diff --git a/analysis/current/analysis-report-hybrid.json b/analysis/current/analysis-report-hybrid.json index 886eae36670f9bc7a423b6b301ff1618a2deb775..99211081567f6240017b8ed8528b7014e1fd37bb 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": "20260510T060019Z", - "generated_at": "2026-05-10T06:03:17Z", + "snapshot_id": "20260510T120026Z", + "generated_at": "2026-05-10T12:08:00Z", "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 95c28fd896e0d289d7282bd98dee50b12a8147e2..b5c6150e84ea28faaa865d6bfa33f9d950698400 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": "20260510T060019Z", - "generated_at": "2026-05-10T06:03:17Z", + "snapshot_id": "20260510T120026Z", + "generated_at": "2026-05-10T12:08:00Z", "model": "gpt-5.4-mini?service_tier=flex", "reviews": [ { @@ -13,8 +13,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-13244-3", "nodes": [ @@ -26,12 +26,12 @@ "issue:13244|issue:16618", "issue:15980|issue:16618" ], - "prepared_review_unit_hash": "6ecb5e0f3a8a83d0a456d71df09f36534fc6fc3f219e64fd95715f0bbf3d4045", + "prepared_review_unit_hash": "33c647724a046cb82f8c07e404fdcd3910777b05e8d41a790ff5636a7e5d1424", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 2477, + "serialized_chars": 2478, "estimated_input_tokens": 620, "estimated_eval_tokens": 1496 }, @@ -39,7 +39,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 2477, + "serialized_chars": 2478, "estimated_input_tokens": 620, "estimated_eval_tokens": 1496 }, @@ -48,24 +48,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These three issues are about different failure modes: Tapas tokenization behavior, a private-model download authentication/error-message problem, and a model-loading failure for bert-base-uncased. They do not look like the same underlying bug cluster.", - "confidence": 0.92, - "canonical_issue_reason": "Issue 16618 is the most active and most general user-facing report in the set, but it is still a different problem from the other two.", + "summary": "These issues are not true duplicates. #16618 is a broad model-loading failure, while #15980 is specifically an authentication/error-message bug for private model downloads, and #13244 is about Tapas tokenization behavior versus TensorFlow code. The only plausible cluster anchor is #16618 by breadth, but it is not the same underlying problem as the others.", + "confidence": 0.96, + "canonical_issue_reason": "#16618 is the broadest and most central-sounding issue in the set, but it still does not actually cover the other reports\u2019 distinct failure modes.", "canonical_pr_reason": null, - "best_issue_reason": "16618 is the strongest representative only by discussion/activity, not because it matches the other issues.", + "best_issue_reason": "#16618 is the best representative only because it is the most generic model-loading report; however, it should not be treated as a duplicate of the other two.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:13244", "right": "issue:16618", "accept": false, - "reason": "Tapas tokenization mismatch is a different code path and symptom than a bert-base-uncased model-loading failure." + "reason": "Tapas tokenization mismatch is a different subsystem and symptom than a generic Bert model-loading failure; no clear shared code-path bug." }, { "left": "issue:15980", "right": "issue:16618", "accept": false, - "reason": "Private-model authentication/error-message handling is distinct from a generic inability to load bert-base-uncased." + "reason": "#15980 is about downloading a private model while logged out and the resulting error message, which is an auth/UX issue, not the same underlying load failure as #16618." } ] }, @@ -84,8 +84,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -208,415 +208,439 @@ "issue:43541|issue:43828", "issue:36010|issue:39290", "issue:44841|issue:45084", - "issue:41093|issue:41762", - "issue:45800|issue:45823" + "issue:45800|issue:45823", + "issue:41093|issue:41762" ], - "prepared_review_unit_hash": "616cd81c38e8480ed7d8f53468498fdf7fd59fd92209ddc88a0170377250f824", + "prepared_review_unit_hash": "86527d6645f7f52158d9fb70f3d41248df1b91bebc892599501571b5c47b9170", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 71, - "serialized_chars": 48875, - "estimated_input_tokens": 12219, - "estimated_eval_tokens": 24694 + "serialized_chars": 48867, + "estimated_input_tokens": 12217, + "estimated_eval_tokens": 24690 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most pairs are false positives from broad lexical similarity. The only clear duplicate is the ZeRO-3 Bert index-out-of-bounds report pair; the rest differ by model, backend, or failure mode.", - "confidence": 0.78, - "canonical_issue_reason": "issue:43596 is the cleanest representative of the real duplicate pair: it describes the ZeRO-3/Bert index-out-of-bounds failure clearly and can serve as the canonical target for issue:43638.", + "summary": "This cluster is very noisy: most pairs are false positives across unrelated model, runtime, and docs issues. The only clear duplicate family is the DeepSpeed ZeRO-3/Bert index-0-out-of-bounds reports.", + "confidence": 0.61, + "canonical_issue_reason": "Issue 43638 is the clearest representative of the only true duplicate family here: the ZeRO-3/BertModel index-0-out-of-bounds failure. Its wording is broad enough to cover the paired report.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43596 is the best issue to anchor duplicate triage because it is specific, reproducible, and has a true near-duplicate in issue:43638.", + "best_issue_reason": "Best overall representative of the cluster\u2019s only confirmed duplicate subtopic; the rest of the cluster is mostly unrelated or too loosely similar.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44322", "right": "issue:45464", "accept": false, - "reason": "Same project area (Qwen3.5) but different failures: config attribute error vs streaming chat/completions failure." + "reason": "Different Qwen3.5 failures: config attribute missing vs streaming chat/completions issue." }, { "left": "issue:45237", "right": "issue:45685", "accept": false, - "reason": "Different models and backends: GPT-OSS on AMD GPUs vs an MPS histogram/int error." + "reason": "Different MoE/runtime paths and different hardware backends; not the same bug." }, { "left": "issue:43716", "right": "issue:45237", "accept": false, - "reason": "Different bugs: Mistral-3 dtype mismatch in preprocessing vs GPT-OSS not working on AMD GPUs." + "reason": "Mistral-3 dtype mismatch is unrelated to GPT-OSS AMD GPU failure." }, { "left": "issue:43828", "right": "issue:45237", "accept": false, - "reason": "Both are dtype/back-end themed, but they are unrelated model-specific failures with no shared code path." + "reason": "Different models and different dtype/device error paths." }, { "left": "issue:45237", "right": "issue:45507", "accept": false, - "reason": "Different model families and symptoms: AMD GPU failure vs invalid method call in GraniteMoEHybrid." + "reason": "GPU runtime failure vs invalid method call in a different MoE model." }, { "left": "issue:44292", "right": "issue:45237", "accept": false, - "reason": "Unrelated failures: Qwen-3 NVFP4 runtime error vs GPT-OSS AMD GPU issue." + "reason": "Different model families and different NVFP4/GPU failure modes." }, { "left": "issue:30277", "right": "issue:35545", "accept": false, - "reason": "Different subsystems entirely: DeepSpeed collective mismatch vs ONNX export failure." + "reason": "DeepSpeed collective mismatch is unrelated to ONNX export." }, { "left": "issue:35545", "right": "issue:42915", "accept": false, - "reason": "Both are model/runtime issues, but one is ModernBERT ONNX export and the other is Qwen3Moe FP8 loading." + "reason": "ONNX export error vs FineGrainedFP8 model failure; unrelated." }, { "left": "issue:35545", "right": "issue:41720", "accept": false, - "reason": "Different concrete problems: ONNX export vs auto device mapping cuda assert." + "reason": "Different model/export path vs auto device mapping CUDA assert." }, { "left": "issue:44509", "right": "issue:44512", "accept": false, - "reason": "Both are v5 docs cleanup items, but they reference different removed docs/features and are not the same underlying issue." + "reason": "Both are docs regressions, but they cover different removed docs topics." }, { "left": "issue:35545", "right": "issue:36010", "accept": false, - "reason": "Different code paths: ONNX export vs missing GenerationMixin import." + "reason": "Export error vs import error for GenerationMixin; unrelated." }, { "left": "issue:44805", "right": "issue:45823", "accept": false, - "reason": "Both involve tensor shape/device problems, but the reports are about different models and different failure modes." + "reason": "Mask/index shape mismatch vs Gemma4 device-mapping mismatch; different bugs." }, { "left": "issue:43638", "right": "issue:44805", "accept": false, - "reason": "Similar error shape, but one is ZeRO-3 Bert index-out-of-bounds and the other is a generic mask/index shape mismatch." + "reason": "Both are IndexError-like, but one is ZeRO-3/Bert loading and the other is a mask-shape mismatch." }, { "left": "issue:43596", "right": "issue:44805", "accept": false, - "reason": "Not the same bug: ZeRO-3 Bert loading failure vs a separate mask shape mismatch." + "reason": "Different failure mode: ZeRO-3 Bert loading vs mask/tensor shape mismatch." }, { "left": "issue:44805", "right": "issue:45161", "accept": false, - "reason": "Generic shape-mismatch IndexError vs a GPT-OSS tensor-parallel issue." + "reason": "Different problems: mask shape mismatch vs GPT-OSS tensor-parallel issue." }, { "left": "issue:44805", "right": "issue:44918", "accept": false, - "reason": "Different failure mechanisms: mask/index shape mismatch vs Qwen3.5 embedding unpacking with TRL SFT." + "reason": "Different models and different SFT/input-embedding failure paths." }, { "left": "issue:34634", "right": "issue:35545", "accept": false, - "reason": "BarkProcessor voice_preset bug is unrelated to ModernBERT ONNX export." + "reason": "BarkProcessor voice_preset issue is unrelated to ModernBERT ONNX export." }, { "left": "issue:30277", "right": "issue:43638", "accept": false, - "reason": "Both involve distributed loading, but the specific failures and model paths are different." + "reason": "Different DeepSpeed symptoms and different models; not the same bug." }, { "left": "issue:43596", "right": "issue:43638", "accept": true, - "reason": "Same underlying ZeRO-3/Bert index-out-of-bounds bug; the titles differ only in model specificity." + "reason": "Same ZeRO-3/Bert index-0-out-of-bounds failure with only minor wording differences." }, { "left": "issue:43039", "right": "issue:45161", "accept": false, - "reason": "Liger Kernel cross-entropy routing issue vs GPT-OSS tensor-parallel failure; not the same bug." + "reason": "Liger Kernel cross-entropy call issue vs GPT-OSS tensor-parallel failure; unrelated." }, { "left": "issue:43975", "right": "issue:44457", "accept": false, - "reason": "DeepSeek detokenization regression and LoRA save/load inconsistency are unrelated." + "reason": "Detokenization bug vs LoRA merge/save/load inconsistency; unrelated." }, { "left": "issue:44292", "right": "issue:44457", "accept": false, - "reason": "Different model/runtime issues: Qwen-3 NVFP4 error vs LoRA merge/reload mismatch." + "reason": "Different model/runtime bug vs LoRA checkpoint consistency issue." }, { "left": "issue:34928", "right": "issue:44805", "accept": false, - "reason": "Both are shape-related, but activation-checkpointing/FSDP mismatch is a different code path from the mask-index error." + "reason": "FSDP checkpointing size mismatch is not the same as mask-index mismatch." }, { "left": "issue:24643", "right": "issue:30277", "accept": false, - "reason": "DeepSpeed training weight-shape error is unrelated to the Jamba collective mismatch." + "reason": "Different DeepSpeed errors and different code paths." }, { "left": "issue:43638", "right": "issue:45161", "accept": false, - "reason": "ZeRO-3 Bert loading failure vs GPT-OSS tensor-parallel issue." + "reason": "ZeRO-3 Bert loading bug is unrelated to GPT-OSS tensor-parallel failure." }, { "left": "issue:34928", "right": "issue:36331", "accept": false, - "reason": "Activation-checkpointing tensor mismatch and compute_loss signature mismatch are unrelated." + "reason": "Activation-checkpointing size mismatch is unrelated to CustomTrainer signature mismatch." + }, + { + "left": "issue:43596", + "right": "issue:45161", + "accept": false, + "reason": "Different model/load bug vs GPT-OSS TP issue." }, { "left": "issue:43716", "right": "issue:45507", "accept": false, - "reason": "Mistral-3 dtype mismatch vs GraniteMoEHybrid invalid method call." + "reason": "Different models and different dtype/method errors." }, { "left": "issue:39290", "right": "issue:45081", "accept": false, - "reason": "Gemma3 vLLM config attribute error is unrelated to the Mistral tokenizer regex patch crash." + "reason": "Gemma3/vLLM config attribute issue is unrelated to Mistral tokenizer regex crash." }, { "left": "issue:43828", "right": "issue:45161", "accept": false, - "reason": "Autocast dtype mismatch in Phi-tiny-MoE is not the same as the GPT-OSS tensor-parallel bug." - }, - { - "left": "issue:43425", - "right": "issue:44292", - "accept": false, - "reason": "Torch version incompatibility and a Qwen-3 NVFP4 runtime issue are different problems." + "reason": "Different models and different dtype/TP failures." }, { "left": "issue:44918", "right": "issue:45161", "accept": false, - "reason": "TRL embedding unpacking issue is unrelated to GPT-OSS MoE tensor-parallel failure." + "reason": "TRL embedding unpacking issue is unrelated to GPT-OSS TP failure." }, { "left": "issue:34928", "right": "issue:41720", "accept": false, - "reason": "FSDP checkpointing mismatch and Qwen3 auto-device-map cuda assert are different failures." + "reason": "FSDP/activation-checkpointing mismatch is unrelated to CUDA assert on auto device mapping." }, { "left": "issue:43716", "right": "issue:45685", "accept": false, - "reason": "Mistral-3 preprocessing dtype mismatch vs MPS histogram/int limitation." + "reason": "Different model-specific dtype issue vs MPS histogram op problem." }, { "left": "issue:43541", "right": "issue:43716", "accept": false, - "reason": "Different model paths and error types: torch dynamo grouped_mm tracing vs image-preprocessor dtype mismatch." + "reason": "Grouped_mm tracing error vs dtype mismatch; not the same bug." }, { "left": "issue:43828", "right": "issue:44292", "accept": false, - "reason": "Different models and failures: autocast dtype mismatch vs Qwen-3 NVFP4 runtime error." + "reason": "Different MoE models and different dtype/runtime failures." }, { "left": "issue:43039", "right": "issue:44918", "accept": false, - "reason": "Liger Kernel cross-entropy call issue is unrelated to Qwen3.5 embedding unpacking." + "reason": "Cross-entropy/Liger Kernel issue is unrelated to TRL embedding unpacking." }, { "left": "issue:41762", "right": "issue:43638", "accept": false, - "reason": "Both mention ZeRO-3 index errors, but Gemma3 loading and Bert loading are different concrete bug reports." + "reason": "Same broad ZeRO-3 theme, but different model and load path; not clearly the same bug." }, { "left": "issue:43827", "right": "issue:44512", "accept": false, - "reason": "Both are docs regressions, but they concern different removed items and are not duplicates." + "reason": "Both are docs cleanups for v5, but they reference different removed content." }, { "left": "issue:43716", "right": "issue:44560", "accept": false, - "reason": "Different bugs: dtype mismatch vs video StopIteration in Qwen3-vl-embedding." + "reason": "Different model/runtime issues with no shared code path." }, { "left": "issue:43828", "right": "issue:45685", "accept": false, - "reason": "Autocast dtype mismatch and MPS histogram/int error are unrelated." + "reason": "Different models, backends, and error types." }, { "left": "issue:44560", "right": "issue:45685", "accept": false, - "reason": "Different models and failure modes: video StopIteration vs an MPS kernel limitation." + "reason": "Qwen3-vl StopIteration is unrelated to MPS histogram Int error." }, { "left": "issue:43039", "right": "issue:43828", "accept": false, - "reason": "Liger Kernel cross-entropy routing issue is not the same as Phi-tiny-MoE autocast dtype mismatch." + "reason": "Different model/runtime issues despite both involving execution failures." }, { "left": "issue:30277", "right": "issue:36331", "accept": false, - "reason": "Distributed collective mismatch and trainer compute_loss kwarg incompatibility are unrelated." + "reason": "DeepSpeed collective mismatch is unrelated to CustomTrainer compute_loss signature." }, { "left": "issue:34928", "right": "issue:36010", "accept": false, - "reason": "Activation checkpointing/FSDP mismatch is unrelated to the GenerationMixin import error." + "reason": "FSDP size mismatch is unrelated to GenerationMixin import failure." }, { "left": "issue:43039", "right": "issue:44805", "accept": false, - "reason": "Liger Kernel cross-entropy issue is unrelated to the generic mask/index shape mismatch." + "reason": "Liger Kernel cross-entropy call issue is not the same as mask-shape IndexError." }, { "left": "issue:43901", "right": "issue:44512", "accept": false, - "reason": "Both are documentation issues, but they refer to different deprecated APIs and different docs pages." + "reason": "Different docs issues: return_all_scores vs removed run command." }, { "left": "issue:41762", "right": "issue:43596", "accept": false, - "reason": "Same broad ZeRO-3 loading area, but Gemma3 and Bert reports are not clearly the same underlying bug." + "reason": "Both are ZeRO-3/load related, but different models and error contexts." }, { "left": "issue:39290", "right": "issue:41720", "accept": false, - "reason": "Gemma3 text-config attribute error and Qwen3 auto-device-map cuda assert are different model-specific bugs." + "reason": "Gemma3/vLLM config bug is unrelated to Qwen3 device-mapping CUDA assert." }, { "left": "issue:44841", - "right": "issue:45084", + "right": "issue:45823", + "accept": false, + "reason": "Processor failure for Voxtral is unrelated to Gemma4 PLE device mismatch." + }, + { + "left": "issue:43039", + "right": "issue:44560", + "accept": false, + "reason": "Liger Kernel cross-entropy issue is unrelated to Qwen3-vl StopIteration." + }, + { + "left": "issue:43866", + "right": "issue:44863", + "accept": false, + "reason": "Corrupted checkpoint vs implementation unable to load checkpoints; different root causes." + }, + { + "left": "issue:43854", + "right": "issue:43866", + "accept": false, + "reason": "Unit-test load failure for GLM-4.7 is unrelated to Ovis2 checkpoint corruption." + }, + { + "left": "issue:39290", + "right": "issue:43531", "accept": false, - "reason": "Processor failure for Voxtral is unrelated to the template-node compilation error." + "reason": "Gemma3/vLLM config attribute issue is unrelated to Qwen3-MoE sliding_window bug." }, { "left": "issue:41093", "right": "issue:43638", "accept": false, - "reason": "Mask/index shape mismatch and ZeRO-3 Bert index-out-of-bounds are different failures." + "reason": "Both are IndexErrors, but one is a mask/tensor shape mismatch and the other is ZeRO-3 Bert loading." }, { "left": "issue:43425", "right": "issue:45070", "accept": false, - "reason": "Torch version incompatibility and pydantic PretrainedConfig field regression are different issues." + "reason": "Torch-version incompatibility vs pydantic PretrainedConfig field regression; unrelated." }, { "left": "issue:43596", "right": "issue:45800", "accept": false, - "reason": "ZeRO-3 Bert loading bug is unrelated to the torch/transformers incompatibility report." + "reason": "ZeRO-3 Bert loading bug is unrelated to torch/transformers version incompatibility." }, { "left": "issue:43792", "right": "issue:45070", "accept": false, - "reason": "Whisper loading/runtime issue is unrelated to the pydantic model regression." + "reason": "Whisper run failure is unrelated to the pydantic model-field regression." }, { "left": "issue:45362", "right": "issue:45464", "accept": false, - "reason": "Both involve Qwen3.5 and chat usage, but one is a model crash in transformers chat and the other is a streaming API failure." + "reason": "Qwen3.5 chat crash vs streaming inference failure; different behavior and likely different bugs." }, { "left": "issue:43404", "right": "issue:43866", "accept": false, - "reason": "Mistral3 lm_head tying bug and Ovis2 checkpoint corruption are unrelated." + "reason": "Mistral3 lm_head tying bug is unrelated to Ovis2 checkpoint corruption." }, { - "left": "issue:43854", - "right": "issue:43866", + "left": "issue:43828", + "right": "issue:45685", "accept": false, - "reason": "GLM-4.7-Flash unit-test loading failure is unrelated to the Ovis2 checkpoint corruption report." + "reason": "Different models and backends; not the same issue." }, { - "left": "issue:43866", - "right": "issue:44863", + "left": "issue:43541", + "right": "issue:45685", "accept": false, - "reason": "Different model families and load failures: corrupted checkpoint vs NemotronH checkpoint incompatibility." + "reason": "Grouped_mm tracing error is unrelated to MPS histogram Int error." }, { - "left": "issue:45507", - "right": "issue:45685", + "left": "issue:43541", + "right": "issue:43828", "accept": false, - "reason": "GraniteMoEHybrid invalid method call is unrelated to the MPS histogram/int kernel issue." + "reason": "Different MoE runtime failures and different models." }, { - "left": "issue:43716", - "right": "issue:44292", + "left": "issue:36010", + "right": "issue:39290", "accept": false, - "reason": "Mistral-3 dtype mismatch and Qwen-3 NVFP4 runtime error are different model-specific bugs." + "reason": "GenerationMixin import failure is unrelated to Gemma3/vLLM config mismatch." }, { - "left": "issue:44292", - "right": "issue:44918", + "left": "issue:44841", + "right": "issue:45084", "accept": false, - "reason": "Qwen-3 NVFP4 runtime error and Qwen3.5/TRL embedding unpacking are unrelated." + "reason": "Processor failure for Voxtral is unrelated to template-node compilation error." }, { - "left": "issue:43039", - "right": "issue:44560", + "left": "issue:45800", + "right": "issue:45823", "accept": false, - "reason": "Liger Kernel cross-entropy routing issue is unrelated to the Qwen3-vl StopIteration bug." + "reason": "Torch/transformers incompatibility is unrelated to Gemma4 device mismatch." }, { - "left": "issue:44841", - "right": "issue:45084", + "left": "issue:41093", + "right": "issue:41762", "accept": false, - "reason": "Voxtral processor failure and template-node compilation failure are different code paths." + "reason": "Both are IndexErrors, but different models and different failure mechanisms." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The 43596/43638 pair is well supported as a near-duplicate ZeRO-3 Bert index-out-of-bounds report, and the other soft-edge rejections avoid overclaiming despite lexical similarity." + "feedback": "The summary is grounded in the packet and stays conservative overall. The only accepted duplicate pair, issue:43596 \u2194 issue:43638, is well supported by the titles and descriptions. The rest of the soft-edge false verdicts are appropriately cautious. Minor caveat: issue:41762 is adjacent to the ZeRO-3 cluster, but treating it as non-duplicate is a defensible conservative call." }, "evaluator_used": true, "retried": false, @@ -632,8 +656,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -722,244 +746,244 @@ "issue:43828|issue:44928", "issue:43866|issue:44403" ], - "prepared_review_unit_hash": "56e0ea3808b14e624c576e290464fb05ccb11227a14cb973ef75bd093dd0c803", + "prepared_review_unit_hash": "0efb76dcc62da0164ea19bdb6b27457dcc36aea720855733fbfb3e1518716a4d", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 34, - "serialized_chars": 37475, - "estimated_input_tokens": 9369, - "estimated_eval_tokens": 18994 + "serialized_chars": 37472, + "estimated_input_tokens": 9368, + "estimated_eval_tokens": 18992 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most items are distinct model- or release-specific bugs. The only clear duplicate set is the v5 non-persistent-buffer regression trio.", + "summary": "This cluster is mostly a set of loosely related Transformers/v5 regression reports, but only the non-persistent-buffer corruption reports form a true duplicate subcluster.", "confidence": 0.84, - "canonical_issue_reason": "Issue 43950 is the clearest, most specific statement of the shared bug: `from_pretrained()` corrupts non-persistent buffers in Transformers 5.x.", + "canonical_issue_reason": "issue:43950 is the clearest root report for the only exact duplicate subcluster here: non-persistent buffers being corrupted on `from_pretrained()` in v5, with direct matches in 43644 and 44534.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43950 is the best representative issue because it names the concrete failing path (`from_pretrained`) and the regression symptom directly.", + "best_issue_reason": "issue:43950 is the best anchor issue because it states the regression most precisely and covers the other matching reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43296", "right": "issue:43366", "accept": false, - "reason": "Different failures: PaddleOCR-VL loading in vLLM vs gpt-oss GGUF architecture support." + "reason": "Different problems: vLLM/PaddleOCR-VL loading failure versus GGUF gpt-oss architecture support." }, { "left": "issue:43572", "right": "issue:45070", "accept": false, - "reason": "Both are v5 config regressions, but the missing `pad_token_idx` and the pydantic `PretrainedConfig` breakage are different bugs." + "reason": "Both are v5 config regressions, but one is missing `pad_token_idx` and the other is a pydantic `PretrainedConfig` field break; not the same bug." }, { "left": "issue:44387", "right": "issue:45005", "accept": false, - "reason": "Different problem spaces: int4 memory/OOM vs tied weights in translation models." + "reason": "Related to v5 and weights, but one is int4 memory/OOM behavior and the other is tied-weights handling on translation models; different code paths." }, { "left": "issue:43541", "right": "issue:45161", "accept": false, - "reason": "Both involve MoE models, but one is a grouped_mm tracing error and the other is a tensor-parallel support issue." + "reason": "Different symptoms and model paths: grouped_mm tracing failure versus GPT-OSS MoE tensor-parallel issue." }, { "left": "issue:43828", "right": "issue:45464", "accept": false, - "reason": "Different code paths and symptoms: autocast dtype mismatch vs chat/completions streaming failure." + "reason": "One is a torch autocast dtype mismatch, the other is a streaming chat/completions failure on Qwen3.5; not the same bug." }, { "left": "issue:44291", "right": "issue:45070", "accept": false, - "reason": "`init_empty_weights` argument handling is unrelated to the `PretrainedConfig` pydantic regression." + "reason": "Both are v5 breakages, but init-empty-weights argument handling is distinct from the pydantic config-field regression." }, { "left": "issue:43366", "right": "issue:45464", "accept": false, - "reason": "Model/architecture support vs streaming API failure; no shared concrete bug." + "reason": "Unrelated model support/report: gpt-oss GGUF versus Qwen3.5 chat streaming failure." }, { "left": "issue:45020", "right": "issue:45800", "accept": false, - "reason": "Generic `remote_code` breakage vs torch/transformers version incompatibility are not the same issue." + "reason": "Broad compatibility issues, but one is remote_code regressions and the other is torch 2.4.1 / transformers 5.8.0 incompatibility; not the same concrete change." }, { "left": "issue:43531", "right": "issue:45070", "accept": false, - "reason": "Qwen3-MoE sliding-window behavior is unrelated to the config-field regression." + "reason": "Qwen3-MoE sliding_window behavior is unrelated to the PretrainedConfig/pydantic regression." }, { "left": "issue:43541", "right": "issue:44589", "accept": false, - "reason": "Tracing/grouped_mm failure is unrelated to missing Float8 storage support." + "reason": "Different runtime failures: torch dynamo grouped_mm tracing versus missing Float8 storage object." }, { "left": "issue:41093", "right": "issue:43596", "accept": false, - "reason": "Both are IndexErrors, but the mask-shape mismatch and zero-sized tensor indexing are different bugs." + "reason": "Both are IndexError-style reports, but they concern different tensors, models, and setup conditions." }, { "left": "issue:43054", "right": "issue:43994", "accept": false, - "reason": "Both mention SigLIP2, but one is about worse text embeddings and the other about nonsensical AutoModel/pipeline outputs; not clearly the same concrete defect." + "reason": "Same model family, but one is a relative-quality complaint and the other is a loading/pipeline correctness report; root cause is not clearly the same." }, { "left": "issue:45542", "right": "issue:45800", "accept": false, - "reason": "TensorBoard-without-TensorFlow backend handling is unrelated to the torch/transformers incompatibility." + "reason": "TensorBoard-without-TensorFlow backend error is unrelated to a torch/transformers version incompatibility." }, { "left": "issue:43716", "right": "issue:44918", "accept": false, - "reason": "Different failures: image preprocessor dtype mismatch vs TRL SFT embedding unpacking." + "reason": "Both mention Qwen-like multimodal models, but one is an image-preprocessor dtype mismatch and the other is a TRL SFT embedding unpacking failure." }, { "left": "issue:44918", "right": "issue:45685", "accept": false, - "reason": "Unpacking embeddings in Qwen3.5 and MPS histogram errors are unrelated MoE issues." + "reason": "Completely different failures: TRL embedding unpacking versus an MPS histogram kernel not implemented for int." }, { "left": "issue:45632", "right": "issue:45698", "accept": false, - "reason": "Both involve custom-code loading, but one is a leaf-directory cache collision and the other is wrong-module selection after save_pretrained; not the same concrete bug." + "reason": "Both involve custom code loading, but one is a cache-path collision and the other is loading the wrong custom module after save/load; not clearly the same bug." }, { "left": "issue:43404", "right": "issue:45005", "accept": false, - "reason": "Both mention tied weights, but the model families and failure modes differ." + "reason": "Related to tied weights, but the affected model classes and failure modes differ enough that they are not the same concrete issue." }, { "left": "issue:43645", "right": "issue:45399", "accept": false, - "reason": "Jupyter custom-model initialization and flash-attn fallback logic are unrelated." + "reason": "Notebook/custom-model initialization breakage is unrelated to flash-attn2 fallback selection logic." }, { "left": "issue:40990", "right": "issue:44960", "accept": false, - "reason": "Unrelated model/perplexity report vs a GLM5 issue." + "reason": "Different model and symptom: perplexity regression on gpt-oss-20b versus a GLM5 issue." }, { "left": "issue:44530", "right": "issue:45005", "accept": false, - "reason": "PagedAttentionCache linear_attention crash is unrelated to tied-weight translation issues." + "reason": "PagedAttentionCache linear_attention crash is unrelated to tied-weights handling." }, { "left": "issue:44360", "right": "issue:44512", "accept": false, - "reason": "An algorithmic DSA-indexer discussion is unrelated to a docs typo." + "reason": "DSA indexer/ReLU discussion is unrelated to docs mentioning the removed run command." }, { "left": "issue:42915", "right": "issue:45399", "accept": false, - "reason": "Qwen3-MoE FP8 failure is unrelated to flash-attn fallback checks." + "reason": "Qwen3-MoE FP8 failure and flash-attn2 fallback gating are different bugs." }, { "left": "issue:44863", "right": "issue:45005", "accept": false, - "reason": "NemotronH checkpoint loading and tied-weight translation regressions are different bugs." + "reason": "NemotronH checkpoint loading failure is unrelated to tied-weights regressions." }, { "left": "issue:43873", "right": "issue:45005", "accept": false, - "reason": "Quantization/offloading behavior is unrelated to tied-weight handling." + "reason": "Quantization/offloading behavior is a different issue from tied-weights handling." }, { "left": "issue:39186", "right": "issue:44918", "accept": false, - "reason": "FSDP 2-D weight error is unrelated to TRL embedding unpacking." + "reason": "FSDP 2-D weight error and TRL embedding unpacking failure are unrelated." }, { "left": "issue:44492", "right": "issue:44512", "accept": false, - "reason": "A cache-strategy typo is unrelated to the removed run-command docs issue." + "reason": "A typo in cache strategies is not the same as stale docs for the removed transformers run command." }, { "left": "issue:44918", "right": "issue:45507", "accept": false, - "reason": "Different failures in different model families; no shared code-path bug is evident." + "reason": "Different model families and failures: TRL embedding unpacking versus GraniteMoEHybrid invalid-method calls." }, { "left": "issue:43975", "right": "issue:44292", "accept": false, - "reason": "Detokenization regression and `init_empty_weights` argument handling are unrelated." + "reason": "DeepSeek detokenization bug and Qwen-3 NVFP4 runtime failure are unrelated." }, { "left": "issue:43950", "right": "issue:44534", "accept": true, - "reason": "Same non-persistent-buffer regression in Transformers v5: both say `from_pretrained`/v5 fills or corrupts non-persistent buffers with junk." + "reason": "Same underlying regression: v5 corrupts non-persistent buffers on load; 44534 is essentially a reworded duplicate of 43950." }, { "left": "issue:43644", "right": "issue:43950", "accept": true, - "reason": "Same underlying bug: Transformers v5 overwrites/corrupts non-persistent buffers during model loading." + "reason": "Same bug: both report v5 filling/corrupting non-persistent buffers, with 43950 giving the clearest root-cause description." }, { "left": "issue:44292", "right": "issue:45507", "accept": false, - "reason": "Unexpected `_is_hf_initialized` argument and invalid method calls on GraniteMoEHybrid are unrelated." + "reason": "Different failures on different models; no shared concrete code-path is evident." }, { "left": "issue:39692", "right": "issue:43054", "accept": false, - "reason": "SigLIP2 docs-example errors and SigLIP2 embedding quality complaints are related only by model family, not the same concrete bug." + "reason": "Both concern SigLIP2, but one is a doc example/model-processor mismatch and quantization failure, while the other is a quality/regression report." }, { "left": "issue:43828", "right": "issue:44928", "accept": false, - "reason": "Both mention dtype/precision problems, but autocast mismatch and RLHF NaN from 3D position_ids are different bugs." + "reason": "Both involve dtype/precision concerns, but the runtime contexts and failure modes are different." }, { "left": "issue:43866", "right": "issue:44403", "accept": false, - "reason": "Checkpoint corruption and loading noise are not the same issue." + "reason": "Ovis2 checkpoint corruption and an unrelated loading-noise issue do not look like the same bug." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet, and the only accepted cluster is well supported by the titles: 43644, 43950, and 44534 all describe the same Transformers v5 non-persistent-buffer corruption/regression. The rejected soft-edge pairs are stated conservatively and do not overclaim duplication from the available evidence." + "feedback": "Grounded overall. The anchor choice and canonical issue are well supported by the packet, and the only accepted pairings (43950/44534 and 43644/43950) look like the same non-persistent-buffer corruption regression. The negative pair verdicts are appropriately conservative and do not overstate evidence." }, "evaluator_used": true, "retried": false, @@ -975,8 +999,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -1040,8 +1064,8 @@ "issue:41553|issue:43881", "issue:42947|issue:44387", "issue:44403|issue:44451", - "issue:43975|issue:45520", "issue:43643|issue:45020", + "issue:43975|issue:45520", "issue:43054|issue:43493", "issue:43975|issue:45237", "issue:44403|issue:44488", @@ -1066,20 +1090,20 @@ "issue:42915|issue:45020", "issue:43638|issue:45823" ], - "prepared_review_unit_hash": "846505c88f5c08bd3a8de30cce1ae3850e2737da3adaec4e0e33fdbaf5a3ad51", + "prepared_review_unit_hash": "cec015b3ff7aed38c924280535f59102455495021ec23cf70845ee5056c18a7b", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 35, - "serialized_chars": 38226, + "serialized_chars": 38227, "estimated_input_tokens": 9557, "estimated_eval_tokens": 19370 }, @@ -1088,231 +1112,231 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose collection of unrelated Transformers regression reports. I only see one plausible duplicate-style link: the two SigLIP2 issues, which both point to an HF-vs-reference implementation mismatch. Most other soft pairs are same broad area but not the same concrete bug.", - "confidence": 0.58, - "canonical_issue_reason": "issue:43493 is the best root-cause report for the only convincing duplicate pair: it explicitly describes the SigLIP2 HF-vs-JAX implementation discrepancy, while issue:43054 reads more like a symptom report about degraded text embeddings.", + "summary": "Most pairs are distinct bugs across different models, loaders, and runtime paths. The two clear soft duplicates are the SigLIP2 implementation-mismatch pair (43054/43493) and the docs pipeline-removal pair (43827/44509). The earlier SigLIP2 quality-only pairing (38175/43054) is not strong enough to accept from titles alone.", + "confidence": 0.82, + "canonical_issue_reason": "issue:45020 is the best open umbrella for the current remote_code-related regression cluster, but it should be treated as a tracker rather than a duplicate of the more specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the strongest cluster anchor overall because it is an open umbrella report for recent remote_code-related breakages and best matches the cluster\u2019s repeated loading/version-regression theme, even though many items are still not true duplicates.", + "best_issue_reason": "issue:45020 is the strongest global issue anchor because it is open, broad, and clearly positioned as a tracker for a recurring family of recent compatibility breakages.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36331", "right": "issue:41762", "accept": false, - "reason": "Different failures: trainer API change vs DeepSpeed ZeRO-3 loading index error." + "reason": "Different failures: trainer signature mismatch vs Gemma3 ZeRO-3 loading index error." }, { "left": "issue:41628", "right": "issue:45800", "accept": false, - "reason": "Import error for AutoImageProcessor is unrelated to torch/transformers compatibility." + "reason": "Different subsystems: missing AutoImageProcessor import vs torch/transformers version incompatibility." }, { "left": "issue:41553", "right": "issue:43883", "accept": false, - "reason": "Different models and code paths: Voxtral AutoTokenizer error vs Molmo tied-weights attribute error." + "reason": "Different model paths and symptoms: Voxtral tokenizer error handling vs Molmo tied-weights attribute error." }, { "left": "issue:43872", "right": "issue:44589", "accept": false, - "reason": "Both are incompatibility reports, but the concrete errors and subsystems differ." + "reason": "Both are compatibility errors, but the concrete bugs differ: bitsandbytes kwarg mismatch vs float8 storage lookup." }, { "left": "issue:38175", "right": "issue:43054", "accept": false, - "reason": "Both involve SigLIP/SigLIP2, but one is zero probabilities and the other is general embedding quality; not clearly the same bug." + "reason": "Same model family, but only related quality symptoms are shown; no shared concrete implementation bug is established." }, { "left": "issue:43054", "right": "issue:45701", "accept": false, - "reason": "SigLIP2 embedding quality issue vs a generic tokenization regression report." + "reason": "SigLIP2 embedding quality regression is unrelated to a generic tokenization-change report." }, { "left": "issue:44910", "right": "issue:45859", "accept": false, - "reason": "Different Qwen3.5 MoE problems: flash-attn packed-sequence handling vs CPU offload incompatibility." + "reason": "Different code paths: Flash Attention packed-sequence crash vs CPU-offload rotary-embedding incompatibility." }, { "left": "issue:41553", "right": "issue:43881", "accept": false, - "reason": "Unrelated model-loading failures for different models." + "reason": "Different model/load failures: Voxtral tokenizer error vs glm-4v-9b loading failure." }, { "left": "issue:42947", "right": "issue:44387", "accept": false, - "reason": "Gradient checkpointing/LoRA issue is not the same as int4 CUDA memory growth." + "reason": "LoRA gradient-checkpointing effectiveness and int4 OOM are separate training/memory issues." }, { "left": "issue:44403", "right": "issue:44451", "accept": false, - "reason": "General loading noise vs a specific model load failure; not the same underlying bug." + "reason": "Generic load-noise report vs a specific model failing to load." }, { - "left": "issue:43975", - "right": "issue:45520", + "left": "issue:43643", + "right": "issue:45020", "accept": false, - "reason": "Tokenizer detokenization regression and flash_attn import KeyError are unrelated." + "reason": "AutoConfig missing-fields under trust_remote_code is a specific bug; 45020 is a broader tracker, not the same defect." }, { - "left": "issue:43643", - "right": "issue:45020", + "left": "issue:43975", + "right": "issue:45520", "accept": false, - "reason": "Both mention remote_code, but one is missing fields in AutoConfig while the other is a broad compatibility report." + "reason": "Tokenizer detokenization regression and Python 3.13 flash_attn import failure are unrelated." }, { "left": "issue:43054", "right": "issue:43493", "accept": true, - "reason": "Both point to the same SigLIP2 implementation discrepancy; one is the symptom, the other the root-cause report." + "reason": "Both describe the same SigLIP2 implementation discrepancy; 43493 explicitly frames it as HF vs original JAX mismatch." }, { "left": "issue:43975", "right": "issue:45237", "accept": false, - "reason": "Deepseek tokenizer behavior vs GPT-OSS-20B on AMD GPUs are unrelated." + "reason": "DeepSeek tokenization regression and AMD GPU runtime support issue are unrelated." }, { "left": "issue:44403", "right": "issue:44488", "accept": false, - "reason": "Both are load-failure-ish, but they involve different models and no shared concrete mechanism." + "reason": "One is about noisy load logs, the other is a concrete model load failure." }, { "left": "issue:42947", "right": "issue:43856", "accept": false, - "reason": "LoRA gradient checkpointing and MoE memory usage are different training issues." + "reason": "Gradient checkpointing/LoRA and Qwen3 MoE memory inefficiency are different training concerns." }, { "left": "issue:43881", "right": "issue:43883", "accept": false, - "reason": "Different model families and different failure modes." + "reason": "Different models and failures: glm-4v-9b load problem vs Molmo tied-weights attribute error." }, { "left": "issue:41762", "right": "issue:42915", "accept": false, - "reason": "Gemma3 ZeRO-3 load error and Qwen3 MoE FP8 failure are unrelated." + "reason": "Both involve model loading, but Gemma3 ZeRO-3 and Qwen3 MoE FP8 are distinct bugs." }, { "left": "issue:42491", "right": "issue:44368", "accept": false, - "reason": "Different Qwen3/Qwen3.5 training warnings and failure modes; not the same bug." + "reason": "Compatibility regression vs configuration warning; not the same concrete defect." }, { "left": "issue:45685", "right": "issue:45823", "accept": false, - "reason": "MPS histogram/int dtype issue is unrelated to Gemma4 device mismatch." + "reason": "MPS histogram-not-implemented error and Gemma4 device-mismatch forward error are unrelated." }, { "left": "issue:43975", "right": "issue:44451", "accept": false, - "reason": "Tokenizer detokenization regression vs model loading failure are not the same issue." + "reason": "DeepSeek tokenizer regression and ScandiBERT load failure are not the same bug." }, { "left": "issue:43824", "right": "issue:43872", "accept": false, - "reason": "Import error for Qwen2.5-VL class is unrelated to bitsandbytes Int8Params incompatibility." + "reason": "Qwen2.5-VL import failure is unrelated to the bitsandbytes Int8Params incompatibility." }, { "left": "issue:45020", "right": "issue:45823", "accept": false, - "reason": "Broad remote_code regressions and Gemma4 device placement are different bugs." + "reason": "A broad remote_code tracker is too general to equate with a Gemma4 device-map forward mismatch." }, { "left": "issue:44991", "right": "issue:45399", "accept": false, - "reason": "Tokenizer loading failure vs flash-attn fallback gating are unrelated." + "reason": "Tokenizer loading regression and flash-attn2 fallback gating are separate issues." }, { "left": "issue:44661", "right": "issue:45020", "accept": false, - "reason": "Tokenizer mapping-name failure is a different code path from remote_code breakages." + "reason": "Registry/add-new-model-like failure is a different bug class than the remote_code regression tracker." }, { "left": "issue:43828", "right": "issue:45823", "accept": false, - "reason": "Autocast dtype mismatch in Phi MoE and Gemma4 PLE device mismatch are different problems." + "reason": "Autocast dtype mismatch in Phi-tiny-MoE is unrelated to Gemma4 device placement." }, { "left": "issue:45020", "right": "issue:45356", "accept": false, - "reason": "Remote_code breakages and Kimi-K2.5 tokenizer regression are related only at a very broad version-regression level." + "reason": "Kimi-K2.5 tokenizer regression is a specific tokenizer bug, not the same as the broader remote_code tracker." }, { - "left": "issue:43792", - "right": "issue:44220", + "left": "issue:43782", + "right": "issue:44960", "accept": false, - "reason": "Whisper runtime failure and fbank feature extraction issue may be related, but not clearly the same underlying bug from titles alone." + "reason": "Qwen3VL weight_only loading error and GLM5 are different model-specific failures." }, { "left": "issue:45081", "right": "issue:45399", "accept": false, - "reason": "Mistral regex patch crash and flash-attn fallback blocking are unrelated." + "reason": "Tokenizer regex crash and flash-attn2 fallback gating are unrelated." }, { "left": "issue:43827", "right": "issue:44509", - "accept": false, - "reason": "Both are docs-related pipeline references, but they are separate documentation cleanup reports." + "accept": true, + "reason": "Both are the same docs regression: pipeline tasks removed in v5 but still referenced in documentation." }, { "left": "issue:43404", "right": "issue:43950", "accept": false, - "reason": "lm_head tying bug and non-persistent buffer corruption are different regression types." + "reason": "Mistral3 tied-weight bug and non-persistent buffer corruption are separate loading issues." }, { "left": "issue:43792", - "right": "issue:43872", + "right": "issue:44220", "accept": false, - "reason": "Whisper/fbank failure and bitsandbytes storage-type incompatibility are unrelated." + "reason": "Whisper runtime failure is too underspecified here to confirm the same concrete bug as the fbank feature issue." }, { "left": "issue:43742", "right": "issue:45399", "accept": false, - "reason": "MobileLLM load error vs flash-attn fallback gating are different issues." + "reason": "MobileLLM key error and flash-attn2 fallback blocking are unrelated." }, { "left": "issue:42915", "right": "issue:45020", "accept": false, - "reason": "Qwen3 MoE FP8 failure and remote_code model-loading regressions are not the same bug." + "reason": "Qwen3 MoE FP8 loading failure is a specific model/config bug, not the same as the broad remote_code tracker." }, { "left": "issue:43638", "right": "issue:45823", "accept": false, - "reason": "ZeRO-3 Bert load failure and Gemma4 device mismatch are unrelated." + "reason": "ZeRO-3/BERT index error and Gemma4 device-mismatch are distinct failures." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is cautious and supported by the packet. The only accepted soft edge, issue:43054 \u2194 issue:43493, is plausibly the same SigLIP2 discrepancy, and the other pair verdicts stay conservative by not over-duplicating broad but distinct regression reports." + "accept": false, + "feedback": "The summary is mostly grounded, but it overstates one soft-edge duplicate claim: issue:38175 vs issue:43054 is not clearly supported by the packet. Both are in the SigLIP2 area, but the titles alone only show related model family/symptoms, not the same concrete bug or implementation mismatch. The 43054 vs 43493 pairing is more defensible, and the docs pair 43827 vs 44509 looks solid. Overall, the conservative verdicts are not consistent enough to accept as-is." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -1325,8 +1349,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -1380,8 +1404,8 @@ ], "soft_pairs": [ "issue:43257|issue:43866", - "issue:43940|issue:45313", "issue:43872|issue:44291", + "issue:43940|issue:45313", "issue:45161|issue:45823", "issue:41553|issue:44991", "issue:42915|issue:43872", @@ -1404,28 +1428,28 @@ "issue:43366|issue:44589", "issue:43950|issue:45325", "issue:44560|issue:44918", - "issue:43824|issue:43957", "issue:42617|issue:43366", + "issue:43824|issue:43957", "issue:42915|issue:43957", "issue:36010|issue:42915", "issue:44661|issue:45823", "issue:38175|issue:43994", "issue:43632|issue:45834" ], - "prepared_review_unit_hash": "f37d05c5d27101ba0be54f7851f3f289cf4c68d912396e16e19eab803a045e2d", + "prepared_review_unit_hash": "96c7d10b6ea0fe2395abedec970d727ab1bf8d6b26b8332735261b4f73b78f73", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 32, - "serialized_chars": 36775, + "serialized_chars": 36776, "estimated_input_tokens": 9194, "estimated_eval_tokens": 18644 }, @@ -1434,210 +1458,210 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The bucket is mostly heterogeneous, with only a few true duplicate subgroups. Clear matches exist for custom model initialization regressions, the _is_hf_initialized / init_empty_weights regression family, the mask-shape mismatch pair, and a couple of Qwen3/Qwen3.5 DeepSpeed or video-input regressions. No single issue cleanly represents the entire cluster.", - "confidence": 0.77, - "canonical_issue_reason": "43646 is the best canonical anchor among the actual duplicate subgroup because it states the core custom-model initialization regression broadly; 43645 is the notebook-specific variant of the same bug.", + "summary": "The set is mostly heterogeneous issue reports with only two strong duplicate-like pairs: the Transformers 5 custom model init regression, and the `_is_hf_initialized` / `init_empty_weights` regression. I rejected the rest because they describe different model families, different symptoms, or only share a broad subsystem.", + "confidence": 0.72, + "canonical_issue_reason": "No single canonical issue cleanly spans the full set because the items are mostly unrelated; if a representative anchor is needed, issue 44291 is the clearest root-cause report for the `_is_hf_initialized` regression.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43646 is the strongest representative of a true duplicate pair in this set, but the overall cluster is not a single coherent duplicate group.", + "best_issue_reason": "Issue 44291 is the strongest standalone issue: it states the concrete failure mode, ties it to the v5 regression, and gives the clearest actionable root cause.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43257", "right": "issue:43866", "accept": false, - "reason": "Different problems: Qwen3 MoE weight conversion under accelerate/deepspeed vs a corrupted Ovis2 checkpoint." - }, - { - "left": "issue:43940", - "right": "issue:45313", - "accept": true, - "reason": "Both describe DeepSpeed ZeRO-3 failing to load Qwen3-family weights, i.e. the same loader-path bug." + "reason": "Different problems: Qwen3 MoE weight conversion under accelerate+deepspeed vs a corrupted Ovis2 checkpoint." }, { "left": "issue:43872", "right": "issue:44291", "accept": true, - "reason": "Same _is_hf_initialized / init_empty_weights regression; 43872 is a bitsandbytes manifestation of the same bug." + "reason": "Same underlying regression around `_is_hf_initialized` / `init_empty_weights` in Transformers 5, with bitsandbytes showing the same incompatibility symptom." + }, + { + "left": "issue:43940", + "right": "issue:45313", + "accept": false, + "reason": "Both involve DeepSpeed ZeRO-3 weight loading, but the models and failure details differ enough that this is not clearly the same bug." }, { "left": "issue:45161", "right": "issue:45823", "accept": false, - "reason": "Both involve model-parallel-ish issues, but the concrete failures differ: TP on GPT-OSS MoE vs Gemma4 PLE device placement." + "reason": "Unrelated model-specific issues: GPT-OSS MoE tensor-parallel behavior vs Gemma4 PLE device placement." }, { "left": "issue:41553", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer-related, but one is a bad error message for Voxtral and the other is a tokenizer load failure for a different model." + "reason": "Both are tokenizer-loading failures, but they concern different models and different failure modes." }, { "left": "issue:42915", "right": "issue:43872", "accept": false, - "reason": "Different failure modes and code paths: Qwen3Moe with FineGrainedFP8Config vs a global _is_hf_initialized/bitsandbytes incompatibility." + "reason": "Different subsystems and symptoms: FineGrainedFP8Config on Qwen3Moe vs the `_is_hf_initialized`/bitsandbytes regression." }, { "left": "issue:43854", "right": "issue:44863", "accept": false, - "reason": "Both are model-loading complaints, but they target different architectures and likely different loader bugs." + "reason": "Different model-loading bugs for different architectures; only the broad theme of loading failure matches." }, { "left": "issue:44928", "right": "issue:45588", "accept": false, - "reason": "Qwen3.5 RLHF/Nan instability from 3D position_ids is unrelated to a flash_attention AttributeError on sink-less models." + "reason": "Qwen3.5 RLHF/SDPA NaN behavior is unrelated to the flash_attention `s_aux=None` crash." }, { "left": "issue:36331", "right": "issue:41093", "accept": false, - "reason": "Unexpected trainer kwarg and a mask/index shape mismatch are different bugs." + "reason": "Different error classes and different code paths: custom trainer API change vs mask/tensor shape mismatch." }, { "left": "issue:43257", "right": "issue:45305", "accept": false, - "reason": "Qwen3 MoE weight conversion and DeepSpeed gradient averaging are separate training issues." + "reason": "Both touch DeepSpeed training, but one is a Qwen MoE weight-conversion problem and the other is a gradient-averaging issue." }, { "left": "issue:41553", "right": "issue:44488", "accept": false, - "reason": "These are different tokenizer-loading problems for different models; no shared concrete code-path is evident." + "reason": "Different tokenizer/loading issues for different models; not the same concrete bug." }, { "left": "issue:43872", "right": "issue:45538", "accept": false, - "reason": "A _is_hf_initialized constructor regression is unrelated to CLIPTokenizer's model_max_length default." + "reason": "bitsandbytes/_is_hf_initialized incompatibility is unrelated to CLIPTokenizer's max length default." }, { "left": "issue:44530", "right": "issue:44863", "accept": false, - "reason": "PagedAttentionCache linear_attention crash and NemotronH checkpoint loading are different subsystems." + "reason": "PagedAttentionCache linear_attention crash and NemotronH checkpoint loading are unrelated." }, { "left": "issue:41093", "right": "issue:44805", - "accept": true, - "reason": "Same indexed-tensor vs mask-length mismatch; 44805 looks like a duplicate report of the same error." + "accept": false, + "reason": "Both are index/shape errors, but they are distinct reports with no evidence of the same root cause." }, { "left": "issue:44479", "right": "issue:44560", - "accept": true, - "reason": "Both are Qwen VL video-input regressions in transformers 5.3.0; 44560 is a specific stopiteration symptom of the broader regression." + "accept": false, + "reason": "Both are video-related Qwen issues, but one is a broad v5.3.0 regression and the other is a specific StopIteration in embedding loading." }, { "left": "issue:24643", "right": "issue:30064", "accept": false, - "reason": "DeepSpeed training weight-shape error vs void segmentation-map processing are unrelated." + "reason": "Training with DeepSpeed and void segmentation-map processing are unrelated." }, { "left": "issue:43366", "right": "issue:43828", "accept": false, - "reason": "GGUF/gpt-oss support and a Phi-tiny-MoE dtype mismatch are unrelated." + "reason": "GPT-OSS architecture support and Phi-tiny-MoE autocast dtype mismatch are separate bugs." }, { "left": "issue:45663", "right": "issue:45823", "accept": false, - "reason": "Both are Gemma4 issues, but one is an FSDP2 shared_kv_states bug and the other is a device_map='auto' forward mismatch." + "reason": "Gemma4 FSDP2 shared_kv_states issue is not the same as the device_map=\"auto\" PLE mismatch." }, { "left": "issue:43366", "right": "issue:43531", "accept": false, - "reason": "GGUF gpt-oss support and Qwen3-MoE sliding_window behavior are unrelated." + "reason": "GPT-OSS support and Qwen3-MoE sliding_window behavior are different issues." }, { "left": "issue:43950", "right": "issue:45834", "accept": false, - "reason": "Silent non-persistent buffer corruption is unrelated to Kosmos2.5 OCR index errors." + "reason": "Non-persistent buffer corruption and Kosmos2.5 OCR index error are unrelated." }, { "left": "issue:43550", "right": "issue:44910", "accept": false, - "reason": "Bamba torch.compile/SDPA failure and Qwen3.5 flash-attention memory access are different bugs." + "reason": "Both mention attention/compile problems, but Bamba SDPA compile failure is unrelated to Qwen3.5 3D position_ids and flash-attention memory access." }, { "left": "issue:43645", "right": "issue:43646", "accept": true, - "reason": "Same custom model initialization regression in Transformers 5.0.0; 43645 just adds the Jupyter notebook context." + "reason": "Same Transformers 5 custom-model initialization regression; one report is just a more specific notebook variant." }, { "left": "issue:43366", "right": "issue:44589", "accept": false, - "reason": "gpt-oss GGUF support and missing Float8 storage type are unrelated." + "reason": "GPT-OSS support request and Float8 storage import failure are unrelated." }, { "left": "issue:43950", "right": "issue:45325", "accept": false, - "reason": "Non-persistent buffer corruption and Qwen2.5-VL rope scaling are different regressions." + "reason": "Non-persistent buffer corruption is unrelated to the Qwen2.5-VL rope-index scaling bug." }, { "left": "issue:44560", "right": "issue:44918", "accept": false, - "reason": "Both mention Qwen3.5, but one is a video-input regression and the other is unpacking embeddings with TRL SFT trainer." + "reason": "Both involve Qwen3.5 video/input handling, but the reported failures and likely code paths differ." }, { - "left": "issue:43824", - "right": "issue:43957", + "left": "issue:42617", + "right": "issue:43366", "accept": false, - "reason": "Missing Qwen2_5_VL export and meta-device loading failures are different issues." + "reason": "3d_parallel execution failure and GPT-OSS architecture support are unrelated." }, { - "left": "issue:42617", - "right": "issue:43366", + "left": "issue:43824", + "right": "issue:43957", "accept": false, - "reason": "A 3d_parallel.py runtime problem is unrelated to GGUF gpt-oss support." + "reason": "Missing Qwen2_5_VL export and `torch.device(\"meta\")` loading regressions are different bugs." }, { "left": "issue:42915", "right": "issue:43957", "accept": false, - "reason": "Qwen3Moe FP8 loading and meta-device model loading are unrelated." + "reason": "Qwen3Moe FP8 failure and meta-device loading regression do not share the same root cause." }, { "left": "issue:36010", "right": "issue:42915", "accept": false, - "reason": "GenerationMixin import failure and Qwen3Moe FP8 loading are unrelated." + "reason": "GenerationMixin import failure and Qwen3Moe FP8 failure are unrelated." }, { "left": "issue:44661", "right": "issue:45823", "accept": false, - "reason": "Tokenizer-mapping failure when adding a new model is unrelated to Gemma4 PLE device mismatch." + "reason": "Tokenizer-mapping registration bug and Gemma4 PLE device mismatch are different issues." }, { "left": "issue:38175", "right": "issue:43994", "accept": false, - "reason": "Both concern siglip2, but one is zero-probabilities behavior and the other is nonsensical outputs with AutoModel/pipeline; not clearly the same bug." + "reason": "Both mention SigLIP2, but one is zero probabilities and the other nonsensical AutoModel/pipeline outputs; not enough to treat as the same bug." }, { "left": "issue:43632", - "right": "issue:43872", - "accept": true, - "reason": "Both stem from the Transformers v5 _is_hf_initialized change; 43872 is the bitsandbytes failure mode of that regression." + "right": "issue:45834", + "accept": false, + "reason": "_is_hf_initialized flag regression and Kosmos2.5 OCR index error are unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "The analyst stayed conservative and the accepted duplicate pairs are well grounded by the packet titles: 43645/43646, 43872/44291, 41093/44805, and the Qwen video/DeepSpeed pairs are plausible within the same bug family. The summary also correctly avoids claiming the whole bucket is one coherent duplicate set." + "feedback": "Grounded and conservative overall. The two accepted soft pairs are plausible duplicate/regression matches, and the rejected pairs are appropriately cautious given the packet. Summary does not overstate the evidence." }, "evaluator_used": true, "retried": false, @@ -1653,8 +1677,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -1717,8 +1741,8 @@ "issue:38175|issue:39692", "issue:43644|issue:44534", "issue:45313|issue:45357", - "issue:45488|issue:45812", "issue:44960|issue:44977", + "issue:45488|issue:45812", "issue:44451|issue:44488", "issue:42491|issue:43720", "issue:43854|issue:43931", @@ -1736,17 +1760,17 @@ "issue:45362|issue:45406", "issue:42491|issue:45292", "issue:44488|issue:44991", - "issue:45276|issue:45335", - "issue:36010|issue:39404" + "issue:36010|issue:39404", + "issue:45276|issue:45335" ], - "prepared_review_unit_hash": "063b2ca454b689f48008dd2c5911e3b451ec8d83734f6fcebac87873c2f5fe0c", + "prepared_review_unit_hash": "87f2872331c988ef42b943937ff0372f82f64e6ea96041ad646db5fa51e88d3f", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, @@ -1761,207 +1785,201 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous, with only a couple of clear duplicate pairs. The strongest duplicate is the Transformers v5 non-persistent-buffer regression; a second plausible duplicate is the resize_token_embeddings failure across model variants.", - "confidence": 0.79, - "canonical_issue_reason": "No cluster-wide canonical issue: the node set spans many unrelated bugs. Among the duplicate candidates, the non-persistent-buffer regression is the clearest shared issue.", + "summary": "Most items are unrelated bug reports. The clearest duplicate remains the non-persistent-buffer regression pair (43644/44534). The `resize_token_embeddings` issues (45276/45335) are plausibly the same propagation bug, but `43644/43645` should be treated conservatively as non-duplicate without body evidence.", + "confidence": 0.81, + "canonical_issue_reason": "Issue 43644 is the best canonical issue: it is a precise, well-scoped regression report and is an exact duplicate of 44534.", "canonical_pr_reason": null, - "best_issue_reason": "44534 is the best representative issue: it states the Transformers v5 regression succinctly and broadly, and is the clearest canonical wording of the duplicate pair with 43644.", + "best_issue_reason": "43644 is the strongest representative issue overall because it has the clearest duplicate evidence and the most specific regression framing.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44155", "right": "issue:45584", "accept": false, - "reason": "Different audio bugs: batched inference token leakage vs Whisper empty-transcription handling." + "reason": "Different audio/model bugs: batched inference track leakage vs Whisper empty-transcription behavior." }, { "left": "issue:44877", "right": "issue:45030", "accept": false, - "reason": "Both are config-validation failures, but for different models and different failure modes." + "reason": "Both involve config loading, but the models and failure modes differ; no clear shared code path." }, { "left": "issue:44560", "right": "issue:44805", "accept": false, - "reason": "Unrelated symptoms: Qwen3-VL video StopIteration vs a generic mask/tensor shape mismatch." + "reason": "Unrelated symptoms and models: video StopIteration in Qwen3-VL vs a mask shape IndexError." }, { "left": "issue:43638", "right": "issue:44661", "accept": false, - "reason": "Different code paths: ZeRO-3/Bert initialization vs tokenizer mapping in add-new-model-like." + "reason": "Deepspeed/BERT zero3 init bug is unrelated to tokenizer-mapping registration failure." }, { "left": "issue:43317", "right": "issue:44756", "accept": false, - "reason": "Both touch loading/offload memory, but one is a dequantized device_map failure and the other is a Strix Halo mmap OOM." + "reason": "Both touch offloading/memory, but one is a dequantized load failure and the other is a Strix Halo mmap OOM fix." }, { "left": "issue:43596", "right": "issue:44661", "accept": false, - "reason": "Different bugs: ZeRO-3 Bert index error vs tokenizer mapping-name handling." + "reason": "Different root causes: zero3/BertModel IndexError vs `add-new-model-like` tokenizer-mapping logic." }, { "left": "issue:44661", "right": "issue:44805", "accept": false, - "reason": "No shared underlying bug; tokenizer mapping logic and mask-shape indexing are unrelated." + "reason": "Tokenizer-mapping registration problem vs a tensor mask shape mismatch are not the same bug." }, { "left": "issue:38175", "right": "issue:39692", "accept": false, - "reason": "Same model family, but one is zero probabilities and the other is a doc-example/model-processor mismatch plus quantization failure." + "reason": "Same model family, but one is bad output probabilities and the other is a docs/example mismatch plus quantization failure." }, { "left": "issue:43644", "right": "issue:44534", "accept": true, - "reason": "Same Transformers v5 regression: non-persistent buffers are filled with junk." + "reason": "Near-identical titles describing the same non-persistent-buffer junk regression; clearly the same underlying bug." }, { "left": "issue:45313", "right": "issue:45357", "accept": false, - "reason": "Both are Qwen3.5 regressions, but one is ZeRO-3 weight loading and the other is bad save_pretrained visual-encoder keys." + "reason": "Load-time ZeRO-3 weight issue and save_pretrained visual-encoder key regression are different code paths." }, { - "left": "issue:45488", - "right": "issue:45812", + "left": "issue:44960", + "right": "issue:44977", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different model families and different concrete tokenization paths." + "reason": "Different models and different flash-attention failures; not the same concrete issue." }, { - "left": "issue:44960", - "right": "issue:44977", + "left": "issue:45488", + "right": "issue:45812", "accept": false, - "reason": "Different models and failures; GLM5 is not the same bug as Qwen3.5 flash-attention generation." + "reason": "Both are tokenizer regressions, but they affect different tokenization layers and different model families." }, { "left": "issue:44451", "right": "issue:44488", "accept": false, - "reason": "Both are model-loading failures, but there is not enough evidence they share the same root cause." + "reason": "Both are model-loading failures, but the titles point to different repos and no shared concrete bug is evident." }, { "left": "issue:42491", "right": "issue:43720", "accept": false, - "reason": "Different domains: Qwen3 MoE LoRA compatibility vs BitNet packed-weight loading." + "reason": "Both involve loading/compatibility, but one is Qwen3 MoE LoRA cross-version use and the other is BitNet packed-weight handling." }, { "left": "issue:43854", "right": "issue:43931", "accept": false, - "reason": "Different model-loading bugs for different architectures and failure types." + "reason": "Different model loading failures with different symptoms and likely different fixes." }, { "left": "issue:41720", "right": "issue:42915", "accept": false, - "reason": "Both involve Qwen3-family failures, but one is device_map/cuda assert and the other is FineGrainedFP8Config." + "reason": "Qwen3 auto device mapping CUDA assert is not the same as Qwen3Moe FineGrainedFP8Config failure." }, { "left": "issue:43708", "right": "issue:44075", "accept": false, - "reason": "Unrelated training scheduling vs optimizer-argument usage bug." + "reason": "Trainer checkpoint resume step calculation and SGD optimizer args being ignored are unrelated." }, { "left": "issue:43257", "right": "issue:43854", "accept": false, - "reason": "Different subsystems and models: Qwen3 MoE accelerate/deepspeed loading vs GLM-4.7 test loading." + "reason": "Different models and different loading paths; no evidence of a common bug." }, { "left": "issue:42491", "right": "issue:45276", "accept": false, - "reason": "Different issues: Qwen3 MoE LoRA compatibility vs Gemma4 embedding resizing." + "reason": "LoRA compatibility across versions is unrelated to `resize_token_embeddings` not updating embeddings." }, { "left": "issue:43856", "right": "issue:43873", "accept": false, - "reason": "Both mention memory, but one is Qwen3 MoE training efficiency and the other is quantization offload behavior." + "reason": "Training memory inefficiency and quantized offload behavior are different problems." }, { "left": "issue:42915", "right": "issue:44756", "accept": false, - "reason": "Different memory-related failures with no shared concrete code path." + "reason": "Qwen3Moe FP8 failure on one side and Strix Halo mmap OOM on the other are unrelated." }, { "left": "issue:42491", "right": "issue:43931", "accept": false, - "reason": "Different model families and bugs: Qwen3 MoE LoRA compatibility vs Qwen3-VL shape mismatch." + "reason": "Different models, different mechanisms, and no shared concrete code-path issue." }, { "left": "issue:39692", "right": "issue:43994", "accept": false, - "reason": "Same SigLIP2 family, but the reported bug details differ enough that this does not look like one duplicate." + "reason": "Same SigLIP2 family, but one is docs/example misuse plus quantization failure while the other is an inference regression." }, { "left": "issue:43856", "right": "issue:44387", "accept": false, - "reason": "Both are memory problems, but one is training inefficiency and the other is int4 reserved-memory OOM regression." + "reason": "Both mention memory, but the causes differ: MoE training efficiency vs increased CUDA reserved memory under int4 quantization." }, { "left": "issue:43644", "right": "issue:43645", "accept": false, - "reason": "Both are v5 regressions, but one is buffer initialization and the other is notebook custom-model initialization." + "reason": "The titles both mention Transformers 5.0.0 regressions, but one is about non-persistent buffers and the other is about custom models in Jupyter notebooks; the packet does not show the same underlying bug." }, { "left": "issue:44910", "right": "issue:45588", "accept": false, - "reason": "Different flash-attention bugs: packed-sequence position_ids vs sink-less-model s_aux handling." + "reason": "Both are flash-attention crashes, but one is a packed-sequence misinterpretation and the other is an `s_aux=None` attribute error." }, { "left": "issue:45362", "right": "issue:45406", "accept": false, - "reason": "Different models and failures: Qwen3.5 chat crash vs Gemma4Processor attribute error in serve." + "reason": "Different crash points: Qwen3.5 chat generation vs Gemma4Processor missing `_tokenizer`." }, { - "left": "issue:42491", - "right": "issue:45292", + "left": "issue:44488", + "right": "issue:44991", "accept": false, - "reason": "Different model-specific loading/compatibility bug vs generic resize_token_embeddings behavior." + "reason": "Both are tokenizer/model loading failures in v5, but the specific repos and failure modes are too different to call duplicates." }, { - "left": "issue:44488", - "right": "issue:44991", + "left": "issue:36010", + "right": "issue:39404", "accept": false, - "reason": "Both are tokenizer-loading regressions, but the evidence only shows separate model-specific failures." + "reason": "Importing `GenerationMixin` and Whisper `return_language` pipeline behavior are unrelated." }, { "left": "issue:45276", "right": "issue:45335", "accept": true, - "reason": "Same underlying resize_token_embeddings regression: embeddings are not updated on model-specific output/decoder embeddings." - }, - { - "left": "issue:36010", - "right": "issue:39404", - "accept": false, - "reason": "Unrelated issues: a GenerationMixin import error vs Whisper pipeline return_language behavior." + "reason": "Both report the same `resize_token_embeddings` propagation bug to model-specific embedding/output modules; one fix could plausibly cover both." } ] }, "evaluator_result": { - "accept": true, - "feedback": "Grounded and conservative overall. The cluster is clearly heterogeneous, and the two accepted duplicate pairs are well supported by the issue titles: 43644/44534 share the same Transformers v5 non-persistent-buffer regression, and 45276/45335 share the resize_token_embeddings regression across model variants. I did not see any soft-edge verdicts that overstate the evidence." + "accept": false, + "feedback": "The cluster summary is mostly grounded, but one soft-edge acceptance is too strong: `issue:43644` vs `issue:43645` is not supported by the packet. `43644`/`44534` clearly match on the non-persistent-buffer junk regression, but `43645` is titled about custom models in Jupyter notebooks and does not obviously describe the same bug. That pair should stay conservative/rejected unless there is direct evidence in the issue bodies. The `resize_token_embeddings` pair looks plausibly related, but the `43644`/`43645` acceptance overstates the evidence." