mismatch do not share a concrete code-path bug." + }, + { + "left": "issue:39404", + "right": "issue:45701", + "accept": false, + "reason": "Whisper pipeline return_language and generic tokenization changes are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:43645", + "accept": false, + "reason": "base_model_prefix loading breakage and notebook custom-model initialization are different custom-model regressions." + }, + { + "left": "issue:43425", + "right": "issue:45823", + "accept": false, + "reason": "Torch compatibility and Gemma4 PLE device mismatch are not the same issue." + }, + { + "left": "issue:43425", + "right": "issue:43716", + "accept": false, + "reason": "Torch version incompatibility and Mistral-3 dtype mismatch are distinct failures." + }, + { + "left": "issue:44530", + "right": "issue:45084", + "accept": false, + "reason": "Qwen3.5 PagedAttentionCache crash and non-template-node compile TypeError are unrelated." + }, + { + "left": "issue:44589", + "right": "issue:45084", + "accept": false, + "reason": "Float8 storage lookup and compile-time template-node errors do not match." + }, + { + "left": "issue:44479", + "right": "issue:44805", + "accept": false, + "reason": "Video-input regression and mask-shape IndexError are different bugs." + }, + { + "left": "issue:43906", + "right": "issue:44530", + "accept": false, + "reason": "Isolated reproduction of another issue and Qwen3.5 cache crash are not clearly the same underlying defect." + }, + { + "left": "issue:44265", + "right": "issue:44898", + "accept": false, + "reason": "torch.export export failures with torch_compilable_check are unrelated to Perceiver interpolation failures." + }, + { + "left": "issue:43901", + "right": "issue:44509", + "accept": false, + "reason": "Both are docs issues, but they concern different pages and different deprecated/changed pipeline text." + }, + { + "left": "issue:43425", + "right": "issue:44918", + "accept": false, + "reason": "Torch incompatibility and TRL SFT input-embedding unpacking are different bugs." + }, + { + "left": "issue:39404", + "right": "issue:44912", + "accept": false, + "reason": "Whisper return_language and MXFP4 quantization fallback are unrelated." + }, + { + "left": "issue:43994", + "right": "issue:44079", + "accept": false, + "reason": "Siglip2 bad outputs and ModelOutput key assignment semantics are different code paths." + }, + { + "left": "issue:43299", + "right": "issue:43931", + "accept": true, + "reason": "Same Qwen3-VL loading regression family; one reports the dev breakage, the other the resulting weight-shape mismatch on a specific checkpoint." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The two accepted pairs are reasonably supported by the issue titles and appear to be same-bug family matches, while the rest of the verdicts are appropriately cautious." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 9, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:32090", + "issue:34567", + "issue:36032", + "issue:39404", + "issue:40444", + "issue:40990", + "issue:42175", + "issue:42796", + "issue:42915", + "issue:43404", + "issue:43421", + "issue:43452", + "issue:43541", + "issue:43582", + "issue:43596", + "issue:43611", + "issue:43632", + "issue:43643", + "issue:43644", + "issue:43646", + "issue:43716", + "issue:43856", + "issue:43906", + "issue:44164", + "issue:44451", + "issue:44462", + "issue:44479", + "issue:44488", + "issue:44530", + "issue:44661", + "issue:44671", + "issue:44683", + "issue:44737", + "issue:44756", + "issue:44811", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45584", + "issue:45759", + "issue:45859", + "issue:46050", + "issue:46178", + "issue:46285", + "issue:46326", + "issue:46402", + "issue:46489" + ], + "soft_pairs": [ + "issue:44671|issue:44991", + "issue:36032|issue:43452", + "issue:46050|issue:46326", + "issue:44164|issue:45759", + "issue:43906|issue:46326", + "issue:43596|issue:44683", + "issue:44991|issue:45081", + "issue:43856|issue:44756", + "issue:44164|issue:44479", + "issue:44488|issue:45081", + "issue:44488|issue:44987", + "issue:44462|issue:46489", + "issue:42796|issue:43404", + "issue:43611|issue:43644", + "issue:44661|issue:44683", + "issue:39404|issue:44671", + "issue:44737|issue:45859", + "issue:32090|issue:39404", + "issue:43716|issue:45084", + "issue:46285|issue:46402", + "issue:36032|issue:42915", + "issue:43582|issue:43632", + "issue:43643|issue:44530", + "issue:40990|issue:43421", + "issue:44451|issue:44987", + "issue:43541|issue:45084", + "issue:42915|issue:44530", + "issue:34567|issue:40444", + "issue:42175|issue:46178", + "issue:43644|issue:43646", + "issue:44811|issue:45584" + ], + "prepared_review_unit_hash": "fd2463489eefb34f84e3b5cbf0d53a8d0f54094548b36df76283bcdd87453332", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37046, + "estimated_input_tokens": 9262, + "estimated_eval_tokens": 18780 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose collection of Transformers v5 regressions, mostly around model/tokenizer loading and initialization, but the soft pairs generally describe distinct bugs rather than one shared fix path.", + "confidence": 0.43, + "canonical_issue_reason": "issue:43646 is the broadest and most representative v5 regression in the set, covering model initialization rather than a narrow model-specific symptom.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 is the best umbrella issue here because it is the most general of the v5 init/load regressions and most likely to represent the cluster\u2019s theme.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44671", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer-load regressions, but they affect different models with different failure modes; not the same underlying bug." + }, + { + "left": "issue:36032", + "right": "issue:43452", + "accept": false, + "reason": "One is a T5 tokenizer method-name conflict, the other is a gguf/from_pretrained loading issue; unrelated code paths." + }, + { + "left": "issue:46050", + "right": "issue:46326", + "accept": false, + "reason": "Quantized save/load corruption and Whisper/mistral_common decoding behavior are different bugs." + }, + { + "left": "issue:44164", + "right": "issue:45759", + "accept": false, + "reason": "extra_state save/load handling and AutoModel text_config unwrapping are separate model-loading problems." + }, + { + "left": "issue:43906", + "right": "issue:46326", + "accept": false, + "reason": "A reproduction of an older issue and a tokenizer continue_final_message bug are not the same change or failure." + }, + { + "left": "issue:43596", + "right": "issue:44683", + "accept": false, + "reason": "Deepspeed ZeRO3/BertModel initialization and torch 2.9 flex_attention compilation are unrelated." + }, + { + "left": "issue:44991", + "right": "issue:45081", + "accept": false, + "reason": "Both involve tokenizer loading, but the specific models and root causes differ." + }, + { + "left": "issue:43856", + "right": "issue:44756", + "accept": false, + "reason": "MoE memory usage and Strix Halo mmap/OOM handling are different performance issues." + }, + { + "left": "issue:44164", + "right": "issue:44479", + "accept": false, + "reason": "save/from_pretrained extra_state handling and a video-input regression are unrelated." + }, + { + "left": "issue:44488", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer load failure and Mistral regex patch crash are different concrete bugs." + }, + { + "left": "issue:44488", + "right": "issue:44987", + "accept": false, + "reason": "Both are load failures for different repos/models, but there is no evidence of one shared defect." + }, + { + "left": "issue:44462", + "right": "issue:46489", + "accept": false, + "reason": "Ignoring tokenizer.json and DeepSeek-Coder output regression are different tokenizer problems." + }, + { + "left": "issue:42796", + "right": "issue:43404", + "accept": false, + "reason": "Tokenizer BPE-marker handling and lm_head weight tying are unrelated issues." + }, + { + "left": "issue:43611", + "right": "issue:43644", + "accept": false, + "reason": "Both are v5 regressions, but one is base_model_prefix loading and the other is non-persistent buffer initialization; distinct bugs." + }, + { + "left": "issue:44661", + "right": "issue:44683", + "accept": false, + "reason": "add-new-model-like mapping logic and flex_attention compilation are unrelated." + }, + { + "left": "issue:39404", + "right": "issue:44671", + "accept": false, + "reason": "Whisper pipeline return_language and CamemBERT masked-LM predictions are different model-specific bugs." + }, + { + "left": "issue:44737", + "right": "issue:45859", + "accept": false, + "reason": "XLNet CPU arange placement and Qwen3_5Moe CPU-offload rotary embedding are not the same issue." + }, + { + "left": "issue:32090", + "right": "issue:39404", + "accept": false, + "reason": "Trainer GPU broadcast NoneType error and Whisper return_language regression are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:45084", + "accept": false, + "reason": "Mistral-3 dtype mismatch and compile non-template nodes are different failures." + }, + { + "left": "issue:46285", + "right": "issue:46402", + "accept": false, + "reason": "CLIP structure inconsistency and Sam3Model weight loading are separate model-architecture regressions." + }, + { + "left": "issue:36032", + "right": "issue:42915", + "accept": false, + "reason": "Tokenizer attribute conflict and Qwen3Moe FP8 configuration failure are unrelated." + }, + { + "left": "issue:43582", + "right": "issue:43632", + "accept": false, + "reason": "Apple Silicon warmup TypeError and _is_hf_initialized flag behavior are different problems." + }, + { + "left": "issue:43643", + "right": "issue:44530", + "accept": false, + "reason": "trust_remote_code field loss and PagedAttentionCache group-type crash are unrelated." + }, + { + "left": "issue:40990", + "right": "issue:43421", + "accept": false, + "reason": "High perplexity on a model and tokenizer backend post-processor updates are not the same bug." + }, + { + "left": "issue:44451", + "right": "issue:44987", + "accept": false, + "reason": "Both are model load failures, but they concern different repositories and no shared concrete failure is shown." + }, + { + "left": "issue:43541", + "right": "issue:45084", + "accept": false, + "reason": "Torch dynamo grouped_mm tracing and compile non-template nodes are different compiler errors." + }, + { + "left": "issue:42915", + "right": "issue:44530", + "accept": false, + "reason": "Qwen3Moe FP8 training and Qwen3.5 PagedAttentionCache crashes are different code paths." + }, + { + "left": "issue:34567", + "right": "issue:40444", + "accept": false, + "reason": "TrainerState token counting and iterable-dataset multi-image finetuning are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:46178", + "accept": false, + "reason": "Missing tensorflow backend in extras and an unsupported tensorflow_text backend mapping are similar packaging areas but not the same bug." + }, + { + "left": "issue:43644", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 init regressions, but non-persistent buffer junk and custom model initialization are distinct failure modes." + }, + { + "left": "issue:44811", + "right": "issue:45584", + "accept": false, + "reason": "Both are Whisper decode-related, but batch_decode skipping special tokens and empty-transcription handling after align_special_tokens 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": 21, + "review_unit_index": 10, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:28530", + "issue:34567", + "issue:36032", + "issue:40444", + "issue:42915", + "issue:43089", + "issue:43262", + "issue:43317", + "issue:43322", + "issue:43377", + "issue:43404", + "issue:43452", + "issue:43482", + "issue:43540", + "issue:43645", + "issue:43646", + "issue:43742", + "issue:43746", + "issue:43819", + "issue:43883", + "issue:43994", + "issue:44062", + "issue:44206", + "issue:44220", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44451", + "issue:44464", + "issue:44466", + "issue:44479", + "issue:44488", + "issue:44534", + "issue:44589", + "issue:44610", + "issue:44683", + "issue:44849", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45216", + "issue:45325", + "issue:45357", + "issue:45507", + "issue:45987", + "issue:46326" + ], + "soft_pairs": [ + "issue:44206|issue:44220", + "issue:43883|issue:44488", + "issue:40444|issue:46326", + "issue:36032|issue:43645", + "issue:43452|issue:43482", + "issue:45084|issue:45507", + "issue:43317|issue:43482", + "issue:44987|issue:44991", + "issue:43262|issue:43377", + "issue:43646|issue:44534", + "issue:43746|issue:46326", + "issue:44464|issue:45987", + "issue:44403|issue:45081", + "issue:42915|issue:45084", + "issue:43645|issue:44534", + "issue:43883|issue:44991", + "issue:43404|issue:44466", + "issue:43994|issue:44610", + "issue:43377|issue:43819", + "issue:44849|issue:45325", + "issue:28530|issue:34567", + "issue:44451|issue:45081", + "issue:43322|issue:43540", + "issue:44987|issue:45081", + "issue:42915|issue:44291", + "issue:44265|issue:44610", + "issue:36032|issue:43742", + "issue:43089|issue:45987", + "issue:42915|issue:44589", + "issue:44062|issue:44987", + "issue:44220|issue:44683", + "issue:45216|issue:45357", + "issue:44479|issue:45325" + ], + "prepared_review_unit_hash": "22aae035c4d531ffd45a94e9e6775c8cf75f41819c5064d66310252248af9360", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 33, + "serialized_chars": 37146, + "estimated_input_tokens": 9287, + "estimated_eval_tokens": 18830 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most pairs are only loosely related by subsystem, not by the same underlying bug. The only clear duplicate is the Qwen3.5 `save_pretrained` checkpoint regression pair.", + "confidence": 0.76, + "canonical_issue_reason": "issue:45216 is the best canonical issue: it describes the regression broadly, includes the version boundary, and captures the `save_pretrained` checkpoint corruption that 45357 appears to continue/specialize.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45216 is the strongest representative issue in this set because it states the shared regression most generally and is the cleanest umbrella for the later follow-up report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44206", + "right": "issue:44220", + "accept": false, + "reason": "Related audio feature-extraction area, but the titles point to different failures (`center` arg vs `_torch_extract_fbank_features()`), so the root cause is not clearly the same." + }, + { + "left": "issue:43883", + "right": "issue:44488", + "accept": false, + "reason": "Both are model-loading complaints, but they target different models and different failure modes; not enough to treat as the same bug." + }, + { + "left": "issue:40444", + "right": "issue:46326", + "accept": false, + "reason": "Different code paths: multimodal iterable-dataset image handling vs tokenizer `continue_final_message` behavior." + }, + { + "left": "issue:36032", + "right": "issue:43645", + "accept": false, + "reason": "Unrelated regressions: tokenizer class method conflict vs custom model initialization in notebooks." + }, + { + "left": "issue:43452", + "right": "issue:43482", + "accept": false, + "reason": "Both involve GGUF/loading, but one is a generic `gguf_file` API break and the other is a specific Qwen2.5-GGUF failure; the common root cause is not clear enough." + }, + { + "left": "issue:45084", + "right": "issue:45507", + "accept": false, + "reason": "Different bugs in different code paths; no strong evidence they are the same underlying issue." + }, + { + "left": "issue:43317", + "right": "issue:43482", + "accept": false, + "reason": "Different failures: device-map/offload dequantized loading vs GGUF loading regression." + }, + { + "left": "issue:44987", + "right": "issue:44991", + "accept": false, + "reason": "Same version family, but different model/tokenizer load failures and no clear shared concrete bug." + }, + { + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Both audio-related, but one is a sampling-rate default issue and the other is a padding-mask batch consistency bug." + }, + { + "left": "issue:43646", + "right": "issue:44534", + "accept": false, + "reason": "Different regressions around custom model handling vs non-persistent buffer initialization." + }, + { + "left": "issue:43746", + "right": "issue:46326", + "accept": false, + "reason": "PEFT/local-checkpoint loading for GraniteSpeech is unrelated to the mistral_common tokenizer message-handling issue." + }, + { + "left": "issue:44464", + "right": "issue:45987", + "accept": false, + "reason": "Compiled chunked generation inconsistency is not the same as the `StaticCache.get_seq_length()` return-type contract bug." + }, + { + "left": "issue:44403", + "right": "issue:45081", + "accept": false, + "reason": "Different problems: generic load-time noise vs Mistral regex patch crash." + }, + { + "left": "issue:42915", + "right": "issue:45084", + "accept": false, + "reason": "Qwen3Moe FP8 config failure and template-node compile error are unrelated." + }, + { + "left": "issue:43645", + "right": "issue:44534", + "accept": false, + "reason": "Custom model init in Jupyter and junk-filled non-persistent buffers are separate issues." + }, + { + "left": "issue:43883", + "right": "issue:44991", + "accept": false, + "reason": "Different model-load/tokenizer-load failures with no concrete shared bug." + }, + { + "left": "issue:43404", + "right": "issue:44466", + "accept": false, + "reason": "Both mention tied weights, but one is a specific model not tying `lm_head` and the other is device-dependent serialization; related theme, not the same bug." + }, + { + "left": "issue:43994", + "right": "issue:44610", + "accept": false, + "reason": "Different vision issues: nonsensical outputs vs processor/model input-size mismatch." + }, + { + "left": "issue:43377", + "right": "issue:43819", + "accept": false, + "reason": "Both are model-output mismatches, but the models and mechanics differ (`padding mask` vs `STE`)." + }, + { + "left": "issue:44849", + "right": "issue:45325", + "accept": false, + "reason": "Different Qwen regressions affecting different features; not the same code-path problem." + }, + { + "left": "issue:28530", + "right": "issue:34567", + "accept": false, + "reason": "Trainer early-stopping behavior and `num_input_tokens_seen` updates are unrelated." + }, + { + "left": "issue:44451", + "right": "issue:45081", + "accept": false, + "reason": "Different failures: loading a specific model vs Mistral regex patch crash." + }, + { + "left": "issue:43322", + "right": "issue:43540", + "accept": false, + "reason": "Different multimodal loading/processing failures; no clear shared underlying bug." + }, + { + "left": "issue:44987", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer/model loading regression vs Mistral tokenizer regex crash are distinct issues." + }, + { + "left": "issue:42915", + "right": "issue:44291", + "accept": false, + "reason": "FP8 configuration failure and `init_empty_weights` argument mismatch are separate loading issues." + }, + { + "left": "issue:44265", + "right": "issue:44610", + "accept": false, + "reason": "torch.export/compilation failure is unrelated to the OmDet-Turbo processor image-size regression." + }, + { + "left": "issue:36032", + "right": "issue:43742", + "accept": false, + "reason": "Different model/tokenizer loading failures with no direct evidence of the same defect." + }, + { + "left": "issue:43089", + "right": "issue:45987", + "accept": false, + "reason": "Generation performance overhead and cache-length return-type mismatch are different problems." + }, + { + "left": "issue:42915", + "right": "issue:44589", + "accept": false, + "reason": "Both mention low-level tensor/storage issues, but the affected paths and root causes are different." + }, + { + "left": "issue:44062", + "right": "issue:44987", + "accept": false, + "reason": "AddedToken keyword collision is not the same as the model-loading failure in 44987." + }, + { + "left": "issue:44220", + "right": "issue:44683", + "accept": false, + "reason": "Different subsystems: audio fbank extraction vs compiled flex_attention on newer torch." + }, + { + "left": "issue:45216", + "right": "issue:45357", + "accept": true, + "reason": "Same Qwen3.5 `save_pretrained` regression; 45357 looks like a later, more specific report of the same incorrect checkpoint/visual-encoder-key bug." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": false, + "reason": "Related Qwen video/rope-index area, but 44479 is a broad video-input regression across models while 45325 is a specific still-image temporal-position bug; not clearly the same underlying defect." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": false, + "reason": "Related Qwen video/rope-index area, but 44479 is a broad video-input regression across models while 45325 is a specific still-image temporal-position bug; not clearly the same underlying defect." + }, + { + "left": "issue:44466", + "right": "issue:45325", + "accept": false, + "reason": "Different Qwen regressions: tied-weight serialization vs temporal position scaling." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The judgment is mostly well grounded in the packet, and the soft-edge calls are conservative overall. The only accepted pair, issue:45216 vs issue:45357, is supported by the titles as the same Qwen3.5 save_pretrained regression family. The summary does not overclaim beyond the evidence provided." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 11, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:33666", + "issue:35707", + "issue:36032", + "issue:40444", + "issue:40990", + "issue:41553", + "issue:41720", + "issue:42175", + "issue:42796", + "issue:43296", + "issue:43317", + "issue:43381", + "issue:43421", + "issue:43425", + "issue:43482", + "issue:43525", + "issue:43531", + "issue:43582", + "issue:43611", + "issue:43632", + "issue:43746", + "issue:43792", + "issue:43818", + "issue:43856", + "issue:44079", + "issue:44291", + "issue:44403", + "issue:44462", + "issue:44534", + "issue:44568", + "issue:44589", + "issue:44756", + "issue:44811", + "issue:44829", + "issue:44877", + "issue:44991", + "issue:45081", + "issue:45325", + "issue:45479", + "issue:45538", + "issue:45542", + "issue:45663", + "issue:45812", + "issue:45923", + "issue:46315", + "issue:46402", + "issue:46491" + ], + "soft_pairs": [ + "issue:43531|issue:44877", + "issue:36032|issue:41720", + "issue:44991|issue:46315", + "issue:43425|issue:44403", + "issue:43611|issue:44534", + "issue:43296|issue:43531", + "issue:43632|issue:44756", + "issue:40444|issue:45325", + "issue:45812|issue:46491", + "issue:43792|issue:44589", + "issue:44403|issue:44991", + "issue:42796|issue:43746", + "issue:33666|issue:40444", + "issue:43381|issue:45923", + "issue:44291|issue:44589", + "issue:44079|issue:45663", + "issue:43632|issue:44534", + "issue:40990|issue:44811", + "issue:43482|issue:43525", + "issue:43482|issue:45538", + "issue:35707|issue:40444", + "issue:43421|issue:44568", + "issue:43317|issue:43856", + "issue:42175|issue:45542", + "issue:41553|issue:43582", + "issue:43531|issue:44291", + "issue:44079|issue:44534", + "issue:43818|issue:46402", + "issue:44829|issue:45479", + "issue:41720|issue:45081", + "issue:44462|issue:45812" + ], + "prepared_review_unit_hash": "32d8bed34e6b37a7e0196e159a460adb5065cde8094b11fb135de975e0ab117e", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36566, + "estimated_input_tokens": 9142, + "estimated_eval_tokens": 18540 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: most items are tokenizer/model-loading regressions or unrelated training/runtime bugs, and the soft-similarity pairs do not look like true duplicates.", + "confidence": 0.34, + "canonical_issue_reason": "issue:44462 is the strongest representative issue in the set: it describes a broad AutoTokenizer regression with repository tokenizer handling, which matches the most central tokenizer-related theme in the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44462 is the best standalone issue here because it is broad, user-visible, and tied to tokenizer loading behavior with external references, making it the most useful anchor for this mixed cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43531", + "right": "issue:44877", + "accept": false, + "reason": "Different models and failure modes: Qwen3-MoE sliding_window vs Granite speech config loading." + }, + { + "left": "issue:36032", + "right": "issue:41720", + "accept": false, + "reason": "Unrelated bugs: T5 tokenizer method conflict vs Qwen3 device-mapping CUDA assert." + }, + { + "left": "issue:44991", + "right": "issue:46315", + "accept": false, + "reason": "Both are tokenizer regressions, but one is loading failure and the other is a convert_tokens_to_ids performance issue." + }, + { + "left": "issue:43425", + "right": "issue:44403", + "accept": false, + "reason": "Torch compatibility is unrelated to noisy model-loading logs." + }, + { + "left": "issue:43611", + "right": "issue:44534", + "accept": false, + "reason": "Different code paths: base_model_prefix loading vs non-persistent buffer initialization." + }, + { + "left": "issue:43296", + "right": "issue:43531", + "accept": false, + "reason": "Different model families and different symptoms: PaddleOCR-VL load failure vs Qwen3-MoE sliding_window." + }, + { + "left": "issue:43632", + "right": "issue:44756", + "accept": false, + "reason": "v5 init flag regression is unrelated to Apple Silicon mmap/OOM behavior." + }, + { + "left": "issue:40444", + "right": "issue:45325", + "accept": false, + "reason": "Same model family, but one is multi-image iterable training and the other is still-image temporal position scaling." + }, + { + "left": "issue:45812", + "right": "issue:46491", + "accept": false, + "reason": "Both involve tokenization, but the concrete bugs differ: Granite token IDs vs CodeLlama leading-space round-trip." + }, + { + "left": "issue:43792", + "right": "issue:44589", + "accept": false, + "reason": "Whisper runtime failure is unrelated to missing Float8 storage support." + }, + { + "left": "issue:44403", + "right": "issue:44991", + "accept": false, + "reason": "Generic loading noise is not the same bug as tokenizer loading for est-roberta." + }, + { + "left": "issue:42796", + "right": "issue:43746", + "accept": false, + "reason": "T5 tokenizer special-token conflict and GraniteSpeech PEFT checkpoint loading are different issues." + }, + { + "left": "issue:33666", + "right": "issue:40444", + "accept": false, + "reason": "Both are vision-model training issues, but the reported failures and code paths differ." + }, + { + "left": "issue:43381", + "right": "issue:45923", + "accept": false, + "reason": "Eval-mode gradient-checkpointing behavior is distinct from a missing supports_gradient_checkpointing flag." + }, + { + "left": "issue:44291", + "right": "issue:44589", + "accept": false, + "reason": "Unexpected _is_hf_initialized argument and missing Float8 storage are unrelated regressions." + }, + { + "left": "issue:44079", + "right": "issue:45663", + "accept": false, + "reason": "ModelOutput key assignment and Gemma4 FSDP shared_kv_states are different internals." + }, + { + "left": "issue:43632", + "right": "issue:44534", + "accept": false, + "reason": "These are different v5 regressions: initialization flag handling vs buffer contents." + }, + { + "left": "issue:40990", + "right": "issue:44811", + "accept": false, + "reason": "Perplexity on WikiText and Whisper batch_decode special-token handling are unrelated." + }, + { + "left": "issue:43482", + "right": "issue:43525", + "accept": false, + "reason": "Different models and different missing pieces: Qwen2.5-GGUF loading vs Llama4Config pad_token_id." + }, + { + "left": "issue:43482", + "right": "issue:45538", + "accept": false, + "reason": "GGUF loading failure is unrelated to CLIPTokenizer max_length behavior." + }, + { + "left": "issue:35707", + "right": "issue:40444", + "accept": false, + "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to multi-image finetuning failures." + }, + { + "left": "issue:43421", + "right": "issue:44568", + "accept": false, + "reason": "Runtime post-processor updates and missing BOS/EOS insertion are related only broadly, not the same bug." + }, + { + "left": "issue:43317", + "right": "issue:43856", + "accept": false, + "reason": "Device-map offload loading and Qwen3 MoE memory efficiency are different problems." + }, + { + "left": "issue:42175", + "right": "issue:45542", + "accept": false, + "reason": "Both touch TensorFlow packaging, but one is missing backend inclusion while the other is a tensorboard-without-tf runtime error." + }, + { + "left": "issue:41553", + "right": "issue:43582", + "accept": false, + "reason": "AutoTokenizer error messaging and Apple Silicon caching allocator TypeError are unrelated." + }, + { + "left": "issue:43531", + "right": "issue:44291", + "accept": false, + "reason": "Qwen3 sliding_window behavior is not the same as init_empty_weights argument handling." + }, + { + "left": "issue:44079", + "right": "issue:44534", + "accept": false, + "reason": "ModelOutput key assignment and junk-filled buffers are different regressions." + }, + { + "left": "issue:43818", + "right": "issue:46402", + "accept": false, + "reason": "Different multimodal model issues: missing temporal attention vs text encoder weights not loading." + }, + { + "left": "issue:44829", + "right": "issue:45479", + "accept": false, + "reason": "Both are classification-training regressions, but the triggers and root causes are different." + }, + { + "left": "issue:41720", + "right": "issue:45081", + "accept": false, + "reason": "Qwen3 device-map CUDA assert is unrelated to Mistral tokenizer regex patching." + }, + { + "left": "issue:44462", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but one is repo tokenizer.json selection and the other is wrong token IDs on Granite models." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 12, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:34634", + "issue:35707", + "issue:39692", + "issue:40444", + "issue:40990", + "issue:41720", + "issue:42175", + "issue:42796", + "issue:42915", + "issue:43262", + "issue:43296", + "issue:43377", + "issue:43454", + "issue:43482", + "issue:43550", + "issue:43606", + "issue:43611", + "issue:43632", + "issue:43644", + "issue:43646", + "issue:43701", + "issue:43746", + "issue:43749", + "issue:43792", + "issue:43818", + "issue:43819", + "issue:43873", + "issue:43881", + "issue:43957", + "issue:43994", + "issue:44079", + "issue:44265", + "issue:44291", + "issue:44466", + "issue:44479", + "issue:44488", + "issue:44877", + "issue:44898", + "issue:45072", + "issue:45325", + "issue:45357", + "issue:45381", + "issue:45491", + "issue:45584", + "issue:45701", + "issue:45800", + "issue:45823", + "issue:45834" + ], + "soft_pairs": [ + "issue:43377|issue:45491", + "issue:43818|issue:45357", + "issue:43644|issue:44079", + "issue:43611|issue:43646", + "issue:44265|issue:45072", + "issue:43749|issue:44079", + "issue:42175|issue:43296", + "issue:45325|issue:45381", + "issue:40444|issue:45381", + "issue:43792|issue:43957", + "issue:43994|issue:44898", + "issue:43454|issue:44466", + "issue:45325|issue:45834", + "issue:42915|issue:43482", + "issue:43632|issue:45072", + "issue:42796|issue:45381", + "issue:43701|issue:45325", + "issue:44291|issue:44877", + "issue:43873|issue:45701", + "issue:44479|issue:45823", + "issue:43296|issue:45800", + "issue:43606|issue:44265", + "issue:43262|issue:43746", + "issue:40990|issue:45584", + "issue:34634|issue:35707", + "issue:35707|issue:41720", + "issue:43454|issue:43550", + "issue:39692|issue:40444", + "issue:43606|issue:43819", + "issue:43881|issue:44488" + ], + "prepared_review_unit_hash": "b12ef12c749a5ba390cc91710caea5dd6f1ca7684a77911e6a619753112482f3", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36672, + "estimated_input_tokens": 9168, + "estimated_eval_tokens": 18592 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly a grab bag of unrelated model-specific bugs. A few Qwen2.5-VL video/position-id reports are superficially close, but none are strong enough to merge as the same concrete bug.", + "confidence": 0.86, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:44479 is the broadest umbrella report among the video-regression items, but the overall set is not a single duplicate cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43377", + "right": "issue:45491", + "accept": false, + "reason": "Both involve batching/padding-related behavior, but they are different models and different failure modes." + }, + { + "left": "issue:43818", + "right": "issue:45357", + "accept": false, + "reason": "One is a Video-LLaVA architecture/weight-sharing bug; the other is a Qwen3.5 save_pretrained key-serialization issue." + }, + { + "left": "issue:43644", + "right": "issue:44079", + "accept": false, + "reason": "Non-persistent buffer junk vs ModelOutput key assignment are unrelated code paths." + }, + { + "left": "issue:43611", + "right": "issue:43646", + "accept": false, + "reason": "Both are Transformers v5 regressions, but one is base_model_prefix loading and the other is custom model initialization; not the same bug." + }, + { + "left": "issue:44265", + "right": "issue:45072", + "accept": false, + "reason": "torch.export with torch_compilable_check and bf16 dtype mismatches are different problems." + }, + { + "left": "issue:43749", + "right": "issue:44079", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and ModelOutput key handling are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:43296", + "accept": false, + "reason": "Packaging/backends issue vs a specific model load failure; no shared underlying bug." + }, + { + "left": "issue:45325", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen2.5-VL position-id regressions, but one is still-image temporal scaling and the other is video vision_position_ids; not clearly the same defect." + }, + { + "left": "issue:40444", + "right": "issue:45381", + "accept": false, + "reason": "Multiple-images finetuning failure and video position-id corruption are different multimodal paths." + }, + { + "left": "issue:43792", + "right": "issue:43957", + "accept": false, + "reason": "A specific model run failure and a meta-device loading regression are not the same bug." + }, + { + "left": "issue:43994", + "right": "issue:44898", + "accept": false, + "reason": "Different models and different symptoms; only a broad image-classification similarity." + }, + { + "left": "issue:43454", + "right": "issue:44466", + "accept": false, + "reason": "Both mention lm_head/tied weights, but one is a missing tie causing garbage generation and the other is serialization inconsistency." + }, + { + "left": "issue:45325", + "right": "issue:45834", + "accept": false, + "reason": "Both are model-specific vision/OCR issues, but the affected models and failure modes differ." + }, + { + "left": "issue:42915", + "right": "issue:43482", + "accept": false, + "reason": "Different models and unrelated failures: FP8 config vs GGUF loading under v5." + }, + { + "left": "issue:43632", + "right": "issue:45072", + "accept": false, + "reason": "_is_hf_initialized handling and bf16 dtype mismatches are separate regressions." + }, + { + "left": "issue:42796", + "right": "issue:45381", + "accept": false, + "reason": "Tokenizer BPE-marker handling is unrelated to Qwen2.5-VL video position ids." + }, + { + "left": "issue:43701", + "right": "issue:45325", + "accept": false, + "reason": "Checkpoint key mismatch and Qwen2.5-VL rope-index scaling are unrelated bugs." + }, + { + "left": "issue:44291", + "right": "issue:44877", + "accept": false, + "reason": "init_empty_weights argument incompatibility and strict config loading are different failures." + }, + { + "left": "issue:43873", + "right": "issue:45701", + "accept": false, + "reason": "Quantization/offloading behavior is unrelated to tokenization changes." + }, + { + "left": "issue:44479", + "right": "issue:45823", + "accept": false, + "reason": "A broad Qwen video regression and a Gemma4 device-mismatch bug are not the same code-path problem." + }, + { + "left": "issue:43296", + "right": "issue:45800", + "accept": false, + "reason": "PaddleOCR-VL loading failure and torch/transformers version incompatibility are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:44265", + "accept": false, + "reason": "CPU offload device mismatch and torch.export compilation failures are different bugs." + }, + { + "left": "issue:43262", + "right": "issue:43746", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting and GraniteSpeech checkpoint loading are unrelated." + }, + { + "left": "issue:40990", + "right": "issue:45584", + "accept": false, + "reason": "High perplexity on one model and Whisper empty-transcription generation failure are unrelated." + }, + { + "left": "issue:34634", + "right": "issue:35707", + "accept": false, + "reason": "Bark voice_preset and inputs_embeds/past_key_values generation are different issues." + }, + { + "left": "issue:35707", + "right": "issue:41720", + "accept": false, + "reason": "Progressive generation and Qwen3 auto device-map asserts are unrelated." + }, + { + "left": "issue:43454", + "right": "issue:43550", + "accept": false, + "reason": "lm_head tying/serialization vs torch.compile SDPA failure are not the same defect." + }, + { + "left": "issue:39692", + "right": "issue:40444", + "accept": false, + "reason": "SigLIP2 doc-example errors and Qwen2.5-VL multi-image finetuning failures are different." + }, + { + "left": "issue:43606", + "right": "issue:43819", + "accept": false, + "reason": "Bark CPU-offload device mismatch and DAC from_latents/forward mismatch are unrelated." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "Two separate model-loading failures with no shared underlying 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": 1, + "component_count": 21, + "review_unit_index": 13, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:26413", + "issue:28530", + "issue:36331", + "issue:40990", + "issue:41720", + "issue:42617", + "issue:42796", + "issue:42915", + "issue:42947", + "issue:43065", + "issue:43257", + "issue:43317", + "issue:43322", + "issue:43377", + "issue:43450", + "issue:43454", + "issue:43526", + "issue:43572", + "issue:43720", + "issue:43749", + "issue:43873", + "issue:43881", + "issue:43883", + "issue:44186", + "issue:44265", + "issue:44387", + "issue:44466", + "issue:44534", + "issue:44568", + "issue:44610", + "issue:44756", + "issue:44877", + "issue:44898", + "issue:44977", + "issue:44991", + "issue:45072", + "issue:45357", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45823", + "issue:45920", + "issue:45923", + "issue:46050", + "issue:46209", + "issue:46305", + "issue:46315", + "issue:46402" + ], + "soft_pairs": [ + "issue:44186|issue:44265", + "issue:41720|issue:45823", + "issue:46305|issue:46315", + "issue:43377|issue:44610", + "issue:42796|issue:46209", + "issue:43454|issue:45072", + "issue:43257|issue:45923", + "issue:44568|issue:45920", + "issue:43377|issue:44898", + "issue:43317|issue:43873", + "issue:42796|issue:44977", + "issue:43454|issue:43720", + "issue:40990|issue:42947", + "issue:43572|issue:45759", + "issue:43322|issue:46050", + "issue:42617|issue:42915", + "issue:43749|issue:44534", + "issue:41720|issue:43322", + "issue:26413|issue:28530", + "issue:43526|issue:43883", + "issue:44387|issue:45704", + "issue:41720|issue:44977", + "issue:44466|issue:45357", + "issue:36331|issue:42617", + "issue:43881|issue:44991", + "issue:36331|issue:42915", + "issue:44877|issue:44991", + "issue:43450|issue:45701", + "issue:40990|issue:45701", + "issue:44186|issue:45072", + "issue:43065|issue:46402", + "issue:43873|issue:44756", + "issue:41720|issue:42617" + ], + "prepared_review_unit_hash": "32c27b3385013b417195f4b9a921e3a62342429979834001700c57630332c8ab", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37626, + "estimated_input_tokens": 9407, + "estimated_eval_tokens": 19070 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These issues are a heterogeneous mix of unrelated regressions across tokenizers, loading, quantization, device mapping, and training utilities. None of the soft pairs look like the same underlying bug/change, so there is no safe duplicate cluster anchor here.", + "confidence": 0.95, + "canonical_issue_reason": "No single canonical issue: the items span different subsystems and failure modes, so there is no safe duplicate representative for the whole set.", + "canonical_pr_reason": null, + "best_issue_reason": "No strong global best issue; if forced, 45920 is the broadest tokenizer regression, but it does not represent the rest of the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44186", + "right": "issue:44265", + "accept": false, + "reason": "Tokenizer NER/padding crash vs torch.export + torch_compilable_check export failure are unrelated code paths." + }, + { + "left": "issue:41720", + "right": "issue:45823", + "accept": false, + "reason": "Both involve device placement, but one is a CUDA assert on Qwen3 auto mapping and the other is a Gemma4 PLE forward device mismatch; different bugs." + }, + { + "left": "issue:46305", + "right": "issue:46315", + "accept": false, + "reason": "Sliding-window mask off-by-one and slow convert_tokens_to_ids are different classes of bug." + }, + { + "left": "issue:43377", + "right": "issue:44610", + "accept": false, + "reason": "Missing padding-mask support in MIMI encoder vs OmDet-Turbo processor shape mismatch are unrelated." + }, + { + "left": "issue:42796", + "right": "issue:46209", + "accept": false, + "reason": "Tokenizer BPE-marker handling vs FP8 inference failure on B200 are not the same bug." + }, + { + "left": "issue:43454", + "right": "issue:45072", + "accept": false, + "reason": "Weight-tying/garbage text issue vs bf16 dtype mismatch are different failure modes." + }, + { + "left": "issue:43257", + "right": "issue:45923", + "accept": false, + "reason": "MoE weight conversion under accelerate+deepspeed is unrelated to a missing gradient_checkpointing flag." + }, + { + "left": "issue:44568", + "right": "issue:45920", + "accept": false, + "reason": "Both are tokenizer regressions, but one is missing BOS/EOS for a specific tokenizer while the other is wrong token IDs across many models; not the same bug." + }, + { + "left": "issue:43377", + "right": "issue:44898", + "accept": false, + "reason": "MIMI batch/padding behavior and Perceiver image-size handling are unrelated model-specific issues." + }, + { + "left": "issue:43317", + "right": "issue:43873", + "accept": false, + "reason": "Dequantized model loading with gpu+cpu offload and generic quantization offloading problems are too broad and not the same concrete defect." + }, + { + "left": "issue:42796", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer BPE marker handling vs flash-attention generation failure are unrelated." + }, + { + "left": "issue:43454", + "right": "issue:43720", + "accept": false, + "reason": "AyaVision lm_head tying and BitNet packed-weight unpacking are different model-loading bugs." + }, + { + "left": "issue:40990", + "right": "issue:42947", + "accept": false, + "reason": "High perplexity on a model benchmark vs ineffective gradient checkpointing with PEFT LoRA are unrelated." + }, + { + "left": "issue:43572", + "right": "issue:45759", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig and unwrapping text_config for composite Qwen configs are different config-loading problems." + }, + { + "left": "issue:43322", + "right": "issue:46050", + "accept": false, + "reason": "Llava Next segfault on load vs sinq quantized save/load breakage are not the same underlying bug." + }, + { + "left": "issue:42617", + "right": "issue:42915", + "accept": false, + "reason": "Failure to run 3d_parallel.py and Qwen3Moe FP8 config failure are unrelated." + }, + { + "left": "issue:43749", + "right": "issue:44534", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and non-persistent buffer corruption are different issues, even if both touch loading/memory." + }, + { + "left": "issue:41720", + "right": "issue:43322", + "accept": false, + "reason": "Qwen3 auto device mapping CUDA assert and Llava Next load segfault are unrelated model/device failures." + }, + { + "left": "issue:26413", + "right": "issue:28530", + "accept": false, + "reason": "Checkpoint resume failing on empty epoch_iterator and early stopping metric lookup are different training-loop bugs." + }, + { + "left": "issue:43526", + "right": "issue:43883", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug vs missing all_tied_weights_keys on Molmo are unrelated." + }, + { + "left": "issue:44387", + "right": "issue:45704", + "accept": false, + "reason": "Int4 quantization OOM from reserved CUDA memory and an apex RMSNorm memory leak are different memory problems." + }, + { + "left": "issue:41720", + "right": "issue:44977", + "accept": false, + "reason": "CUDA assert under auto device mapping and flash-attention generation issues are unrelated." + }, + { + "left": "issue:44466", + "right": "issue:45357", + "accept": false, + "reason": "lm_head serialization inconsistency and incorrect visual encoder keys are different save_pretrained bugs." + }, + { + "left": "issue:36331", + "right": "issue:42617", + "accept": false, + "reason": "Unexpected compute_loss kwarg and 3d_parallel.py failure are unrelated." + }, + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "Model loading failure for glm-4v-9b vs tokenizer loading regression for est-roberta are unrelated." + }, + { + "left": "issue:36331", + "right": "issue:42915", + "accept": false, + "reason": "Unexpected compute_loss kwarg and Qwen3Moe FP8 failure are unrelated." + }, + { + "left": "issue:44877", + "right": "issue:44991", + "accept": false, + "reason": "Strict config loading failure and tokenizer loading regression are different bugs." + }, + { + "left": "issue:43450", + "right": "issue:45701", + "accept": false, + "reason": "Batched video processor shape bug and version-dependent tokenization changes are unrelated." + }, + { + "left": "issue:40990", + "right": "issue:45701", + "accept": false, + "reason": "Perplexity regression and tokenization changes are different symptoms and code paths." + }, + { + "left": "issue:44186", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer crash with NER/padding and bf16 dtype mismatch are unrelated." + }, + { + "left": "issue:43065", + "right": "issue:46402", + "accept": false, + "reason": "Sam3 pixel decoder dummy Conv2d and Sam3 text-encoder weights not loaded are different components and different bugs." + }, + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "Quantization/offloading behavior and Strix Halo mmap OOM are not the same underlying issue." + }, + { + "left": "issue:41720", + "right": "issue:42617", + "accept": false, + "reason": "Qwen3 device-map CUDA assert and inability to run 3d_parallel.py 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": 21, + "review_unit_index": 14, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:28530", + "issue:30064", + "issue:33357", + "issue:33666", + "issue:34634", + "issue:35707", + "issue:36010", + "issue:36331", + "issue:41720", + "issue:42617", + "issue:42915", + "issue:43295", + "issue:43316", + "issue:43317", + "issue:43322", + "issue:43404", + "issue:43526", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43572", + "issue:43577", + "issue:43606", + "issue:43653", + "issue:43708", + "issue:43720", + "issue:43782", + "issue:43873", + "issue:43881", + "issue:44008", + "issue:44186", + "issue:44361", + "issue:44423", + "issue:44610", + "issue:44671", + "issue:44743", + "issue:44898", + "issue:44913", + "issue:45161", + "issue:45357", + "issue:45406", + "issue:45632", + "issue:45684", + "issue:45701", + "issue:45704", + "issue:45823", + "issue:46291" + ], + "soft_pairs": [ + "issue:43550|issue:43720", + "issue:43577|issue:43708", + "issue:43404|issue:45357", + "issue:44008|issue:45823", + "issue:43295|issue:45701", + "issue:43550|issue:43606", + "issue:44186|issue:44610", + "issue:44610|issue:44898", + "issue:44423|issue:45406", + "issue:42915|issue:43317", + "issue:45632|issue:45684", + "issue:36010|issue:42617", + "issue:41720|issue:45161", + "issue:42915|issue:46291", + "issue:43531|issue:43572", + "issue:28530|issue:33666", + "issue:44671|issue:44743", + "issue:36331|issue:41720", + "issue:44186|issue:44361", + "issue:43526|issue:43881", + "issue:42915|issue:43322", + "issue:43316|issue:44913", + "issue:28530|issue:30064", + "issue:43540|issue:43782", + "issue:33357|issue:35707", + "issue:43873|issue:45704", + "issue:30064|issue:34634", + "issue:43550|issue:43653", + "issue:43606|issue:43720" + ], + "prepared_review_unit_hash": "e2f3b02be3a69c72caf9632ff1e39acc335a1cd52bc2df600282827d5fea1ec1", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35831, + "estimated_input_tokens": 8958, + "estimated_eval_tokens": 18172 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly unrelated issues that only share broad subsystem keywords (offloading, multimodal processors, tokenizers, training). The overlaps are not tight enough to treat any soft pair as the same underlying bug, so all soft edges should be rejected.", + "confidence": 0.79, + "canonical_issue_reason": "issue:43873 is the best umbrella representative: it is broad, current, and centered on offloading/quantization behavior, which is the closest common theme among several nearby issues. It is still not a duplicate of the others.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 has the clearest general scope and the strongest chance of serving as a cluster anchor, even though the surrounding issues are distinct bug reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43550", + "right": "issue:43720", + "accept": false, + "reason": "Different failure modes: torch.compile with SDPA vs packed weights not unpacked during accelerate loading." + }, + { + "left": "issue:43577", + "right": "issue:43708", + "accept": false, + "reason": "Unrelated bugs: BLIP-2 dtype handling vs Trainer resume/max_steps calculation." + }, + { + "left": "issue:43404", + "right": "issue:45357", + "accept": false, + "reason": "Both touch multimodal model internals, but one is lm_head tying and the other is save_pretrained visual key serialization." + }, + { + "left": "issue:44008", + "right": "issue:45823", + "accept": false, + "reason": "Different root causes: Gemma3n variable-name collision vs Gemma4 PLE device mismatch." + }, + { + "left": "issue:43295", + "right": "issue:45701", + "accept": false, + "reason": "Custom processor/tokenizer API regression vs version-dependent tokenization changes are not the same bug." + }, + { + "left": "issue:43550", + "right": "issue:43606", + "accept": false, + "reason": "Both mention device/offload contexts, but one is torch.compile+SDPA and the other is Bark CPU offload mismatch." + }, + { + "left": "issue:44186", + "right": "issue:44610", + "accept": false, + "reason": "Tokenizer padding/NER crash vs processor image-size mismatch are different code paths." + }, + { + "left": "issue:44610", + "right": "issue:44898", + "accept": false, + "reason": "Both are vision preprocessing issues, but one is fixed-size processor output and the other is Perceiver interpolation failure." + }, + { + "left": "issue:44423", + "right": "issue:45406", + "accept": false, + "reason": "Both affect serve/multimodal processor paths, but the concrete crashes and root causes differ." + }, + { + "left": "issue:42915", + "right": "issue:43317", + "accept": false, + "reason": "Quantization-related, but Qwen3Moe FineGrainedFP8 and dequantized model offload are different problems." + }, + { + "left": "issue:45632", + "right": "issue:45684", + "accept": false, + "reason": "Cache-path collision and file-permissions propagation are unrelated save/load behaviors." + }, + { + "left": "issue:36010", + "right": "issue:42617", + "accept": false, + "reason": "ImportError for GenerationMixin is unrelated to not being able to run 3d_parallel.py." + }, + { + "left": "issue:41720", + "right": "issue:45161", + "accept": false, + "reason": "Both concern model parallelism/device mapping, but the reported failures and model stacks are different." + }, + { + "left": "issue:42915", + "right": "issue:46291", + "accept": false, + "reason": "FineGrainedFP8 loading bug vs pipelines import dependency error are unrelated." + }, + { + "left": "issue:43531", + "right": "issue:43572", + "accept": false, + "reason": "Qwen3-MoE sliding window behavior and StableLmConfig pad_token_idx regression are different issues." + }, + { + "left": "issue:28530", + "right": "issue:33666", + "accept": false, + "reason": "Early stopping metric selection vs Qwen2-VL multi-GPU training are not the same bug." + }, + { + "left": "issue:44671", + "right": "issue:44743", + "accept": false, + "reason": "CamemBERT masked-LM prediction regression vs Qwen3.5 recurrent state reset are unrelated model bugs." + }, + { + "left": "issue:36331", + "right": "issue:41720", + "accept": false, + "reason": "Trainer compute_loss signature breakage is unrelated to Qwen3 auto device mapping cuda assert." + }, + { + "left": "issue:44186", + "right": "issue:44361", + "accept": false, + "reason": "Different tokenizer/model families and different AttributeErrors; not one underlying defect." + }, + { + "left": "issue:43526", + "right": "issue:43881", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug vs glm-4v-9b loading failure are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:43322", + "accept": false, + "reason": "One is a quantization config failure, the other is a Llava Next segfault during loading." + }, + { + "left": "issue:43316", + "right": "issue:44913", + "accept": false, + "reason": "Gemma3TextConfig API discrepancy and GPTNeoX rotary_pct reload regression are separate config issues." + }, + { + "left": "issue:28530", + "right": "issue:30064", + "accept": false, + "reason": "Early stopping/eval metric handling and void segmentation map processing are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:43782", + "accept": false, + "reason": "Video input processing in Qwen3OmniMoe is not the same as Qwen3VL from_pretrained weight_only failure." + }, + { + "left": "issue:33357", + "right": "issue:35707", + "accept": false, + "reason": "MacOS CLIP bus error and progressive generation with inputs_embeds/past_key_values are different bugs." + }, + { + "left": "issue:43873", + "right": "issue:45704", + "accept": false, + "reason": "Quantization/offload behavior is unrelated to T5's apex.FusedRMSNorm memory leak." + }, + { + "left": "issue:30064", + "right": "issue:34634", + "accept": false, + "reason": "Segmentation-map processing bug vs Bark voice_preset issue are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:43653", + "accept": false, + "reason": "torch.compile/SDPA failure is not the same as tokenizer special-token registration." + }, + { + "left": "issue:43606", + "right": "issue:43720", + "accept": false, + "reason": "CPU offload device mismatch and packed-weight loading are 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": 21, + "review_unit_index": 15, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:29159", + "issue:34567", + "issue:34634", + "issue:35707", + "issue:36032", + "issue:39692", + "issue:41720", + "issue:42490", + "issue:43329", + "issue:43381", + "issue:43388", + "issue:43421", + "issue:43531", + "issue:43606", + "issue:43653", + "issue:43746", + "issue:43844", + "issue:43927", + "issue:44062", + "issue:44077", + "issue:44368", + "issue:44387", + "issue:44451", + "issue:44492", + "issue:44509", + "issue:44514", + "issue:44545", + "issue:44589", + "issue:44610", + "issue:44625", + "issue:44898", + "issue:44918", + "issue:44936", + "issue:44987", + "issue:44991", + "issue:45042", + "issue:45059", + "issue:45072", + "issue:45357", + "issue:45406", + "issue:45464", + "issue:45479", + "issue:45588", + "issue:45701", + "issue:45759", + "issue:45859", + "issue:46402" + ], + "soft_pairs": [ + "issue:29159|issue:36032", + "issue:44387|issue:45701", + "issue:43388|issue:45059", + "issue:44492|issue:44509", + "issue:41720|issue:43746", + "issue:43531|issue:45759", + "issue:42490|issue:43329", + "issue:43606|issue:43653", + "issue:45406|issue:45464", + "issue:45357|issue:46402", + "issue:34634|issue:41720", + "issue:34567|issue:35707", + "issue:43531|issue:44589", + "issue:44451|issue:44991", + "issue:45759|issue:45859", + "issue:44077|issue:44625", + "issue:44625|issue:45479", + "issue:44898|issue:45072", + "issue:45588|issue:45759", + "issue:44062|issue:44451", + "issue:44987|issue:45042", + "issue:44368|issue:45357", + "issue:39692|issue:44610", + "issue:44514|issue:44545", + "issue:43381|issue:43844", + "issue:44918|issue:44936", + "issue:43421|issue:43927" + ], + "prepared_review_unit_hash": "a59eac9bd1462f8c7950f821074af1911e82a0b0d8022c512213938dde15a75a", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35582, + "estimated_input_tokens": 8896, + "estimated_eval_tokens": 18048 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Only one soft duplicate looks convincing: the two Qwen2_5_VLProcessor batched-input crash reports are the same underlying bug. The rest are title-level similarities across different components or failure modes, so they should stay separate.", + "confidence": 0.94, + "canonical_issue_reason": "issue:44514 and issue:44545 describe the same crash in `Qwen2_5_VLProcessor.apply_chat_template` with batched input and `padding=False`. I\u2019d keep 44514 as canonical because it was filed first and appears to have the fuller report.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44514 is the best representative for the duplicated Qwen2_5_VLProcessor bug: it is specific, directly actionable, and matches the exact failure mode in the paired duplicate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:29159", + "right": "issue:36032", + "accept": false, + "reason": "Both are tokenizer-related, but one is slow/fast inconsistency while the other is a T5 tokenizer load-time `add_special_tokens` conflict; different bugs." + }, + { + "left": "issue:44387", + "right": "issue:45701", + "accept": false, + "reason": "Different problems: int4 quantization memory regression vs tokenization changes across versions. Same broad area, not the same bug." + }, + { + "left": "issue:43388", + "right": "issue:45059", + "accept": false, + "reason": "One is a distributed metrics tuple-truncation bug; the other is SAM3 PCS behavior with text and boxes. No concrete overlap." + }, + { + "left": "issue:44492", + "right": "issue:44509", + "accept": false, + "reason": "A typo in cache strategy docs and removed pipeline task docs are both documentation issues, but they are unrelated changes." + }, + { + "left": "issue:41720", + "right": "issue:43746", + "accept": false, + "reason": "Qwen3 auto device mapping CUDA assert and GraniteSpeech PEFT local checkpoint loading are different model-loading failures." + }, + { + "left": "issue:43531", + "right": "issue:45759", + "accept": false, + "reason": "Both mention Qwen3.x config behavior, but one is a sliding_window issue and the other is `from_config` not unwrapping `text_config`; different code paths." + }, + { + "left": "issue:42490", + "right": "issue:43329", + "accept": false, + "reason": "Trailing-slash local model import warning and a multimodal token counting video-branch NameError are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:43653", + "accept": false, + "reason": "Both are Bark-related CI issues, but one is CPU offload device mismatch and the other is a BigBirdTokenizer special-token decode bug." + }, + { + "left": "issue:45406", + "right": "issue:45464", + "accept": false, + "reason": "Gemma4Processor missing `_tokenizer` in serve and Qwen3.5 streaming chat/completions failures are different subsystems and failure modes." + }, + { + "left": "issue:45357", + "right": "issue:46402", + "accept": false, + "reason": "Both concern model saving/loading in multimodal models, but one is Qwen3.5 visual encoder keys on save and the other is Sam3Model text encoder weights not loading." + }, + { + "left": "issue:34634", + "right": "issue:41720", + "accept": false, + "reason": "BarkProcessor voice_preset failure and Qwen3 auto device mapping CUDA assert are unrelated." + }, + { + "left": "issue:34567", + "right": "issue:35707", + "accept": false, + "reason": "TrainerState token counter not updating and progressive generation with inputs_embeds/past_key_values are separate trainer/runtime issues." + }, + { + "left": "issue:43531", + "right": "issue:44589", + "accept": false, + "reason": "Sliding window handling for Qwen3-MoE and a Float8 storage lookup TypeError are unrelated." + }, + { + "left": "issue:44451", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer/load failures for different models, but one is a generic latest-version load regression and the other is an EMBEDDIA tokenizer issue; not the same bug." + }, + { + "left": "issue:45759", + "right": "issue:45859", + "accept": false, + "reason": "Both involve Qwen 3.5 config/model mechanics, but one is `from_config` text_config unwrapping and the other is CPU-offload rotary embedding compatibility." + }, + { + "left": "issue:44077", + "right": "issue:44625", + "accept": false, + "reason": "Optional `post_init` in patchtsmixer and Qwen3.5 `num_labels` propagation are different configuration bugs." + }, + { + "left": "issue:44625", + "right": "issue:45479", + "accept": false, + "reason": "One is config propagation for Qwen3.5; the other is zero-loss behavior for single-label classification across models. Different underlying issues." + }, + { + "left": "issue:44898", + "right": "issue:45072", + "accept": false, + "reason": "Perceiver image classification resolution failure and dtype mismatches in bfloat16 inference are separate model-processing bugs." + }, + { + "left": "issue:45588", + "right": "issue:45759", + "accept": false, + "reason": "Flash attention `s_aux=None` crash and Qwen multimodal config unwrapping are unrelated code paths." + }, + { + "left": "issue:44062", + "right": "issue:44451", + "accept": false, + "reason": "AddedToken special-argument duplication and a model load failure for ScandiBERT are different tokenizer/model issues." + }, + { + "left": "issue:44987", + "right": "issue:45042", + "accept": false, + "reason": "A loading regression for a specific model and a PIL backend torchvision dependency regression are unrelated." + }, + { + "left": "issue:44368", + "right": "issue:45357", + "accept": false, + "reason": "A tie_word_embeddings warning during LoRA fine-tuning and incorrect visual encoder keys on save are not the same failure." + }, + { + "left": "issue:39692", + "right": "issue:44610", + "accept": false, + "reason": "SigLIP2 docs example errors and OmDet-Turbo processor/model resolution mismatch are both vision-related but describe different problems." + }, + { + "left": "issue:44514", + "right": "issue:44545", + "accept": true, + "reason": "Exact same title and same failure mode: `Qwen2_5_VLProcessor.apply_chat_template` crashes on batched input when `padding=False`. These are duplicates." + }, + { + "left": "issue:43381", + "right": "issue:43844", + "accept": false, + "reason": "Gradient checkpointing in eval mode and gradient blow-up with DeepSpeed ZeRO-3 are distinct training issues." + }, + { + "left": "issue:44918", + "right": "issue:44936", + "accept": false, + "reason": "Qwen3.5 input-embedding unpacking with TRL SFT trainer and trainer.evaluate() failing after train() are different trainer bugs." + }, + { + "left": "issue:43421", + "right": "issue:43927", + "accept": false, + "reason": "Runtime post-processor updates for TokenizersBackend and DiaConfig losing custom token IDs after save/load are related to tokens but not the same bug." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The evaluation is grounded in the packet and the soft-edge verdicts are conservative. The only accepted duplicate, issue:44514 vs issue:44545, is directly supported by identical titles and matching failure mode. The remaining pairs are sensibly rejected as different bugs or only loosely related." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 16, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:30064", + "issue:36010", + "issue:39692", + "issue:41720", + "issue:42886", + "issue:43295", + "issue:43421", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43596", + "issue:43618", + "issue:43643", + "issue:43645", + "issue:43653", + "issue:43697", + "issue:43720", + "issue:43746", + "issue:43782", + "issue:43819", + "issue:43873", + "issue:43927", + "issue:43994", + "issue:44008", + "issue:44062", + "issue:44186", + "issue:44220", + "issue:44488", + "issue:44661", + "issue:44671", + "issue:44843", + "issue:44877", + "issue:44898", + "issue:45072", + "issue:45081", + "issue:45356", + "issue:45381", + "issue:45663", + "issue:45704", + "issue:45834", + "issue:45859", + "issue:45907", + "issue:45923", + "issue:46032", + "issue:46050", + "issue:46294", + "issue:46463" + ], + "soft_pairs": [ + "issue:45704|issue:45923", + "issue:43746|issue:46050", + "issue:41720|issue:45704", + "issue:45704|issue:45859", + "issue:43531|issue:43645", + "issue:46032|issue:46463", + "issue:42886|issue:44843", + "issue:43782|issue:45381", + "issue:43421|issue:43653", + "issue:45663|issue:45923", + "issue:39692|issue:43873", + "issue:43994|issue:44671", + "issue:43746|issue:43873", + "issue:43421|issue:44186", + "issue:30064|issue:36010", + "issue:45081|issue:45356", + "issue:43720|issue:43819", + "issue:43596|issue:44220", + "issue:43653|issue:43720", + "issue:45356|issue:45907", + "issue:43540|issue:46294", + "issue:43295|issue:45834", + "issue:43618|issue:43697", + "issue:44220|issue:44661", + "issue:43550|issue:45072", + "issue:43550|issue:43927", + "issue:44062|issue:44488", + "issue:44186|issue:44898", + "issue:43531|issue:43643", + "issue:43653|issue:43819", + "issue:43643|issue:44877", + "issue:43540|issue:44008" + ], + "prepared_review_unit_hash": "2129d5cdb17ed9259c412d778c02f1cacbf0edbe777aabdcfe43c3d82d63a22a", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 37088, + "estimated_input_tokens": 9272, + "estimated_eval_tokens": 18800 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the items span unrelated bugs in tokenizers, multimodal inputs, quantization/offloading, config loading, and model-specific inference paths. The soft pairs are mostly only superficially similar, so none should be merged as duplicates.", + "confidence": 0.89, + "canonical_issue_reason": "No clear canonical duplicate emerges because the cluster does not center on one underlying bug; the issues are spread across distinct code paths and model families.", + "canonical_pr_reason": null, + "best_issue_reason": "If one issue must be used as a representative anchor, issue 43873 is the broadest and most discussion-heavy open report, but it still does not subsume the rest of the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45704", + "right": "issue:45923", + "accept": false, + "reason": "Different bugs: one is an apex RMSNorm memory leak in T5, the other is a missing supports_gradient_checkpointing flag in a Nemotron trust_remote_code variant." + }, + { + "left": "issue:43746", + "right": "issue:46050", + "accept": false, + "reason": "Unrelated failures: PEFT/local checkpoint loading for GraniteSpeech vs broken save/load behavior for sinq-quantized models." + }, + { + "left": "issue:41720", + "right": "issue:45704", + "accept": false, + "reason": "Different code paths and symptoms: Qwen3 auto device mapping cuda assert vs a T5 RMSNorm memory leak." + }, + { + "left": "issue:45704", + "right": "issue:45859", + "accept": false, + "reason": "Both mention runtime issues, but one is an apex leak and the other is CPU-offload incompatibility in Qwen3_5Moe rotary embeddings." + }, + { + "left": "issue:43531", + "right": "issue:43645", + "accept": false, + "reason": "Separate problems: Qwen3-MoE sliding_window behavior vs notebook custom-model initialization breakage in Transformers 5.0." + }, + { + "left": "issue:46032", + "right": "issue:46463", + "accept": false, + "reason": "Different bug classes: incorrect caching results in Mamba2Mixer vs an error-message/validation issue in T5 decoder inputs." + }, + { + "left": "issue:42886", + "right": "issue:44843", + "accept": false, + "reason": "Both involve offline mode, but one is a general tokenizer cache-load regression and the other is a specific Mistral regex patch making an unconditional network call." + }, + { + "left": "issue:43782", + "right": "issue:45381", + "accept": false, + "reason": "Different models and failures: Qwen3VL weight_only loading error vs Qwen2.5-VL video position-id mismatch." + }, + { + "left": "issue:43421", + "right": "issue:43653", + "accept": false, + "reason": "Related tokenization area, but not the same bug: runtime special-token/post-processor updates vs BigBirdTokenizer mask-token registration/empty decode." + }, + { + "left": "issue:45663", + "right": "issue:45923", + "accept": false, + "reason": "Unrelated: Gemma4 FSDP2 shared_kv_states bug vs a missing gradient-checkpointing capability flag in a Nemotron model." + }, + { + "left": "issue:39692", + "right": "issue:43873", + "accept": false, + "reason": "One is a SigLIP2 documentation/model-processor mismatch with quantization issues; the other is a general offloading-with-quantization runtime bug." + }, + { + "left": "issue:43994", + "right": "issue:44671", + "accept": false, + "reason": "Different concerns: SigLIP2 inference quality/pipeline behavior vs add-new-model-like failing when the model is in TOKENIZER_MAPPING_NAMES." + }, + { + "left": "issue:43746", + "right": "issue:43873", + "accept": false, + "reason": "Both touch loading/quantization-adjacent behavior, but they describe different problems: PEFT checkpoint loading vs offloading with quantization." + }, + { + "left": "issue:43421", + "right": "issue:44186", + "accept": false, + "reason": "Both are tokenizer bugs, but the concrete failures differ: runtime special-token updates vs LayoutLMv2 NER/batched padding-truncation crashes." + }, + { + "left": "issue:30064", + "right": "issue:36010", + "accept": false, + "reason": "Completely unrelated: image processor void segmentation map handling vs an ImportError for GenerationMixin." + }, + { + "left": "issue:45081", + "right": "issue:45356", + "accept": false, + "reason": "Both involve Mistral/Kimi tokenizer regex handling, but one is a backend_tokenizer AttributeError and the other is a codec-handling regression with a warning issue." + }, + { + "left": "issue:43720", + "right": "issue:43819", + "accept": false, + "reason": "Different model internals and symptoms: BitNet packed-weight loading vs DAC latent reconstruction missing STE." + }, + { + "left": "issue:43596", + "right": "issue:44220", + "accept": false, + "reason": "No shared underlying bug: DeepSpeed ZeRO3/BertModel init failure vs a feature-extraction helper issue." + }, + { + "left": "issue:43653", + "right": "issue:43720", + "accept": false, + "reason": "Distinct tokenizer-special-token bug vs BitNet accelerate-loading bug; only broad infrastructure similarity." + }, + { + "left": "issue:45356", + "right": "issue:45907", + "accept": false, + "reason": "Unrelated: tokenizer codec/regex regression vs list_repo_files failing to forward revision." + }, + { + "left": "issue:43540", + "right": "issue:46294", + "accept": false, + "reason": "Both concern video inputs, but they are different failures in different Qwen models and code paths." + }, + { + "left": "issue:43295", + "right": "issue:45834", + "accept": false, + "reason": "Different regressions: processor/tokenizer API breakage vs an OCR long-input index error in Kosmos2.5." + }, + { + "left": "issue:43618", + "right": "issue:43697", + "accept": false, + "reason": "Different issues: CLIP attentions field assignment vs RTDetrV2 output divergence in v5." + }, + { + "left": "issue:44220", + "right": "issue:44661", + "accept": false, + "reason": "No shared defect: audio fbank extraction helper vs add-new-model-like failure on TOKENIZER_MAPPING_NAMES." + }, + { + "left": "issue:43550", + "right": "issue:45072", + "accept": false, + "reason": "Different runtime problems: torch.compile/SDPA failure in Bamba vs bfloat16 dtype mismatches in SwitchTransformers and TimmWrapperModel." + }, + { + "left": "issue:43550", + "right": "issue:43927", + "accept": false, + "reason": "Unrelated: torch.compile+SDPA failure vs DiaConfig losing custom token IDs after save/load." + }, + { + "left": "issue:44062", + "right": "issue:44488", + "accept": false, + "reason": "Only loosely tokenizer-related; the AddedToken special-argument error is not the same underlying bug as failing to load cjvt/sleng-bert." + }, + { + "left": "issue:44186", + "right": "issue:44898", + "accept": false, + "reason": "Different model classes and failure modes: LayoutLMv2 tokenizer crashes vs Perceiver image classification failing on non-default resolution." + }, + { + "left": "issue:43531", + "right": "issue:43643", + "accept": false, + "reason": "Different areas: Qwen3-MoE sliding_window behavior vs AutoConfig trust_remote_code returning missing fields." + }, + { + "left": "issue:43653", + "right": "issue:43819", + "accept": false, + "reason": "Different bugs: BigBirdTokenizer special-token registration vs DAC.from_latents mismatch with missing STE." + }, + { + "left": "issue:43643", + "right": "issue:44877", + "accept": false, + "reason": "Both involve config loading, but one is a trust_remote_code field-loss bug and the other is strict config blocking granite_speech loading." + }, + { + "left": "issue:43540", + "right": "issue:44008", + "accept": false, + "reason": "Different multimodal model bugs: Qwen3OmniMoe video processing ValueError vs Gemma 3n variable-name collision causing audio_mel_mask AttributeError." + } + ] + }, + "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": 21, + "review_unit_index": 17, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:30064", + "issue:34634", + "issue:36010", + "issue:36331", + "issue:39692", + "issue:41720", + "issue:43322", + "issue:43335", + "issue:43388", + "issue:43421", + "issue:43540", + "issue:43550", + "issue:43577", + "issue:43582", + "issue:43606", + "issue:43644", + "issue:43716", + "issue:43749", + "issue:43756", + "issue:43818", + "issue:43819", + "issue:43950", + "issue:44008", + "issue:44186", + "issue:44442", + "issue:44448", + "issue:44479", + "issue:44485", + "issue:44509", + "issue:44671", + "issue:44805", + "issue:44811", + "issue:44821", + "issue:44843", + "issue:44936", + "issue:44991", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45183", + "issue:45381", + "issue:45479", + "issue:45584", + "issue:45685", + "issue:45834", + "issue:45987", + "issue:46032", + "issue:46512" + ], + "soft_pairs": [ + "issue:43335|issue:43756", + "issue:43550|issue:43819", + "issue:30064|issue:36331", + "issue:41720|issue:44991", + "issue:44843|issue:45081", + "issue:44821|issue:45084", + "issue:43322|issue:45685", + "issue:39692|issue:45834", + "issue:43388|issue:46032", + "issue:44186|issue:44442", + "issue:45183|issue:45479", + "issue:45584|issue:45834", + "issue:43950|issue:45987", + "issue:43644|issue:43749", + "issue:44448|issue:44671", + "issue:43582|issue:46512", + "issue:43716|issue:44008", + "issue:44485|issue:44509", + "issue:43606|issue:45072", + "issue:34634|issue:36010", + "issue:44479|issue:45834", + "issue:43577|issue:43818", + "issue:44805|issue:44936", + "issue:43540|issue:45381", + "issue:43421|issue:44811" + ], + "prepared_review_unit_hash": "48c1674c00a884c64e1eea96802acdb4d4f293d90d02e9b835872194bb0b80ff", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 25, + "serialized_chars": 35594, + "estimated_input_tokens": 8899, + "estimated_eval_tokens": 18054 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "All soft-similarity links look like false positives: the issues span unrelated models, subsystems, and failure modes, so none should be merged as duplicates. The cluster is noisy and heterogeneous, with no true duplicate pair identified.", + "confidence": 0.91, + "canonical_issue_reason": "issue:45834 is the most connected item in this noisy set and is a concrete, reproducible bug report, but the cluster overall is too heterogeneous to indicate a real canonical duplicate target.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45834 is the best representative by connectivity and specificity, though it is still only a local hub in an otherwise unrelated cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43335", + "right": "issue:43756", + "accept": false, + "reason": "Different model/config bugs: sparse-layer construction vs RoPE-layer dropping." + }, + { + "left": "issue:43550", + "right": "issue:43819", + "accept": false, + "reason": "Unrelated failures in different code paths: torch.compile/SDPA vs DAC latent decoding." + }, + { + "left": "issue:30064", + "right": "issue:36331", + "accept": false, + "reason": "Completely different subsystems: image segmentation preprocessing vs Trainer loss signature." + }, + { + "left": "issue:41720", + "right": "issue:44991", + "accept": false, + "reason": "Different components and symptoms: Qwen3 device-map CUDA assert vs tokenizer loading failure." + }, + { + "left": "issue:44843", + "right": "issue:45081", + "accept": false, + "reason": "Same helper function, but distinct bugs: offline hub call vs missing backend_tokenizer attribute." + }, + { + "left": "issue:44821", + "right": "issue:45084", + "accept": false, + "reason": "No shared underlying bug; image-processor URL loading and template compilation are unrelated." + }, + { + "left": "issue:43322", + "right": "issue:45685", + "accept": false, + "reason": "Different model/runtime failures: Llava Next segfault vs MPS histogram on Int." + }, + { + "left": "issue:39692", + "right": "issue:45834", + "accept": false, + "reason": "Docs/example issues for SigLIP2 are unrelated to Kosmos2.5 OCR indexing errors." + }, + { + "left": "issue:43388", + "right": "issue:46032", + "accept": false, + "reason": "Different execution paths: metric gathering in evaluation vs Mamba2 use_cache correctness." + }, + { + "left": "issue:44186", + "right": "issue:44442", + "accept": false, + "reason": "Tokenizer/padding crash vs AutoTokenizer loading failure; no same bug." + }, + { + "left": "issue:45183", + "right": "issue:45479", + "accept": false, + "reason": "Validation error messaging and classification loss behavior are different problems." + }, + { + "left": "issue:45584", + "right": "issue:45834", + "accept": false, + "reason": "Different models and bugs: Whisper empty-transcription generation vs Kosmos2.5 OCR indexing." + }, + { + "left": "issue:43950", + "right": "issue:45987", + "accept": false, + "reason": "Both touch loading/state, but the bugs are different: buffer corruption vs cache-length return type." + }, + { + "left": "issue:43644", + "right": "issue:43749", + "accept": false, + "reason": "Generic loading regression vs FSDP CPU RAM-efficient loading; not the same concrete defect." + }, + { + "left": "issue:44448", + "right": "issue:44671", + "accept": false, + "reason": "Different model outputs in v5, but no evidence of a shared root cause." + }, + { + "left": "issue:43582", + "right": "issue:46512", + "accept": false, + "reason": "Apple Silicon warmup TypeError and stale config dtype are unrelated dtype/runtime issues." + }, + { + "left": "issue:43716", + "right": "issue:44008", + "accept": false, + "reason": "Image-preprocessor dtype mismatch and Gemma3n variable collision are unrelated bugs." + }, + { + "left": "issue:44485", + "right": "issue:44509", + "accept": false, + "reason": "RoPE implementation discussion is unrelated to stale documentation pipeline-task references." + }, + { + "left": "issue:43606", + "right": "issue:45072", + "accept": false, + "reason": "Bark CPU-offload device mismatch and SwitchTransformers/TimmWrapper dtype mismatch are different bugs." + }, + { + "left": "issue:34634", + "right": "issue:36010", + "accept": false, + "reason": "BarkProcessor voice preset handling and GenerationMixin import regression are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:45834", + "accept": false, + "reason": "Both are video/vision adjacent at most, but one is a Qwen video regression and the other is an OCR index error." + }, + { + "left": "issue:43577", + "right": "issue:43818", + "accept": false, + "reason": "Blip2 dtype loading bug and Video-LLaVA temporal-attention/weight-sharing issue are unrelated." + }, + { + "left": "issue:44805", + "right": "issue:44936", + "accept": false, + "reason": "Mask-shape IndexError and evaluate-after-train failure are not the same concrete problem." + }, + { + "left": "issue:43540", + "right": "issue:45381", + "accept": false, + "reason": "Different Qwen video bugs in different model families and code paths." + }, + { + "left": "issue:43421", + "right": "issue:44811", + "accept": false, + "reason": "Runtime post-processor updates and Whisper batch_decode special-token handling 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": 21, + "review_unit_index": 18, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:28218", + "issue:29159", + "issue:34634", + "issue:36331", + "issue:42175", + "issue:43262", + "issue:43329", + "issue:43526", + "issue:43577", + "issue:43582", + "issue:43653", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43782", + "issue:43819", + "issue:43825", + "issue:43827", + "issue:43994", + "issue:44060", + "issue:44291", + "issue:44368", + "issue:44561", + "issue:44589", + "issue:44610", + "issue:44671", + "issue:44857", + "issue:44912", + "issue:44977", + "issue:44991", + "issue:45003", + "issue:45020", + "issue:45042", + "issue:45072", + "issue:45081", + "issue:45325", + "issue:45372", + "issue:45406", + "issue:45600", + "issue:45701", + "issue:45987", + "issue:46192", + "issue:46285", + "issue:46291", + "issue:46294", + "issue:46326", + "issue:46463" + ], + "soft_pairs": [ + "issue:43825|issue:43994", + "issue:43782|issue:44977", + "issue:44991|issue:46291", + "issue:46294|issue:46463", + "issue:44561|issue:45020", + "issue:44610|issue:45072", + "issue:43720|issue:45072", + "issue:43577|issue:43582", + "issue:45987|issue:46192", + "issue:43825|issue:43827", + "issue:28218|issue:29159", + "issue:43749|issue:44291", + "issue:44671|issue:45701", + "issue:42175|issue:45081", + "issue:43582|issue:44857", + "issue:34634|issue:36331", + "issue:43782|issue:45325", + "issue:45003|issue:46285", + "issue:43262|issue:43819", + "issue:43653|issue:44912", + "issue:43526|issue:43697", + "issue:43742|issue:45020", + "issue:45081|issue:45406", + "issue:44589|issue:45042", + "issue:45372|issue:46326", + "issue:44060|issue:44368", + "issue:43329|issue:45600" + ], + "prepared_review_unit_hash": "5be4629bf838046f6696ed46ab701d02fdde0de3ddd55380165f38d7233e33c4", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 36258, + "estimated_input_tokens": 9065, + "estimated_eval_tokens": 18386 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a loose topical cluster of mostly unrelated Transformers bugs/regressions. None of the soft pairs look like the same concrete underlying issue, so I would not merge any of them.", + "confidence": 0.28, + "canonical_issue_reason": "issue:45020 is the broadest umbrella report in the set, covering a general class of recent `remote_code` breakages rather than a narrow model-specific symptom.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the best cluster anchor because it is the most general and could plausibly subsume several version-related loading regressions.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43825", + "right": "issue:43994", + "accept": false, + "reason": "Both involve pipeline/model behavior, but one is an error-message regression and the other is a model output quality issue; different bugs." + }, + { + "left": "issue:43782", + "right": "issue:44977", + "accept": false, + "reason": "Different models, different failure modes (`weight_only` load error vs flash-attention generation issue)." + }, + { + "left": "issue:44991", + "right": "issue:46291", + "accept": false, + "reason": "Tokenizer loading failure and a pipelines import dependency error are unrelated code paths." + }, + { + "left": "issue:46294", + "right": "issue:46463", + "accept": false, + "reason": "Both touch chat-template/tokenizer logic, but they affect different models and distinct behaviors." + }, + { + "left": "issue:44561", + "right": "issue:45020", + "accept": false, + "reason": "Related broad theme, but one is a specific removed helper causing remote-code failures while the other is a generic recent regression report; not the same concrete bug." + }, + { + "left": "issue:44610", + "right": "issue:45072", + "accept": false, + "reason": "Processor image-size bug vs dtype mismatch; different subsystems and symptoms." + }, + { + "left": "issue:43720", + "right": "issue:45072", + "accept": false, + "reason": "Packed-weight loading failure is unrelated to bfloat16 dtype mismatches." + }, + { + "left": "issue:43577", + "right": "issue:43582", + "accept": false, + "reason": "Blip2 dtype persistence and Apple Silicon allocator warmup TypeError are separate issues." + }, + { + "left": "issue:45987", + "right": "issue:46192", + "accept": false, + "reason": "Cache sequence-length return type bug and `return_tensors` kwargs handling are different API regressions." + }, + { + "left": "issue:43825", + "right": "issue:43827", + "accept": false, + "reason": "One is an incorrect runtime message, the other is stale documentation; related area but not the same bug." + }, + { + "left": "issue:28218", + "right": "issue:29159", + "accept": false, + "reason": "Both are tokenizer complaints, but one is about added-token spacing and the other is a broad slow/fast inconsistency report." + }, + { + "left": "issue:43749", + "right": "issue:44291", + "accept": false, + "reason": "Different loading paths and different exceptions (`FSDP` RAM-efficient loading vs `_is_hf_initialized` argument)." + }, + { + "left": "issue:44671", + "right": "issue:45701", + "accept": false, + "reason": "Masked LM prediction quality regression is not the same as a tokenizer behavior change report." + }, + { + "left": "issue:42175", + "right": "issue:45081", + "accept": false, + "reason": "Backend packaging/dependency issue vs tokenizer regex patch crash; unrelated." + }, + { + "left": "issue:43582", + "right": "issue:44857", + "accept": false, + "reason": "Apple Silicon warmup TypeError and CUDA AMP crash are different platform-specific bugs." + }, + { + "left": "issue:34634", + "right": "issue:36331", + "accept": false, + "reason": "Processor voice preset bug and trainer loss-signature mismatch are unrelated." + }, + { + "left": "issue:43782", + "right": "issue:45325", + "accept": false, + "reason": "Different Qwen VL bugs: load-time `weight_only` failure vs rope-index scaling bug." + }, + { + "left": "issue:45003", + "right": "issue:46285", + "accept": false, + "reason": "Unsafe `sys.modules` access and CLIP class structure inconsistency are different code-path problems." + }, + { + "left": "issue:43262", + "right": "issue:43819", + "accept": false, + "reason": "Audio sampling-rate default bug vs DAC latent reconstruction mismatch; not the same defect." + }, + { + "left": "issue:43653", + "right": "issue:44912", + "accept": false, + "reason": "Tokenizer special-token registration and MXFP4 quantization fallback are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:43697", + "accept": false, + "reason": "Beit label reduction bug and RT-DETR output mismatch are different model/processor regressions." + }, + { + "left": "issue:43742", + "right": "issue:45020", + "accept": false, + "reason": "A model-specific key error is not the same as the broader remote-code regression report." + }, + { + "left": "issue:45081", + "right": "issue:45406", + "accept": false, + "reason": "Both are AttributeErrors, but they occur in different components (`mistral` tokenizer patch vs Gemma4 processor)." + }, + { + "left": "issue:44589", + "right": "issue:45042", + "accept": false, + "reason": "Float8 storage lookup failure and PIL/torchvision dependency regression are unrelated." + }, + { + "left": "issue:45372", + "right": "issue:46326", + "accept": false, + "reason": "Different tokenizer/model-loading failures with different external dependency and API surfaces." + }, + { + "left": "issue:44060", + "right": "issue:44368", + "accept": false, + "reason": "Both mention tied-weights warnings, but one is a mistaken tie across parameters and the other is a training-time warning; not the same bug." + }, + { + "left": "issue:43329", + "right": "issue:45600", + "accept": false, + "reason": "Multimodal token counting bug and auto-mapping config reference cleanup 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": 21, + "review_unit_index": 19, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:36010", + "issue:36331", + "issue:38175", + "issue:39692", + "issue:42175", + "issue:42490", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43450", + "issue:43452", + "issue:43526", + "issue:43531", + "issue:43550", + "issue:43638", + "issue:43643", + "issue:43645", + "issue:43650", + "issue:43696", + "issue:43720", + "issue:43746", + "issue:43749", + "issue:43819", + "issue:43854", + "issue:43906", + "issue:43927", + "issue:43931", + "issue:44387", + "issue:44479", + "issue:44554", + "issue:44610", + "issue:44617", + "issue:44756", + "issue:44811", + "issue:44912", + "issue:44936", + "issue:44977", + "issue:44991", + "issue:45042", + "issue:45084", + "issue:45459", + "issue:45538", + "issue:45759", + "issue:45803", + "issue:46209", + "issue:46291", + "issue:46421", + "issue:46424" + ], + "soft_pairs": [ + "issue:45084|issue:45538", + "issue:43906|issue:46209", + "issue:43526|issue:44554", + "issue:43550|issue:44912", + "issue:43638|issue:44936", + "issue:43452|issue:44991", + "issue:43257|issue:43329", + "issue:43819|issue:43927", + "issue:36010|issue:43645", + "issue:36010|issue:36331", + "issue:42175|issue:46291", + "issue:43531|issue:45042", + "issue:43720|issue:44610", + "issue:38175|issue:44610", + "issue:44387|issue:44617", + "issue:43746|issue:44811", + "issue:39692|issue:44912", + "issue:42490|issue:43257", + "issue:43854|issue:44977", + "issue:43643|issue:45759", + "issue:43262|issue:44811", + "issue:45459|issue:45803", + "issue:43329|issue:44479", + "issue:46421|issue:46424", + "issue:43257|issue:43696", + "issue:43650|issue:43906", + "issue:43329|issue:43450", + "issue:43257|issue:43931", + "issue:43749|issue:44756" + ], + "prepared_review_unit_hash": "04c94ca0608427461d2c13c92fb5e6bbb25e8197671bb564217ed2a158525696", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36472, + "estimated_input_tokens": 9118, + "estimated_eval_tokens": 18492 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab-bag of unrelated Transformer issues. I found one likely true duplicate pair around the Qwen video regression; the rest look distinct and should stay separate.", + "confidence": 0.89, + "canonical_issue_reason": "Issue 44479 is the broadest and most representative report of the Qwen video-input regression across multiple model variants; it captures the user-facing failure more completely than the narrower root-cause-style report in 43329.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44479 is the best single issue to represent the only plausible duplicate set here because it is the most general, clearly scoped, and highest-impact description of the underlying video regression.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45084", + "right": "issue:45538", + "accept": false, + "reason": "Template-compilation failure and tokenizer max_length default are unrelated bugs." + }, + { + "left": "issue:43906", + "right": "issue:46209", + "accept": false, + "reason": "One is a reproduction of an older issue; the other is an FP8 inference/OOM problem on B200. Different failure modes." + }, + { + "left": "issue:43526", + "right": "issue:44554", + "accept": false, + "reason": "BEiT label reduction and MPS attention correctness are separate code paths and symptoms." + }, + { + "left": "issue:43550", + "right": "issue:44912", + "accept": false, + "reason": "torch.compile+SDPA failure in Bamba and MXFP4 quantization fallback in git-oss are different model/runtime issues." + }, + { + "left": "issue:43638", + "right": "issue:44936", + "accept": false, + "reason": "ZeRO3/non-pretrained Bert indexing error is not the same as train-then-evaluate failure." + }, + { + "left": "issue:43452", + "right": "issue:44991", + "accept": false, + "reason": "gguf_file integration breakage and tokenizer loading regression for a specific model are different problems." + }, + { + "left": "issue:43257", + "right": "issue:43329", + "accept": false, + "reason": "Qwen3 MoE weight conversion/loading and multimodal video token counting are different bugs." + }, + { + "left": "issue:43819", + "right": "issue:43927", + "accept": false, + "reason": "DAC latent reconstruction mismatch and DiaConfig custom token loss are unrelated." + }, + { + "left": "issue:36010", + "right": "issue:43645", + "accept": false, + "reason": "GenerationMixin import failure and Jupyter custom-model initialization breakage are distinct issues." + }, + { + "left": "issue:36010", + "right": "issue:36331", + "accept": false, + "reason": "ImportError for GenerationMixin and compute_loss signature regression are not the same bug." + }, + { + "left": "issue:42175", + "right": "issue:46291", + "accept": false, + "reason": "Packaging/backends selection issue is unrelated to kernels dependency/version validation on pipelines import." + }, + { + "left": "issue:43531", + "right": "issue:45042", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE and unintended torchvision requirement in PIL backend processors do not share the same root cause." + }, + { + "left": "issue:43720", + "right": "issue:44610", + "accept": false, + "reason": "Packed-weight loading in BitNet and processor/model input-size mismatch in OmDet-Turbo are different failure classes." + }, + { + "left": "issue:38175", + "right": "issue:44610", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 is unrelated to OmDet-Turbo\u2019s 640px/224px preprocessing mismatch." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Int4 quantization memory regression and Sam3Video OOM are different issues with different triggers." + }, + { + "left": "issue:43746", + "right": "issue:44811", + "accept": false, + "reason": "PEFT local checkpoint loading and Whisper batch_decode special-token handling are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:44912", + "accept": false, + "reason": "Documentation/example errors for SigLIP2 are not the same as MXFP4 quantization loading behavior." + }, + { + "left": "issue:42490", + "right": "issue:43257", + "accept": false, + "reason": "Slash-path import warning and Qwen3 MoE accelerate/deepspeed loading are unrelated." + }, + { + "left": "issue:43854", + "right": "issue:44977", + "accept": false, + "reason": "Model loading test failures for GLM-4.7-Flash and flash-attention generation problems in Qwen3.5 are distinct." + }, + { + "left": "issue:43643", + "right": "issue:45759", + "accept": false, + "reason": "Missing fields in AutoConfig with trust_remote_code and not unwrapping text_config in AutoModelForCausalLM.from_config are related to config handling but not the same concrete bug." + }, + { + "left": "issue:43262", + "right": "issue:44811", + "accept": false, + "reason": "Audio chat-template sample-rate default and Whisper decode token filtering are different processor bugs." + }, + { + "left": "issue:45459", + "right": "issue:45803", + "accept": false, + "reason": "Both involve error handling, but they affect different functions and different swallowed exceptions." + }, + { + "left": "issue:43329", + "right": "issue:44479", + "accept": true, + "reason": "Both describe the same Qwen video-input regression in the multimodal/video code path; 43329 points to the undefined helper bug, while 44479 reports the broader regression across affected model variants." + }, + { + "left": "issue:46421", + "right": "issue:46424", + "accept": false, + "reason": "Lazy cache initialization causing recompilation and max_cache_length handling for quantized cache are separate cache-related bugs." + }, + { + "left": "issue:43257", + "right": "issue:43696", + "accept": false, + "reason": "Qwen3 MoE loading/conversion and GPT-oss-20b OOM are unrelated." + }, + { + "left": "issue:43650", + "right": "issue:43906", + "accept": false, + "reason": "The placeholder issue title and an isolated reproduction of another issue do not indicate the same underlying bug." + }, + { + "left": "issue:43329", + "right": "issue:43450", + "accept": false, + "reason": "Undefined video token helpers and batched video shape mismatch are both video-related, but they are not the same concrete defect." + }, + { + "left": "issue:43257", + "right": "issue:43931", + "accept": false, + "reason": "MoE weight conversion/loading and Qwen3-VL shape mismatch are different model-loading failures." + }, + { + "left": "issue:43749", + "right": "issue:44756", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and mmap-related OOM on Strix Halo are different memory/loading issues." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The one accepted duplicate pair (43329/44479) is plausibly the same Qwen video regression, with 44479 as the broader report. The other soft-edge rejections are mostly well-justified and do not overstate duplication. No mergeability issues apply since these are issue pairs, not PR pairs." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 20, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:24643", + "issue:28282", + "issue:29127", + "issue:29159", + "issue:33357", + "issue:34567", + "issue:34689", + "issue:40990", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43317", + "issue:43404", + "issue:43425", + "issue:43452", + "issue:43454", + "issue:43525", + "issue:43540", + "issue:43577", + "issue:43582", + "issue:43606", + "issue:43697", + "issue:43742", + "issue:43749", + "issue:43824", + "issue:43927", + "issue:44062", + "issue:44291", + "issue:44479", + "issue:44560", + "issue:44568", + "issue:44821", + "issue:44849", + "issue:44912", + "issue:44987", + "issue:44991", + "issue:45042", + "issue:45127", + "issue:45137", + "issue:45356", + "issue:45357", + "issue:45381", + "issue:45405", + "issue:45538", + "issue:45701", + "issue:45859", + "issue:46291", + "issue:46512" + ], + "soft_pairs": [ + "issue:42915|issue:43742", + "issue:43606|issue:45859", + "issue:43454|issue:43577", + "issue:44291|issue:45042", + "issue:34567|issue:40990", + "issue:42915|issue:44821", + "issue:43525|issue:46291", + "issue:43927|issue:44912", + "issue:29159|issue:34567", + "issue:44568|issue:45356", + "issue:44987|issue:45538", + "issue:43697|issue:45357", + "issue:43404|issue:45127", + "issue:42915|issue:45538", + "issue:43425|issue:46291", + "issue:45137|issue:45405", + "issue:28282|issue:34689", + "issue:43577|issue:46512", + "issue:42175|issue:43824", + "issue:33357|issue:34567", + "issue:43452|issue:43582", + "issue:43317|issue:45538", + "issue:43742|issue:44291", + "issue:44849|issue:45701", + "issue:28282|issue:29127", + "issue:44062|issue:44991", + "issue:44479|issue:45381", + "issue:41720|issue:43540", + "issue:43540|issue:44479", + "issue:43540|issue:44560", + "issue:42915|issue:43749", + "issue:24643|issue:29159" + ], + "prepared_review_unit_hash": "49e0176932011d0cc61f5e5e0f5e6faeeade3aaa803f3dad32cb91ac66ded46e", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37342, + "estimated_input_tokens": 9336, + "estimated_eval_tokens": 18928 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: model loading, tokenizer, dtype, offload, and video-processing bugs are mixed together. Only the Qwen VL video-input regression pair looks like a plausible duplicate; the rest are separate issues.", + "confidence": 0.46, + "canonical_issue_reason": "issue:44479 is the best representative issue because it describes the broad 5.3.0 Qwen video-input regression across multiple affected Qwen VL variants, which is the closest thing in this cluster to a shared underlying bug.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44479 is the strongest cluster anchor: it is broad, symptom-focused, and explicitly frames the regression in the same Qwen video-input code path that the closest soft match also reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42915", + "right": "issue:43742", + "accept": false, + "reason": "Both are loading/quantization failures, but they affect different models and report different failures (FineGrainedFP8Config vs. key error on MobileLLM). No clear same bug." + }, + { + "left": "issue:43606", + "right": "issue:45859", + "accept": false, + "reason": "Both mention CPU offload, but one is a Bark-small device mismatch and the other is a Qwen3.5 MoE rotary embedding compatibility issue. Different code paths and models." + }, + { + "left": "issue:43454", + "right": "issue:43577", + "accept": false, + "reason": "One is about tying lm_head weights in AyaVision/Mistral3; the other is about BLIP2 dtype propagation. Similar theme, but not the same underlying bug." + }, + { + "left": "issue:44291", + "right": "issue:45042", + "accept": false, + "reason": "Init-empty-weights argument handling and PIL image-processor torchvision requirements are unrelated." + }, + { + "left": "issue:34567", + "right": "issue:40990", + "accept": false, + "reason": "TrainerState token accounting and perplexity on gpt-oss-20b are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:44821", + "accept": false, + "reason": "Different features and failures: FP8 config loading vs AutoImageProcessor-from-URL behavior." + }, + { + "left": "issue:43525", + "right": "issue:46291", + "accept": false, + "reason": "Llama4 pad_token_id access and kernels package import requirements are unrelated." + }, + { + "left": "issue:43927", + "right": "issue:44912", + "accept": false, + "reason": "DiaConfig token-ID persistence and MXFP4 quantization fallback are separate problems." + }, + { + "left": "issue:29159", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer slow/fast inconsistency is not the same bug as TrainerState num_input_tokens_seen not updating." + }, + { + "left": "issue:44568", + "right": "issue:45356", + "accept": false, + "reason": "Different tokenizer regressions with different models and failure modes." + }, + { + "left": "issue:44987", + "right": "issue:45538", + "accept": false, + "reason": "Loading a specific model and CLIPTokenizer max-length handling are unrelated." + }, + { + "left": "issue:43697", + "right": "issue:45357", + "accept": false, + "reason": "RTDetrV2 output drift in v5 and Qwen3.5 save_pretrained key mapping are different issues." + }, + { + "left": "issue:43404", + "right": "issue:45127", + "accept": false, + "reason": "Both involve tied embeddings, but one is a missing weight-tying bug in a model class and the other is LoRA merge behavior with extended vocab. Not the same concrete bug." + }, + { + "left": "issue:42915", + "right": "issue:45538", + "accept": false, + "reason": "FP8 loading and CLIP tokenizer model_max_length are unrelated." + }, + { + "left": "issue:43425", + "right": "issue:46291", + "accept": false, + "reason": "Torch version compatibility and kernels import requirements are unrelated." + }, + { + "left": "issue:45137", + "right": "issue:45405", + "accept": false, + "reason": "DeepSpeed ZeRO3 deque error and a PEFT minimum-version bump are unrelated." + }, + { + "left": "issue:28282", + "right": "issue:34689", + "accept": false, + "reason": "PyTorch missing from environment and Llama 3.2 90B Vision model loading regression are unrelated." + }, + { + "left": "issue:43577", + "right": "issue:46512", + "accept": false, + "reason": "Both concern dtype state, but one is BLIP2 loading behavior and the other is from_config leaving config.dtype stale. Different API paths and fixes." + }, + { + "left": "issue:42175", + "right": "issue:43824", + "accept": false, + "reason": "TensorFlow extra packaging and missing Qwen2.5-VL class import are separate packaging/import issues." + }, + { + "left": "issue:33357", + "right": "issue:34567", + "accept": false, + "reason": "MacOS bus error on CLIP model loading and TrainerState token count updates are unrelated." + }, + { + "left": "issue:43452", + "right": "issue:43582", + "accept": false, + "reason": "GGUF file loading and Apple Silicon caching allocator warmup are unrelated." + }, + { + "left": "issue:43317", + "right": "issue:45538", + "accept": false, + "reason": "Device-map/offload dequantized model loading and CLIP tokenizer max-length are unrelated." + }, + { + "left": "issue:43742", + "right": "issue:44291", + "accept": false, + "reason": "Key error on model loading and init_empty_weights argument handling are distinct problems." + }, + { + "left": "issue:44849", + "right": "issue:45701", + "accept": false, + "reason": "Qwen3.5 output_hidden_states behavior and tokenizer changes across versions are different issues." + }, + { + "left": "issue:28282", + "right": "issue:29127", + "accept": false, + "reason": "Missing PyTorch import and LayoutLMv3 box-content error are unrelated." + }, + { + "left": "issue:44062", + "right": "issue:44991", + "accept": false, + "reason": "AddedToken special-argument handling and est-roberta tokenizer loading are different tokenizer failures." + }, + { + "left": "issue:44479", + "right": "issue:45381", + "accept": true, + "reason": "Both report the same Qwen VL video-input regression in 5.3.0, with 45381 describing a specific symptom (vision_position_ids) of the broader broken video path in 44479." + }, + { + "left": "issue:41720", + "right": "issue:43540", + "accept": false, + "reason": "Qwen3 auto device mapping cudaErrorAssert and Qwen3OmniMoe video ValueError are different failure modes." + }, + { + "left": "issue:43540", + "right": "issue:44479", + "accept": false, + "reason": "Both are video-related Qwen issues, but they affect different model families and symptoms; not enough evidence they are the same bug." + }, + { + "left": "issue:43540", + "right": "issue:44560", + "accept": false, + "reason": "Qwen3OmniMoe video ValueError and Qwen3-vl-embedding StopIteration are distinct video bugs." + }, + { + "left": "issue:42915", + "right": "issue:43749", + "accept": false, + "reason": "FP8 config loading and FSDP CPU RAM efficient loading are unrelated training/loading issues." + }, + { + "left": "issue:24643", + "right": "issue:29159", + "accept": false, + "reason": "DeepSpeed weight dimensionality error and tokenizer slow/fast inconsistency are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall. The cluster summary is conservative and supported by the packet, and the only accepted soft edge (44479/45381) is plausibly the same Qwen VL 5.3.0 video-input regression. The rejected pairs are mostly clearly distinct by model, API path, or failure mode, without overstating linkage." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 21, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:27957", + "issue:28282", + "issue:34567", + "issue:36331", + "issue:38175", + "issue:42175", + "issue:43257", + "issue:43404", + "issue:43452", + "issue:43493", + "issue:43526", + "issue:43531", + "issue:43577", + "issue:43742", + "issue:43756", + "issue:43906", + "issue:43994", + "issue:44062", + "issue:44291", + "issue:44488", + "issue:44568", + "issue:44589", + "issue:44737", + "issue:44792", + "issue:44821", + "issue:44898", + "issue:44964", + "issue:44987", + "issue:45042", + "issue:45072", + "issue:45081", + "issue:45325", + "issue:45357", + "issue:45381", + "issue:45405", + "issue:45479", + "issue:45488", + "issue:45588", + "issue:45600", + "issue:45823", + "issue:45907", + "issue:45920", + "issue:45987", + "issue:46399", + "issue:46463", + "issue:46468", + "issue:46491" + ], + "soft_pairs": [ + "issue:42175|issue:44488", + "issue:38175|issue:44898", + "issue:43493|issue:43994", + "issue:27957|issue:28282", + "issue:44987|issue:45907", + "issue:44568|issue:45488", + "issue:44291|issue:44821", + "issue:34567|issue:36331", + "issue:44737|issue:46463", + "issue:43257|issue:43756", + "issue:43577|issue:45072", + "issue:44792|issue:44964", + "issue:44062|issue:45081", + "issue:43742|issue:44589", + "issue:45042|issue:45081", + "issue:44737|issue:45588", + "issue:44568|issue:46491", + "issue:43906|issue:46468", + "issue:43756|issue:46399", + "issue:43756|issue:45325", + "issue:43526|issue:45479", + "issue:43404|issue:45987", + "issue:45823|issue:46463", + "issue:43531|issue:44821", + "issue:45405|issue:45600", + "issue:43452|issue:44488", + "issue:43906|issue:45381", + "issue:45357|issue:45920" + ], + "prepared_review_unit_hash": "6db46b3076b74b6d63af50fa7eff74ddab4c13bf557f7e23ca1c374f5f4dd3fe", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35951, + "estimated_input_tokens": 8988, + "estimated_eval_tokens": 18232 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most proposed soft links are false positives: they share a library area or model family, but describe different failures and code paths. The one strong duplicate candidate is the pair of SigLIP2 reports, which both point to the same model producing invalid outputs. For a representative issue, the broad multi-model tokenizer regression is the best anchor.", + "confidence": 0.76, + "canonical_issue_reason": "issue:45920 is the broadest, most user-impacting regression in the set: it describes a v5 tokenizer bug affecting multiple model families, so it is the best representative issue for triage and duplicate collection.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45920 has the widest scope and clearest cross-model impact, making it the strongest single issue to keep visible as the cluster anchor.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42175", + "right": "issue:44488", + "accept": false, + "reason": "Different problems: packaging/backend availability versus a specific model load failure." + }, + { + "left": "issue:38175", + "right": "issue:44898", + "accept": false, + "reason": "Both are vision-model bugs, but they concern different models and different failure modes." + }, + { + "left": "issue:43493", + "right": "issue:43994", + "accept": true, + "reason": "Both describe the same SigLIP2 model producing incorrect/nonsensical outputs, likely the same underlying implementation bug." + }, + { + "left": "issue:27957", + "right": "issue:28282", + "accept": false, + "reason": "Unrelated: Flash Attention support for XLM-Roberta versus a missing PyTorch installation error." + }, + { + "left": "issue:44987", + "right": "issue:45907", + "accept": false, + "reason": "One is a model-loading regression; the other is a repo-file listing API forwarding bug in CI." + }, + { + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different tokenizer implementations and different behaviors." + }, + { + "left": "issue:44291", + "right": "issue:44821", + "accept": false, + "reason": "Different subsystems: init_empty_weights loading error versus loading an image processor from URL." + }, + { + "left": "issue:34567", + "right": "issue:36331", + "accept": false, + "reason": "Both touch trainer internals, but one is token-count tracking and the other is a breaking compute_loss signature change." + }, + { + "left": "issue:44737", + "right": "issue:46463", + "accept": false, + "reason": "Completely different code paths: XLNet positional encoding on CPU versus a T5 error-message bug." + }, + { + "left": "issue:43257", + "right": "issue:43756", + "accept": false, + "reason": "Different model-specific bugs: MoE weight conversion versus RoPE layer dropping." + }, + { + "left": "issue:43577", + "right": "issue:45072", + "accept": false, + "reason": "Both are dtype-related, but they affect different models and likely different forward paths." + }, + { + "left": "issue:44792", + "right": "issue:44964", + "accept": false, + "reason": "Different multimodal model failures with no clear shared code path." + }, + { + "left": "issue:44062", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer construction crash versus Mistral regex patching crash; both are tokenizer-adjacent but distinct." + }, + { + "left": "issue:43742", + "right": "issue:44589", + "accept": false, + "reason": "Different loading failures: a key error for one model versus a Float8 storage type lookup error." + }, + { + "left": "issue:45042", + "right": "issue:45081", + "accept": false, + "reason": "Image-processor dependency issue versus tokenizer regex patch crash." + }, + { + "left": "issue:44737", + "right": "issue:45588", + "accept": false, + "reason": "Different runtime failures in unrelated modules." + }, + { + "left": "issue:44568", + "right": "issue:46491", + "accept": false, + "reason": "Both involve tokenizers, but one is special-token insertion and the other is encode/decode whitespace regression." + }, + { + "left": "issue:43906", + "right": "issue:46468", + "accept": false, + "reason": "Too little evidence they are the same bug; one is an isolated repro and the other is a broken example." + }, + { + "left": "issue:43756", + "right": "issue:46399", + "accept": false, + "reason": "Different model families and different failure modes." + }, + { + "left": "issue:43756", + "right": "issue:45325", + "accept": false, + "reason": "Both are Qwen-family, but one is a RoPE-layer configuration issue and the other is a position-id scaling bug." + }, + { + "left": "issue:43526", + "right": "issue:45479", + "accept": false, + "reason": "Image-label reduction bug versus a sequence-classification loss edge case." + }, + { + "left": "issue:43404", + "right": "issue:45987", + "accept": false, + "reason": "Different core issues: weight tying in Mistral3 versus cache sequence-length return type." + }, + { + "left": "issue:45823", + "right": "issue:46463", + "accept": false, + "reason": "Device placement bug versus T5 validation-message bug." + }, + { + "left": "issue:43531", + "right": "issue:44821", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior and image-processor loading are unrelated." + }, + { + "left": "issue:45405", + "right": "issue:45600", + "accept": false, + "reason": "One is a dependency version pinning problem; the other is stale auto-mapping references." + }, + { + "left": "issue:43452", + "right": "issue:44488", + "accept": false, + "reason": "Both involve model loading, but the reported failures and likely root causes are different." + }, + { + "left": "issue:43906", + "right": "issue:45381", + "accept": false, + "reason": "Both mention Qwen2.5-VL-like behavior only indirectly; the described bugs are not the same code-path problem." + }, + { + "left": "issue:45357", + "right": "issue:45920", + "accept": false, + "reason": "Different areas: save_pretrained visual-encoder key serialization versus AutoTokenizer token-id regression." + }, + { + "left": "issue:46399", + "right": "issue:45920", + "accept": false, + "reason": "Completely different regressions: kernel registration failure versus tokenizer token-id corruption." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall. The canonical anchor at issue:45920 is supported by its broad multi-model tokenizer regression scope, and the soft-edge verdicts stay mostly conservative. The SigLIP2 pair is the only accepted duplicate candidate, and that linkage is reasonable from the titles without overstating evidence. " + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 22, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:24643", + "issue:28282", + "issue:29127", + "issue:32090", + "issue:35141", + "issue:38175", + "issue:41762", + "issue:42898", + "issue:43295", + "issue:43452", + "issue:43531", + "issue:43577", + "issue:43643", + "issue:43653", + "issue:43688", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43819", + "issue:44361", + "issue:44442", + "issue:44466", + "issue:44493", + "issue:44514", + "issue:44625", + "issue:44811", + "issue:44843", + "issue:44877", + "issue:44898", + "issue:44913", + "issue:44936", + "issue:45005", + "issue:45200", + "issue:45216", + "issue:45356", + "issue:45372", + "issue:45375", + "issue:45381", + "issue:45542", + "issue:45823", + "issue:46050", + "issue:46121", + "issue:46170", + "issue:46178", + "issue:46463", + "issue:46491" + ], + "soft_pairs": [ + "issue:44625|issue:44877", + "issue:44913|issue:46121", + "issue:43720|issue:44898", + "issue:43643|issue:45375", + "issue:43697|issue:45216", + "issue:24643|issue:29127", + "issue:28282|issue:32090", + "issue:41762|issue:44936", + "issue:45200|issue:46463", + "issue:28282|issue:35141", + "issue:44493|issue:46050", + "issue:44466|issue:45005", + "issue:43819|issue:44811", + "issue:43452|issue:44843", + "issue:44361|issue:44442", + "issue:43688|issue:43749", + "issue:38175|issue:43295", + "issue:44936|issue:45823", + "issue:43653|issue:44361", + "issue:45372|issue:46170", + "issue:45542|issue:46178", + "issue:43452|issue:43577", + "issue:42898|issue:46491", + "issue:44514|issue:45381", + "issue:43531|issue:43742", + "issue:43295|issue:45356" + ], + "prepared_review_unit_hash": "e9d91b57b93055788946e68e20acc8e35079039326d70e659168744ebb0e909c", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35351, + "estimated_input_tokens": 8838, + "estimated_eval_tokens": 17932 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The proposed soft edges are mostly superficial theme overlaps (configs, tokenizers, tied weights, loading paths) rather than the same concrete bug. I would reject all of them; this set is too heterogeneous to form a single duplicate cluster.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue fits: the items span unrelated subsystems and failure modes, so there isn\u2019t one underlying bug to anchor the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "No global best issue: the cluster is heterogeneous, so any one issue would be a misleading representative of the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44625", + "right": "issue:44877", + "accept": false, + "reason": "Both are config-related, but one is Qwen3.5 num_labels propagation and the other is strict loading for granite_speech; different models and code paths." + }, + { + "left": "issue:44913", + "right": "issue:46121", + "accept": false, + "reason": "Both mention rope/config handling, but one is a persisted config field regression and the other is a TypeError in rope validation for JSON-loaded lists." + }, + { + "left": "issue:43720", + "right": "issue:44898", + "accept": false, + "reason": "Different subsystems entirely: BitNet accelerate loading vs Perceiver image classification resolution handling." + }, + { + "left": "issue:43643", + "right": "issue:45375", + "accept": false, + "reason": "Both concern missing config fields, but one is generic AutoConfig with trust_remote_code and the other is a Qwen3_5MoeVision strict-field issue; not the same bug." + }, + { + "left": "issue:43697", + "right": "issue:45216", + "accept": false, + "reason": "Different regressions: RTDetrV2 output changes vs Qwen3.5 checkpoint save_pretrained corruption." + }, + { + "left": "issue:24643", + "right": "issue:29127", + "accept": false, + "reason": "DeepSpeed training runtime error vs LayoutLMv3 error-message clarity; unrelated failures." + }, + { + "left": "issue:28282", + "right": "issue:32090", + "accept": false, + "reason": "PyTorch import detection vs Trainer broadcast TypeError; unrelated." + }, + { + "left": "issue:41762", + "right": "issue:44936", + "accept": false, + "reason": "Gemma3 loading with ZeRO-3 vs trainer.evaluate after trainer.train; different code paths." + }, + { + "left": "issue:45200", + "right": "issue:46463", + "accept": false, + "reason": "Gemma 4 multimodal token defaults vs T5 decoder input validation messaging; not the same underlying bug." + }, + { + "left": "issue:28282", + "right": "issue:35141", + "accept": false, + "reason": "ImportError for missing PyTorch vs tied-weights reinitialization on resize; unrelated." + }, + { + "left": "issue:44493", + "right": "issue:46050", + "accept": false, + "reason": "Unexpected position-id keys vs sinq quantization save/load breakage; different bug classes." + }, + { + "left": "issue:44466", + "right": "issue:45005", + "accept": false, + "reason": "Both involve tied weights, but one is device-dependent lm_head serialization and the other is translation-model tied-weight issues in v5; too broad to merge." + }, + { + "left": "issue:43819", + "right": "issue:44811", + "accept": false, + "reason": "DAC latent conversion mismatch vs Whisper batch_decode skip_special_tokens behavior; unrelated." + }, + { + "left": "issue:43452", + "right": "issue:44843", + "accept": false, + "reason": "Both involve AutoTokenizer loading, but one is gguf_file support breaking tokenizer/model loads and the other is an offline model_info call in mistral regex patching." + }, + { + "left": "issue:44361", + "right": "issue:44442", + "accept": false, + "reason": "Different tokenizer failures on different tokenizers; not the same concrete bug." + }, + { + "left": "issue:43688", + "right": "issue:43749", + "accept": false, + "reason": "Auxiliary loss normalization in MoE models vs FSDP CPU RAM efficient loading breakage." + }, + { + "left": "issue:38175", + "right": "issue:43295", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 vs processor.tokenizer/image-passing regression; unrelated." + }, + { + "left": "issue:44936", + "right": "issue:45823", + "accept": false, + "reason": "Trainer lifecycle failure vs Gemma4 PLE device mismatch during forward; different issues." + }, + { + "left": "issue:43653", + "right": "issue:44361", + "accept": false, + "reason": "BigBird special-token decode bug vs MLukeTokenizer AttributeError; unrelated tokenizers and failure modes." + }, + { + "left": "issue:45372", + "right": "issue:46170", + "accept": false, + "reason": "Gemma 4 processor import dependency vs Windows ROCm torch.distributed dependency problem; unrelated." + }, + { + "left": "issue:45542", + "right": "issue:46178", + "accept": false, + "reason": "Backend registration/availability issues, but one is tensorflow backend undefined and the other is tensorflow_text not present in BACKENDS_MAPPING; different concrete bugs." + }, + { + "left": "issue:43452", + "right": "issue:43577", + "accept": false, + "reason": "Tokenizer/model loading with gguf_file vs BLIP2 dtype propagation; unrelated." + }, + { + "left": "issue:42898", + "right": "issue:46491", + "accept": false, + "reason": "General v5 cleanup-space behavior change vs CodeLlama leading-space round-trip regression; related area but not the same underlying bug." + }, + { + "left": "issue:44514", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen2.5-VL issues, but one is batched chat-template padding crash and the other is incorrect video vision_position_ids; different code paths." + }, + { + "left": "issue:43531", + "right": "issue:43742", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior vs MobileLLM-125M loading key error; unrelated." + }, + { + "left": "issue:43295", + "right": "issue:45356", + "accept": false, + "reason": "Both are regressions in tokenizer/processor behavior, but one is processor.tokenizer/image handling and the other is Kimi-K2.5 codec handling plus a warning; 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": 21, + "review_unit_index": 23, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:29159", + "issue:32090", + "issue:33357", + "issue:33666", + "issue:36010", + "issue:36331", + "issue:41720", + "issue:43504", + "issue:43550", + "issue:43577", + "issue:43643", + "issue:43646", + "issue:43653", + "issue:43688", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43756", + "issue:43927", + "issue:44361", + "issue:44464", + "issue:44466", + "issue:44479", + "issue:44484", + "issue:44514", + "issue:44554", + "issue:44561", + "issue:44568", + "issue:44610", + "issue:44625", + "issue:44704", + "issue:44734", + "issue:44743", + "issue:44821", + "issue:44843", + "issue:44977", + "issue:44991", + "issue:45290", + "issue:45440", + "issue:45491", + "issue:45701", + "issue:45704", + "issue:46002", + "issue:46050", + "issue:46402", + "issue:46612" + ], + "soft_pairs": [ + "issue:44484|issue:44704", + "issue:43577|issue:43697", + "issue:43646|issue:43749", + "issue:43653|issue:44568", + "issue:44464|issue:44743", + "issue:44843|issue:44991", + "issue:43688|issue:45704", + "issue:29159|issue:33357", + "issue:43653|issue:43927", + "issue:43756|issue:46402", + "issue:43688|issue:45440", + "issue:44479|issue:44625", + "issue:43688|issue:45701", + "issue:43550|issue:44361", + "issue:44734|issue:46612", + "issue:45491|issue:46002", + "issue:33666|issue:36010", + "issue:44464|issue:44977", + "issue:32090|issue:33666", + "issue:43742|issue:44821", + "issue:41720|issue:44464", + "issue:44466|issue:46050", + "issue:43504|issue:43720", + "issue:43504|issue:43550", + "issue:43643|issue:44561", + "issue:44514|issue:45290", + "issue:43577|issue:44554", + "issue:43504|issue:44610", + "issue:43720|issue:43927", + "issue:33666|issue:36331" + ], + "prepared_review_unit_hash": "3eaf698472fa23c64a85e6a6cdda0083d8b5063ddbe338db1cfdf85440e1b027", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36228, + "estimated_input_tokens": 9057, + "estimated_eval_tokens": 18370 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: the items span unrelated tokenizer, loading, generation, training, serialization, and model-specific regression reports. Every soft pair is only lexically similar and should not be merged as duplicates.", + "confidence": 0.88, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44484", + "right": "issue:44704", + "accept": false, + "reason": "max_shard_size/save_pretrained behavior is unrelated to AutoProcessor kwargs forwarding to cached_file." + }, + { + "left": "issue:43577", + "right": "issue:43697", + "accept": false, + "reason": "Blip2 dtype loading bug and RTDetr output divergence affect different model paths and failure modes." + }, + { + "left": "issue:43646", + "right": "issue:43749", + "accept": false, + "reason": "trust_remote_code config object field loss is a different issue from FSDP CPU RAM efficient loading." + }, + { + "left": "issue:43653", + "right": "issue:44568", + "accept": false, + "reason": "Both involve tokenizers, but one is special-token registration and the other is BOS/EOS insertion with add_special_tokens." + }, + { + "left": "issue:44464", + "right": "issue:44743", + "accept": false, + "reason": "Compiled chunked generation inconsistency is not the same as recurrent-state reset with cache in modular_qwen3_5.py." + }, + { + "left": "issue:44843", + "right": "issue:44991", + "accept": false, + "reason": "Offline model_info patching in AutoTokenizer is a different failure than tokenizer loading for est-roberta." + }, + { + "left": "issue:43688", + "right": "issue:45704", + "accept": false, + "reason": "Auxiliary loss normalization is unrelated to the apex FusedRMSNorm memory leak." + }, + { + "left": "issue:29159", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer slow/fast inconsistency is not the same as a MacOS bus error loading a CLIP model." + }, + { + "left": "issue:43653", + "right": "issue:43927", + "accept": false, + "reason": "Special-token registration in BigBirdTokenizer and custom token ID persistence in DiaConfig are distinct bugs." + }, + { + "left": "issue:43756", + "right": "issue:46402", + "accept": false, + "reason": "Smollm3 RoPE layer mismatch and Sam3 text-encoder weight loading are unrelated model-specific problems." + }, + { + "left": "issue:43688", + "right": "issue:45440", + "accept": false, + "reason": "Aux loss normalization does not match the DeepseekV3MoE divergence from remote implementation." + }, + { + "left": "issue:44479", + "right": "issue:44625", + "accept": false, + "reason": "Video-input regression in Qwen VL models is unrelated to num_labels propagation in Qwen3.5 config." + }, + { + "left": "issue:43688", + "right": "issue:45701", + "accept": false, + "reason": "Incorrect auxiliary-loss normalization and tokenization-version differences are different bugs." + }, + { + "left": "issue:43550", + "right": "issue:44361", + "accept": false, + "reason": "torch.compile/SDPA failure in Bamba is not the same as MLukeTokenizer AttributeError on tasks." + }, + { + "left": "issue:44734", + "right": "issue:46612", + "accept": false, + "reason": "Serve KV-cache continuation indexing and beam-search cache reorder omission are different generation code paths." + }, + { + "left": "issue:45491", + "right": "issue:46002", + "accept": false, + "reason": "NaNs from sliding-window attention with mixed-length batches are not the same as NaNs in classification heads after checkout/init." + }, + { + "left": "issue:33666", + "right": "issue:36010", + "accept": false, + "reason": "Qwen2-VL multi-GPU training and the GenerationMixin import error are unrelated failures." + }, + { + "left": "issue:44464", + "right": "issue:44977", + "accept": false, + "reason": "Compiled-forward chunked generation inconsistency is not the same as flash-attention regression in Qwen3.5." + }, + { + "left": "issue:32090", + "right": "issue:33666", + "accept": false, + "reason": "NoneType GPU broadcast in Trainer is a different problem from Qwen2-VL multi-GPU training." + }, + { + "left": "issue:43742", + "right": "issue:44821", + "accept": false, + "reason": "Model loading key error and AutoImageProcessor-from-URL failure involve different loaders." + }, + { + "left": "issue:41720", + "right": "issue:44464", + "accept": false, + "reason": "Qwen3 auto device mapping cuda assert is unrelated to compiled-forward chunked generation output differences." + }, + { + "left": "issue:44466", + "right": "issue:46050", + "accept": false, + "reason": "Inconsistent lm_head.weight serialization is not the same as sinq quantization breaking save/load." + }, + { + "left": "issue:43504", + "right": "issue:43720", + "accept": false, + "reason": "Beit legacy-field loading and BitNet packed-weight unpacking are distinct model-loading defects." + }, + { + "left": "issue:43504", + "right": "issue:43550", + "accept": false, + "reason": "A Beit preset-loading regression is not the same as Bamba torch.compile with SDPA." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "Missing config fields with trust_remote_code and removal of is_torch_fx_available are separate v5 compatibility breaks." + }, + { + "left": "issue:44514", + "right": "issue:45290", + "accept": false, + "reason": "Batched apply_chat_template padding behavior and assistant tool-call/no-content crashes are close but not the same bug." + }, + { + "left": "issue:43504", + "right": "issue:44610", + "accept": false, + "reason": "Legacy-field model loading and processor/model image-size mismatch are unrelated." + }, + { + "left": "issue:43720", + "right": "issue:43927", + "accept": false, + "reason": "Packed-weight accelerate loading and DiaConfig custom token ID persistence are different issues." + }, + { + "left": "issue:33666", + "right": "issue:36331", + "accept": false, + "reason": "Qwen2-VL multi-GPU training and CustomTrainer.compute_loss signature mismatch are unrelated trainer problems." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 24, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:28218", + "issue:33357", + "issue:33666", + "issue:39692", + "issue:43257", + "issue:43262", + "issue:43278", + "issue:43295", + "issue:43299", + "issue:43334", + "issue:43452", + "issue:43493", + "issue:43525", + "issue:43540", + "issue:43572", + "issue:43575", + "issue:43606", + "issue:43646", + "issue:43688", + "issue:43701", + "issue:43720", + "issue:43825", + "issue:43844", + "issue:43927", + "issue:43994", + "issue:44361", + "issue:44451", + "issue:44466", + "issue:44488", + "issue:44554", + "issue:44568", + "issue:44589", + "issue:44625", + "issue:44792", + "issue:44843", + "issue:44849", + "issue:44964", + "issue:44977", + "issue:45357", + "issue:45405", + "issue:45520", + "issue:45584", + "issue:45823", + "issue:46050", + "issue:46192", + "issue:46209", + "issue:46365", + "issue:46489" + ], + "soft_pairs": [ + "issue:43262|issue:43295", + "issue:43606|issue:43927", + "issue:44554|issue:46365", + "issue:44977|issue:46209", + "issue:43525|issue:43572", + "issue:43278|issue:43688", + "issue:43720|issue:44361", + "issue:44964|issue:45405", + "issue:39692|issue:43493", + "issue:43452|issue:44451", + "issue:43493|issue:43825", + "issue:43994|issue:46192", + "issue:45823|issue:46050", + "issue:44625|issue:45584", + "issue:28218|issue:33357", + "issue:44849|issue:46209", + "issue:39692|issue:43295", + "issue:43278|issue:43844", + "issue:43493|issue:43844", + "issue:43646|issue:44466", + "issue:33357|issue:33666", + "issue:44488|issue:44843", + "issue:44849|issue:45357", + "issue:44589|issue:45520", + "issue:43278|issue:43701", + "issue:43540|issue:44977", + "issue:44568|issue:46489", + "issue:43257|issue:43299", + "issue:43575|issue:45405", + "issue:43927|issue:44361", + "issue:43334|issue:43572", + "issue:44792|issue:45405" + ], + "prepared_review_unit_hash": "71e258f99a75fccbe373e919d918b44ea02edaf99f725c9505d1c26c61ef5d64", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37286, + "estimated_input_tokens": 9322, + "estimated_eval_tokens": 18900 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is dominated by unrelated transformers regressions, tokenizer/config bugs, and model-specific loading issues. None of the soft pairs look like the same underlying bug or change, so all soft edges should be rejected. The most representative issue is the broad v5 custom model initialization regression.", + "confidence": 0.39, + "canonical_issue_reason": "issue:43646 is the broadest, most central regression report in the set: it targets Transformers 5.0.0 breaking custom model initialization, which is a wider umbrella than the more specialized tokenizer/config/model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 is the best general representative because it captures a high-level v5 regression affecting custom model init rather than a narrow model-specific symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43262", + "right": "issue:43295", + "accept": false, + "reason": "Audio processor chat-template sampling-rate bug and processor/tokenizer regression are different code paths and symptoms." + }, + { + "left": "issue:43606", + "right": "issue:43927", + "accept": false, + "reason": "CPU offload device mismatch for bark-small is unrelated to DiaConfig losing custom token IDs on save/load." + }, + { + "left": "issue:44554", + "right": "issue:46365", + "accept": false, + "reason": "MPS attention correctness issue is unrelated to importing Trainer breaking because datasets is installed." + }, + { + "left": "issue:44977", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug and FP8 inference-on-B200 issue are distinct backend/hardware problems." + }, + { + "left": "issue:43525", + "right": "issue:43572", + "accept": false, + "reason": "Llama4Config pad_token_id and StableLmConfig pad_token_idx are similar attribute-missing reports but for different configs and fixes." + }, + { + "left": "issue:43278", + "right": "issue:43688", + "accept": false, + "reason": "BF16/FP32 embedding dtype drift in evaluation is unrelated to auxiliary-loss normalization in MoE models." + }, + { + "left": "issue:43720", + "right": "issue:44361", + "accept": false, + "reason": "BitNet packed-weight unpacking during accelerate loading is a different failure from MLukeTokenizer task AttributeError." + }, + { + "left": "issue:44964", + "right": "issue:45405", + "accept": false, + "reason": "Phi-4 multimodal loading and PEFT minimum-version bump are unrelated; one is a model compatibility issue, the other a dependency constraint issue." + }, + { + "left": "issue:39692", + "right": "issue:43493", + "accept": false, + "reason": "SigLIP2 doc/example errors and SigLIP2 implementation discrepancy are both about the same model family but describe different problems and would not be merged as one fix." + }, + { + "left": "issue:43452", + "right": "issue:44451", + "accept": false, + "reason": "gguf_file breaks tokenizer/model loading is unrelated to inability to load a specific model checkpoint." + }, + { + "left": "issue:43493", + "right": "issue:43825", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and pipeline translation-task error message are unrelated subsystems." + }, + { + "left": "issue:43994", + "right": "issue:46192", + "accept": false, + "reason": "SigLIP2 nonsensical outputs and return_tensors being ignored are different behavioral bugs with different APIs." + }, + { + "left": "issue:45823", + "right": "issue:46050", + "accept": false, + "reason": "Gemma4 PLE device mismatch and sinq quantization save/load breakage are different model/runtime issues." + }, + { + "left": "issue:44625", + "right": "issue:45584", + "accept": false, + "reason": "Qwen3.5 num_labels propagation and Whisper empty-transcription failure are unrelated model-specific bugs." + }, + { + "left": "issue:28218", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer extra-space behavior and MacOS bus error with a CLIP model are completely unrelated." + }, + { + "left": "issue:44849", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3.5 hidden-state bug and FP8 B200 inference bug are different execution-path and hardware issues." + }, + { + "left": "issue:39692", + "right": "issue:43295", + "accept": false, + "reason": "SigLIP2 doc/example failures and custom model/tokenizer regression are different models and different bug classes." + }, + { + "left": "issue:43278", + "right": "issue:43844", + "accept": false, + "reason": "Embedding dtype mismatch and ZeRO-3 gradient growth are unrelated training behaviors." + }, + { + "left": "issue:43493", + "right": "issue:43844", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and ZeRO-3 gradient inflation do not share the same underlying bug." + }, + { + "left": "issue:43646", + "right": "issue:44466", + "accept": false, + "reason": "Custom model initialization regression and lm_head serialization inconsistency are different issues in different parts of the stack." + }, + { + "left": "issue:33357", + "right": "issue:33666", + "accept": false, + "reason": "MacOS bus error in a pretrained CLIP model and Qwen2-VL multi-GPU training are unrelated." + }, + { + "left": "issue:44488", + "right": "issue:44843", + "accept": false, + "reason": "Model loading failure for a specific checkpoint and unconditional hub model_info() calls in offline mode are different bugs." + }, + { + "left": "issue:44849", + "right": "issue:45357", + "accept": false, + "reason": "Qwen3.5 hidden-state output bug and incorrect visual encoder key serialization are separate issues." + }, + { + "left": "issue:44589", + "right": "issue:45520", + "accept": false, + "reason": "Float8 storage lookup failure and flash_attn import KeyError on Python 3.13 are different environment/runtime compatibility problems." + }, + { + "left": "issue:43278", + "right": "issue:43701", + "accept": false, + "reason": "Embedding dtype change and resume_from_checkpoint key mismatch are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44977", + "accept": false, + "reason": "Video-processing ValueError in Qwen3OmniMoe and Qwen3.5 flash-attention generation bug are different modalities and failure modes." + }, + { + "left": "issue:44568", + "right": "issue:46489", + "accept": false, + "reason": "mdeberta-v3 add_special_tokens behavior and DeepSeek-Coder tokenizer output regression are tokenizer bugs, but not the same concrete code-path problem." + }, + { + "left": "issue:43257", + "right": "issue:43299", + "accept": false, + "reason": "Qwen3 MOE accelerate+deepspeed weight conversion and Qwen3VL MOE loading breakage are different loading failures with different root causes." + }, + { + "left": "issue:43575", + "right": "issue:45405", + "accept": false, + "reason": "Tensor-parallel OOM on Qwen2-57B and PEFT version pinning are unrelated." + }, + { + "left": "issue:43927", + "right": "issue:44361", + "accept": false, + "reason": "DiaConfig token-ID persistence bug and MLukeTokenizer task AttributeError are unrelated config/tokenizer failures." + }, + { + "left": "issue:43334", + "right": "issue:43572", + "accept": false, + "reason": "Qwen3-VL pad_token_id missing and StableLmConfig pad_token_idx missing are similar at a high level, but they affect different config classes and likely require different fixes." + }, + { + "left": "issue:44792", + "right": "issue:45405", + "accept": false, + "reason": "Janus image-generation test failure and PEFT minimum-version bump 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": 21, + "review_unit_index": 25, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:15354", + "issue:28218", + "issue:29159", + "issue:29942", + "issue:30333", + "issue:32090", + "issue:34567", + "issue:38175", + "issue:42490", + "issue:43065", + "issue:43334", + "issue:43493", + "issue:43504", + "issue:43572", + "issue:43606", + "issue:43653", + "issue:43696", + "issue:43701", + "issue:43824", + "issue:43844", + "issue:43873", + "issue:43994", + "issue:44112", + "issue:44291", + "issue:44423", + "issue:44464", + "issue:44493", + "issue:44554", + "issue:44568", + "issue:44589", + "issue:44734", + "issue:44964", + "issue:45084", + "issue:45200", + "issue:45229", + "issue:45406", + "issue:45446", + "issue:45520", + "issue:45584", + "issue:45588", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:46050", + "issue:46121", + "issue:46285", + "issue:46369", + "issue:46424" + ], + "soft_pairs": [ + "issue:43504|issue:43606", + "issue:44568|issue:45812", + "issue:43334|issue:45759", + "issue:43824|issue:44589", + "issue:43606|issue:44112", + "issue:43572|issue:44291", + "issue:44464|issue:46424", + "issue:45200|issue:45229", + "issue:43504|issue:43653", + "issue:44493|issue:46285", + "issue:44464|issue:45584", + "issue:45229|issue:45823", + "issue:43994|issue:44493", + "issue:43873|issue:46050", + "issue:43844|issue:44964", + "issue:45588|issue:46050", + "issue:29942|issue:30333", + "issue:29159|issue:32090", + "issue:42490|issue:43696", + "issue:43065|issue:46369", + "issue:45084|issue:45520", + "issue:44423|issue:44734", + "issue:44554|issue:45446", + "issue:28218|issue:34567", + "issue:15354|issue:38175", + "issue:45406|issue:46121", + "issue:43493|issue:43701" + ], + "prepared_review_unit_hash": "4f0a869316be7f304e589f0ca809cffe1b2497d649025f73cd93becb06b62ab0", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35991, + "estimated_input_tokens": 8998, + "estimated_eval_tokens": 18252 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly unrelated issue reports. None of the soft pairs look like the same underlying bug/change, though a few are in adjacent tokenizer, multimodal, or quantization areas.