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -1974,11 +1992,12 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ + "issue:39404", "issue:39692", "issue:41720", "issue:42947", @@ -1994,7 +2013,6 @@ "issue:43828", "issue:43854", "issue:43873", - "issue:43906", "issue:43957", "issue:44155", "issue:44164", @@ -2005,7 +2023,6 @@ "issue:44451", "issue:44479", "issue:44488", - "issue:44530", "issue:44589", "issue:44683", "issue:44756", @@ -2022,6 +2039,7 @@ "issue:45663", "issue:45684", "issue:45698", + "issue:45701", "issue:45704", "issue:45759", "issue:45823", @@ -2033,8 +2051,8 @@ "issue:44402|issue:44488", "issue:43425|issue:43957", "issue:39692|issue:44960", - "issue:43335|issue:45663", "issue:42947|issue:45704", + "issue:43335|issue:45663", "issue:43873|issue:44387", "issue:45663|issue:45759", "issue:45663|issue:45859", @@ -2048,8 +2066,8 @@ "issue:45276|issue:45292", "issue:43381|issue:44928", "issue:44164|issue:44756", - "issue:41720|issue:44155", "issue:44479|issue:44877", + "issue:41720|issue:44155", "issue:43646|issue:43708", "issue:45588|issue:45663", "issue:43854|issue:44960", @@ -2062,246 +2080,252 @@ "issue:43425|issue:43828", "issue:44589|issue:44756", "issue:44683|issue:45823", - "issue:43906|issue:44530" + "issue:39404|issue:45701" ], - "prepared_review_unit_hash": "bdece9ceb2d966be418903dee166308c8f9e29bdd0365a29497661cb2bb95d82", + "prepared_review_unit_hash": "0090eedf181c34281bdbe70ecdf4f93a763a26e3850d31d49547d1f25a16289b", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 35, - "serialized_chars": 37684, - "estimated_input_tokens": 9421, - "estimated_eval_tokens": 19098 + "serialized_chars": 37609, + "estimated_input_tokens": 9403, + "estimated_eval_tokens": 19062 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is mostly a grab bag of unrelated model-loading, quantization, and training bugs. The only clear duplicate-style matches are the `resize_token_embeddings` regressions, where the generic issue and the model-specific reports look like the same underlying bug.", - "confidence": 0.87, - "canonical_issue_reason": "issue:45292 is the best canonical issue because it states the underlying `resize_token_embeddings` bug in the most general form (`output_embeddings` not updated), while the other matching reports are model-specific variants of the same failure.", + "summary": "This cluster is mostly heterogeneous model-loading / quantization / training bugs. Only the resize_token_embeddings issues look like true duplicates; most other soft edges are same subsystem at best, but not the same concrete bug.", + "confidence": 0.93, + "canonical_issue_reason": "issue:45292 is the broadest, most generic representative of the only clear duplicate set (`resize_token_embeddings` not updating output/decoder embeddings).", "canonical_pr_reason": null, - "best_issue_reason": "issue:45292 is the best global issue representative for the duplicate cluster, since it captures the shared `resize_token_embeddings` regression more broadly than the model-specific variants.", + "best_issue_reason": "issue:45292 best represents the cluster\u2019s one real duplicate thread and is the most general formulation of that bug.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43404", "right": "issue:43454", "accept": false, - "reason": "Both are lm_head tying bugs, but they hit different model classes and likely different code paths; too broad to treat as the same duplicate." + "reason": "Same symptom class (lm_head tying) but different model families and likely separate fixes." }, { "left": "issue:44164", "right": "issue:44291", "accept": false, - "reason": "One is about `extra_state` in save/load, the other about `init_empty_weights` passing `_is_hf_initialized`; related area, different bug." + "reason": "Both touch loading/saving paths, but one is extra_state handling and the other is init_empty_weights argument compatibility." }, { "left": "issue:44402", "right": "issue:44488", "accept": false, - "reason": "Both are model loading failures, but the titles point to different models and different root causes." + "reason": "Both report load failures for different BERT models, but the titles point to separate model/tokenizer mismatches rather than one bug." }, { "left": "issue:43425", "right": "issue:43957", "accept": false, - "reason": "Torch version incompatibility is not the same as meta-device loading failures; different failure modes." + "reason": "Unrelated failures: Torch version incompatibility vs meta-device loading regression." }, { "left": "issue:39692", "right": "issue:44960", "accept": false, - "reason": "Completely different models and symptoms; no shared code-path evidence." + "reason": "Documentation example errors versus a GLM5 loading issue; no shared underlying bug." }, { - "left": "issue:43335", - "right": "issue:45663", + "left": "issue:42947", + "right": "issue:45704", "accept": false, - "reason": "Different architectures and failures; only a superficial theme of configuration/runtime bugs." + "reason": "Gradient checkpointing/LoRA behavior is unrelated to the T5 apex RMSNorm memory leak." }, { - "left": "issue:42947", - "right": "issue:45704", + "left": "issue:43335", + "right": "issue:45663", "accept": false, - "reason": "Gradient checkpointing/LoRA is unrelated to a T5 RMSNorm memory leak." + "reason": "SwitchTransformers sparse-layer creation bug is unrelated to Gemma4 FSDP shared_kv_states handling." }, { "left": "issue:43873", "right": "issue:44387", "accept": false, - "reason": "Both involve quantization and OOMs, but one is offloading behavior and the other is reserved-memory growth under int4; not the same bug." + "reason": "Both involve quantization/OOM, but one is offloading behavior and the other is reserved-memory growth under int4; different code paths." }, { "left": "issue:45663", "right": "issue:45759", "accept": false, - "reason": "Different features: FSDP2 shared_kv_states rebuild versus multimodal config unwrapping." + "reason": "Gemma4 FSDP attention state rebuild is unrelated to Qwen multimodal from_config text_config unwrapping." }, { "left": "issue:45663", "right": "issue:45859", "accept": false, - "reason": "Different Qwen/Gemma runtime paths and different failures; not the same underlying bug." + "reason": "Different models and failures: Gemma4 FSDP KeyError vs Qwen3_5MoE CPU-offload incompatibility." }, { "left": "issue:44155", "right": "issue:45823", "accept": false, - "reason": "Both are forward-time bugs in multimodal models, but the reported mechanisms are different and not mergeable as one fix." + "reason": "AudioFlamingo batched inference token leak is unrelated to Gemma4 device-mapping mismatch." }, { "left": "issue:43854", "right": "issue:45216", "accept": false, - "reason": "Different model families and different bugs: unit-test loading vs incorrect save_pretrained checkpoint handling." + "reason": "Different model-loading regressions with no shared concrete failure mode." }, { "left": "issue:44387", "right": "issue:44756", "accept": false, - "reason": "Both mention OOM, but the causes are unrelated." + "reason": "OOM-related, but one is int4 quantization memory reservation and the other is mmap on Strix Halo." }, { "left": "issue:44291", "right": "issue:44756", "accept": false, - "reason": "`init_empty_weights` loading error and Strix Halo mmap OOM are unrelated." + "reason": "init_empty_weights compatibility issue is unrelated to mmap/OOM behavior." }, { "left": "issue:44292", "right": "issue:44912", "accept": false, - "reason": "Different quantization formats, models, and symptoms; not the same concrete issue." + "reason": "Both are quantization-related, but they involve different quantization formats and different models." }, { "left": "issue:42947", "right": "issue:43381", "accept": false, - "reason": "One says checkpointing is ineffective, the other says it cannot be used in eval mode; different behaviors." + "reason": "Both mention gradient checkpointing, but one is ineffective LoRA behavior and the other is an eval-mode restriction." }, { "left": "issue:45684", "right": "issue:45698", "accept": false, - "reason": "Both concern custom-model save/load, but one is permissions propagation and the other is wrong-module reloading." + "reason": "Custom-model save/load issues, but permissions propagation and wrong-module loading are distinct bugs." }, { "left": "issue:45276", "right": "issue:45292", "accept": true, - "reason": "Same underlying `resize_token_embeddings` bug; 45276 is the Gemma4-specific manifestation of the broader 45292 report." + "reason": "Same underlying resize_token_embeddings bug: output embeddings are not updated, with 45276 being a model-specific variant." }, { "left": "issue:43381", "right": "issue:44928", "accept": false, - "reason": "Gradient checkpointing in eval mode is unrelated to the Qwen3.5 RLHF NaN/SDPA fallback issue." + "reason": "Gradient checkpointing eval-mode limitation is unrelated to Qwen3.5 RLHF gradient explosion." }, { "left": "issue:44164", "right": "issue:44756", "accept": false, - "reason": "Save/load `extra_state` handling is unrelated to mmap-induced OOM." + "reason": "save/from_pretrained extra_state handling is unrelated to Strix Halo mmap OOM." }, { - "left": "issue:41720", - "right": "issue:44155", + "left": "issue:44479", + "right": "issue:44877", "accept": false, - "reason": "Device-map CUDA assert and AudioFlamingo3 token leak are different bugs." + "reason": "Video-input regression across Qwen variants versus strict config loading for granite_speech; different problems." }, { - "left": "issue:44479", - "right": "issue:44877", + "left": "issue:41720", + "right": "issue:44155", "accept": false, - "reason": "Both are regression-style multimodal issues, but they affect different models and different subsystems." + "reason": "Auto device mapping cudaErrorAssert and cross-track token leak are unrelated." }, { "left": "issue:43646", "right": "issue:43708", "accept": false, - "reason": "Custom model initialization breakage and Trainer checkpoint step calculation are unrelated." + "reason": "Custom model initialization breakage is unrelated to Trainer max_steps recomputation on resume." }, { "left": "issue:45588", "right": "issue:45663", "accept": false, - "reason": "Flash-attention sink handling and Gemma4 FSDP2 shared state rebuilding are unrelated." + "reason": "Flash-attention s_aux=None crash is unrelated to Gemma4 FSDP state rebuilding." }, { "left": "issue:43854", "right": "issue:44960", "accept": false, - "reason": "Both are GLM-related, but one is a specific load/test issue and the other is a broad GLM5 report." + "reason": "Different GLM-related failures with no evidence they are the same bug." + }, + { + "left": "issue:43957", + "right": "issue:44756", + "accept": false, + "reason": "Meta-device loading regression is unrelated to the Strix Halo mmap workaround." }, { "left": "issue:45292", "right": "issue:45335", "accept": true, - "reason": "Same `resize_token_embeddings` regression; 45335 is the T5Gemma-specific manifestation of the broader output-embedding bug." + "reason": "Same resize_token_embeddings bug: output_embeddings and decoder.embed_tokens are both not refreshed." }, { "left": "issue:44402", "right": "issue:44451", "accept": false, - "reason": "Different model load failures with no evidence of a shared root cause." + "reason": "Both are model load failures, but the titles indicate distinct model-specific issues." }, { "left": "issue:43828", "right": "issue:43957", "accept": false, - "reason": "Autocast dtype mismatch and meta-device loading failures are different issues." + "reason": "Autocast dtype mismatch for Phi MoE is unrelated to meta-device loading failures." }, { "left": "issue:43645", "right": "issue:44164", "accept": false, - "reason": "Jupyter/custom-model initialization breakage is not the same as `extra_state` save/load handling." + "reason": "Notebook custom-model initialization and extra_state save/load handling are different failures." }, { "left": "issue:43257", "right": "issue:44863", "accept": false, - "reason": "Both are checkpoint/loading problems, but the models and failure mechanisms differ." + "reason": "Qwen3 MOE deepspeed conversion issue is unrelated to NemotronH checkpoint loading." }, { "left": "issue:43425", "right": "issue:43828", "accept": false, - "reason": "Torch version incompatibility is not the same as the Phi-tiny-MoE autocast dtype mismatch." + "reason": "Torch version incompatibility is unrelated to Phi-tiny-MoE autocast dtype mismatch." }, { "left": "issue:44589", "right": "issue:44756", "accept": false, - "reason": "Float8 storage deserialization and mmap OOM are unrelated." + "reason": "Float8 storage lookup failure is unrelated to Strix Halo mmap OOM." }, { "left": "issue:44683", "right": "issue:45823", "accept": false, - "reason": "Compiled flex_attention on newer torch and Gemma4 device mismatch are unrelated." + "reason": "Compiled flex_attention on torch>=2.9 is unrelated to Gemma4 PLE device mismatch." }, { - "left": "issue:43906", - "right": "issue:44530", + "left": "issue:39404", + "right": "issue:45701", "accept": false, - "reason": "`43906` is only a reproduction of an earlier issue and does not clearly match the specific Qwen3.5 `linear_attention` crash." + "reason": "Whisper pipeline return_language regression is unrelated to tokenization changes in a different issue." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet: issue:45292 is a reasonable canonical representative for the resize_token_embeddings regression, and the accepted soft edges to issue:45276 and issue:45335 are conservative given the very similar titles and the shared output_embedding/decoder.embed_tokens symptom. The other pairwise rejections are generally cautious and do not overstate evidence." + "feedback": "The summary is grounded in the packet and stays appropriately conservative. The only clear duplicate cluster is the resize_token_embeddings set (45276/45292/45335), and the other soft-edge rejections are justified by the titles being model- or bug-specific rather than clearly mergeable into one PR." }, "evaluator_used": true, "retried": false, @@ -2317,8 +2341,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -2341,6 +2365,7 @@ "issue:43716", "issue:43856", "issue:43901", + "issue:43906", "issue:43931", "issue:43994", "issue:44079", @@ -2366,13 +2391,12 @@ "issue:44991", "issue:45081", "issue:45084", - "issue:45701", "issue:45759", "issue:45823", "issue:45859" ], "soft_pairs": [ - "issue:39404|issue:45701", + "issue:43906|issue:44530", "issue:43425|issue:45823", "issue:43611|issue:43645", "issue:43425|issue:43716", @@ -2388,8 +2412,8 @@ "issue:43856|issue:44756", "issue:40990|issue:42947", "issue:39404|issue:44671", - "issue:43994|issue:44079", "issue:44164|issue:45759", + "issue:43994|issue:44079", "issue:36032|issue:43452", "issue:43596|issue:44683", "issue:44671|issue:44991", @@ -2399,239 +2423,242 @@ "issue:44164|issue:44479", "issue:44488|issue:45081", "issue:44661|issue:44683", - "issue:40990|issue:43421", "issue:43582|issue:43632", + "issue:40990|issue:43421", "issue:43611|issue:43644", "issue:44206|issue:44220", "issue:43716|issue:45084" ], - "prepared_review_unit_hash": "ed6af37e5c8fe762aa29e551cc635bf57d571b69fd9db788c6fe29b86914d769", + "prepared_review_unit_hash": "1860ce3a02a5bc3e021b404266db3628f7bbf43aad00c44fcbdd0c99c57769fb", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37279, - "estimated_input_tokens": 9320, - "estimated_eval_tokens": 18896 + "serialized_chars": 37316, + "estimated_input_tokens": 9329, + "estimated_eval_tokens": 18914 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is heterogeneous: most items are unrelated Transformers regressions or feature requests, and none of the soft pairs look like the same underlying bug/change.", - "confidence": 0.63, - "canonical_issue_reason": "issue:43611 is the best representative anchor because it is a concrete v5 model-loading regression with a clear code path and broader scope than the more model-specific reports, though the overall cluster is not a true duplicate set.", + "summary": "Mostly a grab bag of unrelated Transformers regressions and docs bugs. Only the audio feature-extraction pair looks plausibly duplicate; otherwise the soft edges are too broad or only share a subsystem. I\u2019d anchor the cluster on the generic v5 model-loading regression around `base_model_prefix`.", + "confidence": 0.46, + "canonical_issue_reason": "issue:43611 is the cleanest representative of the v5 loading regressions: generic, concrete, and closer to the common model-initialization/loading failure mode than the more specialized neighboring reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43611 is the strongest issue to keep as the cluster anchor: it is specific enough to be actionable, but broad enough to represent the dominant v5 loading-regression theme in this noisy group.", + "best_issue_reason": "issue:43611 is the best overall issue to keep as the canonical issue because it is broad enough to represent the cluster\u2019s main v5 model-loading breakage without being overly model-specific.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:39404", - "right": "issue:45701", + "left": "issue:43906", + "right": "issue:44530", "accept": false, - "reason": "Whisper pipeline return_language regression and a tokenization-behavior report are different problems and code paths." + "reason": "Both involve Qwen-related behavior, but one is an isolated reproduction of a prior issue and the other is a PagedAttentionCache crash on Qwen3.5; different failures and code paths." }, { "left": "issue:43425", "right": "issue:45823", "accept": false, - "reason": "Torch 2.10 incompatibility and Gemma4 device-mismatch during forward are unrelated failures." + "reason": "Torch version incompatibility vs a Gemma4 PLE device-mismatch forward bug; unrelated symptoms and fixes." }, { "left": "issue:43611", "right": "issue:43645", "accept": false, - "reason": "base_model_prefix loading regression and Jupyter custom-model initialization issues are distinct bugs." + "reason": "Both are v5 regressions, but `base_model_prefix` loading failure and custom-model-in-Jupyter initialization are distinct underlying problems." }, { "left": "issue:43425", "right": "issue:43716", "accept": false, - "reason": "A torch version compatibility issue is not the same as a multimodal dtype mismatch." + "reason": "Torch 2.10 incompatibility is unrelated to the Mistral-3 image preprocessor dtype mismatch." }, { "left": "issue:42915", "right": "issue:44863", "accept": false, - "reason": "Different model families and different loading failures; no shared concrete bug." + "reason": "Different model families and failures: FineGrainedFP8Config on Qwen3Moe vs checkpoint-loading failure in NemotronH." }, { "left": "issue:44530", "right": "issue:45084", "accept": false, - "reason": "PagedAttentionCache linear_attention crash and template compilation error are separate code paths." + "reason": "Qwen3.5 cache crash vs template compilation error; no shared underlying bug." }, { "left": "issue:43901", "right": "issue:44509", "accept": false, - "reason": "Docs mentioning return_all_scores and docs mentioning removed pipeline tasks are different documentation issues." + "reason": "Both are docs-related, but one is about `return_all_scores` and the other is about removed pipeline task docs; not the same issue." }, { "left": "issue:39404", "right": "issue:44912", "accept": false, - "reason": "Whisper return_language behavior and MXFP4 quantization fallback are unrelated." + "reason": "Whisper pipeline language return regression and gpt-oss quantization/perplexity are unrelated." }, { "left": "issue:44265", "right": "issue:44898", "accept": false, - "reason": "torch.export / torch_compilable_check failure and Perceiver non-default resolution failure are different bugs." + "reason": "`torch.export`/`torch_compilable_check` failure and Perceiver resolution/interpolation failure are different bugs." }, { "left": "issue:43299", "right": "issue:43931", "accept": false, - "reason": "Qwen3VL-MoE loading breakage and Qwen3-VL-30B weight-shape mismatch are not the same issue." + "reason": "Both concern Qwen3-VL loading, but one is a dev-version breakage for MoE models and the other is a weight-shape mismatch on a specific checkpoint; not enough to treat as the same bug." }, { "left": "issue:44479", "right": "issue:44805", "accept": false, - "reason": "Video-input regression for Qwen VL models and a mask-shape IndexError are not the same underlying bug." + "reason": "Video-input regression across Qwen VL variants vs a generic mask-shape IndexError; different failure modes." }, { "left": "issue:43425", "right": "issue:44918", "accept": false, - "reason": "Torch compatibility and Qwen3.5 embedding unpacking in TRL are unrelated." + "reason": "Torch compatibility and TRL SFT input-embedding unpacking are unrelated." }, { "left": "issue:44589", "right": "issue:45084", "accept": false, - "reason": "Float8 storage lookup failure and non-template-node compile error are different failures." + "reason": "Float8 storage lookup failure is not the same as the template-node compilation error." }, { "left": "issue:43856", "right": "issue:44756", "accept": false, - "reason": "Qwen3 MoE training memory usage and mmap OOM on Strix Halo are different performance problems." + "reason": "Both mention memory/OOM, but one is Qwen3 MoE training memory inefficiency and the other is mmap behavior on Strix Halo; different root causes." }, { "left": "issue:40990", "right": "issue:42947", "accept": false, - "reason": "High perplexity on gpt-oss and gradient checkpointing ineffectiveness with LoRA are unrelated." + "reason": "Perplexity regression on gpt-oss-20b and gradient-checkpointing ineffectiveness with LoRA are separate issues." }, { "left": "issue:39404", "right": "issue:44671", "accept": false, - "reason": "Whisper pipeline regression and CamemBERT masked-LM prediction regression are different model issues." + "reason": "Whisper pipeline language return regression vs CamemBERT masked-LM prediction regression; unrelated model paths." }, { - "left": "issue:43994", - "right": "issue:44079", + "left": "issue:44164", + "right": "issue:45759", "accept": false, - "reason": "SigLIP2 bad outputs and ModelOutput key assignment with None are unrelated." + "reason": "`extra_state` serialization/loading and `text_config` unwrapping for composite Qwen configs are distinct loader issues." }, { - "left": "issue:44164", - "right": "issue:45759", + "left": "issue:43994", + "right": "issue:44079", "accept": false, - "reason": "extra_state save/load handling and Qwen multimodal text_config unwrapping are different serialization paths." + "reason": "SigLIP2 bad outputs with AutoModel/pipeline and ModelOutput key assignment when a value was previously None are different bugs." }, { "left": "issue:36032", "right": "issue:43452", "accept": false, - "reason": "Tokenizer add_special_tokens conflict and gguf_file loading breakage are distinct tokenizer/loading bugs." + "reason": "T5 tokenizer special-token method conflict and gguf-related from_pretrained breakage are unrelated loading failures." }, { "left": "issue:43596", "right": "issue:44683", "accept": false, - "reason": "DeepSpeed zero3 BertModel IndexError and compiled flex_attention torch-version failure are unrelated." + "reason": "DeepSpeed ZeRO3/BertModel index error and compiled flex_attention failure on newer torch versions do not share the same root cause." }, { "left": "issue:44671", "right": "issue:44991", "accept": false, - "reason": "CamemBERT prediction regression and tokenizer loading failure for a different checkpoint are not the same bug." + "reason": "CamemBERT incorrect predictions and tokenizer loading failure for EMBEDDIA/est-roberta are different issues." }, { "left": "issue:44991", "right": "issue:45081", "accept": false, - "reason": "Different tokenizer regressions with different failure modes and different affected models." + "reason": "Both are tokenizer-loading regressions, but one is model-specific tokenizer loading for EMBEDDIA/est-roberta and the other is Mistral regex patching; not the same bug." }, { "left": "issue:44737", "right": "issue:45859", "accept": false, - "reason": "XLNet CPU device placement bug and CPU-offload incompatibility in Qwen3_5Moe are separate issues." + "reason": "Both involve device placement/offload concerns, but XLNet CPU `arange` work and Qwen3_5Moe rotary embedding CPU-offload incompatibility are separate code paths." }, { "left": "issue:32090", "right": "issue:39404", "accept": false, - "reason": "Trainer _gpu_broadcast_one NoneType error and Whisper return_language regression are unrelated." + "reason": "Trainer GPU broadcast TypeError and Whisper `return_language` regression are unrelated." }, { "left": "issue:44164", "right": "issue:44479", "accept": false, - "reason": "Serialization extra_state handling and video input regression are different code paths." + "reason": "Serialization/loading of extra state is unrelated to Qwen VL video-input regression." }, { "left": "issue:44488", "right": "issue:45081", "accept": false, - "reason": "Loading cjvt/sleng-bert and Mistral regex patching failures are distinct tokenizer issues." + "reason": "Loading `cjvt/sleng-bert` and the Mistral regex patch crash are different tokenizer issues." }, { "left": "issue:44661", "right": "issue:44683", "accept": false, - "reason": "Tokenizer mapping-name failure and compiled flex_attention failure are unrelated." + "reason": "`add-new-model-like`/mapping-name handling and compiled flex_attention are unrelated." }, { - "left": "issue:40990", - "right": "issue:43421", + "left": "issue:43582", + "right": "issue:43632", "accept": false, - "reason": "High perplexity on a model and runtime post-processor updates in TokenizersBackend are not the same issue." + "reason": "Apple Silicon allocator warmup TypeError and `_is_hf_initialized` flag breakage are different regressions." }, { - "left": "issue:43582", - "right": "issue:43632", + "left": "issue:40990", + "right": "issue:43421", "accept": false, - "reason": "Apple Silicon caching_allocator_warmup TypeError and _is_hf_initialized flag breakage are different failures." + "reason": "Perplexity regression and runtime special-token post-processor update behavior are not the same bug." }, { "left": "issue:43611", "right": "issue:43644", "accept": false, - "reason": "Model loading with base_model_prefix and junk-filled non-persistent buffers are different regressions." + "reason": "Both are v5 model-loading regressions, but `base_model_prefix` handling and non-persistent buffer initialization are separate underlying defects." }, { "left": "issue:44206", "right": "issue:44220", - "accept": false, - "reason": "Unsupported center argument in LasrFeatureExtractor and _torch_extract_fbank_features are related to audio preprocessing but not clearly the same concrete bug." + "accept": true, + "reason": "These both point at the same audio feature-extraction path: a regression in `LasrFeatureExtractor`/`_torch_extract_fbank_features` around the same internal function call and crash mode." }, { "left": "issue:43716", "right": "issue:45084", "accept": false, - "reason": "Mistral-3 dtype mismatch and non-template-node compilation error are unrelated." + "reason": "Mistral-3 image preprocessor dtype mismatch and template-node compilation errors are unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet and stays conservative on the soft edges. The only accepted pair, issue:44206 and issue:44220, looks plausibly like the same audio feature-extraction bug and mergeable into one fix. Anchoring on issue:43611 as the canonical v5 loading regression is also reasonable given the surrounding reports are mostly model-specific or unrelated." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -2643,8 +2670,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -2668,10 +2695,10 @@ "issue:43644", "issue:43645", "issue:43646", + "issue:43742", "issue:43819", "issue:43883", "issue:43994", - "issue:44062", "issue:44220", "issue:44265", "issue:44291", @@ -2700,8 +2727,8 @@ "soft_pairs": [ "issue:43883|issue:44488", "issue:44488|issue:44987", - "issue:43643|issue:44530", "issue:34567|issue:40444", + "issue:43643|issue:44530", "issue:44811|issue:45584", "issue:43644|issue:43646", "issue:44451|issue:44987", @@ -2726,266 +2753,269 @@ "issue:42915|issue:45084", "issue:44451|issue:45081", "issue:45216|issue:45357", - "issue:44265|issue:44610", "issue:44987|issue:45081", "issue:44220|issue:44683", + "issue:44265|issue:44610", "issue:43404|issue:45127", "issue:43296|issue:43531", "issue:43632|issue:44756", "issue:42915|issue:44291", - "issue:44062|issue:44987" + "issue:36032|issue:43742" ], - "prepared_review_unit_hash": "c5a3b15716260e725c6382fc5e208214d22dd9e0dbb4f62050bd038363741a97", + "prepared_review_unit_hash": "453af618e3a203eec6efd592b0bf03a1d19af376cdcd30578caa3e95087322f8", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 36, - "serialized_chars": 38560, - "estimated_input_tokens": 9640, - "estimated_eval_tokens": 19536 + "serialized_chars": 38525, + "estimated_input_tokens": 9632, + "estimated_eval_tokens": 19520 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a collection of separate Transformers regressions around loading, tokenizer/model init, quantization, and model-specific edge cases. I did not find any soft pair that looks like the same concrete bug strongly enough to merge.", - "confidence": 0.84, - "canonical_issue_reason": "Issue 43646 is the broadest representative of the v5 model-initialization regression subgroup, so it is the best anchor if one must be chosen; however, the cluster is heterogeneous and not a true duplicate set.", + "summary": "Most pairs are unrelated model-specific regressions, not true duplicates. The only clear duplicate-like thread is the Qwen3.5 `save_pretrained` serialization bug, where one issue is the broader report and the other is a narrower symptom.", + "confidence": 0.39, + "canonical_issue_reason": "issue:45216 is the broadest description of the Qwen3.5 `save_pretrained` regression; issue:45357 looks like a narrower symptom about incorrect visual encoder keys.", "canonical_pr_reason": null, - "best_issue_reason": "43646 is the most general issue title among the initialization-related reports and is the clearest representative of the closest mini-cluster, though still not a reliable duplicate canonical.", + "best_issue_reason": "issue:45216 is the best umbrella issue for the one clearly duplicate-like thread in this set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43883", "right": "issue:44488", "accept": false, - "reason": "Both are model-loading failures, but for different models and different error paths; not the same bug." + "reason": "Different bugs: Molmo tied-weights metadata vs loading a specific checkpoint." }, { "left": "issue:44488", "right": "issue:44987", "accept": false, - "reason": "Two separate checkpoint-loading failures for different models; no concrete shared code-path is evident." + "reason": "Both are load failures, but for different models and different failure modes." }, { - "left": "issue:43643", - "right": "issue:44530", + "left": "issue:34567", + "right": "issue:40444", "accept": false, - "reason": "Missing fields from AutoConfig with trust_remote_code is unrelated to PagedAttentionCache group-type handling." + "reason": "Trainer token counting bug vs Qwen2.5-VL iterable dataset/multi-image training bug." }, { - "left": "issue:34567", - "right": "issue:40444", + "left": "issue:43643", + "right": "issue:44530", "accept": false, - "reason": "Trainer token-count tracking and Qwen2.5-VL iterable dataset/image handling are unrelated." + "reason": "Missing fields from `AutoConfig` with `trust_remote_code` is unrelated to Qwen3.5 cache group-type errors." }, { "left": "issue:44811", "right": "issue:45584", "accept": false, - "reason": "Both involve Whisper token handling, but one is batch_decode skip_special_tokens behavior and the other is empty-transcription generation after align_special_tokens; different bugs." + "reason": "Both involve Whisper, but one is `batch_decode` special-token handling and the other is empty-transcription generation after alignment." }, { "left": "issue:43644", "right": "issue:43646", "accept": false, - "reason": "Both are v5 initialization regressions, but one is non-persistent buffer serialization and the other is custom model initialization in notebooks; too different to merge." + "reason": "Both are v5 regressions, but one is non-persistent buffers and the other is custom model initialization." }, { "left": "issue:44451", "right": "issue:44987", "accept": false, - "reason": "Different model-loading failures for different checkpoints; no same underlying defect is shown." + "reason": "Different checkpoints with no shared code-path evidence." }, { "left": "issue:43541", "right": "issue:45084", "accept": false, - "reason": "Torch dynamo grouped_mm tracing failure and template-node compilation error are unrelated." + "reason": "Torch dynamo grouped_mm tracing error is unrelated to template-node compilation." }, { "left": "issue:36032", "right": "issue:42915", "accept": false, - "reason": "Tokenizer attribute conflict and Qwen3Moe FineGrainedFP8Config failure are different subsystems and failure modes." + "reason": "Tokenizer method conflict vs Qwen3Moe FP8 config loading failure." }, { "left": "issue:43452", "right": "issue:43482", "accept": false, - "reason": "Both mention GGUF loading, but one is API breakage around gguf_file and the other is a specific Qwen2.5-GGUF v5 failure; not clearly the same bug." + "reason": "Both mention GGUF loading, but one is a general `gguf_file` API regression and the other is a Qwen2.5-GGUF-specific v5 failure." }, { "left": "issue:42915", "right": "issue:44530", "accept": false, - "reason": "FineGrainedFP8Config loading failure and PagedAttentionCache linear_attention crash are unrelated." + "reason": "Qwen3Moe FP8 config issue and Qwen3.5 linear-attention cache crash are different bugs." }, { "left": "issue:43317", "right": "issue:43482", "accept": false, - "reason": "device_map auto offload failure and GGUF model loading failure are different code paths." + "reason": "Device-map/offload dequantized-model loading issue is not the same as Qwen2.5-GGUF v5 loading failure." }, { "left": "issue:44849", "right": "issue:45325", "accept": false, - "reason": "Qwen3.5 hidden-states output bug and Qwen2.5-VL rope-index scaling bug are unrelated." + "reason": "Different Qwen family regressions: hidden-states output vs RoPE position-id scaling." }, { "left": "issue:43883", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer/model loading complaints, but for different models and different errors." + "reason": "Different problems: missing tied-weight keys vs tokenizer loading for another model." }, { "left": "issue:43404", "right": "issue:44466", "accept": false, - "reason": "Both concern tied weights, but one is a model-specific untied lm_head bug and the other is serialization differing by device; not the same concrete issue." + "reason": "Both concern tied weights, but one is a model-specific untied `lm_head` and the other is save serialization differing by device." }, { "left": "issue:36032", "right": "issue:43645", "accept": false, - "reason": "Tokenizer method conflict and notebook custom-model initialization breakage are unrelated." + "reason": "Tokenizer attribute conflict is unrelated to custom model initialization in notebooks." }, { "left": "issue:45084", "right": "issue:45507", "accept": false, - "reason": "Template compilation error and GraniteMoEHybrid invalid method call are different failures." + "reason": "Template-node compilation error is unrelated to GraniteMoEHybrid invalid-method calls." }, { "left": "issue:43262", "right": "issue:43377", "accept": false, - "reason": "Audio chat-template sampling-rate defaulting and MIMI batched-vs-single output mismatch are different bugs." + "reason": "Audio chat-template sampling-rate bug vs MIMI padding-mask batch inconsistency are different issues." }, { "left": "issue:43646", "right": "issue:44534", "accept": false, - "reason": "Both are v5 initialization-related, but one is about custom model init and the other about non-persistent buffers being filled with junk; related theme, not same bug." + "reason": "Custom model initialization and non-persistent buffer junk are separate regressions." }, { "left": "issue:44987", "right": "issue:44991", "accept": false, - "reason": "Separate model-loading failures for different repositories and error causes." + "reason": "Different model-loading failures with no shared concrete bug." }, { "left": "issue:43994", "right": "issue:44610", "accept": false, - "reason": "Siglip2 nonsensical outputs and OmDet-Turbo image-size mismatch are different model issues." + "reason": "SigLIP2 nonsense outputs vs OmDet-Turbo image-size mismatch are unrelated." }, { "left": "issue:44403", "right": "issue:45081", "accept": false, - "reason": "Generic loading noise and Mistral regex patch crash are unrelated." + "reason": "Generic loading noise is not the same as the Mistral regex patch crash." }, { "left": "issue:43322", "right": "issue:43540", "accept": false, - "reason": "Llava Next segmentation fault and Qwen3OmniMoe video-processing ValueError are distinct multimodal loading/runtime bugs." + "reason": "Llava Next loading segfault and Qwen3OmniMoe video-input ValueError are different code paths." }, { "left": "issue:43377", "right": "issue:43819", "accept": false, - "reason": "Padding-mask support in MIMI encoder and DAC from_latents STE mismatch are separate audio/model-consistency issues." + "reason": "Both are model-consistency bugs, but for different models and different mechanisms." }, { "left": "issue:43645", "right": "issue:44534", "accept": false, - "reason": "Custom-model initialization in notebooks is not the same as non-persistent buffer junk filling." + "reason": "Jupyter custom-model initialization issue is distinct from buffer serialization junk." }, { "left": "issue:42915", "right": "issue:45084", "accept": false, - "reason": "Qwen3Moe FP8 loading failure and template-node compilation error are unrelated." + "reason": "Qwen3Moe FP8 config and template compilation are unrelated." }, { "left": "issue:44451", "right": "issue:45081", "accept": false, - "reason": "Different model-loading failure and tokenizer regex patch crash; no shared concrete bug." + "reason": "Different models and different failure modes." }, { "left": "issue:45216", "right": "issue:45357", - "accept": false, - "reason": "Both are Qwen3.5 save_pretrained regressions, but one is incorrect checkpoint saving broadly and the other is incorrect visual-encoder keys; related family but not the same specific bug." - }, - { - "left": "issue:44265", - "right": "issue:44610", - "accept": false, - "reason": "torch.export exportability with torch_compilable_check is unrelated to OmDet-Turbo input resolution mismatch." + "accept": true, + "reason": "Same underlying Qwen3.5 `save_pretrained` serialization regression; the second issue is a narrower symptom about incorrect visual encoder keys." }, { "left": "issue:44987", "right": "issue:45081", "accept": false, - "reason": "A model-loading regression and a tokenizer regex patch crash are different code paths." + "reason": "Loading a specific model vs Mistral tokenizer regex patch crash are unrelated." }, { "left": "issue:44220", "right": "issue:44683", "accept": false, - "reason": "Feature-extraction issue and compiled flex_attention failure are unrelated." + "reason": "Audio feature extraction bug vs flex_attention compile failure are unrelated." + }, + { + "left": "issue:44265", + "right": "issue:44610", + "accept": false, + "reason": "`torch.export`/`torch_compilable_check` failure is unrelated to the OmDet-Turbo processor size mismatch." }, { "left": "issue:43404", "right": "issue:45127", "accept": false, - "reason": "Both involve tied weights/LM head behavior, but one is missing tying in a specific model and the other is LoRA merge collapse with extended vocab; not the same bug." + "reason": "Both mention tied embeddings, but one is untied weights in a model class and the other is LoRA merge collapse." }, { "left": "issue:43296", "right": "issue:43531", "accept": false, - "reason": "PaddleOCR-VL load failure and Qwen3-MoE sliding_window bug are unrelated." + "reason": "PaddleOCR-VL load failure in vLLM is unrelated to Qwen3-MoE sliding window behavior." }, { "left": "issue:43632", "right": "issue:44756", "accept": false, - "reason": "_is_hf_initialized flag regression and disabling mmap on Strix Halo to avoid OOM are unrelated." + "reason": "`_is_hf_initialized` flag regression is unrelated to Strix Halo mmap OOM behavior." }, { "left": "issue:42915", "right": "issue:44291", "accept": false, - "reason": "FineGrainedFP8 loading and init_empty_weights unexpected _is_hf_initialized argument are different initialization/quantization issues." + "reason": "Different regressions: Qwen3Moe FP8 config vs generic `init_empty_weights` argument error." }, { - "left": "issue:44062", - "right": "issue:44987", + "left": "issue:36032", + "right": "issue:43742", "accept": false, - "reason": "AddedToken special-argument conflict and a model-loading failure for physical-intelligence/fast are unrelated." + "reason": "Tokenizer method-name collision is unrelated to a key error loading MobileLLM-125M." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the one accepted soft-edge pair is supported by the issue titles: 45216 is the broader Qwen3.5 `save_pretrained` regression, while 45357 is a narrower symptom about visual encoder keys. The rejected pairs are phrased conservatively and do not overclaim duplication based on the available evidence." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -2997,8 +3027,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -3022,10 +3052,10 @@ "issue:43611", "issue:43632", "issue:43646", - "issue:43742", "issue:43792", "issue:43818", "issue:43856", + "issue:44062", "issue:44079", "issue:44291", "issue:44403", @@ -3038,6 +3068,7 @@ "issue:44811", "issue:44829", "issue:44877", + "issue:44987", "issue:44991", "issue:45020", "issue:45081", @@ -3051,30 +3082,30 @@ "issue:45812" ], "soft_pairs": [ - "issue:36032|issue:43742", + "issue:44062|issue:44987", "issue:42915|issue:44589", "issue:43611|issue:44534", "issue:43531|issue:44877", "issue:41553|issue:43582", "issue:43317|issue:43856", - "issue:44479|issue:45325", "issue:43425|issue:44403", + "issue:44479|issue:45325", "issue:40990|issue:44811", - "issue:43818|issue:45357", "issue:42175|issue:45542", + "issue:43818|issue:45357", "issue:44462|issue:45812", "issue:36032|issue:41720", - "issue:43421|issue:44568", "issue:44079|issue:45663", + "issue:43421|issue:44568", "issue:43792|issue:44589", "issue:43632|issue:44534", "issue:44403|issue:44991", - "issue:44829|issue:45479", - "issue:43482|issue:45538", "issue:44683|issue:45020", + "issue:43482|issue:45538", + "issue:44829|issue:45479", "issue:33666|issue:40444", - "issue:43482|issue:43525", "issue:44291|issue:44589", + "issue:43482|issue:43525", "issue:40444|issue:45325", "issue:43377|issue:45491", "issue:41720|issue:45081", @@ -3082,220 +3113,223 @@ "issue:43531|issue:44291", "issue:43611|issue:43646" ], - "prepared_review_unit_hash": "6aa89efb21ad358fa4f0c0caaf4b621b257a50a1e8621ffcd75b9ab44bc368ec", + "prepared_review_unit_hash": "ce5136055948fe3fb3a5892ba2a69985498d5ed612770c1672f6f579c2fdda02", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36181, - "estimated_input_tokens": 9046, - "estimated_eval_tokens": 18348 + "serialized_chars": 36777, + "estimated_input_tokens": 9195, + "estimated_eval_tokens": 18646 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: several separate Transformers v5 regressions and model-specific loading/training bugs, but no soft pair looks like a true duplicate. The broadest umbrella issue is the open remote_code regression report.", - "confidence": 0.47, - "canonical_issue_reason": "issue:45020 is the broadest open issue here, covering recent-version breakage in model loading with remote_code, which is the closest umbrella for the loading/regression-heavy subset of this cluster.", + "summary": "The cluster is mostly heterogeneous. The closest-looking pairs are related regressions, but from titles alone they are not confirmed to be the same concrete bug or change, so I would not merge any soft edges.", + "confidence": 0.61, + "canonical_issue_reason": "No single canonical issue stands out: the cluster is mixed, and the nearest matches are only loosely related from titles alone.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the best single issue to keep as the cluster anchor because it is open, broad, and captures the common theme of recent Transformers regressions better than the model-specific reports.", + "best_issue_reason": "Issue 45479 is the best standalone representative because it describes a concrete, broadly applicable classification failure more precisely than most other items.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:36032", - "right": "issue:43742", + "left": "issue:44062", + "right": "issue:44987", "accept": false, - "reason": "Different models and failures: T5 tokenizer method conflict vs a key error while loading MobileLLM." + "reason": "Both are tokenizer/loading regressions, but the titles describe different failure modes and do not establish the same bug." }, { "left": "issue:42915", "right": "issue:44589", "accept": false, - "reason": "Both mention float8-related failures, but one is Qwen3Moe FP8 config handling and the other is a missing Float8 storage type; not the same bug." + "reason": "Both mention FP8-related breakage, but one is a config failure and the other is a storage-object TypeError; not the same issue." }, { "left": "issue:43611", "right": "issue:44534", "accept": false, - "reason": "Separate v5 regressions: base_model_prefix loading failure vs non-persistent buffer corruption." + "reason": "Both are Transformers v5 regressions, but the titles point to different concrete failures and likely different code paths." }, { "left": "issue:43531", "right": "issue:44877", "accept": false, - "reason": "Unrelated model/config bugs: Qwen3-MoE sliding_window issue vs granite_speech strict config loading." + "reason": "Different models and different symptoms; the titles do not show a shared underlying bug." }, { "left": "issue:41553", "right": "issue:43582", "accept": false, - "reason": "Different subsystems entirely: tokenizer error message vs Apple Silicon allocator warmup TypeError." + "reason": "Unrelated subsystems: tokenizer loading/error messaging versus Apple Silicon allocator warmup." }, { "left": "issue:43317", "right": "issue:43856", "accept": false, - "reason": "One is device_map/offload loading failure; the other is Qwen3 MoE training memory inefficiency. Same broad area, different bug." + "reason": "Both involve memory/device behavior, but one is device-map/offload loading and the other is MoE training memory usage." }, { - "left": "issue:44479", - "right": "issue:45325", + "left": "issue:43425", + "right": "issue:44403", "accept": false, - "reason": "Both are Qwen VL-related, but one is a broad video-input regression and the other is a still-image temporal position_ids scaling bug." + "reason": "Version compatibility and loading-noise complaints are not the same concrete bug." }, { - "left": "issue:43425", - "right": "issue:44403", + "left": "issue:44479", + "right": "issue:45325", "accept": false, - "reason": "Incompatible torch/backend issue vs an unrelated logging/noise issue during model loading." + "reason": "These are related Qwen VL/video-position regressions, but the titles alone do not confirm the same root cause or code path." }, { "left": "issue:40990", "right": "issue:44811", "accept": false, - "reason": "Completely different problems: perplexity on GPT-OSS vs Whisper decode skipping special tokens." + "reason": "Different models and different behaviors: perplexity evaluation versus Whisper decoding options." }, { - "left": "issue:43818", - "right": "issue:45357", + "left": "issue:42175", + "right": "issue:45542", "accept": false, - "reason": "Different model families and code paths: Video-LLaVA weight/temporal attention issue vs Qwen3.5 save_pretrained key regression." + "reason": "Both touch backend packaging, but the reported breakages are not the same concrete failure." }, { - "left": "issue:42175", - "right": "issue:45542", + "left": "issue:43818", + "right": "issue:45357", "accept": false, - "reason": "Both involve TensorFlow packaging/backend, but one is missing TF from the torch extra and the other is an undefined tf backend when only tensorboard is installed." + "reason": "Different model families and distinct bugs: missing temporal attention versus incorrect saved key mapping." }, { "left": "issue:44462", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but one ignores tokenizer.json and the other yields wrong Granite token IDs; not the same concrete failure." + "reason": "Both involve AutoTokenizer, but one is repository tokenizer precedence and the other is a Granite token-ID regression." }, { "left": "issue:36032", "right": "issue:41720", "accept": false, - "reason": "Tokenizer method conflict vs Qwen3 auto device mapping cuda assert; unrelated bugs." + "reason": "Tokenizer loading conflict versus CUDA assertion during device mapping; unrelated." }, { - "left": "issue:43421", - "right": "issue:44568", + "left": "issue:44079", + "right": "issue:45663", "accept": false, - "reason": "Both touch special-token handling, but one is runtime post-processor refresh and the other is add_special_tokens not adding BOS/EOS for a specific tokenizer." + "reason": "A ModelOutput key assignment bug and a Gemma-4 FSDP2 KeyError are different issues." }, { - "left": "issue:44079", - "right": "issue:45663", + "left": "issue:43421", + "right": "issue:44568", "accept": false, - "reason": "ModelOutput key assignment bug vs Gemma4 FSDP2 shared_kv_states reconstruction error; different code paths." + "reason": "Both mention special tokens, but one is runtime post-processor updates and the other is BOS/EOS emission for a specific tokenizer." }, { "left": "issue:43792", "right": "issue:44589", "accept": false, - "reason": "Whisper runtime failure vs missing Float8 storage type; unrelated." + "reason": "A Whisper runtime failure and a missing Float8 storage object are unrelated." }, { "left": "issue:43632", "right": "issue:44534", "accept": false, - "reason": "Different Transformers v5 regressions: _is_hf_initialized breakage vs junk-filled non-persistent buffers." + "reason": "Distinct v5 regressions affecting different initialization/buffer mechanisms." }, { "left": "issue:44403", "right": "issue:44991", "accept": false, - "reason": "Broad loading noise report vs tokenizer loading failure for a specific model; not a duplicate." + "reason": "A noise/logging issue and a tokenizer loading failure are not the same bug." }, { - "left": "issue:44829", - "right": "issue:45479", + "left": "issue:44683", + "right": "issue:45020", "accept": false, - "reason": "Both are classification-training regressions, but one is flash_attention_3-induced degeneration and the other is a zero-loss problem from num_labels=1." + "reason": "Compiled flex attention on torch>=2.9 and remote_code loading are separate problems." }, { "left": "issue:43482", "right": "issue:45538", "accept": false, - "reason": "Qwen2.5-GGUF loading failure vs CLIPTokenizer model_max_length handling; unrelated." + "reason": "GGUF loading under v5 and CLIPTokenizer model_max_length are unrelated." }, { - "left": "issue:44683", - "right": "issue:45020", + "left": "issue:44829", + "right": "issue:45479", "accept": false, - "reason": "Torch flex_attention compilation failure vs remote_code model loading regressions; different layers of the stack." + "reason": "These are related degenerate-training reports, but the titles do not confirm the same failure path or code change." }, { "left": "issue:33666", "right": "issue:40444", "accept": false, - "reason": "Both are Qwen-VL training issues, but multi-GPU training and iterable-dataset multi-image finetuning are distinct bugs." + "reason": "Both are Qwen VL training issues, but they affect different workflows and are not the same bug." }, { - "left": "issue:43482", - "right": "issue:43525", + "left": "issue:44291", + "right": "issue:44589", "accept": false, - "reason": "Different loading failures: GGUF/v5 compatibility vs missing pad_token_id on Llama4Config." + "reason": "Different TypeErrors caused by different v5 changes; no shared code-path evidence." }, { - "left": "issue:44291", - "right": "issue:44589", + "left": "issue:43482", + "right": "issue:43525", "accept": false, - "reason": "Init_empty_weights/_is_hf_initialized argument issue vs Float8 storage lookup failure; not the same bug." + "reason": "Different model-loading failures with different symptoms and likely different causes." }, { "left": "issue:40444", "right": "issue:45325", "accept": false, - "reason": "Different Qwen2.5-VL regressions: multi-image iterable dataset failure vs still-image temporal position scaling." + "reason": "Both are Qwen2.5-VL-related, but one is multi-image training and the other is temporal position scaling." }, { "left": "issue:43377", "right": "issue:45491", "accept": false, - "reason": "Different model families and symptoms: MIMI padding-mask issue vs Gemma3 sliding-window NaNs." + "reason": "Both involve batching/masking behavior, but they are separate models and separate bugs." }, { "left": "issue:41720", "right": "issue:45081", "accept": false, - "reason": "Qwen3 A800 cuda assert vs Mistral tokenizer regex patch crash; unrelated." + "reason": "CUDA assertion during device mapping and a Mistral tokenizer regex crash are unrelated." }, { "left": "issue:42175", "right": "issue:43296", "accept": false, - "reason": "TensorFlow packaging/backend issue vs PaddleOCR-VL load failure in vLLM; separate causes." + "reason": "A packaging/backend install issue and a vLLM model-loading failure are different problems." }, { "left": "issue:43531", "right": "issue:44291", "accept": false, - "reason": "Qwen3-MoE sliding_window bug vs init_empty_weights unexpected argument error; unrelated." + "reason": "Sliding-window Qwen3-MoE behavior and an init_empty_weights argument regression are unrelated." }, { "left": "issue:43611", "right": "issue:43646", "accept": false, - "reason": "Both are v5 loading regressions, but one is base_model_prefix handling and the other is custom model initialization; too different to merge." + "reason": "Both are broad Transformers v5 regressions, but the titles describe different concrete failures and not one mergeable fix." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, + "evaluator_result": { + "accept": false, + "feedback": "The cluster-level summary is directionally plausible, but a few claims go beyond what the packet supports. In particular, the two accepted soft pairs are asserted too strongly from titles alone: 44479/45325 is described as the same regression with 45325 as the root cause, and 44829/45479 is called the same zero-loss bug. The packet only gives issue titles, so these should be framed more cautiously as related, possibly overlapping regressions rather than confirmed duplicates or shared code paths." + }, + "evaluator_used": true, + "retried": true, "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null @@ -3308,8 +3342,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -3397,232 +3431,232 @@ "issue:43572|issue:45759", "issue:39692|issue:40444" ], - "prepared_review_unit_hash": "e7f90e29b8ebe9c37644eb9fe77fc3070ae5229e6d700c1559880aedfb7195a2", + "prepared_review_unit_hash": "5b3dcbb3bb817f96037c2dd072d87cd443171bfe63d5680526bc0e933dfe659a", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37595, - "estimated_input_tokens": 9399, - "estimated_eval_tokens": 19054 + "serialized_chars": 37582, + "estimated_input_tokens": 9396, + "estimated_eval_tokens": 19048 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a heterogeneous set of unrelated bug reports. I only found one plausible duplicate pair: the Qwen video-input regression reports (#44479 and #45381). All other soft edges look like different bugs or different code paths.", - "confidence": 0.71, - "canonical_issue_reason": "#44479 is the broadest and most representative report of the Qwen video-input regression; it covers multiple affected Qwen variants, while #45381 is the narrower qwen2.5-vl manifestation.", + "summary": "This cluster is mostly a grab-bag of distinct model-loading, multimodal, quantization, and device-mapping regressions. None of the soft edges look like true duplicates, and there are no PRs to consolidate.", + "confidence": 0.62, + "canonical_issue_reason": "issue:43644 is the broadest systemic regression in the set, centered on a Transformers v5 state/buffer behavior break that is representative of the wider v5-related failures here.", "canonical_pr_reason": null, - "best_issue_reason": "#44479 is the best issue anchor here because it captures the most general version of the only likely duplicate bug in the set and is the clearest representative for triage.", + "best_issue_reason": "issue:43644 is the best umbrella issue candidate because it describes a concrete, cross-cutting v5 regression with broad downstream impact and is likely to be the most reusable reference point.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44265", "right": "issue:45072", "accept": false, - "reason": "Different bugs: export failures around torch_compilable_check vs bfloat16 dtype mismatches during inference." + "reason": "Different code paths and symptoms: torch.export support for torch_compilable_check versus bfloat16 dtype mismatches during inference." }, { "left": "issue:43632", "right": "issue:45072", "accept": false, - "reason": "Different issues: _is_hf_initialized / v5 initialization breakage vs dtype mismatch under bfloat16 inference." + "reason": "One is an initialization flag regression; the other is a dtype-mismatch issue under bfloat16 inference. Not the same bug." }, { "left": "issue:43454", "right": "issue:44466", "accept": false, - "reason": "Related theme (tied weights) but different code paths: one is failing to tie lm_head weights, the other is inconsistent serialization behavior by device." + "reason": "Both involve tied weights, but one is a missing weight-tying bug and the other is inconsistent serialization behavior; distinct underlying failures." }, { "left": "issue:43994", "right": "issue:44898", "accept": false, - "reason": "Different models and failure modes: SigLIP2 AutoModel/pipeline nonsense outputs vs Perceiver non-default-resolution failure." + "reason": "Different models and failure modes: SigLIP2 pipeline/AutoModel mismatch versus Perceiver non-default-resolution classification." }, { "left": "issue:44079", "right": "issue:44534", "accept": false, - "reason": "Different underlying bugs: ModelOutput key assignment vs non-persistent buffer junk in v5." + "reason": "ModelOutput key assignment logic is unrelated to non-persistent buffer initialization/serialization." }, { "left": "issue:45325", "right": "issue:45834", "accept": false, - "reason": "Different Qwen-family problems: still-image rope scaling bug vs Kosmos2.5 long-OCR index error." + "reason": "Both are multimodal indexing bugs, but one is a Qwen2.5-VL temporal RoPE scaling issue and the other is a Kosmos2.5 long-OCR index error." }, { "left": "issue:43792", "right": "issue:43957", "accept": false, - "reason": "Both involve model loading, but the concrete failures differ: whisper runtime failure vs meta-device loading breakage in v5." + "reason": "Different failure classes: Whisper model runtime failure versus meta-device loading breakage across some models." }, { "left": "issue:42915", "right": "issue:43482", "accept": false, - "reason": "Different problems: FineGrainedFP8Config failure vs GGUF loading failure in transformers v5." + "reason": "Different models and mechanisms: FineGrainedFP8Config failure versus GGUF loading failure in v5." }, { "left": "issue:44079", "right": "issue:45072", "accept": false, - "reason": "Different bugs: ModelOutput key handling vs dtype mismatch during bfloat16 inference." + "reason": "ModelOutput key handling is not the same as dtype mismatches in bf16 inference." }, { "left": "issue:40444", "right": "issue:45381", "accept": false, - "reason": "Both are Qwen2.5-VL related, but one is iterable-dataset multi-image finetuning and the other is video position-id computation." + "reason": "Both are Qwen2.5-VL related, but one is training with multiple images per prompt and the other is video vision_position_ids; not the same concrete bug." }, { "left": "issue:45325", "right": "issue:45381", "accept": false, - "reason": "Both concern Qwen2.5-VL positioning, but one is still-image temporal scaling and the other is video-input vision_position_ids." + "reason": "Closely related family, but still different symptoms: still-image temporal position scaling versus video vision_position_id computation." }, { "left": "issue:43644", "right": "issue:44079", "accept": false, - "reason": "Different regressions: non-persistent buffer junk vs ModelOutput key assignment." + "reason": "Buffer initialization junk and ModelOutput key assignment are unrelated regressions." }, { "left": "issue:44387", "right": "issue:44617", "accept": false, - "reason": "Both mention OOM, but one is increased reserved memory under int4 quantization and the other is a model-specific video OOM." + "reason": "Both can end in OOM, but one is a Transformers int4 memory regression and the other is a model-specific video OOM." }, { "left": "issue:43881", "right": "issue:44488", "accept": false, - "reason": "Both are loading failures, but they affect unrelated models with no shared concrete code path evident." + "reason": "Separate model-loading failures for different models; no shared concrete code-path shown." }, { "left": "issue:43296", "right": "issue:45800", "accept": false, - "reason": "Different layers and failures: PaddleOCR-VL load in vLLM vs torch/transformers version incompatibility." + "reason": "Different compatibility problems: vLLM/PaddleOCR-VL loading versus a torch 2.4.1 / Transformers 5.8.0 incompatibility." }, { "left": "issue:44479", "right": "issue:45823", "accept": false, - "reason": "Different bugs: Qwen video-input regression vs Gemma4 PLE device mismatch under device_map=auto." + "reason": "Different multimodal/device issues: Qwen VL video regression versus Gemma4 PLE device mismatch in forward." }, { "left": "issue:35707", "right": "issue:40444", "accept": false, - "reason": "Different generation/training scenarios and model families; no shared concrete bug path." + "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to Qwen2.5-VL multi-image finetuning failures." }, { "left": "issue:44291", "right": "issue:44877", "accept": false, - "reason": "Different loading/config issues: unexpected _is_hf_initialized arg vs strict config blocking granite_speech config." + "reason": "One is an unexpected _is_hf_initialized argument during init_empty_weights; the other is strict config validation blocking granite_speech config loading." }, { "left": "issue:43317", "right": "issue:43873", "accept": false, - "reason": "Related offload/quantization area, but one is dequantized model loading on gpu+cpu offload and the other is a broader offloading complaint." + "reason": "Both mention offloading/quantization, but one is dequantized model loading with auto device maps and the other is a broader offloading complaint; not clearly the same bug." }, { "left": "issue:43606", "right": "issue:44265", "accept": false, - "reason": "Different failure modes and subsystems: CPU offload device mismatch vs torch.export / torch_compilable_check export failure." + "reason": "CPU offload device mismatch and torch.export support for torch_compilable_check are unrelated." }, { "left": "issue:43701", "right": "issue:45325", "accept": false, - "reason": "Completely different problems: checkpoint resume key mismatch vs Qwen2.5-VL rope index scaling." + "reason": "Checkpoint resume key mismatch is unrelated to Qwen2.5-VL RoPE position-id scaling." }, { "left": "issue:44186", "right": "issue:44265", "accept": false, - "reason": "Tokenizer padding/truncation crash vs torch.export failure; unrelated code paths." + "reason": "Tokenizer crashes on NER padding/truncation are unrelated to torch.export failures on compilable checks." }, { "left": "issue:43606", "right": "issue:43819", "accept": false, - "reason": "Different audio-model bugs: CPU offload device mismatch vs DAC latent/STE mismatch." + "reason": "CPU offload device mismatch is unrelated to DAC.from_latents versus forward-pass mismatch." }, { "left": "issue:43262", "right": "issue:43746", "accept": false, - "reason": "Different GraniteSpeech-related issues: audio chat-template sampling rate default vs PEFT adapter local checkpoint loading." + "reason": "Audio chat-template sampling-rate defaults and GraniteSpeech PEFT checkpoint loading are different issues." }, { "left": "issue:41720", "right": "issue:45823", "accept": false, - "reason": "Different device-map failures on different models; no evidence of the same underlying bug." + "reason": "Both are device_map=auto issues, but they affect different models and failures; not the same code-path bug." }, { "left": "issue:43526", "right": "issue:43883", "accept": false, - "reason": "Different regressions: BeitImageProcessorFast label reduction bug vs missing all_tied_weights_keys on Molmo." + "reason": "BEiT label reduction and missing all_tied_weights_keys are unrelated." }, { "left": "issue:43454", "right": "issue:43550", "accept": false, - "reason": "Different bugs: tied lm_head weight issue vs torch.compile SDPA failure in Bamba-9B-v2." + "reason": "Tied-weight loading/serialization versus torch.compile SDPA failure are different bugs." }, { "left": "issue:43377", "right": "issue:44610", "accept": false, - "reason": "Different problems: batching/padding-mask output mismatch in MIMI vs OmDet-Turbo processor/model input-size mismatch." + "reason": "Missing padding mask support in MIMI encoder is unrelated to OmDet-Turbo processor/model resolution mismatch." }, { "left": "issue:43377", "right": "issue:44898", "accept": false, - "reason": "Different bugs in different models: MIMI batching mismatch vs Perceiver resolution handling." + "reason": "Different models and different failures: MIMI batching discrepancy versus Perceiver resolution handling." }, { "left": "issue:43454", "right": "issue:45072", "accept": false, - "reason": "Different bugs: lm_head weight tying vs dtype mismatch during bfloat16 inference." + "reason": "A tied-weights bug is not the same as bf16 dtype mismatch during inference." }, { "left": "issue:40990", "right": "issue:45584", "accept": false, - "reason": "Unrelated failures: perplexity anomaly on gpt-oss-20b vs Whisper empty-transcription generation failure." + "reason": "High perplexity on WikiText-2 is unrelated to Whisper empty-transcription generation after align_special_tokens." }, { "left": "issue:43572", "right": "issue:45759", "accept": false, - "reason": "Different issues: missing pad_token_idx in StableLmConfig vs from_config not unwrapping text_config for multimodal Qwen configs." + "reason": "StableLm pad_token_idx regression and Qwen multimodal config unwrapping are different failures." }, { "left": "issue:39692", "right": "issue:40444", "accept": false, - "reason": "Different models and bugs: SigLIP2 documentation example errors vs Qwen2.5-VL multi-image finetuning failure." + "reason": "SigLIP2 documentation errors are unrelated to Qwen2.5-VL finetuning with multiple images per prompt." } ] }, @@ -3641,8 +3675,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -3655,7 +3689,6 @@ "issue:43296", "issue:43316", "issue:43317", - "issue:43322", "issue:43404", "issue:43450", "issue:43454", @@ -3686,12 +3719,14 @@ "issue:44977", "issue:44991", "issue:45072", + "issue:45081", "issue:45357", "issue:45406", "issue:45632", "issue:45684", "issue:45701", "issue:45704", + "issue:45812", "issue:45823" ], "soft_pairs": [ @@ -3709,245 +3744,239 @@ "issue:44423|issue:45406", "issue:44186|issue:45072", "issue:43577|issue:43708", - "issue:35707|issue:41720", "issue:43873|issue:44756", + "issue:35707|issue:41720", "issue:44877|issue:44991", - "issue:43749|issue:44534", "issue:42617|issue:42915", + "issue:43749|issue:44534", "issue:43450|issue:45701", "issue:43550|issue:43720", "issue:36331|issue:42915", "issue:44610|issue:44898", - "issue:44186|issue:44610", "issue:36331|issue:42617", + "issue:44186|issue:44610", "issue:44387|issue:45704", - "issue:44479|issue:44918", "issue:43526|issue:43881", + "issue:44479|issue:44918", "issue:42915|issue:43317", "issue:43316|issue:44913", "issue:43550|issue:43606", - "issue:41720|issue:43322" + "issue:45081|issue:45812" ], - "prepared_review_unit_hash": "982c472f21e17977c55557050a8d624b5ca3568bc9689ba77260645181c4a213", + "prepared_review_unit_hash": "787f72bbe02d5f4636fa9028e04a7835af16275847bf28f5c847fd25762d9b80", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 36687, - "estimated_input_tokens": 9172, - "estimated_eval_tokens": 18600 + "serialized_chars": 37407, + "estimated_input_tokens": 9352, + "estimated_eval_tokens": 18960 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: the issues span unrelated bugs in loading, configs, generation, offloading, processors, and trainer behavior. I found no true duplicate pairs; all soft matches should be rejected.", - "confidence": 0.93, - "canonical_issue_reason": "No clear duplicate anchor exists. If a representative is required, issue 43873 is the broadest and most referenced offloading/quantization report, but it is not the same underlying bug as the others.", + "summary": "The cluster is mostly heterogeneous: many items are model-specific loading/runtime regressions, and the soft pairs share broad themes but not the same concrete bug. I\u2019d use the broad offloading/quantization issue as the cluster representative.", + "confidence": 0.61, + "canonical_issue_reason": "issue:43873 is the most general and central representative of the cluster\u2019s recurring offloading/quantization/device-loading failures, and it has broader linkage than the more model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43873 is the best representative by breadth and activity, but only as a loose thematic anchor; it should not be treated as a duplicate of the rest.", + "best_issue_reason": "issue:43873 is the best single issue anchor because it is broad, still open, and closest to the repeated loading/offload failure pattern seen across the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43881", "right": "issue:44991", "accept": false, - "reason": "Different failure classes: model loading for glm-4v-9b vs tokenizer loading for a specific RoBERTa model." + "reason": "Different models and different failure surfaces: GLM-4V load failure vs tokenizer regression for EMBEDDIA/est-roberta." }, { "left": "issue:44466", "right": "issue:45357", "accept": false, - "reason": "Both involve serialization, but one is tied lm_head weights in v5 and the other is incorrect visual encoder keys for Qwen3.5." + "reason": "Both involve serialization, but one is tied lm_head handling on device-dependent save, the other is incorrect visual encoder keys in Qwen3.5." }, { "left": "issue:43792", "right": "issue:44683", "accept": false, - "reason": "Unrelated paths: Whisper model runtime failure vs compiled flex_attention breakage on newer torch." + "reason": "Whisper loading failure and compiled flex_attention on torch>=2.9 are unrelated code paths." }, { "left": "issue:43296", "right": "issue:44877", "accept": false, - "reason": "Different loader issues: PaddleOCR-VL model loading vs strict config rejecting granite_speech config." + "reason": "PaddleOCR-VL/vLLM load compatibility and strict config loading for granite_speech are distinct issues." }, { "left": "issue:43454", "right": "issue:43720", "accept": false, - "reason": "Both mention loading and weights, but they are different models and different bugs: tied lm_head weights vs unpacking BitNet packed weights." + "reason": "Mistral3 weight tying and BitNet packed-weight unpacking are different loader/model-conversion bugs." }, { "left": "issue:43873", "right": "issue:45701", "accept": false, - "reason": "Offloading/quantization behavior is unrelated to tokenizer changes across versions." + "reason": "Quantization/offloading behavior and tokenization differences are not the same underlying bug." }, { "left": "issue:41720", "right": "issue:44977", "accept": false, - "reason": "Same model family, but different code paths and symptoms: device-map CUDA assert vs flash-attention generation regression." + "reason": "Qwen3 auto device-mapping cuda assert is a different failure from Qwen3.5 flash-attention generation regressions." }, { "left": "issue:44008", "right": "issue:45823", "accept": false, - "reason": "Different Gemma variants and different bugs: forward variable-name collision vs PLE device mismatch." + "reason": "Gemma3n forward variable collision and Gemma4 PLE device mismatch are separate model-specific bugs." }, { "left": "issue:45632", "right": "issue:45684", "accept": false, - "reason": "Both touch save/cache flows, but one is a trust_remote_code cache collision and the other is permission propagation in save_pretrained." + "reason": "Cache-path collisions and read-only permission propagation are unrelated save/load issues." }, { "left": "issue:43404", "right": "issue:45357", "accept": false, - "reason": "Both involve model serialization/weights, but one is Mistral3 tied weights and the other is Qwen3.5 visual encoder keys." + "reason": "Mistral3 lm_head tying and Qwen3.5 visual encoder key serialization affect different models and mechanisms." }, { "left": "issue:34634", "right": "issue:35707", "accept": false, - "reason": "Completely different areas: Bark voice preset handling vs progressive generation with inputs_embeds/past_key_values." + "reason": "BarkProcessor voice_preset failure and progressive generation with inputs_embeds/past_key_values are unrelated." }, { "left": "issue:44423", "right": "issue:45406", "accept": false, - "reason": "Both are serve/processor crashes, but the concrete failures differ: multimodal .to on str vs Gemma4Processor missing _tokenizer." + "reason": "Continuous-batching multimodal crash and missing _tokenizer on Gemma4Processor are different serve-time failures." }, { "left": "issue:44186", "right": "issue:45072", "accept": false, - "reason": "Different subsystems and symptoms: tokenizer padding/NER crash vs dtype mismatch in inference wrappers." + "reason": "LayoutLMv2Tokenizer NER/padding crash and bfloat16 dtype mismatch are not the same bug." }, { "left": "issue:43577", "right": "issue:43708", "accept": false, - "reason": "BLIP2 dtype loading bug and Trainer max_steps resume math are unrelated." + "reason": "Blip2 dtype propagation and Trainer resume/max_steps calculation are different subsystems." }, { - "left": "issue:35707", - "right": "issue:41720", + "left": "issue:43873", + "right": "issue:44756", "accept": false, - "reason": "Progressive generation bug and device-map CUDA assert are different failure modes with no shared code-path." + "reason": "Both mention memory/offload concerns, but one is quantization/offloading behavior and the other is a Strix Halo mmap OOM fix." }, { - "left": "issue:43873", - "right": "issue:44756", + "left": "issue:35707", + "right": "issue:41720", "accept": false, - "reason": "Both relate to memory/offloading, but one is quantization offload behavior and the other is mmap on Strix Halo causing OOM." + "reason": "Input-embedding generation handling is unrelated to Qwen3 device-map cuda asserts." }, { "left": "issue:44877", "right": "issue:44991", "accept": false, - "reason": "Config strictness blocking granite_speech is not the same as tokenizer loading failing on a specific model in v5." + "reason": "Strict config loading and tokenizer loading regression are separate problems." }, { - "left": "issue:43749", - "right": "issue:44534", + "left": "issue:42617", + "right": "issue:42915", "accept": false, - "reason": "Both concern model loading/state handling, but one is FSDP CPU RAM efficient loading and the other is non-persistent buffers being filled with junk." + "reason": "3d_parallel.py startup/runtime failure and Qwen3Moe FineGrainedFP8Config failure are different concrete bugs." }, { - "left": "issue:42617", - "right": "issue:42915", + "left": "issue:43749", + "right": "issue:44534", "accept": false, - "reason": "Unrelated issues: 3d_parallel.py execution failure vs Qwen3Moe FineGrainedFP8Config failure." + "reason": "FSDP CPU RAM efficient loading and non-persistent buffer junk are both load/init related but not the same code-path problem." }, { "left": "issue:43450", "right": "issue:45701", "accept": false, - "reason": "Video processor batch-shape bug and tokenizer behavior changes are unrelated." + "reason": "Video processor batching shape bug and tokenizer version-dependent tokenization are unrelated." }, { "left": "issue:43550", "right": "issue:43720", "accept": false, - "reason": "Different model/code paths: torch.compile SDPA failure in Bamba vs packed-weight unpacking in BitNet loading." + "reason": "Bamba SDPA torch.compile failure and BitNet packed-weight unpacking are different model/runtime bugs." }, { "left": "issue:36331", "right": "issue:42915", "accept": false, - "reason": "Trainer compute_loss signature mismatch is unrelated to Qwen3Moe FP8 loading." + "reason": "CustomTrainer compute_loss signature mismatch and Qwen3Moe FP8 load failure are unrelated." }, { "left": "issue:44610", "right": "issue:44898", "accept": false, - "reason": "Both involve image preprocessing size, but they are different models and different bugs, not the same underlying issue." + "reason": "Both are vision-processing/resolution issues, but they hit different models and different mismatch logic." }, { - "left": "issue:44186", - "right": "issue:44610", + "left": "issue:36331", + "right": "issue:42617", "accept": false, - "reason": "Tokenizer crash in LayoutLMv2 and OmDet-Turbo resolution mismatch are unrelated." + "reason": "Trainer API mismatch and 3d_parallel.py failure are different concrete issues." }, { - "left": "issue:36331", - "right": "issue:42617", + "left": "issue:44186", + "right": "issue:44610", "accept": false, - "reason": "Trainer API mismatch and 3d_parallel execution failure are unrelated." + "reason": "Tokenizer crash on NER/batched padding and processor output-size mismatch are different preprocessing bugs." }, { "left": "issue:44387", "right": "issue:45704", "accept": false, - "reason": "Both are memory-related, but one is int4 quantization OOM in Transformers 5.x and the other is a T5-apex RMSNorm memory leak." + "reason": "Int4 quantization OOM and apex FusedRMSNorm memory leak are both memory-related but not the same defect." }, { - "left": "issue:44479", - "right": "issue:44918", + "left": "issue:43526", + "right": "issue:43881", "accept": false, - "reason": "Different multimodal failures: video-input regression in Qwen VL variants vs TRL SFT unpacking of input embeddings." + "reason": "BeitImageProcessorFast label reduction bug and GLM-4V loading failure are unrelated." }, { - "left": "issue:43526", - "right": "issue:43881", + "left": "issue:44479", + "right": "issue:44918", "accept": false, - "reason": "BEIT label reduction bug and glm-4v-9b load failure are unrelated." + "reason": "Qwen VL video regression and TRL SFT embedding unpack failure are distinct multimodal/training bugs." }, { "left": "issue:42915", "right": "issue:43317", "accept": false, - "reason": "Both are loading/config issues, but one is Qwen3Moe FP8 config failure and the other is dequantized model loading with gpu+cpu offload." + "reason": "Qwen3Moe FP8 config failure and dequantized model offload loading failure are different failure modes." }, { "left": "issue:43316", "right": "issue:44913", "accept": false, - "reason": "Both are config/API persistence issues, but the concrete fields and models differ: Gemma3TextConfig API discrepancy vs GPTNeoX rotary_pct reload." + "reason": "Gemma3TextConfig API discrepancy and GPTNeoX rotary_pct reload regression are unrelated config issues." }, { "left": "issue:43550", "right": "issue:43606", "accept": false, - "reason": "Unrelated problems: Bamba SDPA torch.compile failure vs Bark-small CPU offload device mismatch." - }, - { - "left": "issue:41720", - "right": "issue:43322", - "accept": false, - "reason": "Different bugs and symptoms: Qwen3 device-map CUDA assert vs Llava Next segmentation fault on load." + "reason": "Bamba SDPA compile failure and Bark CPU-offload device mismatch are different runtime paths." } ] }, @@ -3966,8 +3995,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -3991,6 +4020,7 @@ "issue:43606", "issue:43653", "issue:43720", + "issue:43746", "issue:43782", "issue:43873", "issue:43927", @@ -4009,7 +4039,6 @@ "issue:44991", "issue:45059", "issue:45072", - "issue:45081", "issue:45161", "issue:45357", "issue:45406", @@ -4021,20 +4050,20 @@ "issue:45859" ], "soft_pairs": [ - "issue:45081|issue:45812", - "issue:40990|issue:45701", + "issue:41720|issue:43322", "issue:43295|issue:45701", + "issue:40990|issue:45701", "issue:41720|issue:45161", "issue:43531|issue:43572", - "issue:44671|issue:44743", "issue:43421|issue:45812", + "issue:44671|issue:44743", "issue:41720|issue:42617", "issue:36010|issue:42617", "issue:43540|issue:43782", "issue:44186|issue:44361", "issue:45406|issue:45464", - "issue:43421|issue:45701", "issue:30064|issue:34634", + "issue:43421|issue:45701", "issue:44492|issue:44509", "issue:44779|issue:45812", "issue:42915|issue:43322", @@ -4050,222 +4079,229 @@ "issue:44451|issue:44991", "issue:43873|issue:45704", "issue:45759|issue:45859", - "issue:44898|issue:45072" + "issue:44898|issue:45072", + "issue:41720|issue:43746" ], - "prepared_review_unit_hash": "2d3a809d1d86f0264bcbebdb2821585b30eab274778dae8d34caab27e99722b8", + "prepared_review_unit_hash": "523ca39b6cbd7183aecce575cefe8df0ed2b7f1e51f2a89b419e84e09c0e330e", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36745, - "estimated_input_tokens": 9187, - "estimated_eval_tokens": 18630 + "soft_pair_count": 31, + "serialized_chars": 36984, + "estimated_input_tokens": 9246, + "estimated_eval_tokens": 18748 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Only one soft pair looks like a true duplicate: the two Qwen2_5_VLProcessor.apply_chat_template crash reports. The rest are different models, code paths, or unrelated bug classes.", + "summary": "This cluster is mostly a set of unrelated transformer/model bugs. Only one pair is a clear duplicate: the two Qwen2_5_VLProcessor.apply_chat_template padding=False crashes. All other soft pairs differ in model, subsystem, or failure mode.", "confidence": 0.95, - "canonical_issue_reason": "issue:44514 is the cleaner canonical issue: it was opened earlier and has the more complete report for the same batched-input `padding=False` crash.", + "canonical_issue_reason": "issue:44545 is the best canonical representative for the only true duplicate pair; it has the same exact bug report as 44514 and slightly more discussion/inbound references.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44514 is the best issue overall for the same reason: it is the same bug as 44545 but has the fuller, earlier report.", + "best_issue_reason": "issue:44545 is the strongest issue to keep as the canonical cluster item because it is the clearest, most engaged report for the duplicated Qwen2_5_VLProcessor.apply_chat_template crash.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:45081", - "right": "issue:45812", + "left": "issue:41720", + "right": "issue:43322", "accept": false, - "reason": "Different tokenizer bugs for different model families (Mistral regex patch crash vs Granite token ID regression)." + "reason": "Different failures in different models: Qwen3 device-mapping cuda assert vs Llava Next load segfault." }, { - "left": "issue:40990", + "left": "issue:43295", "right": "issue:45701", "accept": false, - "reason": "Different symptoms and likely roots: perplexity evaluation problem vs tokenization behavior changes across versions." + "reason": "One is a processor/tokenizer API regression; the other is a broad tokenization-version complaint." }, { - "left": "issue:43295", + "left": "issue:40990", "right": "issue:45701", "accept": false, - "reason": "Processor/tokenizer regression for multimodal custom code vs general tokenization change report; not the same bug." + "reason": "Perplexity regression on GPT-OSS is not the same bug as version-dependent tokenization changes." }, { "left": "issue:41720", "right": "issue:45161", "accept": false, - "reason": "Different model/runtime failures: CUDA assert with auto device mapping vs tensor-parallel not working on GPT-OSS MoE." + "reason": "Different Qwen/MoE issues with unrelated symptoms: cudaErrorAssert vs tensor-parallel not working." }, { "left": "issue:43531", "right": "issue:43572", "accept": false, - "reason": "Unrelated issues in different components: Qwen3-MoE sliding window vs StableLmConfig pad_token_idx regression." + "reason": "Sliding-window behavior in Qwen3-MoE and missing pad_token_idx in StableLmConfig are unrelated." }, { - "left": "issue:44671", - "right": "issue:44743", + "left": "issue:43421", + "right": "issue:45812", "accept": false, - "reason": "Different code paths and models: CamemBERT MLM outputs vs recurrent state reset in modular Qwen3.5 cache handling." + "reason": "Both involve tokenizers, but one is runtime special-token post-processing; the other is Granite token IDs regression." }, { - "left": "issue:43421", - "right": "issue:45812", + "left": "issue:44671", + "right": "issue:44743", "accept": false, - "reason": "Runtime special-token post-processing update bug is distinct from Granite tokenizer wrong IDs." + "reason": "CamemBERT masked-LM outputs and Qwen3.5 recurrent cache-state resets are different code paths." }, { "left": "issue:41720", "right": "issue:42617", "accept": false, - "reason": "Different failures in different features; no shared concrete bug path." + "reason": "Qwen3 auto device mapping crash and a 3d_parallel.py execution failure are unrelated." }, { "left": "issue:36010", "right": "issue:42617", "accept": false, - "reason": "ImportError for GenerationMixin is unrelated to a 3d_parallel.py runtime failure." + "reason": "ImportError for GenerationMixin is not the same as a parallel script runtime failure." }, { "left": "issue:43540", "right": "issue:43782", "accept": false, - "reason": "Different multimodal model issues: video input processing vs from_pretrained weight_only loading error." + "reason": "Video-input processing error and Qwen3VL weight_only load error are different bugs." }, { "left": "issue:44186", "right": "issue:44361", "accept": false, - "reason": "Different tokenizer/task crashes; same broad area but not the same underlying bug." + "reason": "Different tokenizer failures on different models and tasks; not the same underlying defect." }, { "left": "issue:45406", "right": "issue:45464", "accept": false, - "reason": "Gemma4Processor missing _tokenizer in serve is unrelated to Qwen3.5 streaming API failure." + "reason": "Gemma4Processor missing _tokenizer in serve is unrelated to Qwen3.5 streaming inference failure." }, { - "left": "issue:43421", - "right": "issue:45701", + "left": "issue:30064", + "right": "issue:34634", "accept": false, - "reason": "TokenizersBackend post-processor sync bug is not the same as a general tokenization-version change report." + "reason": "Image segmentation map processing and BarkProcessor voice_preset are unrelated." }, { - "left": "issue:30064", - "right": "issue:34634", + "left": "issue:43421", + "right": "issue:45701", "accept": false, - "reason": "Completely different components and behaviors: image segmentation maps vs BarkProcessor voice preset." + "reason": "Runtime post-processor update bug is different from a general tokenization regression report." }, { "left": "issue:44492", "right": "issue:44509", "accept": false, - "reason": "Both are docs-related, but one is a typo and the other is removed pipeline-task documentation; not duplicates." + "reason": "A typo in cache strategies and outdated docs for removed pipeline tasks are unrelated." }, { "left": "issue:44779", "right": "issue:45812", "accept": false, - "reason": "Different tokenizer regressions affecting different model families and outputs." + "reason": "Both are tokenizer regressions, but they affect different models and likely different fixes." }, { "left": "issue:42915", "right": "issue:43322", "accept": false, - "reason": "Different model/runtime failures: FP8 config issue vs segmentation fault on Llava Next loading." + "reason": "Qwen3Moe FP8 failure and Llava Next loading segfault are unrelated." }, { "left": "issue:43606", "right": "issue:43720", "accept": false, - "reason": "Both involve loading/offload, but the concrete bugs are different model/code paths and not mergeable as one fix." + "reason": "CPU offload device mismatch and BitNet packed-weight unloading are different accelerate/loading bugs." }, { "left": "issue:43388", "right": "issue:45059", "accept": false, - "reason": "Different subsystems and symptoms: gather_for_metrics label truncation vs SAM3 PCS text/box behavior." + "reason": "Metrics label truncation and SAM3 text/bbox weirdness are different problems." }, { "left": "issue:43550", "right": "issue:43653", "accept": false, - "reason": "Different areas entirely: torch.compile/SDPA vs tokenizer special-token registration." + "reason": "torch.compile/SDPA failure and tokenizer special-token registration are unrelated." }, { "left": "issue:44514", "right": "issue:44545", "accept": true, - "reason": "Same exact crash report: `Qwen2_5_VLProcessor.apply_chat_template` fails on batched input with `padding=False`." + "reason": "Same exact Qwen2_5_VLProcessor.apply_chat_template padding=False crash; these are clear duplicates." }, { "left": "issue:43531", "right": "issue:45759", "accept": false, - "reason": "Different Qwen model/config bugs: sliding window behavior vs `from_config` text_config unwrapping." + "reason": "Qwen3-MoE sliding_window bug and composite-config unwrap behavior are different model/config issues." }, { "left": "issue:43606", "right": "issue:43653", "accept": false, - "reason": "Different bug classes and model families; no shared underlying change." + "reason": "CPU offload mismatch and tokenizer mask-token registration are unrelated CI bugs." }, { "left": "issue:43421", "right": "issue:43927", "accept": false, - "reason": "Runtime special-token update bug is unrelated to DiaConfig save/load token ID loss." + "reason": "Special-token post-processing and DiaConfig token ID persistence are different tokenizer/config defects." }, { "left": "issue:34567", "right": "issue:35707", "accept": false, - "reason": "TrainerState token count tracking is unrelated to progressive generation with embeds and cache." + "reason": "TrainerState token counter not updating and progressive generation with inputs_embeds/past_key_values are unrelated." }, { "left": "issue:44368", "right": "issue:45357", "accept": false, - "reason": "Different Qwen3.5 issues: config warning vs incorrect visual encoder keys on save_pretrained." + "reason": "Warning about tie_word_embeddings is not the same as incorrect visual encoder keys in save_pretrained." }, { "left": "issue:44451", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer-loading failures, but for different models and with no evidence of the same root cause." + "reason": "General model-loading failure for one checkpoint and tokenizer loading regression for another checkpoint are not the same bug." }, { "left": "issue:43873", "right": "issue:45704", "accept": false, - "reason": "Quantization/offload behavior is unrelated to the apex FusedRMSNorm memory leak." + "reason": "Quantization offloading behavior and apex FusedRMSNorm memory leak are unrelated." }, { "left": "issue:45759", "right": "issue:45859", "accept": false, - "reason": "Different Qwen 3.5/3.6 multimodal config handling vs CPU-offload incompatibility in rotary embedding." + "reason": "Composite Qwen config unwrapping and CPU-offload incompatibility in rotary embedding are different issues." }, { "left": "issue:44898", "right": "issue:45072", "accept": false, - "reason": "Different model execution problems: Perceiver positional interpolation vs dtype mismatch in bfloat16 inference." + "reason": "Perceiver non-default resolution failure and dtype mismatches in other models are unrelated." + }, + { + "left": "issue:41720", + "right": "issue:43746", + "accept": false, + "reason": "Qwen3 device-mapping assert and GraniteSpeech PEFT checkpoint loading are different code paths." } ] }, "evaluator_result": { "accept": true, - "feedback": "The conclusions are grounded in the packet. The lone accepted pair, issue:44514/issue:44545, is an exact title match and the earlier issue also has the fuller report, so the canonicalization choice is conservative. The other soft-edge rejections stay within what the titles support and avoid overstating duplicate evidence." + "feedback": "The canonical choice for issue:44545 is well grounded: it is effectively the same report as 44514 and has slightly more inbound attention. The soft-edge rejections are conservative and supported by the titles/body metadata; no pair is overstated as a duplicate or same-bug merge. The summary stays within the packet evidence." }, "evaluator_used": true, "retried": false, @@ -4281,8 +4317,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -4295,6 +4331,7 @@ "issue:41720", "issue:42886", "issue:42915", + "issue:43335", "issue:43381", "issue:43421", "issue:43525", @@ -4304,7 +4341,7 @@ "issue:43618", "issue:43653", "issue:43697", - "issue:43746", + "issue:43756", "issue:43782", "issue:43844", "issue:43873", @@ -4335,7 +4372,6 @@ "issue:45859" ], "soft_pairs": [ - "issue:41720|issue:43746", "issue:44062|issue:44451", "issue:44987|issue:45042", "issue:43531|issue:44589", @@ -4350,8 +4386,8 @@ "issue:43525|issue:43531", "issue:44625|issue:45479", "issue:45704|issue:45859", - "issue:43782|issue:45381", "issue:45081|issue:45356", + "issue:43782|issue:45381", "issue:34634|issue:41720", "issue:43994|issue:44671", "issue:43381|issue:45704", @@ -4361,205 +4397,206 @@ "issue:43540|issue:44008", "issue:43618|issue:43697", "issue:44220|issue:44661", + "issue:30064|issue:36010", "issue:45183|issue:45479", - "issue:30064|issue:36010" + "issue:43335|issue:43756" ], - "prepared_review_unit_hash": "cbafb5631f9b38104abeec8db3b4f4af8f7eaff0dcff30a64587688523acc641", + "prepared_review_unit_hash": "e960158f57910d53a8df1181808acf90b31086782bf4da8ba055c45ce06e878f", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 35665, - "estimated_input_tokens": 8917, - "estimated_eval_tokens": 18090 + "serialized_chars": 36240, + "estimated_input_tokens": 9060, + "estimated_eval_tokens": 18376 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous. The soft links are mostly broad semantic overlaps (tokenizers, Qwen, gradient/checkpointing, image processors) rather than the same concrete bug, so they should be rejected. No PRs are present.", - "confidence": 0.79, - "canonical_issue_reason": "issue:44843 is the most concrete, broadly relevant issue in the cluster and has a clear, actionable tokenizer/offline failure mode; it is also the closest to another soft pair in this set.", + "summary": "The set is highly heterogeneous: tokenizer, multimodal, training, quantization, and docs regressions with no convincing duplicate pairs. The strongest anchor is the Mistral tokenizer regex/offline-loading regression, but it is not a clear duplicate hub for the rest.", + "confidence": 0.91, + "canonical_issue_reason": "Best anchor in this mixed set: explicit tokenizer code-path regression with a concrete crash surface and multiple references.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44843 is a good representative because it describes a specific reproducible regression with a clearly identified code path (_patch_mistral_regex / HF_HUB_OFFLINE) and is more general than the other similarly themed tokenizer issues.", + "best_issue_reason": "Most actionable standalone bug here: precise repro, narrow fix surface, and a clear runtime failure in a core tokenizer-loading path.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:41720", - "right": "issue:43746", - "accept": false, - "reason": "Different models and failures: Qwen3 device-mapping CUDA assert vs GraniteSpeech PEFT checkpoint loading." - }, { "left": "issue:44062", "right": "issue:44451", "accept": false, - "reason": "Different bugs: AddedToken special-kwarg conflict vs failure to load a specific model checkpoint." + "reason": "Tokenizer constructor error vs model-loading failure; no shared underlying bug." }, { "left": "issue:44987", "right": "issue:45042", "accept": false, - "reason": "Unrelated failures: model loading for a specific repo vs PIL image processors requiring torchvision." + "reason": "Different subsystems: specific model load issue vs PIL/torchvision backend dependency regression." }, { "left": "issue:43531", "right": "issue:44589", "accept": false, - "reason": "Different code paths: Qwen3-MoE sliding_window behavior vs missing Float8 storage type." + "reason": "Unrelated Qwen sliding-window config bug vs PyTorch storage-type import/runtime error." }, { "left": "issue:33666", "right": "issue:35707", "accept": false, - "reason": "Qwen2-VL multi-GPU training is unrelated to progressive generation with inputs_embeds and past_key_values." + "reason": "Multi-GPU training problem and generation with inputs_embeds/past_key_values are different failure modes." }, { "left": "issue:43421", "right": "issue:43653", "accept": false, - "reason": "Both involve special tokens, but one is runtime post-processor updates and the other is BigBirdTokenizer mask-token registration." + "reason": "Both involve tokenizers, but one is runtime post-processor updates and the other is mask-token registration/empty decode behavior." }, { "left": "issue:45588", "right": "issue:45759", "accept": false, - "reason": "Different concrete problems: flash_attention s_aux=None crash vs missing text_config unwrap in composite Qwen configs." + "reason": "Different code paths and symptoms: flash-attention sink handling vs composite Qwen config unwrapping." }, { "left": "issue:42886", "right": "issue:44843", "accept": false, - "reason": "Both concern offline behavior, but one is cache loading and the other is an unconditional model_info call in _patch_mistral_regex." + "reason": "Both touch offline loading, but 44843 is a specific mistral-regex model_info call; 42886 is a broader cache-load failure." }, { "left": "issue:43381", "right": "issue:43844", "accept": false, - "reason": "Both mention gradients, but one is eval-mode checkpointing misuse and the other is ZeRO-3 gradient explosion." + "reason": "Checkpointing-in-eval issue vs ZeRO-3 gradient anomaly; unrelated bugs." }, { "left": "issue:44918", "right": "issue:44936", "accept": false, - "reason": "Trainer/SFT issues, but they are distinct failures: unpacking embeddings vs evaluate-after-train behavior." + "reason": "Different trainer issues: SFT embedding unpacking vs train/evaluate lifecycle failure." }, { "left": "issue:39692", "right": "issue:44610", "accept": false, - "reason": "Both are model/processor mismatches, but they affect different models and different preprocessing mistakes." + "reason": "Both are processor/model mismatches, but for different models and different root causes." }, { "left": "issue:43421", "right": "issue:44186", "accept": false, - "reason": "Different tokenizer defects: runtime post-processor sync vs NER/batched padding crashes." + "reason": "Different tokenizer bugs: post-processor refresh vs NER padding/truncation crash." }, { "left": "issue:43525", "right": "issue:43531", "accept": false, - "reason": "Unrelated Qwen3-MoE sliding_window bug vs missing pad_token_id on Llama4Config." + "reason": "Missing config attribute vs sliding_window behavior; not the same defect." }, { "left": "issue:44625", "right": "issue:45479", "accept": false, - "reason": "Both involve num_labels, but one is config propagation and the other is degenerate loss for num_labels=1." + "reason": "Shared num_labels wording, but one is config propagation and the other is classification-loss behavior." }, { "left": "issue:45704", "right": "issue:45859", "accept": false, - "reason": "Different components: T5 apex RMSNorm memory leak vs CPU-offload incompatibility in Qwen3_5MoeTextRotaryEmbedding." + "reason": "Different models and mechanisms: apex RMSNorm leak vs CPU-offload incompatibility in rotary embedding." }, { - "left": "issue:43782", - "right": "issue:45381", + "left": "issue:45081", + "right": "issue:45356", "accept": false, - "reason": "Different multimodal bugs: weight_only loading error vs wrong video vision_position_ids." + "reason": "Same broad regex-patching area, but one is a tokenizer-load crash and the other is a Kimi codec regression; not clearly one bug." }, { - "left": "issue:45081", - "right": "issue:45356", + "left": "issue:43782", + "right": "issue:45381", "accept": false, - "reason": "Related tokenizer area, but not the same bug: one crashes in _patch_mistral_regex, the other is a Kimi-K2.5 regression with codec handling/warning text." + "reason": "Different Qwen VL issues: weight-only loading error vs video vision_position_ids mismatch." }, { "left": "issue:34634", "right": "issue:41720", "accept": false, - "reason": "Different models and behaviors: BarkProcessor voice_preset vs Qwen3 auto device mapping CUDA assert." + "reason": "Bark voice preset bug and Qwen3 auto-device-map CUDA assert are unrelated." }, { "left": "issue:43994", "right": "issue:44671", "accept": false, - "reason": "Different model-specific output regressions: SigLIP2 pipeline/model mismatch vs CamemBERT masked LM predictions." + "reason": "Different model regressions with no shared code path." }, { "left": "issue:43381", "right": "issue:45704", "accept": false, - "reason": "Unrelated gradient-checkpointing mode issue vs T5 RMSNorm memory leak." + "reason": "Eval-mode checkpointing bug vs T5 RMSNorm memory leak; unrelated." }, { "left": "issue:39692", "right": "issue:43873", "accept": false, - "reason": "Different problems: documentation/model mismatch vs quantization offloading behavior." + "reason": "Docs/example mismatch vs quantization offloading behavior; not the same underlying issue." }, { "left": "issue:42915", "right": "issue:45759", "accept": false, - "reason": "Different Qwen families and failures: FineGrainedFP8Config error vs missing text_config unwrap." + "reason": "Different Qwen issues: FP8 config failure vs composite-config unwrapping." }, { "left": "issue:43596", "right": "issue:44220", "accept": false, - "reason": "Different runtime errors in unrelated paths: ZeRO-3/BertModel init vs _torch_extract_fbank_features." + "reason": "Deepspeed/Bert init error vs audio fbank feature extraction; unrelated." }, { "left": "issue:43540", "right": "issue:44008", "accept": false, - "reason": "Different multimodal failures: Qwen3OmniMoe video processing vs Gemma 3n audio_mel_mask name collision." + "reason": "Both are multimodal, but they fail in different modalities and from different root causes." }, { "left": "issue:43618", "right": "issue:43697", "accept": false, - "reason": "Different issues: CLIPOutput attentions assignment regression vs RTDetrV2 output differences in v5." + "reason": "Different model regressions; only both are output-related at a very high level." }, { "left": "issue:44220", "right": "issue:44661", "accept": false, - "reason": "Different subsystems: audio feature extraction vs add-new-model-like/tokenizer mapping logic." + "reason": "Audio feature extraction bug vs tokenizer-mapping extension bug." + }, + { + "left": "issue:30064", + "right": "issue:36010", + "accept": false, + "reason": "Image segmentation processing bug vs GenerationMixin import regression; unrelated." }, { "left": "issue:45183", "right": "issue:45479", "accept": false, - "reason": "Generic validation error messaging vs a sequence-classification loss/label bug; not the same underlying defect." + "reason": "Input-validation XOR error vs single-label classification loss bug; different code paths." }, { - "left": "issue:30064", - "right": "issue:36010", + "left": "issue:43335", + "right": "issue:43756", "accept": false, - "reason": "Completely unrelated: void segmentation map processing vs GenerationMixin import failure." + "reason": "Sparse-layer config bug vs RoPE-layer dropping mismatch; not the same issue." } ] }, @@ -4578,17 +4615,17 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ "issue:30064", "issue:36331", + "issue:39692", "issue:42915", "issue:43295", "issue:43322", - "issue:43335", "issue:43421", "issue:43531", "issue:43540", @@ -4599,7 +4636,6 @@ "issue:43716", "issue:43720", "issue:43746", - "issue:43756", "issue:43782", "issue:43818", "issue:43819", @@ -4611,6 +4647,7 @@ "issue:44077", "issue:44186", "issue:44448", + "issue:44479", "issue:44485", "issue:44488", "issue:44509", @@ -4634,7 +4671,6 @@ ], "soft_pairs": [ "issue:43720|issue:43819", - "issue:43335|issue:43756", "issue:44077|issue:44625", "issue:43653|issue:43720", "issue:43716|issue:44008", @@ -4647,8 +4683,8 @@ "issue:44186|issue:44898", "issue:44843|issue:45081", "issue:44062|issue:44488", - "issue:43531|issue:43643", "issue:43577|issue:43818", + "issue:43531|issue:43643", "issue:43643|issue:44877", "issue:43421|issue:44811", "issue:43540|issue:45381", @@ -4658,213 +4694,221 @@ "issue:45584|issue:45834", "issue:43322|issue:45685", "issue:43825|issue:43994", - "issue:44561|issue:45020", "issue:44821|issue:45084", + "issue:44561|issue:45020", "issue:44485|issue:44509", - "issue:30064|issue:36331" + "issue:30064|issue:36331", + "issue:39692|issue:45834", + "issue:44479|issue:45834" ], - "prepared_review_unit_hash": "581afd0ed426f7995d3c02bdaaa8d01fedf7bd5c76d40daf4317db52df2d72d9", + "prepared_review_unit_hash": "5556f93240bfdb776111a24644ea8792bc4371cb0b8413fc464f09618ec9e96e", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36556, - "estimated_input_tokens": 9139, - "estimated_eval_tokens": 18534 + "soft_pair_count": 30, + "serialized_chars": 36855, + "estimated_input_tokens": 9214, + "estimated_eval_tokens": 18684 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is highly heterogeneous: most items are unrelated model-, tokenizer-, config-, or docs-level regressions. The soft matches mostly share only subsystem vocabulary, not the same concrete bug, so none should be merged as duplicates.", - "confidence": 0.94, - "canonical_issue_reason": "If one issue must anchor the cluster, issue:45020 is the broadest/open report and best matches a general umbrella of recent remote_code regressions.", + "summary": "This cluster is not a true duplicate set: it mixes many unrelated bugs across tokenizers, multimodal processors, quantization/loading, and version regressions. The soft links are mostly title-level similarity, not the same concrete code-path failure.", + "confidence": 0.95, + "canonical_issue_reason": "No single canonical issue fits well; the items are heterogeneous and do not describe one underlying bug.", "canonical_pr_reason": null, - "best_issue_reason": "Issue:45020 is the most representative of the cluster's broad regression theme, but only weakly so because the surrounding items are not true duplicates.", + "best_issue_reason": "issue:45020 is the broadest umbrella complaint in the set (recent transformers versions breaking remote_code models), but it still does not subsume the rest as duplicates.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43720", "right": "issue:43819", "accept": false, - "reason": "Both are loading/runtime failures, but one is BitNet packed-weight unpacking during accelerate loading and the other is DAC from_latents vs forward mismatch from missing STE. Different code paths and bugs." - }, - { - "left": "issue:43335", - "right": "issue:43756", - "accept": false, - "reason": "SwitchTransformers config layer-count bug and Smollm3 RoPE-layer omission are unrelated model/config issues." + "reason": "BitNet accelerate-loading failure vs DAC latent/STE mismatch; unrelated model paths." }, { "left": "issue:44077", "right": "issue:44625", "accept": false, - "reason": "One is about patchtsmixer post_init policy; the other is Qwen3.5 num_labels propagation. Same broad config area, different bugs." + "reason": "PatchTSMixer post-init policy vs Qwen3.5 num_labels propagation; different subsystems." }, { "left": "issue:43653", "right": "issue:43720", "accept": false, - "reason": "Tokenizer special-token decode issue vs BitNet packed-weight loading bug; no shared underlying failure." + "reason": "BigBird tokenizer special-token decode bug vs BitNet weight loading bug; not the same issue." }, { "left": "issue:43716", "right": "issue:44008", "accept": false, - "reason": "Both are multimodal dtype-ish reports, but one is Mistral image preprocessor/model dtype mismatch and the other is a Gemma3n variable-name collision in forward()." + "reason": "Mistral image dtype mismatch vs Gemma3n audio variable collision; different code paths and modalities." }, { "left": "issue:42915", "right": "issue:43746", "accept": false, - "reason": "Qwen3Moe FineGrainedFP8Config failure and GraniteSpeech PEFT checkpoint loading are unrelated." + "reason": "Qwen3Moe FP8 config failure vs GraniteSpeech PEFT/local checkpoint loading; unrelated failures." }, { "left": "issue:43295", "right": "issue:45834", "accept": false, - "reason": "Processor.tokenizer regression and Kosmos2.5 OCR long-input index error are different code paths." + "reason": "Processor/tokenizer regression vs Kosmos2.5 OCR index error; only broad multimodal overlap." }, { "left": "issue:43746", "right": "issue:45381", "accept": false, - "reason": "GraniteSpeech local-checkpoint loading and qwen2.5-vl video position-id bug are unrelated model-specific issues." + "reason": "GraniteSpeech adapter loading vs Qwen2.5-VL video position ids; different models and bugs." }, { "left": "issue:43550", "right": "issue:45072", "accept": false, - "reason": "Both mention dtype/compile concerns, but they affect different models and failures (Bamba torch.compile SDPA vs SwitchTransformers/TimmWrapper bfloat16 mismatch)." + "reason": "torch.compile SDPA issue vs bf16 dtype mismatch; not the same concrete problem." }, { "left": "issue:43746", "right": "issue:43873", "accept": false, - "reason": "GraniteSpeech PEFT loading and quantization offloading behavior are not the same bug." + "reason": "PEFT checkpoint loading vs quantization offloading behavior; distinct failure modes." }, { "left": "issue:43550", "right": "issue:43819", "accept": false, - "reason": "Bamba SDPA torch.compile failure and DAC latent/STE mismatch are unrelated." + "reason": "Bamba SDPA/compile bug vs DAC from_latents mismatch; unrelated." }, { "left": "issue:44186", "right": "issue:44898", "accept": false, - "reason": "LayoutLMv2Tokenizer NER/padding crash and Perceiver interpolation failure are different tokenizer vs vision-model bugs." + "reason": "Tokenizer padding/truncation crash vs image classification/interpolation failure; different components." }, { "left": "issue:44843", "right": "issue:45081", "accept": false, - "reason": "Both touch _patch_mistral_regex, but one is an offline model_info call and the other is a backend_tokenizer AttributeError. Same helper, different failure mode." + "reason": "Same helper function area, but different bugs: offline model_info call vs backend_tokenizer AttributeError." }, { "left": "issue:44062", "right": "issue:44488", "accept": false, - "reason": "AddedToken special-argument collision is unrelated to the sleng-bert loading problem." + "reason": "AddedToken special-arg conflict vs model loading failure; no concrete duplicate evidence." }, { - "left": "issue:43531", - "right": "issue:43643", + "left": "issue:43577", + "right": "issue:43818", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior and trust_remote_code missing-fields config loading are unrelated." + "reason": "Blip2 dtype propagation bug vs Video-LLaVA tower architecture issue; unrelated." }, { - "left": "issue:43577", - "right": "issue:43818", + "left": "issue:43531", + "right": "issue:43643", "accept": false, - "reason": "Blip2 dtype persistence and Video-LLaVA temporal-attention/weight-sharing are different bugs." + "reason": "Qwen3-MoE sliding window bug vs trust_remote_code config field loss; distinct areas." }, { "left": "issue:43643", "right": "issue:44877", "accept": false, - "reason": "Both are config-loading regressions, but one is trust_remote_code missing fields and the other is strict config rejection for granite_speech." + "reason": "Both involve config loading, but one is missing fields and the other is strict GraniteSpeech config rejection." }, { "left": "issue:43421", "right": "issue:44811", "accept": false, - "reason": "Special-token post-processor updates and Whisper skip_special_tokens handling are related only at a high level, not the same bug." + "reason": "Tokenizer backend post-processor sync issue vs Whisper batch_decode skip_special_tokens behavior; not the same bug." }, { "left": "issue:43540", "right": "issue:45381", "accept": false, - "reason": "Both involve video inputs, but Qwen3OmniMoe video processing and qwen2.5-vl vision_position_ids are distinct issues." + "reason": "Different video-input regressions in different Qwen model families; not mergeable as one fix." }, { "left": "issue:43653", "right": "issue:43819", "accept": false, - "reason": "BigBirdTokenizer mask-token decoding and DAC forward/from_latents mismatch are unrelated." + "reason": "BigBird special-token registration bug vs DAC latent mismatch; unrelated." }, { "left": "issue:44448", "right": "issue:44671", "accept": false, - "reason": "Both are v4/v5 behavior regressions, but on different models and code paths; not duplicates." + "reason": "Different models and outputs (Pegasus vs CamemBERT); no shared concrete code-path." }, { "left": "issue:43782", "right": "issue:44977", "accept": false, - "reason": "Qwen3VL weight_only loading error and Qwen3.5 flash-attention generation bug are different Qwen model failures." + "reason": "Qwen3VL from_pretrained weight_only error vs Qwen3.5 flash-attention generation issue; distinct failures." }, { "left": "issue:45584", "right": "issue:45834", "accept": false, - "reason": "Whisper empty-transcription generation and Kosmos2.5 long OCR index error are unrelated." + "reason": "Whisper empty-transcription generation bug vs Kosmos2.5 OCR long-input index error; unrelated." }, { "left": "issue:43322", "right": "issue:45685", "accept": false, - "reason": "Llava Next segfault and MPS histogram Int error are unrelated runtime failures." + "reason": "Llava Next load segfault vs MPS histogram Int limitation; completely different areas." }, { "left": "issue:43825", "right": "issue:43994", "accept": false, - "reason": "Docs/error-message mismatch for removed pipeline tasks and SigLIP2 bad outputs are different issues." + "reason": "Pipeline docs/error-message issue vs SigLIP2 bad predictions; not the same bug." }, { - "left": "issue:44561", - "right": "issue:45020", + "left": "issue:44821", + "right": "issue:45084", "accept": false, - "reason": "Both concern remote_code regressions, but 44561 is a specific missing utility and 45020 is a broad umbrella report; not the same underlying bug." + "reason": "AutoImageProcessor URL loading vs template-node compilation error; unrelated." }, { - "left": "issue:44821", - "right": "issue:45084", + "left": "issue:44561", + "right": "issue:45020", "accept": false, - "reason": "AutoImageProcessor URL loading and template-node compilation errors are unrelated." + "reason": "Related broad theme around remote_code regressions, but these are not the same concrete breakage." }, { "left": "issue:44485", "right": "issue:44509", "accept": false, - "reason": "GLM-5 RoPE implementation discussion and v5 docs cleanup are unrelated." + "reason": "GLM-5 RoPE implementation discussion vs documentation task-name mismatch; no duplicate overlap." }, { "left": "issue:30064", "right": "issue:36331", "accept": false, - "reason": "Void segmentation-map processing and CustomTrainer.compute_loss signature breakage have no common code path." + "reason": "Image segmentation-map processing bug vs CustomTrainer compute_loss signature change; unrelated." + }, + { + "left": "issue:39692", + "right": "issue:45834", + "accept": false, + "reason": "SigLIP2 docs/model mismatch and quantization failure vs Kosmos2.5 OCR indexing bug; different issues." + }, + { + "left": "issue:44479", + "right": "issue:45834", + "accept": false, + "reason": "Qwen video-input regression vs Kosmos2.5 OCR long-input index error; only generic multimodal similarity." } ] }, @@ -4883,14 +4927,13 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ "issue:34634", "issue:36010", - "issue:39692", "issue:42175", "issue:43262", "issue:43329", @@ -4907,6 +4950,7 @@ "issue:43697", "issue:43720", "issue:43742", + "issue:43746", "issue:43749", "issue:43782", "issue:43819", @@ -4917,10 +4961,10 @@ "issue:44186", "issue:44368", "issue:44442", - "issue:44479", "issue:44610", "issue:44671", "issue:44805", + "issue:44811", "issue:44857", "issue:44912", "issue:44936", @@ -4934,12 +4978,9 @@ "issue:45538", "issue:45600", "issue:45701", - "issue:45803", - "issue:45834" + "issue:45803" ], "soft_pairs": [ - "issue:39692|issue:45834", - "issue:44479|issue:45834", "issue:43825|issue:43827", "issue:44186|issue:44442", "issue:43606|issue:45072", @@ -4951,9 +4992,9 @@ "issue:44610|issue:45072", "issue:43720|issue:45072", "issue:45081|issue:45406", + "issue:43404|issue:43782", "issue:44060|issue:44368", "issue:43650|issue:43906", - "issue:43404|issue:43782", "issue:44671|issue:45701", "issue:43582|issue:44857", "issue:43262|issue:43819", @@ -4962,200 +5003,195 @@ "issue:45084|issue:45538", "issue:43644|issue:43749", "issue:43329|issue:45600", - "issue:43653|issue:44912", "issue:43742|issue:45020", - "issue:43638|issue:44936" + "issue:43653|issue:44912", + "issue:43638|issue:44936", + "issue:43746|issue:44811" ], - "prepared_review_unit_hash": "cf251d1549d265c23d29f66c2da4d404863d954f73873b874f1cfd1dd8e15ae9", + "prepared_review_unit_hash": "ceb51bba412330ce799f5e96eb586fb67791c88f7b9984d44673c40e03202c3f", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 27, - "serialized_chars": 36120, - "estimated_input_tokens": 9030, - "estimated_eval_tokens": 18316 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35264, + "estimated_input_tokens": 8816, + "estimated_eval_tokens": 17888 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No soft edge looks like a true duplicate; the pairs are mostly unrelated model-specific regressions or, at most, adjacent documentation/runtime issues. The broadest issue in the set is the open remote_code regression tracker.", - "confidence": 0.7, - "canonical_issue_reason": "issue:45020 is the broadest open umbrella here, covering recent-version regressions that break model loading under remote_code; most other issues are narrower, model-specific bugs.", + "summary": "This cluster is mostly a grab bag of distinct Transformers regressions across tokenizers, processors, loading paths, dtype/device issues, and training/runtime errors. I found no true duplicate pair among the soft edges; the closest matches are only topic-level overlaps, not the same underlying bug.", + "confidence": 0.9, + "canonical_issue_reason": "Use issue 45020 as the canonical issue: it is the broadest umbrella report about recent Transformers versions breaking model behavior, and it best represents the general regression theme in this otherwise heterogeneous cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 best represents the cluster because it is open, general, and captures the shared Transformers-v5 regression theme better than the single-model reports.", + "best_issue_reason": "Issue 45020 is the best representative because it is open, high-level, and covers the widest slice of version-related breakage; the rest are narrower, model-specific bugs that do not collapse cleanly into one concrete root cause.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:39692", - "right": "issue:45834", - "accept": false, - "reason": "SigLIP2 doc example errors and Kosmos2.5 OCR index errors are unrelated models and failure modes." - }, - { - "left": "issue:44479", - "right": "issue:45834", - "accept": false, - "reason": "Qwen VLM video-input regression is a different code path from Kosmos2.5 long-OCR indexing." - }, { "left": "issue:43825", "right": "issue:43827", "accept": false, - "reason": "One is a runtime error message about translation pipelines; the other is documentation still mentioning pipeline()." + "reason": "Both involve pipeline/translation deprecation fallout, but one is an error message bug and the other is outdated docs; different artifacts and fixes." }, { "left": "issue:44186", "right": "issue:44442", "accept": false, - "reason": "LayoutLMv2 tokenizer NER/padding crashes and FastSpeech2Conformer tokenizer loading are unrelated tokenizer bugs." + "reason": "Different tokenizer failures on different models and code paths; only broad tokenizer-loading similarity." }, { "left": "issue:43606", "right": "issue:45072", "accept": false, - "reason": "Bark CPU offload device mismatch and SwitchTransformers/TimmWrapperModel dtype mismatch are different loading failures." + "reason": "Device mismatch in Bark CPU offload vs dtype mismatch in other models; unrelated concrete bugs." }, { "left": "issue:43577", "right": "issue:43582", "accept": false, - "reason": "BLIP2 dtype propagation and Apple Silicon allocator warmup TypeError do not share the same bug." + "reason": "BLIP2 dtype propagation vs Apple Silicon allocator TypeError; no shared underlying defect." }, { "left": "issue:34634", "right": "issue:36010", "accept": false, - "reason": "BarkProcessor voice_preset behavior and GenerationMixin import failure are unrelated." + "reason": "BarkProcessor voice_preset bug is unrelated to a transformers.GenerationMixin import error." }, { "left": "issue:44805", "right": "issue:44936", "accept": false, - "reason": "A tensor mask shape IndexError is not the same failure as trainer.evaluate() after trainer.train()." + "reason": "Different failures: mask/index shape mismatch vs trainer lifecycle failure after train()." }, { "left": "issue:43452", "right": "issue:44991", "accept": false, - "reason": "gguf_file/pretrained-loading breakage and est-roberta tokenizer loading are different loader paths." + "reason": "Both are loading/tokenizer issues, but the models and root causes differ; not the same bug." }, { "left": "issue:42175", "right": "issue:45081", "accept": false, - "reason": "TensorFlow backend packaging and Mistral regex patch crashes are unrelated." + "reason": "Backend dependency packaging issue vs tokenizer regex patch crash; unrelated." }, { "left": "issue:44610", "right": "issue:45072", "accept": false, - "reason": "OmDet-Turbo image-size mismatch is unrelated to dtype mismatch inference bugs." + "reason": "Processor image size mismatch vs inference dtype mismatch; different code paths." }, { "left": "issue:43720", "right": "issue:45072", "accept": false, - "reason": "BitNet packed-weight unpacking and SwitchTransformers/TimmWrapperModel dtype issues are different concrete bugs." + "reason": "Packed-weight loading bug in BitNet vs dtype mismatch in inference; not mergeable as one fix." }, { "left": "issue:45081", "right": "issue:45406", "accept": false, - "reason": "Mistral tokenizer regex patching and Gemma4Processor missing _tokenizer are separate tokenizer/processor failures." + "reason": "Both are attribute errors during loading, but one is Mistral tokenizer regex patching and the other is Gemma4 serve processor state." + }, + { + "left": "issue:43404", + "right": "issue:43782", + "accept": false, + "reason": "Mistral lm_head tying bug vs Qwen3VL weight_only load error; different models and mechanisms." }, { "left": "issue:44060", "right": "issue:44368", "accept": false, - "reason": "Both mention tied-weight warnings, but they concern different models and different underlying causes." + "reason": "Both mention tied-weights warnings, but the underlying models and behavior are different warning/reporting issues." }, { "left": "issue:43650", "right": "issue:43906", "accept": false, - "reason": "A placeholder 'ADD THE DATA' issue is not a duplicate of an isolated reproduction report." - }, - { - "left": "issue:43404", - "right": "issue:43782", - "accept": false, - "reason": "Mistral lm_head tying and Qwen3VL weight_only loading error are unrelated model-specific problems." + "reason": "Placeholder/irrelevant issue vs isolated reproduction of another issue; no duplication." }, { "left": "issue:44671", "right": "issue:45701", "accept": false, - "reason": "CamemBERT masked-LM regression and a general tokenization-change report are not the same bug." + "reason": "CamemBERT masked LM regression vs a general tokenization-change report; only loose version-related similarity." }, { "left": "issue:43582", "right": "issue:44857", "accept": false, - "reason": "Apple Silicon allocator warmup TypeError and CUDA float16 loss crash are unrelated platform-specific issues." + "reason": "Allocator warmup TypeError and AMP crash are separate runtime failures." }, { "left": "issue:43262", "right": "issue:43819", "accept": false, - "reason": "Audio apply_chat_template sampling-rate defaulting and DAC latent reconstruction mismatch are different behaviors." + "reason": "Audio sampling-rate defaulting bug vs DAC latent reconstruction mismatch; unrelated." }, { "left": "issue:43526", "right": "issue:43697", "accept": false, - "reason": "BEiT FastImageProcessor label reduction and RTDetrV2 output drift are unrelated model/processor bugs." + "reason": "BEiT label reduction bug vs RTDetrV2 output drift; different models and fixes." }, { "left": "issue:45459", "right": "issue:45803", "accept": false, - "reason": "Both involve exception handling, but they are different functions with different failure surfaces." + "reason": "Both involve overly broad exception handling, but they are different functions and failure modes." }, { "left": "issue:45084", "right": "issue:45538", "accept": false, - "reason": "Template-compilation TypeError and CLIPTokenizer model_max_length are unrelated tokenizer/template issues." + "reason": "Template compilation TypeError vs tokenizer max_length constant misuse; not the same bug." }, { "left": "issue:43644", "right": "issue:43749", "accept": false, - "reason": "Non-persistent buffer junk and FSDP CPU RAM efficient loading breakage are different loading bugs." + "reason": "Non-persistent buffer corruption in v5 vs FSDP CPU RAM efficient loading breakage; different subsystems." }, { "left": "issue:43329", "right": "issue:45600", "accept": false, - "reason": "Undefined video-token helpers and stale auto_mappings references are unrelated regressions." + "reason": "Undefined multimodal-token helper bug vs stale auto_mappings references; unrelated maintenance issues." }, { - "left": "issue:43653", - "right": "issue:44912", + "left": "issue:43742", + "right": "issue:45020", "accept": false, - "reason": "BigBird special-token decode failure and MXFP4 quantization fallback are unrelated." + "reason": "MobileLLM key error may be version-related, but it is a distinct loading failure from the broad remote_code regression report." }, { - "left": "issue:43742", - "right": "issue:45020", + "left": "issue:43653", + "right": "issue:44912", "accept": false, - "reason": "A key error for one model is too specific to be the same underlying bug as the broad remote_code regression tracker." + "reason": "BigBird tokenizer special-token bug vs MXFP4 quantization fallback; no shared root cause." }, { "left": "issue:43638", "right": "issue:44936", "accept": false, - "reason": "DeepSpeed ZeRO-3 with a non-pretrained Bert model and post-train evaluate failure are different training/runtime issues." + "reason": "Deepspeed zero3 indexing error vs trainer evaluate-after-train failure; separate training bugs." + }, + { + "left": "issue:43746", + "right": "issue:44811", + "accept": false, + "reason": "PEFT/local checkpoint loading issue vs Whisper batch_decode skip_special_tokens bug; different code paths." } ] }, @@ -5174,8 +5210,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -5186,23 +5222,24 @@ "issue:38175", "issue:39692", "issue:40990", - "issue:41720", + "issue:42915", "issue:43257", "issue:43262", "issue:43329", "issue:43450", + "issue:43452", "issue:43454", "issue:43526", "issue:43531", - "issue:43540", "issue:43550", "issue:43577", + "issue:43582", "issue:43606", "issue:43643", "issue:43645", "issue:43697", "issue:43720", - "issue:43746", + "issue:43742", "issue:43749", "issue:43819", "issue:43854", @@ -5228,17 +5265,16 @@ "issue:45859" ], "soft_pairs": [ - "issue:43746|issue:44811", "issue:44589|issue:45042", "issue:43854|issue:44977", "issue:38175|issue:44610", - "issue:43526|issue:44554", "issue:39692|issue:44912", + "issue:43526|issue:44554", "issue:36010|issue:43645", "issue:34634|issue:36331", "issue:43819|issue:43927", - "issue:43257|issue:43329", "issue:43454|issue:43577", + "issue:43257|issue:43329", "issue:43720|issue:44610", "issue:43550|issue:44912", "issue:44568|issue:45488", @@ -5249,231 +5285,235 @@ "issue:36010|issue:36331", "issue:43643|issue:45759", "issue:44849|issue:45701", - "issue:34567|issue:40990", "issue:43329|issue:43450", + "issue:34567|issue:40990", "issue:43329|issue:44479", "issue:43606|issue:45859", "issue:43257|issue:43931", "issue:43697|issue:45357", "issue:45823|issue:45859", "issue:44291|issue:45042", - "issue:41720|issue:43540" + "issue:43452|issue:43582", + "issue:42915|issue:43742" ], - "prepared_review_unit_hash": "34a231f632c19b9b90b7b070c40cdb5bfa1932ccfe4a7e7aa5eebc9620f3db25", + "prepared_review_unit_hash": "db3ecc0d08eadbf788e6b89d0c5c92bc06d6036cfc67a6a90f1178f0f33fcc0c", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36446, - "estimated_input_tokens": 9112, - "estimated_eval_tokens": 18480 + "serialized_chars": 36974, + "estimated_input_tokens": 9244, + "estimated_eval_tokens": 18744 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is heterogeneous: it mixes unrelated Transformers regressions across loading, tokenization, vision/audio processors, and model-specific inference paths. None of the soft pairs looks like a true duplicate; the best representative is the broad tokenizer-regression issue.", - "confidence": 0.31, - "canonical_issue_reason": "issue:45701 is the broadest, most umbrella-style bug here and best matches the recurring tokenizer/version-regression theme; it is more representative than the many narrow model-specific reports.", + "summary": "This cluster is mostly a set of unrelated closed issues from the same repo; only one soft pair looks like the same underlying regression in the video multimodal path.", + "confidence": 0.77, + "canonical_issue_reason": "issue:45701 is the broadest and most reusable issue in the set: it is open, has inbound references, and describes a general tokenization regression that can plausibly anchor related tokenizer behavior reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45701 is the best cluster representative because it describes a version-wide tokenization regression rather than a single model or narrow code path, making it the most reusable anchor for this set.", + "best_issue_reason": "issue:45701 is the best overall issue candidate because it is broad, still open, and likely to attract/organize follow-up reports around tokenizer behavior changes.