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43504", + "right": "issue:43606", + "accept": false, + "reason": "Different bugs: a legacy-field model loading failure versus a CPU offload device-mismatch on a specific model." + }, + { + "left": "issue:44568", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but one is mDeBERTa special-token behavior and the other is Granite token-ID handling; not the same code path." + }, + { + "left": "issue:43334", + "right": "issue:45759", + "accept": false, + "reason": "Related Qwen multimodal area, but one is a missing pad_token_id on load and the other is AutoModelForCausalLM config unwrapping; different fixes." + }, + { + "left": "issue:43824", + "right": "issue:44589", + "accept": false, + "reason": "Completely different failures: an import-name error for Qwen2.5-VL versus a missing Float8 storage type." + }, + { + "left": "issue:43606", + "right": "issue:44112", + "accept": false, + "reason": "One is a model/offload runtime mismatch; the other is a stale CI test in GraniteSpeech." + }, + { + "left": "issue:43572", + "right": "issue:44291", + "accept": false, + "reason": "Both involve model loading regressions after v5.0, but the symptoms and code paths differ: StableLm config field vs init_empty_weights argument handling." + }, + { + "left": "issue:44464", + "right": "issue:46424", + "accept": false, + "reason": "Both mention caching/generation, but one is compiled-forward chunked generation mismatch and the other is max_cache_length handling for quantized cache only." + }, + { + "left": "issue:45200", + "right": "issue:45229", + "accept": false, + "reason": "Different Gemma4 problems: missing default token-type IDs for text-only fine-tuning versus multi-GPU CUDA OOM." + }, + { + "left": "issue:43504", + "right": "issue:43653", + "accept": false, + "reason": "One is a pretrained-model load bug; the other is a tokenizer special-token registration bug with empty decode output." + }, + { + "left": "issue:44493", + "right": "issue:46285", + "accept": false, + "reason": "Different regressions: a position-id key mismatch across many models versus a CLIP text-model structure inconsistency." + }, + { + "left": "issue:44464", + "right": "issue:45584", + "accept": false, + "reason": "Different generation issues: compiled-forward inconsistency versus Whisper empty-transcription failure after align_special_tokens." + }, + { + "left": "issue:45229", + "right": "issue:45823", + "accept": false, + "reason": "One reports OOM, the other a device mismatch during forward; not the same underlying bug." + }, + { + "left": "issue:43994", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 wrong outputs and a position-id state-dict/key issue are not the same concrete failure." + }, + { + "left": "issue:43873", + "right": "issue:46050", + "accept": false, + "reason": "Both touch quantization/offloading, but one is offloading behavior and the other is broken save/load for sinq-quantized models." + }, + { + "left": "issue:43844", + "right": "issue:44964", + "accept": false, + "reason": "Different problems: ZeRO-3 gradient growth versus a multimodal model loading failure." + }, + { + "left": "issue:45588", + "right": "issue:46050", + "accept": false, + "reason": "Different subsystems: flash_attention.py crashing on s_aux=None versus quantized model persistence after save_pretrained/from_pretrained." + }, + { + "left": "issue:29942", + "right": "issue:30333", + "accept": false, + "reason": "A Flash Attention 2 test failure and an MLFlow reporting issue are unrelated." + }, + { + "left": "issue:29159", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer slow/fast inconsistency is unrelated to a Trainer GPU broadcast TypeError." + }, + { + "left": "issue:42490", + "right": "issue:43696", + "accept": false, + "reason": "A local-path import warning is unrelated to a GPT-oss-20b CUDA out-of-memory error." + }, + { + "left": "issue:43065", + "right": "issue:46369", + "accept": false, + "reason": "Same broad SAM3 area, but one is a dummy Conv2d implementation detail and the other is a memory overflow report; not the same bug." + }, + { + "left": "issue:45084", + "right": "issue:45520", + "accept": false, + "reason": "Different failures: template-node compilation error versus Python 3.13 flash_attn import KeyError." + }, + { + "left": "issue:44423", + "right": "issue:44734", + "accept": false, + "reason": "Both are serve crashes, but one is a multimodal .to() failure and the other is a KV-cache continuation indexing bug." + }, + { + "left": "issue:44554", + "right": "issue:45446", + "accept": false, + "reason": "Both mention attention/flex_attention, but one is an MPS correctness issue and the other is a PyTorch version-check import bug." + }, + { + "left": "issue:28218", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer spacing behavior and TrainerState token counting are unrelated." + }, + { + "left": "issue:15354", + "right": "issue:38175", + "accept": false, + "reason": "Different model issues: TorchScript export failure for ViT versus zero probabilities on SigLIP2." + }, + { + "left": "issue:45406", + "right": "issue:46121", + "accept": false, + "reason": "A serve-time processor attribute error and a rope-parameter JSON list TypeError are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:43701", + "accept": false, + "reason": "SigLIP2 implementation mismatch and resume_from_checkpoint key mismatch are different bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 26, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:37428", + "issue:38175", + "issue:39692", + "issue:42490", + "issue:43232", + "issue:43278", + "issue:43295", + "issue:43334", + "issue:43450", + "issue:43452", + "issue:43493", + "issue:43525", + "issue:43540", + "issue:43550", + "issue:43572", + "issue:43575", + "issue:43653", + "issue:43824", + "issue:43844", + "issue:43856", + "issue:43873", + "issue:43950", + "issue:43957", + "issue:44112", + "issue:44186", + "issue:44466", + "issue:44521", + "issue:44554", + "issue:44610", + "issue:44849", + "issue:44898", + "issue:44987", + "issue:45072", + "issue:45127", + "issue:45198", + "issue:45200", + "issue:45290", + "issue:45356", + "issue:45381", + "issue:45412", + "issue:45520", + "issue:45823", + "issue:46178", + "issue:46192", + "issue:46209", + "issue:46285", + "issue:46512" + ], + "soft_pairs": [ + "issue:45072|issue:45198", + "issue:43295|issue:43450", + "issue:43540|issue:45200", + "issue:37428|issue:43957", + "issue:44186|issue:46192", + "issue:45200|issue:45823", + "issue:45823|issue:46192", + "issue:43295|issue:44610", + "issue:43295|issue:43873", + "issue:43232|issue:44849", + "issue:43450|issue:43575", + "issue:43572|issue:44987", + "issue:45520|issue:46178", + "issue:43550|issue:43844", + "issue:44521|issue:45290", + "issue:39692|issue:46192", + "issue:43572|issue:43957", + "issue:44554|issue:45356", + "issue:37428|issue:43824", + "issue:45381|issue:46209", + "issue:42490|issue:43493", + "issue:44466|issue:45127", + "issue:38175|issue:43493", + "issue:43653|issue:44112", + "issue:44898|issue:45412", + "issue:43950|issue:46285", + "issue:43334|issue:43525", + "issue:43278|issue:43856", + "issue:43452|issue:46512" + ], + "prepared_review_unit_hash": "3116d6b5b501261d4a402f31dd3de27dc37c47fecba7ee577d946b1bb9665489", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 36157, + "estimated_input_tokens": 9040, + "estimated_eval_tokens": 18336 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous issue reports with only loose theme overlap around model loading, tokenization, and v5 regressions. I found no soft-edge pair that is clearly the same underlying bug/change, so all candidate edges are rejected.", + "confidence": 0.69, + "canonical_issue_reason": "Issue 43950 is the strongest representative of the cluster: it is a detailed, core loading regression with clear reproduction and broad impact on `from_pretrained()`, whereas the other issues are narrower and mostly unrelated one-off failures.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43950 is the best single issue to anchor the cluster because it targets a central library path, has a well-defined failure mode, and is more broadly reusable as a representative bug than the model-specific edge cases here.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45072", + "right": "issue:45198", + "accept": false, + "reason": "Different failures: SDPA/bfloat16 dtype mismatch in SwitchTransformers/TimmWrapperModel vs Wav2Vec2 save/tokenization failure." + }, + { + "left": "issue:43295", + "right": "issue:43450", + "accept": false, + "reason": "Both involve processors, but one is a regression in `processor.tokenizer` / image passing and the other is batched video-shape output; not the same bug." + }, + { + "left": "issue:43540", + "right": "issue:45200", + "accept": false, + "reason": "Qwen3OmniMoe video-processing ValueError is unrelated to Gemma 4 text-only fine-tuning needing `mm_token_type_ids`." + }, + { + "left": "issue:37428", + "right": "issue:43957", + "accept": false, + "reason": "ImportError for a missing flash-attn helper is not the same as meta-device loading breaking some models." + }, + { + "left": "issue:44186", + "right": "issue:46192", + "accept": false, + "reason": "LayoutLMv2 tokenizer crashes on NER/batched padding, while the other issue is about `return_tensors` being ignored when `text_kwargs` is passed." + }, + { + "left": "issue:45200", + "right": "issue:45823", + "accept": false, + "reason": "Different code paths: Gemma 4 token-type defaults vs Gemma 4 PLE device placement with `device_map=\"auto\"`." + }, + { + "left": "issue:45823", + "right": "issue:46192", + "accept": false, + "reason": "Device-mapping forward mismatch is unrelated to `return_tensors` argument handling." + }, + { + "left": "issue:43295", + "right": "issue:44610", + "accept": false, + "reason": "Processor/tokenizer regression is not the same as OmDet-Turbo\u2019s image-size mismatch." + }, + { + "left": "issue:43295", + "right": "issue:43873", + "accept": false, + "reason": "Custom processor/tokenizer regression and quantization offloading behavior are different bugs." + }, + { + "left": "issue:43232", + "right": "issue:44849", + "accept": false, + "reason": "Generation kwargs update after `sync_gpus` is unrelated to `output_hidden_states=True` in Qwen3.5." + }, + { + "left": "issue:43450", + "right": "issue:43575", + "accept": false, + "reason": "Batched video shape bug and tensor-parallel OOM on Qwen2-57B are distinct issues." + }, + { + "left": "issue:43572", + "right": "issue:44987", + "accept": false, + "reason": "Missing `pad_token_idx` in StableLmConfig is not the same as loading `physical-intelligence/fast` failing on transformers>=5.1.0." + }, + { + "left": "issue:45520", + "right": "issue:46178", + "accept": false, + "reason": "Different import-registry problems: `flash_attn` on Python 3.13 vs missing `tensorflow_text` backend mapping." + }, + { + "left": "issue:43550", + "right": "issue:43844", + "accept": false, + "reason": "One is a torch.compile + SDPA model failure; the other is abnormal gradient growth under ZeRO-3 during training." + }, + { + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Both touch chat templates, but one is zero assistant masks for multimodal inputs and the other is a crash on tool-call assistant messages with no content." + }, + { + "left": "issue:39692", + "right": "issue:46192", + "accept": false, + "reason": "SigLIP2 doc-example errors are unrelated to `return_tensors` being ignored with explicit `text_kwargs`." + }, + { + "left": "issue:43572", + "right": "issue:43957", + "accept": false, + "reason": "A missing pad-token config field and meta-device loading failures are different regression classes." + }, + { + "left": "issue:44554", + "right": "issue:45356", + "accept": false, + "reason": "MPS attention correctness with mismatched head dims is unrelated to the Kimi-K2.5 tokenizer/codec regression." + }, + { + "left": "issue:37428", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols and different models." + }, + { + "left": "issue:45381", + "right": "issue:46209", + "accept": false, + "reason": "Wrong `vision_position_ids` for Qwen2.5-VL video input is unrelated to FP8 inference failure on B200." + }, + { + "left": "issue:42490", + "right": "issue:43493", + "accept": false, + "reason": "Trailing-slash local-path warning and SigLIP2 JAX/HF discrepancy are not the same underlying bug." + }, + { + "left": "issue:44466", + "right": "issue:45127", + "accept": false, + "reason": "Serialization of tied weights by device is unrelated to LoRA merge collapse with extended vocabulary." + }, + { + "left": "issue:38175", + "right": "issue:43493", + "accept": false, + "reason": "Both involve SigLIP2, but one reports zero probabilities and the other a broader HF-vs-JAX implementation discrepancy; not enough to treat as the same bug." + }, + { + "left": "issue:43653", + "right": "issue:44112", + "accept": false, + "reason": "BigBirdTokenizer special-token registration is unrelated to a GraniteSpeech CI stale test failure." + }, + { + "left": "issue:44898", + "right": "issue:45412", + "accept": false, + "reason": "Perceiver image-size interpolation failure and RT-DETR memory-release issues are unrelated." + }, + { + "left": "issue:43950", + "right": "issue:46285", + "accept": false, + "reason": "Non-persistent buffer corruption in `from_pretrained()` is a loading regression; CLIP text-model structural inconsistency is a different API change." + }, + { + "left": "issue:43334", + "right": "issue:43525", + "accept": false, + "reason": "Both mention missing `pad_token_id`, but they affect different model configs and loading paths; no clear shared bug." + }, + { + "left": "issue:43278", + "right": "issue:43856", + "accept": false, + "reason": "Embedding dtype drift between train/eval is unrelated to Qwen3 MoE memory inefficiency." + }, + { + "left": "issue:43452", + "right": "issue:46512", + "accept": false, + "reason": "GGUF handling in `from_pretrained()` is unrelated to stale `config.dtype` after `from_config(dtype=...)`." + } + ] + }, + "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": 21, + "review_unit_index": 27, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:28218", + "issue:29942", + "issue:32090", + "issue:35141", + "issue:37428", + "issue:38175", + "issue:41720", + "issue:42915", + "issue:43262", + "issue:43335", + "issue:43388", + "issue:43425", + "issue:43450", + "issue:43526", + "issue:43550", + "issue:43606", + "issue:43611", + "issue:43653", + "issue:43819", + "issue:43824", + "issue:43827", + "issue:43901", + "issue:43931", + "issue:43975", + "issue:44112", + "issue:44206", + "issue:44442", + "issue:44561", + "issue:44610", + "issue:44792", + "issue:44857", + "issue:45083", + "issue:45198", + "issue:45200", + "issue:45325", + "issue:45412", + "issue:45520", + "issue:45563", + "issue:45584", + "issue:45593", + "issue:45759", + "issue:46170", + "issue:46192", + "issue:46209", + "issue:46234", + "issue:46285", + "issue:46512" + ], + "soft_pairs": [ + "issue:28218|issue:35141", + "issue:42915|issue:43824", + "issue:43262|issue:43450", + "issue:43526|issue:46192", + "issue:43824|issue:43975", + "issue:43388|issue:43526", + "issue:45200|issue:45325", + "issue:44442|issue:45198", + "issue:44792|issue:45563", + "issue:43819|issue:44112", + "issue:44610|issue:44857", + "issue:38175|issue:45412", + "issue:43827|issue:43901", + "issue:44561|issue:46234", + "issue:41720|issue:46209", + "issue:44206|issue:45584", + "issue:45759|issue:46285", + "issue:43335|issue:43550", + "issue:44857|issue:45593", + "issue:43819|issue:45083", + "issue:45520|issue:46170", + "issue:45759|issue:46512", + "issue:43606|issue:45083", + "issue:37428|issue:43425", + "issue:29942|issue:32090", + "issue:43335|issue:43653", + "issue:43550|issue:44112", + "issue:43611|issue:43931" + ], + "prepared_review_unit_hash": "f9cce3f063192b6b2f467ebbb0510067a889060e87bcb306a6e7b7a9a48477d9", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35695, + "estimated_input_tokens": 8924, + "estimated_eval_tokens": 18104 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These soft links are mostly title-level keyword matches across unrelated Transformer bugs. I found no true duplicate pairs, so I rejected all candidate edges.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue stands out: the items span many unrelated bugs, regressions, and docs issues rather than one underlying problem.", + "canonical_pr_reason": null, + "best_issue_reason": "None of the issues is a strong cluster representative because the proposed links do not form a coherent duplicate set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:28218", + "right": "issue:35141", + "accept": false, + "reason": "Both mention token embeddings, but one is a tokenizer spacing bug and the other is output embedding reinitialization after resize; different code paths." + }, + { + "left": "issue:42915", + "right": "issue:43824", + "accept": false, + "reason": "FP8Config failure in Qwen3MoE is unrelated to a missing Qwen2.5-VL import class." + }, + { + "left": "issue:43262", + "right": "issue:43450", + "accept": false, + "reason": "Audio chat-template sampling rate mismatch and batched video shape handling are separate processor bugs." + }, + { + "left": "issue:43526", + "right": "issue:46192", + "accept": false, + "reason": "One is label reduction dropping values; the other is return_tensors being ignored when text_kwargs is passed." + }, + { + "left": "issue:43824", + "right": "issue:43975", + "accept": false, + "reason": "An import error for a model class is not the same as a detokenization regression in v5." + }, + { + "left": "issue:43388", + "right": "issue:43526", + "accept": false, + "reason": "Both touch labels, but one is gather_for_metrics truncation and the other is BeitImageProcessorFast reduce_labels behavior." + }, + { + "left": "issue:45200", + "right": "issue:45325", + "accept": false, + "reason": "Gemma 4 mm_token_type_ids defaults and Qwen2.5-VL RoPE temporal scaling are different multimodal bugs." + }, + { + "left": "issue:44442", + "right": "issue:45198", + "accept": false, + "reason": "FastSpeech2ConformerTokenizer loading failure is unrelated to Wav2Vec2 save_pretrained/tokenization breakage." + }, + { + "left": "issue:44792", + "right": "issue:45563", + "accept": false, + "reason": "A Janus image-generation test failure and a stale paged-generate warning are not the same problem." + }, + { + "left": "issue:43819", + "right": "issue:44112", + "accept": false, + "reason": "DAC latent/forward mismatch and GraniteSpeech CI device-override flakiness are unrelated." + }, + { + "left": "issue:44610", + "right": "issue:44857", + "accept": false, + "reason": "Processor output size mismatch and LwDetrImageLoss AMP crash are different detection issues." + }, + { + "left": "issue:38175", + "right": "issue:45412", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 is unrelated to RT-DETR memory-release behavior." + }, + { + "left": "issue:43827", + "right": "issue:43901", + "accept": false, + "reason": "Both are docs-related, but one is pipeline removal fallout and the other is a stale return_all_scores doc note." + }, + { + "left": "issue:44561", + "right": "issue:46234", + "accept": false, + "reason": "Removed is_torch_fx_available breaking trust_remote_code is not the same as a missing is_falcon_mamba_ssm_available helper." + }, + { + "left": "issue:41720", + "right": "issue:46209", + "accept": false, + "reason": "Auto device mapping cuda assert on A800 and FP8 inference failure on B200 are separate hardware/runtime issues." + }, + { + "left": "issue:44206", + "right": "issue:45584", + "accept": false, + "reason": "Unsupported center arg in LasrFeatureExtractor and Whisper empty-transcription failure after align_special_tokens are unrelated." + }, + { + "left": "issue:45759", + "right": "issue:46285", + "accept": false, + "reason": "Composite Qwen multimodal config unwrapping and CLIP text-model structure inconsistency are different model/config bugs." + }, + { + "left": "issue:43335", + "right": "issue:43550", + "accept": false, + "reason": "SwitchTransformers zero sparse-layer config and Bamba torch.compile SDPA failure do not share the same bug." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "Float16 AMP crash in LwDetrImageLoss is unrelated to D-FINE skipping auxiliary losses when denoising is off." + }, + { + "left": "issue:43819", + "right": "issue:45083", + "accept": false, + "reason": "DAC.from_latents mismatch and qwen3_omni_moe feat-extract length behavior are different helper/model issues." + }, + { + "left": "issue:45520", + "right": "issue:46170", + "accept": false, + "reason": "Python 3.13 flash_attn import KeyError and Windows ROCm torch.distributed dependency problems are unrelated." + }, + { + "left": "issue:45759", + "right": "issue:46512", + "accept": false, + "reason": "Both involve from_config, but one is about unwrapping text_config and the other is stale config.dtype after dtype construction." + }, + { + "left": "issue:43606", + "right": "issue:45083", + "accept": false, + "reason": "CPU offload device mismatch for bark-small is unrelated to qwen3_omni_moe feature-length helper behavior." + }, + { + "left": "issue:37428", + "right": "issue:43425", + "accept": false, + "reason": "Missing _flash_supports_window_size is a specific import break, not the same as broad Torch 2.10 incompatibility." + }, + { + "left": "issue:29942", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention 2 test failures and a Trainer _gpu_broadcast_one NoneType error are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "SwitchTransformers sparse-layer config bug and BigBirdTokenizer special-token registration issue are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:44112", + "accept": false, + "reason": "A torch.compile/SDPA model failure is not the same as a stale CI device-override test." + }, + { + "left": "issue:43611", + "right": "issue:43931", + "accept": false, + "reason": "base_model_prefix loading regression and Qwen3-VL weight-shape mismatch are different model-loading failures." + } + ] + }, + "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": 21, + "review_unit_index": 28, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:26413", + "issue:28282", + "issue:29127", + "issue:31501", + "issue:32090", + "issue:33357", + "issue:34689", + "issue:35141", + "issue:37428", + "issue:39692", + "issue:42490", + "issue:43278", + "issue:43335", + "issue:43504", + "issue:43526", + "issue:43606", + "issue:43720", + "issue:43819", + "issue:43825", + "issue:43854", + "issue:44112", + "issue:44206", + "issue:44351", + "issue:44448", + "issue:44493", + "issue:44514", + "issue:44521", + "issue:44743", + "issue:44779", + "issue:44855", + "issue:44857", + "issue:45072", + "issue:45083", + "issue:45127", + "issue:45200", + "issue:45245", + "issue:45290", + "issue:45310", + "issue:45459", + "issue:45478", + "issue:45588", + "issue:45593", + "issue:45823", + "issue:45859", + "issue:46234", + "issue:46326", + "issue:46431", + "issue:46566" + ], + "soft_pairs": [ + "issue:28282|issue:33357", + "issue:44514|issue:44521", + "issue:44448|issue:44743", + "issue:43335|issue:43504", + "issue:44855|issue:45083", + "issue:45290|issue:46326", + "issue:46431|issue:46566", + "issue:37428|issue:44351", + "issue:29127|issue:32090", + "issue:43278|issue:43825", + "issue:43335|issue:45083", + "issue:43720|issue:44112", + "issue:45588|issue:45859", + "issue:45200|issue:45588", + "issue:43335|issue:44112", + "issue:45127|issue:45245", + "issue:43335|issue:45593", + "issue:39692|issue:43526", + "issue:43819|issue:44493", + "issue:34689|issue:35141", + "issue:26413|issue:31501", + "issue:44206|issue:45823", + "issue:45310|issue:45478", + "issue:43606|issue:44857", + "issue:42490|issue:43854", + "issue:39692|issue:44493", + "issue:45459|issue:46234", + "issue:44448|issue:44779", + "issue:44857|issue:45072" + ], + "prepared_review_unit_hash": "f8b3a01a5252b506ac09545d37cc44586053b7c4af8ac6ef56e3e9eaba7fbe6c", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36746, + "estimated_input_tokens": 9187, + "estimated_eval_tokens": 18630 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly false-positive similarity matches across unrelated bugs. The only clear duplicate pair is the Qwen3.5 MoE `from_pretrained` regression (`45310` / `45478`).", + "confidence": 0.95, + "canonical_issue_reason": "Issue `45310` is the clearest canonical representative because it is a concrete, reproducible bug report and the matching duplicate (`45478`) is the same Qwen3.5 MoE `from_pretrained` failure.", + "canonical_pr_reason": null, + "best_issue_reason": "`45310` is the best issue to keep as the cluster representative: it has the strongest duplicate evidence and the most direct problem statement.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:28282", + "right": "issue:33357", + "accept": false, + "reason": "Different failure modes: missing PyTorch vs MacOS bus error on CLIP loading; same library, not the same bug." + }, + { + "left": "issue:44514", + "right": "issue:44521", + "accept": false, + "reason": "Both involve chat templates, but one is a batched padding crash and the other is incorrect assistant masks for multimodal inputs." + }, + { + "left": "issue:44448", + "right": "issue:44743", + "accept": false, + "reason": "Unrelated model behaviors: Pegasus output drift across versions vs Qwen3 recurrent state reset with cache." + }, + { + "left": "issue:43335", + "right": "issue:43504", + "accept": false, + "reason": "Different models and code paths: SwitchTransformers sparse-layer config bug vs BEiT legacy pretrained-field loading issue." + }, + { + "left": "issue:44855", + "right": "issue:45083", + "accept": false, + "reason": "Python import/parse failure in DebertaV2 is unrelated to the Qwen3 Omni MoE feature-length helper behavior." + }, + { + "left": "issue:45290", + "right": "issue:46326", + "accept": false, + "reason": "Both are tokenizer/chat-template bugs, but one is tool-call content handling and the other is `continue_final_message` for mistral_common." + }, + { + "left": "issue:46431", + "right": "issue:46566", + "accept": false, + "reason": "A missing chat template error is unrelated to DiffusionGemma weights \u10d3\u10d0\u10e0\u10e9ing on `meta` with `device_map='auto'`." + }, + { + "left": "issue:37428", + "right": "issue:44351", + "accept": false, + "reason": "Different missing-symbol import errors (`_flash_supports_window_size` vs `HybridCache`), not the same underlying breakage." + }, + { + "left": "issue:29127", + "right": "issue:32090", + "accept": false, + "reason": "Clarity of an error message in LayoutLMv3 is unrelated to a Trainer `_gpu_broadcast_one` NoneType crash." + }, + { + "left": "issue:43278", + "right": "issue:43825", + "accept": false, + "reason": "Embedding dtype drift during eval is unrelated to an incorrect pipeline translation-task warning." + }, + { + "left": "issue:43335", + "right": "issue:45083", + "accept": false, + "reason": "Different component bugs: SwitchTransformers config construction vs Qwen3 Omni MoE feature-length helper behavior." + }, + { + "left": "issue:43720", + "right": "issue:44112", + "accept": false, + "reason": "Packed-weight loading in BitNet is unrelated to a stale GraniteSpeech CI test." + }, + { + "left": "issue:45588", + "right": "issue:45859", + "accept": false, + "reason": "Flash-attention sink handling and CPU-offload rotary embedding issues are different code paths and symptoms." + }, + { + "left": "issue:45200", + "right": "issue:45588", + "accept": false, + "reason": "Gemma 4 multimodal token-type defaults are unrelated to the flash_attention `s_aux=None` crash." + }, + { + "left": "issue:43335", + "right": "issue:44112", + "accept": false, + "reason": "No shared bug: one is a SwitchTransformers config edge case, the other is a GraniteSpeech CI failure." + }, + { + "left": "issue:45127", + "right": "issue:45245", + "accept": false, + "reason": "LoRA/extended-vocab collapse is unrelated to the category-cardinality runtime limit." + }, + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation and D-FINE auxiliary-loss handling are different issues." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "Docs/example errors for SigLIP2 are unrelated to BeitImageProcessorFast's `reduce_labels` bug." + }, + { + "left": "issue:43819", + "right": "issue:44493", + "accept": false, + "reason": "DAC latent-forward mismatch is unrelated to unexpected position-id keys across many models." + }, + { + "left": "issue:34689", + "right": "issue:35141", + "accept": false, + "reason": "Llama 3.2 Vision model loading regression is not the same as token-embedding resize reinitialization." + }, + { + "left": "issue:26413", + "right": "issue:31501", + "accept": false, + "reason": "Training resume iterator failure and FineWeb training not starting are both training-related, but not the same underlying bug." + }, + { + "left": "issue:44206", + "right": "issue:45823", + "accept": false, + "reason": "Unsupported feature-extractor args and Gemma4 device-map mismatch are unrelated failures." + }, + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Exact duplicate: both report the same `transformers>=5.4.0` Qwen3.5 MoE `from_pretrained` error." + }, + { + "left": "issue:43606", + "right": "issue:44857", + "accept": false, + "reason": "CPU-offload device mismatch in Bark is unrelated to float16 AMP crashes in LwDetrImageLoss." + }, + { + "left": "issue:42490", + "right": "issue:43854", + "accept": false, + "reason": "A local-path import warning is unrelated to the GLM-4.7-Flash unit-test loading failure." + }, + { + "left": "issue:39692", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 documentation issues are not the same as broad position-id key warnings." + }, + { + "left": "issue:45459", + "right": "issue:46234", + "accept": false, + "reason": "Tokenizer protobuf error masking is unrelated to a missing `is_falcon_mamba_ssm_available` import utility symbol." + }, + { + "left": "issue:44448", + "right": "issue:44779", + "accept": false, + "reason": "Pegasus v4/v5 output differences are unrelated to Deepseek tokenizer regressions." + }, + { + "left": "issue:44857", + "right": "issue:45072", + "accept": false, + "reason": "These are distinct numerical/runtime issues: AMP crash in LwDetrImageLoss vs dtype mismatch in SwitchTransformers/TimmWrapperModel inference." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet: the only clearly supported duplicate is 45310/45478, and the other soft-edge rejections stay conservative by citing distinct titles/failure modes rather than overstating certainty. No mergeability issues apply here since these are issue pairs, not PR pairs." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 29, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:10000", + "issue:15354", + "issue:25251", + "issue:27957", + "issue:28218", + "issue:29942", + "issue:32090", + "issue:33290", + "issue:33357", + "issue:34689", + "issue:35141", + "issue:37428", + "issue:38175", + "issue:40444", + "issue:41720", + "issue:42490", + "issue:43278", + "issue:43329", + "issue:43381", + "issue:43526", + "issue:43550", + "issue:43575", + "issue:43701", + "issue:43746", + "issue:43873", + "issue:43927", + "issue:43976", + "issue:44360", + "issue:44448", + "issue:44466", + "issue:44493", + "issue:44561", + "issue:44779", + "issue:44855", + "issue:44857", + "issue:44977", + "issue:44987", + "issue:45083", + "issue:45229", + "issue:45290", + "issue:45381", + "issue:45440", + "issue:45588", + "issue:45698", + "issue:45920", + "issue:46192", + "issue:46209", + "issue:46399" + ], + "soft_pairs": [ + "issue:28218|issue:32090", + "issue:44360|issue:44561", + "issue:43701|issue:45083", + "issue:38175|issue:44493", + "issue:45290|issue:45381", + "issue:32090|issue:33357", + "issue:40444|issue:43526", + "issue:42490|issue:43575", + "issue:29942|issue:35141", + "issue:43550|issue:45083", + "issue:45920|issue:46192", + "issue:44448|issue:44977", + "issue:27957|issue:34689", + "issue:44855|issue:46399", + "issue:44977|issue:45229", + "issue:43278|issue:43329", + "issue:10000|issue:15354", + "issue:44857|issue:45588", + "issue:43927|issue:45698", + "issue:43278|issue:43381", + "issue:43746|issue:46209", + "issue:37428|issue:44987", + "issue:44561|issue:44779", + "issue:43873|issue:45229", + "issue:32090|issue:33290", + "issue:25251|issue:32090", + "issue:41720|issue:45229", + "issue:44448|issue:44466", + "issue:43976|issue:45440" + ], + "prepared_review_unit_hash": "dd23d6346c22253fce35606f2bb5b4a98eecf21e975528d312e4e92eab9c56a2", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36429, + "estimated_input_tokens": 9108, + "estimated_eval_tokens": 18472 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These issues are broadly heterogeneous; the soft links mostly reflect superficial similarity in error types or subsystem names, not the same underlying bug. No soft pair looks safe to merge as a duplicate.", + "confidence": 0.93, + "canonical_issue_reason": "If one issue must anchor the cluster, issue 45920 is the best representative: it is a broad, clearly described v5 tokenizer regression affecting multiple models and has the widest apparent user impact.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45920 is the strongest triage target because it is open, concrete, and summarizes a multi-model regression in a way that can guide follow-up work.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:28218", + "right": "issue:32090", + "accept": false, + "reason": "Added-token spacing in tokenization and a Trainer _gpu_broadcast_one NoneType error are unrelated failures." + }, + { + "left": "issue:44360", + "right": "issue:44561", + "accept": false, + "reason": "DSA indexer ReLU discussion and a v5 trust_remote_code compatibility break are different code paths and symptoms." + }, + { + "left": "issue:43701", + "right": "issue:45083", + "accept": false, + "reason": "Checkpoint key mismatches and a qwen3_omni_moe helper behavior issue are not the same bug." + }, + { + "left": "issue:38175", + "right": "issue:44493", + "accept": false, + "reason": "Zero probabilities in SigLIP2 and unexpected position-id keys are distinct model-output problems." + }, + { + "left": "issue:45290", + "right": "issue:45381", + "accept": false, + "reason": "Chat template crashes on tool-call messages and Qwen2.5-VL video position-id issues are separate failures." + }, + { + "left": "issue:32090", + "right": "issue:33357", + "accept": false, + "reason": "Distributed broadcast NoneType support and a MacOS bus error with CLIP are unrelated." + }, + { + "left": "issue:40444", + "right": "issue:43526", + "accept": false, + "reason": "Multi-image iterable dataset finetuning and BeitImageProcessorFast reduce_labels behavior are different bugs." + }, + { + "left": "issue:42490", + "right": "issue:43575", + "accept": false, + "reason": "Local-path import warning and tensor-parallel OOM are not the same issue." + }, + { + "left": "issue:29942", + "right": "issue:35141", + "accept": false, + "reason": "Flash Attention 2 test failures and post_init reinitializing embeddings are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:45083", + "accept": false, + "reason": "torch.compile/SDPA failure in Bamba and qwen3_omni_moe helper behavior are distinct." + }, + { + "left": "issue:45920", + "right": "issue:46192", + "accept": false, + "reason": "Wrong tokenizer IDs across models and silently ignored return_tensors are different tokenizer regressions." + }, + { + "left": "issue:44448", + "right": "issue:44977", + "accept": false, + "reason": "Pegasus v4/v5 output drift and Qwen3.5 flash-attention generation failure are unrelated." + }, + { + "left": "issue:27957", + "right": "issue:34689", + "accept": false, + "reason": "XLMRoberta Flash Attention 2 and Llama 3.2 Vision loading are separate model-specific problems." + }, + { + "left": "issue:44855", + "right": "issue:46399", + "accept": false, + "reason": "Python 3.13 decorator/comment parsing error is not the same as a missing apply_rotary_pos_emb decorator registration bug." + }, + { + "left": "issue:44977", + "right": "issue:45229", + "accept": false, + "reason": "Flash-attention generation issues and a multi-GPU CUDA OOM are different failure modes." + }, + { + "left": "issue:43278", + "right": "issue:43329", + "accept": false, + "reason": "Embedding dtype drift and a multimodal-token counting video-branch bug are unrelated." + }, + { + "left": "issue:10000", + "right": "issue:15354", + "accept": false, + "reason": "German DistilBertModel and GeneratorExp torch.jit.script export are different historical issues." + }, + { + "left": "issue:44857", + "right": "issue:45588", + "accept": false, + "reason": "LwDetrImageLoss AMP crash and flash_attention.py s_aux=None AttributeError are not the same code-path bug." + }, + { + "left": "issue:43927", + "right": "issue:45698", + "accept": false, + "reason": "Custom token IDs lost on save/load and wrong custom module loading after save_pretrained are related only broadly to serialization, not the same defect." + }, + { + "left": "issue:43278", + "right": "issue:43381", + "accept": false, + "reason": "BF16-to-FP32 embedding dtype change and gradient checkpointing in eval mode are unrelated." + }, + { + "left": "issue:43746", + "right": "issue:46209", + "accept": false, + "reason": "PEFT adapter checkpoint loading and FP8 inference on B200 are separate issues." + }, + { + "left": "issue:37428", + "right": "issue:44987", + "accept": false, + "reason": "Missing _flash_supports_window_size import and a model-loading failure for physical-intelligence/fast are too vague to treat as the same bug." + }, + { + "left": "issue:44561", + "right": "issue:44779", + "accept": false, + "reason": "Removed helper breaking trust_remote_code and DeepSeek tokenizer regression are different failures." + }, + { + "left": "issue:43873", + "right": "issue:45229", + "accept": false, + "reason": "Quantization offloading behavior and Gemma4 multi-GPU OOM are not the same underlying problem." + }, + { + "left": "issue:32090", + "right": "issue:33290", + "accept": false, + "reason": "Trainer broadcast TypeError and Adafactor/DeepSpeed OOM are unrelated." + }, + { + "left": "issue:25251", + "right": "issue:32090", + "accept": false, + "reason": "Pipeline top_k nesting behavior and a distributed broadcast TypeError do not share a code path." + }, + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "Qwen3 auto device mapping cudaErrorAssert and Gemma4 OOM are distinct memory/runtime issues." + }, + { + "left": "issue:44448", + "right": "issue:44466", + "accept": false, + "reason": "Pegasus output differences and lm_head serialization inconsistency are different regressions." + }, + { + "left": "issue:43976", + "right": "issue:45440", + "accept": false, + "reason": "Python version compatibility in Transformers 5.1.0 and DeepseekV3MoE divergence are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 30, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:27957", + "issue:29127", + "issue:29942", + "issue:30064", + "issue:33357", + "issue:34567", + "issue:39692", + "issue:42490", + "issue:42915", + "issue:42994", + "issue:43065", + "issue:43066", + "issue:43278", + "issue:43381", + "issue:43412", + "issue:43575", + "issue:43611", + "issue:43653", + "issue:43701", + "issue:43756", + "issue:43873", + "issue:43994", + "issue:44186", + "issue:44206", + "issue:44265", + "issue:44361", + "issue:44442", + "issue:44462", + "issue:44479", + "issue:44568", + "issue:44743", + "issue:44871", + "issue:44977", + "issue:45198", + "issue:45245", + "issue:45372", + "issue:45406", + "issue:45412", + "issue:45440", + "issue:45459", + "issue:45491", + "issue:45563", + "issue:45588", + "issue:45712", + "issue:45803", + "issue:45823", + "issue:46121", + "issue:46369" + ], + "soft_pairs": [ + "issue:45588|issue:45823", + "issue:42490|issue:43873", + "issue:44568|issue:45491", + "issue:45198|issue:45803", + "issue:43412|issue:45588", + "issue:45372|issue:45588", + "issue:44977|issue:45563", + "issue:43278|issue:43575", + "issue:33357|issue:39692", + "issue:29127|issue:34567", + "issue:43653|issue:44743", + "issue:43381|issue:43701", + "issue:43611|issue:46121", + "issue:44871|issue:44977", + "issue:44361|issue:45198", + "issue:45372|issue:45406", + "issue:44186|issue:45198", + "issue:42915|issue:42994", + "issue:44206|issue:44479", + "issue:43065|issue:45412", + "issue:29942|issue:30064", + "issue:43065|issue:43994", + "issue:45459|issue:45712", + "issue:43756|issue:45440", + "issue:45412|issue:46369", + "issue:43653|issue:44442", + "issue:45245|issue:45563", + "issue:45372|issue:45459", + "issue:43994|issue:45412", + "issue:44265|issue:45198", + "issue:29127|issue:33357", + "issue:43066|issue:44462", + "issue:27957|issue:29942" + ], + "prepared_review_unit_hash": "35ab63a6113f29f36ac92e6467a30125430d064be194df7f65606b5a23c9e870", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37588, + "estimated_input_tokens": 9397, + "estimated_eval_tokens": 19050 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is not a duplicate cluster: the items span many unrelated bugs/regressions across different models, tokenizers, export paths, and runtime behaviors. The soft-similarity pairs mostly share broad subsystem keywords, not the same underlying defect.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the best anchor only because it is the most active open item with broad impact and discussion; it is not a duplicate representative of the other issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45588", + "right": "issue:45823", + "accept": false, + "reason": "Different bugs: flash_attention.py crashes on sink-less models, while Gemma4 PLE has a device-mismatch during forward." + }, + { + "left": "issue:42490", + "right": "issue:43873", + "accept": false, + "reason": "Unrelated: trailing-slash local model import warning vs quantization/offloading behavior." + }, + { + "left": "issue:44568", + "right": "issue:45491", + "accept": false, + "reason": "Different tokenizer special-token regression vs Gemma3 NaN embeddings with mixed-length sliding-window batching." + }, + { + "left": "issue:45198", + "right": "issue:45803", + "accept": false, + "reason": "Wav2Vec2 save/tokenization failure is unrelated to Fuyu batch tensor conversion swallowing errors." + }, + { + "left": "issue:43412", + "right": "issue:45588", + "accept": false, + "reason": "Gemma3n export failure and flash_attention sink-less crash are different code paths and symptoms." + }, + { + "left": "issue:45372", + "right": "issue:45588", + "accept": false, + "reason": "Gemma 4 processor import breakage is unrelated to the flash_attention.py AttributeError." + }, + { + "left": "issue:44977", + "right": "issue:45563", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug vs stale paged-generate warning; not the same defect." + }, + { + "left": "issue:43278", + "right": "issue:43575", + "accept": false, + "reason": "Different runtime issues: BF16/FP32 dtype drift vs tensor-parallel OOM on a specific Qwen model." + }, + { + "left": "issue:33357", + "right": "issue:39692", + "accept": false, + "reason": "MacOS bus error loading a CLIP community model is unrelated to SigLIP2 doc example errors." + }, + { + "left": "issue:29127", + "right": "issue:34567", + "accept": false, + "reason": "LayoutLMv3 box-clarity error message issue vs TrainerState token-count update bug." + }, + { + "left": "issue:43653", + "right": "issue:44743", + "accept": false, + "reason": "BigBirdTokenizer special-token registration bug is unrelated to Qwen3.5 recurrent-state reset behavior." + }, + { + "left": "issue:43381", + "right": "issue:43701", + "accept": false, + "reason": "Gradient checkpointing in eval mode and resume_from_checkpoint key mismatch are different training bugs." + }, + { + "left": "issue:43611", + "right": "issue:46121", + "accept": false, + "reason": "base_model_prefix loading breakage is unrelated to rope-validation JSON list handling." + }, + { + "left": "issue:44871", + "right": "issue:44977", + "accept": false, + "reason": "Gemma-3 eos_token_id mismatch is not the same as Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:44361", + "right": "issue:45198", + "accept": false, + "reason": "MLukeTokenizer task AttributeError differs from Wav2Vec2 save_pretrained/tokenization failure." + }, + { + "left": "issue:45372", + "right": "issue:45406", + "accept": false, + "reason": "Both involve Gemma 4, but one is a missing ReasoningEffort import and the other is a serve-side _tokenizer AttributeError." + }, + { + "left": "issue:44186", + "right": "issue:45198", + "accept": false, + "reason": "LayoutLMv2Tokenizer NER/padding crash is unrelated to Wav2Vec2 save/tokenization failure." + }, + { + "left": "issue:42915", + "right": "issue:42994", + "accept": false, + "reason": "Qwen3Moe FineGrainedFP8Config failure and quantized model saving failure are distinct quantization-related bugs." + }, + { + "left": "issue:44206", + "right": "issue:44479", + "accept": false, + "reason": "Feature extractor unsupported-arg crash is different from the video-input regression in Qwen VL models." + }, + { + "left": "issue:43065", + "right": "issue:45412", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder and RT-DETR memory not released are unrelated model implementation bugs." + }, + { + "left": "issue:29942", + "right": "issue:30064", + "accept": false, + "reason": "Failing Flash Attention 2 tests are not the same as void segmentation map processing failure." + }, + { + "left": "issue:43065", + "right": "issue:43994", + "accept": false, + "reason": "Sam3PixelDecoder dummy layer issue is unrelated to SigLIP2 nonsensical AutoModel/pipeline results." + }, + { + "left": "issue:45459", + "right": "issue:45712", + "accept": false, + "reason": "Tokenizer protobuf error handling is unrelated to leftover dummy classes in dummy_pt_objects.py." + }, + { + "left": "issue:43756", + "right": "issue:45440", + "accept": false, + "reason": "Smollm3 RoPE-layer dropout bug is unrelated to DeepseekV3MoE divergence from remote implementation." + }, + { + "left": "issue:45412", + "right": "issue:46369", + "accept": false, + "reason": "RT-DETR memory leak on deletion is different from Sam3-LiteText memory overflow during use." + }, + { + "left": "issue:43653", + "right": "issue:44442", + "accept": false, + "reason": "BigBirdTokenizer special-token registration bug is unrelated to AutoTokenizer failing to load FastSpeech2ConformerTokenizer." + }, + { + "left": "issue:45245", + "right": "issue:45563", + "accept": false, + "reason": "Category-count RuntimeError and stale num_return_sequences warning are unrelated issues." + }, + { + "left": "issue:45372", + "right": "issue:45459", + "accept": false, + "reason": "Gemma 4 processor import failure is unrelated to protobuf exception masking tokenizer errors." + }, + { + "left": "issue:43994", + "right": "issue:45412", + "accept": false, + "reason": "SigLIP2 model/pipeline mismatch is unrelated to RT-DETR memory not being released." + }, + { + "left": "issue:44265", + "right": "issue:45198", + "accept": false, + "reason": "torch.export failure with torch_compilable_check is unrelated to Wav2Vec2 save/tokenization failure." + }, + { + "left": "issue:29127", + "right": "issue:33357", + "accept": false, + "reason": "LayoutLMv3 error-message clarity and MacOS CLIP bus error are unrelated bugs." + }, + { + "left": "issue:43066", + "right": "issue:44462", + "accept": false, + "reason": "Wrong tokenizer decoder type in v5 and AutoTokenizer ignoring tokenizer.json are different loading/tokenizer issues." + }, + { + "left": "issue:27957", + "right": "issue:29942", + "accept": false, + "reason": "XLMRoberta Flash Attention 2 support and failing Flash Attention 2 tests are not the same underlying bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 31, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:27957", + "issue:29127", + "issue:33290", + "issue:34567", + "issue:34689", + "issue:35141", + "issue:42490", + "issue:42915", + "issue:43262", + "issue:43334", + "issue:43381", + "issue:43421", + "issue:43526", + "issue:43531", + "issue:43575", + "issue:43646", + "issue:43701", + "issue:43761", + "issue:43824", + "issue:43873", + "issue:43931", + "issue:43994", + "issue:44077", + "issue:44186", + "issue:44206", + "issue:44279", + "issue:44387", + "issue:44442", + "issue:44464", + "issue:44743", + "issue:44811", + "issue:44857", + "issue:44898", + "issue:44977", + "issue:45072", + "issue:45245", + "issue:45250", + "issue:45440", + "issue:45491", + "issue:45588", + "issue:45593", + "issue:45600", + "issue:45834", + "issue:45859", + "issue:46121", + "issue:46369", + "issue:46399", + "issue:46640" + ], + "soft_pairs": [ + "issue:45440|issue:45593", + "issue:43575|issue:43701", + "issue:43421|issue:45491", + "issue:29127|issue:35141", + "issue:43526|issue:45834", + "issue:42490|issue:43381", + "issue:45250|issue:46640", + "issue:43575|issue:44387", + "issue:43994|issue:46369", + "issue:43526|issue:44186", + "issue:44442|issue:45072", + "issue:42915|issue:43931", + "issue:43824|issue:44279", + "issue:43526|issue:44857", + "issue:46121|issue:46399", + "issue:27957|issue:29127", + "issue:43334|issue:43531", + "issue:27957|issue:33290", + "issue:43761|issue:45859", + "issue:33290|issue:34689", + "issue:45588|issue:45600", + "issue:33290|issue:35141", + "issue:43262|issue:44206", + "issue:44977|issue:45245", + "issue:33290|issue:34567", + "issue:43381|issue:43873", + "issue:44464|issue:45491", + "issue:43646|issue:44077", + "issue:44743|issue:44811", + "issue:44442|issue:44898", + "issue:42490|issue:44743" + ], + "prepared_review_unit_hash": "1eb57403c1e5dd71b92394a7fb339f5b18dfd732044552946de5e0f92ee97d35", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36765, + "estimated_input_tokens": 9192, + "estimated_eval_tokens": 18640 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: the soft-similarity pairs are mostly broad title overlaps, not the same underlying bug or change. No pull requests are present, and none of the soft edges should be merged.", + "confidence": 0.64, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the strongest representative issue in the set: it has the most discussion/inbound references and a concrete, well-scoped bug report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45440", + "right": "issue:45593", + "accept": false, + "reason": "Different models/features and different failure modes; only a broad ML bug theme is shared." + }, + { + "left": "issue:43575", + "right": "issue:43701", + "accept": false, + "reason": "One is an OOM during tensor-parallel model load, the other is a checkpoint resume key mismatch." + }, + { + "left": "issue:43421", + "right": "issue:45491", + "accept": false, + "reason": "Tokenizer post-processor runtime behavior vs GPU NaNs in Gemma3; unrelated code paths." + }, + { + "left": "issue:29127", + "right": "issue:35141", + "accept": false, + "reason": "LayoutLMV3 error messaging and embedding reinit after resize are different bugs." + }, + { + "left": "issue:43526", + "right": "issue:45834", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug vs Kosmos2.5 OCR index error; not the same issue." + }, + { + "left": "issue:42490", + "right": "issue:43381", + "accept": false, + "reason": "Local-model import warning with trailing slash vs gradient checkpointing eval-mode restriction." + }, + { + "left": "issue:45250", + "right": "issue:46640", + "accept": false, + "reason": "Both mention Flash Attention 2, but the titles are too generic to prove the same concrete bug." + }, + { + "left": "issue:43575", + "right": "issue:44387", + "accept": false, + "reason": "Both mention OOM, but one is TP model loading and the other is int4 quantization memory growth." + }, + { + "left": "issue:43994", + "right": "issue:46369", + "accept": false, + "reason": "SigLIP2 bad outputs vs SAM3-LiteText memory overflow; different models and symptoms." + }, + { + "left": "issue:43526", + "right": "issue:44186", + "accept": false, + "reason": "Image processor label handling and LayoutLMv2 tokenizer crashes are unrelated." + }, + { + "left": "issue:44442", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer loading failure vs dtype mismatch in inference; not the same code-path problem." + }, + { + "left": "issue:42915", + "right": "issue:43931", + "accept": false, + "reason": "FineGrainedFP8Config failure for Qwen3MoE vs weight-shape mismatch loading Qwen3-VL." + }, + { + "left": "issue:43824", + "right": "issue:44279", + "accept": false, + "reason": "Missing Qwen2.5-VL import and a generic dependency issue are not the same bug." + }, + { + "left": "issue:43526", + "right": "issue:44857", + "accept": false, + "reason": "Label-reduction behavior in an image processor vs AMP crash in LwDetrImageLoss." + }, + { + "left": "issue:46121", + "right": "issue:46399", + "accept": false, + "reason": "RoPE validation TypeError vs missing decorator for apply_rotary_pos_emb registration." + }, + { + "left": "issue:27957", + "right": "issue:29127", + "accept": false, + "reason": "Flash Attention support for XLM-Roberta vs LayoutLMv3 error-message clarity; unrelated." + }, + { + "left": "issue:43334", + "right": "issue:43531", + "accept": false, + "reason": "Qwen3-VL pad_token_id attribute error vs Qwen3-MoE sliding-window behavior." + }, + { + "left": "issue:27957", + "right": "issue:33290", + "accept": false, + "reason": "Flash Attention support vs AdaFactor/DeepSpeed OOM; different problems." + }, + { + "left": "issue:43761", + "right": "issue:45859", + "accept": false, + "reason": "CLIP hidden_states regression vs Qwen3_5Moe CPU offload incompatibility." + }, + { + "left": "issue:33290", + "right": "issue:34689", + "accept": false, + "reason": "DeepSpeed/AdaFactor OOM vs model-loading regression for Llama 3.2 Vision." + }, + { + "left": "issue:45588", + "right": "issue:45600", + "accept": false, + "reason": "Flash-attention sink-less crash vs stale auto_mappings references; different subsystems." + }, + { + "left": "issue:33290", + "right": "issue:35141", + "accept": false, + "reason": "Optimizer OOM vs output-embedding reinitialization after resizing token embeddings." + }, + { + "left": "issue:43262", + "right": "issue:44206", + "accept": false, + "reason": "Audio chat-template sampling-rate default vs unsupported center arg in feature extractor." + }, + { + "left": "issue:44977", + "right": "issue:45245", + "accept": false, + "reason": "Qwen3.5 flash-attention generation issue vs category-count limit runtime error." + }, + { + "left": "issue:33290", + "right": "issue:34567", + "accept": false, + "reason": "OOM in DeepSpeed/AdaFactor vs TrainerState token counter not updating." + }, + { + "left": "issue:43381", + "right": "issue:43873", + "accept": false, + "reason": "Gradient checkpointing eval-mode restriction and quantization offloading are distinct behaviors." + }, + { + "left": "issue:44464", + "right": "issue:45491", + "accept": false, + "reason": "Compiled-forward chunked generation inconsistency vs Gemma3 sliding-window NaNs." + }, + { + "left": "issue:43646", + "right": "issue:44077", + "accept": false, + "reason": "Custom model initialization breakage in v5 vs patchtsmixer post_init API restriction." + }, + { + "left": "issue:44743", + "right": "issue:44811", + "accept": false, + "reason": "Qwen3 recurrent-state reset vs Whisper batch_decode skip_special_tokens bug." + }, + { + "left": "issue:44442", + "right": "issue:44898", + "accept": false, + "reason": "Tokenizer loading failure vs Perceiver image-size/interpolation failure; unrelated." + }, + { + "left": "issue:42490", + "right": "issue:44743", + "accept": false, + "reason": "Local path import warning and Qwen3 recurrent-state reset are unrelated 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": 21, + "review_unit_index": 32, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:25251", + "issue:28282", + "issue:34689", + "issue:42175", + "issue:42617", + "issue:42907", + "issue:43232", + "issue:43316", + "issue:43388", + "issue:43493", + "issue:43525", + "issue:43575", + "issue:43611", + "issue:43643", + "issue:43701", + "issue:43720", + "issue:43723", + "issue:43749", + "issue:43761", + "issue:43827", + "issue:43873", + "issue:43931", + "issue:44062", + "issue:44077", + "issue:44079", + "issue:44188", + "issue:44190", + "issue:44442", + "issue:44464", + "issue:44479", + "issue:44496", + "issue:44855", + "issue:44871", + "issue:44913", + "issue:44987", + "issue:45003", + "issue:45137", + "issue:45216", + "issue:45440", + "issue:45479", + "issue:45698", + "issue:45759", + "issue:45803", + "issue:45907", + "issue:46178", + "issue:46234", + "issue:46421", + "issue:46506" + ], + "soft_pairs": [ + "issue:34689|issue:43611", + "issue:45003|issue:45440", + "issue:43720|issue:44442", + "issue:43316|issue:43525", + "issue:42175|issue:46234", + "issue:43723|issue:45479", + "issue:44062|issue:45907", + "issue:43761|issue:44079", + "issue:44077|issue:44479", + "issue:43493|issue:43827", + "issue:43761|issue:44188", + "issue:43873|issue:44871", + "issue:43232|issue:43388", + "issue:43749|issue:43931", + "issue:44987|issue:46178", + "issue:43931|issue:45216", + "issue:44442|issue:45803", + "issue:43316|issue:45759", + "issue:44855|issue:45137", + "issue:43643|issue:44913", + "issue:42907|issue:45698", + "issue:43701|issue:46506", + "issue:44190|issue:44496", + "issue:44464|issue:46421", + "issue:42617|issue:43575", + "issue:25251|issue:28282" + ], + "prepared_review_unit_hash": "dcc1efa9a0ff8ec757c1ad2fe3e28f3b7f196f2f2f175a0b114312bb3a409401", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 26, + "serialized_chars": 35636, + "estimated_input_tokens": 8909, + "estimated_eval_tokens": 18074 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These issues span many distinct bugs across loading, configs, generation, datasets, and docs. Despite superficial subsystem overlap, none of the soft pairs look like the same underlying bug or change, so all should be rejected as duplicates.", + "confidence": 0.92, + "canonical_issue_reason": "Issue 43873 is the broadest and most actively discussed open regression in the set, so it is the least-specific representative if a canonical must be chosen. That said, the cluster is heterogeneous and has no strong true canonical.", + "canonical_pr_reason": null, + "best_issue_reason": "43873 is the best-fit issue only in a weak, administrative sense: it is open, high-activity, and broadly impactful. There is no stronger cluster-wide representative among these unrelated reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:34689", + "right": "issue:43611", + "accept": false, + "reason": "Both are model-loading regressions, but they describe different failures, models, and likely different code paths." + }, + { + "left": "issue:45003", + "right": "issue:45440", + "accept": false, + "reason": "`sys.modules` access safety and DeepSeekV3 parity are unrelated bugs with different root causes." + }, + { + "left": "issue:43720", + "right": "issue:44442", + "accept": false, + "reason": "One is BitNet accelerate-weight loading, the other is tokenizer loading; these are separate components." + }, + { + "left": "issue:43316", + "right": "issue:43525", + "accept": false, + "reason": "Both are config API discrepancies, but for different models and different missing fields, not the same bug." + }, + { + "left": "issue:42175", + "right": "issue:46234", + "accept": false, + "reason": "Package extra/backend inclusion and a missing import_utils availability helper are different issues." + }, + { + "left": "issue:43723", + "right": "issue:45479", + "accept": false, + "reason": "Tokenizer loading in v5 and sequence-classification loss behavior are unrelated." + }, + { + "left": "issue:44062", + "right": "issue:45907", + "accept": false, + "reason": "AddedToken keyword duplication and list_repo_files revision forwarding are unrelated code paths." + }, + { + "left": "issue:43761", + "right": "issue:44079", + "accept": false, + "reason": "One is CLIPVision hidden_states output, the other is generic ModelOutput key assignment; too different to merge." + }, + { + "left": "issue:44077", + "right": "issue:44479", + "accept": false, + "reason": "PatchTSMixer post_init validation and Qwen video-input regression are separate model-specific bugs." + }, + { + "left": "issue:43493", + "right": "issue:43827", + "accept": false, + "reason": "Model implementation parity and docs still referencing pipeline() are unrelated." + }, + { + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "Hidden-state output regression and attention-kernel divergence under torch.compile are different problems." + }, + { + "left": "issue:43873", + "right": "issue:44871", + "accept": false, + "reason": "Quantization/offloading behavior and Gemma eos_token_id configuration mismatch are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:43388", + "accept": false, + "reason": "Generation kwargs handling and gather_for_metrics label dropping are different trainer/generation bugs." + }, + { + "left": "issue:43749", + "right": "issue:43931", + "accept": false, + "reason": "FSDP CPU RAM-efficient loading and Qwen3-VL shape mismatch are both loading-related but have different failure modes and causes." + }, + { + "left": "issue:44987", + "right": "issue:46178", + "accept": false, + "reason": "A model-loading regression and a backend mapping error for tensorflow_text are not the same issue." + }, + { + "left": "issue:43931", + "right": "issue:45216", + "accept": false, + "reason": "Both involve Qwen model persistence/loading, but one is a weight-shape mismatch and the other is a save_pretrained regression." + }, + { + "left": "issue:44442", + "right": "issue:45803", + "accept": false, + "reason": "Tokenizer loading vs Fuyu batch feature tensor conversion error; different functionality and symptoms." + }, + { + "left": "issue:43316", + "right": "issue:45759", + "accept": false, + "reason": "Both touch config handling, but one is a field discrepancy and the other is unwrapping text_config for composite multimodal configs." + }, + { + "left": "issue:44855", + "right": "issue:45137", + "accept": false, + "reason": "Python 3.13 torch.jit parsing and DeepSpeed ZeRO3 deque underflow are unrelated." + }, + { + "left": "issue:43643", + "right": "issue:44913", + "accept": false, + "reason": "Remote-code config field loss and GPTNeoX rotary_pct not persisting are both config-related but not the same concrete bug." + }, + { + "left": "issue:42907", + "right": "issue:45698", + "accept": false, + "reason": "Saving dequantized models and loading the wrong custom module after save_pretrained are different save/load failures." + }, + { + "left": "issue:43701", + "right": "issue:46506", + "accept": false, + "reason": "Checkpoint key mismatch and CPU/CUDA output differences are unrelated." + }, + { + "left": "issue:44190", + "right": "issue:44496", + "accept": false, + "reason": "Dataset loading in a training script and missing model_type in config.json are unrelated." + }, + { + "left": "issue:44464", + "right": "issue:46421", + "accept": false, + "reason": "Both mention chunked generation/prefill, but one is output inconsistency and the other is cache lazy-init recompilation; not the same bug." + }, + { + "left": "issue:42617", + "right": "issue:43575", + "accept": false, + "reason": "A parallel script failure and a tensor-parallel OOM on a specific model are different distributed-runtime issues." + }, + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Pipeline top_k nesting behavior and a PyTorch ImportError are completely 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": 21, + "review_unit_index": 33, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:16998", + "issue:22355", + "issue:27957", + "issue:32090", + "issue:42175", + "issue:42796", + "issue:43065", + "issue:43295", + "issue:43316", + "issue:43352", + "issue:43404", + "issue:43441", + "issue:43452", + "issue:43540", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43646", + "issue:43650", + "issue:43701", + "issue:43716", + "issue:43742", + "issue:43761", + "issue:43825", + "issue:44077", + "issue:44162", + "issue:44479", + "issue:44496", + "issue:44514", + "issue:44534", + "issue:44610", + "issue:44617", + "issue:44792", + "issue:44871", + "issue:44913", + "issue:45290", + "issue:45584", + "issue:45632", + "issue:45803", + "issue:46050", + "issue:46170", + "issue:46202", + "issue:46234", + "issue:46291", + "issue:46365", + "issue:46369", + "issue:46424", + "issue:46640" + ], + "soft_pairs": [ + "issue:27957|issue:32090", + "issue:43550|issue:43701", + "issue:46234|issue:46291", + "issue:43575|issue:43742", + "issue:43606|issue:43701", + "issue:43295|issue:43825", + "issue:42796|issue:45290", + "issue:16998|issue:22355", + "issue:43295|issue:44871", + "issue:44871|issue:45584", + "issue:43316|issue:43716", + "issue:42175|issue:43316", + "issue:45803|issue:46365", + "issue:44610|issue:46369", + "issue:44077|issue:44534", + "issue:43646|issue:44913", + "issue:44871|issue:46050", + "issue:45632|issue:46424", + "issue:45803|issue:46170", + "issue:44162|issue:44792", + "issue:43452|issue:44496", + "issue:43065|issue:44617", + "issue:43650|issue:46202", + "issue:43404|issue:43441", + "issue:43352|issue:46640", + "issue:43540|issue:44514", + "issue:43761|issue:44479" + ], + "prepared_review_unit_hash": "f0a2ce758fce7b43043e439006f3c7343f599fd7c28c4a553794e310d79422bc", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35439, + "estimated_input_tokens": 8860, + "estimated_eval_tokens": 17976 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is a loose mix of mostly unrelated Transformers issues that only share superficial subsystem or keyword overlap; none of the soft pairs look like true duplicates, so no pair should be merged.", + "confidence": 0.92, + "canonical_issue_reason": "issue:43825 is the clearest self-contained, user-facing regression in the set, with explicit expected behavior and inbound references, so it is the best single representative if one must be chosen.