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43746", - "right": "issue:44811", - "accept": false, - "reason": "Different subsystems: PEFT/local checkpoint loading vs Whisper batch decoding behavior." - }, { "left": "issue:44589", "right": "issue:45042", "accept": false, - "reason": "Unrelated bugs: Float8 storage lookup vs PIL backend image-processor dependency handling." + "reason": "Different failures: one is a storage/backend type error, the other is an image-processor dependency regression." }, { "left": "issue:43854", "right": "issue:44977", "accept": false, - "reason": "Both are model-specific regressions, but they affect different models and different code paths (loading vs flash-attention generation)." + "reason": "Different models and code paths: GLM loading/unit-test failure vs Qwen3.5 flash-attention generation behavior." }, { "left": "issue:38175", "right": "issue:44610", "accept": false, - "reason": "Vision-related, but one is a probability/output issue and the other is an input-size preprocessing mismatch." + "reason": "Unrelated subsystems: SigLIP2 probability output issue vs OmDet-Turbo image size mismatch." }, { - "left": "issue:43526", - "right": "issue:44554", + "left": "issue:39692", + "right": "issue:44912", "accept": false, - "reason": "Different areas: Beit image label reduction vs MPS attention correctness." + "reason": "Docs/example errors for SigLIP2 are not the same bug as MXFP4 quantization fallback for git-oss-20b." }, { - "left": "issue:39692", - "right": "issue:44912", + "left": "issue:43526", + "right": "issue:44554", "accept": false, - "reason": "Docs/example issues and quantization fallback are not the same underlying bug." + "reason": "Different components: Beit image label reduction vs an MPS attention correctness issue." }, { "left": "issue:36010", "right": "issue:43645", "accept": false, - "reason": "Both involve loading/import behavior, but one is an import path break and the other is custom model initialization in notebooks." + "reason": "Both are import/model-init regressions in v5, but they affect different mechanisms and are not the same code-path bug." }, { "left": "issue:34634", "right": "issue:36331", "accept": false, - "reason": "Audio processor preset handling and Trainer API incompatibility are separate bugs." + "reason": "BarkProcessor voice preset handling is unrelated to CustomTrainer.compute_loss signature changes." }, { "left": "issue:43819", "right": "issue:43927", "accept": false, - "reason": "Different code paths: DAC latent decoding vs config save/load token IDs." + "reason": "Different model behaviors: DAC latent reconstruction mismatch vs DiaConfig losing custom token IDs on save/load." }, { - "left": "issue:43257", - "right": "issue:43329", + "left": "issue:43454", + "right": "issue:43577", "accept": false, - "reason": "Both mention Qwen3/MoE and multimodal code, but one is weight conversion during loading and the other is a video token-counting bug." + "reason": "Different model families and failures: weight tying in AyaVision vs dtype propagation in Blip2." }, { - "left": "issue:43454", - "right": "issue:43577", + "left": "issue:43257", + "right": "issue:43329", "accept": false, - "reason": "Both are model-loading related, but the failures are different: tied weights vs dtype propagation." + "reason": "Qwen3 MoE weight conversion/loading is unrelated to the multimodal token-counting video-branch undefined-variable bug." }, { "left": "issue:43720", "right": "issue:44610", "accept": false, - "reason": "Packed-weight loading and image processor output shape are unrelated." + "reason": "Packed-weight loading for BitNet is unrelated to the OmDet-Turbo processor input-size regression." }, { "left": "issue:43550", "right": "issue:44912", "accept": false, - "reason": "torch.compile/SDPA failure is unrelated to MXFP4 quantization loading behavior." + "reason": "torch.compile + SDPA failure in Bamba is a different issue than MXFP4 quantization fallback in git-oss-20b." }, { "left": "issue:44568", "right": "issue:45488", "accept": false, - "reason": "Both are tokenization regressions, but they affect different tokenizers and different mechanics (special tokens vs pre-tokenizer override)." + "reason": "Both are tokenizer regressions, but they involve different tokenizers and different failure modes." }, { "left": "issue:43749", "right": "issue:44291", "accept": false, - "reason": "Both are loading-related, but FSDP CPU RAM loading and init_empty_weights have different failure modes and code paths." + "reason": "Related loading stack, but not the same bug: FSDP CPU RAM efficient loading vs init_empty_weights passing an unexpected argument." }, { "left": "issue:43531", "right": "issue:45042", "accept": false, - "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to PIL/torchvision processor dependency issues." + "reason": "Qwen3-MoE sliding-window behavior is unrelated to PIL backend torchvision requirements." }, { "left": "issue:43262", "right": "issue:44811", "accept": false, - "reason": "Audio chat-template sampling-rate defaults and Whisper batch_decode token skipping are different processor behaviors." + "reason": "Both are audio processor issues, but one is chat-template sampling rate and the other is batch_decode skip_special_tokens handling." }, { "left": "issue:44568", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer regressions, but they are distinct bugs affecting different models and failure mechanisms." + "reason": "Different tokenizer regressions affecting different models; not the same underlying bug." }, { "left": "issue:36010", "right": "issue:36331", "accept": false, - "reason": "ImportError from generation module and CustomTrainer signature breakage are unrelated." + "reason": "Importing GenerationMixin and CustomTrainer.compute_loss keyword handling are unrelated failures." }, { "left": "issue:43643", "right": "issue:45759", "accept": false, - "reason": "Both touch config/model loading, but missing custom fields in AutoConfig is not the same as text_config unwrapping for multimodal causal LM creation." + "reason": "AutoConfig remote-code field loss is not the same issue as unwrapping text_config in composite multimodal configs." }, { "left": "issue:44849", "right": "issue:45701", "accept": false, - "reason": "Qwen3.5 hidden-state output regression is not a tokenization-version issue." + "reason": "Qwen3.5 output_hidden_states behavior is a model forward bug, not the broader tokenization regression." }, { - "left": "issue:34567", - "right": "issue:40990", + "left": "issue:43329", + "right": "issue:43450", "accept": false, - "reason": "Trainer token counters and perplexity evaluation are unrelated." + "reason": "Both involve video, but one is token counting with undefined vars and the other is batched processor output shape; different bugs." }, { - "left": "issue:43329", - "right": "issue:43450", + "left": "issue:34567", + "right": "issue:40990", "accept": false, - "reason": "Both are video-related, but one is token counting/undefined names and the other is batched processor output shape." + "reason": "TrainerState token counters and GPT-OSS perplexity are unrelated." }, { "left": "issue:43329", "right": "issue:44479", - "accept": false, - "reason": "Both mention video regressions, but they appear to hit different layers: token counting vs model video-input regression." + "accept": true, + "reason": "Both describe the same video-input regression in Qwen VLM paths, centered on the video multimodal code path used by several Qwen variants." }, { "left": "issue:43606", "right": "issue:45859", "accept": false, - "reason": "Both are CPU-offload/device issues, but they affect different models and different failing operations." + "reason": "Both mention CPU offload, but they affect different models and different failure points." }, { "left": "issue:43257", "right": "issue:43931", "accept": false, - "reason": "Different loading failures: MOE weight conversion vs weight-shape mismatch." + "reason": "Qwen3 MoE loading/conversion and Qwen3-VL weight-shape mismatch are different model-loading problems." }, { "left": "issue:43697", "right": "issue:45357", "accept": false, - "reason": "Different regressions: v5 output drift for RTDetrV2 vs Qwen3.5 save_pretrained key corruption." + "reason": "RTDetrV2 output changes and Qwen3.5 save_pretrained key naming are unrelated regressions." }, { "left": "issue:45823", "right": "issue:45859", "accept": false, - "reason": "Both involve device/offload, but the underlying failures and model code paths are different." + "reason": "Different offload/device-mapping failures in different models; not the same concrete bug." }, { "left": "issue:44291", "right": "issue:45042", "accept": false, - "reason": "init_empty_weights loading and PIL backend torchvision dependency are unrelated." + "reason": "Unexpected init argument during empty-weight loading is unrelated to the torchvision dependency regression." }, { - "left": "issue:41720", - "right": "issue:43540", + "left": "issue:43452", + "right": "issue:43582", + "accept": false, + "reason": "gguf_file tokenizer/model loading and Apple Silicon allocator warmup are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:43742", "accept": false, - "reason": "Both are Qwen3-family issues, but one is auto device mapping/cuda assert and the other is video-input validation in a different model." + "reason": "Fine-grained FP8 loading for Qwen3Moe is unrelated to the MobileLLM key error." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the only accepted soft pair (issue:43329 with issue:44479) is a conservative match: both point to the same Qwen video-input regression area and could plausibly be fixed together. The canonical issue choice is also reasonable because issue:45701 is open, broad, and has inbound references. A minor caveat is that the cluster is not entirely \u2018closed,\u2019 since several open issues are present, but that does not undermine the overall conclusion." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -5485,8 +5525,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -5495,6 +5535,7 @@ "issue:36010", "issue:36331", "issue:38175", + "issue:41720", "issue:42175", "issue:42915", "issue:43257", @@ -5503,7 +5544,6 @@ "issue:43493", "issue:43540", "issue:43577", - "issue:43582", "issue:43643", "issue:43646", "issue:43696", @@ -5539,15 +5579,14 @@ "issue:45823" ], "soft_pairs": [ - "issue:43452|issue:43582", - "issue:42915|issue:43742", + "issue:41720|issue:43540", "issue:43317|issue:45538", "issue:33357|issue:34567", - "issue:43927|issue:44912", - "issue:44987|issue:45538", "issue:42915|issue:44821", - "issue:44479|issue:45381", + "issue:44987|issue:45538", "issue:42915|issue:45538", + "issue:43927|issue:44912", + "issue:44479|issue:45381", "issue:43257|issue:43756", "issue:43540|issue:44560", "issue:43540|issue:44479", @@ -5564,243 +5603,237 @@ "issue:44062|issue:45081", "issue:43749|issue:44756", "issue:34567|issue:36331", - "issue:43643|issue:45375", "issue:36010|issue:42175", "issue:43493|issue:43994", + "issue:43643|issue:45375", "issue:42175|issue:43824", - "issue:43452|issue:44488", "issue:44291|issue:44821", + "issue:43452|issue:44488", "issue:44792|issue:44964" ], - "prepared_review_unit_hash": "f2973b0975d256d41feea2602dfdce8172471a882fc5fe3413cce4f04bc08230", + "prepared_review_unit_hash": "499a1faa77392d06b864adeca38057a338435625c404c8fb93dd0034d81112ab", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 32, - "serialized_chars": 36765, - "estimated_input_tokens": 9192, - "estimated_eval_tokens": 18640 + "soft_pair_count": 31, + "serialized_chars": 36473, + "estimated_input_tokens": 9119, + "estimated_eval_tokens": 18494 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The packet is mostly heterogeneous. The only clear duplicate-worthy pairs are the Qwen video-input regression reports (44479/45381) and the SigLIP2 implementation/output reports (43493/43994). For the global issue anchor, 43824 is the stronger overall representative on the supplied activity totals, slightly ahead of 43749; I would avoid claiming a clearly reproducible regression from metadata alone.", - "confidence": 0.82, - "canonical_issue_reason": "Issue 44479 is the better duplicate anchor for the Qwen video-input pair because its title is the broader regression statement, while 45381 reads like a narrower symptom report for qwen2.5-vl video position IDs. That makes 44479 the more general anchor based on the packet text.", + "summary": "This cluster is largely heterogeneous and contains several unrelated issue families. The only strong duplicate candidates are the Qwen video-input regression pair and the SigLIP2 correctness pair; the rest are not the same underlying bug/change.", + "confidence": 0.78, + "canonical_issue_reason": "No single issue cleanly represents the cluster: it spans unrelated topics (trainer API, tokenizers, model loading, device_map, multimodal video regressions, dtype issues, etc.).", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43824 is the best overall issue anchor in this packet because it has the strongest combined engagement signal in the supplied metadata (discussion plus inbound references), and its import error spans the broader Qwen2.5-VL family rather than a narrower single-symptom report.", + "best_issue_reason": "None of these issues is a reliable global canonical for the cluster; the set is too mixed for one issue to represent all others.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43452", - "right": "issue:43582", - "accept": false, - "reason": "Tokenizer/model-loading failure vs Apple Silicon allocator warmup TypeError; different code paths and symptoms." - }, - { - "left": "issue:42915", - "right": "issue:43742", + "left": "issue:41720", + "right": "issue:43540", "accept": false, - "reason": "Qwen3Moe FP8 config failure vs MobileLLM key error; unrelated models and bugs." + "reason": "Different Qwen models and failures: CUDA assert during auto device mapping vs video-processing ValueError. Same broad area, not the same bug." }, { "left": "issue:43317", "right": "issue:45538", "accept": false, - "reason": "Dequantized offload/device-map issue vs CLIPTokenizer max_length setting; not the same defect." + "reason": "Device-map/offload loading failure and CLIPTokenizer max_length bug are unrelated." }, { "left": "issue:33357", "right": "issue:34567", "accept": false, - "reason": "MacOS CLIP bus error vs TrainerState token counter not updating; different subsystems." + "reason": "Completely different subsystems: MacOS CLIP bus error vs TrainerState token counting." }, { - "left": "issue:43927", - "right": "issue:44912", + "left": "issue:42915", + "right": "issue:44821", "accept": false, - "reason": "Config token-ID loss during save/load vs MXFP4 quantization fallback; distinct failures." + "reason": "Qwen3Moe FP8 loading failure vs AutoImageProcessor URL loading; no shared code path." }, { "left": "issue:44987", "right": "issue:45538", "accept": false, - "reason": "Model-loading regression vs tokenizer max_length behavior; unrelated." + "reason": "Model-loading regression for physical-intelligence/fast vs CLIP tokenizer length issue; unrelated." }, { "left": "issue:42915", - "right": "issue:44821", + "right": "issue:45538", + "accept": false, + "reason": "FP8 Qwen3Moe load failure is not the same as CLIPTokenizer model_max_length behavior." + }, + { + "left": "issue:43927", + "right": "issue:44912", "accept": false, - "reason": "FP8 Qwen3Moe issue vs AutoImageProcessor URL loading; different areas of the library." + "reason": "DiaConfig save/load token ID loss and git-oss quantization fallback are different bugs." }, { "left": "issue:44479", "right": "issue:45381", "accept": true, - "reason": "Both describe the same Qwen video-input regression around incorrect video position IDs in the shared video code path." - }, - { - "left": "issue:42915", - "right": "issue:45538", - "accept": false, - "reason": "Qwen3Moe FP8 config bug vs CLIPTokenizer length default; unrelated." + "reason": "Both describe the same Qwen2.5-VL / related Qwen video-input regression in v5.3.0, centered on incorrect video position handling." }, { "left": "issue:43257", "right": "issue:43756", "accept": false, - "reason": "Qwen MOE weight conversion bug vs Smollm3 RoPE-layer mismatch; different model families and fixes." + "reason": "Qwen3 MoE weight conversion and Smollm3 RoPE-layer dropping are unrelated model-specific issues." }, { "left": "issue:43540", "right": "issue:44560", "accept": false, - "reason": "Both are Qwen video issues, but the model families and concrete failures differ too much to treat as one bug." + "reason": "Both are video-input failures, but in different Qwen model lines with different errors; not clearly one bug." }, { "left": "issue:43540", "right": "issue:44479", "accept": false, - "reason": "Shared video theme, but different Qwen model families and failure modes; too broad to merge." + "reason": "Different model families; 44479 is a broader Qwen2.5-VL/Qwen3-VL regression, while 43540 is Qwen3OmniMoe-specific." }, { "left": "issue:43257", "right": "issue:43696", "accept": false, - "reason": "MOE conversion bug vs CUDA OOM during GPT-oss loading; not the same problem." + "reason": "Qwen3 MoE weight conversion vs GPT-oss OOM are unrelated." }, { "left": "issue:43577", "right": "issue:45072", "accept": false, - "reason": "Different dtype mismatch reports in different models/code paths." + "reason": "Both mention dtype mismatches, but in different models and code paths; not the same concrete bug." }, { "left": "issue:45137", "right": "issue:45405", "accept": false, - "reason": "ZeRO3 deque error vs PEFT version pinning; unrelated." + "reason": "DeepSpeed ZeRO3 deque error and PEFT version bump are unrelated." }, { "left": "issue:38175", "right": "issue:44898", "accept": false, - "reason": "SigLIP2 zero-probability issue vs Perceiver interpolation bug; different models and symptoms." + "reason": "Different vision-model correctness problems (SigLIP2 zero probs vs Perceiver non-default resolution failure)." }, { "left": "issue:44737", "right": "issue:45588", "accept": false, - "reason": "XLNet CPU tensor creation bug vs flash-attention sinkless-model crash; unrelated." + "reason": "XLNet CPU arange bug and flash_attention sink-less crash are unrelated." }, { "left": "issue:44062", "right": "issue:44991", "accept": false, - "reason": "AddedToken kwarg duplication vs tokenizer load failure for a specific model; different causes." + "reason": "Tokenizer AddedToken argument error and est-roberta tokenizer loading failure are not the same bug." }, { "left": "issue:43742", "right": "issue:44291", "accept": false, - "reason": "MobileLLM key error vs init_empty_weights signature regression; different loading failures." + "reason": "Different model-loading failures with different root causes: key error vs unexpected init_empty_weights argument." }, { "left": "issue:43540", "right": "issue:45823", "accept": false, - "reason": "Qwen video-processing failure vs Gemma4 PLE device mismatch; unrelated." + "reason": "Both involve device-related inference issues, but on different models and code paths." }, { "left": "issue:43906", "right": "issue:45381", "accept": false, - "reason": "Isolated reproduction of another issue vs a specific Qwen2.5-VL video-position bug; not the same defect." + "reason": "The isolated reproduction points to another issue, but it is not clearly the same as the qwen2.5-vl video-position-id regression." }, { "left": "issue:43646", "right": "issue:43749", "accept": false, - "reason": "Custom model initialization breakage vs FSDP CPU RAM-efficient loading; different code paths." + "reason": "Custom model initialization breakage and FSDP CPU RAM efficient loading are unrelated." }, { "left": "issue:44062", "right": "issue:45081", "accept": false, - "reason": "Different tokenizer bugs with different error mechanisms." + "reason": "AddedToken duplication and Mistral regex backend_tokenizer crash are different tokenizer problems." }, { "left": "issue:43749", "right": "issue:44756", "accept": false, - "reason": "Both are loading/memory related, but one is FSDP CPU loading and the other is mmap on Strix Halo; different fixes." + "reason": "Both concern loading/memory, but one is an FSDP bug and the other is mmap/OOM on Strix Halo; not the same issue." }, { "left": "issue:34567", "right": "issue:36331", "accept": false, - "reason": "TrainerState token counting bug vs custom compute_loss signature change; unrelated training APIs." - }, - { - "left": "issue:43643", - "right": "issue:45375", - "accept": false, - "reason": "trust_remote_code field handling vs strict-dropped Qwen vision config field; different serialization paths." + "reason": "Both are Trainer-related, but one is token accounting and the other is compute_loss signature incompatibility." }, { "left": "issue:36010", "right": "issue:42175", "accept": false, - "reason": "GenerationMixin import regression vs TensorFlow backend packaging; unrelated." + "reason": "GenerationMixin import failure and missing TensorFlow backend are unrelated packaging/import issues." }, { "left": "issue:43493", "right": "issue:43994", "accept": true, - "reason": "Both are SigLIP2 reports pointing to the same underlying implementation mismatch and output-quality regression for the same model family." + "reason": "Both are about SigLIP2 producing wrong outputs; one is the concrete HF-vs-JAX discrepancy and the other is the user-visible nonsensical-results symptom." }, { - "left": "issue:42175", - "right": "issue:43824", + "left": "issue:43643", + "right": "issue:45375", "accept": false, - "reason": "Backend packaging issue vs Qwen import error; unrelated." + "reason": "Both mention missing fields, but one is trust_remote_code AutoConfig serialization and the other is a specific Qwen3_5MoeVisionConfig field-loss bug." }, { - "left": "issue:43452", - "right": "issue:44488", + "left": "issue:42175", + "right": "issue:43824", "accept": false, - "reason": "gguf_file loading breakage vs a specific model-load failure; different concrete causes." + "reason": "Package backend composition and a missing Qwen2_5_VL import are unrelated." }, { "left": "issue:44291", "right": "issue:44821", "accept": false, - "reason": "init_empty_weights argument bug vs AutoImageProcessor URL loading; unrelated." + "reason": "init_empty_weights/_is_hf_initialized and AutoImageProcessor URL loading are different failures." + }, + { + "left": "issue:43452", + "right": "issue:44488", + "accept": false, + "reason": "gguf_file/from_pretrained breakage and sleng-bert loading failure are not the same underlying bug." }, { "left": "issue:44792", "right": "issue:44964", "accept": false, - "reason": "Janus image-generation test failure vs Phi-4 multimodal load failure; different models and symptoms." + "reason": "Different multimodal model failures; one is a failed janus test, the other is Phi-4-multimodal-instruct loading." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The soft-edge verdicts are mostly conservative, but the top-level ranking overstates the packet. Issue 43749 is not clearly the most-discussed or most-referenced item in the set\u201443824 appears stronger by the provided activity counts\u2014and the phrase \"clear reproducible regression\" is not supported by the packet metadata. The canonical duplicate anchor choice for 44479 vs 45381 is plausible, but the representative-issue rationale needs to be more cautious and fully grounded in the supplied data." + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative: it characterizes the cluster as heterogeneous, and the only accepted soft edges are the two pairs that have clear textual alignment (Qwen video regression and SigLIP2 correctness). The remaining rejections are appropriately cautious and do not overstate duplicate evidence." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -5813,8 +5846,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -5875,11 +5908,11 @@ "issue:45405|issue:45600", "issue:43531|issue:44821", "issue:43697|issue:45216", - "issue:44514|issue:45381", "issue:28282|issue:29127", + "issue:44514|issue:45381", "issue:44484|issue:44704", - "issue:43643|issue:44561", "issue:43582|issue:44977", + "issue:43643|issue:44561", "issue:44625|issue:44877", "issue:43720|issue:44898", "issue:42175|issue:44938", @@ -5894,30 +5927,30 @@ "issue:44936|issue:45823", "issue:24643|issue:29127" ], - "prepared_review_unit_hash": "d0a4ac06784cd0031f239c440bcbf3d6296a765359c14a0c120631f1feb7df61", + "prepared_review_unit_hash": "27eb51972404d92118598cd891b8b3ddd7c4b3f0c8b693a232f5ccc615f3078d", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 26, - "serialized_chars": 35370, - "estimated_input_tokens": 8843, - "estimated_eval_tokens": 17942 + "serialized_chars": 35357, + "estimated_input_tokens": 8840, + "estimated_eval_tokens": 17936 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is not a duplicate group: the items span unrelated loading, tokenizer, processor, model-specific, and backend regression bugs. None of the soft-edge pairs look like the same underlying issue.", - "confidence": 0.98, + "summary": "These items are mostly unrelated Transformers bug reports that only share broad themes like loading, tokenization, config handling, or version regressions. None of the soft pairs looks like the same underlying bug/change, so this cluster should not be deduplicated as one issue set.", + "confidence": 0.92, "canonical_issue_reason": null, "canonical_pr_reason": null, "best_issue_reason": null, @@ -5927,157 +5960,157 @@ "left": "issue:43526", "right": "issue:45479", "accept": false, - "reason": "Different bugs: BeitImageProcessorFast label reduction vs sequence-classification zero-loss behavior across many models." + "reason": "Both concern labels/classification, but one is an image-processor bug and the other is a sequence-classification loss issue; different code paths." }, { "left": "issue:28282", "right": "issue:34689", "accept": false, - "reason": "Both are loading-related, but one is missing PyTorch detection and the other is a specific Llama 3.2 Vision regression." + "reason": "Missing PyTorch installation is an environment error; Llama 3.2 Vision loading is a regression in model loading behavior." }, { "left": "issue:45042", "right": "issue:45081", "accept": false, - "reason": "Both touch image/tokenizer paths, but one is PIL/torchvision dependency handling and the other is a Mistral regex backend attribute crash." + "reason": "One is about PIL backend requiring torchvision; the other is a Mistral regex/tokenizer attribute crash." }, { "left": "issue:43756", "right": "issue:45325", "accept": false, - "reason": "Different model families and failures: Smollm3 RoPE-layer handling vs Qwen2.5-VL still-image temporal position IDs." + "reason": "Both mention vision/rope ideas, but they affect different models and different positional-index logic." }, { "left": "issue:43742", "right": "issue:44589", "accept": false, - "reason": "Different low-level failures: MobileLLM key error on load vs missing Float8 storage object error." + "reason": "Different failures: a key error while loading MobileLLM versus a Float8 storage lookup TypeError." }, { "left": "issue:45405", "right": "issue:45600", "accept": false, - "reason": "PEFT version bump issue is unrelated to auto_mappings referencing removed Sam3LiteText configs." + "reason": "One is a dependency version bump, the other is a stale auto_mappings reference to removed configs." }, { "left": "issue:43531", "right": "issue:44821", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior is unrelated to AutoImageProcessor URL loading." + "reason": "Qwen3-MoE sliding-window behavior and AutoImageProcessor URL loading are unrelated issues." }, { "left": "issue:43697", "right": "issue:45216", "accept": false, - "reason": "Different regressions: RTDetrV2 output mismatch in v5 vs Qwen3.5 checkpoint serialization correctness." - }, - { - "left": "issue:44514", - "right": "issue:45381", - "accept": false, - "reason": "Different Qwen2.5-VL problems: chat-template batching crash vs video vision_position_ids calculation." + "reason": "RTDetrV2 output differences and Qwen3.5 save_pretrained regression are separate bugs." }, { "left": "issue:28282", "right": "issue:29127", "accept": false, - "reason": "One is missing PyTorch in environment; the other is a LayoutLMv3 boxes validation/error-message issue." + "reason": "ImportError for missing PyTorch is not the same as LayoutLMv3 box-validation messaging." }, { - "left": "issue:44484", - "right": "issue:44704", + "left": "issue:44514", + "right": "issue:45381", "accept": false, - "reason": "save_pretrained shard-size behavior is unrelated to AutoProcessor kwargs not reaching cached_file." + "reason": "Both involve chat/template or VL inputs, but they fail in different shapes and model-specific position handling." }, { - "left": "issue:43643", - "right": "issue:44561", + "left": "issue:44484", + "right": "issue:44704", "accept": false, - "reason": "Both involve v5/trust_remote_code compatibility, but one is missing config fields and the other is removal of is_torch_fx_available." + "reason": "Save-pretrained shard-size behavior and AutoProcessor kwarg forwarding are unrelated." }, { "left": "issue:43582", "right": "issue:44977", "accept": false, - "reason": "Apple Silicon warmup TypeError is unrelated to Qwen3.5 flash-attention generation issues." + "reason": "Apple Silicon warmup TypeError and Qwen3.5 flash-attention generation are different runtime problems." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "One is missing fields in AutoConfig with trust_remote_code; the other is removal of a helper breaking remote-code models." }, { "left": "issue:44625", "right": "issue:44877", "accept": false, - "reason": "Qwen3.5 num_labels propagation is a model config bug, not the strict granite_speech config-loading failure." + "reason": "Num-label propagation in Qwen3.5 and strict config rejection for granite_speech are not the same bug." }, { "left": "issue:43720", "right": "issue:44898", "accept": false, - "reason": "BitNet packed-weight unpacking during accelerate loading is unrelated to Perceiver interpolation/resolution failure." + "reason": "BitNet packed-weight loading and Perceiver non-default-resolution behavior are unrelated." }, { "left": "issue:42175", "right": "issue:44938", "accept": false, - "reason": "Backend packaging mismatch for torch install is unrelated to Python 3.14 import/load failure." + "reason": "TensorFlow extra dependency packaging and Python 3.14 load failure are different issues." }, { "left": "issue:43452", "right": "issue:43577", "accept": false, - "reason": "gguf_file tokenizer/model loading breakage is unrelated to BLIP2 dtype propagation." + "reason": "GGUF loading failure and Blip2 dtype propagation do not share the same failing code path." }, { "left": "issue:41762", "right": "issue:44936", "accept": false, - "reason": "Gemma3 ZeRO-3 loading index error is unrelated to trainer.evaluate() failing after train()." + "reason": "Gemma3 ZeRO-3 loading IndexError and trainer.evaluate() after train are unrelated runtime failures." }, { "left": "issue:44514", "right": "issue:45290", "accept": false, - "reason": "Both are chat-template issues, but one is batched padding behavior and the other is tool-call messages with empty content." + "reason": "Both are apply_chat_template crashes, but one is batched padding=False and the other is assistant tool-call handling; not the same bug." }, { "left": "issue:43452", "right": "issue:44843", "accept": false, - "reason": "gguf_file loading bug is unrelated to offline mistral_regex patch calling model_info() unconditionally." + "reason": "GGUF tokenizer/model loading and offline mistral regex patching are distinct problems." }, { "left": "issue:44466", "right": "issue:45005", "accept": false, - "reason": "Both concern tied weights, but they are different serialization/save_pretrained regressions affecting different model classes." + "reason": "Both mention tied weights, but one is a serialization inconsistency and the other is a broader translation-model regression." }, { "left": "issue:43295", "right": "issue:45356", "accept": false, - "reason": "Custom processor.tokenizer regression is unrelated to Kimi-K2.5 codec handling and mistral-regex warning." + "reason": "Processor.tokenizer regression and Kimi-K2.5 codec/fix_mistral_regex issues are different tokenizer bugs." }, { "left": "issue:43819", "right": "issue:44811", "accept": false, - "reason": "DAC from_latents mismatch is unrelated to Whisper batch_decode ignoring skip_special_tokens." + "reason": "DAC latent reconstruction mismatch and Whisper batch_decode skipping special tokens are unrelated." }, { "left": "issue:43653", "right": "issue:44568", "accept": false, - "reason": "BigBirdTokenizer special-token registration bug is unrelated to mdeberta-v3 add_special_tokens not adding BOS/EOS in v5." + "reason": "BigBird special-token registration and mdeberta BOS/EOS handling are separate tokenizer special-token bugs." }, { "left": "issue:44936", "right": "issue:45823", "accept": false, - "reason": "trainer.evaluate() failure after train() is unrelated to Gemma4 device mismatch under device_map=\"auto\"." + "reason": "Trainer evaluation failure and Gemma4 PLE device mismatch are different layers of the stack." }, { "left": "issue:24643", "right": "issue:29127", "accept": false, - "reason": "DeepSpeed weight-dimension runtime error is unrelated to LayoutLMv3 insufficient-box-information validation." + "reason": "DeepSpeed weight-shape RuntimeError and LayoutLMv3 box-info error are not the same underlying issue." } ] }, @@ -6096,8 +6129,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -6152,8 +6185,8 @@ "soft_pairs": [ "issue:30064|issue:33666", "issue:38175|issue:43295", - "issue:44361|issue:44442", "issue:43653|issue:44361", + "issue:44361|issue:44442", "issue:43577|issue:43697", "issue:44464|issue:44743", "issue:44843|issue:44991", @@ -6162,140 +6195,140 @@ "issue:43688|issue:45440", "issue:43531|issue:43742", "issue:43653|issue:43927", - "issue:44464|issue:44977", "issue:43550|issue:43927", + "issue:44464|issue:44977", "issue:43577|issue:45588", "issue:43550|issue:44361", "issue:32090|issue:34567", "issue:43504|issue:43550", - "issue:43504|issue:43720", "issue:43122|issue:43906", + "issue:43504|issue:43720", "issue:28282|issue:35141", "issue:32090|issue:33666", "issue:43504|issue:44610", "issue:43452|issue:44451", "issue:44849|issue:45357", - "issue:44568|issue:45812", "issue:43742|issue:44821", - "issue:41720|issue:44464", + "issue:44568|issue:45812", "issue:33666|issue:36010", - "issue:43525|issue:43572", - "issue:43756|issue:44913", + "issue:41720|issue:44464", "issue:43329|issue:45325", + "issue:43756|issue:44913", + "issue:43525|issue:43572", "issue:43688|issue:43749" ], - "prepared_review_unit_hash": "b909d741ef8fa533178917c30fb1cb5f5af960dae7c0036ae027dc369ac9c79c", + "prepared_review_unit_hash": "60a0de3a9b98b15a138e0f22a5356a8693905288e89a2aeabebdddfca6049dbb", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 33, - "serialized_chars": 37186, - "estimated_input_tokens": 9297, - "estimated_eval_tokens": 18850 + "serialized_chars": 37184, + "estimated_input_tokens": 9296, + "estimated_eval_tokens": 18848 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The candidate pairs are mostly superficial matches by subsystem only; none appear to describe the same underlying bug or change. This looks like a mixed set of unrelated issues rather than one duplicate cluster.", - "confidence": 0.9, - "canonical_issue_reason": "There is no single canonical duplicate here because the issues span many unrelated areas. If forced to pick the broadest representative, issue 45812 is the most general tokenizer regression, but it still does not represent the rest of the set.", + "summary": "The cluster is mostly a loose semantic group of unrelated Transformers bugs, especially tokenizer/model-loading regressions plus several separate training/inference and vision issues. None of the soft-edge pairs look like true duplicates, so I would not merge any of them.", + "confidence": 0.94, + "canonical_issue_reason": "Issue 45812 is the broadest and most user-impactful recent tokenizer regression in the set: it affects all Granite models and describes a silent v4\u2192v5 ID change, making it a strong representative issue for the cluster\u2019s tokenizer-regression slice.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45812 is the strongest global representative only in the narrow sense that it is a broad, open tokenizer regression affecting multiple Granite models. Even so, it is not a true canonical for this heterogeneous group.", + "best_issue_reason": "Issue 45812 is the best single representative because it is specific enough to be actionable yet broad enough to capture the most important common theme in the cluster (tokenizer regressions).", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:30064", "right": "issue:33666", "accept": false, - "reason": "Image processor segmentation-map failure and Qwen2-VL multi-GPU training are different components and different failure modes." + "reason": "Different subsystems: image processor void segmentation-map handling vs Qwen2-VL multi-GPU training." }, { "left": "issue:38175", "right": "issue:43295", "accept": false, - "reason": "Zero-probabilities in SigLIP2 and a processor/tokenizer regression in v4.57.5 are unrelated bugs." + "reason": "Unrelated bugs: SIGLIP2 zero probabilities vs a regression around processor.tokenizer and image passing." }, { - "left": "issue:44361", - "right": "issue:44442", + "left": "issue:43653", + "right": "issue:44361", "accept": false, - "reason": "MLukeTokenizer attribute error and AutoTokenizer failing on FastSpeech2ConformerTokenizer are distinct tokenizer-loading issues." + "reason": "Different tokenizer failures: BigBird special-token decode issue vs MLukeTokenizer attribute error." }, { - "left": "issue:43653", - "right": "issue:44361", + "left": "issue:44361", + "right": "issue:44442", "accept": false, - "reason": "BigBird mask-token registration and MLukeTokenizer task failures do not describe the same bug." + "reason": "Both are tokenizer-related, but one is an MLukeTokenizer task error and the other is AutoTokenizer failing on a different tokenizer class." }, { "left": "issue:43577", "right": "issue:43697", "accept": false, - "reason": "BLIP2 dtype handling and RTDetrV2 output drift are separate model behavior problems." + "reason": "Different model behavior bugs: BLIP2 dtype propagation vs RTDetrV2 output drift in v5." }, { "left": "issue:44464", "right": "issue:44743", "accept": false, - "reason": "Chunked generation with compiled forward and Qwen3.5 recurrent-state reset are different generation bugs." + "reason": "Different code paths: compiled chunked generation inconsistency vs Qwen3.5 modular recurrent-state reset." }, { "left": "issue:44843", "right": "issue:44991", "accept": false, - "reason": "HF_HUB_OFFLINE/model_info patching and tokenizer loading for est-roberta are different tokenizer path regressions." + "reason": "Different tokenizer regressions: offline hub regex patch side effect vs EMBEDDIA/est-roberta loading failure." }, { "left": "issue:41720", "right": "issue:43906", "accept": false, - "reason": "Qwen3 auto device mapping CUDA assert and an isolated reproduction of another issue are not the same concrete failure." + "reason": "Qwen3 auto device mapping CUDA assert is unrelated to an isolated reproduction of another issue." }, { "left": "issue:44977", "right": "issue:45356", "accept": false, - "reason": "Flash-attention generation failure and Kimi-K2.5 tokenizer codec/regex regression are unrelated." + "reason": "Both involve Qwen/Kimi generation/tokenization, but the concrete failures are flash-attention generation vs codec/regex tokenizer regression." }, { "left": "issue:43688", "right": "issue:45440", "accept": false, - "reason": "Aux-loss normalization and DeepseekV3MoE implementation divergence are different algorithmic issues." + "reason": "Different MoE/model-implementation bugs: auxiliary-loss normalization vs DeepseekV3 native divergence." }, { "left": "issue:43531", "right": "issue:43742", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior and MobileLLM loading key errors are unrelated." + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to a key error when loading MobileLLM-125M." }, { "left": "issue:43653", "right": "issue:43927", "accept": false, - "reason": "BigBird special-token registration and DiaConfig custom token-ID loss on save/load are different tokenizer/config bugs." + "reason": "BigBird mask-token decode bug and DiaConfig custom token ID loss are separate tokenizer/config issues." }, { - "left": "issue:44464", - "right": "issue:44977", + "left": "issue:43550", + "right": "issue:43927", "accept": false, - "reason": "Compiled chunked generation inconsistency is not the same as Qwen3.5 flash-attention generation failure." + "reason": "Different failures: Bamba torch.compile SDPA vs DiaConfig save/load token-ID corruption." }, { - "left": "issue:43550", - "right": "issue:43927", + "left": "issue:44464", + "right": "issue:44977", "accept": false, - "reason": "Bamba torch.compile/SDPA failure and DiaConfig token-ID persistence are different failure modes." + "reason": "Both are generation bugs, but compiled forward inconsistency is not the same concrete issue as Qwen3.5 flash-attention generation failure." }, { "left": "issue:43577", @@ -6307,109 +6340,109 @@ "left": "issue:43550", "right": "issue:44361", "accept": false, - "reason": "Bamba compile/SDPA and MLukeTokenizer attribute errors are unrelated areas and symptoms." + "reason": "No shared bug path: Bamba compile/SDPA issue vs MLukeTokenizer task AttributeError." }, { "left": "issue:32090", "right": "issue:34567", "accept": false, - "reason": "Trainer broadcast NoneType error and num_input_tokens_seen not updating are separate trainer-state bugs." + "reason": "Trainer NoneType broadcast error is distinct from TrainerState.num_input_tokens_seen not updating." }, { "left": "issue:43504", "right": "issue:43550", "accept": false, - "reason": "Beit legacy-field loading and Bamba torch.compile/SDPA are different code paths and bugs." + "reason": "Beit pretrained-loading legacy field bug is unrelated to Bamba compile/SDPA behavior." }, { - "left": "issue:43504", - "right": "issue:43720", + "left": "issue:43122", + "right": "issue:43906", "accept": false, - "reason": "Beit legacy-field load failure and BitNet packed-weight unpacking are unrelated loading issues." + "reason": "Tokenizer output differences across versions are not the same as an isolated reproduction of another issue." }, { - "left": "issue:43122", - "right": "issue:43906", + "left": "issue:43504", + "right": "issue:43720", "accept": false, - "reason": "Tokenizer behavior change across versions and an isolated reproduction of another issue are not enough to identify the same bug." + "reason": "Different loading issues: legacy field in Beit vs BitNet packed-weight unpacking during accelerate loading." }, { "left": "issue:28282", "right": "issue:35141", "accept": false, - "reason": "Missing PyTorch ImportError and output embedding reinitialization after resize are unrelated." + "reason": "AutoModel missing PyTorch import and token-embedding resizing/post_init reinitialization are unrelated." }, { "left": "issue:32090", "right": "issue:33666", "accept": false, - "reason": "NoneType broadcast in Trainer and Qwen2-VL multi-GPU training are different training problems." + "reason": "Trainer broadcast NoneType error is a different failure mode from Qwen2-VL multi-GPU training." }, { "left": "issue:43504", "right": "issue:44610", "accept": false, - "reason": "Beit pretrained-load legacy field issue and OmDet-Turbo input-size mismatch are unrelated vision bugs." + "reason": "Beit legacy-field loading bug and OmDet-Turbo processor/model image-size mismatch are different problems." }, { "left": "issue:43452", "right": "issue:44451", "accept": false, - "reason": "gguf_file handling for AutoTokenizer/AutoModelForCausalLM and inability to load ScandiBERT are distinct load failures." + "reason": "gguf_file breaks specific AutoTokenizer/AutoModel loading paths, while ScandiBERT loading failure is a separate model/tokenizer compatibility issue." }, { "left": "issue:44849", "right": "issue:45357", "accept": false, - "reason": "Qwen3.5 hidden-state bug and incorrect visual-encoder keys on save_pretrained are different regressions." - }, - { - "left": "issue:44568", - "right": "issue:45812", - "accept": false, - "reason": "mdeberta special-token insertion and Granite token-ID regression are both tokenizer issues but not the same model-specific bug." + "reason": "Qwen3.5 hidden-states bug and incorrect visual-encoder keys on save_pretrained are different regressions." }, { "left": "issue:43742", "right": "issue:44821", "accept": false, - "reason": "MobileLLM key error on load and AutoImageProcessor-from-URL failure are unrelated loading paths." + "reason": "MobileLLM key error on load is unrelated to AutoImageProcessor loading from URL." }, { - "left": "issue:41720", - "right": "issue:44464", + "left": "issue:44568", + "right": "issue:45812", "accept": false, - "reason": "Qwen3 device-mapping CUDA assert and compiled chunked generation inconsistency are different runtime failures." + "reason": "mdeberta-v3 special-token addition bug is not the same regression as Granite token-ID changes; different tokenizer families and behaviors." }, { "left": "issue:33666", "right": "issue:36010", "accept": false, - "reason": "Qwen2-VL multi-GPU training and GenerationMixin import error are unrelated." + "reason": "Qwen2-VL multi-GPU training and GenerationMixin import failure are unrelated." }, { - "left": "issue:43525", - "right": "issue:43572", + "left": "issue:41720", + "right": "issue:44464", + "accept": false, + "reason": "CUDA assert on auto device mapping is a different failure from compiled-forward generation inconsistency." + }, + { + "left": "issue:43329", + "right": "issue:45325", "accept": false, - "reason": "Llama4Config missing pad_token_id and StableLmConfig missing pad_token_idx are similar in theme but different model/config bugs." + "reason": "Undefined helper/merge_size in video multimodal token counting is distinct from Qwen2.5-VL rope-index scaling for still images." }, { "left": "issue:43756", "right": "issue:44913", "accept": false, - "reason": "Smollm3 RoPE-layer dropping and GPTNeoX rotary_pct reload defaulting are distinct config issues." + "reason": "Smollm3 RoPE layer mismatch and GPTNeoX rotary_pct reload bug are separate configuration issues." }, { - "left": "issue:43329", - "right": "issue:45325", + "left": "issue:43525", + "right": "issue:43572", "accept": false, - "reason": "Undefined video-branch helpers in multimodal token counting and Qwen2.5-VL still-image position scaling are different RoPE/multimodal bugs." + "reason": "Both are config attribute regressions, but they concern different fields and model families (pad_token_id vs pad_token_idx)." }, { "left": "issue:43688", "right": "issue:43749", "accept": false, - "reason": "Aux-loss normalization and broken FSDP CPU RAM efficient loading are unrelated." + "reason": "Aux-loss normalization in MoE models and FSDP_CPU_RAM_EFFICIENT_LOADING are unrelated." } ] }, @@ -6428,8 +6461,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -6491,21 +6524,21 @@ "issue:28282|issue:33357", "issue:43577|issue:44554", "issue:33666|issue:36331", - "issue:43720|issue:44361", "issue:39692|issue:43493", + "issue:43720|issue:44361", "issue:43688|issue:45704", "issue:43334|issue:43572", - "issue:43540|issue:44977", "issue:44964|issue:45405", + "issue:43540|issue:44977", "issue:44479|issue:44625", "issue:43493|issue:43844", "issue:43756|issue:45083", - "issue:45200|issue:45229", "issue:43334|issue:45759", + "issue:45200|issue:45229", "issue:44488|issue:44843", "issue:43257|issue:43299", - "issue:43278|issue:43844", "issue:43824|issue:44589", + "issue:43278|issue:43844", "issue:43646|issue:44466", "issue:45229|issue:45823", "issue:44589|issue:45520", @@ -6517,232 +6550,232 @@ "issue:44423|issue:44734", "issue:43927|issue:44361" ], - "prepared_review_unit_hash": "8a826ea6616c851b6136ceca47fd6d31e00541cdbc8940dc62a9b995a6abac2c", + "prepared_review_unit_hash": "a1b233fedade9e9f596428ccc25679fa4f7c7264a9ec36dcc800c2653271f51c", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37972, - "estimated_input_tokens": 9493, - "estimated_eval_tokens": 19242 + "serialized_chars": 37975, + "estimated_input_tokens": 9494, + "estimated_eval_tokens": 19244 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is highly heterogeneous: the soft-linked pairs mostly share only broad Transformer subsystems or model families, not the same concrete bug. No PRs are present.", - "confidence": 0.96, - "canonical_issue_reason": "issue:43824 is the most prominent standalone issue in the set by discussion and inbound references, with a concrete reproducible loading failure; the rest of the cluster is too diverse to serve as a true duplicate canonical.", + "summary": "The set is highly heterogeneous: most items are distinct Transformers bugs/regressions across unrelated models, loaders, and runtime paths. I found no true duplicate pairs among the soft edges.", + "confidence": 0.94, + "canonical_issue_reason": "Issue 43646 is the broadest, most representative regression report in the set (\u201cTransformers 5.0.0 breaks custom model initialization\u201d) and is phrased as a concrete framework-level breakage rather than a model-specific symptom.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43824 is the best overall issue representative if one must be chosen, because it has the strongest visibility and a clearly scoped bug report; however, it is not a duplicate of most other items here.", + "best_issue_reason": "Issue 43646 is the best global exemplar because it describes a clear, high-level compatibility regression with direct impact and a well-scoped failure mode.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43493", "right": "issue:43825", "accept": false, - "reason": "Different bugs: SigLIP2 implementation mismatch vs a pipeline error-message regression about translation support." + "reason": "Different bugs: SigLIP2 implementation mismatch vs an incorrect pipeline error message about translation tasks." }, { "left": "issue:43262", "right": "issue:43295", "accept": false, - "reason": "Different code paths: audio processor sampling-rate default vs processor.tokenizer/image passthrough regression." + "reason": "Unrelated regressions: audio sampling-rate defaulting vs processor.tokenizer / image passthrough breakage." }, { "left": "issue:43606", "right": "issue:43927", "accept": false, - "reason": "Unrelated failures: CPU offload device mismatch vs DiaConfig custom token IDs lost on save/load." + "reason": "Different code paths: CPU offload device mismatch in Bark vs DiaConfig losing custom token IDs on save/load." }, { "left": "issue:43720", "right": "issue:43927", "accept": false, - "reason": "Different issues: BitNet packed-weight handling during accelerate loading vs DiaConfig token ID persistence." + "reason": "Different models and failures: BitNet packed-weight loading vs DiaConfig token-id persistence." }, { "left": "issue:28282", "right": "issue:32090", "accept": false, - "reason": "Import-time PyTorch missing error is unrelated to Trainer NoneType broadcast failure." + "reason": "ImportError for missing PyTorch is unrelated to Trainer's NoneType broadcast failure." }, { "left": "issue:28282", "right": "issue:33357", "accept": false, - "reason": "PyTorch availability error and MacOS CLIP bus error are distinct bugs." + "reason": "Different underlying problems: missing PyTorch installation vs MacOS bus error loading a CLIP model." }, { "left": "issue:43577", "right": "issue:44554", "accept": false, - "reason": "BLIP2 dtype persistence and an MPS attention correctness bug are different concrete problems." + "reason": "BLIP2 dtype mismatch is not the same as the MPS attention correctness issue." }, { "left": "issue:33666", "right": "issue:36331", "accept": false, - "reason": "Multi-GPU Qwen2-VL training request is unrelated to the custom compute_loss signature error." + "reason": "Multi-GPU training for Qwen2-VL is unrelated to CustomTrainer.compute_loss signature breakage." }, { - "left": "issue:43720", - "right": "issue:44361", + "left": "issue:39692", + "right": "issue:43493", "accept": false, - "reason": "BitNet packed-weight loading failure and MLukeTokenizer task AttributeError are unrelated." + "reason": "Docs/example errors are separate from a SigLIP2 HF-vs-JAX implementation discrepancy." }, { - "left": "issue:39692", - "right": "issue:43493", + "left": "issue:43720", + "right": "issue:44361", "accept": false, - "reason": "Docs/example mistakes are not the same as the SigLIP2 HF-vs-JAX implementation discrepancy." + "reason": "BitNet packed-weight loading and MLukeTokenizer task attribute errors are different bugs." }, { "left": "issue:43688", "right": "issue:45704", "accept": false, - "reason": "Auxiliary-loss normalization and T5 apex RMSNorm memory leak are different model bugs." + "reason": "Auxiliary-loss normalization and T5 apex memory leak are unrelated issues." }, { "left": "issue:43334", "right": "issue:43572", "accept": false, - "reason": "Both involve config fields, but they are different models and different missing fields, not the same bug." + "reason": "Both involve missing config fields, but for different models and different attributes (pad_token_id vs pad_token_idx)." }, { - "left": "issue:43540", - "right": "issue:44977", + "left": "issue:44964", + "right": "issue:45405", "accept": false, - "reason": "Qwen3OmniMoe video-input processing and Qwen3.5 flash-attention generation are separate failures." + "reason": "Model loading failure for Phi-4 multimodal is unrelated to a dependency version bump for PEFT." }, { - "left": "issue:44964", - "right": "issue:45405", + "left": "issue:43540", + "right": "issue:44977", "accept": false, - "reason": "Model load failure and a bumped unreleased dependency requirement are unrelated." + "reason": "Video-input processing failure in Qwen3OmniMoe is unrelated to flash-attention generation behavior in Qwen3.5." }, { "left": "issue:44479", "right": "issue:44625", "accept": false, - "reason": "Video-input regression across Qwen variants is unrelated to num_labels propagation in Qwen3.5." + "reason": "Video regression across Qwen variants is different from num_labels propagation in Qwen3.5 configs." }, { "left": "issue:43493", "right": "issue:43844", "accept": false, - "reason": "SigLIP2 implementation drift and DeepSpeed gradient growth are different bugs." + "reason": "SigLIP2 implementation discrepancy is unrelated to ZeRO-3 gradient growth." }, { "left": "issue:43756", "right": "issue:45083", "accept": false, - "reason": "Smollm3 RoPE-layer drop and qwen3_omni_moe feature-length helper behavior are unrelated." + "reason": "Smollm3 RoPE-layer dropping and a helper length function behaving unexpectedly are different failures." }, { - "left": "issue:45200", - "right": "issue:45229", + "left": "issue:43334", + "right": "issue:45759", "accept": false, - "reason": "Default mm_token_type_ids behavior and Gemma4 multi-GPU OOM are not the same underlying issue." + "reason": "Both are multimodal config issues, but they affect different fields and initialization paths." }, { - "left": "issue:43334", - "right": "issue:45759", + "left": "issue:45200", + "right": "issue:45229", "accept": false, - "reason": "Both involve Qwen multimodal configs, but one is a missing pad_token_id error and the other is from_config not unwrapping text_config." + "reason": "Text-only token-type defaults are unrelated to Gemma 4 multi-GPU CUDA OOM." }, { "left": "issue:44488", "right": "issue:44843", "accept": false, - "reason": "Model load failure for a specific checkpoint is unrelated to HF_HUB_OFFLINE behavior in tokenizer patching." + "reason": "Model-loading failure for a specific checkpoint is not the same as an offline-mode tokenizer patch bug." }, { "left": "issue:43257", "right": "issue:43299", "accept": false, - "reason": "Different Qwen MOE loading regressions with different failure modes and code paths." + "reason": "Qwen3 MoE weight conversion under accelerate/deepspeed is distinct from a general Qwen3VL MoE loading regression." }, { - "left": "issue:43278", - "right": "issue:43844", + "left": "issue:43824", + "right": "issue:44589", "accept": false, - "reason": "Embedding dtype drift in train/eval is unrelated to the ZeRO-3 gradient increase report." + "reason": "Missing Qwen2.5-VL import and missing Float8 storage support are unrelated import/runtime issues." }, { - "left": "issue:43824", - "right": "issue:44589", + "left": "issue:43278", + "right": "issue:43844", "accept": false, - "reason": "Missing Qwen2_5_VL import and Float8 storage lookup failure are unrelated load-time errors." + "reason": "A dtype change in embeddings is not the same bug as exploding gradients under ZeRO-3." }, { "left": "issue:43646", "right": "issue:44466", "accept": false, - "reason": "Custom model initialization breakage and lm_head.weight serialization inconsistency are separate issues." + "reason": "Custom model initialization breakage is different from device-dependent serialization of lm_head.weight." }, { "left": "issue:45229", "right": "issue:45823", "accept": false, - "reason": "Gemma4 inference OOM and PLE device mismatch are different forward-pass failures." + "reason": "OOM during multi-GPU inference is not the same as a device mismatch in PLE with device_map=\"auto\"." }, { "left": "issue:44589", "right": "issue:45520", "accept": false, - "reason": "Float8 storage lookup error is unrelated to the flash_attn KeyError on Python 3.13." + "reason": "Float8 storage lookup failure and Python 3.13 flash_attn KeyError are separate compatibility issues." }, { "left": "issue:43688", "right": "issue:45701", "accept": false, - "reason": "Auxiliary-loss normalization and tokenization changes across versions are not the same bug." + "reason": "Aux-loss normalization and tokenization changes are unrelated." }, { "left": "issue:39692", "right": "issue:43295", "accept": false, - "reason": "Documentation example errors are unrelated to the processor.tokenizer regression in v4.57.5." + "reason": "Documentation example mistakes do not match the processor/tokenizer regression in custom model code." }, { "left": "issue:43572", "right": "issue:44291", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig is a different problem from init_empty_weights passing _is_hf_initialized." + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to init_empty_weights passing an unexpected argument." }, { "left": "issue:33357", "right": "issue:33666", "accept": false, - "reason": "MacOS bus error with a CLIP model and Qwen2-VL multi-GPU training are unrelated." + "reason": "MacOS bus error on CLIP and Qwen2-VL multi-GPU training are different problems." }, { "left": "issue:43504", "right": "issue:43606", "accept": false, - "reason": "Legacy-field load failure in BEiT is unrelated to the Bark device-mismatch CPU offload bug." + "reason": "Legacy-field pretrained loading failure is unrelated to a CPU-offload device mismatch." }, { "left": "issue:44423", "right": "issue:44734", "accept": false, - "reason": "Continuous-batching multimodal crash and KV-cache continuation indexing bug are distinct serve-path issues." + "reason": "Continuous-batching multimodal crash and KV-cache continuation indexing bug are distinct serve-path failures." }, { "left": "issue:43927", "right": "issue:44361", "accept": false, - "reason": "DiaConfig save/load token-ID loss and MLukeTokenizer task AttributeError do not describe the same failure." + "reason": "DiaConfig token-id persistence and MLukeTokenizer task attribute errors do not share the same root cause." } ] }, @@ -6761,8 +6794,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -6823,8 +6856,8 @@ "issue:44792|issue:45405", "issue:43994|issue:44493", "issue:44521|issue:45290", - "issue:44464|issue:45584", "issue:29942|issue:30333", + "issue:44464|issue:45584", "issue:43844|issue:44964", "issue:28218|issue:34567", "issue:45084|issue:45520", @@ -6841,209 +6874,206 @@ "issue:28218|issue:33357", "issue:44554|issue:45356", "issue:45200|issue:45325", - "issue:44466|issue:45127", - "issue:43295|issue:44610" + "issue:43295|issue:44610", + "issue:44466|issue:45127" ], - "prepared_review_unit_hash": "c391cf41884d7611a219891571e059c421c53f2ecc555dd738a1e1fd4216f75d", + "prepared_review_unit_hash": "847b47c86ca43776114df3781461d17c53d64f6629760b84f5df2960691a68ea", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 35870, - "estimated_input_tokens": 8968, - "estimated_eval_tokens": 18192 + "serialized_chars": 35873, + "estimated_input_tokens": 8969, + "estimated_eval_tokens": 18194 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated issues spanning tokenization, generation, multimodal preprocessing, training, and CI. Only one soft edge looks like a plausible duplicate pair: the Qwen composite-config propagation/from_config handling issues.", - "confidence": 0.79, - "canonical_issue_reason": "issue:45759 is the best canonical issue: it states the broader root cause in composite Qwen multimodal config handling, while issue:44625 is a narrower symptom of the same config-propagation path.", + "summary": "This is a mixed cluster of unrelated issues across tokenization, generation, multimodal preprocessing, training state, quantization, and CI. None of the soft pairs look like the same underlying defect, so I would not collapse any of them as duplicates.", + "confidence": 0.92, + "canonical_issue_reason": "No single canonical issue fits the cluster: the items are heterogeneous and do not describe one shared bug.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45759 is the strongest representative because it describes the underlying API/config plumbing problem most clearly and broadly.", + "best_issue_reason": "If a representative issue is needed, issue:44521 is the most active and clearly scoped, but it only represents one narrow multimodal chat-template bug and does not subsume the rest.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43504", "right": "issue:43653", "accept": false, - "reason": "Different subsystems and failures: BEiT legacy config loading vs BigBird tokenizer special-token registration." + "reason": "Legacy field loading for Beit segmentation is unrelated to BigBirdTokenizer special-token registration and empty decode output." }, { "left": "issue:43575", "right": "issue:45405", "accept": false, - "reason": "One is tensor-parallel OOM on a model load; the other is a dependency version bump problem. No shared bug." + "reason": "Tensor-parallel OOM on Qwen2-57B-A14B is unrelated to a PEFT minimum-version bump not yet on PyPI." }, { "left": "issue:44625", "right": "issue:45759", - "accept": true, - "reason": "Both point to the same composite Qwen config plumbing in model/config construction; one is a specific propagation symptom, the other the broader from_config unwrap bug." + "accept": false, + "reason": "Qwen3.5 num_labels propagation and Qwen 3.5/3.6 text_config unwrapping are different config bugs with different failure modes." }, { "left": "issue:43606", "right": "issue:44112", "accept": false, - "reason": "Both mention device mismatches, but they affect different models and test paths; not the same underlying fix." + "reason": "CPU offload device mismatch in bark-small is unrelated to a stale GraniteSpeech CI test." }, { "left": "issue:44792", "right": "issue:45405", "accept": false, - "reason": "Janus image-generation test failure is unrelated to the PEFT minimum-version packaging issue." + "reason": "Janus image-generation test failures are not the same as a PEFT version availability issue." }, { "left": "issue:43994", "right": "issue:44493", "accept": false, - "reason": "Different symptoms and code paths: SigLIP2 wrong outputs vs position-id key handling." + "reason": "SigLIP2 nonsensical outputs are unrelated to unexpected position-id key handling." }, { "left": "issue:44521", "right": "issue:45290", "accept": false, - "reason": "Both involve apply_chat_template, but one is multimodal assistant-mask generation and the other is a crash on tool-call/no-content messages; not clearly the same bug." + "reason": "All-zero assistant masks for multimodal inputs are a different bug from crashing on assistant tool-call messages with no content." }, { - "left": "issue:44464", - "right": "issue:45584", + "left": "issue:29942", + "right": "issue:30333", "accept": false, - "reason": "Compiled-forward chunked generation inconsistency is unrelated to Whisper empty-transcription handling after token alignment." + "reason": "Flash Attention 2 CI failures and MLflow not reporting failed jobs are unrelated subsystems and symptoms." }, { - "left": "issue:29942", - "right": "issue:30333", + "left": "issue:44464", + "right": "issue:45584", "accept": false, - "reason": "Flash Attention CI failures and MLflow reporting on failed jobs are unrelated." + "reason": "Chunked generation inconsistency with compiled forward is unrelated to empty-transcription generation after align_special_tokens." }, { "left": "issue:43844", "right": "issue:44964", "accept": false, - "reason": "ZeRO-3 gradient growth and Phi-4 multimodal load failure are different problems." + "reason": "Gradient inflation under ZeRO-3 is not the same as failing to load a multimodal Phi-4 model." }, { "left": "issue:28218", "right": "issue:34567", "accept": false, - "reason": "Tokenizer added-space behavior and TrainerState token counter updates are unrelated." + "reason": "Tokenizer extra-space insertion and TrainerState num_input_tokens_seen not updating are unrelated bugs." }, { "left": "issue:45084", "right": "issue:45520", "accept": false, - "reason": "Template compilation error and Python 3.13 flash_attn import KeyError are different issues." + "reason": "Compilation of non-template nodes is unrelated to a flash_attn import KeyError on Python 3.13." }, { "left": "issue:44554", "right": "issue:45446", "accept": false, - "reason": "MPS attention correctness and a PyTorch version check bug in flex_attention are not the same code path." + "reason": "An MPS attention correctness issue is unrelated to an incorrect PyTorch version check for AuxRequest import." }, { "left": "issue:43688", "right": "issue:43873", "accept": false, - "reason": "Aux-loss normalization and quantization/offload behavior are separate model/runtime issues." + "reason": "Incorrect auxiliary-loss normalization and offloading/quantization behavior are different code paths and defects." }, { "left": "issue:44568", "right": "issue:45701", "accept": false, - "reason": "Tokenizer special-token regression and generic tokenization changes across versions are too broad to treat as the same bug." + "reason": "A specific add_special_tokens regression for mdeberta-v3-base is not the same as a broad report that tokenization changes across versions." }, { "left": "issue:15354", "right": "issue:38175", "accept": false, - "reason": "TorchScript export support and zero-probability outputs from SigLIP2 are unrelated." + "reason": "TorchScript generator-expression export failure is unrelated to SigLIP2 producing zero probabilities." }, { "left": "issue:43232", "right": "issue:44849", "accept": false, - "reason": "Generation kwargs update after sync_gpus and Qwen3.5 hidden-state output behavior are different bugs." + "reason": "Generation kwargs updates after sync_gpus are unrelated to Qwen3.5 output_hidden_states behavior." }, { "left": "issue:43540", "right": "issue:45200", "accept": false, - "reason": "Video-input processing in Qwen3OmniMoe and Gemma4 text-only token-type defaults are unrelated." + "reason": "Qwen3OmniMoe video-input processing errors are unrelated to Gemma 4 mm_token_type_ids defaults." }, { "left": "issue:45200", "right": "issue:45823", "accept": false, - "reason": "Same model family, but one is missing default mm_token_type_ids and the other is a device-map forward mismatch; not the same fix." + "reason": "Defaulting mm_token_type_ids to zeros is a different issue from Gemma4 PLE device mismatch during forward." }, { "left": "issue:45072", "right": "issue:45198", "accept": false, - "reason": "Dtype mismatch inference failures and Wav2Vec2 save/tokenization failures are unrelated." + "reason": "bfloat16 inference dtype mismatches are unrelated to Wav2Vec2 save_pretrained/tokenization failures." }, { "left": "issue:44315", "right": "issue:44849", "accept": false, - "reason": "Liger Kernel application with model_init is unrelated to Qwen3.5 hidden-state output behavior." + "reason": "Liger Kernel not applied with model_init is unrelated to Qwen3.5 output_hidden_states bugs." }, { "left": "issue:43278", "right": "issue:43701", "accept": false, - "reason": "Training/eval dtype drift and resume_from_checkpoint key mismatch are separate training-state problems." + "reason": "Embedding dtype changing between train/eval is unrelated to resume_from_checkpoint key mismatch." }, { "left": "issue:28218", "right": "issue:33357", "accept": false, - "reason": "Token spacing after added tokens and a MacOS bus error in CLIP loading are unrelated." + "reason": "Added-token spacing is not the same defect as a MacOS bus error in a CLIP community model." }, { "left": "issue:44554", "right": "issue:45356", "accept": false, - "reason": "MPS attention correctness and Kimi tokenizer codec/regression handling are different concerns." + "reason": "MPS attention correctness and Kimi tokenizer codec handling are different issues in different components." }, { - "left": "issue:45200", - "right": "issue:45325", + "left": "issue:43295", + "right": "issue:44610", "accept": false, - "reason": "Gemma4 token-type defaults and Qwen2.5-VL rope position scaling are different multimodal bugs." + "reason": "Processor.tokenizer access regression is unrelated to OmDet-Turbo producing 640px inputs instead of 224px." }, { "left": "issue:44466", "right": "issue:45127", "accept": false, - "reason": "Serialization of tied weights and LoRA merge collapse with extended vocabulary are not the same bug." + "reason": "Device-dependent lm_head serialization is unrelated to LoRA merge collapse with extended vocabulary." }, { - "left": "issue:43295", - "right": "issue:44610", + "left": "issue:45200", + "right": "issue:45325", "accept": false, - "reason": "Processor/tokenizer regression and OmDet-Turbo image-size mismatch are unrelated." + "reason": "mm_token_type_ids defaults are unrelated to the Qwen2.5-VL rope-index scaling bug." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft edge, issue:44625 vs issue:45759, is a plausible duplicate based on both titles pointing to Qwen composite config propagation/from_config handling. The rest of the rejects are cautious and do not overstate shared causality." - }, - "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 }, @@ -7055,8 +7085,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -7071,6 +7101,7 @@ "issue:43450", "issue:43493", "issue:43525", + "issue:43550", "issue:43572", "issue:43575", "issue:43653", @@ -7097,7 +7128,6 @@ "issue:44792", "issue:44857", "issue:44898", - "issue:44936", "issue:44987", "issue:45083", "issue:45127", @@ -7111,20 +7141,20 @@ "soft_pairs": [ "issue:44898|issue:45412", "issue:43493|issue:43701", - "issue:37428|issue:43957", "issue:43572|issue:44987", - "issue:43334|issue:43525", + "issue:37428|issue:43957", "issue:43572|issue:43957", + "issue:43334|issue:43525", "issue:43873|issue:44464", "issue:43295|issue:43450", - "issue:43844|issue:44936", "issue:43824|issue:43975", + "issue:43550|issue:43844", "issue:43295|issue:43873", - "issue:38175|issue:45412", "issue:37428|issue:43824", + "issue:38175|issue:45412", "issue:42915|issue:43824", - "issue:44514|issue:44521", "issue:38175|issue:43493", + "issue:44514|issue:44521", "issue:44442|issue:45198", "issue:43278|issue:43856", "issue:43827|issue:43901", @@ -7132,8 +7162,8 @@ "issue:43450|issue:43493", "issue:43450|issue:43575", "issue:44792|issue:45563", - "issue:45127|issue:45245", "issue:44360|issue:44561", + "issue:45127|issue:45245", "issue:43819|issue:45083", "issue:44206|issue:45584", "issue:43819|issue:44112", @@ -7143,239 +7173,242 @@ "issue:44857|issue:45593", "issue:43278|issue:43825" ], - "prepared_review_unit_hash": "79beb0f0305d5fdef32a6795d8f8bfc594314295aad690874d07040af09fdb25", + "prepared_review_unit_hash": "a48961fed2af815973f8dfb216c2e56e301d431718f9320a5a25e077a2ecec27", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 33, - "serialized_chars": 36920, - "estimated_input_tokens": 9230, - "estimated_eval_tokens": 18716 + "serialized_chars": 36923, + "estimated_input_tokens": 9231, + "estimated_eval_tokens": 18718 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab-bag of unrelated issues across model loading, processors, training, docs, and CI. The soft pairs share broad keywords in some cases, but they do not look like the same underlying bug or change.", - "confidence": 0.96, - "canonical_issue_reason": "Issue 43824 is the strongest standalone anchor: it has a clear, reproducible import traceback, multiple references, and a focused regression signature.", + "summary": "This cluster is mostly heterogeneous; nearly all soft pairs are different bugs or docs/test regressions. The only plausible duplicate-like pair is the shared `pad_token_id` AttributeError regression across Qwen3-VL and Llama4 config loading.", + "confidence": 0.74, + "canonical_issue_reason": "issue:43334 is the clearest representative of the one plausible duplicate theme: config-loading failures caused by missing `pad_token_id` on newer text configs.