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43825 is the strongest global issue candidate because it is concrete, well-scoped, and clearly describes a reproducible API regression.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:27957", + "right": "issue:32090", + "accept": false, + "reason": "Different failures: Flash Attention support in XLM-Roberta vs Trainer GPU broadcast NoneType error." + }, + { + "left": "issue:43550", + "right": "issue:43701", + "accept": false, + "reason": "Unrelated bugs: torch.compile/SDPA failure vs resume_from_checkpoint key mismatch." + }, + { + "left": "issue:46234", + "right": "issue:46291", + "accept": false, + "reason": "Both involve imports, but one is a missing utility symbol and the other is a kernels package version requirement." + }, + { + "left": "issue:43575", + "right": "issue:43742", + "accept": false, + "reason": "Different loading problems: tensor-parallel OOM vs key error loading a small model." + }, + { + "left": "issue:43606", + "right": "issue:43701", + "accept": false, + "reason": "CPU offload device mismatch is unrelated to checkpoint resume key mapping." + }, + { + "left": "issue:43295", + "right": "issue:43825", + "accept": false, + "reason": "Different regressions: processor.tokenizer/images handling vs a pipeline error message about translation tasks." + }, + { + "left": "issue:42796", + "right": "issue:45290", + "accept": false, + "reason": "Tokenizer BPE-marker handling and chat-template crashes are distinct code paths and symptoms." + }, + { + "left": "issue:16998", + "right": "issue:22355", + "accept": false, + "reason": "Model max length question is unrelated to the missing transformers.onnx module import error." + }, + { + "left": "issue:43295", + "right": "issue:44871", + "accept": false, + "reason": "One is a processor/tokenizer regression; the other is an eos_token_id config inconsistency." + }, + { + "left": "issue:44871", + "right": "issue:45584", + "accept": false, + "reason": "Gemma eos_token_id mismatch and Whisper empty-transcription failure are different bugs." + }, + { + "left": "issue:43316", + "right": "issue:43716", + "accept": false, + "reason": "API discrepancy in config classes is unrelated to image preprocessor dtype mismatch." + }, + { + "left": "issue:42175", + "right": "issue:43316", + "accept": false, + "reason": "Packaging/backend dependency issue does not match a config API discrepancy." + }, + { + "left": "issue:45803", + "right": "issue:46365", + "accept": false, + "reason": "Bare except swallowing errors in Fuyu conversion is unrelated to Trainer import failure with datasets." + }, + { + "left": "issue:44610", + "right": "issue:46369", + "accept": false, + "reason": "Processor image-size mismatch and Sam3LiteText memory overflow are different model bugs." + }, + { + "left": "issue:44077", + "right": "issue:44534", + "accept": false, + "reason": "Optional post_init enforcement and junk non-persistent buffers are unrelated v5 issues." + }, + { + "left": "issue:43646", + "right": "issue:44913", + "accept": false, + "reason": "Generic custom model initialization breakage is not the same as GPTNeoX rotary_pct reload regression." + }, + { + "left": "issue:44871", + "right": "issue:46050", + "accept": false, + "reason": "Gemma eos_token_id inconsistency and sinq quantization roundtrip failure are separate serialization/loading problems." + }, + { + "left": "issue:45632", + "right": "issue:46424", + "accept": false, + "reason": "Cache-path collisions and max_cache_length behavior are different cache-related issues." + }, + { + "left": "issue:45803", + "right": "issue:46170", + "accept": false, + "reason": "Fuyu tensor conversion exception handling is unrelated to Windows/ROCm torch.distributed import errors." + }, + { + "left": "issue:44162", + "right": "issue:44792", + "accept": false, + "reason": "ESM2 model breakage and janus image-generation test failure are unrelated." + }, + { + "left": "issue:43452", + "right": "issue:44496", + "accept": false, + "reason": "gguf_file from_pretrained breakage is a different loading path from a missing model_type config error." + }, + { + "left": "issue:43065", + "right": "issue:44617", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder and Sam3Video OOM are not the same underlying problem." + }, + { + "left": "issue:43650", + "right": "issue:46202", + "accept": false, + "reason": "Placeholder issue title with no technical detail is unrelated to Gemma pipeline roundtrip behavior." + }, + { + "left": "issue:43404", + "right": "issue:43441", + "accept": false, + "reason": "Weight tying in Mistral3 is unrelated to FlashAttention support in Ministral-3." + }, + { + "left": "issue:43352", + "right": "issue:46640", + "accept": false, + "reason": "A specific model lacking Flash Attention 2 support is not enough to match a vague generic flash-attention issue." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Video-input ValueError in Qwen3OmniMoe is unrelated to batched chat-template padding crashes." + }, + { + "left": "issue:43761", + "right": "issue:44479", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and Qwen video-input regression affect different models and code paths." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 34, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:35141", + "issue:41950", + "issue:42490", + "issue:43065", + "issue:43388", + "issue:43475", + "issue:43504", + "issue:43525", + "issue:43531", + "issue:43572", + "issue:43644", + "issue:43673", + "issue:43688", + "issue:43761", + "issue:43827", + "issue:43931", + "issue:43994", + "issue:44060", + "issue:44190", + "issue:44242", + "issue:44479", + "issue:44534", + "issue:44610", + "issue:44617", + "issue:44734", + "issue:44743", + "issue:44792", + "issue:44855", + "issue:44913", + "issue:44964", + "issue:44977", + "issue:45127", + "issue:45200", + "issue:45381", + "issue:45406", + "issue:45542", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45600", + "issue:45632", + "issue:45698", + "issue:46178", + "issue:46234", + "issue:46369", + "issue:46421", + "issue:46617" + ], + "soft_pairs": [ + "issue:41950|issue:43475", + "issue:43388|issue:43688", + "issue:43572|issue:46178", + "issue:45127|issue:45200", + "issue:45542|issue:46234", + "issue:43504|issue:45698", + "issue:43673|issue:44743", + "issue:44792|issue:45600", + "issue:44792|issue:44977", + "issue:43525|issue:43531", + "issue:44534|issue:44913", + "issue:44617|issue:45584", + "issue:44734|issue:45406", + "issue:44190|issue:44964", + "issue:35141|issue:42490", + "issue:45632|issue:46617", + "issue:43931|issue:44060", + "issue:45381|issue:46369", + "issue:44855|issue:45588", + "issue:43761|issue:44242", + "issue:43673|issue:46421", + "issue:43827|issue:43994", + "issue:45563|issue:45584", + "issue:44479|issue:44734", + "issue:43644|issue:44913", + "issue:43065|issue:44610" + ], + "prepared_review_unit_hash": "a645066301604423111e5871037bf312c94ef8f96cf08724f3db4bf646c73c4f", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35043, + "estimated_input_tokens": 8761, + "estimated_eval_tokens": 17778 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab-bag of unrelated Transformers issues; the soft pairs share broad vocabulary (cache, loading, vision, v5 regressions) but not the same underlying bug, so I rejected all proposed duplicates. I used a broad, user-facing loading regression as the representative issue.", + "confidence": 0.38, + "canonical_issue_reason": "issue:45698 is the broadest representative here: it is a high-discussion, user-facing loading regression around from_pretrained/save_pretrained and custom modules, which makes it a better cluster anchor than the many narrow model-specific bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45698 is the most suitable single issue to represent the cluster because it is open, broadly impactful, and has the strongest evidence of recurring user impact (discussion/inbound refs) compared with the mostly isolated model- or version-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41950", + "right": "issue:43475", + "accept": false, + "reason": "Both are vision/video-adjacent, but one is a pipeline processor selection bug and the other is a missing attribute in Sam3 video outputs; different code paths." + }, + { + "left": "issue:43388", + "right": "issue:43688", + "accept": false, + "reason": "Metric gathering drops labels in the last batch; the other is auxiliary-loss normalization. Same training context, different bug." + }, + { + "left": "issue:43572", + "right": "issue:46178", + "accept": false, + "reason": "StableLm config missing pad_token_idx is unrelated to a missing tensorflow_text backend mapping." + }, + { + "left": "issue:45127", + "right": "issue:45200", + "accept": false, + "reason": "LoRA merge / tied embeddings corruption is a different problem from Gemma 4 mm_token_type_ids defaulting." + }, + { + "left": "issue:45542", + "right": "issue:46234", + "accept": false, + "reason": "Undefined tf backend due to tensorboard-only install is not the same as a missing import_utils availability helper." + }, + { + "left": "issue:43504", + "right": "issue:45698", + "accept": false, + "reason": "Legacy BEiT preset field failure is a model-config loading issue, but not the same as loading the wrong custom module after save_pretrained." + }, + { + "left": "issue:43673", + "right": "issue:44743", + "accept": false, + "reason": "Both mention cache, but one is chunked_prefill cache initialization and the other is recurrent state reset in a specific modular model." + }, + { + "left": "issue:44792", + "right": "issue:45600", + "accept": false, + "reason": "Janus image-generation test failure and stale auto_mappings entries are unrelated." + }, + { + "left": "issue:44792", + "right": "issue:44977", + "accept": false, + "reason": "Janus test failure is not the same as Qwen3.5 flash-attention generation misbehavior." + }, + { + "left": "issue:43525", + "right": "issue:43531", + "accept": false, + "reason": "Missing pad_token_id in Llama4Config and Qwen3-MoE sliding_window issues are different config bugs." + }, + { + "left": "issue:44534", + "right": "issue:44913", + "accept": false, + "reason": "Non-persistent buffer junk in v5 is unrelated to GPTNeoX rotary_pct not persisting on reload." + }, + { + "left": "issue:44617", + "right": "issue:45584", + "accept": false, + "reason": "Sam3Video OOM and Whisper empty-transcription failure are different model/runtime failures." + }, + { + "left": "issue:44734", + "right": "issue:45406", + "accept": false, + "reason": "KV-cache tensor indexing in transformers serve is unrelated to Gemma4Processor missing _tokenizer." + }, + { + "left": "issue:44190", + "right": "issue:44964", + "accept": false, + "reason": "Local dataset loading in a training script is unrelated to loading a multimodal checkpoint." + }, + { + "left": "issue:35141", + "right": "issue:42490", + "accept": false, + "reason": "Token embedding resize reinit is unrelated to path-with-trailing-slash import warnings." + }, + { + "left": "issue:45632", + "right": "issue:46617", + "accept": false, + "reason": "Both involve trust_remote_code cache usage, but one is cache-path collision and the other is FileNotFoundError from cache loading; not the same bug." + }, + { + "left": "issue:43931", + "right": "issue:44060", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch and Qwen3-Next tied-weights warning point to different model/config problems." + }, + { + "left": "issue:45381", + "right": "issue:46369", + "accept": false, + "reason": "Wrong video position ids in Qwen2.5-VL is unrelated to Sam3-LiteText memory overflow." + }, + { + "left": "issue:44855", + "right": "issue:45588", + "accept": false, + "reason": "Python 3.13 import parsing failure in DebertaV2Model is unrelated to a flash_attention s_aux=None crash." + }, + { + "left": "issue:43761", + "right": "issue:44242", + "accept": false, + "reason": "CLIP hidden_states regression and MoE load-balancing loss omission are different functional bugs." + }, + { + "left": "issue:43673", + "right": "issue:46421", + "accept": false, + "reason": "Both mention chunked prefill/cache, but one is a missing GenerationMixin cache and the other is spurious recompilation from lazy cache init." + }, + { + "left": "issue:43827", + "right": "issue:43994", + "accept": false, + "reason": "Docs still referencing pipeline() is unrelated to SigLIP2 producing bad outputs through AutoModel/pipeline." + }, + { + "left": "issue:45563", + "right": "issue:45584", + "accept": false, + "reason": "Paged generate warning about num_return_sequences is unrelated to Whisper empty transcription failure." + }, + { + "left": "issue:44479", + "right": "issue:44734", + "accept": false, + "reason": "Video-input regression across Qwen VL models is unrelated to KV-cache continuation tensor indexing in transformers serve." + }, + { + "left": "issue:43644", + "right": "issue:44913", + "accept": false, + "reason": "Non-persistent buffer junk in v5 and rotary_pct reset on reload are different serialization/regression issues." + }, + { + "left": "issue:43065", + "right": "issue:44610", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder is not the same as OmDet-Turbo processor/model resolution mismatch." + } + ] + }, + "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": 21, + "review_unit_index": 35, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:25251", + "issue:27957", + "issue:36246", + "issue:41628", + "issue:41720", + "issue:42796", + "issue:43116", + "issue:43295", + "issue:43335", + "issue:43540", + "issue:43550", + "issue:43643", + "issue:44077", + "issue:44112", + "issue:44190", + "issue:44246", + "issue:44462", + "issue:44464", + "issue:44496", + "issue:44521", + "issue:44617", + "issue:44734", + "issue:44737", + "issue:44779", + "issue:44855", + "issue:44877", + "issue:44933", + "issue:44938", + "issue:44964", + "issue:45003", + "issue:45127", + "issue:45245", + "issue:45276", + "issue:45357", + "issue:45362", + "issue:45381", + "issue:45542", + "issue:45563", + "issue:45584", + "issue:45600", + "issue:45632", + "issue:45803", + "issue:45907", + "issue:46082", + "issue:46121", + "issue:46234", + "issue:46369", + "issue:46489" + ], + "soft_pairs": [ + "issue:25251|issue:27957", + "issue:44933|issue:45362", + "issue:41720|issue:45245", + "issue:43335|issue:44855", + "issue:43550|issue:44190", + "issue:44077|issue:44877", + "issue:44464|issue:45563", + "issue:44617|issue:46369", + "issue:43335|issue:44190", + "issue:45127|issue:45357", + "issue:44521|issue:45381", + "issue:44112|issue:45600", + "issue:43116|issue:44190", + "issue:44779|issue:46489", + "issue:43540|issue:44734", + "issue:43295|issue:45276", + "issue:44737|issue:45803", + "issue:44938|issue:45542", + "issue:45276|issue:45584", + "issue:45632|issue:46234", + "issue:44462|issue:45907", + "issue:42796|issue:44521", + "issue:44246|issue:45003", + "issue:36246|issue:41628", + "issue:46082|issue:46121", + "issue:44964|issue:45600", + "issue:45245|issue:45584", + "issue:43643|issue:44496" + ], + "prepared_review_unit_hash": "b00f853ac23f83dea1c1b24c12ea9fe08effb8b2729b5866819cefb1ec75f75d", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36203, + "estimated_input_tokens": 9051, + "estimated_eval_tokens": 18358 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous transformer bug reports with only one clear duplicate-like pair. The only strong match is the DeepSeek tokenizer regression pair; the rest differ by model, subsystem, or failure mode.", + "confidence": 0.91, + "canonical_issue_reason": "issue:44779 is the broader DeepSeek tokenizer regression report, while issue:46489 appears to be a narrower follow-up about the same v5 tokenizer breakage and explicitly references a gap in PR #44801's fix.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44779 is the best representative because it states the underlying regression generically and covers the wider DeepSeek tokenizer failure surface.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:25251", + "right": "issue:27957", + "accept": false, + "reason": "Different problems: pipeline top_k output shape vs XLM-RoBERTa Flash Attention 2 support." + }, + { + "left": "issue:44933", + "right": "issue:45362", + "accept": false, + "reason": "Unrelated: missing image_utils import vs Qwen3.5 chat crash." + }, + { + "left": "issue:41720", + "right": "issue:45245", + "accept": false, + "reason": "Different failure modes and subsystems: Qwen3 device-mapping CUDA assert vs category-limit RuntimeError." + }, + { + "left": "issue:43335", + "right": "issue:44855", + "accept": false, + "reason": "Different code paths: SwitchTransformers sparse-layer config bug vs DebertaV2 Python 3.13 import parsing issue." + }, + { + "left": "issue:43550", + "right": "issue:44190", + "accept": false, + "reason": "Unrelated: torch.compile/SDPA model failure vs example script dataset loading bug." + }, + { + "left": "issue:44077", + "right": "issue:44877", + "accept": false, + "reason": "Both touch config handling, but the concrete bugs are different models and different validation failures." + }, + { + "left": "issue:44464", + "right": "issue:45563", + "accept": false, + "reason": "Different generation issues: compiled forward inconsistency vs stale warning in paged generate." + }, + { + "left": "issue:44617", + "right": "issue:46369", + "accept": false, + "reason": "Same broad model family, but different products and symptoms; not clearly the same underlying bug." + }, + { + "left": "issue:43335", + "right": "issue:44190", + "accept": false, + "reason": "Different subsystems: model config construction vs example-script dataset loading." + }, + { + "left": "issue:45127", + "right": "issue:45357", + "accept": false, + "reason": "Different regression surfaces: LoRA vocab merging collapse vs incorrect visual encoder keys on save_pretrained." + }, + { + "left": "issue:44521", + "right": "issue:45381", + "accept": false, + "reason": "Both multimodal-related, but assistant-mask generation and vision_position_ids are distinct bugs." + }, + { + "left": "issue:44112", + "right": "issue:45600", + "accept": false, + "reason": "Both are CI/test maintenance issues, but they affect different model mappings and different failures." + }, + { + "left": "issue:43116", + "right": "issue:44190", + "accept": false, + "reason": "Different example scripts and different behaviors: multi-label outputs vs image-classification dataset loading." + }, + { + "left": "issue:44779", + "right": "issue:46489", + "accept": true, + "reason": "Same underlying DeepSeek tokenizer regression in v5; 46489 is a narrower report that points to the same breakage and references the prior fix gap." + }, + { + "left": "issue:43540", + "right": "issue:44734", + "accept": false, + "reason": "Different runtime paths: multimodal video input processing vs serve KV-cache continuation indexing." + }, + { + "left": "issue:43295", + "right": "issue:45276", + "accept": false, + "reason": "Both are regressions, but one is processor/tokenizer API breakage and the other is embedding resize propagation." + }, + { + "left": "issue:44737", + "right": "issue:45803", + "accept": false, + "reason": "Different bugs: XLNet device placement inefficiency vs Fuyu exception handling." + }, + { + "left": "issue:44938", + "right": "issue:45542", + "accept": false, + "reason": "Python 3.14 import failure is unrelated to the TensorFlow/tensorboard backend mismatch." + }, + { + "left": "issue:45276", + "right": "issue:45584", + "accept": false, + "reason": "Unrelated: token embedding resize propagation vs Whisper empty-transcription generation failure." + }, + { + "left": "issue:45632", + "right": "issue:46234", + "accept": false, + "reason": "Different areas: trust_remote_code cache collision vs missing import_utils availability helper." + }, + { + "left": "issue:44462", + "right": "issue:45907", + "accept": false, + "reason": "Tokenizer-file resolution bug vs list_repo_files revision forwarding are not the same underlying issue." + }, + { + "left": "issue:42796", + "right": "issue:44521", + "accept": false, + "reason": "Different tokenization/template problems affecting different modalities and APIs." + }, + { + "left": "issue:44246", + "right": "issue:45003", + "accept": false, + "reason": "Both relate to import-time behavior, but one is slowness and the other is unsafe sys.modules access." + }, + { + "left": "issue:36246", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but for different symbols and different model areas." + }, + { + "left": "issue:46082", + "right": "issue:46121", + "accept": false, + "reason": "Different config-validation bugs: Llama head_dim constraints vs rope validation list handling." + }, + { + "left": "issue:44964", + "right": "issue:45600", + "accept": false, + "reason": "Different failures: model loading for Phi-4 multimodal vs removed Sam3LiteText references in auto_mappings CI." + }, + { + "left": "issue:45245", + "right": "issue:45584", + "accept": false, + "reason": "Category-count runtime limit vs empty-transcription generation are unrelated." + }, + { + "left": "issue:43643", + "right": "issue:44496", + "accept": false, + "reason": "Related to config loading, but one is trust_remote_code field loss and the other is unrecognized model/config metadata." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The clustering looks grounded in the packet, and the duplicate call between issue:44779 and issue:46489 is conservative and well-supported by the titles. The other soft-edge rejections stay cautious and do not overclaim cross-issue equivalence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 36, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:27957", + "issue:32090", + "issue:35141", + "issue:36246", + "issue:38617", + "issue:41950", + "issue:42175", + "issue:42907", + "issue:43295", + "issue:43381", + "issue:43404", + "issue:43441", + "issue:43479", + "issue:43526", + "issue:43531", + "issue:43650", + "issue:43673", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44062", + "issue:44112", + "issue:44186", + "issue:44220", + "issue:44279", + "issue:44295", + "issue:44336", + "issue:44496", + "issue:44683", + "issue:44734", + "issue:44792", + "issue:44964", + "issue:44987", + "issue:45042", + "issue:45083", + "issue:45216", + "issue:45335", + "issue:45375", + "issue:45446", + "issue:45588", + "issue:45593", + "issue:45600", + "issue:45759", + "issue:46123", + "issue:46369", + "issue:46424", + "issue:46506" + ], + "soft_pairs": [ + "issue:42175|issue:44279", + "issue:36246|issue:38617", + "issue:44112|issue:44964", + "issue:43854|issue:44964", + "issue:44683|issue:45446", + "issue:27957|issue:35141", + "issue:43404|issue:45216", + "issue:44295|issue:46506", + "issue:44186|issue:44734", + "issue:41950|issue:44016", + "issue:43650|issue:44220", + "issue:43673|issue:46424", + "issue:32090|issue:35141", + "issue:43479|issue:45375", + "issue:44016|issue:44220", + "issue:43531|issue:45600", + "issue:45083|issue:46506", + "issue:43295|issue:45335", + "issue:43976|issue:44987", + "issue:43526|issue:43901", + "issue:44496|issue:45042", + "issue:42907|issue:43441", + "issue:43901|issue:44496", + "issue:43381|issue:45593", + "issue:43673|issue:44336", + "issue:43976|issue:44279", + "issue:43441|issue:45588", + "issue:44016|issue:44062", + "issue:46123|issue:46369", + "issue:44792|issue:46424", + "issue:44496|issue:45759" + ], + "prepared_review_unit_hash": "8f7103a18a80e7f95256b9b8f151542fefbfa825ea79066e96d4a8dd5486d512", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37044, + "estimated_input_tokens": 9261, + "estimated_eval_tokens": 18778 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Loose similarity cluster with many unrelated issues; none of the soft edges look like true duplicates. The set spans different models, configs, pipelines, and runtime failures rather than one underlying bug.", + "confidence": 0.9, + "canonical_issue_reason": "No single canonical duplicate stands out because the cluster is heterogeneous. If a representative is needed, issue 44964 is the closest fit: it is a concrete, well-scoped model-loading regression with a detailed repro.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44964 is the best manual anchor in this set because it is specific, reproducible, and well described, but it does not unify the rest of the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42175", + "right": "issue:44279", + "accept": false, + "reason": "TensorFlow backend packaging vs a generic dependency issue; different failure modes and likely different fixes." + }, + { + "left": "issue:36246", + "right": "issue:38617", + "accept": false, + "reason": "Different missing imports from different modules (`qwen2_5_vl` vs `configuration_utils`); unrelated code paths." + }, + { + "left": "issue:44112", + "right": "issue:44964", + "accept": false, + "reason": "CI test flake in GraniteSpeech vs model loading failure for Phi-4 multimodal; no shared root bug." + }, + { + "left": "issue:43854", + "right": "issue:44964", + "accept": false, + "reason": "Both are model-loading reports, but for different models and different likely causes; not the same bug." + }, + { + "left": "issue:44683", + "right": "issue:45446", + "accept": false, + "reason": "Torch version / compiled flex attention issue vs an incorrect version check for a different import path." + }, + { + "left": "issue:27957", + "right": "issue:35141", + "accept": false, + "reason": "Flash Attention support for XLM-Roberta is unrelated to token embedding reinitialization after resize." + }, + { + "left": "issue:43404", + "right": "issue:45216", + "accept": false, + "reason": "Weight tying bug in Mistral3 vs checkpoint save regression in Qwen3.5; different model families and code paths." + }, + { + "left": "issue:44295", + "right": "issue:46506", + "accept": false, + "reason": "Buffer/position_ids access bug vs CPU/CUDA numerical divergence; unrelated symptoms." + }, + { + "left": "issue:44186", + "right": "issue:44734", + "accept": false, + "reason": "Tokenizer crash on NER padding/truncation vs serve KV-cache indexing bug; different layers and fixes." + }, + { + "left": "issue:41950", + "right": "issue:44016", + "accept": false, + "reason": "Pipeline image-processor lookup bug vs a notebook syntax error; not the same underlying issue." + }, + { + "left": "issue:43650", + "right": "issue:44220", + "accept": false, + "reason": "Vague request for data vs fbank feature extraction bug; no common failure." + }, + { + "left": "issue:43673", + "right": "issue:46424", + "accept": false, + "reason": "Chunked prefill cache regression vs max_cache_length only applying to Quantized Cache; related area but distinct behavior." + }, + { + "left": "issue:32090", + "right": "issue:35141", + "accept": false, + "reason": "Trainer broadcast NoneType error vs output embedding reinitialization after resize; different subsystems." + }, + { + "left": "issue:43479", + "right": "issue:45375", + "accept": false, + "reason": "Different multimodal config validation bugs on different fields/models; not mergeable as one fix." + }, + { + "left": "issue:44016", + "right": "issue:44220", + "accept": false, + "reason": "Notebook syntax error and feature extraction bug are unrelated." + }, + { + "left": "issue:43531", + "right": "issue:45600", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior vs stale auto_mappings references to removed configs; unrelated." + }, + { + "left": "issue:45083", + "right": "issue:46506", + "accept": false, + "reason": "Helper length calculation in qwen3_omni_moe vs CPU/CUDA output mismatch; different bugs." + }, + { + "left": "issue:43295", + "right": "issue:45335", + "accept": false, + "reason": "Processor/tokenizer regression vs `resize_token_embeddings` not affecting decoder embeddings; different code paths." + }, + { + "left": "issue:43976", + "right": "issue:44987", + "accept": false, + "reason": "Python version compatibility issue vs loading a specific model failing on transformers>=5.1.0; not the same bug." + }, + { + "left": "issue:43526", + "right": "issue:43901", + "accept": false, + "reason": "BeitImageProcessorFast label handling vs docs mentioning `return_all_scores`; one is runtime behavior, the other documentation drift." + }, + { + "left": "issue:44496", + "right": "issue:45042", + "accept": false, + "reason": "Unrecognized model config vs PIL image processors requiring torchvision; unrelated model-loading and dependency issues." + }, + { + "left": "issue:42907", + "right": "issue:43441", + "accept": false, + "reason": "Saving dequantized Ministrals/Devstrals vs FlashAttention failure in Ministral-3 RC; same family name but different concrete failures." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "Documentation mismatch vs unrecognized model config; no shared bug." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Gradient checkpointing in eval mode vs missing auxiliary losses when denoising is off; unrelated behavior." + }, + { + "left": "issue:43673", + "right": "issue:44336", + "accept": false, + "reason": "Cache handling bug vs ANSI codes emitted in loading report; different components entirely." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Python support regression vs generic dependency issue; no common fix." + }, + { + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "Both mention flash attention, but one is a model-specific RC failure and the other is an AttributeError on `s_aux=None`; not enough to be the same bug." + }, + { + "left": "issue:44016", + "right": "issue:44062", + "accept": false, + "reason": "Notebook syntax error vs tokenizer `AddedToken` argument conflict; unrelated." + }, + { + "left": "issue:46123", + "right": "issue:46369", + "accept": false, + "reason": "Partial-batch mask generation bug vs memory overflow in sam3-litetext; different symptoms and fixes." + }, + { + "left": "issue:44792", + "right": "issue:46424", + "accept": false, + "reason": "Janus image generation test failure vs cache max length semantics; unrelated." + }, + { + "left": "issue:44496", + "right": "issue:45759", + "accept": false, + "reason": "Unrecognized model type vs `from_config` not unwrapping `text_config`; both loading-related but distinct model/config bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 37, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:38175", + "issue:39692", + "issue:41950", + "issue:42796", + "issue:43295", + "issue:43299", + "issue:43479", + "issue:43525", + "issue:43577", + "issue:43650", + "issue:43653", + "issue:43824", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:44075", + "issue:44242", + "issue:44279", + "issue:44351", + "issue:44368", + "issue:44466", + "issue:44521", + "issue:44617", + "issue:44683", + "issue:44779", + "issue:44933", + "issue:45020", + "issue:45103", + "issue:45200", + "issue:45216", + "issue:45276", + "issue:45278", + "issue:45325", + "issue:45335", + "issue:45357", + "issue:45372", + "issue:45405", + "issue:45542", + "issue:45584", + "issue:45593", + "issue:45812", + "issue:45823", + "issue:46082", + "issue:46209", + "issue:46399", + "issue:46512", + "issue:46515" + ], + "soft_pairs": [ + "issue:43824|issue:44933", + "issue:43299|issue:45216", + "issue:44279|issue:46515", + "issue:45335|issue:45584", + "issue:44075|issue:44368", + "issue:46082|issue:46399", + "issue:44279|issue:45405", + "issue:42796|issue:43650", + "issue:43577|issue:45216", + "issue:39692|issue:43901", + "issue:43295|issue:44779", + "issue:41950|issue:45020", + "issue:45103|issue:45372", + "issue:44351|issue:45278", + "issue:43299|issue:45357", + "issue:43976|issue:45405", + "issue:44242|issue:45200", + "issue:38175|issue:44617", + "issue:46082|issue:46512", + "issue:41950|issue:45823", + "issue:44779|issue:45812", + "issue:44683|issue:45542", + "issue:43299|issue:43854", + "issue:43479|issue:43525", + "issue:43650|issue:46209", + "issue:44466|issue:45276", + "issue:44521|issue:45325", + "issue:43653|issue:45593" + ], + "prepared_review_unit_hash": "acad1401fae057a8be4bf6f9303eb0c260c698c4623abec2a2271a58d683bb57", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35558, + "estimated_input_tokens": 8890, + "estimated_eval_tokens": 18036 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: it mixes unrelated tokenizer, config, model-loading, pipeline, documentation, and dependency/version regression reports. None of the soft edges look like the same underlying bug, so no duplicate merge is justified.", + "confidence": 0.97, + "canonical_issue_reason": "44521 ('apply_chat_template returns all-zero assistant_masks for multimodal inputs') is the closest thing to a canonical issue because it is the most concrete, reproducible, and currently open bug in the set.", + "canonical_pr_reason": null, + "best_issue_reason": "44521 is the best issue to keep as the representative problem: narrow scope, clear failure mode, and actionable reproduction details.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43824", + "right": "issue:44933", + "accept": false, + "reason": "Both are import-related, but they fail on different symbols and different code paths: Qwen2.5-VL class import vs a bad image_utils import." + }, + { + "left": "issue:43299", + "right": "issue:45216", + "accept": false, + "reason": "Both affect Qwen models, but one is a model-loading failure for Qwen3VL MoE and the other is an incorrect save_pretrained checkpoint for Qwen3.5." + }, + { + "left": "issue:44279", + "right": "issue:46515", + "accept": false, + "reason": "These are unrelated dependency/version complaints: a generic transformers dependency issue vs torch 2.4.1 incompatibility." + }, + { + "left": "issue:45335", + "right": "issue:45584", + "accept": false, + "reason": "Different models and different bugs: token embedding resize not propagating in t5gemma vs Whisper generation failure after align_special_tokens." + }, + { + "left": "issue:44075", + "right": "issue:44368", + "accept": false, + "reason": "Optimizer SGD args being ignored is unrelated to a Qwen3.5 LoRA warning about tie_word_embeddings." + }, + { + "left": "issue:46082", + "right": "issue:46399", + "accept": false, + "reason": "LlamaConfig head_dim validation and Qwen3.5 rotary-kernel registration are separate config/code-path bugs." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "A generic dependency issue is not the same as bumping MIN_PEFT_VERSION to an unreleased release." + }, + { + "left": "issue:42796", + "right": "issue:43650", + "accept": false, + "reason": "Tokenizer BPE-marker handling and an extremely vague 'ADD THE DATA' report do not describe the same bug." + }, + { + "left": "issue:43577", + "right": "issue:45216", + "accept": false, + "reason": "Blip2 dtype propagation during loading and Qwen3.5 checkpoint serialization are different regression points." + }, + { + "left": "issue:39692", + "right": "issue:43901", + "accept": false, + "reason": "Both mention docs, but one is a SigLIP2 example with multiple runtime errors while the other is a TextClassificationPipeline docs mismatch." + }, + { + "left": "issue:43295", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer-related regressions, but they involve different model families and distinct failure modes." + }, + { + "left": "issue:41950", + "right": "issue:45020", + "accept": false, + "reason": "A video-classification pipeline processor lookup bug is unrelated to remote_code model breakage." + }, + { + "left": "issue:45103", + "right": "issue:45372", + "accept": false, + "reason": "auto_docstring crashing on future annotations and Gemma 4 processor loading failing on a mistral_common import are separate issues." + }, + { + "left": "issue:44351", + "right": "issue:45278", + "accept": false, + "reason": "A missing HybridCache import and broad import failures after upgrade are related only at a very high level, not the same concrete bug." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Both are Qwen3.5/Qwen3VL regressions, but one is model loading and the other is save_pretrained writing wrong visual encoder keys." + }, + { + "left": "issue:43976", + "right": "issue:45405", + "accept": false, + "reason": "Python version compatibility and an unreleased PEFT version bump are unrelated release/dependency problems." + }, + { + "left": "issue:44242", + "right": "issue:45200", + "accept": false, + "reason": "Missing load-balancing loss when output_router_logits=False is not the same as mm_token_type_ids defaulting to zeros." + }, + { + "left": "issue:38175", + "right": "issue:44617", + "accept": false, + "reason": "SigLIP2 zero probabilities and Sam3Video CUDA OOM are completely different model/runtime problems." + }, + { + "left": "issue:46082", + "right": "issue:46512", + "accept": false, + "reason": "Config head_dim validation and stale model.config.dtype after from_config(dtype=...) are separate config issues." + }, + { + "left": "issue:41950", + "right": "issue:45823", + "accept": false, + "reason": "A pipeline processor lookup bug and Gemma4 PLE device mismatch are not the same underlying defect." + }, + { + "left": "issue:44779", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and do not clearly share the same concrete code-path bug." + }, + { + "left": "issue:44683", + "right": "issue:45542", + "accept": false, + "reason": "Compiled flex_attention on torch>=2.9 and a missing TensorFlow backend check are unrelated." + }, + { + "left": "issue:43299", + "right": "issue:43854", + "accept": false, + "reason": "Both mention Qwen VL models, but one is a load failure and the other is an import-name failure for a different class." + }, + { + "left": "issue:43479", + "right": "issue:43525", + "accept": false, + "reason": "Phi4MultimodalConfig default-init behavior and Llama4Config missing pad_token_id are different configuration bugs." + }, + { + "left": "issue:43650", + "right": "issue:46209", + "accept": false, + "reason": "The first issue is too vague to match, and the second is a specific FP8 inference failure on B200." + }, + { + "left": "issue:44466", + "right": "issue:45276", + "accept": false, + "reason": "Both involve weights/embeddings, but one is tied-weight serialization varying by device and the other is resize_token_embeddings not updating Gemma4 embeddings." + }, + { + "left": "issue:44521", + "right": "issue:45325", + "accept": false, + "reason": "Both are multimodal, but assistant mask generation and rope-index temporal scaling are distinct preprocessing bugs." + }, + { + "left": "issue:43653", + "right": "issue:45593", + "accept": false, + "reason": "BigBirdTokenizer special-token registration and D-FINE auxiliary-loss gating 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": 21, + "review_unit_index": 38, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:33290", + "issue:42175", + "issue:42994", + "issue:43317", + "issue:43441", + "issue:43525", + "issue:43643", + "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:44246", + "issue:44261", + "issue:44263", + "issue:44291", + "issue:44297", + "issue:44336", + "issue:44361", + "issue:44373", + "issue:44462", + "issue:44589", + "issue:44623", + "issue:44671", + "issue:44683", + "issue:44749", + "issue:44779", + "issue:45081", + "issue:45362", + "issue:45397", + "issue:45542", + "issue:45593", + "issue:45600", + "issue:45656", + "issue:45712", + "issue:45800", + "issue:45823", + "issue:45834", + "issue:45907", + "issue:46291", + "issue:46399" + ], + "soft_pairs": [ + "issue:44016|issue:44291", + "issue:44462|issue:45081", + "issue:43441|issue:45834", + "issue:42994|issue:43317", + "issue:33290|issue:45656", + "issue:44361|issue:45593", + "issue:44623|issue:44749", + "issue:44261|issue:44263", + "issue:45600|issue:45712", + "issue:45081|issue:45907", + "issue:44623|issue:45397", + "issue:42175|issue:43976", + "issue:45542|issue:45800", + "issue:43901|issue:43994", + "issue:44671|issue:44779", + "issue:44016|issue:45823", + "issue:43441|issue:44683", + "issue:43723|issue:44162", + "issue:43901|issue:44246", + "issue:43673|issue:43761", + "issue:43525|issue:43881", + "issue:46291|issue:46399", + "issue:43824|issue:43976", + "issue:43525|issue:43643", + "issue:44188|issue:44336", + "issue:44589|issue:45362", + "issue:44297|issue:45397", + "issue:43784|issue:45542", + "issue:44336|issue:44373" + ], + "prepared_review_unit_hash": "cebb9a70916ea6c317f7bfbd6857f4df7ede0c0c082ceb6663d8057068353c59", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36254, + "estimated_input_tokens": 9064, + "estimated_eval_tokens": 18384 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose collection of mostly unrelated Transformers issues across loading, tokenization, attention, packaging, docs, and model-specific regressions. None of the soft pairs appears to be the same underlying bug/change, so no duplicate merges are justified.", + "confidence": 0.91, + "canonical_issue_reason": "No single issue is a clear duplicate hub for this set; if one representative must be chosen, issue 45656 is the clearest current, actionable bug report.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45656 is the best representative because it is a specific, reproducible, currently open deepspeed bug, while most other items are orthogonal or less central.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44016", + "right": "issue:44291", + "accept": false, + "reason": "Notebook syntax error vs init_empty_weights TypeError: different code paths and symptoms." + }, + { + "left": "issue:44462", + "right": "issue:45081", + "accept": false, + "reason": "AutoTokenizer ignoring tokenizer.json is unrelated to Mistral regex patching crash." + }, + { + "left": "issue:43441", + "right": "issue:45834", + "accept": false, + "reason": "Different models and failures: FlashAttention regression vs OCR long-input index error." + }, + { + "left": "issue:42994", + "right": "issue:43317", + "accept": false, + "reason": "Both involve quantization, but save failure and dequantized-load/offload failure are different bugs." + }, + { + "left": "issue:33290", + "right": "issue:45656", + "accept": false, + "reason": "OOM with Adafactor under deepspeed is not the same as optimizer.step being called twice." + }, + { + "left": "issue:44361", + "right": "issue:45593", + "accept": false, + "reason": "Tokenizer AttributeError vs missing auxiliary losses in a vision model." + }, + { + "left": "issue:44623", + "right": "issue:44749", + "accept": false, + "reason": "Processor save_pretrained missing files is unrelated to data filtering slowdown." + }, + { + "left": "issue:44261", + "right": "issue:44263", + "accept": false, + "reason": "Different GLM-related internals: layernorm config precision issue vs torch.split return values." + }, + { + "left": "issue:45600", + "right": "issue:45712", + "accept": false, + "reason": "Both are repo-maintenance issues, but they affect different files and failure modes." + }, + { + "left": "issue:45081", + "right": "issue:45907", + "accept": false, + "reason": "Mistral tokenizer regex crash is unrelated to list_repo_files revision forwarding." + }, + { + "left": "issue:44623", + "right": "issue:45397", + "accept": false, + "reason": "Generic processor save bug does not match gemma-4 zero3 loading failure." + }, + { + "left": "issue:42175", + "right": "issue:43976", + "accept": false, + "reason": "Packaging missing TensorFlow backend vs Python version incompatibility are distinct environment issues." + }, + { + "left": "issue:45542", + "right": "issue:45800", + "accept": false, + "reason": "Missing tf backend from partial install is not the same as torch/transformers version incompatibility." + }, + { + "left": "issue:43901", + "right": "issue:43994", + "accept": false, + "reason": "Documentation mismatch for return_all_scores is unrelated to incorrect SigLIP2 predictions." + }, + { + "left": "issue:44671", + "right": "issue:44779", + "accept": false, + "reason": "CamemBERT masked-LM wrong outputs and DeepSeek tokenizer regression are different subsystems." + }, + { + "left": "issue:44016", + "right": "issue:45823", + "accept": false, + "reason": "Notebook syntax error and Gemma4 PLE device mismatch are unrelated." + }, + { + "left": "issue:43441", + "right": "issue:44683", + "accept": false, + "reason": "FlashAttention/model loading issue is not the same as compiled flex_attention failing on newer torch." + }, + { + "left": "issue:43723", + "right": "issue:44162", + "accept": false, + "reason": "Tokenizer loading regression and ESM2 model breakage are separate bugs." + }, + { + "left": "issue:43901", + "right": "issue:44246", + "accept": false, + "reason": "Docs issue and slow import issue do not share the same root cause." + }, + { + "left": "issue:43673", + "right": "issue:43761", + "accept": false, + "reason": "Chunked prefill cache regression and CLIP hidden_states bug affect different paths." + }, + { + "left": "issue:43525", + "right": "issue:43881", + "accept": false, + "reason": "Different model loading failures; one is pad_token_id missing, the other is a GLM4V load problem." + }, + { + "left": "issue:46291", + "right": "issue:46399", + "accept": false, + "reason": "Kernels revision/version requirement failure is not the same as missing registration decorator in apply_rotary_pos_emb." + }, + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "Qwen2.5-VL import error and Python version incompatibility are unrelated." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": false, + "reason": "Llama4Config pad_token_id error and trust_remote_code missing fields are different API bugs." + }, + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "Torch.compile attention-kernel divergence is unrelated to ANSI codes in loading_report." + }, + { + "left": "issue:44589", + "right": "issue:45362", + "accept": false, + "reason": "Missing Float8 storage support and Qwen3.5 chat crash are different failure modes." + }, + { + "left": "issue:44297", + "right": "issue:45397", + "accept": false, + "reason": "Tokenizer save metadata mismatch is unrelated to gemma-4 zero3 loading." + }, + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "sentence-transformers import NameError and TensorFlow backend install issue are not the same bug." + }, + { + "left": "issue:44336", + "right": "issue:44373", + "accept": false, + "reason": "ANSI color output bug and wrong docstring 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": 21, + "review_unit_index": 39, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:36246", + "issue:41628", + "issue:41950", + "issue:42491", + "issue:42757", + "issue:42886", + "issue:43257", + "issue:43262", + "issue:43299", + "issue:43475", + "issue:43526", + "issue:43531", + "issue:43550", + "issue:43596", + "issue:43673", + "issue:43901", + "issue:43976", + "issue:44162", + "issue:44220", + "issue:44242", + "issue:44246", + "issue:44295", + "issue:44336", + "issue:44351", + "issue:44485", + "issue:44496", + "issue:44561", + "issue:44734", + "issue:44749", + "issue:44779", + "issue:44857", + "issue:44871", + "issue:45003", + "issue:45103", + "issue:45137", + "issue:45200", + "issue:45292", + "issue:45362", + "issue:45397", + "issue:45561", + "issue:45563", + "issue:45593", + "issue:45859", + "issue:46132", + "issue:46369", + "issue:46399", + "issue:46424", + "issue:46506" + ], + "soft_pairs": [ + "issue:41628|issue:43976", + "issue:44246|issue:44496", + "issue:45561|issue:45563", + "issue:45103|issue:46399", + "issue:44242|issue:45593", + "issue:42491|issue:43299", + "issue:44295|issue:45397", + "issue:43526|issue:45003", + "issue:43531|issue:45362", + "issue:44246|issue:44871", + "issue:44336|issue:44485", + "issue:41950|issue:44220", + "issue:43550|issue:46506", + "issue:36246|issue:42757", + "issue:36246|issue:44351", + "issue:44336|issue:46424", + "issue:42886|issue:46132", + "issue:44779|issue:45292", + "issue:44734|issue:46424", + "issue:45859|issue:46399", + "issue:45003|issue:45561", + "issue:43901|issue:45003", + "issue:43475|issue:46369", + "issue:43257|issue:45137", + "issue:44242|issue:44857", + "issue:36246|issue:43596", + "issue:44295|issue:44749", + "issue:44496|issue:45003", + "issue:43673|issue:44561", + "issue:44162|issue:45200", + "issue:43262|issue:44246" + ], + "prepared_review_unit_hash": "588b074d9afce8e533688590c90d8856d29a09ec989fa72619039350270de465", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36905, + "estimated_input_tokens": 9227, + "estimated_eval_tokens": 18710 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous Transformers bugs, import errors, and model-specific regressions. I only see one plausible duplicate pocket: the Qwen3-MoE loading breakage around the hf4.x\u2192hf5.x / v5 transition.", + "confidence": 0.81, + "canonical_issue_reason": "No single canonical issue fits the whole cluster; 42491 is the closest anchor because it is the broadest Qwen3-MoE loading regression report and best matches the only likely duplicate pocket.", + "canonical_pr_reason": null, + "best_issue_reason": "42491 is the best representative issue for the cluster's one near-duplicate area (Qwen3-MoE loading incompatibility on the new branch/version).", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41628", + "right": "issue:43976", + "accept": false, + "reason": "Both are import-related, but one is a missing AutoImageProcessor import and the other is a Python-version/package-compatibility failure; not the same bug." + }, + { + "left": "issue:44246", + "right": "issue:44496", + "accept": false, + "reason": "Import slowdown and unrecognized-model/config parsing are unrelated failure modes." + }, + { + "left": "issue:45561", + "right": "issue:45563", + "accept": false, + "reason": "A pytest-xdist file race and a stale generation warning are unrelated issues." + }, + { + "left": "issue:45103", + "right": "issue:46399", + "accept": false, + "reason": "Auto-docstring annotation handling and Qwen3.5 kernel registration are different code paths and symptoms." + }, + { + "left": "issue:44242", + "right": "issue:45593", + "accept": false, + "reason": "Both mention missing losses, but they are different model families and different auxiliary-loss mechanisms." + }, + { + "left": "issue:42491", + "right": "issue:43299", + "accept": true, + "reason": "Both report Qwen3-MoE loading incompatibility after the hf4.x/main \u2192 hf5.x/v5 transition; they look like the same regression." + }, + { + "left": "issue:44295", + "right": "issue:45397", + "accept": false, + "reason": "A position_ids buffer error and a Gemma-4 ZeRO3 from_pretrained failure are not the same concrete bug." + }, + { + "left": "issue:43526", + "right": "issue:45003", + "accept": false, + "reason": "A BeitImageProcessorFast label reduction bug and unsafe sys.modules access are unrelated." + }, + { + "left": "issue:43531", + "right": "issue:45362", + "accept": false, + "reason": "Both are Qwen3-related, but one is a sliding_window issue and the other is a chat loading crash; too different to merge." + }, + { + "left": "issue:44246", + "right": "issue:44871", + "accept": false, + "reason": "Import performance and Gemma eos_token configuration are unrelated." + }, + { + "left": "issue:44336", + "right": "issue:44485", + "accept": false, + "reason": "ANSI color handling in loading_report and GLM-5 RoPE implementation are unrelated." + }, + { + "left": "issue:41950", + "right": "issue:44220", + "accept": false, + "reason": "Video-classification image-processor lookup and audio fbank extraction are different preprocessing bugs." + }, + { + "left": "issue:43550", + "right": "issue:46506", + "accept": false, + "reason": "torch.compile+SDPA breakage and CPU/CUDA result divergence are distinct model/runtime problems." + }, + { + "left": "issue:36246", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols and likely different packaging regressions." + }, + { + "left": "issue:36246", + "right": "issue:44351", + "accept": false, + "reason": "Different missing exports (Qwen2.5 VL image processor vs HybridCache); not the same underlying issue." + }, + { + "left": "issue:44336", + "right": "issue:46424", + "accept": false, + "reason": "Terminal ANSI output and cache-length semantics are unrelated." + }, + { + "left": "issue:42886", + "right": "issue:46132", + "accept": false, + "reason": "Offline cache loading and AttentionInterface.register behavior are unrelated." + }, + { + "left": "issue:44779", + "right": "issue:45292", + "accept": false, + "reason": "A DeepSeek tokenizer regression and resize_token_embeddings not updating output embeddings are different bugs." + }, + { + "left": "issue:44734", + "right": "issue:46424", + "accept": false, + "reason": "KV-cache continuation indexing and max_cache_length honoring are different cache issues." + }, + { + "left": "issue:45859", + "right": "issue:46399", + "accept": false, + "reason": "Both are Qwen3.5/MoE-related, but one is CPU-offload incompatibility and the other is a registration/decorator crash; not the same bug." + }, + { + "left": "issue:45003", + "right": "issue:45561", + "accept": false, + "reason": "Unsafe sys.modules access and an xdist file race are unrelated." + }, + { + "left": "issue:43901", + "right": "issue:45003", + "accept": false, + "reason": "Documentation drift about return_all_scores is unrelated to sys.modules access." + }, + { + "left": "issue:43475", + "right": "issue:46369", + "accept": false, + "reason": "A missing attribute in Sam3VisionEncoderOutput and a SAM3 LiteText memory overflow are different failures." + }, + { + "left": "issue:43257", + "right": "issue:45137", + "accept": false, + "reason": "Qwen3 MOE weight conversion and DeepSpeed ZeRO3 deque/index errors are different problems." + }, + { + "left": "issue:44242", + "right": "issue:44857", + "accept": false, + "reason": "Missing load-balancing loss and an AMP CUDA crash are unrelated." + }, + { + "left": "issue:36246", + "right": "issue:43596", + "accept": false, + "reason": "A missing import symbol and a ZeRO3 BertModel index error are unrelated." + }, + { + "left": "issue:44295", + "right": "issue:44749", + "accept": false, + "reason": "A position_ids buffer access bug and a slowdown after upgrade are not the same issue." + }, + { + "left": "issue:44496", + "right": "issue:45003", + "accept": false, + "reason": "An unrecognized model/config error and unsafe sys.modules access are unrelated." + }, + { + "left": "issue:43673", + "right": "issue:44561", + "accept": false, + "reason": "Chunked-prefill cache missing and removal of is_torch_fx_available affect different parts of generation/model loading." + }, + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "ESM2 breakage and Gemma-4 mm_token_type_ids defaults are unrelated model bugs." + }, + { + "left": "issue:43262", + "right": "issue:44246", + "accept": false, + "reason": "Audio chat-template sampling-rate handling and import slowdown are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft edge, 42491 \u2194 43299, is plausibly the same Qwen3-MoE loading regression around the hf4.x/hf5.x transition. The remaining pairs are rejected with appropriately narrow reasoning and no overstatement." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 40, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:42886", + "issue:43065", + "issue:43352", + "issue:43408", + "issue:43475", + "issue:43540", + "issue:43653", + "issue:43906", + "issue:43931", + "issue:44038", + "issue:44060", + "issue:44077", + "issue:44162", + "issue:44188", + "issue:44368", + "issue:44466", + "issue:44479", + "issue:44483", + "issue:44561", + "issue:44877", + "issue:44977", + "issue:44987", + "issue:44995", + "issue:45137", + "issue:45200", + "issue:45229", + "issue:45250", + "issue:45292", + "issue:45335", + "issue:45375", + "issue:45397", + "issue:45412", + "issue:45459", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45600", + "issue:45656", + "issue:45685", + "issue:45701", + "issue:45759", + "issue:46123", + "issue:46170", + "issue:46202", + "issue:46399", + "issue:46431", + "issue:46506" + ], + "soft_pairs": [ + "issue:45375|issue:45759", + "issue:44877|issue:45375", + "issue:45292|issue:45584", + "issue:45137|issue:45600", + "issue:44038|issue:46506", + "issue:45292|issue:45701", + "issue:44561|issue:46399", + "issue:43352|issue:45250", + "issue:44466|issue:45335", + "issue:45412|issue:46123", + "issue:44987|issue:45459", + "issue:44479|issue:46170", + "issue:44060|issue:44077", + "issue:43475|issue:43540", + "issue:44995|issue:45200", + "issue:45397|issue:45656", + "issue:44368|issue:44977", + "issue:43653|issue:44162", + "issue:42886|issue:45229", + "issue:44995|issue:45588", + "issue:44877|issue:46170", + "issue:43065|issue:43408", + "issue:44368|issue:45563", + "issue:43906|issue:46202", + "issue:43931|issue:44368", + "issue:45685|issue:46431", + "issue:44188|issue:46202", + "issue:44483|issue:46202" + ], + "prepared_review_unit_hash": "10644e12b8a955241c3200f2a09591a5867207d27a2ae090c989fd8b5c8112d4", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35414, + "estimated_input_tokens": 8854, + "estimated_eval_tokens": 17964 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the items span unrelated model bugs, tokenizer/config issues, serving regressions, and backend-specific failures. I found no pair that clearly looks like the same underlying bug/change, so all soft edges should be rejected.", + "confidence": 0.84, + "canonical_issue_reason": "issue:45229 is the strongest representative issue in the set: it is concrete, high-impact, and well-documented with substantial discussion. It is more substantial than the very broad or highly model-specific reports nearby.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45229 is the best single issue to anchor the cluster because it describes a reproducible, high-severity regression with clear scope and community attention.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45375", + "right": "issue:45759", + "accept": false, + "reason": "Both are about Qwen multimodal configs, but one is a missing config field and the other is AutoModel CausalLM config unwrapping; different failure modes and fixes." + }, + { + "left": "issue:44877", + "right": "issue:45375", + "accept": false, + "reason": "Both mention strict config behavior, but they affect different model families and different config fields; not the same bug." + }, + { + "left": "issue:45292", + "right": "issue:45584", + "accept": false, + "reason": "One is about resize_token_embeddings not updating output embeddings, the other is Whisper failing on empty transcription after token alignment; unrelated code paths." + }, + { + "left": "issue:45137", + "right": "issue:45600", + "accept": false, + "reason": "DeepSpeed ZeRO3 deque pop error and Sam3LiteText auto-mapping references are unrelated issues." + }, + { + "left": "issue:44038", + "right": "issue:46506", + "accept": false, + "reason": "Qwen3-VL-Moe loading/regression and PPDocLayoutV3 CPU/CUDA divergence are different model/backend problems." + }, + { + "left": "issue:45292", + "right": "issue:45701", + "accept": false, + "reason": "Both involve tokenization/embedding behavior, but one is about embedding resizing while the other is version-dependent tokenization output; not the same bug." + }, + { + "left": "issue:44561", + "right": "issue:46399", + "accept": false, + "reason": "Removal of a utility in v5 breaking trust_remote_code models is unrelated to Qwen3.5 kernel registration for apply_rotary_pos_emb." + }, + { + "left": "issue:43352", + "right": "issue:45250", + "accept": false, + "reason": "One is a specific Nemotron Flash Attention 2 incompatibility; the other is a generic Flash Attention 2 topic with no matching concrete bug." + }, + { + "left": "issue:44466", + "right": "issue:45335", + "accept": false, + "reason": "Both touch tied embeddings, but one is device-dependent serialization and the other is resize propagation to decoder embeddings; different defects." + }, + { + "left": "issue:45412", + "right": "issue:46123", + "accept": false, + "reason": "RT-DETR memory release and MaskGenerationPipeline partial-batch dropping are unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45459", + "accept": false, + "reason": "Model loading failure for physical-intelligence/fast and tokenizer decode errors when protobuf is missing are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:46170", + "accept": false, + "reason": "A video-input regression for several Qwen models is unrelated to Windows/ROCm torch.distributed import issues." + }, + { + "left": "issue:44060", + "right": "issue:44077", + "accept": false, + "reason": "Incorrect tied-weights warning and patchtsmixer post_init restriction are different configuration/model-authoring issues." + }, + { + "left": "issue:43475", + "right": "issue:43540", + "accept": false, + "reason": "SAM3 vision encoder attribute error and Qwen3OmniMoe video-processing ValueError are different model-specific bugs." + }, + { + "left": "issue:44995", + "right": "issue:45200", + "accept": false, + "reason": "A stale cache crash on second forward pass is unrelated to Gemma4 needing mm_token_type_ids defaults." + }, + { + "left": "issue:45397", + "right": "issue:45656", + "accept": false, + "reason": "Gemma4 zero3 loading and optimizer step being called twice under deepspeed are different DeepSpeed-related problems." + }, + { + "left": "issue:44368", + "right": "issue:44977", + "accept": false, + "reason": "A tied-word-embeddings warning for Qwen3.5 fine-tuning is not the same as a flash-attention generation failure." + }, + { + "left": "issue:43653", + "right": "issue:44162", + "accept": false, + "reason": "BigBirdTokenizer special-token registration and ESM2 breakage are different tokenizer/model failures." + }, + { + "left": "issue:42886", + "right": "issue:45229", + "accept": false, + "reason": "Offline tokenizer cache loading and Gemma4 multi-GPU CUDA OOM are entirely unrelated." + }, + { + "left": "issue:44995", + "right": "issue:45588", + "accept": false, + "reason": "Stale indexer cache crash and flash_attention.py crashing on s_aux=None are unrelated failures." + }, + { + "left": "issue:44877", + "right": "issue:46170", + "accept": false, + "reason": "Strict granite_speech config loading and Windows/ROCm distributed import errors do not share the same bug." + }, + { + "left": "issue:43065", + "right": "issue:43408", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder and sam3_video vs sam3_tracker model-type mismatch are different SAM3 issues." + }, + { + "left": "issue:44368", + "right": "issue:45563", + "accept": false, + "reason": "Both concern generation warnings, but one is tied embeddings and the other is a stale num_return_sequences warning in paged generate()." + }, + { + "left": "issue:43906", + "right": "issue:46202", + "accept": false, + "reason": "An isolated reproduction of an earlier issue and a Gemma pipeline roundtrip failure on \u2581 are not the same underlying bug." + }, + { + "left": "issue:43931", + "right": "issue:44368", + "accept": false, + "reason": "Qwen3-VL-30B-A3B-Instruct weight shape mismatch is unrelated to the tied-embeddings warning in Qwen3.5 fine-tuning." + }, + { + "left": "issue:45685", + "right": "issue:46431", + "accept": false, + "reason": "MPS histogram not implemented for Int and missing chat template in a processor are unrelated." + }, + { + "left": "issue:44188", + "right": "issue:46202", + "accept": false, + "reason": "torch.compile branching causing diverging attention kernels is unrelated to the Gemma \u2581 roundtrip failure." + }, + { + "left": "issue:44483", + "right": "issue:46202", + "accept": false, + "reason": "A v5.3 chat/completions request rejection and a Gemma tokenization roundtrip failure are different API/model 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": 21, + "review_unit_index": 41, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:28282", + "issue:30990", + "issue:41628", + "issue:42371", + "issue:42994", + "issue:43065", + "issue:43232", + "issue:43475", + "issue:43482", + "issue:43525", + "issue:43643", + "issue:43646", + "issue:43673", + "issue:43723", + "issue:43784", + "issue:44038", + "issue:44162", + "issue:44188", + "issue:44297", + "issue:44361", + "issue:44623", + "issue:44749", + "issue:44869", + "issue:44871", + "issue:44933", + "issue:44964", + "issue:45092", + "issue:45137", + "issue:45230", + "issue:45310", + "issue:45356", + "issue:45375", + "issue:45381", + "issue:45397", + "issue:45406", + "issue:45478", + "issue:45584", + "issue:45656", + "issue:45987", + "issue:46082", + "issue:46123", + "issue:46132", + "issue:46170", + "issue:46202", + "issue:46326", + "issue:46399", + "issue:46421" + ], + "soft_pairs": [ + "issue:45987|issue:46082", + "issue:45584|issue:46170", + "issue:44188|issue:46421", + "issue:43065|issue:46123", + "issue:44297|issue:44623", + "issue:44361|issue:44749", + "issue:44749|issue:45656", + "issue:43646|issue:44162", + "issue:44623|issue:45656", + "issue:44749|issue:45310", + "issue:44871|issue:46132", + "issue:43232|issue:43673", + "issue:42994|issue:43482", + "issue:45230|issue:45397", + "issue:43475|issue:43643", + "issue:44869|issue:45356", + "issue:28282|issue:30990", + "issue:44038|issue:44623", + "issue:45375|issue:45478", + "issue:41628|issue:44933", + "issue:43475|issue:45381", + "issue:42371|issue:45092", + "issue:45310|issue:45375", + "issue:42994|issue:43525", + "issue:44964|issue:45137", + "issue:43784|issue:46399", + "issue:43723|issue:46326", + "issue:45406|issue:46202", + "issue:43232|issue:44188" + ], + "prepared_review_unit_hash": "93ab6ad855f604707df2b88f44d2fdf7f68eb07e46aa50e8855a417489e560d9", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35623, + "estimated_input_tokens": 8906, + "estimated_eval_tokens": 18068 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: it mixes unrelated tokenizer, model-loading, generation, save/load, and performance regressions. None of the soft pairs look like the same concrete bug, so I reject all of them. As a representative issue, I\u2019d anchor the cluster on the broad v5 custom-model-init regression.", + "confidence": 0.64, + "canonical_issue_reason": "issue:43646 is the broadest, most umbrella-like regression report in the set, so it\u2019s the best canonical issue to represent this otherwise mixed cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 has the clearest high-level framing of a general Transformers v5 breakage, making it the most useful issue to summarize the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45987", + "right": "issue:46082", + "accept": false, + "reason": "Return-type contract bug in StaticCache vs LlamaConfig validation error; different code paths and failure modes." + }, + { + "left": "issue:45584", + "right": "issue:46170", + "accept": false, + "reason": "Whisper empty-transcription generation bug vs Windows/ROCm torch.distributed import dependency issue." + }, + { + "left": "issue:44188", + "right": "issue:46421", + "accept": false, + "reason": "Both mention generation/cache behavior, but one is attention-kernel divergence under torch.compile and the other is cache lazy-init recompilation; not the same bug." + }, + { + "left": "issue:43065", + "right": "issue:46123", + "accept": false, + "reason": "SAM3 pixel decoder dummy layer issue vs partial-batch masking pipeline dropping results; unrelated." + }, + { + "left": "issue:44297", + "right": "issue:44623", + "accept": false, + "reason": "Tokenizer save metadata mismatch vs processor save missing files; both are save_pretrained-related but not the same artifact or bug." + }, + { + "left": "issue:44361", + "right": "issue:44749", + "accept": false, + "reason": "MLukeTokenizer AttributeError vs a general throughput regression; different problems." + }, + { + "left": "issue:44749", + "right": "issue:45656", + "accept": false, + "reason": "Performance slowdown regression vs DeepSpeed optimizer.step being called twice; not the same underlying issue." + }, + { + "left": "issue:43646", + "right": "issue:44162", + "accept": false, + "reason": "Generic custom model init breakage vs ESM2-specific breakage; different models and failure paths." + }, + { + "left": "issue:44623", + "right": "issue:45656", + "accept": false, + "reason": "Processor save_pretrained file emission bug vs DeepSpeed optimizer-step duplication; unrelated." + }, + { + "left": "issue:44749", + "right": "issue:45310", + "accept": false, + "reason": "General performance regression vs Qwen3.5 MoE from_pretrained error; different symptoms and fixes." + }, + { + "left": "issue:44871", + "right": "issue:46132", + "accept": false, + "reason": "Gemma eos_token_id inconsistency vs AttentionInterface.register behavior change; not the same bug." + }, + { + "left": "issue:43232", + "right": "issue:43673", + "accept": false, + "reason": "Generation kwargs update after sync_gpus vs missing GenerationMixin cache in chunked prefill; both generation-related but distinct." + }, + { + "left": "issue:42994", + "right": "issue:43482", + "accept": false, + "reason": "Quantized model saving failure vs Qwen2.5-GGUF loading failure; opposite directions and different subsystems." + }, + { + "left": "issue:45230", + "right": "issue:45397", + "accept": false, + "reason": "Generic placeholder bug report vs Gemma-4 zero3 from_pretrained failure; no concrete overlap." + }, + { + "left": "issue:43475", + "right": "issue:43643", + "accept": false, + "reason": "Missing Sam3 vision encoder field vs trust_remote_code AutoConfig field-loss issue; both are missing-field reports but not the same one." + }, + { + "left": "issue:44869", + "right": "issue:45356", + "accept": false, + "reason": "Whisper timestamp decode crash vs Kimi tokenizer codec/regex regression; different models and code paths." + }, + { + "left": "issue:28282", + "right": "issue:30990", + "accept": false, + "reason": "PyTorch-missing ImportError vs Sentence-Transformers loading hang; unrelated import vs runtime stall." + }, + { + "left": "issue:44038", + "right": "issue:44623", + "accept": false, + "reason": "Qwen3-VL-Moe/v5 bug vs processor.save_pretrained missing files; different artifacts and failures." + }, + { + "left": "issue:45375", + "right": "issue:45478", + "accept": false, + "reason": "Missing config field dropped by @strict vs generic Qwen3.5 MoE from_pretrained error; related model family but not the same concrete bug." + }, + { + "left": "issue:41628", + "right": "issue:44933", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols from different paths; too broad to merge." + }, + { + "left": "issue:43475", + "right": "issue:45381", + "accept": false, + "reason": "Sam3VisionEncoderOutput missing attribute vs qwen2.5-vl video position-id error; unrelated model families and failures." + }, + { + "left": "issue:42371", + "right": "issue:45092", + "accept": false, + "reason": "TF32 behavior/API settings issue vs InternVL2 meta-initialization incompatibility; different concerns." + }, + { + "left": "issue:45310", + "right": "issue:45375", + "accept": false, + "reason": "One is a Qwen3.5 MoE from_pretrained failure, the other a missing config field being dropped; not enough evidence they are the same bug." + }, + { + "left": "issue:42994", + "right": "issue:43525", + "accept": false, + "reason": "Quantized save failure vs LlamaConfig pad_token_id validation error; unrelated." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Phi-4 multimodal load failure vs DeepSpeed ZeRO3 deque IndexError; different components and symptoms." + }, + { + "left": "issue:43784", + "right": "issue:46399", + "accept": false, + "reason": "sentence-transformers import NameError vs Qwen3.5 kernel-registration ValueError; unrelated." + }, + { + "left": "issue:43723", + "right": "issue:46326", + "accept": false, + "reason": "AutoTokenizer loading regression in v5 vs continue_final_message bug for mistral_common tokenizers; different tokenizer paths." + }, + { + "left": "issue:45406", + "right": "issue:46202", + "accept": false, + "reason": "transformers serve crash on missing _tokenizer vs Gemma roundtrip failure on a specific character; unrelated." + }, + { + "left": "issue:43232", + "right": "issue:44188", + "accept": false, + "reason": "Generation kwargs update bug vs attention-kernel divergence under torch.compile; both touch generation/attention but are different concrete problems." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 42, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:30333", + "issue:30990", + "issue:33290", + "issue:35141", + "issue:36010", + "issue:41628", + "issue:42175", + "issue:42222", + "issue:42673", + "issue:42831", + "issue:42898", + "issue:42994", + "issue:43232", + "issue:43278", + "issue:43377", + "issue:43479", + "issue:43504", + "issue:43704", + "issue:43784", + "issue:43824", + "issue:43873", + "issue:43976", + "issue:44016", + "issue:44038", + "issue:44079", + "issue:44222", + "issue:44242", + "issue:44351", + "issue:44496", + "issue:44568", + "issue:44623", + "issue:44792", + "issue:44869", + "issue:44933", + "issue:44945", + "issue:45081", + "issue:45103", + "issue:45137", + "issue:45200", + "issue:45310", + "issue:45561", + "issue:45600", + "issue:45712", + "issue:45753", + "issue:45812", + "issue:46077", + "issue:46123", + "issue:46231" + ], + "soft_pairs": [ + "issue:43873|issue:44945", + "issue:41628|issue:44351", + "issue:44945|issue:45753", + "issue:45081|issue:45103", + "issue:42673|issue:43704", + "issue:42994|issue:43824", + "issue:44222|issue:45310", + "issue:44038|issue:45310", + "issue:36010|issue:41628", + "issue:42831|issue:44222", + "issue:43278|issue:45137", + "issue:43232|issue:44792", + "issue:42175|issue:44933", + "issue:44496|issue:45561", + "issue:43504|issue:45561", + "issue:42175|issue:45712", + "issue:45561|issue:45600", + "issue:43976|issue:44933", + "issue:30333|issue:33290", + "issue:46077|issue:46123", + "issue:30990|issue:35141", + "issue:42898|issue:44568", + "issue:43377|issue:44242", + "issue:44623|issue:45310", + "issue:43479|issue:44869", + "issue:42222|issue:45200", + "issue:43479|issue:44079", + "issue:44016|issue:46231", + "issue:42898|issue:45812", + "issue:43784|issue:43824" + ], + "prepared_review_unit_hash": "571db8331624a0872e0d7387beff84fceaa5883c22572b4c31ed99e25ae89a43", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36639, + "estimated_input_tokens": 9160, + "estimated_eval_tokens": 18576 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly unrelated issues. The only strong duplicate-looking pair is the two Qwen3ForCausalLM VRAM leak reports, which share the same title and very similar symptom description; the rest appear to cover different bugs or features.", + "confidence": 0.93, + "canonical_issue_reason": "issue:42673 is the best canonical issue because it is the fuller Qwen3ForCausalLM VRAM-leak report, while issue:43704 appears to repeat the same title and symptom.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:42673 is the best representative of the duplicate pair because it has the more detailed description of the same VRAM leak behavior reported again in issue:43704.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43873", + "right": "issue:44945", + "accept": false, + "reason": "Quantization/offloading behavior is a different bug from incorrect outputs under pipeline parallelism." + }, + { + "left": "issue:41628", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols and different code paths." + }, + { + "left": "issue:44945", + "right": "issue:45753", + "accept": false, + "reason": "Pipeline-parallelism output corruption is unrelated to a model-specific infinite loop on one image." + }, + { + "left": "issue:45081", + "right": "issue:45103", + "accept": false, + "reason": "Different AttributeError sites: tokenizer regex patching vs auto-docstring keyword parsing." + }, + { + "left": "issue:42673", + "right": "issue:43704", + "accept": true, + "reason": "Same title and same underlying symptom: Qwen3ForCausalLM leaking VRAM across multiple dataloader threads." + }, + { + "left": "issue:42994", + "right": "issue:43824", + "accept": false, + "reason": "Quantized model saving failure is not the same bug as an import error for a model class." + }, + { + "left": "issue:44222", + "right": "issue:45310", + "accept": false, + "reason": "FP8 save_pretrained failure and Qwen3.5 MoE from_pretrained failure are different operations with different failure modes." + }, + { + "left": "issue:44038", + "right": "issue:45310", + "accept": false, + "reason": "They both mention Qwen3/MoE, but the reported symptoms and entry points are different enough to treat as separate issues." + }, + { + "left": "issue:36010", + "right": "issue:41628", + "accept": false, + "reason": "Different missing imports from different modules; no sign of the same broken export." + }, + { + "left": "issue:42831", + "right": "issue:44222", + "accept": false, + "reason": "One is an FP8 accuracy regression, the other is an FP8 save_pretrained problem." + }, + { + "left": "issue:43278", + "right": "issue:45137", + "accept": false, + "reason": "Embedding dtype drift in eval is unrelated to a DeepSpeed ZeRO3 deque IndexError." + }, + { + "left": "issue:43232", + "right": "issue:44792", + "accept": false, + "reason": "Generation kwargs handling after sync_gpus is unrelated to a Janus image-generation test failure." + }, + { + "left": "issue:42175", + "right": "issue:44933", + "accept": false, + "reason": "Backend dependency wiring is unrelated to a missing image_utils import." + }, + { + "left": "issue:44496", + "right": "issue:45561", + "accept": false, + "reason": "Unrecognized model/config loading is unrelated to the pytest-xdist captured_info.txt race." + }, + { + "left": "issue:43504", + "right": "issue:45561", + "accept": false, + "reason": "Legacy-field model loading and the test-file race are different failures." + }, + { + "left": "issue:42175", + "right": "issue:45712", + "accept": false, + "reason": "Torch extra backend dependency issues are not the same as leftover dummy classes leaking into dir(transformers)." + }, + { + "left": "issue:45561", + "right": "issue:45600", + "accept": false, + "reason": "A test race and stale auto_mappings references are different CI/test problems." + }, + { + "left": "issue:43976", + "right": "issue:44933", + "accept": false, + "reason": "Python version compatibility failure is unrelated to a missing image_utils import." + }, + { + "left": "issue:30333", + "right": "issue:33290", + "accept": false, + "reason": "MLflow job-state reporting and DeepSpeed Adafactor OOM are unrelated." + }, + { + "left": "issue:46077", + "right": "issue:46123", + "accept": false, + "reason": "Attention-mask construction for Gemma4 is unrelated to final-batch tracking in MaskGenerationPipeline." + }, + { + "left": "issue:30990", + "right": "issue:35141", + "accept": false, + "reason": "A loading hang for Sentence Transformers is unrelated to embedding reinitialization after resize." + }, + { + "left": "issue:42898", + "right": "issue:44568", + "accept": false, + "reason": "Tokenization cleanup behavior changes and missing BOS/EOS insertion are distinct tokenizer regressions." + }, + { + "left": "issue:43377", + "right": "issue:44242", + "accept": false, + "reason": "Missing padding-mask support in MIMI is unrelated to omitted load-balancing loss." + }, + { + "left": "issue:44623", + "right": "issue:45310", + "accept": false, + "reason": "Missing files on processor save_pretrained is unrelated to the Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:43479", + "right": "issue:44869", + "accept": false, + "reason": "Default multimodal config initialization is unrelated to Whisper timestamp decode crashing on a trailing replacement character." + }, + { + "left": "issue:42222", + "right": "issue:45200", + "accept": false, + "reason": "Broken vitpose model loading is unrelated to Gemma 4 mm_token_type_ids defaults." + }, + { + "left": "issue:43479", + "right": "issue:44079", + "accept": false, + "reason": "Default config initialization is unrelated to ModelOutput key assignment for previously-None fields." + }, + { + "left": "issue:44016", + "right": "issue:46231", + "accept": false, + "reason": "A notebook syntax error and a docstring parameter-name mismatch are different documentation issues." + }, + { + "left": "issue:42898", + "right": "issue:45812", + "accept": false, + "reason": "clean_up_tokenization_spaces behavior and Granite token-ID regression are separate tokenizer issues." + }, + { + "left": "issue:43784", + "right": "issue:43824", + "accept": false, + "reason": "Different import failures with different missing symbols; no evidence of a shared root cause." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The soft-edge verdicts are generally conservative, but the issue-level reasoning overstates the evidence: the packet only shows two issues with identical titles, not a confirmed duplicate link or externally confirmed duplicate topic. Phrases like \"best representative of the cluster\u2019s only real duplicate topic\" and \"matches the confirmed duplicate topic exactly\" go beyond the packet. Keep the conclusion limited to \"these two reports appear to be duplicates based on identical titles and similar descriptions.\"" + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 43, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:28282", + "issue:30333", + "issue:34689", + "issue:35141", + "issue:36331", + "issue:41628", + "issue:42757", + "issue:42796", + "issue:42831", + "issue:42898", + "issue:43295", + "issue:43475", + "issue:43479", + "issue:43644", + "issue:43723", + "issue:43874", + "issue:44060", + "issue:44075", + "issue:44222", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44373", + "issue:44448", + "issue:44485", + "issue:44568", + "issue:44623", + "issue:44734", + "issue:44779", + "issue:44933", + "issue:44977", + "issue:45071", + "issue:45092", + "issue:45103", + "issue:45125", + "issue:45245", + "issue:45310", + "issue:45356", + "issue:45405", + "issue:45656", + "issue:45712", + "issue:46082", + "issue:46132", + "issue:46231", + "issue:46305" + ], + "soft_pairs": [ + "issue:43874|issue:45092", + "issue:43723|issue:44568", + "issue:42898|issue:44779", + "issue:44734|issue:45103", + "issue:44075|issue:46231", + "issue:44295|issue:45656", + "issue:42898|issue:45356", + "issue:27957|issue:30333", + "issue:42898|issue:43295", + "issue:30333|issue:34689", + "issue:44060|issue:45125", + "issue:45405|issue:45712", + "issue:44222|issue:44297", + "issue:44933|issue:45712", + "issue:30333|issue:36331", + "issue:30333|issue:35141", + "issue:44297|issue:45656", + "issue:22355|issue:28282", + "issue:42831|issue:44623", + "issue:43479|issue:43644", + "issue:44222|issue:44623", + "issue:44448|issue:46132", + "issue:42796|issue:43723", + "issue:45245|issue:46132", + "issue:41628|issue:42757", + "issue:44297|issue:45310", + "issue:43475|issue:46305", + "issue:45071|issue:46082", + "issue:44373|issue:44485", + "issue:44263|issue:45092", + "issue:43723|issue:44977" + ], + "prepared_review_unit_hash": "d523dab0ac982152e94ea198f25006b82b89a6f7f19a057af939bca55e8a67be", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36367, + "estimated_input_tokens": 9092, + "estimated_eval_tokens": 18440 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is very noisy: most items are only loosely related by broad Transformers subsystem or v5-regression wording, not the same underlying bug. I found no convincing duplicate pairs, so all soft edges are rejected.", + "confidence": 0.24, + "canonical_issue_reason": "Selected `issue:42898` as the most broadly representative tokenizer regression report in v5; it is generic enough to stand in for the tokenizer-regression subset.", + "canonical_pr_reason": null, + "best_issue_reason": "`issue:42898` is the best overall representative of the cluster\u2019s recurring tokenizer/v5 regression theme, but the set is too heterogeneous for true duplicate consolidation.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43874", + "right": "issue:45092", + "accept": false, + "reason": "Different failures: missing image-processor method vs remote-code/meta-init incompatibility." + }, + { + "left": "issue:43723", + "right": "issue:44568", + "accept": false, + "reason": "Tokenizer loading failure is distinct from special-token insertion behavior for a specific model." + }, + { + "left": "issue:42898", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions in v5, but they describe different symptoms and likely different code paths." + }, + { + "left": "issue:44734", + "right": "issue:45103", + "accept": false, + "reason": "Serving KV-cache indexing bug is unrelated to auto-docstring crash under future annotations." + }, + { + "left": "issue:44075", + "right": "issue:46231", + "accept": false, + "reason": "Optimizer argument handling is unrelated to a documentation parameter-name mismatch." + }, + { + "left": "issue:44295", + "right": "issue:45656", + "accept": false, + "reason": "Buffer read error and duplicate optimizer step under deepspeed are different runtime problems." + }, + { + "left": "issue:42898", + "right": "issue:45356", + "accept": false, + "reason": "Both concern tokenizer regressions, but one is clean-up spacing behavior and the other is codec/warning handling for Kimi-K2.5." + }, + { + "left": "issue:27957", + "right": "issue:30333", + "accept": false, + "reason": "Flash Attention 2 model support and MLflow reporting on failed training jobs are unrelated." + }, + { + "left": "issue:42898", + "right": "issue:43295", + "accept": false, + "reason": "Tokenizer clean-up behavior and processor/tokenizer API regression are distinct breakages." + }, + { + "left": "issue:30333", + "right": "issue:34689", + "accept": false, + "reason": "Training-job reporting bug is unrelated to Llama 3.2 Vision model-loading regression." + }, + { + "left": "issue:44060", + "right": "issue:45125", + "accept": false, + "reason": "Incorrect tied-weights warning is not the same as missing tensor-parallel plan metadata." + }, + { + "left": "issue:45405", + "right": "issue:45712", + "accept": false, + "reason": "PEFT version pinning issue is unrelated to dummy class cleanup and repo checks." + }, + { + "left": "issue:44222", + "right": "issue:44297", + "accept": false, + "reason": "Both mention save_pretrained, but one is FP8 MoE serialization and the other is tokenizer_class metadata mismatch." + }, + { + "left": "issue:44933", + "right": "issue:45712", + "accept": false, + "reason": "Nonexistent import from image_utils is unrelated to leftover dummy PT classes." + }, + { + "left": "issue:30333", + "right": "issue:36331", + "accept": false, + "reason": "MLflow reporting on failure is unrelated to CustomTrainer.compute_loss signature changes." + }, + { + "left": "issue:30333", + "right": "issue:35141", + "accept": false, + "reason": "Failed-job reporting and embedding reinitialization are separate training behaviors." + }, + { + "left": "issue:44297", + "right": "issue:45656", + "accept": false, + "reason": "Tokenizer save metadata bug does not match optimizer stepping twice under deepspeed." + }, + { + "left": "issue:22355", + "right": "issue:28282", + "accept": false, + "reason": "Missing transformers.onnx module and missing PyTorch dependency are different import errors." + }, + { + "left": "issue:42831", + "right": "issue:44623", + "accept": false, + "reason": "FP8 accuracy issue and processor save_pretrained missing files are not the same bug." + }, + { + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Default multimodal config initialization and non-persistent buffer junk are unrelated initialization bugs." + }, + { + "left": "issue:44222", + "right": "issue:44623", + "accept": false, + "reason": "Different save_pretrained failures: FP8 MoE serialization vs processor artifact omission." + }, + { + "left": "issue:44448", + "right": "issue:46132", + "accept": false, + "reason": "PEGASUS v4/v5 output drift is unrelated to AttentionInterface.register behavior changes." + }, + { + "left": "issue:42796", + "right": "issue:43723", + "accept": false, + "reason": "Tokenizer marker handling and tokenizer loading/import behavior are different problems." + }, + { + "left": "issue:45245", + "right": "issue:46132", + "accept": false, + "reason": "Category-cardinality runtime error is unrelated to attention registry behavior." + }, + { + "left": "issue:41628", + "right": "issue:42757", + "accept": false, + "reason": "Missing AutoImageProcessor export and missing is_offline_mode import come from different modules and causes." + }, + { + "left": "issue:44297", + "right": "issue:45310", + "accept": false, + "reason": "Tokenizer save metadata mismatch and Qwen3.5 MoE from_pretrained error are separate issues." + }, + { + "left": "issue:43475", + "right": "issue:46305", + "accept": false, + "reason": "SAM 3 Video attribute error and MoonshineStreaming off-by-one mask are unrelated model bugs." + }, + { + "left": "issue:45071", + "right": "issue:46082", + "accept": false, + "reason": "Type-checking regression in PretrainedConfig is different from LlamaConfig head_dim validation." + }, + { + "left": "issue:44373", + "right": "issue:44485", + "accept": false, + "reason": "Docstring typo and GLM-5 RoPE implementation discussion are not the same bug." + }, + { + "left": "issue:44263", + "right": "issue:45092", + "accept": false, + "reason": "torch.split return handling in GLM MoE code is unrelated to remote-code meta-initialization compatibility." + }, + { + "left": "issue:43723", + "right": "issue:44977", + "accept": false, + "reason": "Both are tokenizer issues in v5, but one is loading failure and the other is incorrect Deepseek tokenization output." + } + ] + }, + "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": 21, + "review_unit_index": 44, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:30990", + "issue:38175", + "issue:41628", + "issue:42831", + "issue:42907", + "issue:43122", + "issue:43232", + "issue:43408", + "issue:43479", + "issue:43644", + "issue:43723", + "issue:43824", + "issue:44038", + "issue:44190", + "issue:44206", + "issue:44222", + "issue:44261", + "issue:44263", + "issue:44279", + "issue:44295", + "issue:44360", + "issue:44373", + "issue:44483", + "issue:44484", + "issue:44623", + "issue:44734", + "issue:44829", + "issue:44871", + "issue:44913", + "issue:45092", + "issue:45103", + "issue:45310", + "issue:45357", + "issue:45381", + "issue:45406", + "issue:45834", + "issue:46002", + "issue:46050", + "issue:46082", + "issue:46209", + "issue:46305", + "issue:46328", + "issue:46402", + "issue:46421", + "issue:46424", + "issue:46431" + ], + "soft_pairs": [ + "issue:43122|issue:46209", + "issue:45381|issue:46305", + "issue:44222|issue:44484", + "issue:46209|issue:46431", + "issue:44206|issue:45103", + "issue:43232|issue:46424", + "issue:44734|issue:46421", + "issue:44190|issue:46082", + "issue:44295|issue:44623", + "issue:43232|issue:45834", + "issue:43723|issue:44871", + "issue:44263|issue:44360", + "issue:44279|issue:46328", + "issue:43644|issue:46305", + "issue:30990|issue:38175", + "issue:44261|issue:45092", + "issue:45092|issue:45357", + "issue:42907|issue:46050", + "issue:42831|issue:45310", + "issue:44263|issue:44373", + "issue:44829|issue:46002", + "issue:44483|issue:45406", + "issue:42831|issue:44038", + "issue:22355|issue:27957", + "issue:41628|issue:43824", + "issue:43479|issue:44913", + "issue:43408|issue:46402" + ], + "prepared_review_unit_hash": "34b581b9fee06bf65fbfcac4ffc098574c543037226dfb4d1ff47974bf207617", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35534, + "estimated_input_tokens": 8884, + "estimated_eval_tokens": 18024 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a grab bag of mostly unrelated Transformers issues spanning tokenization, cache/generation, multimodal configs, serving, quantization, and model-specific regressions. None of the proposed soft pairs look like true duplicates of the same bug or change.", + "confidence": 0.79, + "canonical_issue_reason": "There is no strong canonical issue because the set is heterogeneous; if one anchor is needed, issue 46050 is the most concrete open bug with a reproducible save/load regression.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46050 is the best overall triage anchor because it is open, specific, and describes a concrete model quantization save/load failure.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43122", + "right": "issue:46209", + "accept": false, + "reason": "Tokenizer-version mismatch and FP8 inference on B200 are unrelated bugs." + }, + { + "left": "issue:45381", + "right": "issue:46305", + "accept": false, + "reason": "Different model paths and failures: Qwen2.5-VL vision positions vs MoonshineStreaming mask off-by-one." + }, + { + "left": "issue:44222", + "right": "issue:44484", + "accept": false, + "reason": "FP8 save_pretrained/moe serialization is unrelated to max_shard_size default behavior." + }, + { + "left": "issue:46209", + "right": "issue:46431", + "accept": false, + "reason": "FP8 inference on B200 and missing chat template are different serving/model issues." + }, + { + "left": "issue:44206", + "right": "issue:45103", + "accept": false, + "reason": "Feature extractor argument crash and auto_docstring annotations crash are separate code paths." + }, + { + "left": "issue:43232", + "right": "issue:46424", + "accept": false, + "reason": "Generation kwargs handling after sync_gpus is distinct from cache max_length configuration." + }, + { + "left": "issue:44734", + "right": "issue:46421", + "accept": false, + "reason": "KV-cache continuation tensor indexing and lazy cache init recompilation are different bugs." + }, + { + "left": "issue:44190", + "right": "issue:46082", + "accept": false, + "reason": "Local dataset loading failure is unrelated to LlamaConfig head_dim validation." + }, + { + "left": "issue:44295", + "right": "issue:44623", + "accept": false, + "reason": "position_ids buffer access and processor.save_pretrained missing files do not share the same cause." + }, + { + "left": "issue:43232", + "right": "issue:45834", + "accept": false, + "reason": "Generation sync_gpus behavior and Kosmos2.5 OCR index errors are unrelated." + }, + { + "left": "issue:43723", + "right": "issue:44871", + "accept": false, + "reason": "AutoTokenizer load failure and Gemma-3 eos_token_id mismatch are separate issues." + }, + { + "left": "issue:44263", + "right": "issue:44360", + "accept": false, + "reason": "Both mention the DSA indexer, but one is split return handling and the other is a missing ReLU; different defects." + }, + { + "left": "issue:44279", + "right": "issue:46328", + "accept": false, + "reason": "A dependency problem and a padding performance regression are not the same bug." + }, + { + "left": "issue:43644", + "right": "issue:46305", + "accept": false, + "reason": "Non-persistent buffer junk and sliding-window off-by-one are unrelated." + }, + { + "left": "issue:30990", + "right": "issue:38175", + "accept": false, + "reason": "Loading stalls and zero probabilities are different failure modes." + }, + { + "left": "issue:44261", + "right": "issue:45092", + "accept": false, + "reason": "Missing rms_norm_eps precision issue is unrelated to old InternVL2 meta-init incompatibility." + }, + { + "left": "issue:45092", + "right": "issue:45357", + "accept": false, + "reason": "Both are v5 multimodal regressions, but they affect different model families and code paths." + }, + { + "left": "issue:42907", + "right": "issue:46050", + "accept": false, + "reason": "Both involve save/load of quantized models, but dequantized Ministrals/Devstrals and sinq round-trip failures are not the same bug." + }, + { + "left": "issue:42831", + "right": "issue:45310", + "accept": false, + "reason": "FineGrainedFP8 accuracy and Qwen3.5-Moe from_pretrained failure are unrelated." + }, + { + "left": "issue:44263", + "right": "issue:44373", + "accept": false, + "reason": "A DSA indexer tensor bug and a docstring issue for position_ids are not duplicates." + }, + { + "left": "issue:44829", + "right": "issue:46002", + "accept": false, + "reason": "Flash-attention training degeneration and NaNs after checkout/init are different instability reports." + }, + { + "left": "issue:44483", + "right": "issue:45406", + "accept": false, + "reason": "Chat/completions request rejection and Gemma4Processor missing _tokenizer are separate serving bugs." + }, + { + "left": "issue:42831", + "right": "issue:44038", + "accept": false, + "reason": "FP8 accuracy regression and Qwen3-VL-Moe bug are different issues." + }, + { + "left": "issue:22355", + "right": "issue:27957", + "accept": false, + "reason": "Missing transformers.onnx import and XLMRoberta Flash Attention 2 are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but they are missing different exported symbols and likely separate fixes." + }, + { + "left": "issue:43479", + "right": "issue:44913", + "accept": false, + "reason": "Default multimodal config initialization and GPTNeoX rotary_pct reload behavior are unrelated." + }, + { + "left": "issue:43408", + "right": "issue:46402", + "accept": false, + "reason": "sam3_video/sam3_tracker warning and missing Sam3Model text encoder weights are different SAM3-family problems." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 45, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:28218", + "issue:30990", + "issue:33357", + "issue:39401", + "issue:41628", + "issue:42371", + "issue:42757", + "issue:42796", + "issue:42831", + "issue:42886", + "issue:42907", + "issue:42994", + "issue:43066", + "issue:43122", + "issue:43232", + "issue:43408", + "issue:43550", + "issue:43576", + "issue:43646", + "issue:43756", + "issue:43784", + "issue:43937", + "issue:43976", + "issue:44060", + "issue:44117", + "issue:44188", + "issue:44261", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44368", + "issue:44373", + "issue:44488", + "issue:44568", + "issue:44704", + "issue:44945", + "issue:44964", + "issue:45278", + "issue:45593", + "issue:45600", + "issue:45753", + "issue:46002", + "issue:46132", + "issue:46305", + "issue:46369" + ], + "soft_pairs": [ + "issue:22355|issue:28218", + "issue:42831|issue:45753", + "issue:42757|issue:45278", + "issue:39401|issue:44568", + "issue:27957|issue:30990", + "issue:44368|issue:46305", + "issue:42886|issue:43066", + "issue:43232|issue:45593", + "issue:43408|issue:46369", + "issue:42886|issue:42994", + "issue:42796|issue:43122", + "issue:44188|issue:44373", + "issue:43576|issue:43784", + "issue:43646|issue:46002", + "issue:42371|issue:44263", + "issue:43550|issue:46132", + "issue:44261|issue:44373", + "issue:44295|issue:44297", + "issue:42907|issue:44945", + "issue:44117|issue:44704", + "issue:30990|issue:33357", + "issue:43756|issue:44060", + "issue:41628|issue:45278", + "issue:43784|issue:45600", + "issue:43976|issue:44945", + "issue:39401|issue:44488", + "issue:42907|issue:43937", + "issue:42907|issue:44964" + ], + "prepared_review_unit_hash": "3e72d286063f1dfa75f3aa5eccc4a35c0adf7c1e4a2932f5f5860157acc2213f", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35025, + "estimated_input_tokens": 8757, + "estimated_eval_tokens": 17770 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: it mixes unrelated tokenizer, import, generation, attention, and model-specific bugs. No soft edge looks like the same underlying bug/change, so I rejected all candidate merges.", + "confidence": 0.36, + "canonical_issue_reason": "issue:45278 is the broadest representative here: a general import-regression report after upgrading Transformers, which overlaps the most with the other import-related items. It is still not a duplicate of the others, just the best anchor for this noisy cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the best single issue to represent the cluster because it is the most generic and highest-level failure mode among the listed items. It is the closest thing to a cluster hub, even though the overall set is not a true duplicate group.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:22355", + "right": "issue:28218", + "accept": false, + "reason": "Different failure classes: missing module import vs tokenizer spacing behavior." + }, + { + "left": "issue:42831", + "right": "issue:45753", + "accept": false, + "reason": "Unrelated problems: FP8 accuracy regression vs image-specific infinite loop." + }, + { + "left": "issue:42757", + "right": "issue:45278", + "accept": false, + "reason": "Both are import-breakage reports, but one is a specific missing hub symbol and the other is a broad upgrade-wide import failure; not the same concrete bug." + }, + { + "left": "issue:39401", + "right": "issue:44568", + "accept": false, + "reason": "Both concern tokenization, but they affect different models and different symptoms (offset mapping vs special-token insertion)." + }, + { + "left": "issue:27957", + "right": "issue:30990", + "accept": false, + "reason": "Different subsystems and symptoms: Flash Attention 2 on XLM-Roberta vs loading hangs in Sentence Transformers." + }, + { + "left": "issue:44368", + "right": "issue:46305", + "accept": false, + "reason": "Unrelated: a tied-embeddings warning in fine-tuning vs a streaming sliding-window off-by-one bug." + }, + { + "left": "issue:42886", + "right": "issue:43066", + "accept": false, + "reason": "Both are tokenizer-related, but cache/offline loading and decoder-type mismatch are distinct code paths." + }, + { + "left": "issue:43232", + "right": "issue:45593", + "accept": false, + "reason": "Generation kwargs handling and D-FINE auxiliary-loss behavior are unrelated." + }, + { + "left": "issue:43408", + "right": "issue:46369", + "accept": false, + "reason": "Same broad model family (sam3) is not enough; one is a type-mismatch warning, the other is a memory overflow." + }, + { + "left": "issue:42886", + "right": "issue:42994", + "accept": false, + "reason": "Offline tokenizer cache loading and quantized model saving are different bugs." + }, + { + "left": "issue:42796", + "right": "issue:43122", + "accept": false, + "reason": "The former is a tokenizer BPE-marker bug for one model; the latter is a general tokenization-difference regression across versions." + }, + { + "left": "issue:44188", + "right": "issue:44373", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile is unrelated to a docstring issue." + }, + { + "left": "issue:43576", + "right": "issue:43784", + "accept": false, + "reason": "A broken CLI/env command is not the same bug as an import-time NameError in sentence-transformers." + }, + { + "left": "issue:43646", + "right": "issue:46002", + "accept": false, + "reason": "Custom model initialization breakage and NaNs in classification heads are different failure modes." + }, + { + "left": "issue:42371", + "right": "issue:44263", + "accept": false, + "reason": "TF32 API settings guidance and torch.split behavior in GlmMoeDsaIndexer are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:46132", + "accept": false, + "reason": "Both involve attention/compile concepts, but they do not describe the same code-path problem." + }, + { + "left": "issue:44261", + "right": "issue:44373", + "accept": false, + "reason": "A precision/config bug in MLA layernorm is unrelated to a docstring correction." + }, + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "Serialization/buffer handling and tokenizer_class metadata mismatch are separate issues." + }, + { + "left": "issue:42907", + "right": "issue:44945", + "accept": false, + "reason": "Quantized-model save failures and pipeline-parallel output errors are unrelated." + }, + { + "left": "issue:44117", + "right": "issue:44704", + "accept": false, + "reason": "Both are loading-path issues, but one is tokenizer mapping assumptions and the other is AutoProcessor kwargs propagation." + }, + { + "left": "issue:30990", + "right": "issue:33357", + "accept": false, + "reason": "Both mention loading, but they are different models and different failure types (stuck load vs bus error)." + }, + { + "left": "issue:43756", + "right": "issue:44060", + "accept": false, + "reason": "RoPE-layer misconfiguration and a tied-weights warning are not the same bug." + }, + { + "left": "issue:41628", + "right": "issue:45278", + "accept": false, + "reason": "Missing AutoImageProcessor export is a specific API regression; the other is a broad import-error report." + }, + { + "left": "issue:43784", + "right": "issue:45600", + "accept": false, + "reason": "Import-time NameError in sentence-transformers is unrelated to auto_mappings CI breakage." + }, + { + "left": "issue:43976", + "right": "issue:44945", + "accept": false, + "reason": "Python version compatibility and pipeline-parallel output correctness are separate issues." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "Both touch tokenization/model loading, but the reported bugs are different and model-specific." + }, + { + "left": "issue:42907", + "right": "issue:43937", + "accept": false, + "reason": "Quantized save failure and invalid GenerationConfig are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:44964", + "accept": false, + "reason": "Saving a dequantized model and failing to load a multimodal model are 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": 21, + "review_unit_index": 46, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:22355", + "issue:29127", + "issue:29159", + "issue:36296", + "issue:38617", + "issue:39401", + "issue:42175", + "issue:42673", + "issue:42757", + "issue:42898", + "issue:42907", + "issue:43066", + "issue:43116", + "issue:43122", + "issue:43278", + "issue:43408", + "issue:43452", + "issue:43475", + "issue:43493", + "issue:43611", + "issue:43824", + "issue:43937", + "issue:44038", + "issue:44188", + "issue:44222", + "issue:44295", + "issue:44297", + "issue:44351", + "issue:44360", + "issue:44485", + "issue:44521", + "issue:44625", + "issue:44704", + "issue:44779", + "issue:44829", + "issue:44938", + "issue:44945", + "issue:45072", + "issue:45200", + "issue:45216", + "issue:45278", + "issue:45290", + "issue:45478", + "issue:45753", + "issue:46002", + "issue:46326", + "issue:46328", + "issue:46614" + ], + "soft_pairs": [ + "issue:44360|issue:44485", + "issue:44704|issue:45290", + "issue:44625|issue:44829", + "issue:39401|issue:44521", + "issue:43122|issue:46326", + "issue:44038|issue:44297", + "issue:43122|issue:44521", + "issue:43116|issue:45200", + "issue:43408|issue:43475", + "issue:39401|issue:46614", + "issue:44038|issue:44295", + "issue:42907|issue:43452", + "issue:44038|issue:44222", + "issue:22355|issue:29159", + "issue:45216|issue:45478", + "issue:43824|issue:44351", + "issue:43122|issue:44779", + "issue:42757|issue:44351", + "issue:36296|issue:43116", + "issue:43278|issue:44945", + "issue:42175|issue:44938", + "issue:22355|issue:29127", + "issue:43493|issue:45753", + "issue:43937|issue:44485", + "issue:38617|issue:45278", + "issue:44188|issue:46328", + "issue:42898|issue:46614", + "issue:42673|issue:45072", + "issue:43066|issue:44779", + "issue:43611|issue:46002", + "issue:43116|issue:43493" + ], + "prepared_review_unit_hash": "13870aa0146acb441fa079205f9701b9c4e9310ea4b74668a9db7de1f3f0aae6", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36663, + "estimated_input_tokens": 9166, + "estimated_eval_tokens": 18588 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is very heterogeneous: import-surface regressions, tokenizer mismatches, multimodal chat-template bugs, model-family-specific issues, and training/runtime problems. I do not see any soft pair that is clearly the same underlying bug/change, so all soft edges should be rejected. If forced to pick an umbrella issue, the broad 4.57\u21925.5 import-regression report is the most representative, but only for a small subset of the set.", + "confidence": 0.84, + "canonical_issue_reason": "Issue 45278 is the broadest regression report here and is the closest thing to an umbrella for the import-error subset, though the overall cluster is still too mixed to make it a true duplicate hub.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45278 is the most general, regression-oriented issue among the set and is the best single representative for triage, even if it does not cover the full cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "Different bugs in different areas: DSA indexer/ReLU discussion vs GLM-5 RoPE implementation." + }, + { + "left": "issue:44704", + "right": "issue:45290", + "accept": false, + "reason": "Different code paths: cached_file kwargs handling vs apply_chat_template crash on tool-call messages." + }, + { + "left": "issue:44625", + "right": "issue:44829", + "accept": false, + "reason": "Different model behaviors: num_labels propagation vs flash_attention_3 training degeneration." + }, + { + "left": "issue:39401", + "right": "issue:44521", + "accept": false, + "reason": "Both involve tokenization, but one is offset_mapping and the other is assistant_masks for multimodal chat templates." + }, + { + "left": "issue:43122", + "right": "issue:46326", + "accept": false, + "reason": "Both are tokenizer regressions, but the symptoms and affected APIs are different (tokenization output vs continue_final_message)." + }, + { + "left": "issue:44038", + "right": "issue:44297", + "accept": false, + "reason": "Different failures: Qwen3-VL-Moe loading vs tokenizer save_pretrained metadata mismatch." + }, + { + "left": "issue:43122", + "right": "issue:44521", + "accept": false, + "reason": "Different problems: text tokenization drift vs multimodal assistant mask generation." + }, + { + "left": "issue:43116", + "right": "issue:45200", + "accept": false, + "reason": "Unrelated features: example-script multi-label classification vs Gemma 4 mm_token_type_ids handling." + }, + { + "left": "issue:43408", + "right": "issue:43475", + "accept": false, + "reason": "Same model family, but one is a config warning and the other is an attribute error in vision encoder outputs." + }, + { + "left": "issue:39401", + "right": "issue:46614", + "accept": false, + "reason": "Different tokenizers and symptoms: Qwen3 offset mapping vs Wav2Vec2 phonemizer backend state leakage." + }, + { + "left": "issue:44038", + "right": "issue:44295", + "accept": false, + "reason": "Different bug classes: Qwen3-VL-Moe import/loading issue vs position_ids buffer access error." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Different operations and artifacts: dequantized save_pretrained vs gguf_file loading through AutoTokenizer/AutoModel." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "Both mention Qwen3-VL-Moe, but one is a loading/import issue and the other is FP8 save_pretrained behavior." + }, + { + "left": "issue:22355", + "right": "issue:29159", + "accept": false, + "reason": "No real overlap: ONNX module import failure vs slow/fast tokenizer inconsistency." + }, + { + "left": "issue:45216", + "right": "issue:45478", + "accept": false, + "reason": "Same model family/version area, but save_pretrained checkpoint corruption and from_pretrained error are distinct code paths." + }, + { + "left": "issue:43824", + "right": "issue:44351", + "accept": false, + "reason": "Both are missing imports, but for different symbols and likely different export paths." + }, + { + "left": "issue:43122", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions in v5, but the affected models and reported failures are different." + }, + { + "left": "issue:42757", + "right": "issue:44351", + "accept": false, + "reason": "Different missing imports from different modules; no evidence of the same underlying fix." + }, + { + "left": "issue:36296", + "right": "issue:43116", + "accept": false, + "reason": "Completely unrelated: tensor-parallel training bug vs an example script classification issue." + }, + { + "left": "issue:43278", + "right": "issue:44945", + "accept": false, + "reason": "Different execution contexts: dtype drift in eval vs incorrect output under pipeline parallelism." + }, + { + "left": "issue:42175", + "right": "issue:44938", + "accept": false, + "reason": "Unrelated environment/setup issues: backend extras install vs Python 3.14 load failure." + }, + { + "left": "issue:22355", + "right": "issue:29127", + "accept": false, + "reason": "Different failures: missing transformers.onnx module vs LayoutLMv3 error-message clarity." + }, + { + "left": "issue:43493", + "right": "issue:45753", + "accept": false, + "reason": "Both concern image/multimodal models, but one is a SigLIP2 implementation discrepancy and the other is an infinite-loop bug in Qwen3_5." + }, + { + "left": "issue:43937", + "right": "issue:44485", + "accept": false, + "reason": "Same GLM-5 branding, but GenerationConfig validation and RoPE implementation are distinct problems." + }, + { + "left": "issue:38617", + "right": "issue:45278", + "accept": false, + "reason": "Both are import regressions, but one is a specific missing symbol and the other is a broad upgrade-wide import-error report; not enough to call them the same bug." + }, + { + "left": "issue:44188", + "right": "issue:46328", + "accept": false, + "reason": "Different subsystems: torch.compile attention-kernel branching vs SequenceFeatureExtractor.pad performance." + }, + { + "left": "issue:42898", + "right": "issue:46614", + "accept": false, + "reason": "Tokenizer-related, but one is a v5 behavior change in cleanup and the other is stale phonemizer language state." + }, + { + "left": "issue:42673", + "right": "issue:45072", + "accept": false, + "reason": "Different failure modes: VRAM leak in multi-threaded inference vs dtype mismatches in bfloat16 inference." + }, + { + "left": "issue:43066", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions in v5, but the specific behavior changes and affected models are different." + }, + { + "left": "issue:43611", + "right": "issue:46002", + "accept": false, + "reason": "Unrelated: base_model_prefix loading breakage vs NaNs during classification-head init/checkout." + }, + { + "left": "issue:43116", + "right": "issue:43493", + "accept": false, + "reason": "Different domains: text classification example behavior vs multimodal model discrepancy." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 47, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:28348", + "issue:33453", + "issue:39401", + "issue:42222", + "issue:42371", + "issue:42673", + "issue:42796", + "issue:43064", + "issue:43116", + "issue:43441", + "issue:43450", + "issue:43673", + "issue:43825", + "issue:43827", + "issue:43874", + "issue:43906", + "issue:43937", + "issue:43976", + "issue:43986", + "issue:43992", + "issue:44038", + "issue:44075", + "issue:44117", + "issue:44291", + "issue:44355", + "issue:44380", + "issue:44466", + "issue:44625", + "issue:44704", + "issue:44779", + "issue:44936", + "issue:44977", + "issue:45071", + "issue:45083", + "issue:45356", + "issue:45357", + "issue:45405", + "issue:45446", + "issue:45478", + "issue:45563", + "issue:45588", + "issue:46002", + "issue:46077", + "issue:46231", + "issue:46328", + "issue:46614", + "issue:46640" + ], + "soft_pairs": [ + "issue:44355|issue:46231", + "issue:42371|issue:43874", + "issue:43976|issue:45446", + "issue:44038|issue:45478", + "issue:43992|issue:44704", + "issue:44625|issue:46002", + "issue:44075|issue:44936", + "issue:28348|issue:46640", + "issue:45357|issue:45478", + "issue:42222|issue:43441", + "issue:45083|issue:45446", + "issue:43116|issue:43825", + "issue:42673|issue:42796", + "issue:43064|issue:43673", + "issue:43937|issue:45563", + "issue:43827|issue:45446", + "issue:44380|issue:46077", + "issue:33453|issue:39401", + "issue:45356|issue:46614", + "issue:44117|issue:45588", + "issue:45071|issue:46614", + "issue:43673|issue:46328", + "issue:44291|issue:45071", + "issue:43450|issue:43986", + "issue:44466|issue:46002", + "issue:42673|issue:43906", + "issue:39401|issue:44779", + "issue:45405|issue:45446", + "issue:43937|issue:44977" + ], + "prepared_review_unit_hash": "029cb6b74f68d35ab787ade6b15dafd840cb7eb270354ea87312c3c8479cdc5e", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35669, + "estimated_input_tokens": 8918, + "estimated_eval_tokens": 18092 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly false-positive similarity links: the pairs span unrelated tokenizer, attention, processor, training, docs, and config bugs. I would not merge any soft edge. If one issue must represent the set, issue:45478 is the clearest concrete runtime regression in the Qwen/MoE loading-failure subset.", + "confidence": 0.39, + "canonical_issue_reason": "issue:45478 is the most concrete, user-visible runtime failure here, with a clearly scoped from_pretrained regression and a model-specific failure path.", + "canonical_pr_reason": null, + "best_issue_reason": "It is specific, reproducible, and easy to triage as a real bug; among these heterogeneous issues it is the strongest standalone representative.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44355", + "right": "issue:46231", + "accept": false, + "reason": "Compiled-file import errors and a docs parameter-name mismatch are unrelated." + }, + { + "left": "issue:42371", + "right": "issue:43874", + "accept": false, + "reason": "TF32 behavior settings and a missing image-processor method affect different code paths." + }, + { + "left": "issue:43976", + "right": "issue:45446", + "accept": false, + "reason": "Python-version compatibility and a flex_attention import/version check are not the same bug." + }, + { + "left": "issue:44038", + "right": "issue:45478", + "accept": false, + "reason": "Both touch Qwen MoE models, but the affected models, APIs, and failure modes differ." + }, + { + "left": "issue:43992", + "right": "issue:44704", + "accept": false, + "reason": "Missing model weights and AutoProcessor kwarg forwarding are unrelated loading issues." + }, + { + "left": "issue:44625", + "right": "issue:46002", + "accept": false, + "reason": "Label propagation and NaNs in classification heads are different defects." + }, + { + "left": "issue:44075", + "right": "issue:44936", + "accept": false, + "reason": "Optimizer args being ignored and evaluate-after-train failure are separate trainer bugs." + }, + { + "left": "issue:28348", + "right": "issue:46640", + "accept": false, + "reason": "Feature request for GPT2 FlashAttention 2 support is not the same as a vague Flash Attention bug report." + }, + { + "left": "issue:45357", + "right": "issue:45478", + "accept": false, + "reason": "Both are Qwen3.5-related, but one is save_pretrained key serialization and the other is from_pretrained loading failure." + }, + { + "left": "issue:42222", + "right": "issue:43441", + "accept": false, + "reason": "VitPose breakage and a Ministral flash-attention failure are unrelated." + }, + { + "left": "issue:45083", + "right": "issue:45446", + "accept": false, + "reason": "Feature-extractor length math and a PyTorch version guard in flex_attention are different issues." + }, + { + "left": "issue:43116", + "right": "issue:43825", + "accept": false, + "reason": "Multi-label example output failure and a pipeline translation error message are not the same bug." + }, + { + "left": "issue:42673", + "right": "issue:42796", + "accept": false, + "reason": "A VRAM leak in dataloader threads is unrelated to tokenizer BPE-marker handling." + }, + { + "left": "issue:43064", + "right": "issue:43673", + "accept": false, + "reason": "FSDP/PEFT optimizer-state corruption and GenerationMixin cache handling are different code paths." + }, + { + "left": "issue:43937", + "right": "issue:45563", + "accept": false, + "reason": "An invalid GenerationConfig error and a stale warning for num_return_sequences are different symptoms." + }, + { + "left": "issue:43827", + "right": "issue:45446", + "accept": false, + "reason": "Documentation cleanup after pipeline removals does not match a flex_attention version-check bug." + }, + { + "left": "issue:44380", + "right": "issue:46077", + "accept": false, + "reason": "GPT2 backend scaling being ignored and Gemma4 bidirectional-mask construction are distinct attention bugs." + }, + { + "left": "issue:33453", + "right": "issue:39401", + "accept": false, + "reason": "Both are tokenizer regressions, but loading regression and offset_mapping correctness are different defects." + }, + { + "left": "issue:45356", + "right": "issue:46614", + "accept": false, + "reason": "Different tokenizer/model families and different failure modes; not the same underlying bug." + }, + { + "left": "issue:44117", + "right": "issue:45588", + "accept": false, + "reason": "Tokenizer mapping None and a flash_attention crash on s_aux=None are unrelated." + }, + { + "left": "issue:45071", + "right": "issue:46614", + "accept": false, + "reason": "PretrainedConfig type-checking and stale phonemizer backend state are unrelated." + }, + { + "left": "issue:43673", + "right": "issue:46328", + "accept": false, + "reason": "Generation cache absence and a padding performance issue are not the same bug." + }, + { + "left": "issue:44291", + "right": "issue:45071", + "accept": false, + "reason": "Unexpected init_empty_weights argument handling and config type-checking are different regressions." + }, + { + "left": "issue:43450", + "right": "issue:43986", + "accept": false, + "reason": "Batched video-processor shape issues and missing torchvision crash behavior are related only loosely by area." + }, + { + "left": "issue:44466", + "right": "issue:46002", + "accept": false, + "reason": "Serialization of tied lm_head weights and NaNs in classification heads are unrelated." + }, + { + "left": "issue:42673", + "right": "issue:43906", + "accept": false, + "reason": "An isolated reproduction of another issue is not enough to show the same VRAM-leak defect." + }, + { + "left": "issue:39401", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer bugs, but Qwen3 offset mapping and DeepSeek incorrect results are different model-specific regressions." + }, + { + "left": "issue:45405", + "right": "issue:45446", + "accept": false, + "reason": "A PEFT version bump and a flex_attention import/version check are unrelated." + }, + { + "left": "issue:43937", + "right": "issue:44977", + "accept": false, + "reason": "Invalid GenerationConfig and flash-attention generation failure are not the same concrete bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 48, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:38617", + "issue:39401", + "issue:41628", + "issue:42175", + "issue:42222", + "issue:42371", + "issue:42490", + "issue:42617", + "issue:42673", + "issue:43066", + "issue:43097", + "issue:43122", + "issue:43408", + "issue:43673", + "issue:43742", + "issue:43937", + "issue:44075", + "issue:44117", + "issue:44220", + "issue:44261", + "issue:44263", + "issue:44410", + "issue:44484", + "issue:44485", + "issue:44493", + "issue:44496", + "issue:44829", + "issue:44877", + "issue:44933", + "issue:45071", + "issue:45081", + "issue:45216", + "issue:45278", + "issue:45310", + "issue:45341", + "issue:45362", + "issue:45478", + "issue:45542", + "issue:45561", + "issue:45656", + "issue:46178", + "issue:46209", + "issue:46231", + "issue:46298", + "issue:46328", + "issue:46512", + "issue:46617" + ], + "soft_pairs": [ + "issue:44410|issue:45310", + "issue:44829|issue:45216", + "issue:45341|issue:46328", + "issue:45341|issue:45561", + "issue:45071|issue:46512", + "issue:38617|issue:41628", + "issue:44117|issue:45081", + "issue:43122|issue:46298", + "issue:42617|issue:45341", + "issue:42222|issue:46178", + "issue:43408|issue:43742", + "issue:45362|issue:46328", + "issue:42371|issue:44261", + "issue:44877|issue:46231", + "issue:42490|issue:42673", + "issue:42371|issue:46617", + "issue:44484|issue:45310", + "issue:39401|issue:46209", + "issue:43937|issue:44496", + "issue:44933|issue:45278", + "issue:38617|issue:44220", + "issue:44410|issue:45478", + "issue:44075|issue:45656", + "issue:43066|issue:43122", + "issue:42222|issue:45542", + "issue:44263|issue:44485", + "issue:43097|issue:44493", + "issue:42175|issue:42222", + "issue:43673|issue:45341" + ], + "prepared_review_unit_hash": "304be38acb463c53e76698a036be56eea7d21d2001466e19d8e404a611e64227", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35382, + "estimated_input_tokens": 8846, + "estimated_eval_tokens": 17948 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly a grab bag of unrelated bugs: tokenizer regressions, import errors, backend/config loading issues, testing-utils problems, and model-specific failures. I don\u2019t see any soft pair that clearly describes the same underlying bug or change.", + "confidence": 0.9, + "canonical_issue_reason": "Issue 45561 is the most clearly scoped and actionable open bug in the set: a concrete pytest-xdist race in patched testing utilities, with reproducible behavior and active discussion.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45561 is the best representative issue because it has the clearest failure mode, strong reproducibility, and multiple references; it is more triage-worthy than the other loosely related reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44410", + "right": "issue:45310", + "accept": false, + "reason": "Both involve Qwen-family models, but one is a missing projection bug in qwen3next and the other is a Qwen3.5 MoE from_pretrained failure." + }, + { + "left": "issue:44829", + "right": "issue:45216", + "accept": false, + "reason": "Different problems: flash_attention_3 training degeneration vs a Qwen3.5 save_pretrained checkpoint regression." + }, + { + "left": "issue:45341", + "right": "issue:46328", + "accept": false, + "reason": "Unrelated: a small testing_utils bug vs a SequenceFeatureExtractor.pad performance issue." + }, + { + "left": "issue:45341", + "right": "issue:45561", + "accept": false, + "reason": "Both touch testing utils, but one is a generic bug and the other is a specific xdist file race; not the same concrete failure." + }, + { + "left": "issue:45071", + "right": "issue:46512", + "accept": false, + "reason": "Different regressions: PreTrainedConfig type checking vs stale model.config.dtype after from_config(dtype=...)." + }, + { + "left": "issue:38617", + "right": "issue:41628", + "accept": false, + "reason": "Separate import errors for different symbols: layer_type_validation vs AutoImageProcessor." + }, + { + "left": "issue:44117", + "right": "issue:45081", + "accept": false, + "reason": "Both are tokenizer-related, but one is a TOKENIZER_MAPPING_NAMES None-handling bug and the other is a Mistral regex patch crash." + }, + { + "left": "issue:43122", + "right": "issue:46298", + "accept": false, + "reason": "Tokenization regression vs a noisy logging message; no shared underlying bug." + }, + { + "left": "issue:42617", + "right": "issue:45341", + "accept": false, + "reason": "A runtime failure running 3d_parallel.py is unrelated to a testing_utils bug." + }, + { + "left": "issue:42222", + "right": "issue:46178", + "accept": false, + "reason": "Vitpose model breakage and a tensorflow_text backend-mapping error are different issues." + }, + { + "left": "issue:43408", + "right": "issue:43742", + "accept": false, + "reason": "Different model-loading problems: sam3_video/sam3_tracker warning vs MobileLLM config/key error." + }, + { + "left": "issue:45362", + "right": "issue:46328", + "accept": false, + "reason": "Qwen3.5 chat crash and a padding performance issue do not share the same code-path." + }, + { + "left": "issue:42371", + "right": "issue:44261", + "accept": false, + "reason": "TF32 API guidance is unrelated to the MLA rms_norm_eps precision bug." + }, + { + "left": "issue:44877", + "right": "issue:46231", + "accept": false, + "reason": "A strict config loading bug and a documentation parameter-name typo are not the same bug." + }, + { + "left": "issue:42490", + "right": "issue:42673", + "accept": false, + "reason": "Local-path import warning vs VRAM leak in multi-dataloader-thread use are unrelated." + }, + { + "left": "issue:42371", + "right": "issue:46617", + "accept": false, + "reason": "TF32 backend settings and trust_remote_code cache loading failures are unrelated." + }, + { + "left": "issue:44484", + "right": "issue:45310", + "accept": false, + "reason": "A save_pretrained shard-size question is unrelated to a Qwen3.5 MoE loading error." + }, + { + "left": "issue:39401", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3 tokenizer offset_mapping bug and FP8 inference failure on B200 are different problems." + }, + { + "left": "issue:43937", + "right": "issue:44496", + "accept": false, + "reason": "Invalid GenerationConfig and unrecognized model/config loading are distinct failures." + }, + { + "left": "issue:44933", + "right": "issue:45278", + "accept": false, + "reason": "The latter is a broad upgrade-related import regression report; the former is one specific import issue, not clearly the same bug." + }, + { + "left": "issue:38617", + "right": "issue:44220", + "accept": false, + "reason": "An import failure and an _torch_extract_fbank_features issue are unrelated." + }, + { + "left": "issue:44410", + "right": "issue:45478", + "accept": false, + "reason": "Similar model family, but the concrete bugs differ: missing projections vs a Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:44075", + "right": "issue:45656", + "accept": false, + "reason": "Optimizer SGD args being ignored is different from optimizer.step being called twice under deepspeed." + }, + { + "left": "issue:43066", + "right": "issue:43122", + "accept": false, + "reason": "Both mention tokenizer behavior in v5, but one is a wrong decoder-type issue and the other is changed tokenization output; not enough evidence they are the same code-path bug." + }, + { + "left": "issue:42222", + "right": "issue:45542", + "accept": false, + "reason": "Vitpose model breakage is unrelated to a missing TensorFlow backend setup." + }, + { + "left": "issue:44263", + "right": "issue:44485", + "accept": false, + "reason": "torch.split return semantics and GLM-5 RoPE implementation are unrelated." + }, + { + "left": "issue:43097", + "right": "issue:44493", + "accept": false, + "reason": "API removal around tie_embeddings_and_encoder_decoder is unrelated to unexpected position-id keys." + }, + { + "left": "issue:42175", + "right": "issue:42222", + "accept": false, + "reason": "Backend packaging/install issue vs vitpose model breakage are different problems." + }, + { + "left": "issue:43673", + "right": "issue:45341", + "accept": false, + "reason": "Generation cache behavior and a testing_utils bug do not share 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": 21, + "review_unit_index": 49, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:27957", + "issue:38617", + "issue:41628", + "issue:42757", + "issue:42890", + "issue:43012", + "issue:43064", + "issue:43097", + "issue:43352", + "issue:43441", + "issue:43452", + "issue:43531", + "issue:43577", + "issue:43673", + "issue:43874", + "issue:43931", + "issue:43986", + "issue:44038", + "issue:44117", + "issue:44261", + "issue:44263", + "issue:44315", + "issue:44351", + "issue:44360", + "issue:44410", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44829", + "issue:44933", + "issue:44987", + "issue:45003", + "issue:45071", + "issue:45341", + "issue:45372", + "issue:45375", + "issue:45440", + "issue:45464", + "issue:45478", + "issue:45542", + "issue:45588", + "issue:45753", + "issue:46050", + "issue:46123", + "issue:46172", + "issue:46231", + "issue:46298", + "issue:46369" + ], + "soft_pairs": [ + "issue:43012|issue:43577", + "issue:41628|issue:45071", + "issue:43874|issue:44261", + "issue:45071|issue:45542", + "issue:43931|issue:44410", + "issue:44263|issue:46172", + "issue:44360|issue:46172", + "issue:44483|issue:45464", + "issue:44484|issue:45478", + "issue:27957|issue:43352", + "issue:43441|issue:44829", + "issue:43931|issue:44038", + "issue:43097|issue:43452", + "issue:45003|issue:45341", + "issue:43986|issue:45341", + "issue:45753|issue:46369", + "issue:44933|issue:46231", + "issue:43874|issue:44263", + "issue:44261|issue:44485", + "issue:43673|issue:46123", + "issue:44117|issue:44987", + "issue:43986|issue:44484", + "issue:43531|issue:45341", + "issue:45440|issue:46172", + "issue:43064|issue:44315", + "issue:43986|issue:45372", + "issue:38617|issue:42757", + "issue:43097|issue:45588", + "issue:43097|issue:46050", + "issue:44410|issue:45375", + "issue:44038|issue:44410", + "issue:38617|issue:44351", + "issue:42890|issue:46298" + ], + "prepared_review_unit_hash": "b22905402ef03e3d5232fd2fa63a3696c62ecf10cbcface323d4cd6b02977540", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37375, + "estimated_input_tokens": 9344, + "estimated_eval_tokens": 18944 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are mostly heterogeneous bug reports with only superficial similarity in a few v5/import/model-loading themes. None of the soft pairs look like the same underlying bug or change, so I would not merge any of them as duplicates.", + "confidence": 0.88, + "canonical_issue_reason": "issue:43097 is the broadest, most central v5-regression report in the set and has the strongest signal to serve as an anchor if a single canonical issue must be chosen.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43097 is the best representative of the recurring compatibility-breakage theme, but the cluster is not cohesive enough to justify treating the others as duplicates of it.