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43824 is the best representative issue because it is concrete, well-discussed, and easy to triage or search by its exact failing import path.", + "best_issue_reason": "issue:43334 is the best cluster representative because it captures the common `pad_token_id` regression pattern and has enough discussion to anchor triage.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44898", "right": "issue:45412", "accept": false, - "reason": "Different subsystems and failures: Perceiver resolution handling vs RT-DETR memory release." + "reason": "Different problems: Perceiver image sizing vs RT-DETR memory reclamation." }, { "left": "issue:43493", "right": "issue:43701", "accept": false, - "reason": "Different bugs: SigLIP2 implementation discrepancy vs checkpoint key mismatch." - }, - { - "left": "issue:37428", - "right": "issue:43957", - "accept": false, - "reason": "Both are loading-related, but they describe different root causes and model paths." + "reason": "SigLIP2 output discrepancy and resume checkpoint key mismatch are unrelated." }, { "left": "issue:43572", "right": "issue:44987", "accept": false, - "reason": "Unrelated: missing pad_token_idx in one config vs loading failure for a specific model repo." + "reason": "Different models and failures: StableLm pad token attr vs loading `physical-intelligence/fast`." }, { - "left": "issue:43334", - "right": "issue:43525", + "left": "issue:37428", + "right": "issue:43957", "accept": false, - "reason": "Same symptom class, but different model configs and no evidence of one shared code defect." + "reason": "ImportError for flash support is unrelated to meta-device model loading." }, { "left": "issue:43572", "right": "issue:43957", "accept": false, - "reason": "Different issues: tokenizer config field regression vs meta-device loading problems." + "reason": "No shared code path beyond general loading issues." + }, + { + "left": "issue:43334", + "right": "issue:43525", + "accept": true, + "reason": "Both are the same `pad_token_id` AttributeError regression during model loading; likely one fix." }, { "left": "issue:43873", "right": "issue:44464", "accept": false, - "reason": "Different areas: quantization/offloading vs compiled-forward chunked generation consistency." + "reason": "Quantization/offloading behavior is unrelated to compiled chunked generation outputs." }, { "left": "issue:43295", "right": "issue:43450", "accept": false, - "reason": "Different modalities and failures: processor/tokenizer regression vs batched video-shape bug." + "reason": "Different processor bugs: tokenizer/image regression vs batched video shape handling." }, { - "left": "issue:43844", - "right": "issue:44936", + "left": "issue:43824", + "right": "issue:43975", "accept": false, - "reason": "Training instability with ZeRO-3 is unrelated to trainer.evaluate() breaking after train()." + "reason": "ImportError for Qwen2.5-VL is unrelated to DeepSeek detokenization." }, { - "left": "issue:43824", - "right": "issue:43975", + "left": "issue:43550", + "right": "issue:43844", "accept": false, - "reason": "Different models and failure modes: import error vs detokenization regression." + "reason": "torch.compile SDPA failure and DeepSpeed gradient growth are different failures." }, { "left": "issue:43295", "right": "issue:43873", "accept": false, - "reason": "Processor API regression and quantization offloading are unrelated." + "reason": "Processor regression vs quantization offloading are not the same bug." }, { - "left": "issue:38175", - "right": "issue:45412", + "left": "issue:37428", + "right": "issue:43824", "accept": false, - "reason": "Different components: SigLIP2 zero probabilities vs RT-DETR memory leak." + "reason": "Different import failures in different model families." }, { - "left": "issue:37428", - "right": "issue:43824", + "left": "issue:38175", + "right": "issue:45412", "accept": false, - "reason": "Both are import errors, but they involve different symbols, modules, and root causes." + "reason": "SigLIP2 zero probabilities and RT-DETR memory leak are unrelated." }, { "left": "issue:42915", "right": "issue:43824", "accept": false, - "reason": "Different model families and different failure signatures." + "reason": "Qwen3MoE FP8 failure is unrelated to Qwen2.5-VL import error." }, { - "left": "issue:44514", - "right": "issue:44521", + "left": "issue:38175", + "right": "issue:43493", "accept": false, - "reason": "Same general processor area, but one crashes on batching and the other returns wrong masks; not the same bug." + "reason": "Both are SigLIP2-related, but the reported symptoms are not the same concrete bug." }, { - "left": "issue:38175", - "right": "issue:43493", + "left": "issue:44514", + "right": "issue:44521", "accept": false, - "reason": "Both mention SigLIP2, but one is zero probabilities while the other is a HF-vs-JAX discrepancy." + "reason": "Same API area, but one is a crash and the other is wrong assistant masks; not enough to treat as one bug." }, { "left": "issue:44442", "right": "issue:45198", "accept": false, - "reason": "Tokenizer load failure and save/tokenization failure are separate model-specific issues." + "reason": "Different tokenizer/save_pretrained failures in different speech models." }, { "left": "issue:43278", "right": "issue:43856", "accept": false, - "reason": "Embedding dtype mismatch in eval is unrelated to MoE training memory usage." + "reason": "Embedding dtype drift and MoE memory usage are different training issues." }, { "left": "issue:43827", "right": "issue:43901", "accept": false, - "reason": "Both touch docs/pipeline behavior, but they concern different API topics and fixes." + "reason": "Both are docs-related, but they cover different pipeline/API deprecations." }, { "left": "issue:43653", "right": "issue:44112", "accept": false, - "reason": "Tokenizer special-token registration bug vs CI test staleness are unrelated." + "reason": "BigBirdTokenizer special-token registration and GraniteSpeech CI test staleness are unrelated." }, { "left": "issue:43450", "right": "issue:43493", "accept": false, - "reason": "Batched video processor shape bug is unrelated to SigLIP2 model discrepancy." + "reason": "Video batch-shape bug is unrelated to SigLIP2 implementation discrepancy." }, { "left": "issue:43450", "right": "issue:43575", "accept": false, - "reason": "Video preprocessing shape issue and tensor-parallel OOM are not the same bug." + "reason": "Video processor shape handling and TP OOM on Qwen2-57B are unrelated." }, { "left": "issue:44792", "right": "issue:45563", "accept": false, - "reason": "Different generation issues: failing Janus test vs stale warning for paged generate()." + "reason": "Janus image-generation test failure and stale paged-generate warning are unrelated." }, { - "left": "issue:45127", - "right": "issue:45245", + "left": "issue:44360", + "right": "issue:44561", "accept": false, - "reason": "LoRA merge collapse and category-count runtime error have no shared code path." + "reason": "DSA indexer activation issue and removed helper breaking trust_remote_code are unrelated." }, { - "left": "issue:44360", - "right": "issue:44561", + "left": "issue:45127", + "right": "issue:45245", "accept": false, - "reason": "A missing ReLU in one component is unrelated to removal of a torch-fx helper." + "reason": "LoRA merge collapse and category cardinality overflow are unrelated." }, { "left": "issue:43819", "right": "issue:45083", "accept": false, - "reason": "Different numerical/helper-function bugs in different models." + "reason": "DAC latent mismatch and qwen3_omni_moe length helper behavior are different code paths." }, { "left": "issue:44206", "right": "issue:45584", "accept": false, - "reason": "Unsupported feature-extractor arg and empty-transcription Whisper failure are unrelated." + "reason": "Feature-extractor arg regression and Whisper empty transcription bug are unrelated." }, { "left": "issue:43819", "right": "issue:44112", "accept": false, - "reason": "DAC latent mismatch and a stale CI test are unrelated." + "reason": "DAC latent mismatch is unrelated to GraniteSpeech CI device override staleness." }, { "left": "issue:43262", "right": "issue:43450", "accept": false, - "reason": "Audio chat-template sampling-rate default and batched video shape are unrelated." + "reason": "Audio chat-template sampling-rate default bug is unrelated to video processor batching." }, { "left": "issue:44610", "right": "issue:44857", "accept": false, - "reason": "Processor resize mismatch vs AMP/CUDA loss crash are different bugs." + "reason": "Processor output resolution mismatch and AMP loss crash are different problems." }, { "left": "issue:37428", "right": "issue:43425", "accept": false, - "reason": "ImportError on a missing symbol is unrelated to Torch 2.10 compatibility concerns." + "reason": "Flash-support import error is unrelated to Torch 2.10 compatibility." }, { "left": "issue:44857", "right": "issue:45593", "accept": false, - "reason": "Different training bugs: AMP/CUDA crash vs auxiliary-loss omission." + "reason": "AMP/CUDA loss crash and missing auxiliary losses are different training bugs." }, { "left": "issue:43278", "right": "issue:43825", "accept": false, - "reason": "Embedding dtype regression is unrelated to pipeline error messaging." + "reason": "Embedding dtype regression and pipeline translation-message docs mismatch are unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The writeup is grounded in the packet and stays conservative on the soft-edge pairs. The only clearly supported duplicate theme is the shared `pad_token_id` AttributeError during config/model loading (`issue:43334` / `issue:43525`), and the remaining pair rejections are phrased narrowly enough for the evidence available." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -7387,8 +7420,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -7445,9 +7478,9 @@ "issue:45290|issue:45381", "issue:43606|issue:45083", "issue:43335|issue:43550", - "issue:45200|issue:45588", "issue:44448|issue:44743", "issue:37428|issue:44351", + "issue:45200|issue:45588", "issue:43611|issue:43931", "issue:39692|issue:43526", "issue:44462|issue:44561", @@ -7460,9 +7493,9 @@ "issue:44855|issue:45083", "issue:45588|issue:45859", "issue:43550|issue:44112", - "issue:43335|issue:45083", "issue:29942|issue:32090", "issue:43335|issue:43504", + "issue:43335|issue:45083", "issue:39692|issue:44493", "issue:44292|issue:45520", "issue:38175|issue:44493", @@ -7473,255 +7506,255 @@ "issue:43720|issue:44112", "issue:44977|issue:45229", "issue:44448|issue:44977", - "issue:44568|issue:45491", "issue:43606|issue:44857", + "issue:44568|issue:45491", "issue:44857|issue:45072", "issue:45459|issue:45712" ], - "prepared_review_unit_hash": "ba8689049f719a6e53eba3d2b8eda9f35a2ef6a03e8d3a8bd06279350cc2cdbb", + "prepared_review_unit_hash": "ab2b5d43213756b27829843f65cf00cbcf03e43d2aed5b0bcda4835a2d6020fb", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 35, - "serialized_chars": 38458, - "estimated_input_tokens": 9615, - "estimated_eval_tokens": 19486 + "serialized_chars": 38456, + "estimated_input_tokens": 9614, + "estimated_eval_tokens": 19484 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most items are unrelated issue reports; only one soft pair looks like a true duplicate. The two Qwen3.5 Moe from_pretrained reports are the same bug, while the rest are different failure modes or subsystems.", - "confidence": 0.95, - "canonical_issue_reason": "issue:45310 is the best canonical issue because it is the earlier of the two exact duplicate reports for the same Qwen3.5 Moe from_pretrained regression.", + "summary": "This cluster is mostly false-positive similarity: the only clear duplicate is the repeated Qwen3.5 MoE from_pretrained error. The rest are related only at a broad subsystem level (tokenizers, flash-attn, loading, vision, CI) but describe different bugs.", + "confidence": 0.93, + "canonical_issue_reason": "issue:45310 is the best canonical issue because it is the earlier report of the exact same Qwen3.5 MoE from_pretrained failure later repeated as issue:45478.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45310 is the strongest issue candidate overall: it has the exact duplicate match, is earlier, and best represents the confirmed duplicate bug in this set.", + "best_issue_reason": "issue:45310 is the strongest representative because it has an exact duplicate and a concrete, narrowly scoped failure mode with clear reproducibility.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45290", "right": "issue:45381", "accept": false, - "reason": "Different bugs: chat template/tool-call crash vs qwen2.5-vl video position-id mismatch." + "reason": "Different bugs: chat template/tool-call handling vs Qwen2.5-VL video position IDs." }, { "left": "issue:43606", "right": "issue:45083", "accept": false, - "reason": "Different subsystems and failures: CPU offload device mismatch vs helper length computation in qwen3_omni_moe." + "reason": "Unrelated failures in different models and code paths." }, { "left": "issue:43335", "right": "issue:43550", "accept": false, - "reason": "Different concrete problems: SwitchTransformers sparse-layer config bug vs Bamba SDPA torch.compile failure." - }, - { - "left": "issue:45200", - "right": "issue:45588", - "accept": false, - "reason": "Different bugs: Gemma4 mm_token_type_ids defaults vs flash_attention s_aux=None crash." + "reason": "Sparse-layer config bug is unrelated to Bamba torch.compile/SDPA failure." }, { "left": "issue:44448", "right": "issue:44743", "accept": false, - "reason": "Different models and code paths: Pegasus v4/v5 output drift vs qwen3_5 recurrent-state reset." + "reason": "Different model families and different regressions (output drift vs recurrent state reset)." }, { "left": "issue:37428", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for different missing symbols and different code paths." + "reason": "Both are import errors, but for different missing symbols and different modules." + }, + { + "left": "issue:45200", + "right": "issue:45588", + "accept": false, + "reason": "Gemma4 token-type defaults and flash_attention s_aux handling are separate issues." }, { "left": "issue:43611", "right": "issue:43931", "accept": false, - "reason": "Different loading failures: base_model_prefix handling vs Qwen3-VL weight shape mismatch." + "reason": "Both loading errors, but one is a base_model_prefix regression and the other is a shape mismatch." }, { "left": "issue:39692", "right": "issue:43526", "accept": false, - "reason": "Different BEiT/SigLIP2 issues: documentation example errors vs BeitImageProcessorFast reduce_labels bug." + "reason": "Different image-processing problems in different models; not the same underlying bug." }, { "left": "issue:44462", "right": "issue:44561", "accept": false, - "reason": "Different regressions: AutoTokenizer ignoring tokenizer.json vs removal of is_torch_fx_available breaking trust_remote_code." + "reason": "Tokenizer-file selection and removal of is_torch_fx_available are unrelated." }, { "left": "issue:44206", "right": "issue:45823", "accept": false, - "reason": "Different model bugs: unsupported center arg in LasrFeatureExtractor vs Gemma4 PLE device mismatch." + "reason": "Feature extractor arg validation vs Gemma4 device-map mismatch are not the same code-path." }, { "left": "issue:44561", "right": "issue:44779", "accept": false, - "reason": "Unrelated breakages: trust_remote_code import regression vs Deepseek tokenizer behavior change." + "reason": "A missing utility symbol and a Deepseek tokenizer regression are different failures." }, { "left": "issue:43388", "right": "issue:43526", "accept": false, - "reason": "Different data handling bugs: gather_for_metrics last-batch label truncation vs BeitImageProcessorFast reduce_labels output." + "reason": "Both concern labels, but one is metric gathering and the other is Beit label reduction." }, { "left": "issue:43335", "right": "issue:43653", "accept": false, - "reason": "Different issues: SwitchTransformers sparse-layer creation vs BigBirdTokenizer special-token registration." + "reason": "SwitchTransformers sparse-layer config and BigBird tokenizer special-token registration are unrelated." }, { "left": "issue:43819", "right": "issue:44493", "accept": false, - "reason": "Different bugs: DAC.from_latents mismatch vs unexpected position-id key warnings across models." + "reason": "DAC latent conversion mismatch and unexpected position-id keys are different bugs." }, { "left": "issue:45310", "right": "issue:45478", "accept": true, - "reason": "Exact same Qwen3.5 Moe from_pretrained error; same title and same regression report." + "reason": "Exact same Qwen3.5 MoE from_pretrained error; clear duplicate." }, { "left": "issue:44855", "right": "issue:45083", "accept": false, - "reason": "Different failures: Python 3.13 import indentation error vs qwen3_omni_moe feature-length helper behavior." + "reason": "Python 3.13 JIT parsing error is unrelated to qwen3_omni_moe feature-length behavior." }, { "left": "issue:45588", "right": "issue:45859", "accept": false, - "reason": "Different flash-attention-related bugs: s_aux=None crash vs CPU-offload incompatibility in text rotary embedding." + "reason": "Both touch flash/offload paths, but the concrete failure modes differ." }, { "left": "issue:43550", "right": "issue:44112", "accept": false, - "reason": "Different problems: Bamba torch.compile/SDPA failure vs GraniteSpeech stale device override test." - }, - { - "left": "issue:43335", - "right": "issue:45072", - "accept": false, - "reason": "Different failures: SwitchTransformers sparse-layer config bug vs dtype mismatch in bfloat16 inference." + "reason": "Bamba SDPA compile failure and GraniteSpeech CI device-override stale test are unrelated." }, { "left": "issue:29942", "right": "issue:32090", "accept": false, - "reason": "Different issues: Flash Attention 2 tests failing vs Trainer NoneType broadcast error." + "reason": "Flash Attention 2 test failures and Trainer _gpu_broadcast_one TypeError are different issues." }, { "left": "issue:43335", "right": "issue:43504", "accept": false, - "reason": "Different model/config bugs: sparse-layer creation vs BEiT legacy-field loading failure." + "reason": "Sparse-layer creation bug is unrelated to Beit pretrained preset loading." + }, + { + "left": "issue:43335", + "right": "issue:45083", + "accept": false, + "reason": "Different model families and different helper/config bugs." }, { "left": "issue:39692", "right": "issue:44493", "accept": false, - "reason": "Different symptom sets: SigLIP2 doc/model mismatch vs unexpected key with position id." + "reason": "SigLIP2 docs/model mismatch is unrelated to widespread position-id key warnings." }, { "left": "issue:44292", "right": "issue:45520", "accept": false, - "reason": "Different runtime failures: Qwen-3-8B-NVFP4 error vs flash_attn import KeyError on Python 3.13." + "reason": "Qwen-3-8B-NVFP4 runtime failure and flash_attn import KeyError are different problems." }, { "left": "issue:38175", "right": "issue:44493", "accept": false, - "reason": "Different model behaviors: zero probabilities in SigLIP2 vs unexpected position-id key warnings." + "reason": "Unexpected zero probabilities in SigLIP2 is not the same as position-id key warnings." }, { "left": "issue:44448", "right": "issue:44779", "accept": false, - "reason": "Different models and regressions: Pegasus output mismatch vs Deepseek tokenizer breakage in v5." + "reason": "Both are regressions across versions, but in different models and failure modes." }, { "left": "issue:28218", "right": "issue:35141", "accept": false, - "reason": "Different tokenizer/model initialization bugs: added-token spacing vs output embedding reinitialization after resizing." + "reason": "Tokenizer spacing bug and embedding reinitialization on resize are unrelated." }, { "left": "issue:43335", "right": "issue:45072", "accept": false, - "reason": "Different failures: SwitchTransformers sparse-layer creation vs dtype mismatch during bfloat16 inference." + "reason": "Sparse-layer config bug and bfloat16 dtype mismatch are different code paths." }, { "left": "issue:43335", "right": "issue:45593", "accept": false, - "reason": "Different model issues: SwitchTransformers config bug vs D-FINE auxiliary-loss handling." + "reason": "SwitchTransformers config issue is unrelated to D-FINE auxiliary-loss behavior." }, { "left": "issue:43720", "right": "issue:44112", "accept": false, - "reason": "Different issues: BitNet packed-weight loading vs GraniteSpeech device override test flakiness." + "reason": "Packed-weight loading bug and GraniteSpeech CI stale test are unrelated." }, { "left": "issue:44977", "right": "issue:45229", "accept": false, - "reason": "Different bugs: Qwen3.5 flash-attention generation failure vs Gemma4 multi-GPU CUDA OOM." + "reason": "Qwen3.5 flash-attention generation bug and Gemma4 CUDA OOM are different failures." }, { "left": "issue:44448", "right": "issue:44977", "accept": false, - "reason": "Different models and code paths: Pegasus v4/v5 output difference vs Qwen3.5 flash-attention failure." + "reason": "Different models and different output-generation regressions." }, { - "left": "issue:44568", - "right": "issue:45491", + "left": "issue:43606", + "right": "issue:44857", "accept": false, - "reason": "Different tokenizer/model bugs: mdeberta add_special_tokens regression vs Gemma3 mixed-length NaN embeddings." + "reason": "CPU offload device mismatch and float16 AMP crash are separate issues." }, { - "left": "issue:43606", - "right": "issue:44857", + "left": "issue:44568", + "right": "issue:45491", "accept": false, - "reason": "Different failures: CPU offload device mismatch vs float16 AMP crash in LwDetrImageLoss." + "reason": "Tokenizer special-token behavior and Gemma3 NaN embeddings are unrelated." }, { "left": "issue:44857", "right": "issue:45072", "accept": false, - "reason": "Both involve mixed precision, but the models and failure mechanisms are unrelated." + "reason": "Float16 AMP crash and bfloat16 dtype mismatch are not the same bug." }, { "left": "issue:45459", "right": "issue:45712", "accept": false, - "reason": "Different bugs: tokenizer error masking with protobuf absent vs leftover dummy classes leaking into API surface." + "reason": "Tokenizer error masking and leftover dummy classes failing checks are unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded in the packet: issue:45310 and issue:45478 share the same title and clearly point to the same Qwen3.5 Moe from_pretrained regression, so choosing the earlier report as canonical is reasonable. The other soft pairs are handled conservatively as non-matches based on distinct titles/subsystems." + "feedback": "The judgment is conservative and grounded in the packet. The only clearly supported duplicate is issue:45310/issue:45478, which share the same title and Qwen3.5 MoE from_pretrained failure. The remaining soft-edge rejections stay at the safe side of the evidence and do not overclaim duplication or mergeability." }, "evaluator_used": true, "retried": false, @@ -7737,8 +7770,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -7758,7 +7791,6 @@ "issue:43278", "issue:43381", "issue:43412", - "issue:43421", "issue:43526", "issue:43550", "issue:43653", @@ -7768,6 +7800,7 @@ "issue:43976", "issue:43994", "issue:44186", + "issue:44361", "issue:44448", "issue:44451", "issue:44462", @@ -7775,6 +7808,7 @@ "issue:44493", "issue:44743", "issue:44857", + "issue:44871", "issue:44977", "issue:45083", "issue:45198", @@ -7783,7 +7817,6 @@ "issue:45412", "issue:45440", "issue:45459", - "issue:45491", "issue:45520", "issue:45563", "issue:45588", @@ -7793,14 +7826,14 @@ ], "soft_pairs": [ "issue:29127|issue:32090", - "issue:44977|issue:45563", "issue:43927|issue:45698", - "issue:45198|issue:45803", "issue:34689|issue:35141", + "issue:45198|issue:45803", + "issue:44977|issue:45563", "issue:40444|issue:43526", "issue:43701|issue:45083", - "issue:45372|issue:45588", "issue:43550|issue:45083", + "issue:45372|issue:45588", "issue:43412|issue:45588", "issue:45245|issue:45563", "issue:29942|issue:35141", @@ -7811,219 +7844,226 @@ "issue:44857|issue:45588", "issue:44448|issue:44466", "issue:43065|issue:45412", - "issue:43278|issue:43381", - "issue:43653|issue:44743", "issue:45588|issue:45823", + "issue:43278|issue:43381", "issue:25251|issue:32090", + "issue:43653|issue:44743", "issue:44451|issue:45520", "issue:43976|issue:45440", "issue:44186|issue:45198", "issue:42915|issue:42994", "issue:43994|issue:45412", - "issue:43421|issue:45491" + "issue:44361|issue:45198", + "issue:44871|issue:44977" ], - "prepared_review_unit_hash": "6a30dfacdda4571cabe2203178f2a1ff8f77005a093002709f8c7459526bb7b8", + "prepared_review_unit_hash": "0c711cf7c5b119f011fb4034dd9be5c0e0b579d01792d37d5929e3dc21fc3be9", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36380, - "estimated_input_tokens": 9095, - "estimated_eval_tokens": 18446 + "soft_pair_count": 30, + "serialized_chars": 36619, + "estimated_input_tokens": 9155, + "estimated_eval_tokens": 18566 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is mostly a collection of unrelated singleton bugs; none of the soft pairs look like true duplicates or mergeable same-bug reports.", - "confidence": 0.94, - "canonical_issue_reason": "issue:43873 is the most active, broadly scoped, still-open bug report in the set, so it is the best anchor even though the items are largely unrelated.", + "summary": "This set is not a duplicate cluster; the soft links are mostly superficial lexical matches across unrelated bugs. I found no PRs and no convincing canonical duplicate issue.", + "confidence": 0.86, + "canonical_issue_reason": "No single issue is a reliable canonical representative here: the linked items span unrelated code paths (tokenizers, generation, flash-attention, quantization, loading, eval/training behavior) and do not describe the same underlying bug.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 has the strongest combination of breadth, discussion, and ongoing relevance, making it the best representative issue here.", + "best_issue_reason": "If one issue must be selected as the closest thing to a representative, issue 45588 is the most concrete, reproducible core crash report in a central integration path; however, it is not a duplicate of the others.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:29127", "right": "issue:32090", "accept": false, - "reason": "Different subsystems and failure modes: LayoutLMv3 error messaging vs Trainer GPU broadcast NoneType." + "reason": "Different subsystems and failures: LayoutLMv3 box validation/error messaging vs Trainer GPU broadcast rejecting NoneType." }, { - "left": "issue:44977", - "right": "issue:45563", + "left": "issue:43927", + "right": "issue:45698", "accept": false, - "reason": "Both involve generation, but one is a flash-attention correctness bug and the other is a stale warning in paged generate()." + "reason": "Both involve save/load behavior, but one is about custom token IDs being lost and the other about loading the wrong custom module; different root causes." }, { - "left": "issue:43927", - "right": "issue:45698", + "left": "issue:34689", + "right": "issue:35141", "accept": false, - "reason": "Both are save/load related, but one loses custom token IDs while the other loads the wrong custom module; different code paths." + "reason": "Model-loading breakage for a specific vision model is unrelated to embedding reinitialization after resizing token embeddings." }, { "left": "issue:45198", "right": "issue:45803", "accept": false, - "reason": "Wav2Vec2 save/tokenization failure is unrelated to Fuyu's broad exception swallowing in tensor conversion." + "reason": "Both touch tokenization/saving, but the reported failures are on different models and different code paths." }, { - "left": "issue:34689", - "right": "issue:35141", + "left": "issue:44977", + "right": "issue:45563", "accept": false, - "reason": "Model loading regression for Llama vision is not the same as post_init reinitializing untied output embeddings." + "reason": "Flash-attention generation failure and a stale warning about num_return_sequences are unrelated bugs." }, { "left": "issue:40444", "right": "issue:43526", "accept": false, - "reason": "Multiple images per prompt in Qwen2.5-VL and BeitImageProcessorFast label reduction are unrelated bugs." + "reason": "IterableDataset multi-image prompt failure for Qwen2.5-VL is unrelated to BeitImageProcessorFast reduce_labels behavior." }, { "left": "issue:43701", "right": "issue:45083", "accept": false, - "reason": "Checkpoint key mismatch is unrelated to qwen3_omni_moe feature-length helper behavior." + "reason": "Checkpoint resume key mismatch is not the same bug as qwen3_omni_moe feature-length helper behavior." }, { - "left": "issue:45372", - "right": "issue:45588", + "left": "issue:43550", + "right": "issue:45083", "accept": false, - "reason": "Gemma4 processor import failure and flash_attention s_aux=None crash are different errors in different components." + "reason": "torch.compile with SDPA on Bamba-9B-v2 is a different concrete failure than qwen3_omni_moe feature-length calculation." }, { - "left": "issue:43550", - "right": "issue:45083", + "left": "issue:45372", + "right": "issue:45588", "accept": false, - "reason": "torch.compile/SDPA failure in Bamba is unrelated to qwen3_omni_moe feature-length computation." + "reason": "Gemma 4 processor import failure due to missing ReasoningEffort is unrelated to flash_attention.py crashing on s_aux=None." }, { "left": "issue:43412", "right": "issue:45588", "accept": false, - "reason": "ExecuTorch export guard/support issue is unrelated to sink-less flash_attention AttributeError." + "reason": "Executorch export issues in gemma3n are a different path than sink-less flash-attention integration crashing on s_aux=None." }, { "left": "issue:45245", "right": "issue:45563", "accept": false, - "reason": "Category-cardinality runtime error and stale generate warning are not the same bug." + "reason": "Category cardinality overflow and a stale generation warning are unrelated." }, { "left": "issue:29942", "right": "issue:35141", "accept": false, - "reason": "Flash Attention 2 test failures do not match embedding reinitialization after resize." + "reason": "Flash Attention 2 test failures do not match an embedding reinitialization bug in post_init." }, { "left": "issue:43873", "right": "issue:44493", "accept": false, - "reason": "Quantization/offloading behavior and unexpected position-id keys are distinct issues." + "reason": "Quantization offloading problems are unrelated to unexpected position-id key warnings." }, { "left": "issue:36010", "right": "issue:37428", "accept": false, - "reason": "Both are import errors, but they concern different symbols and different modules." + "reason": "Two distinct ImportError regressions involving different symbols and modules." }, { "left": "issue:43066", "right": "issue:44462", "accept": false, - "reason": "Tokenizer decoder type mismatch in v5 is not the same as AutoTokenizer ignoring tokenizer.json." + "reason": "Wrong tokenizer decoder type in v5 is not the same as AutoTokenizer ignoring tokenizer.json from the repo." }, { "left": "issue:45372", "right": "issue:45459", "accept": false, - "reason": "Different tokenizer-related failures: missing ReasoningEffort vs protobuf exception masking." + "reason": "One is a missing import dependency for Gemma 4 processing; the other is exception masking when protobuf is absent." }, { "left": "issue:44857", "right": "issue:45588", "accept": false, - "reason": "AMP/CUDA loss crash in LwDetr is unrelated to flash_attention sink handling." + "reason": "CUDA float16 loss crash in LwDetrImageLoss is unrelated to flash_attention.py's s_aux=None AttributeError." }, { "left": "issue:44448", "right": "issue:44466", "accept": false, - "reason": "One is output differences across versions; the other is inconsistent weight serialization by device." + "reason": "Different issues around model output/serialization across versions; they do not share the same underlying bug." }, { "left": "issue:43065", "right": "issue:45412", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder and RT-DETR memory reclamation are unrelated." + "reason": "Dummy Conv2d in Sam3PixelDecoder and RT-DETR memory release are separate model lifecycle issues." }, { - "left": "issue:43278", - "right": "issue:43381", + "left": "issue:45588", + "right": "issue:45823", "accept": false, - "reason": "Eval-mode dtype drift and gradient-checkpointing restrictions are different symptoms and fixes." + "reason": "Both involve Gemma 4/flash-attention adjacent code, but one is a sink-less flash-attention crash and the other is a device mismatch in PLE forward; not the same bug." }, { - "left": "issue:43653", - "right": "issue:44743", + "left": "issue:43278", + "right": "issue:43381", "accept": false, - "reason": "Tokenizer special-token registration and recurrent-state reset with cache are unrelated." + "reason": "Training/eval dtype drift is not the same as gradient checkpointing being disallowed in eval mode." }, { - "left": "issue:45588", - "right": "issue:45823", + "left": "issue:25251", + "right": "issue:32090", "accept": false, - "reason": "flash_attention None-handling crash and Gemma4 device mismatch are different forward-path bugs." + "reason": "Pipeline top_k nesting behavior and Trainer NoneType GPU broadcast are unrelated." }, { - "left": "issue:25251", - "right": "issue:32090", + "left": "issue:43653", + "right": "issue:44743", "accept": false, - "reason": "Pipeline top_k nesting behavior is unrelated to Trainer GPU broadcast NoneType handling." + "reason": "Tokenizer special-token registration causing empty decode is unrelated to recurrent states resetting with cache in qwen3_omni_moe." }, { "left": "issue:44451", "right": "issue:45520", "accept": false, - "reason": "Model loading failure for ScandiBERT and Python 3.13 flash_attn import KeyError are unrelated." + "reason": "ScandiBERT loading failure and Python 3.13 flash_attn import key error are separate environment/model-loading problems." }, { "left": "issue:43976", "right": "issue:45440", "accept": false, - "reason": "Python-version compatibility issue and DeepSeekV3MoE divergence are not the same problem." + "reason": "Python version compatibility and DeepseekV3MoE divergence are different issues." }, { "left": "issue:44186", "right": "issue:45198", "accept": false, - "reason": "LayoutLMv2 tokenizer padding/truncation crash and Wav2Vec2 save/tokenization failure are different tokenizer bugs." + "reason": "LayoutLMv2 NER/batched padding crashes are not the same as Wav2Vec2 save_pretrained/tokenization failures." }, { - "left": "issue:42915", - "right": "issue:42994", + "left": "issue:44361", + "right": "issue:45198", "accept": false, - "reason": "Quantization config failure on Qwen3Moe and quantized saving failure are related only broadly, not the same bug." + "reason": "MLukeTokenizer AttributeError on tasks is unrelated to Wav2Vec2 save/tokenization failure." }, { - "left": "issue:43994", - "right": "issue:45412", + "left": "issue:44871", + "right": "issue:44977", "accept": false, - "reason": "SigLIP2 inference weirdness and RT-DETR memory release are unrelated." + "reason": "EOS token configuration inconsistency is unrelated to flash-attention generation behavior in Qwen3.5." }, { - "left": "issue:43421", - "right": "issue:45491", + "left": "issue:45083", + "right": "issue:45588", + "accept": false, + "reason": "Feature-extraction length helper behavior in qwen3_omni_moe is unrelated to a flash_attention integration crash." + }, + { + "left": "issue:45588", + "right": "issue:45823", "accept": false, - "reason": "Runtime special-token post-processor updates and Gemma3 NaN embeddings are different components and failure modes." + "reason": "Different Gemma 4 failures: one is an AttributeError in flash-attention integration, the other is a device placement mismatch during forward." } ] }, @@ -8042,8 +8082,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -8061,6 +8101,7 @@ "issue:43329", "issue:43334", "issue:43381", + "issue:43421", "issue:43526", "issue:43531", "issue:43575", @@ -8076,13 +8117,11 @@ "issue:44206", "issue:44265", "issue:44279", - "issue:44361", "issue:44442", "issue:44479", "issue:44554", "issue:44610", "issue:44734", - "issue:44871", "issue:44977", "issue:45103", "issue:45198", @@ -8092,25 +8131,25 @@ "issue:45406", "issue:45412", "issue:45440", + "issue:45491", "issue:45593", "issue:45834", "issue:45859" ], "soft_pairs": [ - "issue:44361|issue:45198", - "issue:44871|issue:44977", + "issue:43421|issue:45491", "issue:44206|issue:44479", "issue:44265|issue:45198", - "issue:44734|issue:45406", "issue:29942|issue:30064", + "issue:44734|issue:45406", "issue:43065|issue:43994", "issue:33357|issue:39692", "issue:29127|issue:34567", "issue:44610|issue:45412", "issue:45103|issue:45372", "issue:43756|issue:45440", - "issue:43723|issue:44554", "issue:43873|issue:45229", + "issue:43723|issue:44554", "issue:45372|issue:45406", "issue:43278|issue:43575", "issue:43761|issue:45859", @@ -8118,217 +8157,211 @@ "issue:43381|issue:43701", "issue:43653|issue:44442", "issue:45440|issue:45593", - "issue:41720|issue:45229", "issue:43278|issue:43329", + "issue:41720|issue:45229", "issue:29942|issue:36010", "issue:32090|issue:33357", "issue:43334|issue:43531", - "issue:44977|issue:45245", "issue:43526|issue:44186", + "issue:44977|issue:45245", "issue:43824|issue:44279" ], - "prepared_review_unit_hash": "b3241eac8893d05a59e4a3fa0cc7ea0e5622638483496ff8fa8065dd58ad3f61", + "prepared_review_unit_hash": "7aa0c03a362c4c4c2027fc81afdbb50f3d44f6a435f8f890e4ce1dc92e4a2f01", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36452, - "estimated_input_tokens": 9113, - "estimated_eval_tokens": 18482 + "soft_pair_count": 28, + "serialized_chars": 36197, + "estimated_input_tokens": 9050, + "estimated_eval_tokens": 18356 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The items are a heterogeneous set of unrelated issues, and none of the soft pairs look like true duplicates of the same underlying bug or change. I rejected all soft edges and found no sensible single canonical issue/PR for this cluster.", - "confidence": 0.94, - "canonical_issue_reason": "No clear canonical issue: the set spans many unrelated subsystems (tokenizers, vision models, trainers, serving, offloading, docs), so there is no single representative duplicate target.", + "summary": "This set is heterogeneous rather than a duplicate cluster. The soft pairs span distinct bugs across tokenizers, vision models, training, serving, and multimodal processors, so none should be merged as the same underlying issue.", + "confidence": 0.84, + "canonical_issue_reason": "Issue 43873 is the closest thing to a cluster anchor: it is open, broadly framed, and has the strongest discussion/inbound-reference footprint. It is still not a duplicate of the others, just the least-specific representative.", "canonical_pr_reason": null, - "best_issue_reason": "No strong global best issue for triage; the cluster is too mixed to nominate one issue as the representative duplicate anchor.", + "best_issue_reason": "Issue 43873 is the best representative of the set because it is broad and active, with multiple inbound references, but the cluster does not contain a true duplicate family.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44361", - "right": "issue:45198", - "accept": false, - "reason": "Both are tokenizer-related failures, but they involve different models and different failure modes (`AttributeError` on tasks vs save/tokenization failure), so they are not the same bug." - }, - { - "left": "issue:44871", - "right": "issue:44977", + "left": "issue:43421", + "right": "issue:45491", "accept": false, - "reason": "Different models and different problem types: EOS token config mismatch vs flash-attention generation failure." + "reason": "Runtime special-token post-processing bug vs Gemma3 NaN embeddings/sliding-window attention; different code paths and symptoms." }, { "left": "issue:44206", "right": "issue:44479", "accept": false, - "reason": "One is an unsupported argument in a feature extractor; the other is a video-input regression across multiple VLMs. Not the same code-path bug." + "reason": "Unsupported `center` arg in a feature extractor vs Qwen VL video-input regression; unrelated subsystems." }, { "left": "issue:44265", "right": "issue:45198", "accept": false, - "reason": "`torch.export.export` with `torch_compilable_check` is unrelated to Wav2Vec2 tokenizer/save_pretrained failures." + "reason": "`torch.export` / `torch_compilable_check` failure vs Wav2Vec2 save/tokenization failure; not the same bug." }, { - "left": "issue:44734", - "right": "issue:45406", + "left": "issue:29942", + "right": "issue:30064", "accept": false, - "reason": "Serving KV-cache tensor indexing and Gemma4 processor `_tokenizer` initialization are different failures in different code paths." + "reason": "Flash Attention 2 test failures vs void segmentation-map processing; different failure modes." }, { - "left": "issue:29942", - "right": "issue:30064", + "left": "issue:44734", + "right": "issue:45406", "accept": false, - "reason": "Flash Attention 2 test failures are unrelated to image processor segmentation-map handling." + "reason": "KV-cache continuation tensor indexing bug in serve vs Gemma4 processor missing `_tokenizer`; unrelated." }, { "left": "issue:43065", "right": "issue:43994", "accept": false, - "reason": "A dummy Conv2d in Sam3PixelDecoder is unrelated to SigLIP2 AutoModel/pipeline correctness." + "reason": "Dummy Conv2d in Sam3PixelDecoder vs SigLIP2 AutoModel/pipeline nonsense outputs; different models and defects." }, { "left": "issue:33357", "right": "issue:39692", "accept": false, - "reason": "MacOS bus error with a community CLIP model is not the same bug as documentation example model/processor mismatch and quantization errors." + "reason": "MacOS bus error on community CLIP model vs SigLIP2 docs/model-processor mismatch and quantization example failure; not one change." }, { "left": "issue:29127", "right": "issue:34567", "accept": false, - "reason": "LayoutLMv3 error messaging and TrainerState token-count tracking are unrelated issues." + "reason": "LayoutLMv3 error-message clarity issue vs TrainerState token counter not updating; unrelated features." }, { "left": "issue:44610", "right": "issue:45412", "accept": false, - "reason": "Processor image-size mismatch and RT-DETR memory not being released are distinct bugs." + "reason": "OmDet-Turbo image-size mismatch vs RT-DETR memory release issue; distinct bugs." }, { "left": "issue:45103", "right": "issue:45372", "accept": false, - "reason": "Auto-docstring annotation handling and Gemma4 processor loading/import dependencies are unrelated." + "reason": "auto_docstring annotation handling vs Gemma4 processor import dependency failure; different layers entirely." }, { "left": "issue:43756", "right": "issue:45440", "accept": false, - "reason": "Both concern model implementation differences, but they affect different models and different mechanisms; not the same concrete bug." + "reason": "Smollm3 RoPE-layer mismatch vs DeepSeekV3MoE divergence from remote impl; not the same code-path." }, { - "left": "issue:43723", - "right": "issue:44554", + "left": "issue:43873", + "right": "issue:45229", "accept": false, - "reason": "Tokenizer loading in v5 and an MPS attention correctness issue are unrelated code paths." + "reason": "General quantization/offloading behavior vs Gemma4 multi-GPU CUDA OOM; too broad and not the same defect." }, { - "left": "issue:43873", - "right": "issue:45229", + "left": "issue:43723", + "right": "issue:44554", "accept": false, - "reason": "Quantization/offloading behavior and multi-GPU CUDA OOM are different failure modes and not the same fix." + "reason": "AutoTokenizer loading regression vs MPS attention correctness issue; unrelated." }, { "left": "issue:45372", "right": "issue:45406", "accept": false, - "reason": "Both mention Gemma4, but one is an import-time dependency issue and the other is a serving-time processor attribute error." + "reason": "Processor-loading import error vs serve crash from missing `_tokenizer`; different surfaces and fixes." }, { "left": "issue:43278", "right": "issue:43575", "accept": false, - "reason": "Training/eval dtype drift is unrelated to tensor-parallel load OOM." + "reason": "Training/eval dtype drift vs tensor-parallel OOM on Qwen2-57B; not the same problem." }, { "left": "issue:43761", "right": "issue:45859", "accept": false, - "reason": "CLIPVisionModel hidden-state regression and Qwen3_5MoE CPU-offload incompatibility are different model/component bugs." + "reason": "CLIPVisionModel hidden_states regression vs CPU-offload incompatibility in Qwen3_5Moe rotary embedding; unrelated." }, { "left": "issue:43526", "right": "issue:45834", "accept": false, - "reason": "BeitImageProcessorFast label reduction and Kosmos2.5 OCR long-input indexing are unrelated processor bugs." + "reason": "BeitImageProcessorFast `reduce_labels` bug vs Kosmos2.5 long-OCR index error; different processors and failure modes." }, { "left": "issue:43381", "right": "issue:43701", "accept": false, - "reason": "Gradient-checkpointing eval-mode behavior and resume-from-checkpoint key mismatch are not the same underlying issue." + "reason": "Gradient checkpointing in eval mode vs resume-from-checkpoint key mismatch; not the same bug." }, { "left": "issue:43653", "right": "issue:44442", "accept": false, - "reason": "BigBirdTokenizer special-token registration and FastSpeech2ConformerTokenizer loading failure are different tokenizer bugs." + "reason": "BigBirdTokenizer special-token registration/decoding bug vs AutoTokenizer loading FastSpeech2ConformerTokenizer; distinct tokenizer issues." }, { "left": "issue:45440", "right": "issue:45593", "accept": false, - "reason": "Different models and different implementation gaps: DeepseekV3MoE divergence vs missing D-FINE auxiliary losses." + "reason": "DeepSeekV3MoE implementation divergence vs D-FINE auxiliary-loss behavior; different model logic." }, { - "left": "issue:41720", - "right": "issue:45229", + "left": "issue:43278", + "right": "issue:43329", "accept": false, - "reason": "A device-mapping cuda assert and a multi-GPU OOM are distinct runtime failures." + "reason": "Embedding dtype mismatch vs undefined variables in multimodal-token counting video branch; unrelated." }, { - "left": "issue:43278", - "right": "issue:43329", + "left": "issue:41720", + "right": "issue:45229", "accept": false, - "reason": "Embedding dtype changes and multimodal token counting/video-branch undefined symbols are unrelated." + "reason": "Qwen3 auto device-mapping assert vs Gemma4 multi-GPU OOM; different models and root causes." }, { "left": "issue:29942", "right": "issue:36010", "accept": false, - "reason": "Flash Attention 2 test regressions are unrelated to an import error for `GenerationMixin`." + "reason": "Flash Attention 2 tests vs `GenerationMixin` import error; no shared code path." }, { "left": "issue:32090", "right": "issue:33357", "accept": false, - "reason": "Trainer broadcast TypeError and MacOS bus error in CLIP are unrelated crashes." + "reason": "Trainer NoneType broadcast error vs MacOS bus error on CLIP model; unrelated runtime failures." }, { "left": "issue:43334", "right": "issue:43531", "accept": false, - "reason": "Qwen3-VL loading failure due to `pad_token_id` and Qwen3-MoE sliding-window behavior are different model-specific bugs." + "reason": "Qwen3-VL missing `pad_token_id` vs Qwen3-MoE sliding-window issue; both Qwen-family but different defects." }, { - "left": "issue:44977", - "right": "issue:45245", + "left": "issue:43526", + "right": "issue:44186", "accept": false, - "reason": "Flash-attention generation failure and category-count limit error are unrelated." + "reason": "Beit fast label-reduction bug vs LayoutLMv2Tokenizer NER/padding crash; different token/image processing logic." }, { - "left": "issue:43526", - "right": "issue:44186", + "left": "issue:44977", + "right": "issue:45245", "accept": false, - "reason": "Beit label reduction and LayoutLMv2 tokenizer padding/truncation crashes are different components and failures." + "reason": "Qwen3.5 flash-attention generation bug vs category-count ceiling error; unrelated." }, { "left": "issue:43824", "right": "issue:44279", "accept": false, - "reason": "A specific model import error is not the same as a vague dependency issue with transformers." + "reason": "Specific Qwen2_5_VL import failure vs generic dependency issue; too little evidence of the same underlying bug." } ] }, @@ -8347,8 +8380,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -8405,228 +8438,228 @@ "issue:43526|issue:44857", "issue:43262|issue:44206", "issue:43441|issue:45588", - "issue:44464|issue:45491", "issue:42915|issue:43931", - "issue:44743|issue:44811", "issue:44442|issue:45072", + "issue:44464|issue:45491", + "issue:44743|issue:44811", "issue:43575|issue:43701", "issue:43646|issue:44077", "issue:29127|issue:35141", "issue:43575|issue:44387", - "issue:44442|issue:45803", - "issue:44977|issue:45584", "issue:33357|issue:38175", + "issue:44977|issue:45584", + "issue:44442|issue:45803", "issue:43316|issue:43525", "issue:44964|issue:45440", "issue:44479|issue:45290", - "issue:43761|issue:44079", "issue:28218|issue:29127", + "issue:43761|issue:44079", "issue:42907|issue:45698", "issue:28218|issue:32090", "issue:43931|issue:45216", "issue:33290|issue:34567", "issue:34689|issue:43611", - "issue:39692|issue:45491", "issue:44442|issue:44898", + "issue:39692|issue:45491", "issue:43761|issue:44188", "issue:43381|issue:43873" ], - "prepared_review_unit_hash": "ca3e7191d5861bc6959c4833b3f5e20f18d83b5a019c000e32bee11e7fd43f47", + "prepared_review_unit_hash": "036b88c991f5f8ba7a120b8f0b42fd14aa328fdafcc33624dfaf2d5d4602f421", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36130, - "estimated_input_tokens": 9033, - "estimated_eval_tokens": 18322 + "serialized_chars": 36120, + "estimated_input_tokens": 9030, + "estimated_eval_tokens": 18316 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are heterogeneous and the proposed soft pairs do not appear to be the same underlying bug or change. Most only share a broad subsystem name (e.g. flash attention, loading/saving, OOM) but differ in concrete code path and symptom.", - "confidence": 0.93, - "canonical_issue_reason": "No single canonical issue stands out because the cluster is not a true duplicate set; the items span unrelated bugs across tokenization, processors, model loading, quantization, training, and serialization.", + "summary": "These items are a heterogeneous mix of unrelated Transformers issues. None of the soft pairs appear to be the same underlying bug/change, so all soft edges should be rejected.", + "confidence": 0.92, + "canonical_issue_reason": "No single issue is clearly canonical for this set: the candidates span unrelated tokenizer, audio, vision, training, quantization, and model-loading problems.", "canonical_pr_reason": null, - "best_issue_reason": "No global best issue is suitable: none of these issues represent the shared problem across the set, so picking one would be arbitrary.", + "best_issue_reason": "No global best issue stands out as a representative duplicate target because the cluster is not a single coherent problem area.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43526", "right": "issue:44857", "accept": false, - "reason": "Different failures: label reduction bug in BEiT fast processor vs AMP/CUDA crash in image loss." + "reason": "Different bugs in different components: BeitImageProcessorFast label reduction vs AMP/CUDA crash in LwDetrImageLoss." }, { "left": "issue:43262", "right": "issue:44206", "accept": false, - "reason": "Audio chat-template sampling-rate default mismatch is unrelated to LasrFeatureExtractor's unsupported argument crash." + "reason": "Both are audio-related, but one is a sampling-rate default in apply_chat_template and the other is an unsupported argument passed by LasrFeatureExtractor." }, { "left": "issue:43441", "right": "issue:45588", "accept": false, - "reason": "Both involve flash attention, but one is a Ministral-3 v5 RC regression and the other is a sink-less model AttributeError on s_aux=None." + "reason": "Both mention flash attention, but they involve different models and different failures: Ministral-3 vs sink-less models with s_aux=None." }, { - "left": "issue:44464", - "right": "issue:45491", + "left": "issue:42915", + "right": "issue:43931", "accept": false, - "reason": "Compiled-forward chunked generation inconsistency is a different bug from Gemma3 NaN embeddings on mixed-length sliding-window batches." + "reason": "Different model/config issues: FineGrainedFP8Config failure for Qwen3Moe vs weight-shape mismatch loading Qwen3-VL-30B." }, { - "left": "issue:42915", - "right": "issue:43931", + "left": "issue:44442", + "right": "issue:45072", "accept": false, - "reason": "Qwen3Moe FP8Config failure and Qwen3-VL weight-shape mismatch are different model/load paths with different symptoms." + "reason": "Unrelated problems: tokenizer loading failure vs dtype mismatches in bfloat16 inference." }, { - "left": "issue:44743", - "right": "issue:44811", + "left": "issue:44464", + "right": "issue:45491", "accept": false, - "reason": "Recurrent-state reset in modular_qwen3_5 is unrelated to Whisper batch_decode ignoring skip_special_tokens." + "reason": "Different generation/runtime bugs: compiled-forward chunked generation inconsistency vs NaNs from sliding-window attention on mixed-length batches." }, { - "left": "issue:44442", - "right": "issue:45072", + "left": "issue:44743", + "right": "issue:44811", "accept": false, - "reason": "Tokenizer load failure and dtype mismatches in inference are unrelated bugs." + "reason": "Different code paths and symptoms: Qwen3.5 recurrent state reset vs Whisper batch_decode skip_special_tokens handling." }, { "left": "issue:43575", "right": "issue:43701", "accept": false, - "reason": "Tensor-parallel OOM while loading a model is not the same as a resume_from_checkpoint key mismatch." + "reason": "Different classes of failures: tensor-parallel OOM during model load vs resume_from_checkpoint key mismatch." }, { "left": "issue:43646", "right": "issue:44077", "accept": false, - "reason": "Broad custom initialization regression vs a specific patchtsmixer post_init policy issue; not the same bug." + "reason": "Related to initialization, but not the same bug: generic custom model init regression vs patchtsmixer-specific post_init allowance." }, { "left": "issue:29127", "right": "issue:35141", "accept": false, - "reason": "LayoutLMv3 error-message clarity issue is unrelated to embedding reinitialization after resize_token_embeddings." + "reason": "Different subsystems: LayoutLMv3 error-message clarity vs embedding reinitialization after resize when tie_word_embedding is False." }, { "left": "issue:43575", "right": "issue:44387", "accept": false, - "reason": "Both are OOM-related, but one is TP load OOM and the other is increased CUDA reserved memory under int4 quantization." + "reason": "Both are OOM reports, but one is TP loading OOM and the other is increased reserved memory under int4 quantization." }, - { - "left": "issue:44442", - "right": "issue:45803", + { + "left": "issue:33357", + "right": "issue:38175", "accept": false, - "reason": "AutoTokenizer load failure is unrelated to FuyuBatchFeature's bare except swallowing interrupts/errors." + "reason": "Different models and failure modes: MacOS bus error on CLIP vs unexpected zero probabilities on SigLIP2." }, { "left": "issue:44977", "right": "issue:45584", "accept": false, - "reason": "Qwen3.5 flash-attention generation issue and Whisper empty-transcription failure after align_special_tokens are different code paths." + "reason": "Both affect generation, but the causes are different: flash-attention behavior in Qwen3.5 vs empty-transcription handling after align_special_tokens in Whisper." }, { - "left": "issue:33357", - "right": "issue:38175", + "left": "issue:44442", + "right": "issue:45803", "accept": false, - "reason": "MacOS bus error with CLIP model loading is not the same as SigLIP2 zero-probability outputs." + "reason": "Tokenizer loading failure vs a bare-except swallowing errors in FuyuBatchFeature; not the same bug." }, { "left": "issue:43316", "right": "issue:43525", "accept": false, - "reason": "Gemma3TextConfig API discrepancy and Llama4Config missing pad_token_id are distinct config-contract bugs." + "reason": "Both are config/API discrepancies, but they concern different models and different missing attributes." }, { "left": "issue:44964", "right": "issue:45440", "accept": false, - "reason": "Phi-4 multimodal loading failure is not the same as DeepseekV3MoE diverging from the remote implementation." + "reason": "Different problems: loading Phi-4-multimodal-instruct fails, while DeepseekV3MoE diverges from the remote implementation." }, { "left": "issue:44479", "right": "issue:45290", "accept": false, - "reason": "Video input regression across Qwen VL models is unrelated to apply_chat_template crashing on tool-call assistant messages with no content." + "reason": "Different regressions: video-input handling for Qwen VL models vs chat-template crashes on assistant tool-call messages." }, { - "left": "issue:43761", - "right": "issue:44079", + "left": "issue:28218", + "right": "issue:32090", "accept": false, - "reason": "CLIPVisionModel hidden_states regression is unrelated to ModelOutput key assignment when prior values were None." + "reason": "Tokenizer spacing bug vs Trainer GPU broadcast TypeError; unrelated." }, { - "left": "issue:28218", - "right": "issue:29127", + "left": "issue:43761", + "right": "issue:44079", "accept": false, - "reason": "Tokenizer spacing after added token is unrelated to layoutlmv3 error-message clarity." + "reason": "CLIPVisionModel hidden_states regression vs ModelOutput key-assignment bug; distinct code paths and symptoms." }, { "left": "issue:42907", "right": "issue:45698", "accept": false, - "reason": "Saving dequantized Ministral/Devstral models and loading the wrong custom module after save_pretrained are different serialization failures." + "reason": "Quantization/offloading issue vs save/load custom-module mismatch; not the same underlying bug." }, { "left": "issue:28218", "right": "issue:32090", "accept": false, - "reason": "Tokenizer extra-space behavior has no relation to Trainer _gpu_broadcast_one NoneType TypeError." + "reason": "Tokenizer spacing bug vs Trainer GPU broadcast TypeError; unrelated." }, { "left": "issue:43931", "right": "issue:45216", "accept": false, - "reason": "Model weight-shape mismatch on load is different from Qwen3.5 checkpoint corruption after save_pretrained." + "reason": "Loading weight-shape mismatch in Qwen3-VL vs incorrect checkpoint saving in Qwen3.5; different stages of the model lifecycle." }, { "left": "issue:33290", "right": "issue:34567", "accept": false, - "reason": "Adafactor OOM in deepspeed is unrelated to TrainerState.num_input_tokens_seen not updating." + "reason": "OOM with Adafactor in DeepSpeed vs TrainerState token counter not updating; unrelated." }, { "left": "issue:34689", "right": "issue:43611", "accept": false, - "reason": "Llama 3.2 Vision loading regression and base_model_prefix loading breakage are both load-related but not the same concrete defect." + "reason": "Both are loading regressions, but they target different versions and different causes." }, { - "left": "issue:39692", - "right": "issue:45491", + "left": "issue:44442", + "right": "issue:44898", "accept": false, - "reason": "SigLIP2 documentation example errors are unrelated to Gemma3 NaN embeddings during mixed-length batching." + "reason": "Tokenizer loading failure vs Perceiver non-default-resolution failure; unrelated." }, { - "left": "issue:44442", - "right": "issue:44898", + "left": "issue:39692", + "right": "issue:45491", "accept": false, - "reason": "Tokenizer loading failure is unrelated to Perceiver non-default resolution failure despite both being model/processor issues." + "reason": "Documentation/example errors vs a runtime NaN bug; not the same change or bug." }, { "left": "issue:43761", "right": "issue:44188", "accept": false, - "reason": "Hidden-states regression in CLIPVisionModel is unrelated to attention-kernel divergence under torch.compile." + "reason": "Different model/runtime issues: CLIPVisionModel hidden_states regression vs attention-kernel divergence under torch.compile." }, { "left": "issue:43381", "right": "issue:43873", "accept": false, - "reason": "Gradient checkpointing eval-mode restriction is unrelated to offloading behavior under quantization." + "reason": "Gradient checkpointing in eval mode vs quantization offloading behavior; different defects." } ] }, @@ -8645,8 +8678,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -8700,8 +8733,8 @@ "issue:45759" ], "soft_pairs": [ - "issue:43550|issue:44743", "issue:43493|issue:43827", + "issue:43550|issue:44743", "issue:44568|issue:45479", "issue:43316|issue:45759", "issue:41628|issue:45372", @@ -8716,26 +8749,26 @@ "issue:44077|issue:44479", "issue:33290|issue:45656", "issue:29127|issue:33357", - "issue:43316|issue:44877", - "issue:43540|issue:44514", "issue:43381|issue:43742", + "issue:43540|issue:44514", + "issue:43316|issue:44877", "issue:42175|issue:43316", "issue:33290|issue:34689", "issue:44162|issue:44792", "issue:43404|issue:43441", "issue:25251|issue:28282", - "issue:43232|issue:43388", "issue:41950|issue:43475", - "issue:43295|issue:43825" + "issue:43295|issue:43825", + "issue:43232|issue:43388" ], - "prepared_review_unit_hash": "745476ddb7da07127ec60f30a78904010a3494cbf3de5757bfe7501ec902afc7", + "prepared_review_unit_hash": "e536a6e270e6323be3306990329e8df0772ab1a277bb102732135f22307e76d9", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, @@ -8750,174 +8783,174 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous transformer bugs/regressions, not a tight duplicate set. The soft-similarity pairs are largely false positives across different subsystems and failure modes.", - "confidence": 0.39, - "canonical_issue_reason": "No true duplicate cluster emerges here; if one issue must anchor the set, issue 41950 is the most generally representative user-facing regression around pipeline/processor handling, but it is not a strong canonical duplicate target.", + "summary": "The set is a broad mix of unrelated bug reports across pipelines, config loading, multimodal models, and distributed training. None of the soft pairs look like the same underlying bug/change, so I reject all proposed duplicate links.", + "confidence": 0.91, + "canonical_issue_reason": "issue:41950 is the only open item and is phrased as a concrete, reproducible pipeline regression, making it the most useful anchor in this set.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 41950 is the best representative issue because it is a concrete, current regression with a clear behavioral failure and broad user impact; however, it still does not subsume the rest of the cluster.", + "best_issue_reason": "issue:41950 is the clearest and most actionable issue here: specific failure mode, current open status, and a well-scoped title.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43550", - "right": "issue:44743", + "left": "issue:43493", + "right": "issue:43827", "accept": false, - "reason": "Different models and failures: torch.compile/SDPA vs cache/state-reset logic." + "reason": "Different problems: SigLIP2 implementation mismatch vs stale docs referencing removed pipelines." }, { - "left": "issue:43493", - "right": "issue:43827", + "left": "issue:43550", + "right": "issue:44743", "accept": false, - "reason": "One is a model-implementation discrepancy; the other is documentation/pipeline deprecation cleanup." + "reason": "Different model/code paths and failures: torch.compile+SDPA issue vs cache/recurrent-state reset bug." }, { "left": "issue:44568", "right": "issue:45479", "accept": false, - "reason": "Tokenizer special-token regression is unrelated to sequence-classification zero-loss behavior." + "reason": "Tokenizer special-token behavior and sequence-classification loss behavior are unrelated." }, { "left": "issue:43316", "right": "issue:45759", "accept": false, - "reason": "Both concern config/model loading, but for different composite-model API shapes and different bugs." + "reason": "Both are config/AutoModel-adjacent, but they concern different models and different failure mechanisms." }, { "left": "issue:41628", "right": "issue:45372", "accept": false, - "reason": "Both are import-related, but they involve different missing symbols and different processors/models." + "reason": "Different import errors in different components; no shared code-path or symptom beyond import-time failure." }, { "left": "issue:43720", "right": "issue:44442", "accept": false, - "reason": "BitNet packed-weight loading and tokenizer loading are unrelated code paths." + "reason": "BitNet weight-loading bug and tokenizer loading bug are unrelated subsystems." }, { "left": "issue:43643", "right": "issue:44913", "accept": false, - "reason": "One is missing fields on trust_remote_code config load; the other is config value loss on reload." + "reason": "One is trust_remote_code returning incomplete config fields; the other is a GPTNeoX config persistence bug." }, { "left": "issue:32090", "right": "issue:33290", "accept": false, - "reason": "Trainer broadcast NoneType error and DeepSpeed Adafactor OOM are different runtime issues." + "reason": "Both involve Trainer/DeepSpeed, but one is a NoneType broadcast error and the other is an optimizer OOM." }, { "left": "issue:44190", "right": "issue:44496", "accept": false, - "reason": "Dataset loading in a script vs model-type/config recognition are separate failures." + "reason": "Dataset loading in a script and unrecognized model config loading are different issues." }, { "left": "issue:43873", "right": "issue:44871", "accept": false, - "reason": "Quantization offloading behavior and Gemma eos_token_id mismatch are unrelated." + "reason": "Quantization/offloading behavior and Gemma eos_token_id mismatch are unrelated." }, { "left": "issue:34689", "right": "issue:36010", "accept": false, - "reason": "Model-loading regression for a vision model and a GenerationMixin import error are different breakages." + "reason": "Model-loading regression for a specific model family vs a missing GenerationMixin import are different failures." }, { "left": "issue:43381", "right": "issue:43856", "accept": false, - "reason": "Gradient checkpointing eval-mode behavior is unrelated to MoE memory usage during training." + "reason": "Gradient checkpointing in eval mode and MoE training memory inefficiency are distinct problems." }, { "left": "issue:44855", "right": "issue:45137", "accept": false, - "reason": "Python 3.13 torch.jit parse/import issue is unrelated to DeepSpeed ZeRO3 deque underflow." + "reason": "Python 3.13 import parsing failure vs DeepSpeed ZeRO3 deque underflow are unrelated." }, { "left": "issue:44077", "right": "issue:44479", "accept": false, - "reason": "PatchTSMixer post_init API policy and video-input regression for Qwen models are different changes." + "reason": "PatchTSMixer post_init policy and Qwen video-input regression do not share the same bug." }, { "left": "issue:33290", "right": "issue:45656", "accept": false, - "reason": "Both mention optimizer/DeepSpeed, but one is OOM while the other is double optimizer stepping." + "reason": "Both mention DeepSpeed/optimizers, but OOM and double optimizer-step are different code-path failures." }, { "left": "issue:29127", "right": "issue:33357", "accept": false, - "reason": "LayoutLMv3 box-content error messaging and MacOS CLIP bus error are unrelated." + "reason": "LayoutLMv3 error messaging issue and CLIP bus error on macOS are unrelated." }, { - "left": "issue:43316", - "right": "issue:44877", + "left": "issue:43381", + "right": "issue:43742", "accept": false, - "reason": "Gemma3TextConfig API mismatch is not the same as strict config blocking granite_speech loading." + "reason": "Gradient checkpointing/eval issue is not the same as a MobileLLM key error on load." }, { "left": "issue:43540", "right": "issue:44514", "accept": false, - "reason": "Qwen3OmniMoE video processing ValueError and Qwen2_5_VL batched chat-template crash are distinct paths." + "reason": "Both are video/multimodal, but the failing inputs and processor code paths are different." }, { - "left": "issue:43381", - "right": "issue:43742", + "left": "issue:43316", + "right": "issue:44877", "accept": false, - "reason": "Gradient-checkpointing eval behavior does not match a MobileLLM config load key error." + "reason": "Different config-loading bugs for different model families; not the same underlying defect." }, { "left": "issue:42175", "right": "issue:43316", "accept": false, - "reason": "Packaging/backends issue is unrelated to Gemma3TextConfig API shape." + "reason": "Dependency/backend packaging issue vs Gemma config API discrepancy are unrelated." }, { "left": "issue:33290", "right": "issue:34689", "accept": false, - "reason": "DeepSpeed Adafactor OOM and Llama 3.2 Vision loading regression are unrelated." + "reason": "DeepSpeed optimizer OOM and a model-loading regression are different failure modes." }, { "left": "issue:44162", "right": "issue:44792", "accept": false, - "reason": "ESM2 model breakage and janus test failure are different artifacts and failure modes." + "reason": "ESM2 model breakage and a janus test failure are not the same bug." }, { "left": "issue:43404", "right": "issue:43441", "accept": false, - "reason": "Mistral3 weight tying bug and Ministral FlashAttention failure are separate model-specific regressions." + "reason": "Weight tying bug in one model class and FlashAttention failure in another are separate issues." }, { "left": "issue:25251", "right": "issue:28282", "accept": false, - "reason": "Pipeline top_k nested-list behavior is unrelated to PyTorch import absence." - }, - { - "left": "issue:43232", - "right": "issue:43388", - "accept": false, - "reason": "Generation kwargs handling after sync_gpus and metric gathering label truncation are different code paths." + "reason": "Pipeline top_k nesting bug is unrelated to missing PyTorch import error." }, { "left": "issue:41950", "right": "issue:43475", "accept": false, - "reason": "Video-classification processor selection and SAM3 video encoder attribute error are different bugs." + "reason": "Pipeline selecting image processors and SAM 3 video output attribute error are different model/pipeline bugs." }, { "left": "issue:43295", "right": "issue:43825", "accept": false, - "reason": "Processor/tokenizer regression is unrelated to pipeline task-support messaging." + "reason": "Processor/tokenizer regression and pipeline error-message regression are different issues." + }, + { + "left": "issue:43232", + "right": "issue:43388", + "accept": false, + "reason": "Generation kwargs update after sync_gpus and metric gathering last-batch label dropping are unrelated distributed-training bugs." } ] }, @@ -8936,8 +8969,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -8991,8 +9024,8 @@ "issue:45698" ], "soft_pairs": [ - "issue:45127|issue:45200", "issue:44871|issue:45584", + "issue:45127|issue:45200", "issue:43550|issue:43701", "issue:43606|issue:43701", "issue:43295|issue:44871", @@ -9001,242 +9034,242 @@ "issue:33290|issue:35141", "issue:33290|issue:36331", "issue:43452|issue:44496", - "issue:43673|issue:44743", "issue:43575|issue:43742", + "issue:43673|issue:44743", "issue:43646|issue:44913", - "issue:44936|issue:45698", "issue:42617|issue:43575", + "issue:44936|issue:45698", "issue:43761|issue:44242", "issue:43575|issue:43873", "issue:44077|issue:44534", "issue:43931|issue:44060", "issue:44464|issue:44871", "issue:44479|issue:44734", - "issue:43906|issue:45290", "issue:44617|issue:45584", + "issue:43906|issue:45290", "issue:44792|issue:44977", "issue:43065|issue:44610", "issue:43827|issue:43994", "issue:44534|issue:44913", - "issue:41720|issue:45245", "issue:44190|issue:44964", + "issue:41720|issue:45245", "issue:44792|issue:45600", "issue:44855|issue:45588" ], - "prepared_review_unit_hash": "18db44438cc98f47617e69d2dc5701a0eda09f813bc8cc575dfdc14c0dacde61", + "prepared_review_unit_hash": "8add0b8968d643f55cc1a8945bde92962636969ddc51e8c974b4742a4183b4bb", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36971, - "estimated_input_tokens": 9243, - "estimated_eval_tokens": 18742 + "serialized_chars": 36968, + "estimated_input_tokens": 9242, + "estimated_eval_tokens": 18740 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is not a duplicate set; it mixes unrelated bugs across loading, generation, trainer, quantization, and model-specific regressions. All suggested soft edges look like superficial theme matches rather than the same underlying issue.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "The set is a grab bag of unrelated model-loading, training, generation, and regression reports. None of the soft pairs look like the same underlying bug/change closely enough to merge.", + "confidence": 0.92, + "canonical_issue_reason": "issue:43646 is the broadest, most central umbrella here: it describes a generic custom-model initialization breakage in Transformers 5.0.0, which is more representative than the many narrower, model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "issue:43646 is the best representative issue to keep because it is generic, user-facing, and broad enough to subsume closely related custom-model init/load regressions.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:45127", - "right": "issue:45200", + "left": "issue:44871", + "right": "issue:45584", "accept": false, - "reason": "Both involve model/config behavior, but one is LoRA vocabulary merging on tied embeddings and the other is Gemma 4 multimodal token-type defaults; different code paths and failures." + "reason": "Different bugs: Gemma eos token config mismatch vs Whisper empty-transcription generation failure." }, { - "left": "issue:44871", - "right": "issue:45584", + "left": "issue:45127", + "right": "issue:45200", "accept": false, - "reason": "Both mention token/eos alignment, but one is a Gemma config inconsistency and the other is Whisper empty-transcription generation after special-token alignment; not the same bug." + "reason": "Both involve model behavior around tokens/embeddings, but one is LoRA merging collapse and the other is Gemma 4 multimodal token-type defaults." }, { "left": "issue:43550", "right": "issue:43701", "accept": false, - "reason": "Torch.compile SDPA failure in Bamba is unrelated to a resume_from_checkpoint key mismatch; no shared underlying code-path." + "reason": "Unrelated failures: torch.compile+SDPA on Bamba versus resume_from_checkpoint key mismatch." }, { "left": "issue:43606", "right": "issue:43701", "accept": false, - "reason": "CPU offload device-mismatch in bark-small is a runtime placement bug, while resume_from_checkpoint key mismatch is a checkpoint/key-loading issue." + "reason": "CPU offload device mismatch is a different problem from checkpoint key mapping." }, { "left": "issue:43295", "right": "issue:44871", "accept": false, - "reason": "v4.57.5 processor.tokenizer regression and Gemma eos_token_id inconsistency are both config-ish, but they are different regressions in different components." + "reason": "One is custom model/processor API regression; the other is eos_token_id config inconsistency." }, { "left": "issue:43065", "right": "issue:44617", "accept": false, - "reason": "Sam3PixelDecoder dummy Conv2d and Sam3Video CUDA OOM are both Sam3-related, but one is architecture/code correctness and the other is memory usage; not the same bug." + "reason": "Same family only at a high level (Sam3), but one is a dummy Conv2d concern and the other is a CUDA OOM in video inference." }, { "left": "issue:43761", "right": "issue:44479", "accept": false, - "reason": "CLIPVisionModel hidden_states regression and Qwen video-input regression affect different models and different forward paths." + "reason": "Different model paths and symptoms: CLIP hidden_states omission versus video-input regression in Qwen VL models." }, { "left": "issue:33290", "right": "issue:35141", "accept": false, - "reason": "Adafactor+DeepSpeed OOM and embedding resize/post_init reinitialization are unrelated training vs model-initialization problems." + "reason": "OOM in DeepSpeed Adafactor is unrelated to token-embedding reinitialization after resize." }, { "left": "issue:33290", "right": "issue:36331", "accept": false, - "reason": "OOM in DeepSpeed with Adafactor is unrelated to CustomTrainer.compute_loss receiving an unexpected keyword argument." + "reason": "Optimizer memory blow-up is not the same as a Trainer compute_loss signature TypeError." }, { "left": "issue:43452", "right": "issue:44496", "accept": false, - "reason": "Both are model-loading failures, but one is gguf_file breaking specific AutoTokenizer/AutoModel loading and the other is an unrecognized model/config.json problem; different failure modes." + "reason": "Tokenizer/model GGUF loading breakage is unrelated to an unrecognized custom model config error." }, { - "left": "issue:43673", - "right": "issue:44743", + "left": "issue:43575", + "right": "issue:43742", "accept": false, - "reason": "Both involve cache/state handling, but one is a missing GenerationMixin cache in chunked_prefill and the other is recurrent state reset in modular_qwen3_5; not mergeable as one fix." + "reason": "Both are loading failures, but one is TP memory pressure on a large model and the other is a key error on a small model." }, { - "left": "issue:43575", - "right": "issue:43742", + "left": "issue:43673", + "right": "issue:44743", "accept": false, - "reason": "Qwen2-57B-A14B-Instruct TP OOM and MobileLLM-125M key error are different loading failures with different root causes." + "reason": "Both mention cache/state handling, but one is missing GenerationMixin cache in chunked prefill and the other is Qwen3.5 recurrent state reset." }, { "left": "issue:43646", "right": "issue:44913", "accept": false, - "reason": "Custom model initialization breakage in Transformers 5.0.0 is not the same as GPTNeoXConfig rotary_pct not persisting on reload." + "reason": "Generic custom-model initialization breakage is not the same as GPTNeoX rotary_pct not persisting on reload." }, { - "left": "issue:44936", - "right": "issue:45698", + "left": "issue:42617", + "right": "issue:43575", "accept": false, - "reason": "trainer.evaluate() after train and wrong custom module loaded after save_pretrained are unrelated training-loop vs serialization/loading bugs." + "reason": "3d_parallel.py execution failure is not the same underlying issue as TP loading OOM." }, { - "left": "issue:42617", - "right": "issue:43575", + "left": "issue:44936", + "right": "issue:45698", "accept": false, - "reason": "3d_parallel.py execution failure and Qwen2 TP OOM are unrelated; only the broad theme of model scaling overlaps." + "reason": "Trainer evaluate-after-train failure differs from wrong custom module being loaded after save_pretrained." }, { "left": "issue:43761", "right": "issue:44242", "accept": false, - "reason": "CLIPVisionModel hidden_states regression is unrelated to load balancing loss not being added when output_router_logits=False." + "reason": "CLIPVisionModel hidden_states regression and load-balancing loss omission are separate code paths." }, { "left": "issue:43575", "right": "issue:43873", "accept": false, - "reason": "A TP loading OOM and quantization offloading not working are both memory-adjacent, but they are distinct problems in different mechanisms." + "reason": "OOM under tensor parallelism is different from offloading/quantization not behaving as expected." }, { "left": "issue:44077", "right": "issue:44534", "accept": false, - "reason": "patchtsmixer post_init allowance and non-persistent buffers filled with junk are both v5 initialization issues, but they are not the same concrete bug." + "reason": "PatchTSMixer post_init allowance and non-persistent buffer junk are distinct initialization regressions." }, { "left": "issue:43931", "right": "issue:44060", "accept": false, - "reason": "Weight-shape mismatch when loading Qwen3-VL and incorrect tied-weights warning in Qwen3-Next involve different model families and different config logic." + "reason": "Weight-shape mismatch while loading Qwen3-VL is unrelated to a bad tied-weights warning in Qwen3-Next." }, { "left": "issue:44464", "right": "issue:44871", "accept": false, - "reason": "Chunked generation inconsistency under compiled forward is unrelated to Gemma eos_token_id configuration." + "reason": "Chunked generation inconsistency with compiled forward is unrelated to Gemma eos token config." }, { "left": "issue:44479", "right": "issue:44734", "accept": false, - "reason": "Video-input regression for Qwen multimodal models and /v1/responses KV-cache continuation crash are separate runtime paths." + "reason": "Video input regression in Qwen VL models is different from KV-cache continuation indexing in serve /v1/responses." }, { - "left": "issue:43906", - "right": "issue:45290", + "left": "issue:44617", + "right": "issue:45584", "accept": false, - "reason": "An isolated reproduction of issue 38071 is too generic to match the specific apply_chat_template crash on assistant tool-call messages with no content." + "reason": "CUDA OOM in Sam3Video and Whisper empty-transcription failure are unrelated runtime issues." }, { - "left": "issue:44617", - "right": "issue:45584", + "left": "issue:43906", + "right": "issue:45290", "accept": false, - "reason": "Sam3Video CUDA OOM and Whisper empty transcription after align_special_tokens are unrelated model-specific failures." + "reason": "Isolated reproduction of issue 38071 is too vague and not clearly the same as tool-call chat template crashes." }, { "left": "issue:44792", "right": "issue:44977", "accept": false, - "reason": "A failing Janus image-generation test and Qwen3.5 flash-attention generation failure are different models and likely different regressions." + "reason": "Janus image-generation test failure and Qwen3.5 flash-attention generation bug are different models and failure modes." }, { "left": "issue:43065", "right": "issue:44610", "accept": false, - "reason": "Sam3PixelDecoder dummy layer and OmDet-Turbo processor producing 640px instead of 224px are different preprocessing/model-shape issues." + "reason": "Dummy Conv2d in Sam3PixelDecoder does not match OmDet-Turbo processor output-size mismatch." }, { "left": "issue:43827", "right": "issue:43994", "accept": false, - "reason": "Docs still referencing pipeline() and SigLIP2 producing nonsensical outputs with AutoModel/pipeline are not the same bug; one is documentation, the other is model behavior." + "reason": "Docs still mentioning pipeline() is not the same as nonsensical outputs from AutoModel/pipeline on SigLIP2." }, { "left": "issue:44534", "right": "issue:44913", "accept": false, - "reason": "Non-persistent buffer junk and rotary_pct not persisting on reload are both persistence-related, but they affect different configuration/state mechanisms." + "reason": "Non-persistent buffer initialization junk and GPTNeoX rotary_pct reload mismatch are unrelated." }, { - "left": "issue:41720", - "right": "issue:45245", + "left": "issue:44190", + "right": "issue:44964", "accept": false, - "reason": "Qwen3 auto device mapping cudaErrorAssert and category-count ceiling are unrelated runtime vs data-shape constraints." + "reason": "Local dataset loading in a no-trainer script is a different bug from loading Phi-4 multimodal-instruct." }, { - "left": "issue:44190", - "right": "issue:44964", + "left": "issue:41720", + "right": "issue:45245", "accept": false, - "reason": "Local dataset loading in a no-trainer example and loading Phi-4 multimodal instruct are separate loading failures." + "reason": "Qwen3 auto-device-map cudaErrorAssert is unrelated to the 2^24 category limit." }, { "left": "issue:44792", "right": "issue:45600", "accept": false, - "reason": "Janus image-generation test failure and stale auto_mappings references to removed Sam3LiteText configs are unrelated CI/model issues." + "reason": "Janus generate-images test failure is unrelated to stale auto_mappings entries for removed Sam3LiteText configs." }, { "left": "issue:44855", "right": "issue:45588", "accept": false, - "reason": "Python 3.13 torch.jit parsing/IndentationError in DebertaV2Model is unrelated to flash_attention crashing on s_aux=None." + "reason": "Python 3.13 TorchScript parsing failure in DebertaV2Model is unrelated to flash_attention s_aux=None crashes." } ] }, @@ -9255,8 +9288,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -9329,218 +9362,218 @@ "issue:45245|issue:45584", "issue:43479|issue:45375", "issue:36246|issue:41628", - "issue:44464|issue:45563", "issue:44933|issue:45362", "issue:44186|issue:44734", "issue:43404|issue:45216", + "issue:44464|issue:45563", "issue:44683|issue:45446", "issue:38175|issue:43065", "issue:44734|issue:45632", "issue:44561|issue:45632", "issue:44466|issue:44913" ], - "prepared_review_unit_hash": "bcd0a56d1874d66dd55bdda76bd36546446118e34e22fa8b5c36ef035f240540", + "prepared_review_unit_hash": "5222244dc94fbace13dcf65f79e2713a592886ca484687a11a521c21985ce478", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 35958, - "estimated_input_tokens": 8990, - "estimated_eval_tokens": 18236 + "serialized_chars": 35963, + "estimated_input_tokens": 8991, + "estimated_eval_tokens": 18238 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a grab-bag of unrelated Transformers issues, mostly import/config/regression bugs. I found no soft-edge pair that looks like the same underlying bug or change, so no duplicates should be merged here.", - "confidence": 0.93, - "canonical_issue_reason": "Selected issue 44521 as the best representative because it is a concrete, user-facing API regression in a common multimodal path and has the strongest discussion/reference signal among these otherwise unrelated issues.", + "summary": "This cluster is mostly noisy and heterogeneous: the soft-pair candidates do not appear to share the same underlying bug or change, so I reject all of them. The strongest issue in the set is the multimodal `apply_chat_template` bug because it is the most active, well-scoped, and well-referenced report.", + "confidence": 0.87, + "canonical_issue_reason": "issue:44521 is the most actionable and best-supported report in the set, with the highest discussion/inbound activity and a clear reproducible multimodal bug.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44521 is the most broadly useful anchor for the cluster: it has a clear repro, affects multimodal chat template behavior, and is more representative of a widely used code path than the more niche import/config bugs.", + "best_issue_reason": "issue:44521 is the best representative issue here because it is concrete, broadly useful, and has the strongest evidence trail among otherwise unrelated reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44737", "right": "issue:45803", "accept": false, - "reason": "Different bugs in different code paths: XLNet positional encoding device placement vs. Fuyu tensor conversion exception handling." + "reason": "Different subsystems and failure modes: XLNet CPU placement bug vs Fuyu tensor conversion exception handling." }, { "left": "issue:44521", "right": "issue:45381", "accept": false, - "reason": "Both are multimodal, but one breaks assistant masks in chat templates while the other is a Qwen2.5-VL video position-id issue; not the same defect." + "reason": "Both involve multimodal inputs, but one is assistant mask generation and the other is video position IDs in Qwen2.5-VL; root causes do not match." }, { "left": "issue:45127", "right": "issue:45357", "accept": false, - "reason": "Both involve save/weight handling, but they affect different model families and different serialization failures." + "reason": "Distinct regression areas: LoRA merge/tied embeddings vs incorrect visual-encoder keys in `save_pretrained`." }, { "left": "issue:43335", "right": "issue:44855", "accept": false, - "reason": "Unrelated: SwitchTransformers sparse-layer config bug vs. DebertaV2 import failure on Python 3.13." + "reason": "Unrelated models and bugs: SwitchTransformers config layer creation vs DebertaV2 import parsing on Python 3.13." }, { "left": "issue:44938", "right": "issue:45542", "accept": false, - "reason": "Python 3.14 import/load failure is not the same as a TensorBoard-without-TF backend error." + "reason": "Both are environment/load issues, but one is Python 3.14 import failure and the other is a missing TensorFlow backend fallback; not the same bug." }, { "left": "issue:43976", "right": "issue:44855", "accept": false, - "reason": "Both mention Python compatibility, but they are different breakages and different root causes." + "reason": "Different Python-version regressions with unrelated mechanisms and code paths." }, { "left": "issue:43550", "right": "issue:44190", "accept": false, - "reason": "Different subsystems entirely: torch.compile/SDPA model failure vs. local dataset loading in an example script." + "reason": "Completely different domains: torch.compile SDPA model failure vs dataset loading in an example script." }, { "left": "issue:43644", "right": "issue:44913", "accept": false, - "reason": "Non-persistent buffer initialization junk is unrelated to GPTNeoX config reload behavior." + "reason": "Both touch model state/config, but one is non-persistent buffer junk in v5 and the other is a GPTNeoX config reload issue." }, { "left": "issue:44077", "right": "issue:44877", "accept": false, - "reason": "Both touch configs, but one is a model-specific post_init policy issue and the other is strict config loading for granite_speech." + "reason": "Both mention config handling, but one is about forbidding optional `post_init` and the other is strict config loading for a specific model." }, { "left": "issue:44496", "right": "issue:45803", "accept": false, - "reason": "Unrecognized-model/config metadata error is unrelated to Fuyu's broad exception handling bug." + "reason": "Unrelated: model-type/config loading error vs a bare `except` swallowing errors in Fuyu tensor conversion." }, { "left": "issue:43540", "right": "issue:44734", "accept": false, - "reason": "Different multimodal/video-serving failures with different code paths and symptoms." + "reason": "Different code paths: Qwen3OmniMoe video preprocessing vs serve KV-cache continuation indexing." }, { "left": "issue:45276", "right": "issue:45584", "accept": false, - "reason": "Token-embedding resize propagation and Whisper empty-transcription handling are unrelated." + "reason": "Different bugs in different models: Gemma4 embedding resize propagation vs Whisper empty transcription handling." }, { "left": "issue:44112", "right": "issue:45600", "accept": false, - "reason": "Both affect CI, but they are distinct failures in different files and due to different causes." + "reason": "Both are CI-related, but one is a stale test override and the other is stale autogenerated mappings; not the same root cause." }, { "left": "issue:43650", "right": "issue:44220", "accept": false, - "reason": "No meaningful overlap: a placeholder issue vs. an fbank feature extraction bug." + "reason": "No shared underlying bug; the titles are unrelated and provide no common code path." }, { "left": "issue:43335", "right": "issue:44190", "accept": false, - "reason": "Different bug classes: SwitchTransformers config construction vs. image-classification example dataset loading." + "reason": "SwitchTransformers config initialization and example dataset loading are unrelated." }, { "left": "issue:43116", "right": "issue:44190", "accept": false, - "reason": "Both are example-script problems, but one is multi-label classification output logic and the other is local dataset loading." + "reason": "Different example-script failures: multi-label classification output bug vs local dataset loading." }, { "left": "issue:43295", "right": "issue:45276", "accept": false, - "reason": "Custom processor/tokenizer regression is unrelated to token embedding resize behavior." + "reason": "Different multimodal/model-management regressions; no shared code path or fixable overlap." }, { "left": "issue:45245", "right": "issue:45584", "accept": false, - "reason": "Category-count limit error is unrelated to Whisper post-processing on empty transcription." + "reason": "Completely unrelated error conditions: category cardinality limit vs Whisper transcription edge case." }, { "left": "issue:43479", "right": "issue:45375", "accept": false, - "reason": "Both are multimodal config issues, but they concern different models and different missing/default fields." + "reason": "Both are config-schema issues, but they affect different model families and different missing-field behavior." }, { "left": "issue:36246", "right": "issue:41628", "accept": false, - "reason": "Both are import errors, but for different symbols and different modules; not the same missing-export bug." - }, - { - "left": "issue:44464", - "right": "issue:45563", - "accept": false, - "reason": "Compiled-forward chunked generation inconsistency is unrelated to a stale warning about num_return_sequences." + "reason": "Both are import errors, but they refer to different symbols and likely different export/registry problems." }, { "left": "issue:44933", "right": "issue:45362", "accept": false, - "reason": "A missing import from image_utils is not the same as a Qwen3.5 chat crash." + "reason": "A missing import in `image_utils` is not the same as a chat crash in Qwen3.5." }, { "left": "issue:44186", "right": "issue:44734", "accept": false, - "reason": "Tokenizer padding/truncation crash and KV-cache continuation tensor indexing are different failures." + "reason": "Tokenizer padding/truncation crash vs server KV-cache continuation indexing; distinct bugs." }, { "left": "issue:43404", "right": "issue:45216", "accept": false, - "reason": "Both involve model weight handling, but one is lm_head tying in Mistral3 and the other is incorrect checkpoint saving in Qwen3.5." + "reason": "Both involve model weights, but one is tied `lm_head` in Mistral3 and the other is Qwen3.5 checkpoint saving." + }, + { + "left": "issue:44464", + "right": "issue:45563", + "accept": false, + "reason": "Generation inconsistency under compilation vs a stale warning about `num_return_sequences`; not the same issue." }, { "left": "issue:44683", "right": "issue:45446", "accept": false, - "reason": "Shared flex_attention area, but one is a compiled-runtime failure on newer torch while the other is a version-check/import gating bug." + "reason": "Both are in flex_attention, but one is a compiled-kernel failure on newer torch and the other is a version-check/import bug for AuxRequest." }, { "left": "issue:38175", "right": "issue:43065", "accept": false, - "reason": "Siglip2 zero probabilities and Sam3PixelDecoder dummy Conv2d are unrelated model-specific bugs." + "reason": "Different models and failure modes: zero probabilities in SigLIP2 vs a dummy Conv2d in Sam3 pixel decoder." }, { "left": "issue:44734", "right": "issue:45632", "accept": false, - "reason": "Serve KV-cache continuation indexing bug is unrelated to trust_remote_code cache path collisions." + "reason": "Serve KV-cache indexing and trust_remote_code cache path collisions are unrelated." }, { "left": "issue:44561", "right": "issue:45632", "accept": false, - "reason": "Both mention trust_remote_code, but one is a removed-helper compatibility break and the other is a cache-path collision." + "reason": "Different regression classes: removed helper breaking trust_remote_code vs cache-path collision for local models." }, { "left": "issue:44466", "right": "issue:44913", "accept": false, - "reason": "Serialized lm_head/tied-weight inconsistency is unrelated to GPTNeoX rotary_pct reload behavior." + "reason": "Serialization of tied weights depending on device vs GPTNeoX config reload behavior; not the same bug." } ] }, @@ -9559,8 +9592,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -9617,8 +9650,8 @@ "issue:44246|issue:45003", "issue:36246|issue:38617", "issue:43643|issue:44496", - "issue:45563|issue:45584", "issue:45406|issue:45632", + "issue:45563|issue:45584", "issue:41950|issue:44016", "issue:44964|issue:45600", "issue:43388|issue:43688", @@ -9626,8 +9659,8 @@ "issue:44112|issue:44964", "issue:43295|issue:45335", "issue:43854|issue:44964", - "issue:43673|issue:44336", "issue:44016|issue:44220", + "issue:43673|issue:44336", "issue:44496|issue:45042", "issue:43299|issue:45216", "issue:43531|issue:45600", @@ -9640,227 +9673,230 @@ "issue:43901|issue:44496", "issue:44496|issue:45759", "issue:44351|issue:45278", - "issue:44242|issue:45200", "issue:44016|issue:44062", + "issue:44242|issue:45200", "issue:43388|issue:43827", - "issue:43976|issue:45405", - "issue:41950|issue:45020" + "issue:41950|issue:45020", + "issue:43976|issue:45405" ], - "prepared_review_unit_hash": "f27d8f975a3b0c0b2e4bf7f1b34a4ce60d05a41780c656dfa62a905080b0527d", + "prepared_review_unit_hash": "26caec4c4b1c2a0d3ce2f4f1f8c327e109c9fd30d7f9e5f97cee67eb504c3d71", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37050, - "estimated_input_tokens": 9263, - "estimated_eval_tokens": 18782 + "serialized_chars": 37045, + "estimated_input_tokens": 9262, + "estimated_eval_tokens": 18780 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No duplicate cluster here: the issues span many unrelated regressions, import errors, model-loading problems, docs, and training/runtime bugs. None of the soft pairs look like the same underlying defect.", - "confidence": 0.97, - "canonical_issue_reason": "No single canonical issue is appropriate; the set is too heterogeneous to treat one report as representative of the rest.", + "summary": "This cluster is mostly heterogeneous, but it has one plausible duplicate axis around post-upgrade import/export regressions in Transformers v5. Most other pairs are different bugs in different code paths and should stay separate.", + "confidence": 0.58, + "canonical_issue_reason": "issue:45278 is the broadest representative report in the set: it describes widespread import errors after upgrading to 5.5.0 and can plausibly subsume concrete missing-symbol cases like issue:44351.", "canonical_pr_reason": null, - "best_issue_reason": "No strong best issue either; the broadest report is still not a true umbrella for the others, so merging them as duplicates would be misleading.", + "best_issue_reason": "issue:45278 is the best umbrella issue because it captures the shared upgrade/regression theme with the widest scope and open discussion, making it the most useful canonical target for related import failures.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44246", "right": "issue:45003", "accept": false, - "reason": "One is intermittent import slowness; the other is an unsafe sys.modules access bug. Different symptoms and code paths." + "reason": "Both are import-related, but one is a performance complaint about importing Transformers and the other is a specific sys.modules access bug; different symptoms and likely different fixes." }, { "left": "issue:36246", "right": "issue:38617", "accept": false, - "reason": "Both are ImportErrors, but they are for different missing symbols in different modules with no shared concrete fix." + "reason": "Both are ImportError reports, but they concern different missing symbols from different modules, so they are not the same underlying bug." }, { "left": "issue:43643", "right": "issue:44496", "accept": false, - "reason": "Remote-code config field loss and unrecognized model_type are different loading failures, not the same bug." + "reason": "Both involve config/model-loading failures, but one is about trust_remote_code returning incomplete configs and the other is an unrecognized model/config issue; different root causes." }, { - "left": "issue:45563", - "right": "issue:45584", + "left": "issue:45406", + "right": "issue:45632", "accept": false, - "reason": "Stale paging warning and Whisper empty-transcription failure are unrelated generation issues." + "reason": "Both mention trust_remote_code behavior, but one is a missing _tokenizer attribute on Gemma4Processor and the other is a cache-path collision for local models; unrelated bugs." }, { - "left": "issue:45406", - "right": "issue:45632", + "left": "issue:45563", + "right": "issue:45584", "accept": false, - "reason": "Processor _tokenizer attribute crash and trust_remote_code cache collision are separate problems." + "reason": "Both are generation-related, but a stale warning on paged generate and a Whisper empty-transcription failure after token alignment are different code paths." }, { "left": "issue:41950", "right": "issue:44016", "accept": false, - "reason": "Video pipeline processor lookup bug and a notebook syntax error are unrelated." + "reason": "A video-classification pipeline bug and a notebook syntax error are clearly unrelated." }, { "left": "issue:44964", "right": "issue:45600", "accept": false, - "reason": "Phi-4 multimodal loading and stale auto_mappings references to removed configs are different breakages." + "reason": "Both are model-loading/CI issues, but they affect different models and different failure modes; no shared concrete bug." }, { "left": "issue:43388", "right": "issue:43688", "accept": false, - "reason": "Batch label dropping in gather_for_metrics and auxiliary-loss normalization are unrelated training/eval bugs." + "reason": "These concern different training utilities: one drops labels in gather_for_metrics, the other misnormalizes auxiliary loss." }, { "left": "issue:42907", "right": "issue:43441", "accept": false, - "reason": "Saving dequantized models and FlashAttention loading failures in Ministral are distinct code paths." + "reason": "Both are about Ministral/Devstral families, but saving dequantized weights and FlashAttention model loading are distinct problems." }, { "left": "issue:44112", "right": "issue:44964", "accept": false, - "reason": "A stale GraniteSpeech CI test is not the same as a Phi-4 multimodal model-loading failure." + "reason": "A stale GraniteSpeech CI test and a Phi-4 multimodal loading failure are unrelated." }, { "left": "issue:43295", "right": "issue:45335", "accept": false, - "reason": "Processor/tokenizer regression and T5Gemma embedding resize behavior are different model API bugs." + "reason": "Both are regressions, but one is about processor.tokenizer/image handling in custom multimodal code and the other is resize_token_embeddings not updating decoder embeddings." }, { "left": "issue:43854", "right": "issue:44964", "accept": false, - "reason": "GLM-4.7-Flash unit-test loading and Phi-4 multimodal loading are separate model-specific failures." + "reason": "Both are model-load failures, but they involve different models and different root causes." }, { - "left": "issue:43673", - "right": "issue:44336", + "left": "issue:44016", + "right": "issue:44220", "accept": false, - "reason": "Generation cache missing and ANSI output formatting are unrelated subsystems." + "reason": "A notebook syntax error and an audio feature-extraction bug are unrelated." }, { - "left": "issue:44016", - "right": "issue:44220", + "left": "issue:43673", + "right": "issue:44336", "accept": false, - "reason": "Notebook syntax error and _torch_extract_fbank_features failure do not share a concrete defect." + "reason": "Chunked-prefill cache handling and ANSI coloring in loading_report are unrelated." }, { "left": "issue:44496", "right": "issue:45042", "accept": false, - "reason": "Unrecognized model config and PIL backend requiring torchvision are different loading/runtime issues." + "reason": "An unrecognized OLMo model config and a PIL/torchvision dependency regression are different issues." }, { "left": "issue:43299", "right": "issue:45216", "accept": false, - "reason": "Qwen3VL MoE model loading and Qwen3.5 save_pretrained checkpoint regression are separate bugs." + "reason": "Both involve Qwen3.5/Qwen3VL regressions, but one is a loading failure for MoE models and the other is an incorrect save_pretrained checkpoint layout; different code paths." }, { "left": "issue:43531", "right": "issue:45600", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior and removed Sam3LiteText config references are unrelated." + "reason": "Sliding_window behavior for Qwen3-MoE and removed auto_mappings references for Sam3LiteText are unrelated." }, { "left": "issue:43526", "right": "issue:43901", "accept": false, - "reason": "BeitImageProcessorFast label reduction and text-classification docs mismatch are not the same issue." + "reason": "One is an image-processor label reduction bug, the other is a docs/behavior mismatch in text classification; not the same bug." }, { "left": "issue:42175", "right": "issue:44279", "accept": false, - "reason": "A missing TensorFlow backend dependency and a generic dependency issue are too broad and not the same concrete failure." + "reason": "A missing TensorFlow backend in the torch extra and a generic dependency issue are too vague and not the same concrete failure." }, { "left": "issue:45335", "right": "issue:45584", "accept": false, - "reason": "Embedding resize behavior and Whisper empty transcription generation failure are unrelated." + "reason": "These are different functional areas: embedding resize behavior vs Whisper generation after token alignment." }, { "left": "issue:44075", "right": "issue:44368", "accept": false, - "reason": "Optimizer SGD args ignored and a tie_word_embeddings warning are different config/runtime concerns." + "reason": "Optimizer argument handling and a warning about tie_word_embeddings in Qwen fine-tuning are unrelated." }, { "left": "issue:43381", "right": "issue:45593", "accept": false, - "reason": "Gradient checkpointing in eval mode and D-FINE auxiliary-loss omission are unrelated behaviors." + "reason": "Gradient checkpointing in eval mode and missing auxiliary losses in D-FINE are different training bugs." }, { "left": "issue:43299", "right": "issue:45357", "accept": false, - "reason": "These are both Qwen3.5 regressions, but one is model loading and the other is save_pretrained key serialization; not one bug." + "reason": "Both are Qwen3.5/Qwen3VL regressions, but the reported failures are different: loading vs incorrect save_pretrained output." }, { "left": "issue:43901", "right": "issue:44496", "accept": false, - "reason": "Docs mentioning return_all_scores and an unrecognized model config error do not share a fix." + "reason": "TextClassificationPipeline docs/behavior and OLMo model loading have no shared underlying bug." }, { "left": "issue:44351", "right": "issue:45278", - "accept": false, - "reason": "A specific HybridCache import error is not the same concrete bug as a broad report of many import errors." + "accept": true, + "reason": "These look like the same upgrade-induced import/export regression: 45278 is a broad report of many import errors after moving to 5.5.0, and 44351 is a concrete missing-symbol instance (HybridCache)." }, { - "left": "issue:44242", - "right": "issue:45200", + "left": "issue:44016", + "right": "issue:44062", "accept": false, - "reason": "Dropped label elements in metrics gathering and Gemma 4 mm_token_type_ids defaults are unrelated." + "reason": "Notebook syntax and AddedToken keyword-argument errors are unrelated." }, { - "left": "issue:44016", - "right": "issue:44062", + "left": "issue:44242", + "right": "issue:45200", "accept": false, - "reason": "Notebook syntax error and AddedToken keyword conflict are unrelated." + "reason": "Load-balancing loss gating and Gemma 4 mm_token_type_ids defaults are different model/training issues." }, { "left": "issue:43388", "right": "issue:43827", "accept": false, - "reason": "Metrics label handling and docs still referencing pipeline() after removals are different issues." + "reason": "Metric label truncation and outdated docs references are unrelated." }, { - "left": "issue:43976", - "right": "issue:45405", + "left": "issue:41950", + "right": "issue:45020", "accept": false, - "reason": "Python version incompatibility and an unreleased PEFT minimum version bump are unrelated." + "reason": "A pipeline image-processor bug and broad remote_code breakage are both version regressions, but not the same concrete code-path problem." }, { - "left": "issue:41950", - "right": "issue:45020", + "left": "issue:43976", + "right": "issue:45405", "accept": false, - "reason": "Video-classification pipeline image-processor lookup and remote_code breakage are different bugs." + "reason": "Python version compatibility and an unreleased peft version bump are unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is grounded in the packet: it correctly treats the set as mostly heterogeneous while identifying a narrow import-regression axis centered on issue:45278. The one accepted soft pair, issue:44351 and issue:45278, is a conservative match on an upgrade-induced import failure theme. The other pairwise rejections are also suitably cautious and do not overclaim shared root causes." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -9872,8 +9908,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -9934,11 +9970,11 @@ "issue:43299|issue:43854", "issue:44683|issue:45542", "issue:44279|issue:45405", - "issue:39692|issue:43901", "issue:38175|issue:44617", + "issue:39692|issue:43901", "issue:32090|issue:35141", - "issue:44462|issue:45081", "issue:43295|issue:44779", + "issue:44462|issue:45081", "issue:44297|issue:44462", "issue:41950|issue:43650", "issue:43976|issue:44279", @@ -9948,8 +9984,8 @@ "issue:44521|issue:45325", "issue:43441|issue:45834", "issue:43450|issue:44279", - "issue:44261|issue:44263", "issue:42994|issue:43317", + "issue:44261|issue:44263", "issue:43065|issue:44898", "issue:44466|issue:45276", "issue:43650|issue:45290", @@ -9957,208 +9993,208 @@ "issue:43723|issue:44162", "issue:44623|issue:45397" ], - "prepared_review_unit_hash": "2bf4344e4347861c0358e0a8d2d62c372cce169323e8c84f1db7d0aa059c1845", + "prepared_review_unit_hash": "e91c99e1aadaa4b401e935290d8eb83732e108927e85e6ff954de4d07f8b55e7", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36327, - "estimated_input_tokens": 9082, - "estimated_eval_tokens": 18420 + "serialized_chars": 36323, + "estimated_input_tokens": 9081, + "estimated_eval_tokens": 18418 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are a mixed bag of mostly unrelated Transformers bugs across tokenizers, multimodal processing, loading/saving, and backend failures. None of the soft-similarity pairs looks like the same concrete bug or fix, so I reject all soft edges. Best canonical issue is the clearest, most self-contained tokenizer-loading regression.", - "confidence": 0.84, - "canonical_issue_reason": "#44462 is a well-scoped, reproducible core API bug with a clear failure mode in AutoTokenizer loading from the repo.", + "summary": "This cluster is mostly a loose collection of Transformers regression reports around tokenizers/processors, multimodal loading, and save/load behavior, but the soft pairs generally describe different bugs rather than true duplicates.", + "confidence": 0.66, + "canonical_issue_reason": "issue:44462 is the most representative tokenizer-loading regression here: it is concrete, user-facing, and sits near several other tokenizer-related reports without being overly narrow.", "canonical_pr_reason": null, - "best_issue_reason": "#44462 is the strongest representative issue here because it isolates one concrete code path and is better specified than the more model-specific or documentation-oriented reports.", + "best_issue_reason": "issue:44462 best captures the cluster\u2019s strongest recurring theme\u2014v5 tokenizer loading/selection regressions\u2014and has the clearest chance of serving as the umbrella issue for related tokenizer bugs.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43824", "right": "issue:44933", "accept": false, - "reason": "Different import failures in unrelated modules: Qwen2.5-VL class export vs a missing image_utils symbol." + "reason": "Both are import/load problems in the multimodal/image stack, but one is a missing Qwen2.5-VL symbol and the other is a nonexistent image_utils import; different concrete failures." }, { "left": "issue:43577", "right": "issue:45216", "accept": false, - "reason": "Both involve loading behavior, but one is dtype propagation in BLIP-2 and the other is Qwen3.5 checkpoint saving corruption." + "reason": "Both involve save/load serialization, but BLIP2 dtype handling and Qwen3.5 checkpoint corruption are different bugs in different code paths." }, { "left": "issue:43479", "right": "issue:43525", "accept": false, - "reason": "Different model config bugs: default multimodal sub-config initialization vs missing pad_token_id on Llama4Config." + "reason": "One is default multimodal config initialization from None, the other is a missing pad_token_id attribute; related configuration surface, not the same defect." }, { "left": "issue:41950", "right": "issue:45823", "accept": false, - "reason": "Unrelated multimodal areas: video-classification processor lookup vs Gemma4 device_map forward mismatch." + "reason": "Video-classification processor lookup and Gemma4 PLE device mismatch are unrelated failures despite both being pipeline/model issues." }, { "left": "issue:43299", "right": "issue:43854", "accept": false, - "reason": "Different model-loading regressions affecting different architectures and code paths." + "reason": "Both are model-loading failures, but they concern different models and different breakpoints." }, { "left": "issue:44683", "right": "issue:45542", "accept": false, - "reason": "Completely different failures: compiled flex_attention on torch 2.9+ vs tf backend detection without TensorFlow." + "reason": "Compiled flex_attention on torch>=2.9 and tensorboard/TensorFlow backend detection are unrelated." }, { "left": "issue:44279", "right": "issue:45405", "accept": false, - "reason": "General dependency complaint vs a specific PEFT version bump; not the same bug." + "reason": "A vague dependency issue and a specific PEFT version bump problem are not the same bug." }, { - "left": "issue:39692", - "right": "issue:43901", + "left": "issue:38175", + "right": "issue:44617", "accept": false, - "reason": "Docs example/model-processor mismatch and quantization errors are unrelated to a stale pipeline docs parameter mention." + "reason": "Unexpected zero probabilities in SigLIP2 and CUDA OOM in Sam3Video are clearly different problems." }, { - "left": "issue:38175", - "right": "issue:44617", + "left": "issue:39692", + "right": "issue:43901", "accept": false, - "reason": "Different symptoms and subsystems: zero probabilities in SigLIP2 vs Sam3Video OOM." + "reason": "SigLIP2 doc example errors and TextClassificationPipeline docs mismatch are documentation issues in different areas, not duplicates." }, { "left": "issue:32090", "right": "issue:35141", "accept": false, - "reason": "Trainer broadcast TypeError is unrelated to embedding resize causing post_init reinitialization." + "reason": "Trainer GPU broadcast of NoneType and embedding reinitialization after resize_token_embeddings are distinct model/training bugs." }, { - "left": "issue:44462", - "right": "issue:45081", + "left": "issue:43295", + "right": "issue:44779", "accept": false, - "reason": "Both touch tokenizers, but one is repo tokenizer discovery and the other is a Mistral regex patch crash." + "reason": "A regression in custom model code using processor.tokenizer/images and Deepseek tokenizer output corruption are different tokenizer-facing failures." }, { - "left": "issue:43295", - "right": "issue:44779", + "left": "issue:44462", + "right": "issue:45081", "accept": false, - "reason": "Different v5 tokenizer regressions with different APIs and models; not one concrete underlying bug." + "reason": "Ignoring tokenizer.json during AutoTokenizer loading is not the same as Mistral regex patch crashing on backend_tokenizer access." }, { "left": "issue:44297", "right": "issue:44462", "accept": false, - "reason": "Tokenizer class metadata on save_pretrained is distinct from AutoTokenizer ignoring tokenizer.json on load." + "reason": "Tokenizer class metadata written by save_pretrained and repository tokenizer.json selection are related serialization/loading topics but separate bugs." }, { "left": "issue:41950", "right": "issue:43650", "accept": false, - "reason": "Video-classification processor bug is unrelated to the placeholder 'ADD THE DATA' issue." + "reason": "Video-classification processor lookup and a nonspecific 'ADD THE DATA' issue are unrelated." }, { "left": "issue:43976", "right": "issue:44279", "accept": false, - "reason": "Python version compatibility failure is not the same as a generic dependency issue." + "reason": "Python version support for Transformers 5.1.0 and a dependency issue are not the same underlying defect." }, { "left": "issue:43976", "right": "issue:44987", "accept": false, - "reason": "Python-version support and model-load failure are different problems." + "reason": "Python compatibility and loading physical-intelligence/fast are different issues." }, { "left": "issue:45600", "right": "issue:45712", "accept": false, - "reason": "Both are CI cleanup issues, but they target different files and different defects." + "reason": "Broken auto_mappings references to Sam3LiteText configs and leftover dummy classes are different cleanup/CI problems." }, { "left": "issue:44462", "right": "issue:44779", "accept": false, - "reason": "Both tokenizer-related, but one is loading repo tokenizer assets and the other is incorrect Deepseek tokenization logic." + "reason": "AutoTokenizer ignoring tokenizer.json and Deepseek tokenizer producing wrong results are both tokenizer regressions, but they affect different models and mechanisms." }, { "left": "issue:44521", "right": "issue:45325", "accept": false, - "reason": "Multimodal chat masks and Qwen2.5-VL rope indexing are separate bugs in different code paths." + "reason": "Assistant masks being all-zero in multimodal chat templates is unrelated to Qwen2.5-VL rope index scaling." }, { "left": "issue:43441", "right": "issue:45834", "accept": false, - "reason": "Ministral FlashAttention failure and Kosmos2.5 OCR index error are unrelated." + "reason": "FlashAttention failure in Ministral-3 and Kosmos2.5 OCR long-input index error are different model-specific runtime bugs." }, { "left": "issue:43450", "right": "issue:44279", "accept": false, - "reason": "Video processor shape bug is unrelated to the dependency complaint." + "reason": "Incorrect batched video processor output shape and a generic dependency issue are unrelated." }, { - "left": "issue:44261", - "right": "issue:44263", + "left": "issue:42994", + "right": "issue:43317", "accept": false, - "reason": "Different model internals: MLA layernorm epsilon config vs torch.split return handling in GlmMoeDsaIndexer." + "reason": "Quantized model saving failure and dequantized model loading with device_map=auto are different quantization lifecycle bugs." }, { - "left": "issue:42994", - "right": "issue:43317", + "left": "issue:44261", + "right": "issue:44263", "accept": false, - "reason": "Quantized model saving and dequantized model offload loading are different lifecycle failures." + "reason": "Missing rms_norm_eps in MLA q_a_layernorm and torch.split return values in GlmMoeDsaIndexer are distinct math/indexing problems." }, { "left": "issue:43065", "right": "issue:44898", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder is unrelated to Perceiver non-default-resolution failure." + "reason": "Dummy Conv2d in Sam3PixelDecoder and Perceiver non-default-resolution classification failure are unrelated vision-model bugs." }, { "left": "issue:44466", "right": "issue:45276", "accept": false, - "reason": "Both mention weights, but one is device-dependent serialization and the other is resize_token_embeddings not updating Gemma4 outputs." + "reason": "Inconsistent lm_head.weight serialization and Gemma4 resize_token_embeddings not updating embeddings are both weight/resize issues, but they are different model families and code paths." }, { "left": "issue:43650", "right": "issue:45290", "accept": false, - "reason": "Placeholder issue has no relation to chat template crashes with empty tool-call assistant messages." + "reason": "A placeholder issue and chat template crashes with tool-call assistant messages are not duplicates." }, { "left": "issue:43441", "right": "issue:44683", "accept": false, - "reason": "FlashAttention-specific model failure is not the same as a torch-version compiled flex_attention failure." + "reason": "Ministral FlashAttention failure and compiled flex_attention failing on torch>=2.9 are both attention-related but clearly separate failure modes." }, { "left": "issue:43723", "right": "issue:44162", "accept": false, - "reason": "Tokenizer loading in v5 and ESM2 model breakage are separate regressions." + "reason": "Tokenizer loading regression in v5 and ESM2 broken workflow are unrelated." }, { "left": "issue:44623", "right": "issue:45397", "accept": false, - "reason": "Missing processor save files and Gemma4 zero3 from_pretrained are different save/load bugs." + "reason": "Missing processor.save_pretrained files and Gemma4 zero3 from_pretrained are different save/load issues." } ] }, @@ -10177,8 +10213,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -10233,8 +10269,8 @@ "soft_pairs": [ "issue:44623|issue:44749", "issue:43653|issue:45593", - "issue:43784|issue:45542", "issue:44671|issue:44779", + "issue:43784|issue:45542", "issue:44016|issue:45823", "issue:44361|issue:45593", "issue:44987|issue:45542", @@ -10242,220 +10278,220 @@ "issue:43122|issue:43650", "issue:44016|issue:44291", "issue:43262|issue:44521", - "issue:44188|issue:44336", "issue:43673|issue:43761", + "issue:44188|issue:44336", "issue:43525|issue:43881", "issue:44297|issue:45397", - "issue:43525|issue:43643", "issue:43450|issue:44617", + "issue:43525|issue:43643", "issue:42491|issue:43299", "issue:44589|issue:45362", "issue:41628|issue:43976", "issue:43901|issue:44246", "issue:44336|issue:44373", "issue:44242|issue:45593", - "issue:43352|issue:45250", "issue:16998|issue:22355", + "issue:43352|issue:45250", "issue:44295|issue:45397", "issue:44246|issue:44496", "issue:41950|issue:44220" ], - "prepared_review_unit_hash": "fa53bef158986be33eb12bbf2e87691f694b80c6a51df29d435538d71f03dafa", + "prepared_review_unit_hash": "d0520b907a9db231559c86ca2e314ba0215eb2d8ea5319e55ab533e30ddd6a50", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 35147, - "estimated_input_tokens": 8787, - "estimated_eval_tokens": 17830 + "serialized_chars": 35153, + "estimated_input_tokens": 8789, + "estimated_eval_tokens": 17834 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly noise: the soft-linked issues span unrelated subsystems (tokenizers, model loading, processors, docs, and performance), so almost none look like true duplicates.", - "confidence": 0.82, - "canonical_issue_reason": "issue:44521 is the best representative because it is the most active open issue in the processor/chat_template area and has the strongest inbound interest among the set.", + "summary": "This cluster is mostly a grab bag of unrelated Transformers issues across tokenization, model loading, processors, docs, and performance. None of the proposed soft pairs look like the same underlying bug/change; they are at best loosely similar by subsystem or version.", + "confidence": 0.92, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 is the most suitable global issue target for this cluster, but only as a representative theme, not because the linked items are actually the same bug.", + "best_issue_reason": "issue:44521 is the strongest standalone report: it describes a concrete, reproducible multimodal chat-template regression with broader user impact than the more niche or vague reports here.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44623", "right": "issue:44749", "accept": false, - "reason": "One is a processor save_pretrained file-generation bug; the other is a data-filtering slowdown regression. Different symptoms and code paths." + "reason": "Processor save/load files regression vs a tokenization slowdown report; different symptoms and code paths." }, { "left": "issue:43653", "right": "issue:45593", "accept": false, - "reason": "BigBirdTokenizer special-token decode corruption is unrelated to D-FINE auxiliary-loss handling." + "reason": "BigBirdTokenizer special-token registration bug vs D-FINE auxiliary-loss behavior; unrelated models and failures." }, { - "left": "issue:43784", - "right": "issue:45542", + "left": "issue:44671", + "right": "issue:44779", "accept": false, - "reason": "Import-time NameError in sentence-transformers integration does not match a TensorFlow/tensorboard backend selection bug." + "reason": "Different tokenizer/model correctness bugs affecting different model families; not the same underlying defect." }, { - "left": "issue:44671", - "right": "issue:44779", + "left": "issue:43784", + "right": "issue:45542", "accept": false, - "reason": "Both affect outputs, but one is CamemBERT MLM prediction quality and the other is Deepseek tokenization correctness; different models and root causes." + "reason": "Import-time NameError in sentence-transformers integration vs TensorFlow backend detection issue; unrelated." }, { "left": "issue:44016", "right": "issue:45823", "accept": false, - "reason": "Notebook syntax error is unrelated to a Gemma4 device-mismatch forward failure." + "reason": "Notebook syntax error vs Gemma4 device-mismatch runtime bug; completely different problem types." }, { "left": "issue:44361", "right": "issue:45593", "accept": false, - "reason": "MLukeTokenizer task AttributeError and D-FINE auxiliary-loss omission are unrelated." + "reason": "MLukeTokenizer task AttributeError vs D-FINE auxiliary-loss omission; unrelated subsystems." }, { "left": "issue:44987", "right": "issue:45542", "accept": false, - "reason": "Model loading regression on transformers>=5.1.0 is not the same as a missing/undefined tf backend error." + "reason": "Model loading failure for a specific checkpoint vs backend-import error; no shared code-path evidence." }, { "left": "issue:43901", "right": "issue:43994", "accept": false, - "reason": "Documentation mismatch about return_all_scores is not the same as an incorrect model/pipeline result bug." + "reason": "Docs mismatch about return_all_scores vs a model inference quality bug; not the same issue." }, { "left": "issue:43122", "right": "issue:43650", "accept": false, - "reason": "Tokenization behavior change and a vague \"ADD THE DATA\" report do not describe the same concrete defect." + "reason": "Tokenizer output changed across versions vs a nonsensical placeholder issue title; no technical overlap." }, { "left": "issue:44016", "right": "issue:44291", "accept": false, - "reason": "Notebook syntax error and init_empty_weights argument handling are unrelated." + "reason": "Notebook syntax error vs init_empty_weights argument mismatch; unrelated." }, { "left": "issue:43262", "right": "issue:44521", "accept": false, - "reason": "Both touch apply_chat_template, but one is an audio sampling-rate default and the other is multimodal assistant_masks; same API surface, different bug." - }, - { - "left": "issue:44188", - "right": "issue:44336", - "accept": false, - "reason": "torch.compile attention-kernel divergence is unrelated to ANSI codes emitted by loading_report." + "reason": "Both touch chat/processor behavior, but one is audio sampling-rate defaulting and the other is multimodal assistant-mask generation; different bugs." }, { "left": "issue:43673", "right": "issue:43761", "accept": false, - "reason": "Generation cache missing in chunked_prefill and CLIPVision hidden_states handling are different model-path regressions." + "reason": "Generation cache/chunked_prefill regression vs CLIPVision hidden_states handling; unrelated models and code paths." + }, + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "torch.compile attention-kernel divergence vs ANSI escaping in loading_report; unrelated." }, { "left": "issue:43525", "right": "issue:43881", "accept": false, - "reason": "Missing pad_token_id on Llama4Config and a glm-4v-9b loading failure are separate model-specific issues." + "reason": "Missing config attribute on Llama4Config vs GLM-4V loading failure; different model-specific loading issues." }, { "left": "issue:44297", "right": "issue:45397", "accept": false, - "reason": "Tokenizer save_pretrained metadata mismatch is unrelated to Gemma-4 zero3 loading." + "reason": "Tokenizer save_pretrained metadata mismatch vs Gemma-4 zero3 from_pretrained failure; unrelated." }, { - "left": "issue:43525", - "right": "issue:43643", + "left": "issue:43450", + "right": "issue:44617", "accept": false, - "reason": "A missing config attribute and trust_remote_code dropping fields are different config-loading bugs." + "reason": "Incorrect batched video-processor shape vs CUDA OOM; one is a shape bug, the other is resource exhaustion." }, { - "left": "issue:43450", - "right": "issue:44617", + "left": "issue:43525", + "right": "issue:43643", "accept": false, - "reason": "Incorrect batched video-processor shape and CUDA OOM are not the same failure." + "reason": "Missing pad_token_id on one config vs missing fields when trust_remote_code=True; both are config-related but not the same bug." }, { "left": "issue:42491", "right": "issue:43299", "accept": false, - "reason": "Both mention Qwen/MoE on v5, but one is LoRA reuse incompatibility and the other is model loading failure; too different to merge." + "reason": "Both involve Qwen3 MoE loading, but one is a LoRA compatibility break and the other is base-model loading in v5; too different to merge." }, { "left": "issue:44589", "right": "issue:45362", "accept": false, - "reason": "Float8 storage lookup failure is unrelated to Qwen3.5 chat crashes." + "reason": "Float8 storage deserialization error vs Qwen3.5 chat crash; no shared failure mode." }, { "left": "issue:41628", "right": "issue:43976", "accept": false, - "reason": "Missing AutoImageProcessor import and a Python-version compatibility regression are different issues." + "reason": "Missing AutoImageProcessor export vs Python-version compatibility regression; unrelated." }, { "left": "issue:43901", "right": "issue:44246", "accept": false, - "reason": "A docs mismatch about return_all_scores is unrelated to intermittent import slowness." + "reason": "Documentation mismatch vs slow import performance; different issue classes." }, { "left": "issue:44336", "right": "issue:44373", "accept": false, - "reason": "ANSI-color output in loading_report and a wrong docstring for position_ids are separate documentation/UX issues." + "reason": "ANSI codes in loading_report vs wrong position_ids docstring; unrelated docs/utility issues." }, { "left": "issue:44242", "right": "issue:45593", "accept": false, - "reason": "Both involve losses, but one is MoE load-balancing loss gating and the other is D-FINE auxiliary-loss omission." + "reason": "Load-balancing loss gating bug vs D-FINE auxiliary-loss omission; both are loss-related but not the same architecture or code path." }, { - "left": "issue:43352", - "right": "issue:45250", + "left": "issue:16998", + "right": "issue:22355", "accept": false, - "reason": "A lack of Flash Attention 2 support and a generic Flash Attention issue are too broad to be the same concrete bug." + "reason": "Model_max_length question for DeBERTa vs missing transformers.onnx module import; unrelated." }, { - "left": "issue:16998", - "right": "issue:22355", + "left": "issue:43352", + "right": "issue:45250", "accept": false, - "reason": "model_max_length confusion on DeBERTa-V3 is unrelated to the missing transformers.onnx module." + "reason": "Specific model not supporting Flash Attention 2.0 yet vs a generic Flash Attention 2.0 issue; not the same concrete bug." }, { "left": "issue:44295", "right": "issue:45397", "accept": false, - "reason": "position_ids buffer reading error and Gemma-4 zero3 loading are different failures." + "reason": "position_ids buffer access error vs Gemma-4 zero3 loading failure; unrelated." }, { "left": "issue:44246", "right": "issue:44496", "accept": false, - "reason": "Import slowdown and an unrecognized-model/config model_type error are unrelated." + "reason": "Import performance complaint vs unrecognized model/config missing model_type; different failure modes." }, { "left": "issue:41950", "right": "issue:44220", "accept": false, - "reason": "Video-classification image-processor lookup and _torch_extract_fbank_features() are different subsystems (video vs audio)." + "reason": "Video-classification pipeline using image processors vs _torch_extract_fbank_features() bug; both are pipeline/helper issues but not the same defect." } ] }, @@ -10474,14 +10510,13 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ "issue:36246", "issue:42757", - "issue:42886", "issue:43065", "issue:43232", "issue:43257", @@ -10501,6 +10536,7 @@ "issue:44295", "issue:44336", "issue:44351", + "issue:44466", "issue:44485", "issue:44496", "issue:44561", @@ -10515,8 +10551,8 @@ "issue:45003", "issue:45137", "issue:45200", - "issue:45229", "issue:45292", + "issue:45335", "issue:45362", "issue:45375", "issue:45459", @@ -10534,8 +10570,8 @@ "issue:43673|issue:44561", "issue:43824|issue:43976", "issue:43976|issue:44964", - "issue:44242|issue:44857", "issue:44246|issue:44871", + "issue:44242|issue:44857", "issue:45362|issue:45375", "issue:44162|issue:45200", "issue:44295|issue:44749", @@ -10543,8 +10579,8 @@ "issue:44877|issue:45375", "issue:45292|issue:45584", "issue:45292|issue:45701", - "issue:45561|issue:45563", "issue:43901|issue:45003", + "issue:45561|issue:45563", "issue:44987|issue:45459", "issue:36246|issue:43596", "issue:43232|issue:44336", @@ -10556,217 +10592,217 @@ "issue:44995|issue:45200", "issue:43262|issue:44246", "issue:43475|issue:43540", - "issue:42886|issue:45229", + "issue:44466|issue:45335", "issue:44995|issue:45588" ], - "prepared_review_unit_hash": "2111255e8cfdd61ea0f1eeb367652ca0b3b8c070da1bb66989f5f72fc79e4a6a", + "prepared_review_unit_hash": "c6642a9653d9ee3307a10ba8b671e78efa09ee5e5fe99163974bc31e10133f7d", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36095, - "estimated_input_tokens": 9024, - "estimated_eval_tokens": 18304 + "serialized_chars": 36195, + "estimated_input_tokens": 9049, + "estimated_eval_tokens": 18354 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a mix of unrelated Transformers bugs that share superficial similarity around imports, configs, generation, and model-loading regressions. No soft pair looks like a true duplicate fix path, so I rejected all edges. Among the issues, the clearest representative is the Qwen2.5-VL import failure because it has the strongest cross-reference signal and a concrete user-facing regression.", - "confidence": 0.86, - "canonical_issue_reason": "issue:43824 is the strongest representative of the cluster: it is a concrete, user-visible import regression, has the most inbound references, and the title/body pattern is central to several nearby import-failure reports.", + "summary": "This cluster is heterogeneous: the paired issues share at most superficial wording (imports, generation, config, embeddings, or model-family names) but describe different failures and code paths. I would not merge any of the soft pairs as duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "No single issue is a true duplicate hub for this cluster; if one must be selected, issue 45292 is the closest generic core-library bug because it concerns a reusable modeling utility regression and has the broadest overlap with nearby embedding-related reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43824 is the best overall issue to anchor the cluster because it is the most broadly referenced and most actionable concrete bug in the set.", + "best_issue_reason": "Issue 45292 is the best representative issue only in a loose sense: it is a concrete framework-level bug with a reusable API impact, whereas most other items are model-specific or unrelated. Still, the cluster overall is too mixed for a strong canonical choice.