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43012", + "right": "issue:43577", + "accept": false, + "reason": "Both involve dtype/precision behavior, but one is a compilation warning and the other is a model-loading dtype bug for a different architecture." + }, + { + "left": "issue:41628", + "right": "issue:45071", + "accept": false, + "reason": "These are unrelated failures: one is an AutoImageProcessor import error, the other is a PreTrainedConfig type-checking regression." + }, + { + "left": "issue:43874", + "right": "issue:44261", + "accept": false, + "reason": "Different models and different failures: a missing image-processor method versus an MLA layernorm config precision issue." + }, + { + "left": "issue:45071", + "right": "issue:45542", + "accept": false, + "reason": "A config type-checking issue and a TensorFlow backend detection issue are not the same bug." + }, + { + "left": "issue:43931", + "right": "issue:44410", + "accept": false, + "reason": "Both are Qwen-related, but one is a weight-shape mismatch on model load and the other is a missing projection set in a different model variant." + }, + { + "left": "issue:44263", + "right": "issue:46172", + "accept": false, + "reason": "These hit different code paths in different DeepSeek/GLM-style models; no evidence they share the same defect." + }, + { + "left": "issue:44360", + "right": "issue:46172", + "accept": false, + "reason": "A missing ReLU in one indexer and a dtype/transpose bug in another model are distinct issues." + }, + { + "left": "issue:44483", + "right": "issue:45464", + "accept": false, + "reason": "Both mention chat/completions, but one is a general request rejection in v5 and the other is a model-specific streaming failure." + }, + { + "left": "issue:44484", + "right": "issue:45478", + "accept": false, + "reason": "One concerns save_pretrained sharding defaults; the other is a from_pretrained load error for Qwen3.5 MoE." + }, + { + "left": "issue:27957", + "right": "issue:43352", + "accept": false, + "reason": "Both are Flash Attention 2 topics, but they are model-specific support/compatibility reports, not the same underlying bug." + }, + { + "left": "issue:43441", + "right": "issue:44829", + "accept": false, + "reason": "A FlashAttention compatibility failure and a FlashAttention-3 degenerate-training bug are different defects." + }, + { + "left": "issue:43931", + "right": "issue:44038", + "accept": false, + "reason": "These are different Qwen model families with different symptoms: weight mismatch versus MoE load failure." + }, + { + "left": "issue:43097", + "right": "issue:43452", + "accept": false, + "reason": "A removed config field and a gguf loading regression are unrelated code paths." + }, + { + "left": "issue:45003", + "right": "issue:45341", + "accept": false, + "reason": "An unsafe sys.modules access bug is unrelated to the testing_utils issue." + }, + { + "left": "issue:43986", + "right": "issue:45341", + "accept": false, + "reason": "A processor-loading crash without torchvision and a test utility bug are not the same bug." + }, + { + "left": "issue:45753", + "right": "issue:46369", + "accept": false, + "reason": "Different model families and different failure modes: infinite loop versus memory overflow." + }, + { + "left": "issue:44933", + "right": "issue:46231", + "accept": false, + "reason": "A missing import and a documentation parameter-name mismatch are unrelated." + }, + { + "left": "issue:43874", + "right": "issue:44263", + "accept": false, + "reason": "Different missing-method / tensor-splitting failures in different model components." + }, + { + "left": "issue:44261", + "right": "issue:44485", + "accept": false, + "reason": "A missing rms_norm_eps precision issue is not the same as a GLM-5 RoPE implementation bug." + }, + { + "left": "issue:43673", + "right": "issue:46123", + "accept": false, + "reason": "One is a GenerationMixin cache regression; the other is a mask-generation batching bug." + }, + { + "left": "issue:44117", + "right": "issue:44987", + "accept": false, + "reason": "A tokenizer mapping None-handling bug and a package load failure are too different to treat as duplicates." + }, + { + "left": "issue:43986", + "right": "issue:44484", + "accept": false, + "reason": "These affect different surfaces entirely: processor loading versus model sharding defaults." + }, + { + "left": "issue:43531", + "right": "issue:45341", + "accept": false, + "reason": "A sliding-window bug in Qwen3-MoE is unrelated to the testing_utils issue." + }, + { + "left": "issue:45440", + "right": "issue:46172", + "accept": false, + "reason": "Both involve DeepSeek-style models, but the reported problems are different code paths and symptoms." + }, + { + "left": "issue:43064", + "right": "issue:44315", + "accept": false, + "reason": "A distributed optimizer-state bug and a Liger Kernel application issue are not the same change." + }, + { + "left": "issue:43986", + "right": "issue:45372", + "accept": false, + "reason": "A missing torchvision crash and a mistral_common import error affect different processor-loading paths." + }, + { + "left": "issue:38617", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but for different symbols and different dependency breakages." + }, + { + "left": "issue:43097", + "right": "issue:45588", + "accept": false, + "reason": "A removed config field and a flash_attention.py None-handling crash are unrelated." + }, + { + "left": "issue:43097", + "right": "issue:46050", + "accept": false, + "reason": "A config/API removal issue and a quantization save/load breakage are distinct bugs." + }, + { + "left": "issue:44410", + "right": "issue:45375", + "accept": false, + "reason": "Both mention Qwen variants, but one is missing attention projections while the other is a config field being dropped." + }, + { + "left": "issue:44038", + "right": "issue:44410", + "accept": false, + "reason": "These are different Qwen model families with different load/config defects." + }, + { + "left": "issue:38617", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but they stem from different missing symbols and are not the same underlying regression." + }, + { + "left": "issue:42890", + "right": "issue:46298", + "accept": false, + "reason": "A flaky integration test from missing set_seed calls is unrelated to a noisy logging message." + } + ] + }, + "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": 21, + "review_unit_index": 50, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:36683", + "issue:42175", + "issue:42490", + "issue:42757", + "issue:42907", + "issue:43012", + "issue:43064", + "issue:43066", + "issue:43097", + "issue:43208", + "issue:43352", + "issue:43502", + "issue:43526", + "issue:43550", + "issue:43582", + "issue:43824", + "issue:43825", + "issue:43873", + "issue:43906", + "issue:43976", + "issue:43986", + "issue:43992", + "issue:44016", + "issue:44062", + "issue:44077", + "issue:44117", + "issue:44162", + "issue:44222", + "issue:44368", + "issue:44380", + "issue:44484", + "issue:44655", + "issue:44683", + "issue:44704", + "issue:44936", + "issue:44938", + "issue:45103", + "issue:45478", + "issue:45479", + "issue:45584", + "issue:45712", + "issue:45753", + "issue:46132", + "issue:46172", + "issue:46178", + "issue:46298", + "issue:46365", + "issue:46421" + ], + "soft_pairs": [ + "issue:44062|issue:44117", + "issue:43097|issue:43873", + "issue:43502|issue:43992", + "issue:44936|issue:46365", + "issue:44704|issue:45103", + "issue:44380|issue:46421", + "issue:43986|issue:44117", + "issue:43986|issue:44016", + "issue:44938|issue:46178", + "issue:43208|issue:43550", + "issue:42907|issue:44655", + "issue:42907|issue:45753", + "issue:43825|issue:44655", + "issue:43012|issue:43550", + "issue:43992|issue:44484", + "issue:43012|issue:44368", + "issue:45712|issue:46298", + "issue:43066|issue:43906", + "issue:43097|issue:46298", + "issue:43352|issue:44380", + "issue:43352|issue:43976", + "issue:44162|issue:46172", + "issue:45584|issue:46365", + "issue:44380|issue:46132", + "issue:44077|issue:46365", + "issue:44222|issue:45478", + "issue:44380|issue:45479", + "issue:36683|issue:43582", + "issue:43064|issue:44380", + "issue:42175|issue:46365", + "issue:43208|issue:43526", + "issue:43012|issue:44683", + "issue:42490|issue:43012", + "issue:42757|issue:43824" + ], + "prepared_review_unit_hash": "dc47ce412cdd2d039edef94e5310f04b084add8de980e7925ae82e55b4dcc4b0", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37715, + "estimated_input_tokens": 9429, + "estimated_eval_tokens": 19114 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is heterogeneous: the soft pairs mostly share broad subsystem language (tokenizers, loading, saving, compilation, processors) but describe different failure modes and code paths, so none should be deduplicated.", + "confidence": 0.96, + "canonical_issue_reason": "issue:43097 is the broadest, most central regression report in the set, with downstream references and a clear, actionable compatibility break.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43097 is the most representative user-facing regression among these otherwise unrelated issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44062", + "right": "issue:44117", + "accept": false, + "reason": "Both are tokenizer-related, but one is an AddedToken argument conflict and the other is a TOKENIZER_MAPPING_NAMES None assumption; different bugs." + }, + { + "left": "issue:43097", + "right": "issue:43873", + "accept": false, + "reason": "A v5 API removal warning is unrelated to quantized offloading behavior; different feature areas and symptoms." + }, + { + "left": "issue:43502", + "right": "issue:43992", + "accept": false, + "reason": "One reports network/API calls despite local_files_only, the other a missing embed_tokens weight in UMT5Encoder; not the same failure." + }, + { + "left": "issue:44936", + "right": "issue:46365", + "accept": false, + "reason": "trainer.evaluate() breaking after train is unrelated to importing Trainer failing because of datasets; different code paths." + }, + { + "left": "issue:44704", + "right": "issue:45103", + "accept": false, + "reason": "AutoProcessor kwarg propagation to cached_file and an auto_docstring annotations crash are unrelated bugs." + }, + { + "left": "issue:44380", + "right": "issue:46421", + "accept": false, + "reason": "GPT2 attention scaling being ignored is separate from cache lazy initialization causing recompilation; not the same issue." + }, + { + "left": "issue:43986", + "right": "issue:44117", + "accept": false, + "reason": "Missing torchvision during AutoProcessor load is unrelated to tokenizer mapping returning None." + }, + { + "left": "issue:43986", + "right": "issue:44016", + "accept": false, + "reason": "A runtime dependency/import crash is not the same as a notebook syntax error." + }, + { + "left": "issue:44938", + "right": "issue:46178", + "accept": false, + "reason": "Python 3.14 load failure and an unknown tensorflow_text backend are different compatibility problems." + }, + { + "left": "issue:43208", + "right": "issue:43550", + "accept": false, + "reason": "xLSTM training bugs smaller than 7B and Bamba torch.compile/SDPA failure are distinct model/backend issues." + }, + { + "left": "issue:42907", + "right": "issue:44655", + "accept": false, + "reason": "Failed dequantized save behavior and inability to save Pipeline objects are different serialization paths." + }, + { + "left": "issue:42907", + "right": "issue:45753", + "accept": false, + "reason": "Dequantized save failure is unrelated to a Qwen3.5 image loop during generation." + }, + { + "left": "issue:43825", + "right": "issue:44655", + "accept": false, + "reason": "Pipeline translation support messaging and saving Pipeline objects are unrelated pipeline issues." + }, + { + "left": "issue:43012", + "right": "issue:43550", + "accept": false, + "reason": "A bf16 compile warning is not the same as a Bamba SDPA compile failure." + }, + { + "left": "issue:43992", + "right": "issue:44484", + "accept": false, + "reason": "Missing embed_tokens weights and max_shard_size defaults are different save/load concerns." + }, + { + "left": "issue:43012", + "right": "issue:44368", + "accept": false, + "reason": "A PyTorch precision warning during compile is unrelated to a tie_word_embeddings config warning." + }, + { + "left": "issue:45712", + "right": "issue:46298", + "accept": false, + "reason": "Leftover dummy classes in repo checks are not the same as a noisy logging message." + }, + { + "left": "issue:43066", + "right": "issue:43906", + "accept": false, + "reason": "Wrong tokenizer decoder type in v5 and an isolated reproduction of another issue are not the same bug." + }, + { + "left": "issue:43097", + "right": "issue:46298", + "accept": false, + "reason": "A breaking API removal is unrelated to a noisy logging message." + }, + { + "left": "issue:43352", + "right": "issue:44380", + "accept": false, + "reason": "Nemotron Flash Attention unsupported and GPT2 attention scaling ignored are different attention/backend problems." + }, + { + "left": "issue:43352", + "right": "issue:43976", + "accept": false, + "reason": "Flash Attention support for a model and Python-version compatibility are unrelated." + }, + { + "left": "issue:44162", + "right": "issue:46172", + "accept": false, + "reason": "ESM2 breakage and a DeepSeek tensor transpose bug are separate model-specific failures." + }, + { + "left": "issue:45584", + "right": "issue:46365", + "accept": false, + "reason": "Whisper empty-transcription failure is unrelated to importing Trainer being broken by datasets." + }, + { + "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:44077", + "right": "issue:46365", + "accept": false, + "reason": "PatchTSMixer post_init validation and Trainer import failure are unrelated." + }, + { + "left": "issue:44222", + "right": "issue:45478", + "accept": false, + "reason": "FP8 save_pretrained issues and a Qwen3.5 MoE from_pretrained error are similar only at a high level, but not the same concrete bug." + }, + { + "left": "issue:44380", + "right": "issue:45479", + "accept": false, + "reason": "Attention backend config loss and degenerate zero-loss classification are unrelated." + }, + { + "left": "issue:36683", + "right": "issue:43582", + "accept": false, + "reason": "Gemma3Config missing vocab_size and an Apple Silicon allocator TypeError are unrelated runtime issues." + }, + { + "left": "issue:43064", + "right": "issue:44380", + "accept": false, + "reason": "FSDP2/PEFT optimizer state corruption is a training-distributed bug, not an attention scaling regression." + }, + { + "left": "issue:42175", + "right": "issue:46365", + "accept": false, + "reason": "Backend extras packaging and Trainer import interference from datasets are different dependency problems." + }, + { + "left": "issue:43208", + "right": "issue:43526", + "accept": false, + "reason": "xLSTM training bugs and BeitImageProcessorFast label reduction are separate components and failures." + }, + { + "left": "issue:43012", + "right": "issue:44683", + "accept": false, + "reason": "A compile-time precision warning is not the same as compiled flex_attention breaking on torch>=2.9." + }, + { + "left": "issue:42490", + "right": "issue:43012", + "accept": false, + "reason": "Trailing-slash local model import warning and bf16 compile warning are unrelated issues." + }, + { + "left": "issue:42757", + "right": "issue:43824", + "accept": false, + "reason": "Missing is_offline_mode import and missing Qwen2_5_VLForConditionalGeneration import are both import errors, but from different modules and causes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 51, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:36010", + "issue:36683", + "issue:38617", + "issue:42757", + "issue:42898", + "issue:42994", + "issue:43010", + "issue:43064", + "issue:43066", + "issue:43299", + "issue:43352", + "issue:43475", + "issue:43502", + "issue:43525", + "issue:43611", + "issue:43618", + "issue:43756", + "issue:43761", + "issue:43867", + "issue:43901", + "issue:43937", + "issue:43976", + "issue:43992", + "issue:44038", + "issue:44075", + "issue:44077", + "issue:44230", + "issue:44380", + "issue:44483", + "issue:44485", + "issue:44492", + "issue:44514", + "issue:44625", + "issue:44655", + "issue:44938", + "issue:45216", + "issue:45276", + "issue:45278", + "issue:45290", + "issue:45310", + "issue:45600", + "issue:45698", + "issue:46077", + "issue:46082", + "issue:46172", + "issue:46298", + "issue:46365" + ], + "soft_pairs": [ + "issue:45600|issue:46298", + "issue:44483|issue:44514", + "issue:38617|issue:43502", + "issue:44038|issue:44230", + "issue:45698|issue:46172", + "issue:43618|issue:44380", + "issue:43618|issue:43761", + "issue:43299|issue:43867", + "issue:42994|issue:44655", + "issue:36683|issue:43475", + "issue:44483|issue:45290", + "issue:43756|issue:44485", + "issue:44625|issue:46365", + "issue:42898|issue:43066", + "issue:45276|issue:46077", + "issue:36010|issue:42757", + "issue:43352|issue:43937", + "issue:43611|issue:44655", + "issue:43901|issue:44655", + "issue:36683|issue:44077", + "issue:43867|issue:45216", + "issue:43976|issue:44938", + "issue:44625|issue:46077", + "issue:46077|issue:46082", + "issue:36683|issue:43525", + "issue:43618|issue:45216", + "issue:36683|issue:44625", + "issue:43992|issue:45310", + "issue:43064|issue:44492", + "issue:44075|issue:46365", + "issue:43010|issue:45278" + ], + "prepared_review_unit_hash": "0fef355932228ce142a3289efaef5cf644e00c0d2b160d45b94aa1c3cf3d51cc", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 35893, + "estimated_input_tokens": 8974, + "estimated_eval_tokens": 18204 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly distinct regression reports. The similarity is often only at the subsystem level (imports, save/load, CLIP outputs, tokenizer behavior), but the concrete bugs differ enough that none of the soft edges should be merged.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 45698 is the strongest umbrella candidate: it is open, highly discussed, and centers on a broad from_pretrained/save_pretrained deserialization mismatch that is more representative of the save/load regressions here than the narrower model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45698 is the best representative issue in this mixed set because it has the most discussion and captures a broadly applicable save/load failure mode.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45600", + "right": "issue:46298", + "accept": false, + "reason": "Different bugs: one is a CI break from removed Sam3LiteText mappings, the other is a noisy logging message." + }, + { + "left": "issue:44483", + "right": "issue:44514", + "accept": false, + "reason": "Both affect chat APIs, but one is request validation for /v1/chat/completions and the other is a batched apply_chat_template crash with padding=False." + }, + { + "left": "issue:38617", + "right": "issue:43502", + "accept": false, + "reason": "Unrelated import-error vs network-request behavior: layer_type_validation import failure is not the same as local_files_only still making API calls." + }, + { + "left": "issue:44038", + "right": "issue:44230", + "accept": false, + "reason": "Same model family, but different features: Qwen3-VL-Moe loading failure vs fp8 inference support request." + }, + { + "left": "issue:45698", + "right": "issue:46172", + "accept": false, + "reason": "Different code paths and symptoms: wrong custom module loading after save_pretrained is not the deepseek-v4 dtype/transpose bug." + }, + { + "left": "issue:43618", + "right": "issue:44380", + "accept": false, + "reason": "Different models and regressions: missing CLIP attentions vs GPT2 attention scaling ignored under SDPA/FlashAttention." + }, + { + "left": "issue:43618", + "right": "issue:43761", + "accept": false, + "reason": "Both are CLIP output regressions, but one is attentions missing and the other is hidden_states missing; not enough evidence of one underlying bug." + }, + { + "left": "issue:43299", + "right": "issue:43867", + "accept": false, + "reason": "Different loading failures: Qwen3VL MoE model loading vs a generic state_dict-sorting load error." + }, + { + "left": "issue:42994", + "right": "issue:44655", + "accept": false, + "reason": "Both mention saving, but quantized model saving and pipeline save_pretrained are different bugs and likely different fixes." + }, + { + "left": "issue:36683", + "right": "issue:43475", + "accept": false, + "reason": "Different model configs and missing attributes: Gemma3Config vocab_size vs SAM 3 Video fpn_position_embeddings." + }, + { + "left": "issue:44483", + "right": "issue:45290", + "accept": false, + "reason": "Different request surfaces: chat completion validation vs assistant tool-call template crashes." + }, + { + "left": "issue:43756", + "right": "issue:44485", + "accept": false, + "reason": "Both mention RoPE, but the Smollm3 layer-count issue and the GLM-5 RoPE implementation issue are model-specific and not the same bug." + }, + { + "left": "issue:44625", + "right": "issue:46365", + "accept": false, + "reason": "Unrelated issues: Qwen3.5 num_labels propagation vs Trainer import failure caused by datasets." + }, + { + "left": "issue:42898", + "right": "issue:43066", + "accept": false, + "reason": "Both are tokenizer-related v5 reports, but clean_up_tokenization_spaces behavior and decoder type selection are different code paths." + }, + { + "left": "issue:45276", + "right": "issue:46077", + "accept": false, + "reason": "Both are Gemma4-related, but one is embedding resize propagation and the other is bidirectional attention mask construction." + }, + { + "left": "issue:36010", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but from different packages and symbols: GenerationMixin from transformers vs is_offline_mode from huggingface_hub." + }, + { + "left": "issue:43352", + "right": "issue:43937", + "accept": false, + "reason": "Different failures: Flash Attention 2.0 unsupported on NemotronH vs invalid GenerationConfig for GLM-5." + }, + { + "left": "issue:43611", + "right": "issue:44655", + "accept": false, + "reason": "Different bugs: base_model_prefix loading regression vs inability to save Pipeline objects." + }, + { + "left": "issue:43901", + "right": "issue:44655", + "accept": false, + "reason": "One is a docs mismatch about return_all_scores; the other is a runtime save_pretrained failure." + }, + { + "left": "issue:36683", + "right": "issue:44077", + "accept": false, + "reason": "Both are model/config related, but Gemma3Config vocab_size and patchtsmixer post_init enforcement are unrelated." + }, + { + "left": "issue:43867", + "right": "issue:45216", + "accept": false, + "reason": "Generic load-model state_dict ordering error is not the same as Qwen3.5 save_pretrained checkpoint corruption." + }, + { + "left": "issue:43976", + "right": "issue:44938", + "accept": false, + "reason": "Both are Python-version compatibility problems, but they target different versions and failure modes." + }, + { + "left": "issue:44625", + "right": "issue:46077", + "accept": false, + "reason": "Different Qwen3.5 config propagation vs Gemma4 attention-mask behavior." + }, + { + "left": "issue:46077", + "right": "issue:46082", + "accept": false, + "reason": "Unrelated model/config issues: Gemma4 bidirectional attention masks vs LlamaConfig head_dim validation." + }, + { + "left": "issue:36683", + "right": "issue:43525", + "accept": false, + "reason": "Both are AttributeError reports on configs, but the missing attributes and model families are different." + }, + { + "left": "issue:43618", + "right": "issue:45216", + "accept": false, + "reason": "CLIP output-field regression and Qwen3.5 checkpoint saving corruption are unrelated." + }, + { + "left": "issue:36683", + "right": "issue:44625", + "accept": false, + "reason": "Different config propagation bugs in different models: Gemma3 vocab_size vs Qwen3.5 num_labels." + }, + { + "left": "issue:43992", + "right": "issue:45310", + "accept": false, + "reason": "Different from_pretrained/model-loading failures on different model families." + }, + { + "left": "issue:43064", + "right": "issue:44492", + "accept": false, + "reason": "Training-state optimizer mismatch is unrelated to a typo in cache strategy naming." + }, + { + "left": "issue:44075", + "right": "issue:46365", + "accept": false, + "reason": "Optimizer SGD arguments not applied is unrelated to Trainer import failure from datasets." + }, + { + "left": "issue:43010", + "right": "issue:45278", + "accept": false, + "reason": "Cache update decoration/no_grad behavior is unrelated to broad import errors after upgrading Transformers." + } + ] + }, + "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": 21, + "review_unit_index": 52, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:28282", + "issue:30333", + "issue:33453", + "issue:36296", + "issue:36683", + "issue:41669", + "issue:42890", + "issue:43010", + "issue:43064", + "issue:43065", + "issue:43502", + "issue:43575", + "issue:43576", + "issue:43618", + "issue:43723", + "issue:43824", + "issue:43867", + "issue:43874", + "issue:43976", + "issue:43992", + "issue:44062", + "issue:44077", + "issue:44112", + "issue:44222", + "issue:44246", + "issue:44355", + "issue:44360", + "issue:44387", + "issue:44485", + "issue:44556", + "issue:44734", + "issue:44855", + "issue:44857", + "issue:44938", + "issue:44964", + "issue:45071", + "issue:45125", + "issue:45357", + "issue:45375", + "issue:45440", + "issue:45478", + "issue:45561", + "issue:45600", + "issue:45850", + "issue:46132", + "issue:46202", + "issue:46285", + "issue:46617" + ], + "soft_pairs": [ + "issue:44355|issue:44855", + "issue:44938|issue:45071", + "issue:43867|issue:45375", + "issue:43867|issue:46202", + "issue:36683|issue:43064", + "issue:42890|issue:45561", + "issue:43010|issue:45850", + "issue:41669|issue:44246", + "issue:36296|issue:43575", + "issue:43867|issue:45357", + "issue:43010|issue:43976", + "issue:43618|issue:44077", + "issue:43992|issue:45478", + "issue:28282|issue:33453", + "issue:43618|issue:46285", + "issue:43618|issue:44857", + "issue:44360|issue:45440", + "issue:42890|issue:44964", + "issue:42890|issue:44112", + "issue:42890|issue:45600", + "issue:43502|issue:44062", + "issue:43874|issue:46617", + "issue:43723|issue:44556", + "issue:44222|issue:45125", + "issue:44387|issue:44556", + "issue:44734|issue:46617", + "issue:43010|issue:43065", + "issue:43576|issue:46132", + "issue:43502|issue:44485", + "issue:43824|issue:46617", + "issue:45125|issue:45478", + "issue:30333|issue:33453" + ], + "prepared_review_unit_hash": "94492bbf8a27b7f4265251a5e669b22b2f4bf3a31ee5cdaedb3254f531e41df8", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 36969, + "estimated_input_tokens": 9243, + "estimated_eval_tokens": 18742 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a heterogeneous issue-only cluster. The soft-similarity pairs mostly overlap on broad subsystems (loading, imports, tests, tensor-parallelism) but do not describe the same underlying bug, so no dedupe merges are justified.", + "confidence": 0.98, + "canonical_issue_reason": "No single canonical issue fits this cluster: it spans many unrelated regressions and test flakes rather than one shared bug.", + "canonical_pr_reason": null, + "best_issue_reason": "No issue stands out as a representative duplicate anchor; the cluster is too mixed to pick a meaningful best issue.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44355", + "right": "issue:44855", + "accept": false, + "reason": "Different failures: compiled Python file import errors vs a torch.jit decorator/comment parsing issue." + }, + { + "left": "issue:44938", + "right": "issue:45071", + "accept": false, + "reason": "One is a Python 3.14 load failure; the other is a PretrainedConfig type-checking regression." + }, + { + "left": "issue:43867", + "right": "issue:45375", + "accept": false, + "reason": "Generic load/state_dict sorting issue vs missing Qwen3.5 vision config field." + }, + { + "left": "issue:43867", + "right": "issue:46202", + "accept": false, + "reason": "Different model paths and symptoms: load failure vs Gemma pipeline roundtrip token issue." + }, + { + "left": "issue:36683", + "right": "issue:43064", + "accept": false, + "reason": "Gemma3Config missing `vocab_size` is unrelated to the FSDP2+PEFT optimizer-state bug." + }, + { + "left": "issue:42890", + "right": "issue:45561", + "accept": false, + "reason": "Both are test-related, but one is missing `set_seed()` coverage and the other is a pytest-xdist file race." + }, + { + "left": "issue:43010", + "right": "issue:45850", + "accept": false, + "reason": "A specific `@torch.no_grad` cache/layer update fix is not the same as a vague broad regression report." + }, + { + "left": "issue:41669", + "right": "issue:44246", + "accept": false, + "reason": "Both mention import slowness, but one is about removing `import *` and the other is an intermittent slow import report." + }, + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Tensor-parallel training bug vs OOM when loading a specific Qwen2 model with TP." + }, + { + "left": "issue:43867", + "right": "issue:45357", + "accept": false, + "reason": "Different save/load paths: state_dict sorting vs incorrect visual encoder keys on save_pretrained." + }, + { + "left": "issue:43010", + "right": "issue:43976", + "accept": false, + "reason": "Cache-update no-grad semantics are unrelated to Python version compatibility." + }, + { + "left": "issue:43618", + "right": "issue:44077", + "accept": false, + "reason": "CLIP attentions assignment regression vs patchtsmixer `post_init` validation are different bugs." + }, + { + "left": "issue:43992", + "right": "issue:45478", + "accept": false, + "reason": "UMT5 missing `embed_tokens.weight` is unrelated to Qwen3.5 MoE from_pretrained errors." + }, + { + "left": "issue:28282", + "right": "issue:33453", + "accept": false, + "reason": "Missing PyTorch backend import error is unrelated to tokenizer-loading regression." + }, + { + "left": "issue:43618", + "right": "issue:46285", + "accept": false, + "reason": "Same general CLIP area, but different concrete breakages: missing attentions vs inconsistent model structure." + }, + { + "left": "issue:43618", + "right": "issue:44857", + "accept": false, + "reason": "CLIPOutput attentions regression is unrelated to the LwDetrImageLoss AMP crash." + }, + { + "left": "issue:44360", + "right": "issue:45440", + "accept": false, + "reason": "DSA indexer ReLU discussion and DeepseekV3MoE divergence are unrelated code paths." + }, + { + "left": "issue:42890", + "right": "issue:44964", + "accept": false, + "reason": "A flaky integration test is not the same as a model-loading failure for Phi-4 multimodal." + }, + { + "left": "issue:42890", + "right": "issue:44112", + "accept": false, + "reason": "Both are CI/test issues, but one is missing seeding and the other is a stale GraniteSpeech device override test." + }, + { + "left": "issue:42890", + "right": "issue:45600", + "accept": false, + "reason": "Test flake vs auto_mappings CI break from removed Sam3LiteText configs." + }, + { + "left": "issue:43502", + "right": "issue:44062", + "accept": false, + "reason": "Local-files-only network leak and `AddedToken()` keyword collision are different bugs." + }, + { + "left": "issue:43874", + "right": "issue:46617", + "accept": false, + "reason": "Missing image-patch helper in GLM46V is unrelated to cache FileNotFoundError with trust_remote_code." + }, + { + "left": "issue:43723", + "right": "issue:44556", + "accept": false, + "reason": "Tokenizer loading regression and checkpoint reload-after-upgrade failure are different load paths." + }, + { + "left": "issue:44222", + "right": "issue:45125", + "accept": false, + "reason": "FP8 save_pretrained issue is unrelated to missing tensor-parallel plan metadata." + }, + { + "left": "issue:44387", + "right": "issue:44556", + "accept": false, + "reason": "Quantization memory/OOM regression is unrelated to checkpoint reload compatibility." + }, + { + "left": "issue:44734", + "right": "issue:46617", + "accept": false, + "reason": "KV cache continuation crash and cache FileNotFoundError are different failure modes." + }, + { + "left": "issue:43010", + "right": "issue:43065", + "accept": false, + "reason": "`@torch.no_grad` on cache updates and the dummy Conv2d in Sam3PixelDecoder are unrelated changes." + }, + { + "left": "issue:43576", + "right": "issue:46132", + "accept": false, + "reason": "Broken `transformers env` command and changed `AttentionInterface.register` behavior are separate issues." + }, + { + "left": "issue:43502", + "right": "issue:44485", + "accept": false, + "reason": "Local-files-only network leakage is unrelated to GLM-5 RoPE implementation details." + }, + { + "left": "issue:43824", + "right": "issue:46617", + "accept": false, + "reason": "Qwen2.5-VL import error and trust_remote_code cache FileNotFoundError are not the same bug." + }, + { + "left": "issue:45125", + "right": "issue:45478", + "accept": false, + "reason": "Missing `_tp_plan` for tensor parallelism is different from the Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:30333", + "right": "issue:33453", + "accept": false, + "reason": "MLflow job-status reporting and tokenizer-loading regression 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": 21, + "review_unit_index": 53, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:27957", + "issue:30333", + "issue:30990", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:36296", + "issue:38617", + "issue:41628", + "issue:42548", + "issue:42757", + "issue:43010", + "issue:43232", + "issue:43299", + "issue:43475", + "issue:43502", + "issue:43526", + "issue:43576", + "issue:43673", + "issue:43824", + "issue:43827", + "issue:43901", + "issue:44222", + "issue:44230", + "issue:44315", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44485", + "issue:44492", + "issue:44556", + "issue:44623", + "issue:44655", + "issue:44749", + "issue:44908", + "issue:44938", + "issue:44961", + "issue:45071", + "issue:45125", + "issue:45230", + "issue:45362", + "issue:45375", + "issue:45468", + "issue:45850", + "issue:46143", + "issue:46144", + "issue:46209", + "issue:46617" + ], + "soft_pairs": [ + "issue:44355|issue:44749", + "issue:42548|issue:44351", + "issue:45125|issue:45375", + "issue:43010|issue:43232", + "issue:43502|issue:44222", + "issue:30990|issue:31795", + "issue:43502|issue:44961", + "issue:43827|issue:44961", + "issue:43299|issue:44556", + "issue:30333|issue:36296", + "issue:44355|issue:45071", + "issue:44355|issue:44655", + "issue:33453|issue:34689", + "issue:27957|issue:33453", + "issue:42757|issue:43576", + "issue:43576|issue:45071", + "issue:44315|issue:44360", + "issue:43576|issue:43824", + "issue:31795|issue:38617", + "issue:44961|issue:45468", + "issue:44351|issue:46617", + "issue:43576|issue:44938", + "issue:44485|issue:44961", + "issue:44908|issue:44961", + "issue:44222|issue:44230", + "issue:36296|issue:43901", + "issue:46143|issue:46144", + "issue:36296|issue:43526", + "issue:41628|issue:42548", + "issue:43475|issue:46617", + "issue:44315|issue:46209", + "issue:45362|issue:45850", + "issue:43673|issue:44492", + "issue:44623|issue:45230" + ], + "prepared_review_unit_hash": "70b8d53f40524fc89b34bb0541a07a6ec8d8dd7754549664e96914829ce7e7fc", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37020, + "estimated_input_tokens": 9255, + "estimated_eval_tokens": 18766 + }, + "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, import errors, loading issues, and docs problems. None of the soft pairs look like true duplicates; the closest ones are still too different in concrete code path or symptom. The best representative issue is the broad v5 breakage report.", + "confidence": 0.63, + "canonical_issue_reason": "issue:45850 is the broadest umbrella in the set: it captures the general 'v5 breaks many models' theme that recurs across the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45850 is the best global representative because it spans the widest set of user-facing v5 regressions, even though it is too generic to be a precise duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44355", + "right": "issue:44749", + "accept": false, + "reason": "Different failures: compiled Python file errors vs slower filtering after upgrade." + }, + { + "left": "issue:42548", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different symbols and likely different fixes." + }, + { + "left": "issue:45125", + "right": "issue:45375", + "accept": false, + "reason": "Different Qwen3.5-MoE problems: missing tensor-parallel plan vs dropped config field." + }, + { + "left": "issue:43010", + "right": "issue:43232", + "accept": false, + "reason": "Both touch generation internals, but the underlying bugs and code paths differ." + }, + { + "left": "issue:43502", + "right": "issue:44222", + "accept": false, + "reason": "Local-files-only network leakage vs FP8 save_pretrained support are unrelated." + }, + { + "left": "issue:30990", + "right": "issue:31795", + "accept": false, + "reason": "Loading stall vs documentation confusion; not the same bug." + }, + { + "left": "issue:43502", + "right": "issue:44961", + "accept": false, + "reason": "API requests during load and the placeholder issue 'racoon' are unrelated." + }, + { + "left": "issue:43827", + "right": "issue:44961", + "accept": false, + "reason": "Docs referencing removed pipeline() vs unrelated placeholder issue." + }, + { + "left": "issue:43299", + "right": "issue:44556", + "accept": false, + "reason": "Both are version-regression loading issues, but for different models and failure modes." + }, + { + "left": "issue:30333", + "right": "issue:36296", + "accept": false, + "reason": "MLflow reporting on failed jobs is unrelated to tensor-parallel training." + }, + { + "left": "issue:44355", + "right": "issue:45071", + "accept": false, + "reason": "Compiled-file execution errors and PretrainedConfig type-checking are different problems." + }, + { + "left": "issue:44355", + "right": "issue:44655", + "accept": false, + "reason": "Compiled-file errors vs save_pretrained on Pipeline objects are unrelated." + }, + { + "left": "issue:33453", + "right": "issue:34689", + "accept": false, + "reason": "Both mention loading regressions, but they affect different models and likely different loaders." + }, + { + "left": "issue:27957", + "right": "issue:33453", + "accept": false, + "reason": "Flash Attention 2 support for XLMRoberta is unrelated to tokenizer loading regressions." + }, + { + "left": "issue:42757", + "right": "issue:43576", + "accept": false, + "reason": "Missing hub import and the broken transformers env command are different failures." + }, + { + "left": "issue:43576", + "right": "issue:45071", + "accept": false, + "reason": "Broken env command and type-checking breakage are not the same issue." + }, + { + "left": "issue:44315", + "right": "issue:44360", + "accept": false, + "reason": "Liger-kernel/model_init integration and DSA ReLU discussion are unrelated." + }, + { + "left": "issue:43576", + "right": "issue:43824", + "accept": false, + "reason": "Env-command breakage and Qwen2.5-VL import failure are different code paths." + }, + { + "left": "issue:31795", + "right": "issue:38617", + "accept": false, + "reason": "Docs confusion in model.forward is unrelated to an import error for layer_type_validation." + }, + { + "left": "issue:44961", + "right": "issue:45468", + "accept": false, + "reason": "Placeholder issue vs Gemma-4 audio positional encoding bug are unrelated." + }, + { + "left": "issue:44351", + "right": "issue:46617", + "accept": false, + "reason": "HybridCache import failure and trust_remote_code cache FileNotFoundError are different issues." + }, + { + "left": "issue:43576", + "right": "issue:44938", + "accept": false, + "reason": "Broken env command in v5 vs Python 3.14 load failure are not the same underlying bug." + }, + { + "left": "issue:44485", + "right": "issue:44961", + "accept": false, + "reason": "GLM-5 RoPE implementation and placeholder issue are unrelated." + }, + { + "left": "issue:44908", + "right": "issue:44961", + "accept": false, + "reason": "Scheduler kwargs bug vs placeholder issue are unrelated." + }, + { + "left": "issue:44222", + "right": "issue:44230", + "accept": false, + "reason": "FP8 save_pretrained behavior and FP8 inference support are distinct problems." + }, + { + "left": "issue:36296", + "right": "issue:43901", + "accept": false, + "reason": "Tensor-parallel training bug and pipeline docs mismatch are unrelated." + }, + { + "left": "issue:46143", + "right": "issue:46144", + "accept": false, + "reason": "Closely related RoFormer issues, but not the same concrete bug." + }, + { + "left": "issue:36296", + "right": "issue:43526", + "accept": false, + "reason": "Tensor-parallel training and BeitImageProcessorFast label reduction are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:42548", + "accept": false, + "reason": "Different missing exports: AutoImageProcessor vs PreTrainedModel." + }, + { + "left": "issue:43475", + "right": "issue:46617", + "accept": false, + "reason": "SAM 3 Video missing attribute and cache FileNotFoundError are unrelated." + }, + { + "left": "issue:44315", + "right": "issue:46209", + "accept": false, + "reason": "Model_init Liger-kernel issue and FP8 inference on B200 are different bugs." + }, + { + "left": "issue:45362", + "right": "issue:45850", + "accept": false, + "reason": "Specific Qwen3.5 chat crash is not the same as the broad 'v5 breaks many models' complaint." + }, + { + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "A concrete save_pretrained file bug and an unspecified 'Bug report' are not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 54, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:31795", + "issue:33453", + "issue:41669", + "issue:42371", + "issue:42890", + "issue:43352", + "issue:43408", + "issue:43525", + "issue:43572", + "issue:43576", + "issue:43673", + "issue:43827", + "issue:43976", + "issue:44038", + "issue:44077", + "issue:44230", + "issue:44246", + "issue:44273", + "issue:44295", + "issue:44297", + "issue:44315", + "issue:44351", + "issue:44360", + "issue:44393", + "issue:44492", + "issue:44556", + "issue:44655", + "issue:44704", + "issue:44829", + "issue:44861", + "issue:44908", + "issue:45003", + "issue:45071", + "issue:45092", + "issue:45125", + "issue:45230", + "issue:45278", + "issue:45310", + "issue:45341", + "issue:45468", + "issue:45478", + "issue:45542", + "issue:45753", + "issue:45850", + "issue:46095", + "issue:46218", + "issue:46328" + ], + "soft_pairs": [ + "issue:44273|issue:45542", + "issue:44038|issue:45125", + "issue:44273|issue:45003", + "issue:44393|issue:45071", + "issue:44393|issue:45753", + "issue:43576|issue:44556", + "issue:43525|issue:45850", + "issue:41669|issue:43673", + "issue:44230|issue:46095", + "issue:44556|issue:45092", + "issue:43572|issue:45850", + "issue:45278|issue:45850", + "issue:44492|issue:45850", + "issue:42371|issue:44492", + "issue:41669|issue:46218", + "issue:43408|issue:44315", + "issue:42890|issue:44908", + "issue:43352|issue:44393", + "issue:43576|issue:44908", + "issue:44246|issue:44273", + "issue:43976|issue:44393", + "issue:44295|issue:45230", + "issue:44704|issue:44908", + "issue:44908|issue:45071", + "issue:44297|issue:45230", + "issue:41669|issue:44351", + "issue:44492|issue:46328", + "issue:41669|issue:45341", + "issue:43976|issue:45850", + "issue:44393|issue:45478", + "issue:43976|issue:46218", + "issue:44077|issue:44861", + "issue:43827|issue:44908", + "issue:31795|issue:33453", + "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:43576|issue:46218" + ], + "prepared_review_unit_hash": "b09c14bbdbaf7438408ac2f81dc1ce09b26b28754e1985dce5b254fb330a3a9b", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 41, + "serialized_chars": 38850, + "estimated_input_tokens": 9713, + "estimated_eval_tokens": 19682 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab bag of Transformers v5 regression reports and model-specific compatibility bugs. The titles are superficially similar, but the concrete failure modes differ enough that none of the soft edges look like true duplicates.", + "confidence": 0.39, + "canonical_issue_reason": "issue:45850 is the broadest umbrella report here, capturing the general 'v5 breaks too many models' theme that many of the other reports orbit around.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45850 is the best cluster anchor by scope, but it is still too vague to represent the concrete bugs in the other issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Lazy loading/import performance is a different problem from the undefined TensorFlow backend error." + }, + { + "left": "issue:44038", + "right": "issue:45125", + "accept": false, + "reason": "Both involve Qwen3/Moe, but one is a broader model bug and the other is missing tensor-parallel metadata." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Lazy loading not working and unsafe sys.modules access are related to imports, but not the same bug." + }, + { + "left": "issue:44393", + "right": "issue:45071", + "accept": false, + "reason": "Qwen3-VL bounding-box errors are unrelated to PretrainedConfig type-checking regressions." + }, + { + "left": "issue:44393", + "right": "issue:45753", + "accept": false, + "reason": "Both are Qwen3-VL-family issues, but one is bbox output failure 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 incompatibility are both v5 regressions, but they are distinct code paths." + }, + { + "left": "issue:43525", + "right": "issue:45850", + "accept": false, + "reason": "Missing pad_token_id in Llama4Config is a specific config bug, not the broad model-breakage complaint." + }, + { + "left": "issue:41669", + "right": "issue:43673", + "accept": false, + "reason": "Import-star slowdown during model import is different from missing GenerationMixin cache during chunked prefill." + }, + { + "left": "issue:44230", + "right": "issue:46095", + "accept": false, + "reason": "FP8 support for Qwen3-VL/Moe and a DeepSeek conversion-script question are not the same issue." + }, + { + "left": "issue:44556", + "right": "issue:45092", + "accept": false, + "reason": "Both mention upgrading to v5, but one is general checkpoint reload breakage and the other is remote-code meta-init incompatibility." + }, + { + "left": "issue:43572", + "right": "issue:45850", + "accept": false, + "reason": "StableLmConfig pad_token_idx is a specific config migration bug, not the broad v5 breakage report." + }, + { + "left": "issue:45278", + "right": "issue:45850", + "accept": false, + "reason": "Both are broad v5 regression complaints, but one is specifically about import errors after upgrade." + }, + { + "left": "issue:44492", + "right": "issue:45850", + "accept": false, + "reason": "A cache-strategy typo is a narrow docs/wording issue, not the general model-breakage problem." + }, + { + "left": "issue:42371", + "right": "issue:44492", + "accept": false, + "reason": "TF32 API guidance and cache-strategy typo are unrelated fixes." + }, + { + "left": "issue:41669", + "right": "issue:46218", + "accept": false, + "reason": "Import-star performance regression is not the same as failing to import Qwen3_5ForCausalLM." + }, + { + "left": "issue:43408", + "right": "issue:44315", + "accept": false, + "reason": "sam3_tracker config warning and Liger Kernel not applied with model_init are unrelated issues." + }, + { + "left": "issue:42890", + "right": "issue:44908", + "accept": false, + "reason": "Missing set_seed in tests and inverse_sqrt kwargs being ignored are unrelated test/scheduler problems." + }, + { + "left": "issue:43352", + "right": "issue:44393", + "accept": false, + "reason": "Flash Attention 2 support for NemotronH and Qwen3-VL bbox output errors are different model-specific bugs." + }, + { + "left": "issue:43576", + "right": "issue:44908", + "accept": false, + "reason": "The env command regression is not the same underlying bug as scheduler kwargs being ignored." + }, + { + "left": "issue:44246", + "right": "issue:44273", + "accept": false, + "reason": "Both concern import behavior, but one is intermittent slow import and the other is lazy loading failure." + }, + { + "left": "issue:43976", + "right": "issue:44393", + "accept": false, + "reason": "Python-version incompatibility is unrelated to the Qwen3-VL hallucination/bbox issue." + }, + { + "left": "issue:44295", + "right": "issue:45230", + "accept": false, + "reason": "A position_ids buffer access error is too specific to be merged with a generic bug report." + }, + { + "left": "issue:44704", + "right": "issue:44908", + "accept": false, + "reason": "AutoProcessor cached_file kwargs handling is unrelated to inverse_sqrt scheduler kwargs." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Scheduler kwargs plumbing and PretrainedConfig type checking are different failures." + }, + { + "left": "issue:44297", + "right": "issue:45230", + "accept": false, + "reason": "Tokenizer save_pretrained mismatch is a concrete bug; the other item is too generic to match." + }, + { + "left": "issue:44315", + "right": "issue:44829", + "accept": false, + "reason": "Liger Kernel application in model_init and degenerate flash_attention_3 training are different code-path problems." + }, + { + "left": "issue:41669", + "right": "issue:44351", + "accept": false, + "reason": "Import-star slowdown is not the same as a missing HybridCache export." + }, + { + "left": "issue:44492", + "right": "issue:46328", + "accept": false, + "reason": "Cache-strategy typo and SequenceFeatureExtractor.pad performance are unrelated." + }, + { + "left": "issue:41669", + "right": "issue:45341", + "accept": false, + "reason": "Model import slowdown and a testing_utils bug are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:45850", + "accept": false, + "reason": "Python 3.9 incompatibility is a different regression class from broad model breakage." + }, + { + "left": "issue:44393", + "right": "issue:45478", + "accept": false, + "reason": "Qwen3-VL bbox hallucinations and Qwen3.5 Moe from_pretrained errors are different model failures." + }, + { + "left": "issue:43976", + "right": "issue:46218", + "accept": false, + "reason": "Python-version support issues are unrelated to importing Qwen3_5ForCausalLM." + }, + { + "left": "issue:44077", + "right": "issue:44861", + "accept": false, + "reason": "Optional post_init policy and _get_tied_weight_keys crashing are different model-initialization bugs." + }, + { + "left": "issue:43827", + "right": "issue:44908", + "accept": false, + "reason": "Docs still referencing pipeline() is unrelated to scheduler kwargs handling." + }, + { + "left": "issue:31795", + "right": "issue:33453", + "accept": false, + "reason": "Model.forward docs confusion and tokenizer loading regression are unrelated issues." + }, + { + "left": "issue:44315", + "right": "issue:44829", + "accept": false, + "reason": "Model-init integration for Liger Kernel is not the same as flash_attention_3 causing degenerate training." + }, + { + "left": "issue:43576", + "right": "issue:44273", + "accept": false, + "reason": "The env command regression is different from lazy loading not working properly." + }, + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "DSA indexer activation discussion and Gemma-4 audio positional encoding bug are unrelated." + }, + { + "left": "issue:44655", + "right": "issue:44861", + "accept": false, + "reason": "Pipeline save_pretrained failure and tied-weight-key crashes are separate serialization/init bugs." + }, + { + "left": "issue:45230", + "right": "issue:45310", + "accept": false, + "reason": "A generic bug report does not identify the same Qwen3.5 Moe from_pretrained error." + }, + { + "left": "issue:44273", + "right": "issue:44908", + "accept": false, + "reason": "Lazy loading/import behavior and inverse_sqrt scheduler kwargs are unrelated." + }, + { + "left": "issue:43576", + "right": "issue:46218", + "accept": false, + "reason": "Broken env command and missing Qwen3_5ForCausalLM import are different import-related 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": 21, + "review_unit_index": 55, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:28348", + "issue:30333", + "issue:31515", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:36296", + "issue:38617", + "issue:41669", + "issue:42371", + "issue:42913", + "issue:42936", + "issue:43066", + "issue:43240", + "issue:43298", + "issue:43352", + "issue:43502", + "issue:43519", + "issue:43525", + "issue:43572", + "issue:43606", + "issue:43618", + "issue:43668", + "issue:43704", + "issue:43717", + "issue:43813", + "issue:44031", + "issue:44230", + "issue:44261", + "issue:44263", + "issue:44351", + "issue:44485", + "issue:44492", + "issue:44623", + "issue:44861", + "issue:44908", + "issue:44938", + "issue:44998", + "issue:45250", + "issue:45290", + "issue:45310", + "issue:45468", + "issue:45850", + "issue:46018", + "issue:46095", + "issue:46144", + "issue:46172", + "issue:46218" + ], + "soft_pairs": [ + "issue:43519|issue:44485", + "issue:43519|issue:44908", + "issue:41669|issue:43704", + "issue:46018|issue:46172", + "issue:31795|issue:34689", + "issue:31515|issue:31795", + "issue:43704|issue:44485", + "issue:44485|issue:45468", + "issue:45850|issue:46218", + "issue:42371|issue:43704", + "issue:43502|issue:43519", + "issue:43519|issue:44623", + "issue:43519|issue:43704", + "issue:44263|issue:45468", + "issue:43298|issue:46095", + "issue:43606|issue:45468", + "issue:44261|issue:45468", + "issue:44351|issue:46218", + "issue:43352|issue:46218", + "issue:42913|issue:43066", + "issue:28348|issue:45250", + "issue:44908|issue:44998", + "issue:44492|issue:44998", + "issue:43618|issue:44998", + "issue:44998|issue:45850", + "issue:44938|issue:44998", + "issue:44861|issue:45290", + "issue:43525|issue:44861", + "issue:31515|issue:38617", + "issue:43572|issue:44861", + "issue:28348|issue:43352", + "issue:44230|issue:44623", + "issue:30333|issue:31515", + "issue:31515|issue:36296", + "issue:44230|issue:45310", + "issue:31515|issue:33453", + "issue:43298|issue:44230", + "issue:43717|issue:46144", + "issue:42936|issue:46018", + "issue:43668|issue:44031", + "issue:42936|issue:43240", + "issue:43240|issue:46144", + "issue:42936|issue:44031", + "issue:42936|issue:46144", + "issue:42936|issue:43813", + "issue:43240|issue:43668", + "issue:43668|issue:43813" + ], + "prepared_review_unit_hash": "7eb27afa28ecbce1b6fb1896d7597741d032b3bd6ef31d752230824f4241bad0", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 47, + "serialized_chars": 40842, + "estimated_input_tokens": 10211, + "estimated_eval_tokens": 20678 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a loose grab-bag of Transformers issues: model-specific regressions, tokenizer/loading bugs, performance problems, and import/API breakages. There is no strong single underlying defect tying the cluster together; the closest thing to an umbrella topic is broad v5 regression fallout.", + "confidence": 0.41, + "canonical_issue_reason": "issue:45850 is the broadest umbrella-style report in the set ('Since 5.0 version it breaks too many models'), so it best represents the cluster despite being vague.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45850 is the most cluster-like representative because it covers multiple model breakages rather than one narrow symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43519", + "right": "issue:44485", + "accept": false, + "reason": "Different model families and different bugs: timestamp calculation vs RoPE implementation." + }, + { + "left": "issue:43519", + "right": "issue:44908", + "accept": false, + "reason": "Processor timestamp math and scheduler kwargs are unrelated code paths." + }, + { + "left": "issue:41669", + "right": "issue:43704", + "accept": false, + "reason": "Import-star performance issue and VRAM leak are unrelated symptoms." + }, + { + "left": "issue:46018", + "right": "issue:46172", + "accept": false, + "reason": "Both mention deepseek-v4, but one is gating logic and the other is a tensor dtype/transpose bug." + }, + { + "left": "issue:31795", + "right": "issue:34689", + "accept": false, + "reason": "Documentation confusion for forward args vs a model-loading regression are not the same bug." + }, + { + "left": "issue:31515", + "right": "issue:31795", + "accept": false, + "reason": "Slow from_pretrained loading and forward-doc confusion are different issues." + }, + { + "left": "issue:43704", + "right": "issue:44485", + "accept": false, + "reason": "VRAM leak under dataloader threads does not match RoPE implementation concerns." + }, + { + "left": "issue:44485", + "right": "issue:45468", + "accept": false, + "reason": "Different models and different numerical/encoding problems." + }, + { + "left": "issue:45850", + "right": "issue:46218", + "accept": false, + "reason": "Broad version-breakage complaint vs a specific import error; not the same failure mode." + }, + { + "left": "issue:42371", + "right": "issue:43704", + "accept": false, + "reason": "TF32 configuration API guidance is unrelated to a VRAM leak." + }, + { + "left": "issue:43502", + "right": "issue:43519", + "accept": false, + "reason": "local_files_only network access and timestamp calculation are different bugs." + }, + { + "left": "issue:43519", + "right": "issue:44623", + "accept": false, + "reason": "Processor timestamp math and save_pretrained missing files are separate processor issues." + }, + { + "left": "issue:43519", + "right": "issue:43704", + "accept": false, + "reason": "Unrelated processor bug vs multi-thread VRAM leak." + }, + { + "left": "issue:44263", + "right": "issue:45468", + "accept": false, + "reason": "Different model internals and different arithmetic/shape issues." + }, + { + "left": "issue:43298", + "right": "issue:46095", + "accept": false, + "reason": "Continuous batching audio support and a DeepSeek-V4 weight-conversion question are not the same bug." + }, + { + "left": "issue:43606", + "right": "issue:45468", + "accept": false, + "reason": "CPU offload device mismatch vs Gemma-4 positional encoding are unrelated." + }, + { + "left": "issue:44261", + "right": "issue:45468", + "accept": false, + "reason": "Layernorm epsilon precision bug and Gemma-4 positional encoding are different code paths." + }, + { + "left": "issue:44351", + "right": "issue:46218", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols/modules." + }, + { + "left": "issue:43352", + "right": "issue:46218", + "accept": false, + "reason": "Flash Attention support gap for Nemotron is unrelated to importing Qwen3_5ForCausalLM." + }, + { + "left": "issue:42913", + "right": "issue:43066", + "accept": false, + "reason": "Both are tokenizer v5-related, but one is a broad behavioral regression and the other a specific decoder-type issue." + }, + { + "left": "issue:28348", + "right": "issue:45250", + "accept": false, + "reason": "GPT2LMHeadModel-specific Flash Attention support request vs a generic Flash Attention topic." + }, + { + "left": "issue:44908", + "right": "issue:44998", + "accept": false, + "reason": "Scheduler kwargs bug and an unrelated issue titled 'Unemployment' are clearly not the same." + }, + { + "left": "issue:44492", + "right": "issue:44998", + "accept": false, + "reason": "A typo in cache strategy docs/code is unrelated to the other issue." + }, + { + "left": "issue:43618", + "right": "issue:44998", + "accept": false, + "reason": "CLIPOutput attentions regression is unrelated to the other issue." + }, + { + "left": "issue:44998", + "right": "issue:45850", + "accept": false, + "reason": "The latter is a broad Transformers v5 regression; 'Unemployment' is unrelated noise." + }, + { + "left": "issue:44938", + "right": "issue:44998", + "accept": false, + "reason": "Python 3.14 load failure is unrelated to the other issue." + }, + { + "left": "issue:44861", + "right": "issue:45290", + "accept": false, + "reason": "Tied-weights key handling and chat template tool-call crashes are different bugs." + }, + { + "left": "issue:43525", + "right": "issue:44861", + "accept": false, + "reason": "Missing pad_token_id on Llama4Config is unrelated to tied-weight key handling." + }, + { + "left": "issue:31515", + "right": "issue:38617", + "accept": false, + "reason": "Slow checkpoint loading and an import error for layer_type_validation are unrelated." + }, + { + "left": "issue:43572", + "right": "issue:44861", + "accept": false, + "reason": "StableLmConfig pad_token_idx regression is not the same as tied-weight key crashes." + }, + { + "left": "issue:28348", + "right": "issue:43352", + "accept": false, + "reason": "Different models and different Flash Attention 2.0 support gaps." + }, + { + "left": "issue:44230", + "right": "issue:44623", + "accept": false, + "reason": "FP8 support request and missing processor save files are unrelated." + }, + { + "left": "issue:30333", + "right": "issue:31515", + "accept": false, + "reason": "MLFlow reporting on failed jobs and slow from_pretrained loading are different problems." + }, + { + "left": "issue:31515", + "right": "issue:36296", + "accept": false, + "reason": "Slow checkpoint loading and tensor-parallel training bug are unrelated." + }, + { + "left": "issue:44230", + "right": "issue:45310", + "accept": false, + "reason": "FP8 support request and Qwen3.5 MoE from_pretrained error are different issues." + }, + { + "left": "issue:31515", + "right": "issue:33453", + "accept": false, + "reason": "Load slowness and tokenizer-loading regression are not the same bug." + }, + { + "left": "issue:43298", + "right": "issue:44230", + "accept": false, + "reason": "Audio continuous batching support and FP8 inference support are separate feature gaps." + }, + { + "left": "issue:43717", + "right": "issue:46144", + "accept": false, + "reason": "Init-weights behavior in Mamba models and RoFormer attention interface usage are unrelated." + }, + { + "left": "issue:42936", + "right": "issue:46018", + "accept": false, + "reason": "Mask2Former ignore_value misuse and DeepSeek-V4 shared-expert gating are different model-specific bugs." + }, + { + "left": "issue:43668", + "right": "issue:44031", + "accept": false, + "reason": "Incorrect type hint and regex warning regression are unrelated." + }, + { + "left": "issue:42936", + "right": "issue:43240", + "accept": false, + "reason": "Mask2Former ignore_value and loss kwargs propagation are different issues." + }, + { + "left": "issue:43240", + "right": "issue:46144", + "accept": false, + "reason": "Loss kwargs propagation and RoFormer attention interface usage are unrelated." + }, + { + "left": "issue:42936", + "right": "issue:44031", + "accept": false, + "reason": "Different subsystems: Mask2Former config/use bug vs tokenizer regex warnings." + }, + { + "left": "issue:42936", + "right": "issue:46144", + "accept": false, + "reason": "Different model families and unrelated bugs." + }, + { + "left": "issue:42936", + "right": "issue:43813", + "accept": false, + "reason": "Mask2Former ignore_value issue is unrelated to the typo in conversion code." + }, + { + "left": "issue:43240", + "right": "issue:43668", + "accept": false, + "reason": "Loss-kwargs forwarding and a config type-hint typo are unrelated." + }, + { + "left": "issue:43668", + "right": "issue:43813", + "accept": false, + "reason": "Type-hint correction and conversion typo are different trivial fixes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 21, + "review_unit_index": 56, + "review_unit_count": 56, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-469", + "nodes": [ + "issue:37934", + "issue:42936", + "issue:43668" + ], + "soft_pairs": [ + "issue:37934|issue:43668", + "issue:42936|issue:43668" + ], + "prepared_review_unit_hash": "f1ebd3f31807d351532933f488be4730514dd107d0089edacfe24b9b0a47be73", + "component_budget": { + "node_count": 469, + "item_count": 469, + "soft_pair_count": 1718, + "serialized_chars": 891778, + "estimated_input_tokens": 222945, + "estimated_eval_tokens": 446146 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2312, + "estimated_input_tokens": 578, + "estimated_eval_tokens": 1412 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These three issues are unrelated: one asks about model design, one reports an unused ignore_value in Mask2Former, and one reports an incorrect type hint in ModernBERTConfig. No pair looks like the same underlying bug or change.", + "confidence": 0.93, + "canonical_issue_reason": "issue:42936 is the most concrete and actionable bug report: it identifies a specific runtime/behavioral problem (ignore_value defined but not used) with a clear code path in Mask2Former.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:42936 is the best representative issue because it describes an actual functional defect rather than a design question or a type annotation mismatch.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:37934", + "right": "issue:43668", + "accept": false, + "reason": "Different subjects and failure modes: Llama4TextL2Norm normalization semantics vs. ModernBERTConfig type annotation. No shared code path or concrete same-bug evidence." + }, + { + "left": "issue:42936", + "right": "issue:43668", + "accept": false, + "reason": "One is a Mask2Former behavioral bug about ignore_value not being used; the other is a ModernBERTConfig type-hint issue. They are distinct and not mergeable as one underlying fix." + } + ] + }, + "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": 21, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10519-3", + "nodes": [ + "issue:10519", + "issue:11005", + "issue:11307" + ], + "soft_pairs": [ + "issue:10519|issue:11307", + "issue:10519|issue:11005", + "issue:11005|issue:11307" + ], + "prepared_review_unit_hash": "96f90a9a0e55dc267a3f6fe181e4a74fa8c9261c1f0dbd3b8fa0fce7a92c0b0d", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 2874, + "estimated_input_tokens": 719, + "estimated_eval_tokens": 1694 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 2874, + "estimated_input_tokens": 719, + "estimated_eval_tokens": 1694 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The three issues are about unrelated feature requests: truncation direction in tokenization, ReduceLROnPlateau-like scheduling behavior, and Wav2Vec2 word time offsets. There is no credible duplicate relationship among them.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:10519", + "right": "issue:11307", + "accept": false, + "reason": "Different domains and code paths: token truncation behavior vs. Wav2Vec2 word boundary offsets." + }, + { + "left": "issue:10519", + "right": "issue:11005", + "accept": false, + "reason": "Unrelated requests: truncation strategy in preprocessing vs. learning-rate scheduler functionality." + }, + { + "left": "issue:11005", + "right": "issue:11307", + "accept": false, + "reason": "No shared underlying bug or change; scheduler feature request and Wav2Vec2 timing offsets 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": 3, + "component_count": 21, + "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": "b7c348bdc4f8ada873d1df621d290ccd67e5a5ba6ccc08a4efb235fd5c539a10", + "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": "This cluster is not cohesive: issue #13244 is about Tapas tokenization and is unrelated to the model-loading/login problems in #15980 and #16618. The title overlap between #15980 and #16618 is too weak to treat them as the same underlying bug without stronger body/diff evidence.", + "confidence": 0.94, + "canonical_issue_reason": "If forced to pick one, #16618 is the broadest and most discussion-heavy model-loading report, but it is still only a loose fit and does not represent the whole cluster well.", + "canonical_pr_reason": null, + "best_issue_reason": "No strong global best issue exists because the cluster mixes unrelated problems; #16618 is only the closest representative by breadth, not by true semantic match.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:13244", + "right": "issue:16618", + "accept": false, + "reason": "Different problem domains: Tapas tokenization vs. model loading/authentication. Same-subsystem wording is not enough to make them the same bug." + }, + { + "left": "issue:15980", + "right": "issue:16618", + "accept": false, + "reason": "Both involve model loading, but one is specifically about a bad error when downloading a private model without login, while the other is a generic failure to load bert-base-uncased. Not enough evidence they are the same concrete issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 4, + "component_count": 21, + "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": "94e44515163ddc2cddbe9fa99ae3c1d7a222c982c5d8241a5f88497ce36b03d3", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1590, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1590, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are not duplicates: one tracks Flash Attention 2 support for more architectures, the other tracks GGUF support for more architectures. They are separate feature efforts with different implementation paths.", + "confidence": 0.98, + "canonical_issue_reason": "Issue 26350 is the older and more-discussed tracker, so it is the better representative if one issue must stand in for the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 26350 has higher discussion activity and broader momentum, making it the best issue representative here.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:26350", + "right": "issue:33260", + "accept": false, + "reason": "Different features and code paths: Flash Attention 2 support vs GGUF support. They share a broad \u2018community contribution\u2019 framing only, not the same underlying change or bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 5, + "component_count": 21, + "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-27088-8", + "nodes": [ + "issue:27088", + "issue:28936", + "issue:38435", + "issue:43834", + "issue:44365", + "issue:44370", + "issue:46222", + "issue:46226" + ], + "soft_pairs": [ + "issue:27088|issue:44370", + "issue:27088|issue:44365", + "issue:27088|issue:43834", + "issue:27088|issue:38435", + "issue:27088|issue:28936", + "issue:28936|issue:43834", + "issue:28936|issue:44365", + "issue:28936|issue:38435", + "issue:28936|issue:44370", + "issue:38435|issue:44370", + "issue:38435|issue:44365", + "issue:38435|issue:43834", + "issue:38435|issue:46222", + "issue:38435|issue:46226", + "issue:43834|issue:44370", + "issue:43834|issue:44365", + "issue:43834|issue:46222", + "issue:43834|issue:46226", + "issue:44365|issue:44370", + "issue:44365|issue:46222", + "issue:44365|issue:46226", + "issue:44370|issue:46222", + "issue:44370|issue:46226", + "issue:46222|issue:46226" + ], + "prepared_review_unit_hash": "395f65a190c06ba61c1a7773c94a719b488dd2a95eb86a5087ec0eced2bca8ef", + "component_budget": { + "node_count": 8, + "item_count": 8, + "soft_pair_count": 24, + "serialized_chars": 13529, + "estimated_input_tokens": 3383, + "estimated_eval_tokens": 7022 + }, + "budget": { + "node_count": 8, + "item_count": 8, + "soft_pair_count": 24, + "serialized_chars": 13529, + "estimated_input_tokens": 3383, + "estimated_eval_tokens": 7022 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly a mix of similar-looking docs-translation requests across different languages. Only the exact placeholder-template duplicates and the identical Thai pair look like true duplicates; the concrete Turkish/Spanish/Romanian issues are not the same underlying issue.", + "confidence": 0.79, + "canonical_issue_reason": "issue:43834 is the oldest exact-match member of the repeated placeholder-template subgroup, so it is the best canonical target for that duplicated text.