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36246", "right": "issue:42757", "accept": false, - "reason": "Different missing symbols from different packages/modules; same broad symptom, but not the same underlying bug." + "reason": "Both are ImportError reports, but for different missing symbols in different packages; unrelated failures." }, { "left": "issue:36246", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for unrelated classes/symbols and different failure paths." + "reason": "Qwen2.5-VL image processor import failure is unrelated to missing HybridCache export." }, { "left": "issue:44336", "right": "issue:44485", "accept": false, - "reason": "ANSI/color output bug in loading_report vs a GLM-5 RoPE modeling discussion; unrelated." + "reason": "ANSI color output in loading_report and GLM-5 RoPE implementation are unrelated subsystems and bugs." }, { "left": "issue:43673", "right": "issue:44561", "accept": false, - "reason": "Both are v5 regressions, but one is generation cache/chunked_prefill and the other is a removed utility breaking trust_remote_code; different code paths." + "reason": "Chunked-prefill cache handling and removal of is_torch_fx_available are different generation/runtime issues." }, { "left": "issue:43824", "right": "issue:43976", "accept": false, - "reason": "Specific Qwen2.5-VL import failure vs a Python/version compatibility problem; not the same bug." + "reason": "Missing Qwen2_5_VLForConditionalGeneration export is unrelated to Python version compatibility." }, { "left": "issue:43976", "right": "issue:44964", "accept": false, - "reason": "Both mention loading/version issues, but one is Python support and the other is a multimodal model load failure." + "reason": "Python 3.9/3.10 support mismatch is unrelated to loading Phi-4 multimodal." }, { - "left": "issue:44242", - "right": "issue:44857", + "left": "issue:44246", + "right": "issue:44871", "accept": false, - "reason": "Different model/loss bugs: router-loss gating vs AMP/CUDA crash in LwDetrImageLoss." + "reason": "Slow imports and inconsistent eos_token_id configuration are different problems." }, { - "left": "issue:44246", - "right": "issue:44871", + "left": "issue:44242", + "right": "issue:44857", "accept": false, - "reason": "Import latency bug vs Gemma EOS-token config inconsistency; unrelated." + "reason": "MoE load-balancing loss logic and LwDetrImageLoss AMP crash do not share the same code path." }, { "left": "issue:45362", "right": "issue:45375", "accept": false, - "reason": "Same Qwen family, but different concrete problems: chat crash vs a missing config field being dropped." + "reason": "Both mention Qwen3.5, but one is a chat crash and the other is a config field being dropped; not the same bug." }, { "left": "issue:44162", "right": "issue:45200", "accept": false, - "reason": "ESM2 breakage and Gemma4 mm_token_type_ids behavior are distinct model-specific issues." + "reason": "ESM2 breakage and Gemma-4 mm_token_type_ids defaulting are unrelated model-specific issues." }, { "left": "issue:44295", "right": "issue:44749", "accept": false, - "reason": "One is a position_ids buffer access error; the other is a tokenizer/performance regression across versions." + "reason": "Position_ids buffer access error and preprocessing slowdown after upgrade are different failures." }, { "left": "issue:44779", "right": "issue:45292", "accept": false, - "reason": "Tokenizer output regression and resize_token_embeddings not updating output embeddings are different bugs." + "reason": "Tokenizer output changes and resize_token_embeddings not updating output embeddings are distinct regressions." }, { "left": "issue:44877", "right": "issue:45375", "accept": false, - "reason": "Both involve config strictness, but they affect different model configs and different failure modes." + "reason": "Strict config rejection for granite_speech is unrelated to the Qwen3_5MoeVisionConfig field-drop bug." }, { "left": "issue:45292", "right": "issue:45584", "accept": false, - "reason": "Embedding resize bug vs Whisper empty-transcription generation failure; not the same change." + "reason": "Embedding resize propagation and Whisper empty-transcription generation failure are not the same underlying bug." }, { "left": "issue:45292", "right": "issue:45701", "accept": false, - "reason": "Resize-token-embeddings behavior and tokenization changes are separate concerns." + "reason": "resize_token_embeddings behavior and general tokenization changes across versions are not the same code-path problem." }, { - "left": "issue:45561", - "right": "issue:45563", + "left": "issue:43901", + "right": "issue:45003", "accept": false, - "reason": "Test-worker file race vs a stale generation warning; unrelated." + "reason": "Pipeline docs mismatch and unsafe sys.modules access are unrelated." }, { - "left": "issue:43901", - "right": "issue:45003", + "left": "issue:45561", + "right": "issue:45563", "accept": false, - "reason": "Docs mismatch for a pipeline vs unsafe sys.modules access in modeling_utils; different issues entirely." + "reason": "xdist race on captured_info.txt and stale warning in paged generate are separate testing vs generation issues." }, { "left": "issue:44987", "right": "issue:45459", "accept": false, - "reason": "Model load failure on transformers>=5.1.0 vs tokenizer error masking when protobuf is absent; different root causes." + "reason": "Model loading failure with transformers>=5.1.0 and tokenizer error masking without protobuf are different bugs." }, { "left": "issue:36246", "right": "issue:43596", "accept": false, - "reason": "Qwen import failure vs DeepSpeed ZeRO3/BertModel initialization crash; not the same bug." + "reason": "Qwen2.5-VL import error is unrelated to the DeepSpeed ZeRO3 BertModel index error." }, { "left": "issue:43232", "right": "issue:44336", "accept": false, - "reason": "Generation kwargs/update bug vs ANSI escape output problem; unrelated." + "reason": "Generation kwargs update after sync_gpus and ANSI codes in loading_report are unrelated." }, { "left": "issue:43526", "right": "issue:45003", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug vs sys.modules import handling; different code paths." + "reason": "BeitImageProcessorFast label reduction bug is unrelated to sys.modules access in modeling_utils." }, { "left": "issue:45375", "right": "issue:45759", "accept": false, - "reason": "Both concern Qwen multimodal config handling, but one is a missing field dropped by strict config parsing and the other is text_config unwrapping in from_config; not the same fix." + "reason": "Both are about Qwen multimodal configs, but one drops a new field under strict config and the other fails to unwrap text_config from composite configs." }, { "left": "issue:44496", "right": "issue:45003", "accept": false, - "reason": "Unrecognized model/config.json problem vs unsafe sys.modules access; unrelated." + "reason": "Unrecognized model/config loading and unsafe sys.modules access are different load-time issues." }, { "left": "issue:43065", "right": "issue:43408", "accept": false, - "reason": "Same SAM3 area, but one is a dummy Conv2d design choice and the other is a model-type warning; not the same underlying bug." + "reason": "Dummy conv in Sam3PixelDecoder and a sam3_video/sam3_tracker warning are not the same defect." }, { "left": "issue:43257", "right": "issue:45137", "accept": false, - "reason": "Different DeepSpeed-related failures: Qwen3 MOE weight conversion vs ZeRO3 deque underflow." + "reason": "Qwen3 MOE weight conversion under accelerate+deepspeed and ZeRO3 deque underflow are different DeepSpeed problems." }, { "left": "issue:44995", "right": "issue:45200", "accept": false, - "reason": "GlmMoeDsa stale cache crash and Gemma4 token-type defaults are unrelated." + "reason": "GlmMoeDsa stale indexer cache crash and Gemma-4 token-id defaults are unrelated." }, { "left": "issue:43262", "right": "issue:44246", "accept": false, - "reason": "Audio chat-template sampling-rate default bug vs import-time performance issue; unrelated." + "reason": "Audio chat-template sampling rate behavior and import slowness are unrelated." }, { "left": "issue:43475", "right": "issue:43540", "accept": false, - "reason": "Different multimodal model failures: missing vision encoder output attribute vs Qwen3OmniMoe video-input ValueError." + "reason": "Missing fpn_position_embeddings in SAM3 video output and Qwen3OmniMoe video-input ValueError are different bugs." }, { - "left": "issue:42886", - "right": "issue:45229", + "left": "issue:44466", + "right": "issue:45335", "accept": false, - "reason": "Offline cache/tokenizer loading bug vs multi-GPU CUDA OOM; not the same issue." + "reason": "Inconsistent serialization of tied weights and resize_token_embeddings not reaching decoder.embed_tokens are different weight-handling issues." }, { "left": "issue:44995", "right": "issue:45588", "accept": false, - "reason": "Stale indexer cache crash in GlmMoeDsa vs flash_attention AttributeError for sink-less models; different bugs." + "reason": "Stale indexer cache in GlmMoeDsa and flash_attention AttributeError on s_aux=None are different runtime paths." } ] }, @@ -10785,8 +10821,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -10794,6 +10830,7 @@ "issue:30990", "issue:41628", "issue:42371", + "issue:42886", "issue:42994", "issue:43232", "issue:43452", @@ -10813,7 +10850,6 @@ "issue:44297", "issue:44361", "issue:44368", - "issue:44466", "issue:44623", "issue:44734", "issue:44749", @@ -10825,9 +10861,9 @@ "issue:45092", "issue:45103", "issue:45137", + "issue:45229", "issue:45230", "issue:45310", - "issue:45335", "issue:45356", "issue:45375", "issue:45381", @@ -10840,7 +10876,7 @@ "issue:45656" ], "soft_pairs": [ - "issue:44466|issue:45335", + "issue:42886|issue:45229", "issue:43653|issue:44162", "issue:44368|issue:45563", "issue:44368|issue:44977", @@ -10870,233 +10906,236 @@ "issue:44038|issue:44623", "issue:41628|issue:44933", "issue:45081|issue:45103", - "issue:42994|issue:43525", - "issue:45375|issue:45478" + "issue:45375|issue:45478", + "issue:42994|issue:43525" ], - "prepared_review_unit_hash": "74a998a8e668aaddb40e7143cbdd829488f5ffc5e62997087cab130c6f72e06b", + "prepared_review_unit_hash": "852c1418e25fa469c9f2c60fe559407a6afeb135920adbe61dd15aa30fe64976", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37312, - "estimated_input_tokens": 9328, - "estimated_eval_tokens": 18912 + "serialized_chars": 37218, + "estimated_input_tokens": 9305, + "estimated_eval_tokens": 18866 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous false-positive similarity: many issues are in the same project/version band but describe different bugs. The only pair that looks plausibly like the same underlying regression is the Qwen3.5-Moe from_pretrained/config field case.", - "confidence": 0.58, - "canonical_issue_reason": "issue:45310 is the cleanest representative of the Qwen3.5-Moe from_pretrained regression; it matches the later same-title report and is the most direct anchor for the only plausible duplicate mini-cluster.", + "summary": "This cluster is heterogeneous: most items are distinct Transformers regressions, model-specific loading bugs, or docs/tests issues rather than duplicates. No soft edge looks like the same underlying bug/change.", + "confidence": 0.71, + "canonical_issue_reason": "issue:43646 is the broadest cross-cutting regression here: Transformers 5.0 breaking custom model initialization. It is the best anchor among otherwise unrelated, model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45310 is the best issue candidate because it is an exact, concrete regression report rather than a broad symptom summary, and it best represents the one likely duplicate topic in this set.", + "best_issue_reason": "issue:43646 is the most representative and general issue in the set, with the widest apparent impact on downstream custom models.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44466", - "right": "issue:45335", + "left": "issue:42886", + "right": "issue:45229", "accept": false, - "reason": "Both touch tied embeddings, but one is serialization/device-dependent while the other is resize_token_embeddings behavior; different code paths." + "reason": "Offline tokenizer cache failure and Gemma4 multi-GPU OOM are unrelated bugs." }, { "left": "issue:43653", "right": "issue:44162", "accept": false, - "reason": "Different models and failures: BigBirdTokenizer special-token registration vs ESM2 breakage." + "reason": "BigBirdTokenizer special-token decode bug vs. an ESM2 breakage; different code paths and failures." }, { "left": "issue:44368", "right": "issue:45563", "accept": false, - "reason": "Both are warnings, but one is about tie_word_embeddings and the other is a stale num_return_sequences warning; not the same bug." + "reason": "Both mention warnings, but one is a Qwen3.5 LoRA tie-weights warning and the other is a stale generate warning for num_return_sequences." }, { "left": "issue:44368", "right": "issue:44977", "accept": false, - "reason": "Different symptoms and subsystems: tied-weights warning vs flash-attention generation failure." + "reason": "Qwen3.5 tie-weights warning and flash-attention generation failure are distinct issues." }, { "left": "issue:45397", "right": "issue:45656", "accept": false, - "reason": "Gemma-4 Zero3 loading and DeepSpeed optimizer double-step are unrelated runtime issues." + "reason": "Gemma-4 ZeRO3 from_pretrained failure and a deepspeed double-step bug are not the same bug." }, { "left": "issue:44060", "right": "issue:44077", "accept": false, - "reason": "Different model/config validation bugs; no shared failing code path is evident." + "reason": "Qwen3-Next tied-weights warning and patchtsmixer post_init compatibility are unrelated." }, { "left": "issue:45003", "right": "issue:45561", "accept": false, - "reason": "sys.modules access in modeling_utils and xdist captured_info.txt races are unrelated." + "reason": "Unsafe sys.modules access and pytest-xdist race on captured_info.txt are different failures." }, { "left": "issue:43673", "right": "issue:44734", "accept": false, - "reason": "Both involve generation/cache, but one is chunked_prefill cache handling and the other is serve KV-cache continuation indexing; not the same bug." + "reason": "Both are generation/cache-related, but one is a chunked_prefill cache regression and the other is a serve KV-cache indexing crash." }, { "left": "issue:45137", "right": "issue:45600", "accept": false, - "reason": "DeepSpeed deque underflow and auto_mappings references to removed configs are unrelated." + "reason": "DeepSpeed ZeRO3 deque pop error and a stale auto_mappings reference are unrelated." }, { "left": "issue:43931", "right": "issue:44368", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch is unrelated to the tied-embeddings warning issue." + "reason": "Qwen3-VL weight-shape mismatch and Qwen3.5 tie-word-embeddings warning are different bugs." }, { "left": "issue:44933", "right": "issue:45459", "accept": false, - "reason": "Removed image_utils import and protobuf tokenization error masking are different import-time failures." + "reason": "Missing image_utils import and tokenizer error masking when protobuf is absent are separate issues." }, { "left": "issue:44869", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer-related, but they affect different models and different decode/codec paths." + "reason": "Whisper word-timestamp decode crash and Kimi-K2.5 tokenizer codec regression are not the same underlying bug." }, { "left": "issue:44297", "right": "issue:44623", "accept": false, - "reason": "Both are save_pretrained regressions, but tokenizer metadata mismatch and processor missing files are distinct serialization bugs." + "reason": "Tokenizer save metadata mismatch and processor save missing files affect different save_pretrained paths." }, { "left": "issue:43646", "right": "issue:44162", "accept": false, - "reason": "Broad v5 model-init regressions, but the reports describe different affected models and failure modes." + "reason": "Generic custom model initialization breakage is broader, but ESM2 is a separate model-specific regression." }, { "left": "issue:44361", "right": "issue:44749", "accept": false, - "reason": "Tokenizer AttributeError and a performance regression are not the same underlying issue." + "reason": "MLukeTokenizer AttributeError and a multilingual text slowdown regression are unrelated." }, { "left": "issue:43232", "right": "issue:43673", "accept": false, - "reason": "Both mention generation internals, but sync_gpus kwargs update and chunked_prefill cache missing are different bugs." + "reason": "Different generation internals: sync_gpus update bug vs. chunked_prefill cache regression." }, { "left": "issue:43673", "right": "issue:43827", "accept": false, - "reason": "Generation cache internals vs docs referencing removed pipeline APIs; unrelated." + "reason": "Generation cache regression and docs still referencing removed pipeline() API are unrelated." }, { "left": "issue:43452", "right": "issue:45003", "accept": false, - "reason": "gguf_file loading failures and unsafe sys.modules access are separate code paths." + "reason": "GGUF loading failures and unsafe sys.modules access are different code paths." }, { "left": "issue:44623", "right": "issue:45656", "accept": false, - "reason": "Processor serialization and DeepSpeed optimizer stepping are unrelated." + "reason": "Processor save regression and deepspeed double optimizer step are unrelated." }, { "left": "issue:28282", "right": "issue:30990", "accept": false, - "reason": "PyTorch-missing ImportError and Sentence Transformers loading hang are different environment/runtime problems." + "reason": "Missing PyTorch import error and Sentence Transformers loading hang are not the same bug." }, { "left": "issue:42994", "right": "issue:43482", "accept": false, - "reason": "Quantized model saving and GGUF loading failures are related only loosely by format, not the same bug." + "reason": "Quantized model saving failure and GGUF loading failure are separate quantization-related issues." }, { "left": "issue:44749", "right": "issue:45310", "accept": false, - "reason": "Performance regression after upgrade vs Qwen3.5-Moe from_pretrained error are different issues." + "reason": "A slowdown regression after upgrade and a Qwen3.5 Moe from_pretrained error are different failures." }, { "left": "issue:44749", "right": "issue:45656", "accept": false, - "reason": "Data filtering slowdown and DeepSpeed optimizer double-step are unrelated." + "reason": "Performance regression in data filtering and deepspeed optimizer stepping are unrelated." }, { "left": "issue:42371", "right": "issue:45092", "accept": false, - "reason": "TF32 API guidance and remote-code meta-init incompatibility are different topics." + "reason": "TF32 settings guidance and remote-code meta-initialization incompatibility are unrelated." }, { "left": "issue:43475", "right": "issue:43643", "accept": false, - "reason": "SAM3 output attribute missing and trust_remote_code missing fields are distinct model-loading issues." + "reason": "SAM3 output attribute error and missing fields in trust_remote_code AutoConfig are different bugs." + }, + { + "left": "issue:43475", + "right": "issue:45381", + "accept": false, + "reason": "SAM3 vision encoder missing attribute and qwen2.5-vl video position IDs are different model-specific issues." }, { "left": "issue:45230", "right": "issue:45397", "accept": false, - "reason": "Generic bug report and Gemma-4 Zero3 loading failure are not the same underlying problem." + "reason": "Generic bug report placeholder and Gemma-4 ZeRO3 from_pretrained failure are not duplicates." }, { "left": "issue:44038", "right": "issue:44623", "accept": false, - "reason": "Qwen3-VL-Moe from_pretrained and processor.save_pretrained missing files are different regressions." + "reason": "Qwen3-VL-Moe v5 breakage and processor.save_pretrained missing files are unrelated." }, { "left": "issue:41628", "right": "issue:44933", "accept": false, - "reason": "Missing AutoImageProcessor import and a nonexistent image_utils import are different API breakages." + "reason": "AutoImageProcessor import failure and a missing image_utils import are similar only superficially; they are different symbols and failure points." }, { "left": "issue:45081", "right": "issue:45103", "accept": false, - "reason": "Tokenizer regex patch crash and docstring kwargs processing crash are unrelated AttributeErrors." + "reason": "Mistral regex/tokenizer backend crash and auto-docstring future-annotations crash are unrelated." }, { - "left": "issue:42994", - "right": "issue:43525", + "left": "issue:45375", + "right": "issue:45478", "accept": false, - "reason": "Quantized saving failure and Llama4Config missing pad_token_id are separate model/config bugs." + "reason": "Missing config field in Qwen3_5MoeVisionConfig and Qwen3.5 Moe from_pretrained error are not the same concrete bug." }, { - "left": "issue:45375", - "right": "issue:45478", - "accept": true, - "reason": "Both point at the Qwen3.5-Moe from_pretrained regression; the missing deepstack_visual_indexes field plausibly explains the reported load failure." + "left": "issue:42994", + "right": "issue:43525", + "accept": false, + "reason": "Quantized save failure and Llama4Config missing pad_token_id are different issues." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "Grounded and conservative overall. The false-edge rejections are well supported by the titles, and the one accepted edge stays cautious: the Qwen3.5-Moe from_pretrained report and the missing deepstack_visual_indexes config-field report plausibly describe the same regression family without overstating certainty. The canonical issue choice is also reasonable given the exact same-title report." - }, - "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 }, @@ -11108,8 +11147,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -11176,8 +11215,8 @@ "issue:43377|issue:44242", "issue:44038|issue:45310", "issue:44496|issue:45561", - "issue:45561|issue:45600", "issue:44964|issue:45137", + "issue:45561|issue:45600", "issue:43901|issue:45561", "issue:43232|issue:44792", "issue:42898|issue:44568", @@ -11195,226 +11234,226 @@ "issue:44060|issue:45125", "issue:43874|issue:45092" ], - "prepared_review_unit_hash": "ae3ee36c606e26e059a8fa9534d1befe06832dee9070f12cb0b83a8d2742410a", + "prepared_review_unit_hash": "cd2428df64605287eabe893b99733196faf47804324479cd473e957e01530f0c", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37038, - "estimated_input_tokens": 9260, - "estimated_eval_tokens": 18776 + "serialized_chars": 37036, + "estimated_input_tokens": 9259, + "estimated_eval_tokens": 18774 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "One soft duplicate pair is a clear exact-match bug report about the same Qwen3ForCausalLM VRAM leak; the rest are different issues spanning unrelated import, tokenizer, multimodal, training, and testing regressions.", - "confidence": 0.93, - "canonical_issue_reason": "issue:42673 is the better canonical: it is the earlier report of the same Qwen3ForCausalLM multi-dataloader VRAM leak and has substantially more discussion activity.", + "summary": "This set is mostly a loose collection of unrelated Transformers issues; the only clear duplicate pair is the two identical Qwen3ForCausalLM VRAM-leak reports. Everything else looks like adjacent but distinct bugs, regressions, or docs/test issues.", + "confidence": 0.88, + "canonical_issue_reason": "issue:42673 is the best canonical issue because it is the older of the only exact duplicate pair and the titles/body indicate the same concrete bug as issue:43704.", "canonical_pr_reason": null, - "best_issue_reason": "issue:42673 is the strongest representative issue in this set because it is a concrete, high-signal bug report with a direct duplicate and good engagement.", + "best_issue_reason": "issue:42673 is the best overall representative because it is the clearest duplicate target in the cluster and anchors the only strong same-bug pairing.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42673", "right": "issue:43704", "accept": true, - "reason": "Exact same title and same Qwen3ForCausalLM VRAM leak symptom; clearly the same underlying bug." + "reason": "Exact same title and same reported VRAM leak in Qwen3ForCausalLM with multiple dataloader threads; these are the same underlying bug." }, { "left": "issue:43475", "right": "issue:44661", "accept": false, - "reason": "Different failures: missing Sam3VisionEncoderOutput attribute vs tokenizer mapping/add-new-model logic." + "reason": "Different failures: one is a Sam3VisionEncoderOutput attribute error, the other is an auto-mapping/config handling failure for tokenizer mapping names." }, { "left": "issue:45310", "right": "issue:45375", "accept": false, - "reason": "Same model family, but one is a from_pretrained error and the other is a missing config field; not the same concrete bug." + "reason": "Both involve Qwen3.5 MoE, but one is a from_pretrained error while the other is a missing config field dropped by strict loading; not the same code-path." }, { "left": "issue:41628", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for different symbols and different code paths." + "reason": "Both are import errors, but for different symbols and likely different export paths; not the same missing-import bug." }, { "left": "issue:42831", "right": "issue:44222", "accept": false, - "reason": "One is an accuracy regression, the other is a save_pretrained FP8/moe issue." + "reason": "Accuracy with FineGrainedFP8 and FP8 save_pretrained for moe are different problems affecting different code paths." }, { "left": "issue:42222", "right": "issue:45200", "accept": false, - "reason": "Different model areas and symptoms: vitpose breakage vs Gemma 4 token-type defaults." + "reason": "VitPose breakage and Gemma 4 mm_token_type_ids defaults are unrelated model-specific bugs." }, { "left": "issue:44222", "right": "issue:45310", "accept": false, - "reason": "Related to Qwen3.5 Moe, but save_pretrained FP8 and from_pretrained loading are distinct failures." + "reason": "Shared MoE/FP8 context is too broad; one is save_pretrained behavior, the other is a from_pretrained error in a different path." }, { "left": "issue:42994", "right": "issue:43824", "accept": false, - "reason": "Quantized saving failure and missing import for a Qwen2.5 VL class are unrelated." + "reason": "Quantized model saving failure and missing Qwen2_5_VLForConditionalGeneration import are unrelated issues." }, { "left": "issue:42175", "right": "issue:45712", "accept": false, - "reason": "Backend dependency installation issue vs leftover dummy class cleanup; not the same bug." + "reason": "TensorFlow backend packaging and leftover dummy classes are different maintenance/packaging problems." }, { "left": "issue:36010", "right": "issue:41628", "accept": false, - "reason": "Different missing imports: GenerationMixin vs AutoImageProcessor." + "reason": "Both are import-name errors, but for different classes from different modules; not the same missing export." }, { "left": "issue:43377", "right": "issue:44242", "accept": false, - "reason": "MIMI padding-mask mismatch and MoE load-balancing-loss logic are unrelated." + "reason": "MIMI padding-mask batching bug and MoE load-balancing loss omission are unrelated model behaviors." }, { "left": "issue:44038", "right": "issue:45310", "accept": false, - "reason": "Both mention Qwen/Moe, but one is a Transformers 5 compatibility problem and the other is a specific loading error." + "reason": "Different Qwen model families and different failure modes; too broad to be the same underlying bug." }, { "left": "issue:44496", "right": "issue:45561", "accept": false, - "reason": "Model config loading failure vs xdist race in testing utilities; no shared bug." + "reason": "Unrecognized model loading and xdist captured-info race are unrelated." }, { - "left": "issue:45561", - "right": "issue:45600", + "left": "issue:44964", + "right": "issue:45137", "accept": false, - "reason": "Captured-info race in xdist vs stale auto_mappings references; different failure modes." + "reason": "Phi-4 multimodal load failure and DeepSpeed ZeRO3 deque error are unrelated." }, { - "left": "issue:44964", - "right": "issue:45137", + "left": "issue:45561", + "right": "issue:45600", "accept": false, - "reason": "Multimodal model loading and DeepSpeed ZeRO3 deque error are unrelated." + "reason": "A pytest-xdist race on captured_info.txt is not the same as stale auto_mappings references to removed configs." }, { "left": "issue:43901", "right": "issue:45561", "accept": false, - "reason": "Docs mismatch for TextClassificationPipeline vs test-worker file race." + "reason": "Documentation mismatch for return_all_scores and xdist worker file race are unrelated." }, { "left": "issue:43232", "right": "issue:44792", "accept": false, - "reason": "Generation kwargs/sync_gpus bug and janus test failure are unrelated." + "reason": "Generation KV/sync_gpus handling and Janus image-generation test failure are different issues." }, { "left": "issue:42898", "right": "issue:44568", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different APIs and model-specific behavior." + "reason": "Both concern tokenizer behavior in v5, but one is cleanup spacing semantics while the other is BOS/EOS insertion for a specific tokenizer." }, { "left": "issue:30990", "right": "issue:35141", "accept": false, - "reason": "Sentence-transformers loading hang and embedding reinitialization on resize are different bugs." + "reason": "Sentence Transformers loading stall and token embedding resize reinitialization are unrelated." }, { "left": "issue:43232", "right": "issue:44188", "accept": false, - "reason": "Generation kwarg update issue vs torch.compile attention-kernel divergence; different code paths." + "reason": "Generation kwargs update after sync_gpus and attention-kernel divergence under torch.compile are different code paths." }, { "left": "issue:42898", "right": "issue:45812", "accept": false, - "reason": "Different tokenizer regressions affecting different behaviors and model families." + "reason": "Different tokenizer regressions affecting different behaviors and models." }, { "left": "issue:44188", "right": "issue:44373", "accept": false, - "reason": "Kernel branching bug vs incorrect docstring; not the same underlying issue." + "reason": "Attention kernel branching bug and docstring typo/mismatch for position_ids are unrelated." }, { "left": "issue:43784", "right": "issue:43824", "accept": false, - "reason": "Different import-time failures affecting different modules and symbols." + "reason": "Sentence-transformers import NameError and missing Qwen2.5-VL class import are different import failures." }, { "left": "issue:44623", "right": "issue:45310", "accept": false, - "reason": "Processor save_pretrained missing files and Qwen3.5 Moe loading error are unrelated." + "reason": "Processor save_pretrained file emission and Qwen3.5 MoE from_pretrained error are separate failures." }, { "left": "issue:43278", "right": "issue:45137", "accept": false, - "reason": "Dtype change on evaluation vs DeepSpeed deque underflow; different bugs." + "reason": "Embedding dtype drift in evaluate and DeepSpeed ZeRO3 deque underflow are unrelated." }, { "left": "issue:44734", "right": "issue:45103", "accept": false, - "reason": "KV cache continuation crash and auto_docstring annotation crash are unrelated." + "reason": "KV cache continuation crash in serve and auto_docstring crash with future annotations are different subsystems." }, { "left": "issue:30333", "right": "issue:33290", "accept": false, - "reason": "MLFlow reporting on failed jobs vs OOM in deepspeed/adafactor are not the same issue." + "reason": "MLflow reporting on failed training jobs and OOM with adafactor in deepspeed are unrelated." }, { "left": "issue:43723", "right": "issue:44568", "accept": false, - "reason": "Tokenizer loading regression vs missing BOS/EOS insertion for a specific tokenizer." + "reason": "General tokenizer loading issue and a specific add_special_tokens/BOS-EOS regression are not the same bug." }, { "left": "issue:42898", "right": "issue:45356", "accept": false, - "reason": "General v5 tokenization cleanup change vs Kimi-K2.5 codec regression; different problems." + "reason": "Different tokenizer regressions: cleanup spacing behavior versus codec handling and warning text for Kimi-K2.5." }, { "left": "issue:44060", "right": "issue:45125", "accept": false, - "reason": "Incorrect tied-weights warning and missing tensor-parallel plan are separate model-infrastructure issues." + "reason": "Incorrect tied-weights warning and missing tensor-parallel plan are unrelated Qwen3.5/Moe issues." }, { "left": "issue:43874", "right": "issue:45092", "accept": false, - "reason": "Missing image-patch helper and remote-code/meta-init incompatibility are different failures." + "reason": "Missing image-patch counting method and meta-init incompatibility for InternVL2 are different multimodal loading problems." } ] }, "evaluator_result": { "accept": true, - "feedback": "The conclusions are grounded in the packet. The canonical choice for issue:42673 over issue:43704 is reasonable because they share the exact title, same symptom, and the earlier issue has more discussion. The soft-edge verdicts are mostly conservative and do not overclaim equivalence for loosely related reports." + "feedback": "The writeup is grounded in the packet: the only clearly supported duplicate pair is issue:42673 \u2194 issue:43704, which do have identical titles and matching VRAM-leak descriptions. The remaining pair judgments stay conservative and do not overclaim equivalence based on broad thematic similarity alone. " }, "evaluator_used": true, "retried": false, @@ -11430,8 +11469,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -11498,19 +11537,19 @@ "issue:43479|issue:44869", "issue:44297|issue:45656", "issue:44206|issue:45103", - "issue:42831|issue:44623", "issue:44222|issue:44623", + "issue:42831|issue:44623", "issue:41628|issue:42757", "issue:30333|issue:36331", - "issue:43232|issue:45834", "issue:30333|issue:34689", + "issue:43232|issue:45834", "issue:43723|issue:44977", "issue:44222|issue:44484", - "issue:43873|issue:44945", "issue:44297|issue:45310", + "issue:43873|issue:44945", "issue:28282|issue:30333", - "issue:44263|issue:45092", "issue:30333|issue:35141", + "issue:44263|issue:45092", "issue:44263|issue:44360", "issue:44483|issue:45406", "issue:44373|issue:44485", @@ -11518,235 +11557,247 @@ "issue:44295|issue:44623", "issue:43479|issue:43644", "issue:44261|issue:45092", - "issue:45092|issue:45357", - "issue:43723|issue:44871" + "issue:43723|issue:44871", + "issue:45092|issue:45357" ], - "prepared_review_unit_hash": "0d0ebba687cc62327e38e04fa2a985b3bc5ba21063610837179af4d7d0edf95a", + "prepared_review_unit_hash": "34e8d5637beb7c0e9b9c38ec411cf49564020d061a571554407fa99967276cbb", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 35, - "serialized_chars": 38022, - "estimated_input_tokens": 9506, - "estimated_eval_tokens": 19268 + "serialized_chars": 38020, + "estimated_input_tokens": 9505, + "estimated_eval_tokens": 19266 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated issues across import errors, tokenizer/model regressions, trainer behavior, and save/load edge cases. The soft-similarity links are thematic at best; none look like the same underlying bug.", - "confidence": 0.91, - "canonical_issue_reason": "No single canonical issue: the items span many unrelated subsystems and versions, so a dedup canonical would be misleading.", + "summary": "This cluster is mostly a loose topical grouping of unrelated Transformers issues (tokenizers, save/load, model loading, training, backend imports, docs, and assorted regressions). None of the soft pairs look like the same underlying defect, so I reject all merges.", + "confidence": 0.82, + "canonical_issue_reason": "issue:44297 is the most concrete, self-contained bug report in the set (tokenizer.save_pretrained writes a mismatched tokenizer_class), so it is a reasonable anchor if one issue must represent the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "No global best issue for dedupe; the closest candidates are only superficially similar, not the same bug.", + "best_issue_reason": "issue:44297 is the clearest actionable report here: specific failure mode, specific API, and easier to triage than the broader version-wide or import-related regressions.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42898", "right": "issue:44779", "accept": false, - "reason": "Both concern tokenizer behavior in v5, but one is about cleanup spacing and the other is a DeepSeek output regression; different bugs." + "reason": "Both are tokenizer-related, but one is a v5 cleanup-space behavior change and the other is a Deepseek tokenization regression; different code paths and symptoms." }, { "left": "issue:45405", "right": "issue:45712", "accept": false, - "reason": "One is a dependency version bump, the other is cleanup of leftover dummy classes; unrelated changes." + "reason": "Unrelated: a dependency version bump vs leftover dummy classes leaking into dir() and repo checks." }, { "left": "issue:42898", "right": "issue:44448", "accept": false, - "reason": "Both mention v5 output changes, but they affect different models and code paths." + "reason": "One is a tokenizer cleanup-space behavior change; the other is a model output drift report for Pegasus. Same broad area, not the same bug." }, { "left": "issue:44188", "right": "issue:44964", "accept": false, - "reason": "Torch.compile attention divergence and Phi-4 model loading failure are different problems." + "reason": "Attention-kernel branching under torch.compile vs a Phi-4 multimodal model-loading failure are different defects." }, { "left": "issue:44295", "right": "issue:45656", "accept": false, - "reason": "Buffer registration/read failure and Deepspeed double optimizer step are unrelated." + "reason": "A position_ids buffer-read issue is unrelated to deepspeed calling optimizer.step() twice." }, { "left": "issue:43479", "right": "issue:44079", "accept": false, - "reason": "Config default initialization and ModelOutput key assignment are different code paths." + "reason": "Config initialization bug for Phi4MultimodalConfig vs ModelOutput key assignment bug; distinct objects and failure modes." }, { "left": "issue:44933", "right": "issue:45712", "accept": false, - "reason": "Import/export failure from image_utils is not the same as removing dummy PT classes." + "reason": "Import from image_utils vs leftover dummy class leakage; different modules and errors." }, { "left": "issue:42175", "right": "issue:44933", "accept": false, - "reason": "Packaging backend omission and a missing image_utils import are different issues." + "reason": "Packaging/backend omission with pip install .[torch] vs a missing/nonexistent image_utils import." }, { "left": "issue:44222", "right": "issue:44297", "accept": false, - "reason": "Both involve save_pretrained, but one is FP8 MoE saving and the other is tokenizer config mismatch." + "reason": "Both mention save_pretrained, but one is FP8 MoE persistence and the other is tokenizer config metadata; not the same bug." }, { "left": "issue:43976", "right": "issue:44933", "accept": false, - "reason": "Python version support regression is unrelated to an import-from-image_utils issue." + "reason": "Python 3.9/3.10 compatibility regression vs an image_utils import failure; unrelated." }, { "left": "issue:43479", "right": "issue:44869", "accept": false, - "reason": "A multimodal config defaulting bug is not the same as Whisper word timestamp decode crashing." + "reason": "Config default initialization vs Whisper word-timestamp decode crash; different subsystems." }, { "left": "issue:44297", "right": "issue:45656", "accept": false, - "reason": "Tokenizer save metadata mismatch and Deepspeed optimizer stepping are unrelated." + "reason": "Tokenizer save metadata mismatch vs deepspeed optimizer double-step; no shared code path." }, { "left": "issue:44206", "right": "issue:45103", "accept": false, - "reason": "Feature extractor argument regression and auto-docstring annotation handling are different bugs." + "reason": "Feature extractor center-arg regression vs auto_docstring annotation crash; unrelated areas." }, { - "left": "issue:42831", + "left": "issue:44222", "right": "issue:44623", "accept": false, - "reason": "FP8 accuracy problems and missing processor save_pretrained files are different failure modes." + "reason": "FP8 save_pretrained on MoE models vs processor.save_pretrained missing files; same broad save API, different concrete defects." }, { - "left": "issue:44222", + "left": "issue:42831", "right": "issue:44623", "accept": false, - "reason": "Both touch save_pretrained, but one is about MoE FP8 serialization and the other about processor artifacts missing." + "reason": "FP8 accuracy problem vs missing processor files on save; not the same failure." }, { "left": "issue:41628", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but they involve different symbols and likely different dependency breakages." + "reason": "Different missing imports: AutoImageProcessor vs is_offline_mode." }, { "left": "issue:30333", "right": "issue:36331", "accept": false, - "reason": "MLflow reporting on failed jobs and Trainer.compute_loss signature mismatch are unrelated." + "reason": "MLflow reporting after failed training jobs vs a Trainer compute_loss signature mismatch." }, { - "left": "issue:43232", - "right": "issue:45834", + "left": "issue:30333", + "right": "issue:34689", "accept": false, - "reason": "Generation kwargs update behavior and Kosmos2.5 OCR index errors are unrelated." + "reason": "Training-job reporting/MLflow issue vs Llama 3.2 vision model-loading regression." }, { - "left": "issue:30333", - "right": "issue:34689", + "left": "issue:43232", + "right": "issue:45834", "accept": false, - "reason": "MLflow run reporting and a model-loading regression are different issues." + "reason": "Generation kwarg handling after sync_gpus vs long OCR input index error; different models and code paths." }, { "left": "issue:43723", "right": "issue:44977", "accept": false, - "reason": "Tokenizer loading regression and flash-attention generation issues are not the same bug." + "reason": "Tokenizer loading regression in v5 vs Qwen3.5 generation issues with flash-attention." }, { "left": "issue:44222", "right": "issue:44484", "accept": false, - "reason": "A save_pretrained FP8 bug and a max_shard_size question are unrelated." + "reason": "FP8 save_pretrained bug vs a question about max_shard_size in save_pretrained; not the same defect." }, { - "left": "issue:43873", - "right": "issue:44945", + "left": "issue:44297", + "right": "issue:45310", "accept": false, - "reason": "Quantization/offloading behavior and pipeline-parallel output correctness are different problems." + "reason": "Tokenizer save_pretrained metadata mismatch vs Qwen3.5 MoE from_pretrained regression; different layers of the stack." }, { - "left": "issue:44297", - "right": "issue:45310", + "left": "issue:43873", + "right": "issue:44945", "accept": false, - "reason": "Tokenizer save metadata mismatch is unrelated to Qwen3.5 MoE from_pretrained failure." + "reason": "Quantized offloading behavior vs incorrect pipeline-parallel output; related broad area, but distinct bugs." }, { "left": "issue:28282", "right": "issue:30333", "accept": false, - "reason": "Missing PyTorch import support and MLflow job reporting are unrelated." + "reason": "PyTorch-missing import error vs MLflow job-reporting problem." }, { - "left": "issue:44263", - "right": "issue:45092", + "left": "issue:30333", + "right": "issue:35141", "accept": false, - "reason": "GlmMoeDsaIndexer split-value behavior and InternVL meta-init incompatibility are different code paths." + "reason": "MLflow/reporting issue vs embedding reinitialization after resizing token embeddings." }, { - "left": "issue:30333", - "right": "issue:35141", + "left": "issue:44263", + "right": "issue:45092", "accept": false, - "reason": "MLflow reporting on failure and output embedding reinitialization are unrelated." + "reason": "torch.split return-value bug in a DSA indexer vs InternVL2 meta-init incompatibility; unrelated." }, { "left": "issue:44263", "right": "issue:44360", "accept": false, - "reason": "Both mention the DSA indexer, but one is about torch.split return handling and the other about a missing ReLU; not the same bug." + "reason": "Both touch the DSA indexer, but one is a torch.split shape/return issue and the other is a missing ReLU; same subsystem, different bugs." }, { "left": "issue:44483", "right": "issue:45406", "accept": false, - "reason": "Chat completions request acceptance and Gemma4Processor missing _tokenizer are different failures." + "reason": "Chat/completions API rejection vs Gemma4Processor missing _tokenizer; distinct failures." }, { "left": "issue:44373", "right": "issue:44485", "accept": false, - "reason": "Wrong docstring and GLM-5 RoPE implementation are not the same underlying issue." + "reason": "Wrong docstring for position_ids vs GLM-5 RoPE implementation; one is documentation, the other is model math." }, { "left": "issue:44514", "right": "issue:45103", "accept": false, - "reason": "Batched chat-template crash and future-annotations auto-docstring crash are unrelated." + "reason": "Batched chat-template crash vs auto_docstring annotation crash; unrelated functionality." }, { "left": "issue:44295", - "right": "issue:45092", + "right": "issue:44623", "accept": false, - "reason": "Reading a registered buffer and remote-code meta initialization incompatibility are different problems." + "reason": "Reading a registered buffer issue vs processor.save_pretrained missing files; different APIs and symptoms." }, { - "left": "issue:45092", - "right": "issue:45357", + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Phi4Multimodal config defaulting bug vs non-persistent buffers being filled with junk in Transformers 5.0.0." + }, + { + "left": "issue:44261", + "right": "issue:45092", "accept": false, - "reason": "Both are version-related model regressions, but they affect different models and different save/load paths." + "reason": "MLA q_a_layernorm precision/config bug vs old InternVL2 remote-code meta-init incompatibility; not the same underlying defect." }, { "left": "issue:43723", "right": "issue:44871", "accept": false, - "reason": "Tokenizer loading regression and eos_token_id config inconsistency are different issues." + "reason": "Tokenizer loading failure in v5 vs Gemma-3 eos_token_id configuration mismatch." + }, + { + "left": "issue:45092", + "right": "issue:45357", + "accept": false, + "reason": "InternVL2 meta-init compatibility bug vs Qwen3.5 save_pretrained writing incorrect visual encoder keys." } ] }, @@ -11765,8 +11816,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -11821,15 +11872,15 @@ "soft_pairs": [ "issue:30990|issue:38175", "issue:41628|issue:43824", - "issue:43479|issue:44913", - "issue:42886|issue:42994", "issue:39401|issue:44568", - "issue:42831|issue:45310", + "issue:43479|issue:44913", "issue:22355|issue:28282", + "issue:42831|issue:45310", + "issue:42886|issue:42994", "issue:44263|issue:44373", "issue:44945|issue:45753", - "issue:42831|issue:44038", "issue:42757|issue:45278", + "issue:42831|issue:44038", "issue:43576|issue:43784", "issue:42886|issue:43066", "issue:43232|issue:45593", @@ -11841,29 +11892,29 @@ "issue:44704|issue:45290", "issue:43122|issue:44514", "issue:44295|issue:44297", - "issue:43232|issue:45245", + "issue:41628|issue:45278", + "issue:43723|issue:44779", "issue:43723|issue:44442", "issue:43784|issue:45600", - "issue:43723|issue:44779", - "issue:41628|issue:45278", "issue:42907|issue:43937", + "issue:43232|issue:45245", "issue:44261|issue:44373", "issue:43408|issue:43475" ], - "prepared_review_unit_hash": "eede6dedc36343b6f3259edee00d5993b2cae0dc09d904cba39772aa2ea941a3", + "prepared_review_unit_hash": "ff07c3af84ba9bd3a9fd71850e6018ee92f5b36966b4109f03d7dfbc1e3cf667", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 35654, + "serialized_chars": 35655, "estimated_input_tokens": 8914, "estimated_eval_tokens": 18084 }, @@ -11872,192 +11923,192 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly separate bug reports with only broad thematic overlap (imports, tokenization, config, generation, quantization). None of the soft pairs look like true duplicates of the same concrete bug/change. If one issue has to represent the set, 45278 is the broadest umbrella, but the cluster is too heterogeneous for a strong canonical duplicate.", - "confidence": 0.78, - "canonical_issue_reason": "issue:45278 is the broadest representative because it covers multiple import regressions after the 4.57.0\u21925.5.0 upgrade, but it is only a coarse umbrella and not a clean duplicate of the rest.", + "summary": "This cluster is a loose collection of mostly unrelated bug reports, with some superficial similarity around imports/tokenizers/model-specific regressions but no single duplicate family.", + "confidence": 0.95, + "canonical_issue_reason": "No clear canonical issue: the items span many distinct subsystems and failure modes, so they do not form one duplicate cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the best single representative for the cluster\u2019s loose import-regression theme, though the underlying reports remain largely distinct.", + "best_issue_reason": "If one issue must be highlighted, issue 45278 is the broadest umbrella for upgrade-related import errors, but it still does not represent the whole mixed cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:30990", "right": "issue:38175", "accept": false, - "reason": "Sentence-Transformers loading hang vs. SigLIP2 zero-probability output; different symptoms and codepaths." + "reason": "Both are model-loading / inference complaints, but they affect different models and different symptoms." }, { "left": "issue:41628", "right": "issue:43824", "accept": false, - "reason": "Missing AutoImageProcessor import vs. missing Qwen2_5_VLForConditionalGeneration import; different symbols and failures." + "reason": "Both are import errors, but for different missing symbols and likely different code paths." }, { - "left": "issue:43479", - "right": "issue:44913", + "left": "issue:39401", + "right": "issue:44568", "accept": false, - "reason": "Phi4Multimodal default config initialization vs. GPTNeoX rotary_pct reload behavior; unrelated config bugs." + "reason": "Both are tokenizer issues, but one is offset mapping and the other is special-token insertion behavior." }, { - "left": "issue:42886", - "right": "issue:42994", + "left": "issue:43479", + "right": "issue:44913", "accept": false, - "reason": "Offline tokenizer cache loading vs. quantized model saving failure; not the same bug." + "reason": "Different config persistence bugs in different model classes; not the same defect." }, { - "left": "issue:39401", - "right": "issue:44568", + "left": "issue:22355", + "right": "issue:28282", "accept": false, - "reason": "Wrong offset_mapping vs. add_special_tokens not adding BOS/EOS; different tokenizer behaviors." + "reason": "One is a missing transformers.onnx module, the other is an AutoModel PyTorch dependency error." }, { "left": "issue:42831", "right": "issue:45310", "accept": false, - "reason": "FineGrainedFP8 accuracy issue vs. Qwen3.5 MoE from_pretrained error; different problems." + "reason": "One is an FP8 accuracy issue, the other is a Qwen3.5 MoE from_pretrained failure." }, { - "left": "issue:22355", - "right": "issue:28282", + "left": "issue:42886", + "right": "issue:42994", "accept": false, - "reason": "transformers.onnx module missing vs. PyTorch dependency import error; different import failures." + "reason": "Offline tokenizer cache loading and quantized model saving are unrelated code paths." }, { "left": "issue:44263", "right": "issue:44373", "accept": false, - "reason": "torch.split return value issue vs. wrong docstring; one is a code bug, the other documentation." + "reason": "A tensor-splitting bug and a docstring issue are not the same underlying problem." }, { "left": "issue:44945", "right": "issue:45753", "accept": false, - "reason": "Pipeline-parallelism output bug vs. Qwen3_5 image-specific infinite loop; unrelated inference failures." + "reason": "Pipeline-parallel output corruption and a Qwen3.5 image loop are different model/runtime failures." }, { - "left": "issue:42831", - "right": "issue:44038", + "left": "issue:42757", + "right": "issue:45278", "accept": false, - "reason": "FineGrainedFP8 accuracy regression vs. Qwen3-VL-Moe bug; different model/codepaths." + "reason": "The first is one missing import from huggingface_hub; the second is a broad report of many import errors, but not the same concrete bug." }, { - "left": "issue:42757", - "right": "issue:45278", + "left": "issue:42831", + "right": "issue:44038", "accept": false, - "reason": "Specific huggingface_hub API import error vs. broad set of import errors after upgrade; overlap is too vague for a duplicate." + "reason": "Different model- and version-specific issues with no shared failing code path." }, { "left": "issue:43576", "right": "issue:43784", "accept": false, - "reason": "Broken transformers env CLI vs. sentence-transformers import NameError; unrelated failure modes." + "reason": "Broken CLI env command and a NameError during sentence-transformers import are different failures." }, { "left": "issue:42886", "right": "issue:43066", "accept": false, - "reason": "Offline cache loading issue vs. wrong tokenizer decoder type in v5; not the same underlying bug." + "reason": "Offline cache loading and tokenizer decoder type mismatch are unrelated tokenizer/config bugs." }, { "left": "issue:43232", "right": "issue:45593", "accept": false, - "reason": "Generation after sync_gpus issue vs. D-FINE auxiliary-loss handling; unrelated features." + "reason": "Generation kwargs handling and D-FINE auxiliary loss behavior are different subsystems." }, { "left": "issue:42371", "right": "issue:44263", "accept": false, - "reason": "TF32 settings warning vs. torch.split return values; no shared bug." + "reason": "TF32 configuration guidance and torch.split return values are unrelated." }, { "left": "issue:43756", "right": "issue:44060", "accept": false, - "reason": "Smollm3 RoPE layer count mismatch vs. Qwen3-Next tied-weights warning; different models and defects." + "reason": "Different model architecture/config bugs; same broad family only." }, { "left": "issue:44117", "right": "issue:44704", "accept": false, - "reason": "TOKENIZER_MAPPING_NAMES None in from_pretrained vs. AutoProcessor kwargs forwarding; different loading paths." + "reason": "Tokenizer mapping fallback and AutoProcessor cached_file kwargs are different loader issues." }, { "left": "issue:30990", "right": "issue:33357", "accept": false, - "reason": "Loading hang vs. MacOS bus error; distinct runtime failures." + "reason": "Both mention loading/inference, but one is a Sentence Transformers hang and the other is a macOS bus error with a CLIP model." }, { "left": "issue:42907", "right": "issue:44945", "accept": false, - "reason": "Saving dequantized models vs. incorrect pipeline-parallel output; unrelated codepaths." + "reason": "Quantized model saving and pipeline-parallel output errors are unrelated." }, { "left": "issue:44704", "right": "issue:45290", "accept": false, - "reason": "AutoProcessor kwargs passthrough vs. chat template crash with tool calls; different APIs and bugs." + "reason": "AutoProcessor kwargs forwarding and chat template tool-call crashing are distinct behaviors." }, { "left": "issue:43122", "right": "issue:44514", "accept": false, - "reason": "General v4\u2192v5 tokenization change vs. Qwen2_5_VL chat template batched crash; not the same issue." + "reason": "Tokenizer output changes and processor chat-template batching failures do not indicate the same bug." }, { "left": "issue:44295", "right": "issue:44297", "accept": false, - "reason": "position_ids buffer read error vs. tokenizer_class mismatch on save_pretrained; different serialization issues." + "reason": "Reading a registered buffer and saving tokenizer metadata are different serialization issues." }, { - "left": "issue:43232", - "right": "issue:45245", + "left": "issue:41628", + "right": "issue:45278", "accept": false, - "reason": "Generation update bug vs. category-count limit runtime error; unrelated." + "reason": "One specific missing import is not enough to treat it as the same defect as a broad multi-import regression report." }, { "left": "issue:43723", - "right": "issue:44442", - "accept": false, - "reason": "Generic AutoTokenizer loading regression vs. failure for a specific tokenizer class; not enough evidence of the same bug." - }, - { - "left": "issue:43784", - "right": "issue:45600", + "right": "issue:44779", "accept": false, - "reason": "sentence-transformers import NameError vs. Sam3LiteText config references breaking CI; unrelated." + "reason": "Both involve tokenizer loading/behavior, but they target different failure modes and likely different tokenizer implementations." }, { "left": "issue:43723", - "right": "issue:44779", + "right": "issue:44442", "accept": false, - "reason": "Tokenizer loading issue vs. wrong Deepseek tokenization output; different failure classes." + "reason": "AutoTokenizer loading failure and a specific FastSpeech2ConformerTokenizer failure are not the same issue." }, { - "left": "issue:41628", - "right": "issue:45278", + "left": "issue:43784", + "right": "issue:45600", "accept": false, - "reason": "Specific AutoImageProcessor import error vs. broad import-error umbrella; too general to treat as one duplicate bug." + "reason": "Sentence-transformers import NameError and stale auto_mappings references break for different reasons." }, { "left": "issue:42907", "right": "issue:43937", "accept": false, - "reason": "Quantized model saving failure vs. invalid GenerationConfig in GLM-5; unrelated." + "reason": "Model saving and invalid GenerationConfig validation are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:45245", + "accept": false, + "reason": "Generation sync behavior and category-count limits are different code paths." }, { "left": "issue:44261", "right": "issue:44373", "accept": false, - "reason": "MLA precision/config issue vs. docstring typo; clearly not duplicates." + "reason": "A precision/config bug and a wrong docstring are not duplicates." }, { "left": "issue:43408", "right": "issue:43475", "accept": false, - "reason": "SAM3 model-type warning vs. missing fpn_position_embeddings attribute; related area but different concrete bugs." + "reason": "Both touch SAM 3 video, but one is a model-type warning and the other is a missing attribute crash; not clearly the same bug." } ] }, @@ -12076,8 +12127,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -12122,11 +12173,11 @@ "issue:44936", "issue:44964", "issue:44977", + "issue:45083", "issue:45200", "issue:45216", "issue:45446", "issue:45478", - "issue:45563", "issue:45753" ], "soft_pairs": [ @@ -12139,14 +12190,14 @@ "issue:43122|issue:44779", "issue:44038|issue:44297", "issue:42907|issue:43452", - "issue:44038|issue:44295", "issue:39401|issue:44488", + "issue:44038|issue:44295", "issue:43824|issue:44351", "issue:36296|issue:43116", "issue:44038|issue:44222", "issue:39401|issue:44521", - "issue:45216|issue:45478", "issue:42757|issue:44351", + "issue:45216|issue:45478", "issue:43066|issue:44779", "issue:22355|issue:28218", "issue:42907|issue:44222", @@ -12158,246 +12209,252 @@ "issue:42371|issue:43874", "issue:43976|issue:45446", "issue:43493|issue:45753", - "issue:42831|issue:45753", "issue:44075|issue:44936", - "issue:44038|issue:45478", + "issue:42831|issue:45753", "issue:43116|issue:43825", + "issue:44038|issue:45478", "issue:44222|issue:45478", "issue:43116|issue:43493", - "issue:43937|issue:45563" + "issue:45083|issue:45446" ], - "prepared_review_unit_hash": "8c6fe6ef11874691f36e3691ed0a24f388a3072f71ed64d1aa429cd311d82105", + "prepared_review_unit_hash": "deee57f29e32d9952ddadc204d11b6ac04eca99e944ccf1262d4287023215377", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 35, - "serialized_chars": 37059, - "estimated_input_tokens": 9265, - "estimated_eval_tokens": 18786 + "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 items are mostly unrelated standalone bugs. The soft pairs share broad keywords (tokenizer, save_pretrained, Qwen/GLM, etc.) but do not describe the same concrete defect or fix path, so none should be merged as duplicates.", - "confidence": 0.88, - "canonical_issue_reason": "issue:43122 is the broadest tokenizer-regression report in the set and one of the few nodes that plausibly sits near several tokenizer-related soft hits, making it the best single anchor if a canonical issue must be chosen.", + "summary": "Mostly false-positive similarity matches: the items span unrelated tokenizer, model-loading, training, export, and config bugs, so there is no cohesive duplicate cluster.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue fits well; the cluster is heterogeneous and the soft links appear to connect unrelated bugs rather than one underlying defect.", "canonical_pr_reason": null, - "best_issue_reason": "Despite the weak cohesion of the cluster, issue:43122 is the most reusable representative because it captures the shared tokenizer-regression theme more generically than the other candidates.", + "best_issue_reason": "If forced to pick a representative, issue:43122 is the broadest tokenizer-regression report, but it still does not subsume the rest of the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44360", "right": "issue:44485", "accept": false, - "reason": "Different subsystems and failures: DSA/ReLU discussion vs GLM-5 RoPE implementation." + "reason": "Unrelated DSA/indexer discussion versus GLM-5 RoPE implementation." }, { "left": "issue:43122", "right": "issue:44521", "accept": false, - "reason": "Both mention tokenization-adjacent behavior, but one is a version-to-version tokenizer regression and the other is a multimodal chat-template assistant_mask bug." + "reason": "Tokenizer regression and multimodal assistant-mask bug are different code paths." }, { "left": "issue:44625", "right": "issue:44829", "accept": false, - "reason": "Qwen3.5 num_labels propagation and FlashAttention-3 training degeneration are different code paths and symptoms." + "reason": "num_labels propagation issue is unrelated to flash_attention_3 training degeneration." }, { "left": "issue:43116", "right": "issue:45200", "accept": false, - "reason": "Example-script multi-label output issue vs Gemma 4 mm_token_type_ids defaulting problem; not the same bug." + "reason": "Example-script classification output bug is unrelated to mm_token_type_ids defaults." }, { "left": "issue:29127", "right": "issue:30990", "accept": false, - "reason": "Clarified error messaging in layoutlmv3 is unrelated to Sentence Transformers hanging on load." + "reason": "LayoutLMv3 error messaging and Sentence Transformers loading hang are unrelated." }, { "left": "issue:42907", "right": "issue:44964", "accept": false, - "reason": "Dequantized save failure for Ministral/Devstral is unrelated to Phi-4 multimodal model loading." + "reason": "Save-dequantized failure and Phi-4 multimodal load failure are different problems." }, { "left": "issue:43122", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer regressions, but they describe different model-specific behavior and no shared concrete failing path is evident." + "reason": "Both are tokenization regressions, but on different models with different symptoms and code paths." }, { "left": "issue:44038", "right": "issue:44297", "accept": false, - "reason": "A Qwen3-VL-Moe load/runtime issue is unrelated to tokenizer_class metadata written by save_pretrained." + "reason": "Qwen3-VL-Moe loading failure is unrelated to tokenizer_class metadata saved by save_pretrained." }, { "left": "issue:42907", "right": "issue:43452", "accept": false, - "reason": "Saving dequantized models and loading via gguf_file are different code paths and failure modes." + "reason": "Dequantized-save bug and gguf_file/pretrained loading bug are not the same defect." }, { - "left": "issue:44038", - "right": "issue:44295", + "left": "issue:39401", + "right": "issue:44488", "accept": false, - "reason": "Qwen3-VL-Moe failure and position_ids buffer access error are distinct bugs." + "reason": "offset_mapping bug and model loading failure are unrelated." }, { - "left": "issue:39401", - "right": "issue:44488", + "left": "issue:44038", + "right": "issue:44295", "accept": false, - "reason": "Wrong offset_mapping in Qwen3 tokenizer is unrelated to failing to load a specific model." + "reason": "Qwen3-VL-Moe loading issue and position_ids buffer access error are unrelated." }, { "left": "issue:43824", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but they concern different missing symbols from different packages." + "reason": "Both are import errors, but for different symbols and different causes." }, { "left": "issue:36296", "right": "issue:43116", "accept": false, - "reason": "Tensor-parallel training bug and empty multi-label example output are unrelated." + "reason": "Tensor-parallel training bug is unrelated to the example script's empty multi-label results." }, { "left": "issue:44038", "right": "issue:44222", "accept": false, - "reason": "Qwen3-VL-Moe loading/version problem is not the same as FP8 save_pretrained for MoE." + "reason": "Model loading failure and FP8 save_pretrained bug affect different code paths." }, { "left": "issue:39401", "right": "issue:44521", "accept": false, - "reason": "Wrong tokenizer offset_mapping and zero assistant_masks are separate tokenizer-related bugs." + "reason": "Tokenizer offset_mapping and multimodal assistant_masks are unrelated tokenizer features." }, { - "left": "issue:45216", - "right": "issue:45478", + "left": "issue:42757", + "right": "issue:44351", "accept": false, - "reason": "One is a save_pretrained regression and the other is a from_pretrained regression; related model family, but not the same concrete failure." + "reason": "Different import failures from different packages/symbols; not the same bug." }, { - "left": "issue:42757", - "right": "issue:44351", + "left": "issue:45216", + "right": "issue:45478", "accept": false, - "reason": "Missing is_offline_mode from huggingface_hub and missing HybridCache from transformers are different import/export issues." + "reason": "Same model family, but save_pretrained checkpoint corruption and from_pretrained failure are distinct bugs." }, { "left": "issue:43066", "right": "issue:44779", "accept": false, - "reason": "Wrong tokenizer decoder type in v5 and Deepseek tokenization regression are different tokenizer-path problems." + "reason": "Wrong decoder type and Deepseek tokenization regression are different issues." }, { "left": "issue:22355", "right": "issue:28218", "accept": false, - "reason": "ONNX module import failure is unrelated to added-token spacing behavior." + "reason": "Missing transformers.onnx module and extra tokenizer space are unrelated." }, { "left": "issue:42907", "right": "issue:44222", "accept": false, - "reason": "Both involve saving models, but one is dequantized Ministral/Devstral saving and the other is FP8 MoE saving." + "reason": "Both touch saving, but they are model-specific quantization/save bugs with different failure modes." }, { "left": "issue:43122", "right": "issue:44977", "accept": false, - "reason": "Tokenizer regression and flash-attention generation failure are unrelated." + "reason": "Tokenizer regression and Qwen3.5 flash-attention generation bug are unrelated." }, { "left": "issue:43992", "right": "issue:44704", "accept": false, - "reason": "Missing embed_tokens.weight in UMT5Encoder is not the same as AutoProcessor dropping kwargs to cached_file." + "reason": "Missing embed_tokens.weight and cached_file kwargs forwarding are unrelated." }, { "left": "issue:43784", "right": "issue:45446", "accept": false, - "reason": "sentence-transformers import NameError and flex_attention PyTorch version check are unrelated import/version bugs." + "reason": "sentence-transformers import NameError and flex_attention version check are unrelated." }, { "left": "issue:43937", "right": "issue:44485", "accept": false, - "reason": "Invalid GenerationConfig and GLM-5 RoPE implementation concern different failure modes." + "reason": "Invalid GenerationConfig and RoPE implementation are unrelated." }, { "left": "issue:42222", "right": "issue:43441", "accept": false, - "reason": "Broken vitpose model imports and Ministral flash-attention failure do not share a concrete code path." + "reason": "Vitpose model breakage and Ministral FlashAttention failure are unrelated." }, { "left": "issue:42371", "right": "issue:43874", "accept": false, - "reason": "TF32 API guidance and missing get_number_of_image_patches are unrelated." + "reason": "TF32 settings warning and missing image-patch helper are unrelated." }, { "left": "issue:43976", "right": "issue:45446", "accept": false, - "reason": "Python version compatibility and a torch-import version check are different compatibility problems." + "reason": "Python version compatibility issue and flex_attention version check are unrelated." }, { "left": "issue:43493", "right": "issue:45753", "accept": false, - "reason": "SigLIP2 implementation discrepancy and a Qwen3_5 image-specific infinite loop are distinct bugs." + "reason": "SigLIP2 implementation discrepancy and Qwen3_5 infinite-loop bug are unrelated." + }, + { + "left": "issue:44075", + "right": "issue:44936", + "accept": false, + "reason": "Optimizer argument handling and trainer.evaluate-after-train are different trainer bugs." }, { "left": "issue:42831", "right": "issue:45753", "accept": false, - "reason": "FP8 accuracy degradation is unrelated to an image-triggered generation loop." + "reason": "FineGrainedFP8 accuracy issue and image-specific infinite loop are unrelated." }, { - "left": "issue:44075", - "right": "issue:44936", + "left": "issue:43116", + "right": "issue:43825", "accept": false, - "reason": "SGD arguments being ignored and trainer.evaluate failing after training are separate training issues." + "reason": "Example-script classification bug and pipeline translation error message are unrelated." }, { "left": "issue:44038", "right": "issue:45478", "accept": false, - "reason": "Different Qwen3 family variants and different failure surfaces; not the same concrete bug." + "reason": "Related model area, but Qwen3-VL-Moe load failure and Qwen3.5 MoE from_pretrained error are not the same concrete bug." }, { - "left": "issue:43116", - "right": "issue:43825", + "left": "issue:44222", + "right": "issue:45478", "accept": false, - "reason": "Example-script classification output bug is unrelated to the pipeline translation-task warning message." + "reason": "FP8 save_pretrained bug and Qwen3.5 MoE from_pretrained failure are distinct save/load problems." }, { "left": "issue:43116", "right": "issue:43493", "accept": false, - "reason": "Multi-label classification output issue and SigLIP2 implementation discrepancy are unrelated." + "reason": "Classification example bug and SigLIP2 discrepancy are unrelated." }, { - "left": "issue:43937", - "right": "issue:45563", + "left": "issue:45083", + "right": "issue:45446", "accept": false, - "reason": "GenerationConfig validation failure is unrelated to a stale warning in paged generate()." + "reason": "qwen3_omni_moe feature-length helper bug and flex_attention import/version check are unrelated." } ] }, @@ -12416,8 +12473,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -12459,7 +12516,6 @@ "issue:45071", "issue:45072", "issue:45081", - "issue:45083", "issue:45216", "issue:45278", "issue:45310", @@ -12467,15 +12523,16 @@ "issue:45446", "issue:45464", "issue:45542", + "issue:45563", "issue:45588", "issue:45656" ], "soft_pairs": [ "issue:43064|issue:43673", - "issue:45083|issue:45446", + "issue:43937|issue:45563", "issue:38617|issue:45278", - "issue:33453|issue:39401", "issue:39401|issue:44779", + "issue:33453|issue:39401", "issue:22355|issue:29127", "issue:44117|issue:45588", "issue:43827|issue:45446", @@ -12495,214 +12552,202 @@ "issue:43408|issue:43742", "issue:43937|issue:44496", "issue:42175|issue:42222", - "issue:44483|issue:45464", "issue:44075|issue:45656", + "issue:44483|issue:45464", "issue:44964|issue:45278", "issue:43012|issue:43577", "issue:43066|issue:43122" ], - "prepared_review_unit_hash": "7a700f14098d723ef94248f3f0db46bd35dba2d2640c4192995fc39e36ecebbf", + "prepared_review_unit_hash": "d4d76187492f6c77e4f04646167697b1e8c80d3ff4c2300fd4339d5566f22b38", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36363, - "estimated_input_tokens": 9091, - "estimated_eval_tokens": 18438 + "serialized_chars": 36332, + "estimated_input_tokens": 9083, + "estimated_eval_tokens": 18422 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is a collection of unrelated Transformers issues with some title-level similarity, but none of the soft edges look like true duplicates. I rejected all pairings and found no safe canonical issue/PR representative.", - "confidence": 0.88, - "canonical_issue_reason": "No canonical issue: the items span many different bug classes and model-specific regressions, so none is a reliable duplicate target for the whole cluster.", + "summary": "These issues are mostly unrelated individual regressions; none of the soft edges look like true duplicates or mergeable PR-equivalent changes.", + "confidence": 0.94, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "No best issue: there is no single issue that meaningfully represents the cluster, since the candidates are about distinct subsystems, models, and failure modes.