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43834 is the clearest representative of the exact duplicate subgroup; there is no single best issue for the whole mixed cluster because several language-specific requests are distinct.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:27088", + "right": "issue:44370", + "accept": false, + "reason": "Different language-targeted docs requests; same broad theme is not enough to make them duplicates." + }, + { + "left": "issue:27088", + "right": "issue:44365", + "accept": false, + "reason": "Turkish translation request vs. a generic placeholder-title issue; not the same concrete issue." + }, + { + "left": "issue:27088", + "right": "issue:43834", + "accept": false, + "reason": "Concrete Turkish request does not match the generic placeholder-template issue." + }, + { + "left": "issue:27088", + "right": "issue:38435", + "accept": false, + "reason": "Different target languages, so these are separate translation requests." + }, + { + "left": "issue:27088", + "right": "issue:28936", + "accept": false, + "reason": "Turkish and Spanish docs translation requests are distinct issues." + }, + { + "left": "issue:28936", + "right": "issue:43834", + "accept": false, + "reason": "Spanish request is concrete; the other item is a generic placeholder-template issue." + }, + { + "left": "issue:28936", + "right": "issue:44365", + "accept": false, + "reason": "Spanish translation request vs. placeholder-template issue; not the same underlying bug/change." + }, + { + "left": "issue:28936", + "right": "issue:38435", + "accept": false, + "reason": "Spanish and Romanian docs translation requests are different language targets." + }, + { + "left": "issue:28936", + "right": "issue:44370", + "accept": false, + "reason": "Concrete Spanish request does not match the generic placeholder-template issue." + }, + { + "left": "issue:38435", + "right": "issue:44370", + "accept": false, + "reason": "Romanian request vs. generic placeholder-template issue; not enough to merge." + }, + { + "left": "issue:38435", + "right": "issue:44365", + "accept": false, + "reason": "Romanian request is concrete, while the other item is a generic placeholder-template issue." + }, + { + "left": "issue:38435", + "right": "issue:43834", + "accept": false, + "reason": "Different issue contents: a specific Romanian request versus a generic placeholder title." + }, + { + "left": "issue:38435", + "right": "issue:46222", + "accept": false, + "reason": "Romanian translation request and the Thai-text issue are unrelated." + }, + { + "left": "issue:38435", + "right": "issue:46226", + "accept": false, + "reason": "Romanian translation request and the Thai-text issue are unrelated." + }, + { + "left": "issue:43834", + "right": "issue:44370", + "accept": true, + "reason": "Same placeholder translation-doc title and near-identical body; looks like a duplicate issue." + }, + { + "left": "issue:43834", + "right": "issue:44365", + "accept": true, + "reason": "Same placeholder title and almost identical issue text; likely the same underlying issue." + }, + { + "left": "issue:43834", + "right": "issue:46222", + "accept": false, + "reason": "Different title/content; high superficial similarity is not enough to call these duplicates." + }, + { + "left": "issue:43834", + "right": "issue:46226", + "accept": false, + "reason": "Different title/content; this is not the same underlying issue." + }, + { + "left": "issue:44365", + "right": "issue:44370", + "accept": true, + "reason": "Identical placeholder title and highly similar body; these appear to be duplicates." + }, + { + "left": "issue:44365", + "right": "issue:46222", + "accept": false, + "reason": "Placeholder translation issue vs. unrelated Thai-text issue; not a duplicate pair." + }, + { + "left": "issue:44365", + "right": "issue:46226", + "accept": false, + "reason": "Placeholder translation issue vs. unrelated Thai-text issue; not a duplicate pair." + }, + { + "left": "issue:44370", + "right": "issue:46222", + "accept": false, + "reason": "Placeholder translation issue vs. unrelated Thai-text issue; not a duplicate pair." + }, + { + "left": "issue:44370", + "right": "issue:46226", + "accept": false, + "reason": "Placeholder translation issue vs. unrelated Thai-text issue; not a duplicate pair." + }, + { + "left": "issue:46222", + "right": "issue:46226", + "accept": true, + "reason": "Same Thai title and effectively identical content; clear duplicate pair." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary stays grounded in the packet: it distinguishes the language-specific translation requests from the repeated placeholder-template subgroup and the identical Thai pair, without overclaiming duplicates among the mixed-language items. The soft-edge verdicts are conservative, and the canonical/best-issue choice is reasonable for the duplicated placeholder subgroup." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 6, + "component_count": 21, + "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": "49b5b513bf6c9f6fdd3acab8fe7b33d88bb694b2e0d4b77778793abe879fab77", + "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 set of distinct model-support requests (different architectures and modalities), not true duplicates. The broadest umbrella issue is 42738, but it still does not subsume the rest as the same underlying request.", + "confidence": 0.91, + "canonical_issue_reason": "No single issue is a true duplicate anchor; 42738 is the closest umbrella because it is the most general model-family request, but the others are separate model additions.", + "canonical_pr_reason": null, + "best_issue_reason": "42738 is the best representative only in a loose thematic sense, since it is broader than the other titles; however, it should not be treated as a duplicate of the rest.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36817", + "right": "issue:41211", + "accept": false, + "reason": "Different model requests: EuroBert vs DEIMv2 are unrelated additions." + }, + { + "left": "issue:36817", + "right": "issue:43671", + "accept": false, + "reason": "BERT-style config support vs Qwen3-TTS audio support are different code paths and model types." + }, + { + "left": "issue:36817", + "right": "issue:36879", + "accept": false, + "reason": "EuroBert config support and RF-DETR model support are separate model additions." + }, + { + "left": "issue:36879", + "right": "issue:41211", + "accept": false, + "reason": "RF-DETR and DEIMv2 are different vision models, not the same change." + }, + { + "left": "issue:36879", + "right": "issue:43909", + "accept": false, + "reason": "RF-DETR vs LFM2.5 Audio are different modalities and implementations." + }, + { + "left": "issue:36879", + "right": "issue:44121", + "accept": false, + "reason": "RF-DETR detection support and weight-sparse transformer support are unrelated requests." + }, + { + "left": "issue:36879", + "right": "issue:43837", + "accept": false, + "reason": "RF-DETR and Qwen3-ASR are different model families and tasks." + }, + { + "left": "issue:36879", + "right": "issue:43671", + "accept": false, + "reason": "RF-DETR vision support and Qwen3-TTS audio support are not the same underlying change." + }, + { + "left": "issue:37015", + "right": "issue:43837", + "accept": false, + "reason": "NeoBERT and Qwen3-ASR target different models and code paths." + }, + { + "left": "issue:37015", + "right": "issue:44121", + "accept": false, + "reason": "NeoBERT support is unrelated to circuit-sparsity transformer support." + }, + { + "left": "issue:37015", + "right": "issue:43909", + "accept": false, + "reason": "NeoBERT and LFM2.5 Audio are distinct model requests." + }, + { + "left": "issue:36879", + "right": "issue:37015", + "accept": false, + "reason": "RF-DETR and NeoBERT are unrelated additions." + }, + { + "left": "issue:37015", + "right": "issue:42503", + "accept": false, + "reason": "NeoBERT and ModernVBERT are different model families." + }, + { + "left": "issue:41115", + "right": "issue:43837", + "accept": false, + "reason": "MiniCPM3 architecture support and Qwen3-ASR are different tasks." + }, + { + "left": "issue:41115", + "right": "issue:44121", + "accept": false, + "reason": "MiniCPM3 architecture support and weight-sparse transformer support are unrelated." + }, + { + "left": "issue:41115", + "right": "issue:42503", + "accept": false, + "reason": "MiniCPM3 and ModernVBERT are different model additions." + }, + { + "left": "issue:41115", + "right": "issue:45027", + "accept": false, + "reason": "MiniCPM3 architecture support and Voxtral TTS support are separate changes." + }, + { + "left": "issue:41211", + "right": "issue:43671", + "accept": false, + "reason": "DEIMv2 and Qwen3-TTS are different model types and tasks." + }, + { + "left": "issue:42503", + "right": "issue:43837", + "accept": false, + "reason": "ModernVBERT and Qwen3-ASR are unrelated model support requests." + }, + { + "left": "issue:42503", + "right": "issue:43671", + "accept": false, + "reason": "ModernVBERT and Qwen3-TTS are different model families." + }, + { + "left": "issue:42503", + "right": "issue:42738", + "accept": false, + "reason": "ModernVBERT and BERT-like RoPE support overlap only loosely, but are not the same concrete request." + }, + { + "left": "issue:42738", + "right": "issue:45027", + "accept": false, + "reason": "BERT-like RoPE support and Voxtral TTS support are not the same underlying bug or change." + }, + { + "left": "issue:43671", + "right": "issue:43837", + "accept": false, + "reason": "Qwen3-TTS and Qwen3-ASR are related by naming only; they are separate audio tasks and fixes." + }, + { + "left": "issue:43671", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-TTS and Voxtral TTS are different model integrations." + }, + { + "left": "issue:43671", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-TTS and DeepSeek-OCR2 are unrelated model requests." + }, + { + "left": "issue:43671", + "right": "issue:44121", + "accept": false, + "reason": "TTS support and weight-sparse transformer support are different code paths." + }, + { + "left": "issue:43671", + "right": "issue:43909", + "accept": false, + "reason": "Qwen3-TTS and LFM2.5 Audio are separate audio model integrations, not duplicates." + }, + { + "left": "issue:43837", + "right": "issue:43909", + "accept": false, + "reason": "Qwen3-ASR and LFM2.5 Audio are different audio models/tasks." + }, + { + "left": "issue:43837", + "right": "issue:45322", + "accept": false, + "reason": "Qwen3-ASR and EUPE are unrelated model requests." + }, + { + "left": "issue:43837", + "right": "issue:44121", + "accept": false, + "reason": "ASR support and weight-sparse transformer support are not the same concrete change." + }, + { + "left": "issue:43909", + "right": "issue:44205", + "accept": false, + "reason": "LFM2.5 Audio and SAM3-LiteText are different model families." + }, + { + "left": "issue:44121", + "right": "issue:45322", + "accept": false, + "reason": "Weight-sparse transformer support and EUPE support are unrelated." + }, + { + "left": "issue:43837", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-ASR and Voxtral TTS are different audio model support requests." + }, + { + "left": "issue:37015", + "right": "issue:45027", + "accept": false, + "reason": "NeoBERT and Voxtral TTS are unrelated model additions." + }, + { + "left": "issue:43837", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-ASR and DeepSeek-OCR2 are different model integrations." + }, + { + "left": "issue:45322", + "right": "issue:46133", + "accept": false, + "reason": "EUPE and TIPSv2 are distinct vision-language model requests, not one 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": 21, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-38740-2", + "nodes": [ + "issue:38740", + "issue:46443" + ], + "soft_pairs": [ + "issue:38740|issue:46443" + ], + "prepared_review_unit_hash": "cdb241b76377d1a93f968eab2d84fd884e0688596722d682adab66dc339a3d64", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1505, + "estimated_input_tokens": 377, + "estimated_eval_tokens": 1010 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1505, + "estimated_input_tokens": 377, + "estimated_eval_tokens": 1010 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are unrelated: one is a docs request to add a framework listing, the other is a technical vision RoPE refactor. They should not be clustered as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": "Issue #46443 is the stronger representative because it is the substantive technical problem in the cluster; #38740 is a narrow documentation request and does not describe the same kind of work.", + "canonical_pr_reason": null, + "best_issue_reason": "#46443 is the best issue overall because it describes a concrete codebase refactor with clear implementation scope, unlike the docs-only request in #38740.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:38740", + "right": "issue:46443", + "accept": false, + "reason": "Different domains and intents: one asks for documentation to add 'pruna' as an optimization framework, while the other concerns a Vision RoPE refactor. They do not appear to be the same underlying bug or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 8, + "component_count": 21, + "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": "0fa2efda354708653686c0d7369b3fc0e2090a19be97ae32e45f8291ede0d39f", + "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 appear unrelated: one is about block decoding behavior, the other is a model-specific support request for DeepSeek v3.2. No duplicate relationship is evident.", + "confidence": 0.96, + "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": "Titles and implied scopes differ substantially; there is no shared code path, explicit target, or description evidence that these are the same bug/request." + } + ] + }, + "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": 21, + "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": "b74e521f9aa4728e9792c3ac755c444c30eb3723a7d6fb1fff1906ec659ea0fd", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1584, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1584, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are thematically related (custom trainers / Trainer extensibility), but they describe different problems: one is about `remove_unused_column` behavior, the other about FLOPS estimation for custom trainers. I would not treat them as the same underlying bug.", + "confidence": 0.91, + "canonical_issue_reason": "Issue #43598 is the broader and more directly API-facing Trainer customizability request, so it is the better representative issue for this small cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "#43598 has the clearer, more general problem statement and is more likely to anchor discussion around Trainer customization than the narrower FLOPS-estimation concern.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43598", + "right": "issue:43600", + "accept": false, + "reason": "They share the general Trainer/custom-trainer theme, but the concrete problems differ: `remove_unused_column` customization vs. FLOPS counting logic for custom trainers. No indication they are the same bug or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 10, + "component_count": 21, + "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": "537345de5da0b91be24c3f059528467f3c1f6249e67479c4ae853409052c6e3a", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1612, + "estimated_input_tokens": 403, + "estimated_eval_tokens": 1062 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1612, + "estimated_input_tokens": 403, + "estimated_eval_tokens": 1062 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related but not duplicates: one requests broad Qwen3-Omni registration in AutoModel/AutoModelForConditionalGeneration, while the other asks for sequence-classification support for Qwen3.5. They target different model families and different auto-class heads.", + "confidence": 0.93, + "canonical_issue_reason": "Issue #43684 is the broader, more central request about registering Qwen3-Omni with core auto classes; it better represents the cluster\u2019s general model-registration theme.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #43684 is the best representative issue because it covers a wider API-registration gap rather than a single task-specific head.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43684", + "right": "issue:44405", + "accept": false, + "reason": "Different underlying requests: Qwen3-Omni auto-model registration vs Qwen3.5 sequence-classification support. Same broad area, but not the same bug/change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 11, + "component_count": 21, + "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": "751b68aadb4e8cdbebd650e3483155214fd1746936b8075bacefc52dc58bc75e", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4265, + "estimated_input_tokens": 1067, + "estimated_eval_tokens": 2390 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4265, + "estimated_input_tokens": 1067, + "estimated_eval_tokens": 2390 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "One clear outlier is issue 43835 ('Fett'). The three 'huggingface ' issues look like a near-duplicate cluster by title alone, but there isn\u2019t enough content to prove they report the exact same underlying problem.", + "confidence": 0.58, + "canonical_issue_reason": "issue:43845 is the best representative of the likely duplicate 'huggingface ' set because it is the earliest of the three matching-title reports and has the most context available among them.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43845 is the most suitable cluster anchor for the same reason: earliest matching-title report, while 43835 appears unrelated.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43835", + "right": "issue:43845", + "accept": false, + "reason": "Different titles ('Fett' vs 'huggingface '), no shared files/targets, and no body evidence tying them to the same issue." + }, + { + "left": "issue:43835", + "right": "issue:43846", + "accept": false, + "reason": "Different titles with no concrete overlap beyond generic similarity; insufficient evidence of the same underlying bug." + }, + { + "left": "issue:43835", + "right": "issue:43859", + "accept": false, + "reason": "Different titles and no shared targets/files or descriptive details linking them." + }, + { + "left": "issue:43845", + "right": "issue:43846", + "accept": false, + "reason": "Identical truncated titles are suggestive, but the bodies are too sparse to confirm the same underlying bug rather than separate reports." + }, + { + "left": "issue:43845", + "right": "issue:43859", + "accept": false, + "reason": "Same short title, but no concrete problem description, target, or code-path evidence to safely merge as duplicates." + }, + { + "left": "issue:43846", + "right": "issue:43859", + "accept": false, + "reason": "Same short title only; the available metadata is too thin to prove these are the same issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 12, + "component_count": 21, + "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": "ab715269cc3debbd73d875564acc153da8495549378cc777816b031c2ed859e0", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 9, + "serialized_chars": 5791, + "estimated_input_tokens": 1448, + "estimated_eval_tokens": 3152 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 9, + "serialized_chars": 5791, + "estimated_input_tokens": 1448, + "estimated_eval_tokens": 3152 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are not duplicates of a single real bug/change; they look like unrelated low-quality or spammy issues, with only one vaguely substantive documentation request.", + "confidence": 0.96, + "canonical_issue_reason": "No issue clearly represents a shared underlying problem. If a representative must be chosen, #44418 is the least spam-like and the only one with a meaningful topic.", + "canonical_pr_reason": null, + "best_issue_reason": "#44418 is the best representative only because it is the only non-spam, semantically interpretable issue title; it is still too vague to serve as a strong canonical for the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44371", + "right": "issue:44572", + "accept": false, + "reason": "Both are generic spam placeholders; no evidence they describe the same bug or change." + }, + { + "left": "issue:44371", + "right": "issue:44573", + "accept": false, + "reason": "Both are generic spam placeholders; no shared underlying issue is described." + }, + { + "left": "issue:44371", + "right": "issue:44418", + "accept": false, + "reason": "One is spam, the other is a vague documentation request; not the same underlying problem." + }, + { + "left": "issue:44371", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder versus a specific-looking title about Google LLC; unrelated." + }, + { + "left": "issue:44418", + "right": "issue:44572", + "accept": false, + "reason": "Vague documentation suggestion versus spam placeholder; no concrete overlap." + }, + { + "left": "issue:44418", + "right": "issue:44573", + "accept": false, + "reason": "Vague documentation suggestion versus spam placeholder; not duplicative." + }, + { + "left": "issue:44572", + "right": "issue:44573", + "accept": false, + "reason": "Both are spam placeholders, but there is no evidence they refer to the same actual issue." + }, + { + "left": "issue:44572", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder versus unrelated specific title; not the same bug/change." + }, + { + "left": "issue:44573", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder versus unrelated specific title; not the same bug/change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 13, + "component_count": 21, + "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": "21ec7d7a715eafe58d5e092d7593867b6b3cc4176b2d25d79021ed0b0ddfcfc1", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1644, + "estimated_input_tokens": 411, + "estimated_eval_tokens": 1078 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1644, + "estimated_input_tokens": 411, + "estimated_eval_tokens": 1078 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two closed issues are highly similar and likely duplicates: they have identical titles and very similar body sizes, but the evidence is only soft similarity, so I would not treat the duplicate relationship as definitive. #45145 is the better canonical representative because it has the slightly stronger linkage history.", + "confidence": 0.84, + "canonical_issue_reason": "#45145 is a reasonable canonical issue because it matches #45151 exactly in title and has one inbound reference, giving it slightly better linkage history.", + "canonical_pr_reason": null, + "best_issue_reason": "#45145 is the best issue representative for the cluster because it is textually identical to #45151 and has the stronger reference signal.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45145", + "right": "issue:45151", + "accept": false, + "reason": "The issues look highly similar, but the evidence is only soft similarity with no explicit cross-links, shared targets, or diffs, and the pair is marked non-deterministic. That supports likely duplication, not a confident merge-level duplicate verdict." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The packet supports strong similarity: the titles are identical, body lengths are close, and issue #45145 has one inbound reference while #45151 has none. However, the soft pair is marked deterministic_accept=false, so accepting the duplicate verdict is too strong on the available evidence. The summary also overstates certainty by saying they are definitively the same underlying issue report; a more conservative phrasing would be that they appear highly similar or likely duplicates. Canonical selection of #45145 is reasonable, but the soft-edge verdict should be more cautious." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 14, + "component_count": 21, + "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": 15, + "component_count": 21, + "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": "aec920a2d4fee0abe927c9c4cdd112ca03cbf77210bc256ad8435337a16f56d7", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1672, + "estimated_input_tokens": 418, + "estimated_eval_tokens": 1092 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1672, + "estimated_input_tokens": 418, + "estimated_eval_tokens": 1092 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two separate 4-bit inference bugs: one specific to Gemma4 audio modules with uint8 weights, and one in BitsAndBytesConfig module-exclusion handling. They are related by subsystem but not the same underlying issue.", + "confidence": 0.91, + "canonical_issue_reason": "Issue #45674 is the better cluster representative because it describes a broader, reusable configuration bug with a clear code-path in BitsAndBytesConfig, while #45672 is a model-specific failure mode.", + "canonical_pr_reason": null, + "best_issue_reason": "#45674 is the most generally useful issue to keep as the representative report: it has a clearer root-cause pattern, broader impact, and is less tied to one model family than #45672.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45672", + "right": "issue:45674", + "accept": false, + "reason": "Do not merge: the failures are different. #45672 is a torch.finfo TypeError triggered by Gemma4 audio modules using uint8 weights in NF4 inference; #45674 is an AssertionError caused by llm_int8_skip_modules clearing the default lm_head exclusion. Same broad area, different bug and fix path." + } + ] + }, + "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": 21, + "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": "9aa01dd876a025d5929f54c6ae227cdcdce7822630e0d0e6b37d94fff716dc52", + "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": "Two closed issues with unrelated person-name titles and no body/context; the soft similarity signal is not enough to treat them as duplicates.", + "confidence": 0.22, + "canonical_issue_reason": "Issue 45762 is only marginally earlier and otherwise just as sparse as 45763, so it is the least arbitrary representative among the issues.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45762 is the best issue candidate only by timestamp; neither issue provides enough context to indicate a stronger representative.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45762", + "right": "issue:45763", + "accept": false, + "reason": "The titles refer to different people and dates, with no shared body, target, or problem description; they do not look like the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 17, + "component_count": 21, + "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": "7cc8ac1b1f2ca7423097a338d38cdd18104a78854d877ac7802373a8fdd2c79f", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two closed issues appear unrelated: both are standalone title-only entries with different personal names and dates, and there is no body, linkage, or shared technical context to indicate a duplicate relationship.", + "confidence": 0.98, + "canonical_issue_reason": "issue:45764 is a reasonable canonical only by ordering; it has no substantive advantage over issue:45765 because both lack content and context.", + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a strong global canonical candidate since both are empty and non-technical, but issue:45764 is the slightly better placeholder due to being first.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45764", + "right": "issue:45765", + "accept": false, + "reason": "The titles are different names/dates with no shared technical subject, body, or links; this does not look like the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 18, + "component_count": 21, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45810-2", + "nodes": [ + "issue:45810", + "issue:46295" + ], + "soft_pairs": [ + "issue:45810|issue:46295" + ], + "prepared_review_unit_hash": "717d0d846340051464630a2075807245d2fe5de794e4bb90c19d672873c73dc0", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1541, + "estimated_input_tokens": 386, + "estimated_eval_tokens": 1028 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1541, + "estimated_input_tokens": 386, + "estimated_eval_tokens": 1028 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These issues are not duplicates: one requests adding a specific token classification model class, the other asks for FP8 training support. They target different features and code paths.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45810 is the more concrete, narrowly scoped feature request and is easier to route or implement independently, but it is not a duplicate of 46295.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45810", + "right": "issue:46295", + "accept": false, + "reason": "Different underlying requests: model-class support for Qwen3_5 token classification vs FP8 training support. They do not describe the same bug or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 19, + "component_count": 21, + "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": "276b799f8799f5c797b95a925522dc2c78020d9ecdebd9979fa921234cfc78a6", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5724, + "estimated_input_tokens": 1431, + "estimated_eval_tokens": 3118 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5724, + "estimated_input_tokens": 1431, + "estimated_eval_tokens": 3118 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The cluster contains two likely duplicate issue pairs: 45870/45871 and 45872/45873. Issue 45869 is only thematically related to the financial-data domain and appears distinct from both request pairs.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 45870 is a good canonical representative for the first likely duplicate pair because it is the earlier of the two identical-title reports and cleanly anchors that request.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45870 is the best single issue to keep as the cluster anchor because it is one of the clearest duplicate pair representatives and it predates 45871; the 45872/45873 pair looks like a separate request, and 45869 is unrelated.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45869", + "right": "issue:45870", + "accept": false, + "reason": "Same broad financial domain, but one is a dataset bundle pitch and the other is an example request; not the same underlying issue." + }, + { + "left": "issue:45869", + "right": "issue:45871", + "accept": false, + "reason": "The dataset bundle and the example request are only loosely related by topic, not duplicates." + }, + { + "left": "issue:45870", + "right": "issue:45871", + "accept": true, + "reason": "Same title and matching body length with high similarity; these appear to be duplicates for the same example request." + }, + { + "left": "issue:45870", + "right": "issue:45872", + "accept": false, + "reason": "They share a financial-time-series theme, but the request scope differs: example prediction versus a scraping-service ask." + }, + { + "left": "issue:45870", + "right": "issue:45873", + "accept": false, + "reason": "Different issue text and scope; not the same underlying request." + }, + { + "left": "issue:45871", + "right": "issue:45872", + "accept": false, + "reason": "The identical example request is still distinct from the scraping-service-focused request." + }, + { + "left": "issue:45871", + "right": "issue:45873", + "accept": false, + "reason": "Different request scope and wording; only loosely related by domain." + }, + { + "left": "issue:45872", + "right": "issue:45873", + "accept": true, + "reason": "Same title and matching body length with high similarity; these appear to be duplicates for the same request." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The clustering is mostly grounded, but the summary overstates the evidence by calling 45870/45871 and 45872/45873 \u201cexact duplicate\u201d pairs. The packet only shows identical titles, equal body lengths, and soft-similarity scores\u2014not the full bodies\u2014so the conservative wording should be \u201clikely duplicate\u201d or \u201cappears to be a duplicate.\u201d The soft-edge verdicts themselves are mostly conservative, but the duplicated/exact language is too strong for the available evidence." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 20, + "component_count": 21, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46439-2", + "nodes": [ + "issue:46439", + "issue:46441" + ], + "soft_pairs": [ + "issue:46439|issue:46441" + ], + "prepared_review_unit_hash": "4d02d81009ed5786f80a59f9c6b1b1b34f3bf654cd8f2bfbad4a4f0140fe45b5", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1635, + "estimated_input_tokens": 409, + "estimated_eval_tokens": 1074 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1635, + "estimated_input_tokens": 409, + "estimated_eval_tokens": 1074 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open issues both involve static cache and linear-attention in Qwen3.5, but they describe different failure modes: one is a shape corruption/RuntimeError during `early_initialization`, the other is a `KeyError: 'linear_attention'` when using `cache_implementation=\"static\"`.", + "confidence": 0.83, + "canonical_issue_reason": "Issue 46439 is the better canonical issue because it describes the more specific low-level failure in static-cache initialization and directly names the linear-attention shape corruption path.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46439 is the strongest representative of the cluster since it is more diagnostic and points to the likely underlying code-path bug, while 46441 reads more like a separate symptom or configuration lookup failure.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46439", + "right": "issue:46441", + "accept": false, + "reason": "Do not merge: both mention static cache and linear-attention, but one reports shape corruption/RuntimeError during early initialization and the other reports a KeyError for a missing `linear_attention` entry. Those are not clearly 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": 21, + "component_count": 21, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46551-2", + "nodes": [ + "issue:46551", + "issue:46552" + ], + "soft_pairs": [ + "issue:46551|issue:46552" + ], + "prepared_review_unit_hash": "c8fd468e2a28e027b0fe0e8e2d146114acf9ef010e3c43f3fc3a97421385706a", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1707, + "estimated_input_tokens": 427, + "estimated_eval_tokens": 1110 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1707, + "estimated_input_tokens": 427, + "estimated_eval_tokens": 1110 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are very likely duplicates: they have identical titles, matching body lengths, and a soft-similarity score of 1.0. The open issue looks like the better canonical representative.", + "confidence": 0.94, + "canonical_issue_reason": "issue:46552 is the better canonical issue because it is open and already has inbound references, and it appears to describe the same report as issue:46551 based on the identical title and matching body length.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:46552 is the strongest cluster representative since it is the active/open copy of the report and has received inbound references.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46551", + "right": "issue:46552", + "accept": true, + "reason": "The pair has the same title, the same body length, and a soft-similarity score of 1.0, which strongly indicates the same underlying issue even though the body text is not shown here." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "Reject: the conclusion is plausible, but the response overstates evidence from the packet. The packet shows identical titles, equal body lengths, and a soft-similarity score of 1.0, but it does not actually display the bodies, so claiming they have the \"same body\" or are a \"clear duplicate pair\" is stronger than the available evidence. The canonical/open-issue reasoning is also slightly overstated as \"already being referenced\" is only supported for issue:46552 via inbound references, not for the pair as a whole." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 1, + "component_count": 38, + "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-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:46267" + ], + "soft_pairs": [ + "pull_request:39895|pull_request:43424", + "pull_request:41855|pull_request:43424", + "pull_request:41855|pull_request:43823", + "pull_request:42668|pull_request:43823", + "pull_request:43251|pull_request:43823", + "pull_request:43251|pull_request:44827", + "pull_request:43424|pull_request:44827", + "pull_request:43424|pull_request:43823", + "pull_request:43823|pull_request:45655", + "pull_request:43823|pull_request:43995", + "pull_request:43995|pull_request:45213", + "pull_request:43995|pull_request:44827", + "pull_request:44827|pull_request:45170", + "pull_request:44827|pull_request:46267", + "pull_request:44827|pull_request:46048", + "pull_request:44827|pull_request:45655", + "pull_request:44827|pull_request:45213", + "pull_request:43823|pull_request:45213", + "pull_request:43251|pull_request:43254" + ], + "prepared_review_unit_hash": "432b24ac1562f3ea5032bfcd9365e339c12bac77b73c2f33aa020d0248f2abf6", + "component_budget": { + "node_count": 14, + "item_count": 14, + "soft_pair_count": 19, + "serialized_chars": 38088, + "estimated_input_tokens": 9522, + "estimated_eval_tokens": 19300 + }, + "budget": { + "node_count": 14, + "item_count": 14, + "soft_pair_count": 19, + "serialized_chars": 38088, + "estimated_input_tokens": 9522, + "estimated_eval_tokens": 19300 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Only one true duplicate-like PR pair appears here: two independent patches against `fixed_cross_entropy` in `loss_utils.py`. The rest are unrelated model additions, docs, refactors, or tests that only share broad subsystem overlap.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "`pull_request:43251` is the better canonical PR: it directly targets issue `43240`, makes the core `fixed_cross_entropy` API change in one place, and has more discussion/review context.", + "best_issue_reason": null, + "best_pr_reason": "`pull_request:43251` is the best representative PR for the duplicate cluster because it contains the same concrete `loss_utils.py` fix as `43254` but is more explicitly anchored to the tracked issue.", + "soft_edge_verdicts": [ + { + "left": "pull_request:39895", + "right": "pull_request:43424", + "accept": false, + "reason": "Different changes: VideoPrism model integration vs an executorch dynamic-shape export test. Same broad testing/modeling ecosystem, but not the same bug or patch." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43424", + "accept": false, + "reason": "Unrelated work: Mistral tokenizer API additions vs executorch export tests. No shared codepath or same fix." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43823", + "accept": false, + "reason": "Different model families and changes: tokenizer missing methods vs adding MobileLLM. Not the same underlying change." + }, + { + "left": "pull_request:42668", + "right": "pull_request:43823", + "accept": false, + "reason": "Both touch processor/model loading areas, but they fix different models and different issues; not mergeable as one PR." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43823", + "accept": false, + "reason": "Loss function kwargs fix vs adding MobileLLM model support. Completely different codepaths." + }, + { + "left": "pull_request:43251", + "right": "pull_request:44827", + "accept": false, + "reason": "`fixed_cross_entropy` kwargs support is unrelated to Mistral4 test fixes and auto-model mapping changes." + }, + { + "left": "pull_request:43424", + "right": "pull_request:44827", + "accept": false, + "reason": "Executorch dynamic-shape export test vs Mistral4 test/modeling adjustments; only both are tests, not the same bug." + }, + { + "left": "pull_request:43424", + "right": "pull_request:43823", + "accept": false, + "reason": "Exportability test vs adding a new model. No shared concrete defect." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45655", + "accept": false, + "reason": "New MobileLLM model addition is unrelated to `cls.config` resolution order in `modeling_utils.py`." + }, + { + "left": "pull_request:43823", + "right": "pull_request:43995", + "accept": false, + "reason": "MobileLLM addition vs Falcon output-interface refactor; different models and different implementation goals." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45213", + "accept": false, + "reason": "Falcon refactor is unrelated to the model-creation skill draft; no shared bug or codepath." + }, + { + "left": "pull_request:43995", + "right": "pull_request:44827", + "accept": false, + "reason": "Falcon output-collection refactor and Mistral4 tests are separate model-specific changes." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45170", + "accept": false, + "reason": "Mistral4 test fixes are unrelated to the `layrnorm` typo cleanup across CLIP-like models." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46267", + "accept": false, + "reason": "Mistral4 tests vs serve chat parsing updates; different subsystems and objectives." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46048", + "accept": false, + "reason": "Mistral4 tests/model mapping changes are unrelated to adding the TDT loss kernel." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45655", + "accept": false, + "reason": "Mistral4-specific test/model changes are not the same as the global config-resolution fix." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45213", + "accept": false, + "reason": "Mistral4 fixes and the new-model skill are entirely different changes." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45213", + "accept": false, + "reason": "New MobileLLM model support vs a model-creation skill scaffold; not the same underlying change." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43254", + "accept": true, + "reason": "Both patch `src/transformers/loss/loss_utils.py` to make `fixed_cross_entropy` forward supported `nn.functional.cross_entropy` kwargs for the same issue `43240`; these are the same underlying fix." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the only accepted pair (43251/43254) is well supported by the shared file and issue target. The other pairwise rejections stay conservative and do not overclaim shared bugs or mergeability." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 2, + "component_count": 38, + "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-40546-2", + "nodes": [ + "pull_request:40546", + "pull_request:42781" + ], + "soft_pairs": [ + "pull_request:40546|pull_request:42781" + ], + "prepared_review_unit_hash": "a482cc24fe193fb2a6bc92c86f960b36d3b6805e0a96d89e2ad7a455c5cabdc6", + "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": "Two VibeVoice PRs overlap heavily in filenames, but they are not duplicates: one is the base VibeVoice integration and the other adds the Realtime variant plus acoustic-tokenizer changes.", + "confidence": 0.84, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #40546 is the broader, original VibeVoice implementation and has substantial review/discussion activity; it is the cleaner representative of the core feature set.", + "best_issue_reason": null, + "best_pr_reason": "PR #40546 is the best overall representative because it is the stable base VibeVoice integration, while #42781 looks like a follow-on extension for the realtime path rather than the same change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40546", + "right": "pull_request:42781", + "accept": false, + "reason": "Shared VibeVoice files alone are not enough: #42781 adds a realtime variant and acoustic-tokenizer/gen-utils work, so it is a related extension, 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": 3, + "component_count": 38, + "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-40755-2", + "nodes": [ + "pull_request:40755", + "pull_request:41224" + ], + "soft_pairs": [ + "pull_request:40755|pull_request:41224" + ], + "prepared_review_unit_hash": "18ed3ce69d70cb67904c6a45ed7e498e1dec5dab64feb9c6141e5cce5ccf2f28", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4675, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4675, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two unrelated feature PRs: one adds TimesFM covariate forecasting support, the other adds DINOv3 image classification support. They do not fix the same code path or implement the same change.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 41224 is the tighter, more self-contained feature addition and is a better cluster representative than the broader TimesFM covariate work in 40755.", + "best_issue_reason": null, + "best_pr_reason": "PR 41224 is the best single representative here because it has a narrower scope and a clearer end-user feature surface.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40755", + "right": "pull_request:41224", + "accept": false, + "reason": "Different models and different functionality: TimesFM covariate forecasting vs DINOv3 image classification. No shared concrete bug or mergeable combined 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": 4, + "component_count": 38, + "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": 38, + "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-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": "3f3ec1e03e6fc7dbdcf052d0cfada52260922572859048c9c06b62f7f14104d3", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10434, + "estimated_input_tokens": 2609, + "estimated_eval_tokens": 5474 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10434, + "estimated_input_tokens": 2609, + "estimated_eval_tokens": 5474 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "All three pull requests are separate model-addition efforts with similar scaffolding, but they target different models and codepaths. They are not duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: DeepSeek OCR, Isaac, and Kimi2-6 each add distinct model packages, docs, and auto-registration entries rather than fixing the same underlying change.", + "best_issue_reason": null, + "best_pr_reason": "None of the PRs is a duplicate of the others; if forced to pick a cluster representative, 45630 is the most complete/newest submission, but it is still a separate model addition.", + "soft_edge_verdicts": [ + { + "left": "pull_request:41797", + "right": "pull_request:45186", + "accept": false, + "reason": "Different models (DeepSeek OCR vs Isaac) with different new model files and docs; similarity is only from shared Hugging Face model-integration boilerplate." + }, + { + "left": "pull_request:45186", + "right": "pull_request:45630", + "accept": false, + "reason": "Different models (Isaac vs Kimi2-6) and different implementation files; this is not the same bug or change, only the same kind of model onboarding work." + } + ] + }, + "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": 38, + "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-42975-2", + "nodes": [ + "pull_request:42975", + "pull_request:42976" + ], + "soft_pairs": [ + "pull_request:42975|pull_request:42976" + ], + "prepared_review_unit_hash": "23f5063957e877742c0a290b6df69cbec99118c3dee3f51e396810ab56683cd5", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5934, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5934, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related GitHub Actions maintenance PRs, but they change different actions and solve different upgrade paths: one targets Node 24 compatibility for several core actions, while the other bumps docker/build-push-action across Docker workflows. They are similar in theme, not the same underlying change.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 42975 is the better canonical representative because it has the clearest stated purpose (Node 24 compatibility) and updates foundational workflow actions used across many jobs.", + "best_issue_reason": null, + "best_pr_reason": "PR 42975 is the strongest standalone representative of the cluster: broader impact, explicit compatibility motivation, and a more general workflow-infrastructure upgrade scope.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42975", + "right": "pull_request:42976", + "accept": false, + "reason": "Both are workflow dependency bumps, but they update different GitHub Actions for different reasons (Node 24 compatibility vs. docker/build-push-action versioning). They do not look like 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": 7, + "component_count": 38, + "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-42978-6", + "nodes": [ + "pull_request:42978", + "pull_request:43448", + "pull_request:43451", + "pull_request:45490", + "pull_request:46562", + "pull_request:46641" + ], + "soft_pairs": [ + "pull_request:42978|pull_request:43448", + "pull_request:42978|pull_request:43451", + "pull_request:42978|pull_request:46641", + "pull_request:42978|pull_request:45490", + "pull_request:43448|pull_request:43451", + "pull_request:43448|pull_request:46641", + "pull_request:43448|pull_request:45490", + "pull_request:43451|pull_request:46641", + "pull_request:43451|pull_request:45490", + "pull_request:45490|pull_request:46562", + "pull_request:45490|pull_request:46641" + ], + "prepared_review_unit_hash": "ec6fd3b6c9ce0853c43095d444f0ff90e1941a4856c9279a33a04dc185299fee", + "component_budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 11, + "serialized_chars": 20403, + "estimated_input_tokens": 5101, + "estimated_eval_tokens": 10458 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 11, + "serialized_chars": 20403, + "estimated_input_tokens": 5101, + "estimated_eval_tokens": 10458 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are unrelated model-addition PRs plus one docs typo fix. The overlap is mostly generic new-model scaffolding, not the same underlying bug or change, so all soft edges should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No true canonical duplicate PR: these submissions cover different models or an unrelated docs typo fix. If a representative is needed, PR 43451 is the broadest and most complete standalone change.", + "best_issue_reason": null, + "best_pr_reason": "PR 43451 is the strongest standalone representative because it has the broadest implementation and test surface, but it is still not a duplicate of the others.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42978", + "right": "pull_request:43448", + "accept": false, + "reason": "Different model additions (ViT NEPA vs Molmo); shared auto/docs files are generic scaffolding, not the same change." + }, + { + "left": "pull_request:42978", + "right": "pull_request:43451", + "accept": false, + "reason": "Different model additions (ViT NEPA vs Molmo2); overlap is only in shared registration/docs boilerplate." + }, + { + "left": "pull_request:42978", + "right": "pull_request:46641", + "accept": false, + "reason": "A model-addition PR and a punctuation-only docs fix are unrelated." + }, + { + "left": "pull_request:42978", + "right": "pull_request:45490", + "accept": false, + "reason": "Different model additions (ViT NEPA vs CTSM); same subsystem templates do not indicate the same bug or feature." + }, + { + "left": "pull_request:43448", + "right": "pull_request:43451", + "accept": false, + "reason": "Molmo and Molmo2 are distinct model integrations with different code paths and files." + }, + { + "left": "pull_request:43448", + "right": "pull_request:46641", + "accept": false, + "reason": "A model integration and a docs comma fix are not the same underlying change." + }, + { + "left": "pull_request:43448", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo and CTSM are unrelated model additions; shared repo plumbing is too broad to merge as duplicates." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46641", + "accept": false, + "reason": "Molmo2 model integration and a docs comma fix are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo2 and CTSM are different model additions with no shared concrete bug or fix path." + }, + { + "left": "pull_request:45490", + "right": "pull_request:46562", + "accept": false, + "reason": "CTSM and Step 3.7 Flash are separate model additions touching different architectures and code paths." + }, + { + "left": "pull_request:45490", + "right": "pull_request:46641", + "accept": false, + "reason": "CTSM model integration and a docs comma fix are unrelated changes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 8, + "component_count": 38, + "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-43291-14", + "nodes": [ + "pull_request:43291", + "pull_request:43498", + "pull_request:43665", + "pull_request:44815", + "pull_request:44958", + "pull_request:45114", + "pull_request:45415", + "pull_request:45425", + "pull_request:46380", + "pull_request:46539", + "pull_request:46579", + "pull_request:46635", + "pull_request:46636", + "pull_request:46637" + ], + "soft_pairs": [ + "pull_request:43291|pull_request:46636", + "pull_request:43291|pull_request:43665", + "pull_request:43291|pull_request:44815", + "pull_request:43291|pull_request:46380", + "pull_request:43498|pull_request:46579", + "pull_request:43665|pull_request:46539", + "pull_request:43665|pull_request:44815", + "pull_request:43665|pull_request:46380", + "pull_request:43665|pull_request:46579", + "pull_request:44815|pull_request:46539", + "pull_request:44815|pull_request:46380", + "pull_request:44815|pull_request:46579", + "pull_request:44958|pull_request:46637", + "pull_request:44958|pull_request:46635", + "pull_request:45114|pull_request:46579", + "pull_request:45415|pull_request:45425", + "pull_request:45415|pull_request:46637", + "pull_request:45415|pull_request:46635", + "pull_request:46380|pull_request:46539", + "pull_request:46380|pull_request:46579", + "pull_request:46635|pull_request:46636", + "pull_request:46635|pull_request:46637", + "pull_request:46636|pull_request:46637" + ], + "prepared_review_unit_hash": "f2f2c957730567cebca0effc04cbf85c08733c57329f39edfab426d359d5676e", + "component_budget": { + "node_count": 14, + "item_count": 14, + "soft_pair_count": 23, + "serialized_chars": 38601, + "estimated_input_tokens": 9651, + "estimated_eval_tokens": 19558 + }, + "budget": { + "node_count": 14, + "item_count": 14, + "soft_pair_count": 23, + "serialized_chars": 38601, + "estimated_input_tokens": 9651, + "estimated_eval_tokens": 19558 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous PRs: unrelated test fixes, typing cleanup, typo-only maintenance, and a few integration changes. Only one pair looks like a likely duplicate/follow-up within the same typing effort; the rest are too broad or touch different concrete changes.", + "confidence": 0.63, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "PR 46579 is the best overall representative of the typo-cleanup subgroup: it is broad, multi-file, and overlaps the same kind of code hygiene changes seen in the nearby typo PRs.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43291", + "right": "pull_request:46636", + "accept": false, + "reason": "Different features: Whisper test/tokenizer work vs a typo fix in generation/configuration and tests." + }, + { + "left": "pull_request:43291", + "right": "pull_request:43665", + "accept": false, + "reason": "Both are test-related, but they target different models and different failures." + }, + { + "left": "pull_request:43291", + "right": "pull_request:44815", + "accept": false, + "reason": "Whisper test/tokenizer changes are unrelated to the FP8 dequantization fix." + }, + { + "left": "pull_request:43291", + "right": "pull_request:46380", + "accept": false, + "reason": "Unrelated changes: Whisper tests versus generation/configuration validation and model test updates." + }, + { + "left": "pull_request:43498", + "right": "pull_request:46579", + "accept": false, + "reason": "Backward-compatibility alias for tie_weights is unrelated to typo cleanup across code comments and filenames." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46539", + "accept": false, + "reason": "Different model test suites and different issues; only broad \"fix\" similarity." + }, + { + "left": "pull_request:43665", + "right": "pull_request:44815", + "accept": false, + "reason": "CLIP/ViT test fixes are unrelated to the FP8 dequantization change." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46380", + "accept": false, + "reason": "These touch different model tests and different code paths; not the same bug." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46579", + "accept": false, + "reason": "A broad typo cleanup does not match the specific CLIP/ViT test fixes." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46539", + "accept": false, + "reason": "FP8 dequantization logic and Phimoe integration-test CI fixes are different changes." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46380", + "accept": false, + "reason": "Different scopes: dequantization internals versus generation/Whisper test behavior." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46579", + "accept": false, + "reason": "A typo cleanup PR is not the same underlying change as the FP8 dequantization fix." + }, + { + "left": "pull_request:44958", + "right": "pull_request:46637", + "accept": false, + "reason": "Import-error fix for PILImageResampling is unrelated to a typo fix in generation/configuration or Mask2Former." + }, + { + "left": "pull_request:44958", + "right": "pull_request:46635", + "accept": false, + "reason": "Different problems: import handling versus a comment typo in generation/configuration." + }, + { + "left": "pull_request:45114", + "right": "pull_request:46579", + "accept": false, + "reason": "Doctest cleanup is unrelated to code typo cleanup in multiple source files." + }, + { + "left": "pull_request:45415", + "right": "pull_request:45425", + "accept": true, + "reason": "Both are part of the same typing/type-checking refactor around modeling_utils and related helpers; they address the same underlying typing effort and could plausibly be merged together." + }, + { + "left": "pull_request:45415", + "right": "pull_request:46637", + "accept": false, + "reason": "General type-checking work is unrelated to a specific typo fix in Mask2Former/configuration files." + }, + { + "left": "pull_request:45415", + "right": "pull_request:46635", + "accept": false, + "reason": "Broad typing changes do not match a single comment-typo cleanup." + }, + { + "left": "pull_request:46380", + "right": "pull_request:46539", + "accept": false, + "reason": "Generation/configuration validation fixes and Phimoe CI/integration fixes are different underlying changes." + }, + { + "left": "pull_request:46380", + "right": "pull_request:46579", + "accept": false, + "reason": "Generation validation fixes are not the same as broad typo cleanup across code files." + }, + { + "left": "pull_request:46635", + "right": "pull_request:46636", + "accept": false, + "reason": "They overlap on one configuration_utils typo, but each PR also fixes a different additional typo; not the same underlying change." + }, + { + "left": "pull_request:46635", + "right": "pull_request:46637", + "accept": false, + "reason": "Same overlap on one typo, but the extra fixes differ, so these are separate cleanup PRs." + }, + { + "left": "pull_request:46636", + "right": "pull_request:46637", + "accept": false, + "reason": "They only share a common config_utils typo; the rest of each PR is different." + } + ] + }, + "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": 38, + "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-43488-2", + "nodes": [ + "pull_request:43488", + "pull_request:43532" + ], + "soft_pairs": [ + "pull_request:43488|pull_request:43532" + ], + "prepared_review_unit_hash": "1ae19fb11f042748c1cb2fd46f3f7b87c61ba524a65bf8db4745fd6b1cd02ecb", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4562, + "estimated_input_tokens": 1141, + "estimated_eval_tokens": 2538 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4562, + "estimated_input_tokens": 1141, + "estimated_eval_tokens": 2538 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs are only superficially similar: both touch a vit test file, but one is a synthetic repo-bot check and the other is a workflow/testing-utils change. They are not the same underlying change and should not be clustered as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43532 is the more substantive and representative change in the cluster; PR 43488 is a deliberate '[don't merge]' placeholder/bot-format test with unrelated setup/version tweaks.", + "best_issue_reason": null, + "best_pr_reason": "PR 43532 is the best candidate to keep because it contains real workflow and test infrastructure changes, while PR 43488 is clearly a throwaway validation PR.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43488", + "right": "pull_request:43532", + "accept": false, + "reason": "Reject: the diffs are unrelated. 43488 changes setup/versioning and adds pytest-random-order for a bot-format test, while 43532 rewrites CI/workflow files and testing utilities. Shared test filename and soft similarity are not enough to indicate the same fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 10, + "component_count": 38, + "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-43612-2", + "nodes": [ + "pull_request:43612", + "pull_request:43613" + ], + "soft_pairs": [ + "pull_request:43612|pull_request:43613" + ], + "prepared_review_unit_hash": "4d51492e2e2a683e6f515c678345598f1a0ceaaff376328e3eb4b12e8902f978", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6082, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6082, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two separate PRs add different new pipelines with shared documentation/registration scaffolding, but they implement distinct features: Promptable Concept Segmentation vs Promptable Visual Segmentation. They are related in theme, not the same underlying change.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #43613 is the broader and more feature-complete implementation in this cluster, with additional model support and tests beyond the minimal pipeline/docs wiring.", + "best_issue_reason": null, + "best_pr_reason": "PR #43613 is the best representative because it covers the more extensive pipeline addition and touches more of the surrounding integration surface.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43612", + "right": "pull_request:43613", + "accept": false, + "reason": "Different features and code paths: concept segmentation vs visual segmentation. Shared docs/registry/pipeline scaffolding is not enough to treat them as one mergeable PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 11, + "component_count": 38, + "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-43651-2", + "nodes": [ + "pull_request:43651", + "pull_request:43663" + ], + "soft_pairs": [ + "pull_request:43651|pull_request:43663" + ], + "prepared_review_unit_hash": "c9fef565c6fff6b419d51198d17595a9e2789ddd1c527ca2d3adeb1a765374bb", + "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": "Two open PRs touch the same trainer.py file but target different Trainer extension points: one changes gradient-accumulation loss scaling, the other refactors signature-column filtering for PEFT/custom trainers. They are not the same underlying change.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43651 is the more concrete behavior change: it adjusts training_step loss scaling and adds a new hook for custom trainers. It is the better representative of a distinct trainer behavior fix/change.", + "best_issue_reason": null, + "best_pr_reason": "PR 43651 is the strongest standalone candidate because it makes an explicit training-step behavior change with a clear API hook; PR 43663 is a separate refactor of data-column filtering and should not be merged with it.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43651", + "right": "pull_request:43663", + "accept": false, + "reason": "Different code paths and goals: 43651 changes gradient-accumulation loss scaling in training_step, while 43663 extracts signature-column logic for column filtering. Shared file alone is not enough to treat them as the same change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 12, + "component_count": 38, + "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-43656-2", + "nodes": [ + "pull_request:43656", + "pull_request:43842" + ], + "soft_pairs": [ + "pull_request:43656|pull_request:43842" + ], + "prepared_review_unit_hash": "9148111fbf4200069b2a39eaf0288ab4b8cecd1f5b3e69262996d5049e2a0bd4", + "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 are about annotation-related `NameError` risk, but they fix different concrete problems: one is a broader `__future__`/typing cleanup, the other is a targeted `TypeAdapter` import/annotation fix for missing `pydantic`. Same surface area, not the same change.", + "confidence": 0.82, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43842 is the most focused representative: it directly addresses the documented `TypeAdapter` `NameError` when `pydantic` is not installed, with a tight, concrete fix.", + "best_issue_reason": null, + "best_pr_reason": "PR 43842 is the best PR candidate overall because it targets the specific runtime failure mode described by its issue link and is less broad than the other PR.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43656", + "right": "pull_request:43842", + "accept": false, + "reason": "Reject as duplicates: they overlap in the same file and general symptom, but 43656 is a broader postponed-annotations/type-hint cleanup while 43842 specifically fixes `TypeAdapter` resolution when `pydantic` is absent. Different underlying bugs, not one mergeable change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 13, + "component_count": 38, + "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": 38, + "review_unit_index": 1, + "review_unit_count": 2, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "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:44019|pull_request:44085", + "pull_request:44086|pull_request:44722", + "pull_request:44018|pull_request:44019", + "pull_request:44019|pull_request:44071", + "pull_request:44068|pull_request:44116", + "pull_request:44018|pull_request:44116", + "pull_request:44068|pull_request:44722", + "pull_request:44071|pull_request:44722", + "pull_request:44018|pull_request:44085", + "pull_request:44018|pull_request:44086", + "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:44018|pull_request:44071", + "pull_request:44059|pull_request:44722", + "pull_request:44017|pull_request:44161", + "pull_request:44017|pull_request:44073", + "pull_request:44017|pull_request:44722", + "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:44073|pull_request:44161", + "pull_request:44056|pull_request:44101", + "pull_request:44059|pull_request:44071", + "pull_request:44018|pull_request:44059", + "pull_request:44059|pull_request:44068", + "pull_request:44101|pull_request:44161", + "pull_request:43997|pull_request:44017", + "pull_request:43997|pull_request:44056", + "pull_request:44071|pull_request:44076", + "pull_request:44074|pull_request:44101", + "pull_request:44056|pull_request:44161", + "pull_request:43997|pull_request:44018", + "pull_request:43997|pull_request:44161", + "pull_request:44074|pull_request:44161", + "pull_request:44018|pull_request:44161", + "pull_request: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:44010|pull_request:44018", + "pull_request:44071|pull_request:44129", + "pull_request:43996|pull_request:44161", + "pull_request:43996|pull_request:44017", + "pull_request:44010|pull_request:44044", + "pull_request:44044|pull_request:44074", + "pull_request:44066|pull_request:44722", + "pull_request:44085|pull_request:44722", + "pull_request:44085|pull_request:44129", + "pull_request:43996|pull_request:44007", + "pull_request:43996|pull_request:44129", + "pull_request:44044|pull_request:44129", + "pull_request:44044|pull_request:44154", + "pull_request:44024|pull_request:44076" + ], + "prepared_review_unit_hash": "2b0fe4aad508d14b2b7a26028a367ecad6d7e16cdb402fee64ecbf8edafca8c0", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 77, + "serialized_chars": 84509, + "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": "Mostly a broad cluster of independent model-by-model output-tracing refactors. The only true overlaps here are the GPT-J/CodeGen variants and the GPT-Neo variants; the rest share only the same umbrella theme or issue target.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "44066 is the clearest representative of the cluster\u2019s core change: a multi-file GPT-J/CodeGen migration to standardized output tracing with an explicit issue link and direct code-path changes.", + "best_issue_reason": null, + "best_pr_reason": "44066 is the best single anchor because it covers the core GPT-J/CodeGen code path, has the cleanest explicit linkage to the tracking issue, and is broad enough to subsume the narrower GPT-J-only variants.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43996", + "right": "pull_request:44085", + "accept": false, + "reason": "Different model families; same umbrella issue is not enough to treat them as duplicates." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44044", + "accept": false, + "reason": "Different models and code paths; both are output-tracing refactors, but not the same change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44085", + "accept": true, + "reason": "Same GPT-J output-tracing refactor in the same code path; one is a narrower overlapping variant of the other." + }, + { + "left": "pull_request:44007", + "right": "pull_request:44072", + "accept": false, + "reason": "Different architectures; only the generic output-tracing theme matches." + }, + { + "left": "pull_request:44072", + "right": "pull_request:44722", + "accept": false, + "reason": "Different model families and code paths; not the same underlying bug/change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44072", + "accept": false, + "reason": "GPT-J/CodeGen versus EfficientNet; too different to be one duplicate change." + }, + { + "left": "pull_request:44013", + "right": "pull_request:44044", + "accept": false, + "reason": "MobileNetV2 and DeBERTa are unrelated code paths despite the shared refactor pattern." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44086", + "accept": false, + "reason": "Different models and different output-tracing implementations; not mergeable as one fix." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-J/CodeGen versus MPT; same theme, different concrete change." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44068", + "accept": true, + "reason": "Both are GPT-Neo output-tracing refactors over the same model code path and look like overlapping variants of the same change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-J/CodeGen versus GPT-Neo; related theme but different concrete model changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44066", + "accept": false, + "reason": "Different models and code paths; not a duplicate despite similar refactor language." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44722", + "accept": false, + "reason": "ResNet versus GPT-J/CodeGen; only the abstract refactor style is shared." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44085", + "accept": false, + "reason": "Different model families and unrelated code paths." + }, + { + "left": "pull_request:44086", + "right": "pull_request:44722", + "accept": false, + "reason": "MGP-STR and GPT-J are different concrete changes; shared issue target would not make them duplicates." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44019", + "accept": false, + "reason": "GPT-Neo and ResNet are unrelated code paths." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44071", + "accept": false, + "reason": "ResNet and MPT are unrelated models; not the same underlying bug/change." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo versus Flaubert/XLM; different model-specific refactors." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert are unrelated concrete changes." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-Neo and GPT-J/CodeGen are different model-specific migrations." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44722", + "accept": false, + "reason": "MPT and GPT-J/CodeGen are different code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo versus GPT-J; same refactor family, different underlying changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo and MGP-STR are unrelated concrete fixes." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44086", + "accept": false, + "reason": "Different models and different output-capture implementations." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT2 and MGP-STR are unrelated; same theme alone is insufficient." + }, + { + "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 unrelated model-specific changes." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44161", + "accept": false, + "reason": "SegFormer and LongT5 are different underlying refactors." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44073", + "accept": false, + "reason": "SegFormer and VisualBert are unrelated concrete fixes." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44722", + "accept": false, + "reason": "Different model families; not the same bug/change." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo and MPT are different model-specific changes." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44101", + "accept": false, + "reason": "VisualBert and Flaubert/XLM are distinct code paths." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44101", + "accept": false, + "reason": "MPNet and Flaubert/XLM are unrelated implementations." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT2 and MPT are different concrete changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44059", + "accept": false, + "reason": "GPT-Neo and GPT2 are different model-specific refactors." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT2 versus GPT-Neo; related patterns, but not the same underlying fix." + }, + { + "left": "pull_request:44101", + "right": "pull_request:44161", + "accept": false, + "reason": "Flaubert/XLM and LongT5 are different model code paths." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44017", + "accept": false, + "reason": "RegNet and SegFormer are unrelated model-specific changes." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44056", + "accept": false, + "reason": "RegNet and MPNet are different concrete fixes." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44076", + "accept": false, + "reason": "MPT and ImageGPT are different model families." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44101", + "accept": false, + "reason": "TextNet and Flaubert/XLM are unrelated code paths." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44161", + "accept": false, + "reason": "MPNet and LongT5 are different underlying changes." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44018", + "accept": false, + "reason": "RegNet and GPT-Neo are unrelated model-specific refactors." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44161", + "accept": false, + "reason": "RegNet and LongT5 are different code paths." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44161", + "accept": false, + "reason": "TextNet and LongT5 are unrelated concrete changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44161", + "accept": false, + "reason": "GPT-Neo and LongT5 are different model-specific changes." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44071", + "accept": false, + "reason": "SqueezeBert and MPT are unrelated implementations." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44073", + "accept": false, + "reason": "SqueezeBert and VisualBert are different code paths." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44722", + "accept": false, + "reason": "SqueezeBert and GPT-J/CodeGen are unrelated concrete fixes." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44086", + "accept": false, + "reason": "MPT and MGP-STR are different models and changes." + }, + { + "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 different code paths." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44129", + "accept": false, + "reason": "MPT and SpeechT5 are unrelated concrete changes." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44161", + "accept": false, + "reason": "FNet/CVT and LongT5 are different model families." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44017", + "accept": false, + "reason": "FNet/CVT and SegFormer are unrelated changes." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44044", + "accept": false, + "reason": "SqueezeBert and DeBERTa are different model-specific fixes." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44074", + "accept": false, + "reason": "DeBERTa and TextNet are unrelated concrete changes." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J/CodeGen output-tracing migration in the same files; these could be merged as one PR." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44722", + "accept": true, + "reason": "Both target the same GPT-J output-tracing code path and overlap enough to be one combined PR." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-J and SpeechT5 are unrelated model-specific changes." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44007", + "accept": false, + "reason": "Different model families; only the shared output-tracing theme matches." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44129", + "accept": false, + "reason": "FNet/CVT and SpeechT5 are unrelated concrete fixes." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44129", + "accept": false, + "reason": "DeBERTa and SpeechT5 are different code paths." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44154", + "accept": false, + "reason": "DeBERTa and VITS are unrelated model-specific changes." + }, + { + "left": "pull_request:44024", + "right": "pull_request:44076", + "accept": false, + "reason": "FocalNet and ImageGPT are different models; same refactor pattern is not enough." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44074", + "accept": false, + "reason": "ResNet and TextNet are unrelated changes." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44059", + "accept": false, + "reason": "MPNet and GPT2 are different model-specific refactors." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44071", + "accept": false, + "reason": "MPNet and MPT are unrelated code paths." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and the soft-edge calls are mostly conservative. The accepted overlaps are limited to pairs that share the same model code path and look mergeable as one bugfix, while the rejects generally stay within different model families. I don\u2019t see a clear overstatement of evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 14, + "component_count": 38, + "review_unit_index": 2, + "review_unit_count": 2, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "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": "949a5af4497d5a4035dad3d4e6808c49f6f022f46afba5e3280da2fa74a36a3a", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 77, + "serialized_chars": 84509, + "estimated_input_tokens": 21128, + "estimated_eval_tokens": 42512 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 5, + "serialized_chars": 13067, + "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 thematically similar migrations to standardized output tracing/capture utilities, but they target different model implementations and code paths, so they are not duplicates.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No issue is present; as a representative PR, 44129 is the clearest concrete change because it has an explicit issue target and a substantial SpeechT5-specific tracing refactor.", + "best_issue_reason": null, + "best_pr_reason": "44129 is the strongest single representative of the cluster: it has an explicit linked issue, a focused model-specific code change, and clear output-tracing refactoring scope.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44073", + "right": "pull_request:44074", + "accept": false, + "reason": "Both are output-tracing refactors, but one is for VisualBert and the other for TextNet; they touch different model code paths and are not the same underlying bug/change." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44073", + "accept": false, + "reason": "MPNet and VisualBert are separate model implementations with distinct encoder logic; this is the same refactor pattern, not a shared duplicate fix." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44074", + "accept": false, + "reason": "These apply the same decorator/capture pattern to different models (MPNet vs TextNet), so they are not mergeable as one duplicate PR." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44010", + "accept": false, + "reason": "RegNet and SqueezeBert are unrelated model paths; despite similar migration wording, they are separate PRs for different implementations." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44129", + "accept": false, + "reason": "SqueezeBert and SpeechT5 changes are on different architectures and code paths; they share a refactor theme but not the same concrete bug or fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 15, + "component_count": 38, + "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-43998-13", + "nodes": [ + "pull_request:43998", + "pull_request:43999", + "pull_request:44000", + "pull_request:44001", + "pull_request:44002", + "pull_request:44003", + "pull_request:44004", + "pull_request:44025", + "pull_request:44026", + "pull_request:44027", + "pull_request:44028", + "pull_request:44029", + "pull_request:44030" + ], + "soft_pairs": [ + "pull_request:43998|pull_request:44002", + "pull_request:43998|pull_request:44000", + "pull_request:43998|pull_request:44001", + "pull_request:43998|pull_request:44003", + "pull_request:43998|pull_request:44029", + "pull_request:43999|pull_request:44001", + "pull_request:43999|pull_request:44002", + "pull_request:43999|pull_request:44025", + "pull_request:43999|pull_request:44026", + "pull_request:43999|pull_request:44027", + "pull_request:44000|pull_request:44001", + "pull_request:44000|pull_request:44002", + "pull_request:44000|pull_request:44003", + "pull_request:44000|pull_request:44025", + "pull_request:44001|pull_request:44002", + "pull_request:44001|pull_request:44003", + "pull_request:44001|pull_request:44025", + "pull_request:44002|pull_request:44029", + "pull_request:44002|pull_request:44003", + "pull_request:44003|pull_request:44028", + "pull_request:44003|pull_request:44004", + "pull_request:43998|pull_request:44004", + "pull_request:44001|pull_request:44004", + "pull_request:44002|pull_request:44004", + "pull_request:44000|pull_request:44004", + "pull_request:44025|pull_request:44026", + "pull_request:44025|pull_request:44027", + "pull_request:44025|pull_request:44028", + "pull_request:44025|pull_request:44029", + "pull_request:44025|pull_request:44030", + "pull_request:44026|pull_request:44029", + "pull_request:44026|pull_request:44027", + "pull_request:44026|pull_request:44028", + "pull_request:44026|pull_request:44030", + "pull_request:44027|pull_request:44028", + "pull_request:44027|pull_request:44029", + "pull_request:44027|pull_request:44030", + "pull_request:44028|pull_request:44029", + "pull_request:44028|pull_request:44030", + "pull_request:44029|pull_request:44030" + ], + "prepared_review_unit_hash": "ee3d603beabe81dfc19e67dcbf53ee5e2da29c732dccf9ef55c95dc1029b9b8a", + "component_budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40170, + "estimated_input_tokens": 10043, + "estimated_eval_tokens": 20342 + }, + "budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40170, + "estimated_input_tokens": 10043, + "estimated_eval_tokens": 20342 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are separate open PRs that all follow the same output-tracing refactor pattern, but they target different model families and code paths; I do not see true duplicates among them.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "44003 is the broadest single example of the shared theme, updating both FalconMamba and Mamba with the new output-capturing plumbing.", + "best_issue_reason": null, + "best_pr_reason": "44003 is the strongest representative of the cluster: it has the largest, most substantive refactor and spans two closely related implementations in the same change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43998", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model families and forward paths; only the refactor style is shared." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44000", + "accept": false, + "reason": "Unrelated model code paths; both are output-tracing cleanup, not the same bug or change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44001", + "accept": false, + "reason": "Separate architectures with no shared concrete code-path issue." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44003", + "accept": false, + "reason": "Different subsystems and implementations; same refactor theme only." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44029", + "accept": false, + "reason": "Timm backbone vs RWKV are unrelated code paths." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44001", + "accept": false, + "reason": "MobileNetV1 and UnivNet are different model families; no duplicate bug/change." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model architectures and forward logic." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44025", + "accept": false, + "reason": "MobileNetV1 and DepthAnything are unrelated implementations." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44026", + "accept": false, + "reason": "No shared concrete code path; only similar output-tracing refactors." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44027", + "accept": false, + "reason": "Speech/vision encoder-decoder code is distinct from MobileNetV1." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44001", + "accept": false, + "reason": "Vision-text dual encoder and UnivNet are unrelated model paths." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model families; same pattern, not same underlying change." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44003", + "accept": false, + "reason": "Vision-text dual encoder vs Mamba/RWKV stateful models are unrelated." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44025", + "accept": false, + "reason": "Different architectures and code paths." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44002", + "accept": false, + "reason": "UnivNet and UperNet do not share the same concrete bug." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44003", + "accept": false, + "reason": "Different implementations; no plausible single merged PR." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44025", + "accept": false, + "reason": "UnivNet vs DepthAnything are unrelated model families." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44029", + "accept": false, + "reason": "UperNet and RWKV are different code paths and fixes." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44003", + "accept": false, + "reason": "No shared underlying bug; only the refactor pattern matches." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44028", + "accept": false, + "reason": "Mamba/RWKV-style stateful models vs SuperPoint are unrelated." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44004", + "accept": false, + "reason": "Stateful Mamba-family code and CodeGen are different concrete paths." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44004", + "accept": false, + "reason": "Timm backbone and CodeGen are unrelated; same refactor theme only." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "UnivNet and CodeGen do not fix the same issue." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44004", + "accept": false, + "reason": "UperNet and CodeGen are distinct code paths." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44004", + "accept": false, + "reason": "Vision-text dual encoder and CodeGen are unrelated implementations." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44026", + "accept": false, + "reason": "DepthAnything and VisionEncoderDecoder are different model families and fixes." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44027", + "accept": false, + "reason": "DepthAnything and SpeechEncoderDecoder are unrelated code paths." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44028", + "accept": false, + "reason": "Depth estimation and SuperPoint are not the same underlying change." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44029", + "accept": false, + "reason": "DepthAnything and RWKV are unrelated model implementations." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44030", + "accept": false, + "reason": "DepthAnything and DPR are different concrete code paths." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44029", + "accept": false, + "reason": "VisionEncoderDecoder and RWKV are unrelated." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44027", + "accept": false, + "reason": "Different encoder-decoder wrappers, but not the same bug/change." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44028", + "accept": false, + "reason": "VisionEncoderDecoder and SuperPoint are unrelated code paths." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44030", + "accept": false, + "reason": "Different model families and concrete fixes." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44028", + "accept": false, + "reason": "SpeechEncoderDecoder and SuperPoint do not share the same underlying issue." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44029", + "accept": false, + "reason": "SpeechEncoderDecoder and RWKV are unrelated." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44030", + "accept": false, + "reason": "SpeechEncoderDecoder and DPR are distinct code paths." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44029", + "accept": false, + "reason": "SuperPoint and RWKV are different model families." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44030", + "accept": false, + "reason": "SuperPoint and DPR are unrelated fixes." + }, + { + "left": "pull_request:44029", + "right": "pull_request:44030", + "accept": false, + "reason": "RWKV and DPR are different concrete implementations, not one duplicate bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 16, + "component_count": 38, + "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-44662-2", + "nodes": [ + "pull_request:44662", + "pull_request:45401" + ], + "soft_pairs": [ + "pull_request:44662|pull_request:45401" + ], + "prepared_review_unit_hash": "0e1cbc32ac848bff16902445ac21f647b7e48180d072b7d51a7301e5b8756b86", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6552, + "estimated_input_tokens": 1638, + "estimated_eval_tokens": 3532 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6552, + "estimated_input_tokens": 1638, + "estimated_eval_tokens": 3532 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two unrelated model-addition PRs were loosely similar because they both register new model packages and touch the auto/docs plumbing, but they implement different models (PenguinVL vs Voxtral TTS) and different code paths.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45401 is the better cluster representative because it has an explicit linked issue target and a self-contained, narrowly scoped Voxtral TTS implementation.", + "best_issue_reason": null, + "best_pr_reason": "PR #45401 is the strongest standalone PR in this cluster: it is tied to a concrete issue and has a focused implementation for one model family.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44662", + "right": "pull_request:45401", + "accept": false, + "reason": "They are both model implementation PRs, but for different models and different functionality; shared registration/docs files alone are not enough to treat them as the same change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 17, + "component_count": 38, + "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-44979-2", + "nodes": [ + "pull_request:44979", + "pull_request:45363" + ], + "soft_pairs": [ + "pull_request:44979|pull_request:45363" + ], + "prepared_review_unit_hash": "e416cebb38dec7350e92c0b66eb5c9f8f8b1b74448628d8af773ad3c51911d7e", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4407, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4407, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two pull requests are related in the broad area of module/kernel fusion, but they are not the same underlying change: #44979 adds a generic Module Fusion API, while #45363 wires in n-to-1 kernel fusion via KernelConfig and adds an example. They should not be merged as duplicates.", + "confidence": 0.77, + "canonical_issue_reason": null, + "canonical_pr_reason": "#44979 is the cleaner representative because it introduces the core generic Module Fusion API in a focused way; #45363 is a more specific KernelConfig-based implementation built on top of that idea.", + "best_issue_reason": null, + "best_pr_reason": "#44979 is the best cluster anchor since it captures the foundational abstraction rather than a narrower application of fusion to one kernel path.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44979", + "right": "pull_request:45363", + "accept": false, + "reason": "Related fusion work, but different concrete changes and code paths: generic API vs KernelConfig-driven n-to-1 kernel fusion. Not the same PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 18, + "component_count": 38, + "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-45086-2", + "nodes": [ + "pull_request:45086", + "pull_request:45317" + ], + "soft_pairs": [ + "pull_request:45086|pull_request:45317" + ], + "prepared_review_unit_hash": "c72e40f4420ff2fc82e093c44b1b14d75703fecc17f64225ae998f921cd9677f", + "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 PRs implement the same fix for the Mistral regex AttributeError in the same code path; 45317 is a superset of 45086 because it includes the same patch plus tests.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45317 is the stronger canonical choice: it contains the same core fix as 45086 and adds test coverage, making it the more complete and review-ready version of the duplicate pair.", + "best_issue_reason": null, + "best_pr_reason": "PR 45317 is best overall because it preserves the exact bug fix while also adding a test file, so it is the most complete representation of the change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45086", + "right": "pull_request:45317", + "accept": true, + "reason": "Accept: both PRs make the same concrete fix in src/transformers/tokenization_utils_tokenizers.py for _patch_mistral_regex, with identical changed lines; 45317 simply adds tests, so they are duplicate implementations of the same underlying bug fix and could 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": 19, + "component_count": 38, + "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-45115-4", + "nodes": [ + "pull_request:45115", + "pull_request:45569", + "pull_request:46332", + "pull_request:46565" + ], + "soft_pairs": [ + "pull_request:45115|pull_request:45569", + "pull_request:45569|pull_request:46565", + "pull_request:46332|pull_request:46565" + ], + "prepared_review_unit_hash": "a068da8e06f08bdf257729741b98fd4246c90698e2ac37cc7c532ad8dcc13bab", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 3, + "serialized_chars": 13472, + "estimated_input_tokens": 3368, + "estimated_eval_tokens": 6992 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 3, + "serialized_chars": 13472, + "estimated_input_tokens": 3368, + "estimated_eval_tokens": 6992 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is heterogeneous: one NemotronH refactor, one NemotronH dense/sparse split, and two separate ASR streaming additions. The shared filenames are mostly auto/docs plumbing, not evidence of the same underlying change, so I would not merge any pair as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR fits the whole cluster cleanly; if one must be chosen, PR 46332 is the most self-contained and feature-complete standalone addition in the set.", + "best_issue_reason": null, + "best_pr_reason": "PR 46332 is the strongest representative because it implements a full Nemotron ASR streaming stack with core losses, model wiring, and documentation, whereas the others are either a refactor or a different Nemotron variant.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45115", + "right": "pull_request:45569", + "accept": false, + "reason": "Same NemotronH area, but 45115 is a refactor toward GraniteMoeHybrid while 45569 adds/docks separate NemotronH dense and sparse variants; different concrete changes, not one fix." + }, + { + "left": "pull_request:45569", + "right": "pull_request:46565", + "accept": false, + "reason": "Overlap is limited to shared auto/docs/model registry files; one is NemotronH architecture splitting, the other is Nemotron ASR streaming for a different model family." + }, + { + "left": "pull_request:46332", + "right": "pull_request:46565", + "accept": false, + "reason": "Both are ASR-related, but they target different model versions and code paths (Nemotron ASR vs Nemotron 3.5 ASR). They are related features, not the same underlying bug/change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 20, + "component_count": 38, + "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-45332-2", + "nodes": [ + "pull_request:45332", + "pull_request:45333" + ], + "soft_pairs": [ + "pull_request:45332|pull_request:45333" + ], + "prepared_review_unit_hash": "8d4b3690f1f568bd00cceba162e8f1a98bcf0a75262dc3fff6e1870b2fbdfb12", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5466, + "estimated_input_tokens": 1367, + "estimated_eval_tokens": 2990 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5466, + "estimated_input_tokens": 1367, + "estimated_eval_tokens": 2990 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are related pieces of a larger heterogeneous-model feature, but they are not duplicates: one adds per-layer config plumbing, the other adds model/cache support that consumes it.", + "confidence": 0.87, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45332 is the broader umbrella change for heterogeneous model support, covering the actual model/cache plumbing; PR 45333 is a narrower config-foundation piece.", + "best_issue_reason": null, + "best_pr_reason": "PR 45332 is the most complete and central implementation of the feature cluster, while 45333 is supporting configuration groundwork.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45332", + "right": "pull_request:45333", + "accept": false, + "reason": "Reject: these are complementary feature increments rather than the same underlying bug/change. 45333 introduces per-layer heterogeneous config support, while 45332 adds heterogeneous model/cache handling that depends on that config plumbing; they would not plausibly collapse into one duplicate PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 21, + "component_count": 38, + "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-45513-2", + "nodes": [ + "pull_request:45513", + "pull_request:46084" + ], + "soft_pairs": [ + "pull_request:45513|pull_request:46084" + ], + "prepared_review_unit_hash": "5e556153e9ce5aa9cd357b868a893b6fa0c2bf3482e91519f9068f6911aa56f7", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4948, + "estimated_input_tokens": 1237, + "estimated_eval_tokens": 2730 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4948, + "estimated_input_tokens": 1237, + "estimated_eval_tokens": 2730 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both items are pull requests about cached forward behavior with seq_len > 1, but they fix different model families and different code paths: one is Mamba2 cache handling, the other is Qwen/OLMo hybrid linear-attention cached forward. They are related by theme, not the same underlying change.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #46084 is the better canonical PR because it has an explicit issue target (#46032) and a narrowly defined fix for Mamba2 use_cache behavior when seq_len > 1.", + "best_issue_reason": null, + "best_pr_reason": "PR #46084 is the strongest representative of the cluster due to its explicit tracking issue and focused scope; PR #45513 spans several model files and a different attention path, so it is less suitable as the canonical representative.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45513", + "right": "pull_request:46084", + "accept": false, + "reason": "Reject: they address different concrete bugs in different model implementations. #45513 fixes multi-token cached forward for Qwen3.5/OLMo hybrid linear attention, while #46084 fixes Mamba2 cache state handling when seq_len > 1. Same broad theme, not the same underlying code-path problem." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 22, + "component_count": 38, + "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-45548-2", + "nodes": [ + "pull_request:45548", + "pull_request:45662" + ], + "soft_pairs": [ + "pull_request:45548|pull_request:45662" + ], + "prepared_review_unit_hash": "014b009d7b0021f03e3b3542e18bfa5a08c846e700180f6da15b1cb7946368ca", + "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": "The two pull requests are related by expert-parallel loading, but they fix different bugs in different code paths: PR 45548 addresses EP + DeepSpeed ZeRO-3 loading/serialization of distributed_config, while PR 45662 addresses EP + FSDP2 experts being overwritten by rank-0 broadcast. Similar subsystem, not the same underlying change.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45662 is the stronger representative: it targets a concrete EP/FSDP2 loading corruption bug with a clear failure mode and a multi-file fix, rather than the narrower serialization/loading compatibility change in 45548.", + "best_issue_reason": null, + "best_pr_reason": "PR 45662 is the best standalone candidate because it fixes the more explicit data-loss bug in the loading path and has the clearest end-user impact.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45548", + "right": "pull_request:45662", + "accept": false, + "reason": "Do not merge: 45548 fixes EP + DeepSpeed ZeRO-3 loading by preserving distributed_config and adding has_ep, while 45662 fixes EP + FSDP2 expert tensors being overwritten after rank-0 broadcast. Same broad subsystem, different concrete bugs/code paths." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 23, + "component_count": 38, + "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-45639-2", + "nodes": [ + "pull_request:45639", + "pull_request:45645" + ], + "soft_pairs": [ + "pull_request:45639|pull_request:45645" + ], + "prepared_review_unit_hash": "60c8c673abc7a83cf89730e9539d94aba6c53f2a010061dcf854ec4a17a1fa82", + "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": "These two PRs address the same xdist/captured_info collision problem; #45645 is the broader superset, while #45639 is a narrower implementation of the same underlying fix.", + "confidence": 0.87, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45645 is the best canonical PR because it includes the xdist-safe output-file change plus the CI/workflow update needed to consume multiple captured_info files, making it the more complete fix.", + "best_issue_reason": null, + "best_pr_reason": "PR #45645 is the best representative of the cluster since it covers the full bug fix and downstream log handling, whereas #45639 only covers the library-side portion.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45639", + "right": "pull_request:45645", + "accept": true, + "reason": "Same underlying xdist collision/captured_info bug; #45645 appears to subsume #45639 and both could 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": 24, + "component_count": 38, + "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-45670-3", + "nodes": [ + "pull_request:45670", + "pull_request:46320", + "pull_request:46401" + ], + "soft_pairs": [ + "pull_request:45670|pull_request:46320", + "pull_request:45670|pull_request:46401", + "pull_request:46320|pull_request:46401" + ], + "prepared_review_unit_hash": "f21fd8b7726af5cc6888bb33ddec04735983789ff1621d60d476d3cb5649e2d9", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7525, + "estimated_input_tokens": 1882, + "estimated_eval_tokens": 4020 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7525, + "estimated_input_tokens": 1882, + "estimated_eval_tokens": 4020 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are three separate PRs: one adds AutoModel registration for GLM-ASR, and two add attn_implementation dispatch support for different model families (RecurrentGemma vs BLIP-2/InstructBLIP). They are thematically similar but not duplicates of the same underlying change.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No issue is present; for a representative PR, pull request 46320 is the cleanest single-model example of the attn_implementation dispatch work with direct code, docs, and tests tied to RecurrentGemma.", + "best_issue_reason": null, + "best_pr_reason": "Pull request 46320 is the best representative of the cluster\u2019s main technical theme because it is a focused, self-contained implementation for one model family. Pull request 46401 is broader but still a separate change set, and 45670 is unrelated to attn_implementation dispatch.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45670", + "right": "pull_request:46320", + "accept": false, + "reason": "Different changes and code paths: GLM-ASR AutoModel registration versus RecurrentGemma attention-dispatch support. They are not the same bug or mergeable fix." + }, + { + "left": "pull_request:45670", + "right": "pull_request:46401", + "accept": false, + "reason": "Different changes and code paths: GLM-ASR AutoModel registration versus BLIP-2/InstructBLIP attention-dispatch support. Shared ML model context is too broad to be a duplicate." + }, + { + "left": "pull_request:46320", + "right": "pull_request:46401", + "accept": false, + "reason": "Both add attn_implementation support, but for different model families and distinct model-specific implementations (RecurrentGemma vs BLIP-2/InstructBLIP/InstructBLIPVideo). They are related features, 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": 25, + "component_count": 38, + "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-45720-2", + "nodes": [ + "pull_request:45720", + "pull_request:46259" + ], + "soft_pairs": [ + "pull_request:45720|pull_request:46259" + ], + "prepared_review_unit_hash": "8580486856f359a38a5d90cc1c746c5b23f32a9f05e07deb8b07a622008a1052", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7248, + "estimated_input_tokens": 1812, + "estimated_eval_tokens": 3880 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7248, + "estimated_input_tokens": 1812, + "estimated_eval_tokens": 3880 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two unrelated linter-driven transformer PRs: one changes model forward return-tuple handling (TRF003), the other changes weight initialization logic (TRF018). They overlap only because they touch many of the same model files, not the same bug or change.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45720 is the better representative: it has a narrower, concrete scope centered on forward signatures/return behavior, making it the cleaner canonical PR for this cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR #45720 is the strongest standalone PR: it targets one specific code-path behavior change across model forwards, whereas #46259 is a separate initialization-rule sweep.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45720", + "right": "pull_request:46259", + "accept": false, + "reason": "Different underlying changes: #45720 updates forward output/return_dict handling, while #46259 refactors model weight initialization. Shared model files are incidental and not evidence of 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": 26, + "component_count": 38, + "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": 38, + "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-45932-2", + "nodes": [ + "pull_request:45932", + "pull_request:46100" + ], + "soft_pairs": [ + "pull_request:45932|pull_request:46100" + ], + "prepared_review_unit_hash": "69d93e5b37ade1d528471bdb787f99716fbaec84a5b7659a1a48501303b00a5e", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5082, + "estimated_input_tokens": 1271, + "estimated_eval_tokens": 2798 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5082, + "estimated_input_tokens": 1271, + "estimated_eval_tokens": 2798 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are unrelated: 45932 is a broad experimental/test-ci plus model/config change set, while 46100 is a focused ViT test migration. They do not fix the same underlying problem and should not be merged as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: the PRs target different concerns and are not duplicates.", + "best_issue_reason": null, + "best_pr_reason": "46100 is the more self-contained and coherent PR if a representative must be chosen, but it is not a duplicate of 45932.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45932", + "right": "pull_request:46100", + "accept": false, + "reason": "Different scopes and code paths: 45932 touches CI workflow and multiple model/config files, while 46100 adds/migrates ViT tests. They are not the same change and would not plausibly be merged into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 28, + "component_count": 38, + "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-46073-3", + "nodes": [ + "pull_request:46073", + "pull_request:46154", + "pull_request:46196" + ], + "soft_pairs": [ + "pull_request:46073|pull_request:46154", + "pull_request:46073|pull_request:46196", + "pull_request:46154|pull_request:46196" + ], + "prepared_review_unit_hash": "d2db3300a43d3c5de966ee52c486a8cdd13b6668f4eaf095e31a418290d56315", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 8400, + "estimated_input_tokens": 2100, + "estimated_eval_tokens": 4456 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 8400, + "estimated_input_tokens": 2100, + "estimated_eval_tokens": 4456 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are thematically similar (all add SDPA/attn_implementation dispatch), but they target different model families and distinct code paths, so they are not duplicates and should not be merged together.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR: CTRL, T5, and wav2vec2/audio stack changes are separate model-specific implementations of the same feature pattern, not the same underlying change.", + "best_issue_reason": null, + "best_pr_reason": "PR 46196 is the broadest and most central representative of the feature family, but it still addresses a different set of models/code paths than the other two.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46073", + "right": "pull_request:46154", + "accept": false, + "reason": "Same feature class, but different model families (CTRL vs T5) and different implementation paths; not the same concrete bug/change." + }, + { + "left": "pull_request:46073", + "right": "pull_request:46196", + "accept": false, + "reason": "Both relate to attention dispatch, but CTRL and wav2vec2/audio models are separate code paths with different fixes; not mergeable as one PR." + }, + { + "left": "pull_request:46154", + "right": "pull_request:46196", + "accept": false, + "reason": "T5 and wav2vec2/audio dispatch support are distinct model-specific changes, not the same underlying PR or bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 29, + "component_count": 38, + "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-46126-2", + "nodes": [ + "pull_request:46126", + "pull_request:46269" + ], + "soft_pairs": [ + "pull_request:46126|pull_request:46269" + ], + "prepared_review_unit_hash": "d01dbaf1fbc37cdd90c49db3048b079ffcd4b93f19f658f8fd729b28250277e7", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5571, + "estimated_input_tokens": 1393, + "estimated_eval_tokens": 3042 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5571, + "estimated_input_tokens": 1393, + "estimated_eval_tokens": 3042 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two PRs are not duplicates: #46126 is a focused EP/Trainer integration change around tensor-parallel routing, while #46269 is a broad distributed-branch base PR touching many unrelated distributed components. The shared filenames are too coarse to imply the same concrete fix.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "#46126 is the more canonical PR because it has a narrow, concrete scope and a clear code-path purpose (EP routing / Trainer integration) rather than serving as a broad branch-base umbrella.", + "best_issue_reason": null, + "best_pr_reason": "#46126 is the best representative of the cluster because it is the only focused change set; #46269 is too large and mixed to stand in as a single duplicate target.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46126", + "right": "pull_request:46269", + "accept": false, + "reason": "Reject: they do not appear to fix the same underlying bug or implement the same change. #46126 is a targeted EP routing integration patch; #46269 is a wide distributed branch-base PR with many unrelated edits, so they could not plausibly be merged as one duplicate PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 30, + "component_count": 38, + "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-46138-9", + "nodes": [ + "pull_request:46138", + "pull_request:46432", + "pull_request:46501", + "pull_request:46502", + "pull_request:46503", + "pull_request:46504", + "pull_request:46554", + "pull_request:46558", + "pull_request:46563" + ], + "soft_pairs": [ + "pull_request:46138|pull_request:46554", + "pull_request:46138|pull_request:46558", + "pull_request:46138|pull_request:46563", + "pull_request:46138|pull_request:46504", + "pull_request:46138|pull_request:46432", + "pull_request:46432|pull_request:46503", + "pull_request:46432|pull_request:46501", + "pull_request:46432|pull_request:46502", + "pull_request:46432|pull_request:46554", + "pull_request:46432|pull_request:46558", + "pull_request:46501|pull_request:46503", + "pull_request:46501|pull_request:46502", + "pull_request:46501|pull_request:46504", + "pull_request:46501|pull_request:46554", + "pull_request:46502|pull_request:46503", + "pull_request:46502|pull_request:46504", + "pull_request:46502|pull_request:46554", + "pull_request:46503|pull_request:46504", + "pull_request:46503|pull_request:46554", + "pull_request:46504|pull_request:46554", + "pull_request:46504|pull_request:46558", + "pull_request:46554|pull_request:46558", + "pull_request:46554|pull_request:46563", + "pull_request:46558|pull_request:46563", + "pull_request:46504|pull_request:46563", + "pull_request:46432|pull_request:46563" + ], + "prepared_review_unit_hash": "d74b452130e367ccfaba5ff926981ea17029b1105d6452c567f71d551115dbb3", + "component_budget": { + "node_count": 9, + "item_count": 9, + "soft_pair_count": 26, + "serialized_chars": 23361, + "estimated_input_tokens": 5841, + "estimated_eval_tokens": 11938 + }, + "budget": { + "node_count": 9, + "item_count": 9, + "soft_pair_count": 26, + "serialized_chars": 23361, + "estimated_input_tokens": 5841, + "estimated_eval_tokens": 11938 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are all small, unrelated workflow-maintenance PRs. The only high-similarity group is the three pr-ci-security-gate edits, but they change different aspects of the workflow and are not the same underlying fix/change.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR: the cluster spans distinct workflow updates across different files, and the similar pr-ci-security-gate PRs still address different changes rather than one duplicate fix.", + "best_issue_reason": null, + "best_pr_reason": "PR 46554 is the strongest representative because it makes the most concrete behavior change in pr-ci-security-gate.yml, but it is still not a duplicate of the other PRs.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46138", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflows and different concerns; only a generic workflow-maintenance similarity." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46558", + "accept": false, + "reason": "Unrelated workflow updates; no shared concrete bug or change." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46563", + "accept": false, + "reason": "Different files and different workflow behavior; not the same change." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46504", + "accept": false, + "reason": "Separate workflow edits in different CI paths; not duplicative." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46432", + "accept": false, + "reason": "Both are workflow tweaks, but they fix different scripts and different failure modes." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46503", + "accept": false, + "reason": "Different workflow files and unrelated trigger logic; not the same underlying change." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46501", + "accept": false, + "reason": "Different workflows; similarity is only at the 'chore update' level." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46502", + "accept": false, + "reason": "Distinct workflow maintenance in separate files; not mergeable as one duplicate fix." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46554", + "accept": false, + "reason": "The SSH runner hardening and pr-ci-security-gate change address unrelated problems." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46558", + "accept": false, + "reason": "Different workflow behavior and different file targets; not a duplicate." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46503", + "accept": false, + "reason": "Different workflows with different triggers; no same underlying bug/change." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46502", + "accept": false, + "reason": "Separate workflow files and independent changes; similarity is superficial." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46504", + "accept": false, + "reason": "Both are CI workflow updates, but they modify different jobs and logic." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46554", + "accept": false, + "reason": "Unrelated workflow changes; no plausible single-PR duplicate." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46503", + "accept": false, + "reason": "Different workflow targets and behavior; not the same change." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46504", + "accept": false, + "reason": "Distinct workflow maintenance changes in separate files." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflow concerns; not the same underlying fix." + }, + { + "left": "pull_request:46503", + "right": "pull_request:46504", + "accept": false, + "reason": "Both are workflow edits, but they change unrelated CI behavior." + }, + { + "left": "pull_request:46503", + "right": "pull_request:46554", + "accept": false, + "reason": "Benchmark workflow trigger logic is unrelated to pr-ci-security-gate hardening." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflows and different bug classes; not duplicates." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46558", + "accept": false, + "reason": "Build-tag parsing and security-gate permissions are unrelated changes." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46558", + "accept": false, + "reason": "Same file, but one adds checkout/security checks while the other changes permissions and approval gating; different changes." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46563", + "accept": false, + "reason": "Same workflow file, but one changes security checks and the other only adds documentation/comments around a reusable workflow." + }, + { + "left": "pull_request:46558", + "right": "pull_request:46563", + "accept": false, + "reason": "Same file, but permissions/approval gating is a different change from the reusable-workflow note." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46563", + "accept": false, + "reason": "Unrelated workflow files and behavior; no duplicate fix." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46563", + "accept": false, + "reason": "Different workflow changes with no shared concrete code-path problem." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 31, + "component_count": 38, + "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-46152-2", + "nodes": [ + "pull_request:46152", + "pull_request:46207" + ], + "soft_pairs": [ + "pull_request:46152|pull_request:46207" + ], + "prepared_review_unit_hash": "0ca1c05dac8f4e66b345b5131f8e416e507634e9db00d4f39d912833809e8d05", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4276, + "estimated_input_tokens": 1069, + "estimated_eval_tokens": 2394 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4276, + "estimated_input_tokens": 1069, + "estimated_eval_tokens": 2394 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two PRs are related by a broad theme of avoiding direct source-file opens, but they fix different code paths in different modules and should not be deduplicated.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46207 is the better canonical-style representative only in the loose sense that it is a narrower, test-backed fix for a specific bug in model attention-implementation detection; PR 46152 is a separate import-utils change.", + "best_issue_reason": null, + "best_pr_reason": "PR 46207 is the strongest PR in the cluster for triage because it addresses a concrete runtime path and includes tests, while PR 46152 is an independent source-loading refactor in a different subsystem.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46152", + "right": "pull_request:46207", + "accept": false, + "reason": "Different underlying fixes: 46152 changes import_utils source resolution to use importlib.resources/path traversal, while 46207 changes modeling_utils to use inspect.getsource for _can_set_attn_implementation. Same theme, but not the same concrete bug or mergeable PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 32, + "component_count": 38, + "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-46173-3", + "nodes": [ + "pull_request:46173", + "pull_request:46436", + "pull_request:46474" + ], + "soft_pairs": [ + "pull_request:46436|pull_request:46474", + "pull_request:46173|pull_request:46474" + ], + "prepared_review_unit_hash": "4d9e827d7fbad904c8449b6164c8045e4f091c3d7c4d8af7d34a354dedd81dac", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 5921, + "estimated_input_tokens": 1481, + "estimated_eval_tokens": 3218 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 5921, + "estimated_input_tokens": 1481, + "estimated_eval_tokens": 3218 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two PRs are near-identical duplicates fixing the StaticLayer.get_seq_length return type annotation in cache_utils.py for issue #45987. The third PR is a different static-cache generation config change and should not be clustered with them.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #46173 is the best canonical representative: it is the earlier, direct fix for #45987, with the same cache_utils.py type-hint change as the duplicate PR #46474.", + "best_issue_reason": null, + "best_pr_reason": "PR #46173 is the strongest cluster anchor because it exactly matches the duplicated code change, explicitly targets the shared issue, and is the earlier of the two identical fixes.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46436", + "right": "pull_request:46474", + "accept": false, + "reason": "Different bugs and code paths: #46436 changes generation cache sizing logic in generation/utils.py, while #46474 only adjusts StaticLayer.get_seq_length type hints in cache_utils.py." + }, + { + "left": "pull_request:46173", + "right": "pull_request:46474", + "accept": true, + "reason": "Same underlying fix: both make the StaticLayer.get_seq_length return type accept int | torch.Tensor in cache_utils.py and target the same issue #45987." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The analysis is grounded in the packet and stays conservative. It correctly treats PRs 46173 and 46474 as duplicate fixes to the same cache_utils.py type-hint issue (#45987), and it correctly keeps PR 46436 separate as a different generation/static-cache config change. No overstatement found." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 33, + "component_count": 38, + "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-46236-2", + "nodes": [ + "pull_request:46236", + "pull_request:46621" + ], + "soft_pairs": [ + "pull_request:46236|pull_request:46621" + ], + "prepared_review_unit_hash": "ba2822deff9e0ea526f12ab078d4852d8982ddf4ef115bae0df93d56d600e225", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5922, + "estimated_input_tokens": 1481, + "estimated_eval_tokens": 3218 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5922, + "estimated_input_tokens": 1481, + "estimated_eval_tokens": 3218 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two pull requests are unrelated documentation localization changes: one adds a Bengali docs translation set, while the other adds a Chinese translation for a single kernels page. They do not fix the same code-path or represent the same change request.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46236 is the better cluster representative because it is the broader, explicitly tracked documentation translation effort and has an issue link, while PR 46621 is a small single-page translation.", + "best_issue_reason": null, + "best_pr_reason": "PR 46236 is the strongest standalone PR in this cluster because it covers a larger, coherent docs localization bundle rather than a narrow page addition.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46236", + "right": "pull_request:46621", + "accept": false, + "reason": "Different translation scopes and languages: Bengali docs set vs. Chinese kernels page. They are not the same underlying change and would not plausibly be merged into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 34, + "component_count": 38, + "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-46310-2", + "nodes": [ + "pull_request:46310", + "pull_request:46546" + ], + "soft_pairs": [ + "pull_request:46310|pull_request:46546" + ], + "prepared_review_unit_hash": "382ec8a17ee59ae74aef12656fb0ecafddee66c10c9d101f8a6f8755992c8538", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4850, + "estimated_input_tokens": 1213, + "estimated_eval_tokens": 2682 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4850, + "estimated_input_tokens": 1213, + "estimated_eval_tokens": 2682 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs fix the same right-window off-by-one boundary bug in Moonshine streaming masking; PR 46546 is the broader, more complete version and includes tests plus the same fix in Gemma4.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46546 is the more complete representative: it applies the same boundary fix to Moonshine streaming, extends it to Gemma4, and adds tests.", + "best_issue_reason": null, + "best_pr_reason": "PR 46546 is the strongest standalone fix because it covers the same bug with broader model coverage and test support.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46310", + "right": "pull_request:46546", + "accept": true, + "reason": "Both PRs correct the same off-by-one right-window boundary in the same mask helper; 46310 is essentially a narrower precursor/subset of the fix in 46546." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 35, + "component_count": 38, + "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-46376-2", + "nodes": [ + "pull_request:46376", + "pull_request:46417" + ], + "soft_pairs": [ + "pull_request:46376|pull_request:46417" + ], + "prepared_review_unit_hash": "6176c53a7ab7cd2765c9ac86379204ba190bc984e588d713452f5f1d4963f642", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5939, + "estimated_input_tokens": 1485, + "estimated_eval_tokens": 3226 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5939, + "estimated_input_tokens": 1485, + "estimated_eval_tokens": 3226 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These pull requests are not duplicates: one is a draft infrastructure change for continuous batching multimodal support, while the other is a finished model integration for HunYuan VL with docs and tests. The overlap is only superficial.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "If one PR must represent the cluster, 46417 is the more complete and mature standalone change: it is non-draft, adds a full model integration, and includes docs/tests.", + "best_issue_reason": null, + "best_pr_reason": "46417 is the best overall artifact here because it is a self-contained feature addition with substantial implementation and validation, whereas 46376 is still a draft infra PR.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46376", + "right": "pull_request:46417", + "accept": false, + "reason": "Different underlying changes: continuous batching multimodal infrastructure vs. HunYuan VL model integration. They do not fix the same code-path problem and would not plausibly merge as one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 36, + "component_count": 38, + "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-46603-2", + "nodes": [ + "pull_request:46603", + "pull_request:46604" + ], + "soft_pairs": [ + "pull_request:46603|pull_request:46604" + ], + "prepared_review_unit_hash": "0470ab71e2609ac1e43c7c7b182609f4589eafeb27808cadf25189132e813523", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5495, + "estimated_input_tokens": 1374, + "estimated_eval_tokens": 3004 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5495, + "estimated_input_tokens": 1374, + "estimated_eval_tokens": 3004 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two Mistral-related PRs overlap in files and subsystem, but they appear to be separate changes: one refactors/relocates MistralConverter, the other adds tekken tokenizer detection and save/conversion utilities. Not a duplicate pair.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46604 is the broader, more substantive Mistral tokenizer change and is the better representative of the cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR 46604 best captures the main functional direction here (tekken tokenizer detection/conversion/save support), whereas 46603 is mainly a code move/refactor.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46603", + "right": "pull_request:46604", + "accept": false, + "reason": "Same subsystem and shared files, but different underlying changes: 46603 moves MistralConverter into an integrations package, while 46604 adds tekken tokenizer detection, conversion, and save utilities. They are related but not the same concrete change and would not reasonably collapse 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": 37, + "component_count": 38, + "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-46615-2", + "nodes": [ + "pull_request:46615", + "pull_request:46639" + ], + "soft_pairs": [ + "pull_request:46615|pull_request:46639" + ], + "prepared_review_unit_hash": "623cb87b3fb1e339140884de2d0b2fa181f4d4c8dc3edc0cba163a9b3726af9e", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4336, + "estimated_input_tokens": 1084, + "estimated_eval_tokens": 2424 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4336, + "estimated_input_tokens": 1084, + "estimated_eval_tokens": 2424 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs target the same generation/beam-search cache reordering bug in transformers and point to the same issue, with 46615 being the more complete fix and 46639 a narrower variant.", + "confidence": 0.92, + "canonical_issue_reason": "Issue 46612 is the shared explicit target for both PRs and appears to be the underlying bug report driving the work.", + "canonical_pr_reason": "PR 46615 is the stronger canonical PR because it implements the broader cache-reorder fix with helper logic and tests, covering non-standard cache types directly.", + "best_issue_reason": "Issue 46612 is the only issue in the cluster and is the central reference for the bug being fixed.", + "best_pr_reason": "PR 46615 is the best representative because it provides the more complete, test-backed implementation for the beam-search cache reorder problem.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46615", + "right": "pull_request:46639", + "accept": true, + "reason": "Both PRs fix the same underlying beam-search cache reordering bug for non-standard cache types in generation/utils.py, share the same issue target 46612, and could plausibly be merged into one change set." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: both PRs explicitly target issue 46612, both modify generation/utils.py to address beam-search cache reordering, and 46615 is reasonably characterized as the more complete, test-backed fix. The soft-edge accept is conservative enough, and the pair looks mergeable as alternate implementations of the same bugfix." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 38, + "component_count": 38, + "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-46616-2", + "nodes": [ + "pull_request:46616", + "pull_request:46638" + ], + "soft_pairs": [ + "pull_request:46616|pull_request:46638" + ], + "prepared_review_unit_hash": "488cccfc5f03b1e05ab9a04e4ff9c4566630de2406b6b1e39904bc4d88b8d779", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4653, + "estimated_input_tokens": 1164, + "estimated_eval_tokens": 2584 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4653, + "estimated_input_tokens": 1164, + "estimated_eval_tokens": 2584 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs target the same Wav2Vec2 phonemizer language-state bug and the same issue, with nearly identical test coverage; 46638 is the better representative because it updates the cached language after backend reinitialization, matching the intended restore flow.", + "confidence": 0.87, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46638 is the cleaner canonical PR: it fixes the same stale phonemizer_lang state bug but places the assignment after init_backend, which better matches restoring language state when switching back.", + "best_issue_reason": null, + "best_pr_reason": "PR 46638 most clearly and plausibly fixes the concrete code path in phonemize by restoring self.phonemizer_lang after backend switching; 46616 is the same bug theme but uses a less convincing assignment order.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46616", + "right": "pull_request:46638", + "accept": true, + "reason": "Both PRs address the same stale phonemizer_lang bug in the same method, touch the same files, and are alternative implementations of one fix." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: both PRs explicitly target issue 46614, touch the same tokenizer and test file, and differ only in the ordering of the stale `phonemizer_lang` update relative to `init_backend()`. The soft-edge accept is conservative because these look like alternative implementations of the same bugfix and can reasonably be treated as one PR cluster." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + } + ] +} diff --git a/snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/manifest.json b/snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..5ad6ad58c65226453ffdf738a87c596184bc20c4 --- /dev/null +++ b/snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/manifest.json @@ -0,0 +1,14 @@ +{ + "analysis_id": "hybrid-model-20260615t000032z", + "artifacts": { + "hybrid": "snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "model": null, + "published_at": "2026-06-15T00:08:24Z", + "repo": "huggingface/transformers", + "schema_version": 1, + "snapshot_id": "20260615T000032Z", + "variant": "hybrid" +} diff --git a/snapshots/20260615T000032Z/manifest.json b/snapshots/20260615T000032Z/manifest.json index 98d3d6d7950fa3e15271746ddc4e386379ee7959..526c153c1b7d1bb7ceafb4f4b5576b34c0345b2f 100644 --- a/snapshots/20260615T000032Z/manifest.json +++ b/snapshots/20260615T000032Z/manifest.json @@ -32,6 +32,24 @@ "timeline_events": 94 }, "extracted_at": "2026-06-15T00:00:32Z", + "published_analysis": { + "canonical_analysis_id": "hybrid-model-20260615t000032z", + "runs": { + "hybrid-model-20260615t000032z": { + "analysis_id": "hybrid-model-20260615t000032z", + "artifacts": { + "hybrid": "snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "manifest_path": "snapshots/20260615T000032Z/analysis-runs/hybrid-model-20260615t000032z/manifest.json", + "model": null, + "published_at": "2026-06-15T00:08:24Z", + "variant": "hybrid" + } + }, + "schema_version": 1 + }, "repo": "huggingface/transformers", "snapshot_id": "20260615T000032Z", "watermark": {