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43064", "right": "issue:43673", "accept": false, - "reason": "Different code paths: distributed training optimizer-state corruption vs generation cache missing in chunked prefill." + "reason": "Different failures: distributed optimizer state corruption vs missing GenerationMixin cache during chunked prefill." }, { - "left": "issue:45083", - "right": "issue:45446", + "left": "issue:43937", + "right": "issue:45563", "accept": false, - "reason": "Unrelated bugs: qwen3_omni_moe feature behavior vs a flex_attention PyTorch version check." + "reason": "Both mention generation, but one is invalid GenerationConfig and the other is a stale warning for num_return_sequences." }, { "left": "issue:38617", "right": "issue:45278", "accept": false, - "reason": "One is a specific missing import symbol; the other is a broad report of many import errors after upgrade." + "reason": "Same broad import-error theme, but different missing symbols and likely different regressions." }, { - "left": "issue:33453", - "right": "issue:39401", + "left": "issue:39401", + "right": "issue:44779", "accept": false, - "reason": "Tokenizer regression vs wrong offset_mapping for Qwen3; related domain but different concrete failures." + "reason": "Tokenizer-related, but Qwen3 wrong offset mapping and Deepseek incorrect tokenization are different bugs." }, { - "left": "issue:39401", - "right": "issue:44779", + "left": "issue:33453", + "right": "issue:39401", "accept": false, - "reason": "Different model-specific tokenizer bugs: offset_mapping error vs incorrect DeepSeek tokenization." + "reason": "Both are tokenizer regressions, but one is generic loading regression and the other is a model-specific offset_mapping bug." }, { "left": "issue:22355", "right": "issue:29127", "accept": false, - "reason": "Completely different problems: missing transformers.onnx module vs a LayoutLMv3 error message issue." + "reason": "Completely different problems: missing transformers.onnx module vs LayoutLMv3 error-message clarity." }, { "left": "issue:44117", "right": "issue:45588", "accept": false, - "reason": "Tokenizer mapping lookup failure vs flash_attention crash on s_aux=None; not the same bug." + "reason": "Unrelated code paths: tokenizer mapping lookup vs flash_attention crash on s_aux=None." }, { "left": "issue:43827", "right": "issue:45446", "accept": false, - "reason": "Docs referencing removed pipeline() vs an attention version-check bug; no shared code-path defect." + "reason": "Docs reference cleanup and a PyTorch version check bug are not the same underlying issue." }, { "left": "issue:43381", "right": "issue:44945", "accept": false, - "reason": "Gradient checkpointing in eval mode and pipeline parallelism output errors are distinct failures." + "reason": "Gradient checkpointing eval-mode behavior is unrelated to pipeline-parallelism output corruption." }, { "left": "issue:43937", "right": "issue:44977", "accept": false, - "reason": "Different model regressions: invalid GenerationConfig for GLM-5 vs Qwen3.5 flash-attention generation failure." + "reason": "Different concrete bugs: invalid generation config vs Qwen3.5 flash-attention generation failure." }, { "left": "issue:45405", "right": "issue:45446", "accept": false, - "reason": "PEFT dependency version mismatch is unrelated to the flex_attention import/version-check problem." + "reason": "Packaging/version pin issue versus an import-version check bug; not the same fix." }, { "left": "issue:43450", "right": "issue:43986", "accept": false, - "reason": "Video batch shape handling vs AutoProcessor crash without torchvision; different root causes." + "reason": "Batched video shape bug and AutoProcessor crash without torchvision are different failure modes." }, { "left": "issue:42673", "right": "issue:45072", "accept": false, - "reason": "VRAM leak in Qwen3ForCausalLM multithreaded loading vs dtype mismatch in CI; unrelated." + "reason": "VRAM leak in Qwen3ForCausalLM threads is unrelated to dtype mismatches in other models." }, { "left": "issue:44117", "right": "issue:45081", "accept": false, - "reason": "Both are tokenizer loading issues, but one is mapping lookup and the other is a Mistral regex patch crash." + "reason": "TOKENIZER_MAPPING_NAMES None handling is unrelated to the Mistral regex patch crash." }, { "left": "issue:38617", "right": "issue:41628", "accept": false, - "reason": "Different missing-import regressions for different symbols/modules, not one underlying bug." + "reason": "Both are import errors, but for different exported names and likely unrelated causes." }, { "left": "issue:39401", "right": "issue:43421", "accept": false, - "reason": "Offset mapping bug and runtime post-processor update behavior are separate tokenizer issues." + "reason": "Wrong offset mapping and stale post-processor updates after special-token edits are different tokenizer bugs." }, { "left": "issue:44410", "right": "issue:45310", "accept": false, - "reason": "Missing projections in qwen3next layer 0 vs Qwen3.5 MoE from_pretrained error; not the same defect." + "reason": "Different Qwen-family issues: missing projections in qwen3next vs from_pretrained failure in Qwen3.5 MoE." }, { "left": "issue:44829", "right": "issue:45216", "accept": false, - "reason": "Degenerate training with flash_attention_3 is unrelated to a wrong saved checkpoint from save_pretrained()." + "reason": "Flash-attention training degeneration is unrelated to incorrect checkpoint saving in Qwen3.5." }, { "left": "issue:44291", "right": "issue:45071", "accept": false, - "reason": "Unexpected _is_hf_initialized arg during init_empty_weights vs PretrainedConfig type checking; different failures." + "reason": "init_empty_weights argument mismatch and PretrainedConfig type-checking breakage are separate v5 regressions." }, { "left": "issue:44484", "right": "issue:45310", "accept": false, - "reason": "A question about max_shard_size is not the same as a Qwen3.5 MoE loading error." + "reason": "A save_pretrained shard-size question is unrelated to the Qwen3.5 MoE loading error." }, { "left": "issue:42222", "right": "issue:45542", "accept": false, - "reason": "VitPose model breakage and TensorBoard-without-TF backend errors are unrelated." + "reason": "Different subsystems entirely: VitPose model breakage vs missing TensorFlow backend when only tensorboard is installed." }, { "left": "issue:43408", "right": "issue:43742", "accept": false, - "reason": "Model-type warning for sam3_tracker vs key error loading MobileLLM-125M; different issues." - }, - { - "left": "issue:43937", - "right": "issue:44496", - "accept": false, - "reason": "Two different model loading problems: invalid generation config vs unrecognized model/config missing model_type." + "reason": "SAM3 model-type warning and MobileLLM load key error are not the same bug." }, { - "left": "issue:42175", - "right": "issue:42222", + "left": "issue:44075", + "right": "issue:45656", "accept": false, - "reason": "Dependency installation/backends issue vs broken VitPose models; no shared underlying bug." + "reason": "SGD args ignored and optimizer.step() called twice under deepspeed are different optimizer bugs." }, { "left": "issue:44483", "right": "issue:45464", "accept": false, - "reason": "Both mention chat/completions, but one is a critical API request rejection and the other is streaming inference failure on a specific model." - }, - { - "left": "issue:44075", - "right": "issue:45656", - "accept": false, - "reason": "Optimizer args being ignored is different from optimizer.step() being called twice under deepspeed." + "reason": "General chat/completions API failure is not the same as the Qwen3.5 streaming-specific failure." }, { "left": "issue:44964", "right": "issue:45278", "accept": false, - "reason": "Specific Phi-4 multimodal loading failure vs a general report of many import errors after upgrade." + "reason": "Model-loading failure for Phi-4 multimodal is unrelated to the broad import-error report." }, { "left": "issue:43012", "right": "issue:43577", "accept": false, - "reason": "Both involve dtype/precision, but one is a compile-time bf16 warning and the other is incorrect model dtype after loading Blip2." + "reason": "One is a PyTorch warning during bf16 compilation; the other is incorrect dtype state when loading BLIP2." }, { "left": "issue:43066", "right": "issue:43122", "accept": false, - "reason": "Tokenizer decoder type mismatch and changed tokenization output are related to v5, but not clearly the same concrete bug." + "reason": "Both are tokenizer-v5 regressions, but decoder-type mismatch and changed tokenization output are not the same concrete bug." } ] }, @@ -12721,8 +12766,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -12740,7 +12785,6 @@ "issue:43408", "issue:43441", "issue:43452", - "issue:43577", "issue:43874", "issue:43906", "issue:43931", @@ -12767,6 +12811,7 @@ "issue:44987", "issue:45071", "issue:45103", + "issue:45200", "issue:45245", "issue:45341", "issue:45375", @@ -12778,8 +12823,8 @@ "soft_pairs": [ "issue:42371|issue:44261", "issue:38617|issue:44220", - "issue:43874|issue:44261", "issue:44410|issue:45478", + "issue:43874|issue:44261", "issue:44484|issue:45478", "issue:43097|issue:44493", "issue:44263|issue:44485", @@ -12787,9 +12832,9 @@ "issue:43012|issue:43408", "issue:43441|issue:44829", "issue:41628|issue:45071", - "issue:42371|issue:42831", - "issue:45071|issue:45542", "issue:42673|issue:44977", + "issue:45071|issue:45542", + "issue:42371|issue:42831", "issue:42673|issue:43906", "issue:43931|issue:44038", "issue:44117|issue:44987", @@ -12807,257 +12852,263 @@ "issue:43097|issue:45588", "issue:44062|issue:44117", "issue:44038|issue:44410", - "issue:42673|issue:43577", + "issue:42673|issue:45200", "issue:44964|issue:45071", "issue:44704|issue:45103", - "issue:43931|issue:44410", "issue:44261|issue:44485", + "issue:43931|issue:44410", "issue:44964|issue:45341" ], - "prepared_review_unit_hash": "5e824c0973ecd4ad487ffcae8f23bc9e844b2d8ba868bd6ebdfb81c4c5449f5e", + "prepared_review_unit_hash": "e0e09981cbd70bb07b71ecfed1ed921744b12c27e3b1d6beec78ebc61d22e8a7", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 37, - "serialized_chars": 38783, - "estimated_input_tokens": 9696, - "estimated_eval_tokens": 19648 + "serialized_chars": 38752, + "estimated_input_tokens": 9688, + "estimated_eval_tokens": 19632 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These soft pairs are mostly thematic, not true duplicates: import errors, model-specific bugs, precision warnings, and trainer/config regressions are distinct code paths. No PRs are present. If a single issue must represent the cluster, 44261 is the best fit because it is a concrete, well-scoped numerical bug with multiple nearby soft links.", + "summary": "This is a heterogeneous cluster of unrelated Transformers bug reports; none of the soft pairs look like the same underlying defect, so no duplicate merge is warranted.", "confidence": 0.71, - "canonical_issue_reason": "issue:44261 is the most substantive and central of the provided issues, with the most soft links and a narrowly described precision/numerics bug; it is a better representative than the more generic import/error reports.", + "canonical_issue_reason": "No single canonical issue fits this cluster: it mixes unrelated import errors, model-specific regressions, config warnings, and training/runtime bugs.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44261 is the best representative issue overall: it is concrete, reproducible, and closest to multiple adjacent precision-related reports, even though the cluster is not truly duplicate-heavy.", + "best_issue_reason": "issue:44704 is the cleanest representative because it is a concrete core API bug with a clear failure mode and broader potential impact.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42371", "right": "issue:44261", "accept": false, - "reason": "TF32 backend-setting discussion and MLA rms_norm_eps precision bug are different features and failure modes." + "reason": "TF32 settings guidance is unrelated to the MLA rms_norm_eps precision bug." }, { "left": "issue:38617", "right": "issue:44220", "accept": false, - "reason": "Two unrelated errors in different code paths: an import error vs an audio feature extraction issue." + "reason": "Different import/pathology: a missing config symbol vs an audio feature-extraction issue." }, { - "left": "issue:43874", - "right": "issue:44261", + "left": "issue:44410", + "right": "issue:45478", "accept": false, - "reason": "Missing image-patch method in a vision processor is unrelated to the MLA layernorm precision/config problem." + "reason": "Different Qwen3 model families and different failure modes." }, { - "left": "issue:44410", - "right": "issue:45478", + "left": "issue:43874", + "right": "issue:44261", "accept": false, - "reason": "Different Qwen model families and different failures; not the same concrete bug." + "reason": "Image patch-count method bug is unrelated to the MLA precision/config issue." }, { "left": "issue:44484", "right": "issue:45478", "accept": false, - "reason": "Save-pretrained shard sizing is unrelated to Qwen3.5 MoE from_pretrained loading failures." + "reason": "Save-pretrained shard sizing is unrelated to Qwen3.5 MoE loading." }, { "left": "issue:43097", "right": "issue:44493", "accept": false, - "reason": "One is an embeddings-API removal issue; the other is a position-id warning. Different regressions." + "reason": "Removed tie_embeddings API warning is not the same as unexpected position-id keys." }, { "left": "issue:44263", "right": "issue:44485", "accept": false, - "reason": "Torch split/indexing behavior and GLM-5 RoPE math are different bugs, not one shared code-path issue." + "reason": "Both are GLM-related, but they describe different code paths and bugs." }, { "left": "issue:42222", "right": "issue:45459", "accept": false, - "reason": "VitPose model breakage and tokenizer protobuf error masking are unrelated subsystems." + "reason": "VitPose breakage is unrelated to tokenizer protobuf error masking." }, { "left": "issue:43012", "right": "issue:43408", "accept": false, - "reason": "A PyTorch bfloat16 warning and a SAM3 model-type warning are not the same bug." + "reason": "bfloat16 compile warning and sam3 model-type warning are different issues." }, { "left": "issue:43441", "right": "issue:44829", "accept": false, - "reason": "Both mention flash attention, but they affect different models and symptoms, so they are not mergeable as one PR/bug." + "reason": "Both mention FlashAttention, but one is inference failure and the other is degenerate training." }, { "left": "issue:41628", "right": "issue:45071", "accept": false, - "reason": "AutoImageProcessor import failure and PreTrainedConfig type-checking regression are unrelated." + "reason": "Missing AutoImageProcessor import is unrelated to PretrainedConfig type checking." }, { - "left": "issue:42371", - "right": "issue:42831", + "left": "issue:42673", + "right": "issue:44977", "accept": false, - "reason": "TF32 API guidance and FineGrainedFP8 accuracy problems are distinct precision topics, not the same bug." + "reason": "VRAM leak in dataloader threads is not the same as flash-attention generation failure." }, { "left": "issue:45071", "right": "issue:45542", "accept": false, - "reason": "Type-checking regression and undefined tf backend error are unrelated failure modes." + "reason": "Type-checking regression is unrelated to the TensorFlow backend import error." }, { - "left": "issue:42673", - "right": "issue:44977", + "left": "issue:42371", + "right": "issue:42831", "accept": false, - "reason": "VRAM leakage in dataloader threads and flash-attention generation failure are different Qwen3 issues." + "reason": "TF32 settings control is unrelated to FineGrainedFP8 accuracy." }, { "left": "issue:42673", "right": "issue:43906", "accept": false, - "reason": "No evidence these describe the same underlying bug; one is a memory leak, the other a reproduction of a different issue." + "reason": "These are different Qwen3-related reports with no shared concrete failure path." }, { "left": "issue:43931", "right": "issue:44038", "accept": false, - "reason": "Both involve Qwen3-VL, but one is a weight-shape mismatch and the other is a model-loading bug for a different variant." + "reason": "Different model variants and different loading failures." }, { "left": "issue:44117", "right": "issue:44987", "accept": false, - "reason": "Tokenizer mapping returning None and a physical-intelligence/fast load failure are not the same code-path problem." + "reason": "Tokenizer mapping returning None is not clearly the same as that model-loading failure." }, { "left": "issue:43097", "right": "issue:43452", "accept": false, - "reason": "Embeddings/tie-weights removal and gguf_file loading errors are unrelated." + "reason": "API removal warning and gguf_file loading breakage are different bugs." }, { "left": "issue:43335", "right": "issue:45071", "accept": false, - "reason": "SwitchTransformers sparse-layer config bug and PreTrainedConfig type checking are different areas." + "reason": "Sparse-layer config creation bug is unrelated to type-checking behavior." }, { "left": "issue:43976", "right": "issue:44945", "accept": false, - "reason": "Python-version compatibility and pipeline-parallel output corruption are unrelated issues." + "reason": "Python-version compatibility and pipeline-parallel output corruption are unrelated." }, { "left": "issue:43408", "right": "issue:44368", "accept": false, - "reason": "A model-type mismatch warning and a tie_word_embeddings warning are different config warnings." + "reason": "Different warnings in different models; not the same underlying bug." }, { "left": "issue:43874", "right": "issue:44263", "accept": false, - "reason": "Vision-processor API omission and torch.split return handling are unrelated." + "reason": "Both mention GLM, but one is a missing processor method and the other is a torch.split handling issue." }, { "left": "issue:43986", "right": "issue:44484", "accept": false, - "reason": "AutoProcessor torchvision dependency crash is unrelated to save_pretrained shard sizing." + "reason": "AutoProcessor torchvision crash is unrelated to save_pretrained sharding." }, { "left": "issue:44829", "right": "issue:45245", "accept": false, - "reason": "Flash-attention training degradation and category-count limit are unrelated." + "reason": "FlashAttention training degeneration is unrelated to category-cardinality limits." }, { "left": "issue:38617", "right": "issue:42757", "accept": false, - "reason": "Different import errors from different symbols; same symptom class, but not the same bug." + "reason": "Different missing imports from different modules/symbols." }, { "left": "issue:43064", "right": "issue:44315", "accept": false, - "reason": "Wrong optimizer-state handling in FSDP2/PEFT and Liger Kernel application with model_init are different training bugs." + "reason": "Wrong optimizer state distribution is not the same as Liger Kernel not being applied." }, { "left": "issue:44410", "right": "issue:45375", "accept": false, - "reason": "Missing qkv/gate projections and a missing config field in vision MoE are not the same issue." + "reason": "Missing projections in qwen3next is unrelated to a silently dropped config field." }, { "left": "issue:38617", "right": "issue:44351", "accept": false, - "reason": "ImportError for layer_type_validation and HybridCache are separate missing-symbol regressions." + "reason": "Both are import errors, but they involve different missing symbols and code paths." }, { "left": "issue:43097", "right": "issue:45588", "accept": false, - "reason": "Embeddings tie-weights API removal and flash_attention.py s_aux handling are unrelated." + "reason": "Config API removal and flash_attention.py s_aux=None crash are different bugs." }, { "left": "issue:44062", "right": "issue:44117", "accept": false, - "reason": "Tokenizer AddedToken construction error and tokenizer-mapping None handling are different tokenizer bugs." + "reason": "Token AddedToken kwarg duplication is unrelated to tokenizer mapping None handling." }, { "left": "issue:44038", "right": "issue:44410", "accept": false, - "reason": "Qwen3-VL-Moe loading/bug and qwen3next missing projections are different model implementations and failure modes." + "reason": "Different Qwen3-family model issues with different symptoms." }, { "left": "issue:42673", - "right": "issue:43577", + "right": "issue:45200", "accept": false, - "reason": "Memory leak in Qwen3ForCausalLM and Blip2 dtype propagation are unrelated." + "reason": "VRAM leak in multi-thread loading is unrelated to Gemma 4 mm_token_type_ids defaults." }, { "left": "issue:44964", "right": "issue:45071", "accept": false, - "reason": "Phi-4 multimodal loading failure and PreTrainedConfig type-checking regression are not the same code-path issue." + "reason": "Phi-4 multimodal load failure is unrelated to a type-checking regression." }, { "left": "issue:44704", "right": "issue:45103", "accept": false, - "reason": "AutoProcessor kwargs forwarding and auto_docstring annotation handling are unrelated." + "reason": "cached_file kwargs forwarding and auto_docstring annotation crashes are unrelated." }, { "left": "issue:44261", "right": "issue:44485", "accept": false, - "reason": "Layernorm epsilon precision in MLA and GLM-5 RoPE implementation are different numerical bugs." + "reason": "MLA layernorm epsilon precision and GLM-5 RoPE implementation are different bugs." + }, + { + "left": "issue:43931", + "right": "issue:44410", + "accept": false, + "reason": "Qwen3-VL shape mismatch and qwen3next missing projections are separate model issues." }, { "left": "issue:44964", "right": "issue:45341", "accept": false, - "reason": "Model loading failure and a testing_utils bug are unrelated." + "reason": "Phi-4 multimodal loading is unrelated to a testing_utils bug." } ] }, @@ -13076,8 +13127,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -13131,8 +13182,8 @@ ], "soft_pairs": [ "issue:43502|issue:43992", - "issue:43986|issue:44117", "issue:45341|issue:45859", + "issue:43986|issue:44117", "issue:43097|issue:43873", "issue:43066|issue:43906", "issue:43986|issue:45290", @@ -13155,8 +13206,8 @@ "issue:43012|issue:44683", "issue:42757|issue:43824", "issue:38617|issue:43502", - "issue:44038|issue:44230", "issue:44483|issue:45290", + "issue:44038|issue:44230", "issue:43618|issue:43761", "issue:43064|issue:44380", "issue:43299|issue:43867", @@ -13166,20 +13217,20 @@ "issue:43867|issue:45216", "issue:42994|issue:44655" ], - "prepared_review_unit_hash": "ba1a159d1039fa4457baaca0dcba78c36971ee3b1874dcd87d0a962b88881b7f", + "prepared_review_unit_hash": "f958b5cdfbb50b31faa9c1d4ce91eab86b9575f3707d780adf56079b2aeb15f2", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 35, - "serialized_chars": 37417, + "serialized_chars": 37419, "estimated_input_tokens": 9355, "estimated_eval_tokens": 18966 }, @@ -13188,222 +13239,222 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated Transformers issues that only share high-level themes like imports, loading, saving, quantization, and v5 regressions. I don\u2019t see any soft pair that is clearly the same underlying bug or change.", - "confidence": 0.43, - "canonical_issue_reason": "issue:43873 is the best umbrella candidate: it is open, has substantial discussion/inbound references, and describes a broad quantization/offloading failure mode that can plausibly collect related reports better than the more narrow, model-specific issues.", + "summary": "No soft pair looks like a safe duplicate merge; the set is mostly unrelated regression reports across different models and subsystems.", + "confidence": 0.68, + "canonical_issue_reason": "No safe canonical issue emerges here: the candidate pairs are mostly unrelated, so choosing one as the duplicate anchor would be arbitrary.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 is the strongest global issue choice because it is active, broadly phrased, and sits in a recurring failure area (quantization/offloading) that could absorb adjacent reports if a canonical issue is needed.", + "best_issue_reason": "issue:43873 is the strongest standalone anchor among these issues: it is open, broad in scope (quantization/offloading), and has the most surrounding discussion/references.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43502", "right": "issue:43992", "accept": false, - "reason": "Both involve loading/saving behavior, but one is local_files_only network leakage and the other is a missing weight during UMT5 loading; different code paths and fixes." + "reason": "Different bugs: local_files_only network leakage vs a missing embed_tokens weight during UMT5 loading." }, { - "left": "issue:43986", - "right": "issue:44117", + "left": "issue:45341", + "right": "issue:45859", "accept": false, - "reason": "Different failures: video AutoProcessor setup without torchvision vs tokenizer mapping returning None during from_pretrained." + "reason": "Unrelated subsystems: a testing_utils bug vs Qwen3_5MoeTextRotaryEmbedding CPU offload compatibility." }, { - "left": "issue:45341", - "right": "issue:45859", + "left": "issue:43986", + "right": "issue:44117", "accept": false, - "reason": "Testing-utils bug versus Qwen3_5Moe CPU-offload incompatibility; unrelated subsystems and symptoms." + "reason": "Both are loading-related, but one is a missing torchvision dependency path and the other is a tokenizer mapping None-handling bug." }, { "left": "issue:43097", "right": "issue:43873", "accept": false, - "reason": "One is an API removal/documentation issue, the other is a quantization/offloading runtime bug; not the same underlying problem." + "reason": "Different problems: a removed API warning vs quantization/offloading behavior." }, { "left": "issue:43066", "right": "issue:43906", "accept": false, - "reason": "Wrong tokenizer decoder type and an isolated reproduction of a different issue are not enough to establish the same concrete bug." + "reason": "No clear match: tokenizer decoder type regression vs an isolated reproduction of a different issue." }, { "left": "issue:43986", "right": "issue:45290", "accept": false, - "reason": "Both touch multimodal/chat processing, but they fail in different places and with different preconditions." + "reason": "Different areas: AutoProcessor video-model import failure vs chat template crashes on tool-call assistant messages." }, { "left": "issue:45341", "right": "issue:45561", "accept": false, - "reason": "A test utility bug and an xdist file-race are related only by testing infrastructure, not the same defect." + "reason": "Shared file name is not enough; one is a generic testing_utils bug and the other is a specific xdist race on captured_info.txt." }, { "left": "issue:43986", "right": "issue:45372", "accept": false, - "reason": "Different missing dependency/import problems in different processor/model-loading paths." + "reason": "Different import failures from different dependencies and models." }, { "left": "issue:43097", "right": "issue:43901", "accept": false, - "reason": "Removed embedding-tie API versus stale docs about return_all_scores; separate issue classes." + "reason": "Completely unrelated: removed tie_embeddings API vs stale pipeline docs about return_all_scores." }, { "left": "issue:42617", "right": "issue:45341", "accept": false, - "reason": "A script execution problem and a testing-utils bug are unrelated beyond being bugs in the repo." + "reason": "Different failures: 3d_parallel.py execution vs a testing_utils bug." }, { "left": "issue:43986", "right": "issue:45341", "accept": false, - "reason": "Video processor import/setup failure versus testing-utils behavior; no shared code path." + "reason": "Different domains: video-model loading/import path vs testing infrastructure." }, { "left": "issue:43825", "right": "issue:44655", "accept": false, - "reason": "Pipeline task messaging and pipeline save_pretrained are distinct issues affecting different APIs." + "reason": "Different surface areas: pipeline task messaging vs saving Pipeline objects." }, { "left": "issue:45341", "right": "issue:45588", "accept": false, - "reason": "Testing utilities and flash_attention sink handling are separate code paths with different failure modes." + "reason": "Testing-utils race bug vs a flash_attention AttributeError in sink-less models." }, { "left": "issue:43992", "right": "issue:44484", "accept": false, - "reason": "Missing embed_tokens weight during load is unrelated to the max_shard_size default question in save_pretrained." + "reason": "Different save/load concerns: missing embed weight on load vs max_shard_size default behavior." }, { "left": "issue:43012", "right": "issue:43550", "accept": false, - "reason": "Precision warning during compile and Bamba SDPA compile failure are different runtime behaviors." + "reason": "One is a warning about bfloat16 compilation precision; the other is an actual torch.compile SDPA failure on Bamba." }, { "left": "issue:43208", "right": "issue:43550", "accept": false, - "reason": "xLSTM training bugs versus Bamba torch.compile+SDPA failure; not the same bug." + "reason": "xLSTM training bugs are unrelated to a Bamba SDPA compile regression." }, { "left": "issue:43012", "right": "issue:44368", "accept": false, - "reason": "Both mention warnings/config, but one is a PyTorch precision warning and the other is an embedding-tying warning in a different model family." + "reason": "Both mention warnings, but they are about different configs and different underlying behaviors." }, { "left": "issue:42886", "right": "issue:43381", "accept": false, - "reason": "Offline cache loading and gradient-checkpointing-in-eval are unrelated." + "reason": "Offline cache loading and gradient checkpointing in eval mode are separate bugs." }, { "left": "issue:42907", "right": "issue:45753", "accept": false, - "reason": "Saving dequantized models and an infinite-generation loop on a specific image are different defects." + "reason": "Saving dequantized models and an infinite-generation loop are unrelated." }, { "left": "issue:42907", "right": "issue:44655", "accept": false, - "reason": "Model dequantization save failure is not the same as saving Pipeline objects." + "reason": "Both involve saving, but one is quantized model serialization and the other is Pipeline save_pretrained support." }, { "left": "issue:43208", "right": "issue:43526", "accept": false, - "reason": "xLSTM training issues versus BeitImageProcessorFast label reduction bug; different components." + "reason": "Different components: xLSTM training bugs vs BeitImageProcessorFast reduce_labels handling." }, { "left": "issue:44483", "right": "issue:44514", "accept": false, - "reason": "OpenAI-style chat endpoint rejection and Qwen2_5_VLProcessor batching crash are separate request paths." + "reason": "Different chat-related failures: OpenAI-style endpoint rejection vs batched apply_chat_template crash." }, { "left": "issue:43012", "right": "issue:44683", "accept": false, - "reason": "Both are compile-related, but one is a bf16 warning and the other is a flex_attention failure on newer torch versions." + "reason": "A warning about bfloat16 precision is not the same bug as flex_attention compilation failure on newer torch." }, { "left": "issue:42757", "right": "issue:43824", "accept": false, - "reason": "Both are import errors, but they involve different missing symbols from different modules and different fixes." + "reason": "Both are ImportErrors, but for different missing symbols and different code paths." }, { "left": "issue:38617", "right": "issue:43502", "accept": false, - "reason": "Missing layer_type_validation import and local_files_only making API requests are unrelated issues." + "reason": "Unrelated imports: layer_type_validation removal vs local_files_only still making API calls." }, { - "left": "issue:44038", - "right": "issue:44230", + "left": "issue:44483", + "right": "issue:45290", "accept": false, - "reason": "Qwen3-VL-Moe version breakage and fp8 inference support are not the same concrete bug." + "reason": "Different chat-stack regressions: request validation vs chat template handling of tool-call messages." }, { - "left": "issue:44483", - "right": "issue:45290", + "left": "issue:44038", + "right": "issue:44230", "accept": false, - "reason": "Different user-facing APIs: chat completions endpoint handling versus apply_chat_template tokenization crash." + "reason": "Both mention Qwen3-VL/MoE, but one is a v5 bug report and the other is fp8 inference support." }, { "left": "issue:43618", "right": "issue:43761", "accept": false, - "reason": "Both are CLIP regressions, but one concerns attentions not being assigned and the other hidden_states being None; not the same defect." + "reason": "Both are CLIP regressions, but one is missing attentions and the other is hidden_states=None; not the same concrete failure." }, { "left": "issue:43064", "right": "issue:44380", "accept": false, - "reason": "Trainer/FSDP2 optimizer-state corruption is unrelated to GPT2 attention scaling being ignored under SDPA/FlashAttention." + "reason": "Trainer/FSDP optimizer-state corruption is unrelated to GPT2 attention scaling being ignored under SDPA/FlashAttention." }, { "left": "issue:43299", "right": "issue:43867", "accept": false, - "reason": "A Qwen3VL-Moe loading regression and a generic state_dict-sorting load error do not look like one underlying bug." + "reason": "Different model-loading bugs: Qwen3VL-Moe load failure vs state_dict sorting load error." }, { "left": "issue:44483", "right": "issue:45381", "accept": false, - "reason": "Chat-completions API rejection and Qwen2.5-VL video position-id corruption are different paths and symptoms." + "reason": "OpenAI chat-completions request handling is unrelated to qwen2.5-vl vision_position_ids for video input." }, { "left": "issue:43901", "right": "issue:44655", "accept": false, - "reason": "Documentation drift for return_all_scores does not match a pipeline serialization failure." + "reason": "Docs drift about return_all_scores is unrelated to Pipeline save_pretrained failures." }, { "left": "issue:36683", "right": "issue:43475", "accept": false, - "reason": "Gemma3Config missing vocab_size and SAM3VisionEncoderOutput missing fpn_position_embeddings are unrelated attribute errors." + "reason": "Different missing-attribute errors on different model classes: Gemma3Config vocab_size vs Sam3VisionEncoderOutput fpn_position_embeddings." }, { "left": "issue:43867", "right": "issue:45216", "accept": false, - "reason": "Generic state_dict sorting load failure and Qwen3.5 save_pretrained regression are distinct save/load bugs." + "reason": "Both involve checkpoints, but one is a load-time state_dict ordering error and the other is a save_pretrained regression." }, { "left": "issue:42994", "right": "issue:44655", "accept": false, - "reason": "Quantized model saving failure and saving Pipeline objects are different save_pretrained problems with different object types." + "reason": "Both are save-related, but one is quantized model saving and the other is Pipeline object serialization." } ] }, @@ -13422,8 +13473,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -13483,22 +13534,22 @@ "issue:44220|issue:44655", "issue:44263|issue:45440", "issue:36683|issue:43525", - "issue:43906|issue:45753", "issue:43618|issue:45216", + "issue:43906|issue:45753", "issue:36683|issue:44242", "issue:43299|issue:45753", "issue:36683|issue:43881", "issue:43352|issue:43976", "issue:43441|issue:44380", - "issue:43352|issue:44380", "issue:43992|issue:45310", + "issue:43352|issue:44380", "issue:43867|issue:45375", "issue:44380|issue:45479", "issue:41669|issue:44246", "issue:43352|issue:45478", "issue:43618|issue:44380", - "issue:43064|issue:44492", "issue:44355|issue:44855", + "issue:43064|issue:44492", "issue:36683|issue:44117", "issue:42890|issue:45561", "issue:43097|issue:43352", @@ -13511,238 +13562,238 @@ "issue:44077|issue:44380", "issue:43867|issue:44484" ], - "prepared_review_unit_hash": "c8fbfbfe3bae75b9fdde37ca4a045a597112123365f40c942155cb6e65e1686b", + "prepared_review_unit_hash": "fef0bb11ca8d90e301e94cad4a8a138fb3319207779285c5dd52bd6f9a8653cc", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 34, - "serialized_chars": 36866, - "estimated_input_tokens": 9217, - "estimated_eval_tokens": 18690 + "serialized_chars": 36869, + "estimated_input_tokens": 9218, + "estimated_eval_tokens": 18692 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: most items are unrelated model-specific, training, import, or test issues that only look superficially similar by keyword. I would not merge any of the soft pairs as duplicates. The strongest standalone issue here is the import-time performance regression in issue 41669.", - "confidence": 0.41, - "canonical_issue_reason": "Issue 41669 is the most concrete, actionable, and broadly impactful regression in the set: it describes a measurable import-time slowdown with a plausible single root cause. The rest of the cluster is mostly a mix of unrelated model/config/runtime bugs.", + "summary": "The cluster is heterogeneous: most items are distinct Transformers regressions spanning different models, versions, and subsystems. The soft-similarity edges are mostly false positives rather than true duplicates.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue cleanly subsumes this set; the items split into unrelated model-specific loading, config, attention, and tooling regressions. If forced to pick a representative, issue 43097 is the broadest v5 regression, but it does not duplicate the others.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 41669 is the best anchor for this noisy cluster because it is specific, reproducible, and high-signal compared with the many narrow model-version failures around it.", + "best_issue_reason": "issue:43097 is the best representative issue only because it is a broad, user-facing v5 breakage with clear impact. It is still not a good canonical for the whole cluster, since the rest are mostly separate bugs.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43352", "right": "issue:43937", "accept": false, - "reason": "Different bugs: FlashAttention support for a model class vs a generation config validation error." + "reason": "Different failures: FlashAttention support gap for Nemotron vs invalid GenerationConfig for GLM-5." }, { "left": "issue:36010", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but they involve different packages and different missing symbols." + "reason": "Both are import errors, but for different symbols from different packages and unrelated code paths." }, { "left": "issue:43066", "right": "issue:45381", "accept": false, - "reason": "Tokenizer decoder type mismatch and Qwen2.5-VL vision position ids are unrelated failures." + "reason": "Wrong tokenizer decoder type vs Qwen2.5-VL video position ids; different model and bug." }, { "left": "issue:43756", "right": "issue:44485", "accept": false, - "reason": "Different model families and different problems: RoPE layer count vs RoPE implementation details." + "reason": "Smollm3 RoPE layer mismatch vs GLM-5 RoPE implementation discussion; distinct model behaviors." }, { "left": "issue:44220", "right": "issue:44655", "accept": false, - "reason": "Audio feature extraction bug and pipeline save_pretrained behavior are unrelated." + "reason": "Audio feature extraction issue vs pipeline save_pretrained failure; unrelated subsystems." }, { "left": "issue:44263", "right": "issue:45440", "accept": false, - "reason": "GlmMoeDsaIndexer torch.split behavior and DeepseekV3MoE divergence are distinct code paths." + "reason": "DSA torch.split behavior vs DeepseekV3MoE divergence; not the same bug." }, { "left": "issue:36683", "right": "issue:43525", "accept": false, - "reason": "Two separate config-attribute omissions on different models; not the same bug." + "reason": "Gemma3 missing vocab_size vs Llama4 missing pad_token_id; same symptom class, different model/config fields." }, { - "left": "issue:43906", - "right": "issue:45753", + "left": "issue:43618", + "right": "issue:45216", "accept": false, - "reason": "An isolated reproduction of an older issue is not the same as a Qwen3.5 infinite-loop regression." + "reason": "CLIPOutput attentions omission vs Qwen3.5 save_pretrained regression; unrelated fixes." }, { - "left": "issue:43618", - "right": "issue:45216", + "left": "issue:43906", + "right": "issue:45753", "accept": false, - "reason": "Missing CLIP attentions and incorrect Qwen3.5 checkpoint saving are different regressions." + "reason": "Isolated reproduction of a prior issue vs Qwen3_5 infinite loop on a specific image; not duplicates." }, { "left": "issue:36683", "right": "issue:44242", "accept": false, - "reason": "Missing Gemma3Config attribute and load-balancing-loss omission are unrelated." + "reason": "Gemma3 config attribute error vs load balancing loss omission in MoE routing; different code paths." }, { "left": "issue:43299", "right": "issue:45753", "accept": false, - "reason": "Qwen3VL MoE loading failure and Qwen3_5 image infinite loop are different symptoms and likely different causes." + "reason": "Qwen3VL MoE model loading breakage vs Qwen3_5 image-specific infinite loop; separate problems." }, { "left": "issue:36683", "right": "issue:43881", "accept": false, - "reason": "Gemma3Config attribute error and GLM-4V-9B load failure are not the same underlying issue." + "reason": "Gemma3 config attribute error vs GLM-4V-9B loading failure; different models and failures." }, { "left": "issue:43352", "right": "issue:43976", "accept": false, - "reason": "FlashAttention support bug and Python-version compatibility failure are unrelated." + "reason": "FlashAttention support bug vs Python version compatibility issue; unrelated." }, { "left": "issue:43441", "right": "issue:44380", "accept": false, - "reason": "Ministral FlashAttention failure and GPT2 attention-scaling configuration are separate code paths." + "reason": "Ministral FlashAttention failure vs GPT2 attention scaling ignored under SDPA/FlashAttention; different models and fixes." }, { - "left": "issue:43352", - "right": "issue:44380", + "left": "issue:43992", + "right": "issue:45310", "accept": false, - "reason": "Different models and different behaviors; one is unsupported FlashAttention, the other is ignored scaling config." + "reason": "Missing embed_tokens.weight in UMT5Encoder vs Qwen3.5 Moe from_pretrained error; distinct loading bugs." }, { - "left": "issue:43992", - "right": "issue:45310", + "left": "issue:43352", + "right": "issue:44380", "accept": false, - "reason": "UMT5 weight loading and Qwen3.5 MoE from_pretrained failure do not match." + "reason": "Both involve attention backends, but one is unsupported model/backend, the other ignores a config scaling parameter." }, { "left": "issue:43867", "right": "issue:45375", "accept": false, - "reason": "State_dict ordering/load error and a missing config field dropped by @strict are not the same bug." + "reason": "State_dict ordering load error vs missing Qwen3_5MoeVisionConfig field; different failure mechanisms." }, { "left": "issue:44380", "right": "issue:45479", "accept": false, - "reason": "Attention scaling in GPT2 and a degenerate zero-loss classification bug are unrelated." + "reason": "Attention scaling backend bug vs degenerate zero-loss classification bug; unrelated." }, { "left": "issue:41669", "right": "issue:44246", "accept": false, - "reason": "Both mention import slowness, but one is a specific import-star regression and the other is a general intermittent slowdown report." + "reason": "Import-time slowdown from import * usage vs intermittent import slowdown; same broad symptom, but not the same concrete bug." }, { "left": "issue:43352", "right": "issue:45478", "accept": false, - "reason": "Same broad theme, but different models and distinct failures." + "reason": "Different Qwen/Nemotron style issue vs Qwen3.5 Moe from_pretrained regression; unrelated model-specific breakages." }, { "left": "issue:43618", "right": "issue:44380", "accept": false, - "reason": "Missing attentions output and attention scaling being ignored are different issues." + "reason": "CLIP attentions field regression vs GPT2 attention scaling ignored; different affected code and behavior." }, { - "left": "issue:43064", - "right": "issue:44492", + "left": "issue:44355", + "right": "issue:44855", "accept": false, - "reason": "A distributed training optimizer-state bug and a cache-strategy typo are not duplicates." + "reason": "Compiled Python files import errors vs Python 3.13 IndentationError in DebertaV2Model; distinct root causes." }, { - "left": "issue:44355", - "right": "issue:44855", + "left": "issue:43064", + "right": "issue:44492", "accept": false, - "reason": "Compiled-file import errors and a Python 3.13 decorator/comment parsing failure are different root causes." + "reason": "FSDP2 optimizer state corruption vs a typo in cache strategies; unrelated." }, { "left": "issue:36683", "right": "issue:44117", "accept": false, - "reason": "Config attribute missing and tokenizer mapping returning None are unrelated." + "reason": "Model config missing vocab_size vs tokenizer mapping sometimes returning None; different layers of the stack." }, { "left": "issue:42890", "right": "issue:45561", "accept": false, - "reason": "Missing set_seed in tests and xdist file-race corruption are both test issues but not the same bug." + "reason": "Flaky seed-dependent integration tests vs xdist race on captured_info.txt; both test issues but not the same bug." }, { "left": "issue:43097", "right": "issue:43352", "accept": false, - "reason": "A removed config field and FlashAttention support are completely different topics." + "reason": "Broad removed-API regression vs model-specific FlashAttention support issue; not one underlying change." }, { "left": "issue:43867", "right": "issue:45357", "accept": false, - "reason": "Both involve loading/saving state, but the concrete failure modes differ." + "reason": "Generic state_dict loading error vs Qwen3.5 visual encoder key regression; different failure path." }, { "left": "issue:43976", "right": "issue:44938", "accept": false, - "reason": "These are separate Python-version compatibility reports affecting different releases and environments." + "reason": "Python 3.9 compatibility regression vs Python 3.14 load failure; version issues only, not same bug." }, { "left": "issue:43992", "right": "issue:45478", "accept": false, - "reason": "Different model families and different missing-weight / from_pretrained failures." + "reason": "UMT5Encoder missing weight vs Qwen3.5 Moe from_pretrained error; unrelated loading bugs." }, { "left": "issue:44360", "right": "issue:45440", "accept": false, - "reason": "DSA indexer math and DeepSeek MoE implementation divergence are unrelated." + "reason": "Indexer missing ReLU vs DeepseekV3MoE divergence; different algorithmic issues." }, { "left": "issue:43064", "right": "issue:44655", "accept": false, - "reason": "Distributed optimizer-state corruption and Pipeline save_pretrained support are separate areas." + "reason": "Optimizer-state corruption in training vs pipeline save_pretrained failure; unrelated subsystems." }, { "left": "issue:36296", "right": "issue:43575", "accept": false, - "reason": "Both mention tensor parallelism, but one is a training bug and the other is a loading/OOM report." + "reason": "Tensor-parallel training bug vs Qwen2-57B-A14B-Instruct OOM on load; different symptoms and likely causes." }, { "left": "issue:44077", "right": "issue:44380", "accept": false, - "reason": "PatchTSMixer post_init constraints and GPT2 attention scaling are not the same code-path problem." + "reason": "Optional post_init policy change vs GPT2 attention scaling backend bug; unrelated." }, { "left": "issue:43867", "right": "issue:44484", "accept": false, - "reason": "State_dict load-order failure and the default max_shard_size question are unrelated." + "reason": "State_dict loading error vs save_pretrained shard size question; not the same bug." } ] }, @@ -13761,8 +13812,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -13828,8 +13879,8 @@ "issue:43502|issue:44062", "issue:43010|issue:44075", "issue:42890|issue:45341", - "issue:44222|issue:45125", "issue:42548|issue:44351", + "issue:44222|issue:45125", "issue:43502|issue:44485", "issue:44355|issue:44749", "issue:45125|issue:45478", @@ -13847,20 +13898,20 @@ "issue:42831|issue:43874", "issue:30333|issue:36296", "issue:44355|issue:45071", - "issue:44355|issue:44655", "issue:44222|issue:44230", + "issue:44355|issue:44655", "issue:43827|issue:44961", "issue:33453|issue:34689", "issue:42757|issue:43576" ], - "prepared_review_unit_hash": "d8f28d6806844c6ebf80142298da36d9466144be7a7039775e88452e7bc784a4", + "prepared_review_unit_hash": "a36944e34c21615e3afea26aca77962ef1353149f7aa24154a38ffae30e94db3", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, @@ -13875,228 +13926,228 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a grab-bag of unrelated issues across loading, modeling, tests, docs, and performance. None of the soft pairs are clearly the same underlying bug/change from titles alone.", - "confidence": 0.56, - "canonical_issue_reason": "Issue 45278 is the broadest umbrella for the visible v5 import-regression theme, so it is the most useful single issue to anchor this set.", + "summary": "This cluster is mostly a loose collection of Transformers v5-era regression/import/load issues, but the proposed soft pairs generally point to different concrete bugs. I would use the broad import-regression report as the representative issue, and reject all soft edges as duplicates.", + "confidence": 0.83, + "canonical_issue_reason": "issue:45278 is the broadest and most representative of the cluster\u2019s recurring theme: breakages after upgrading to v5, especially import/loading regressions.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45278 is the best global tracker here because it is open, recent, and broad enough to subsume the generic import-failure cluster better than the narrower reports.", + "best_issue_reason": "issue:45278 best captures the dominant, user-facing problem pattern in this set and is broad enough to serve as the cluster anchor.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:28282", "right": "issue:33453", "accept": false, - "reason": "Different failures: PyTorch missing import vs tokenizer loading regression." + "reason": "Different failure classes: missing PyTorch dependency vs tokenizer loading regression." }, { "left": "issue:43618", "right": "issue:44857", "accept": false, - "reason": "Unrelated model/output bug versus AMP/CUDA crash in a different loss path." + "reason": "Unrelated symptoms and code paths; CLIP output attribute handling is not the same as LwDetr AMP crash." }, { "left": "issue:43618", "right": "issue:44977", "accept": false, - "reason": "Different code paths: CLIP output attentions vs Qwen3.5 flash-attention generation." + "reason": "Same broad model family only; one is CLIPOutput attentions, the other is Qwen3.5 flash-attention generation." }, { "left": "issue:43618", "right": "issue:45588", "accept": false, - "reason": "Different failures in different attention code paths; not the same bug." + "reason": "Both touch attention code, but one is missing attentions assignment and the other is an s_aux=None crash in flash_attention.py." }, { "left": "issue:43901", "right": "issue:44938", "accept": false, - "reason": "Docs mismatch for a pipeline argument vs Python 3.14 load failure." + "reason": "Documentation mismatch vs Python 3.14 import/load failure; not the same bug." }, { "left": "issue:44655", "right": "issue:44938", "accept": false, - "reason": "Saving Pipeline objects is unrelated to Python 3.14 import/load issues." + "reason": "Saving pipeline objects and loading transformers on Python 3.14 are different issues." }, { "left": "issue:42890", "right": "issue:44964", "accept": false, - "reason": "Test seeding flake versus inability to load a specific multimodal model." + "reason": "Test seeding flakiness is unrelated to a multimodal model load failure." }, { "left": "issue:42890", "right": "issue:44112", "accept": false, - "reason": "Both are test failures, but they describe different fixtures and root causes." + "reason": "Both are CI/test problems, but they affect different tests and different causes." }, { "left": "issue:42890", "right": "issue:45600", "accept": false, - "reason": "Different CI regressions: seed flakiness vs stale auto-mapping references." + "reason": "Test seed flakiness is not the same as stale auto_mappings config references breaking CI." }, { "left": "issue:43502", "right": "issue:44062", "accept": false, - "reason": "Offline-mode request bug is unrelated to AddedToken keyword collision." + "reason": "local_files_only network leakage and AddedToken keyword collision are unrelated." }, { "left": "issue:43010", "right": "issue:44075", "accept": false, - "reason": "No-grad cache update bug is unrelated to optimizer argument handling." + "reason": "Cache no_grad decoration and SGD args not being used are different code paths." }, { "left": "issue:42890", "right": "issue:45341", "accept": false, - "reason": "Test seed flakiness is not the same as a helper bug in testing_utils.py." + "reason": "A missing set_seed in tests is not the same as a separate testing_utils bug." }, { - "left": "issue:44222", - "right": "issue:45125", + "left": "issue:42548", + "right": "issue:44351", "accept": false, - "reason": "FP8 save_pretrained issue and missing tensor-parallel plan are different model bugs." + "reason": "Both are import errors, but for different symbols and likely different export fixes." }, { - "left": "issue:42548", - "right": "issue:44351", + "left": "issue:44222", + "right": "issue:45125", "accept": false, - "reason": "Both are import errors, but for different symbols and likely different breakages." + "reason": "FP8 save_pretrained bug and missing tensor-parallel plan are distinct model/serialization problems." }, { "left": "issue:43502", "right": "issue:44485", "accept": false, - "reason": "Local-files-only network leak is unrelated to GLM-5 RoPE implementation." + "reason": "local_files_only leakage is unrelated to GLM-5 RoPE implementation." }, { "left": "issue:44355", "right": "issue:44749", "accept": false, - "reason": "Compiled-file errors and slowdown after upgrade are not the same defect." + "reason": "Compiled Python file errors and slowed filtering after upgrade are different regressions." }, { "left": "issue:45125", "right": "issue:45478", "accept": false, - "reason": "Same model family, but tensor-parallel metadata and from_pretrained loading are distinct paths." + "reason": "Both involve Qwen3.5 MoE, but one is a tensor-parallel plan issue and the other is a broader from_pretrained failure." }, { "left": "issue:30990", "right": "issue:31795", "accept": false, - "reason": "Both mention loading/docs, but one is Sentence Transformers hanging and the other is documentation confusion." + "reason": "Sentence Transformers loading hang and forward() docs confusion are unrelated." }, { "left": "issue:45125", "right": "issue:45375", "accept": false, - "reason": "Missing _tp_plan and missing config field are different Qwen3.5 MoE issues." + "reason": "Missing tensor-parallel plan and missing config field are different Qwen3.5 MoE defects." }, { "left": "issue:44387", "right": "issue:44556", "accept": false, - "reason": "Quantization memory regression is not the same as checkpoint reload incompatibility." + "reason": "Int4 CUDA memory/OOM behavior is not the same as checkpoint reload failure after upgrade." }, { "left": "issue:30333", "right": "issue:33453", "accept": false, - "reason": "MLflow reporting on failed training jobs is unrelated to tokenizer loading regression." + "reason": "MLflow reporting after failed training and tokenizer loading regression are unrelated." }, { "left": "issue:43010", "right": "issue:45278", "accept": false, - "reason": "Cache no_grad behavior is unrelated to broad import failures after upgrade." + "reason": "A cache update no_grad request is not the same as broad import errors after v5.5.0." }, { "left": "issue:43723", "right": "issue:44556", "accept": false, - "reason": "Tokenizer loading and checkpoint reload problems overlap only loosely; likely different bugs." + "reason": "Tokenizer loading regression and checkpoint reload regression may rhyme, but they are not the same concrete bug." }, { "left": "issue:43502", "right": "issue:44222", "accept": false, - "reason": "Network-blocking bug and FP8 save_pretrained bug affect different code paths." + "reason": "local_files_only network calls and FP8 save_pretrained for MoE are unrelated." }, { "left": "issue:43010", "right": "issue:44368", "accept": false, - "reason": "Cache update gradient handling is unrelated to a tied-embeddings warning." + "reason": "No_grad on cache updates and a tied-embeddings warning in LoRA fine-tuning are different issues." }, { "left": "issue:44315", "right": "issue:44360", "accept": false, - "reason": "Liger Kernel application bug and DSA indexer ReLU discussion are unrelated." + "reason": "Liger Kernel application and a DSA indexer ReLU discussion have no shared bug path." }, { "left": "issue:43010", "right": "issue:45850", "accept": false, - "reason": "Specific cache/layer no_grad issue is not the same as a broad complaint about many model breakages." + "reason": "A specific cache mutation bug is not the same as a vague report that v5 breaks too many models." }, { "left": "issue:43502", "right": "issue:44961", "accept": false, - "reason": "Local-files-only networking bug is unrelated to the placeholder 'racoon' issue." + "reason": "Network requests with local_files_only and an unrelated placeholder issue are not duplicates." }, { "left": "issue:42831", "right": "issue:43874", "accept": false, - "reason": "FP8 accuracy issue and missing multimodal token-count method are different defects." + "reason": "FineGrainedFP8 accuracy and GLM46V image patch counting are separate model bugs." }, { "left": "issue:30333", "right": "issue:36296", "accept": false, - "reason": "MLflow reporting failure is unrelated to tensor-parallel training bugs." + "reason": "Failed-job MLflow reporting and tensor-parallel training are unrelated." }, { "left": "issue:44355", "right": "issue:45071", "accept": false, - "reason": "Compiled-file load errors and PretrainedConfig type-checking regression are different issues." + "reason": "Compiled-file errors and PretrainedConfig type-checking breakage are different regressions." }, { - "left": "issue:44355", - "right": "issue:44655", + "left": "issue:44222", + "right": "issue:44230", "accept": false, - "reason": "Compiled-file execution errors and Pipeline save_pretrained failures are not the same bug." + "reason": "Both are fp8-themed, but one is MoE save_pretrained and the other is fp8 inference support." }, { - "left": "issue:44222", - "right": "issue:44230", + "left": "issue:44355", + "right": "issue:44655", "accept": false, - "reason": "Both involve FP8 and MoE, but one is saving and the other is inference support." + "reason": "Compiled Python file failures and pipeline save_pretrained are unrelated." }, { "left": "issue:43827", "right": "issue:44961", "accept": false, - "reason": "Docs still referencing pipeline() is unrelated to the unrelated 'racoon' issue." + "reason": "Documentation referencing pipeline() and the placeholder racoon issue are not duplicates." }, { "left": "issue:33453", "right": "issue:34689", "accept": false, - "reason": "Tokenizer loading regression and Llama 3.2 Vision model-loading breakage are different concrete bugs." + "reason": "Tokenizer loading regression and Llama 3.2 vision model loading breakage are different model-loading bugs." }, { "left": "issue:42757", "right": "issue:43576", "accept": false, - "reason": "These are plausibly related import/runtime regressions, but the titles do not show the same concrete bug path strongly enough to merge." + "reason": "Missing is_offline_mode import from huggingface_hub and broken transformers env command are different failures." } ] }, @@ -14115,8 +14166,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -14173,12 +14224,12 @@ "issue:43299|issue:44556", "issue:43010|issue:45656", "issue:33453|issue:36296", - "issue:43576|issue:45071", "issue:43576|issue:43824", + "issue:43576|issue:45071", "issue:44961|issue:45468", "issue:31795|issue:38617", - "issue:38617|issue:43576", "issue:41628|issue:42548", + "issue:38617|issue:43576", "issue:43010|issue:43976", "issue:43576|issue:44351", "issue:44908|issue:44961", @@ -14186,16 +14237,16 @@ "issue:41628|issue:44355", "issue:36296|issue:43526", "issue:44261|issue:44360", - "issue:44273|issue:45003", "issue:44623|issue:45230", "issue:44273|issue:45542", + "issue:44273|issue:45003", "issue:44393|issue:45753", "issue:43576|issue:44556", "issue:44038|issue:45125", "issue:44393|issue:45071", "issue:43525|issue:45850", - "issue:41669|issue:43673", "issue:44556|issue:45092", + "issue:41669|issue:43673", "issue:44373|issue:44961", "issue:43572|issue:45850", "issue:45362|issue:45850", @@ -14211,268 +14262,274 @@ "issue:44295|issue:45230", "issue:41669|issue:44351" ], - "prepared_review_unit_hash": "c4eb10c9fbcd48733eed069d34fb48f698788d37eb66002f5b5c298130e1000c", + "prepared_review_unit_hash": "dd171e46d16a7c9cd02795734c092479224567693fff408c7645f63e5777f74f", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 40, - "serialized_chars": 38941, - "estimated_input_tokens": 9736, - "estimated_eval_tokens": 19728 + "serialized_chars": 38936, + "estimated_input_tokens": 9734, + "estimated_eval_tokens": 19724 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is very noisy: most pairs are only loosely related by version-5 regressions or import/lazy-loading symptoms, but they do not look like the same concrete bug or fix. I found no strong duplicate pair to merge, so the best anchor is a broad umbrella issue rather than a specific regression.", - "confidence": 0.38, - "canonical_issue_reason": "Issue 45850 is the broadest umbrella-style report in the set and best matches the noisy cluster of v5 breakage complaints, even though it is vague.", + "summary": "No soft pair looks like a safe duplicate merge on titles alone. Most are broad v5 regressions, import/export failures, or unrelated subsystem bugs that share only a loose theme.", + "confidence": 0.58, + "canonical_issue_reason": "issue:44246 is the best standalone representative here: it is a concrete, actionable import-regression report, while most other candidates are either vague or narrower symptoms of different bugs.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45850 is the most suitable cluster representative because it is broad enough to cover multiple v5-related regressions without pretending the specific bugs are identical.", + "best_issue_reason": "issue:44246 is the strongest overall issue to anchor the cluster because it describes a specific user-visible failure with a plausible shared area (import/lazy loading) and enough detail to be useful for triage.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43299", "right": "issue:44556", "accept": false, - "reason": "Both mention v5 loading breakage, but one is Qwen3VL MoE loading and the other is general checkpoint reload incompatibility; not the same concrete failure." + "reason": "Both are v5 checkpoint-loading regressions, but they hit different model families and likely different failure paths; too broad to treat as the same bug." }, { "left": "issue:43010", "right": "issue:45656", "accept": false, - "reason": "One is about cache/layer update no_grad decoration, the other about deepspeed calling optimizer.step twice; different code paths and symptoms." + "reason": "Cache update no_grad decoration and deepspeed double optimizer-step are unrelated code paths." }, { "left": "issue:33453", "right": "issue:36296", "accept": false, - "reason": "Tokenizer loading regression and tensor-parallel training bug are unrelated subsystems and failures." + "reason": "Tokenizer-loading regression and tensor-parallel training bug are different problems." }, { "left": "issue:43576", - "right": "issue:45071", + "right": "issue:43824", "accept": false, - "reason": "Broken env command and PretrainedConfig type-checking regression are separate issues with no shared code path." + "reason": "Broken env command and missing top-level Qwen VL export are unrelated symptoms." }, { "left": "issue:43576", - "right": "issue:43824", + "right": "issue:45071", "accept": false, - "reason": "Env command failure is unrelated to a missing Qwen2.5-VL top-level import." + "reason": "A broken env command is not the same as PreTrainedConfig type-checking breakage." }, { "left": "issue:44961", "right": "issue:45468", "accept": false, - "reason": "The titles are unrelated and provide no evidence of the same bug or change." + "reason": "No meaningful overlap beyond both being generic issue reports." }, { "left": "issue:31795", "right": "issue:38617", "accept": false, - "reason": "Documentation confusion for forward args is unrelated to an ImportError for layer_type_validation." + "reason": "Documentation confusion and an ImportError for a missing symbol are not the same bug." }, { - "left": "issue:38617", - "right": "issue:43576", + "left": "issue:41628", + "right": "issue:42548", "accept": false, - "reason": "An import error for a specific symbol and a broken CLI env command are different problems." + "reason": "Both are import failures, but for different missing symbols and likely different export fixes." }, { - "left": "issue:41628", - "right": "issue:42548", + "left": "issue:38617", + "right": "issue:43576", "accept": false, - "reason": "Both are import errors, but for different top-level classes; the titles do not indicate the same missing-export bug." + "reason": "Missing layer_type_validation import and the broken env command are unrelated." }, { "left": "issue:43010", "right": "issue:43976", "accept": false, - "reason": "A no_grad/decorator request and a Python-version compatibility issue are not the same defect." + "reason": "Decorating cache updates with no_grad and Python-version incompatibility are unrelated." }, { "left": "issue:43576", "right": "issue:44351", "accept": false, - "reason": "Broken env command and missing HybridCache import are unrelated." + "reason": "Both concern imports, but they refer to different public symbols and different regressions." }, { "left": "issue:44908", "right": "issue:44961", "accept": false, - "reason": "Scheduler kwargs handling and a nonsensical titled issue are unrelated." + "reason": "Scheduler kwargs handling and an unrelated issue titled 'racoon' do not match." }, { "left": "issue:36296", "right": "issue:43901", "accept": false, - "reason": "Tensor-parallel training and text-classification docs behavior are not the same underlying problem." + "reason": "Tensor-parallel training and a pipeline docs mismatch are unrelated." }, { "left": "issue:41628", "right": "issue:44355", "accept": false, - "reason": "A missing AutoImageProcessor import and errors running compiled Python files do not point to the same bug." + "reason": "Top-level import failure and compiled-file execution errors do not point to the same defect." }, { "left": "issue:36296", "right": "issue:43526", "accept": false, - "reason": "Tensor parallel training bug and BeitImageProcessorFast reduce_labels bug are unrelated." + "reason": "Tensor parallel training and BeitImageProcessorFast label reduction are unrelated." }, { "left": "issue:44261", "right": "issue:44360", "accept": false, - "reason": "MLA q_a_layernorm precision/config issue and missing ReLU in the DSA indexer are different model changes." - }, - { - "left": "issue:44273", - "right": "issue:45003", - "accept": false, - "reason": "Lazy loading issues and unsafe sys.modules access may be adjacent, but the titles do not establish the same concrete bug." + "reason": "A precision/config issue in MLA q_a_layernorm is not the same as a DSA indexer activation omission." }, { "left": "issue:44623", "right": "issue:45230", "accept": false, - "reason": "Processor.save_pretrained missing files is unrelated to a generic bug-report placeholder." + "reason": "A missing processor.save_pretrained output bug and an empty 'Bug report' title are not enough to establish sameness." }, { "left": "issue:44273", "right": "issue:45542", "accept": false, - "reason": "Lazy loading and a tf backend error from tensorboard-only installs are different problems." + "reason": "Lazy loading failures and a TensorBoard-without-TensorFlow backend error are different import-time problems." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Both touch import/lazy-loading machinery, but the reported failures and likely fixes are different." }, { "left": "issue:44393", "right": "issue:45753", "accept": false, - "reason": "Both are Qwen3-related, but one is bounding-box hallucination and the other is an infinite loop on a specific image; too different to merge." + "reason": "Same model family, but different symptoms and repros; not safe to merge as one bug." }, { "left": "issue:43576", "right": "issue:44556", "accept": false, - "reason": "Broken env command and old checkpoint reload incompatibility are separate regressions." + "reason": "Broken env command and checkpoint reload regressions are unrelated." }, { "left": "issue:44038", "right": "issue:45125", "accept": false, - "reason": "Both touch Qwen3 MoE, but one is a general v5/Qwen3-VL-Moe issue and the other is a missing tensor-parallel plan; not the same code-path problem." + "reason": "Qwen3-VL/Moe loading trouble and missing tensor-parallel plan are different concrete issues." }, { "left": "issue:44393", "right": "issue:45071", "accept": false, - "reason": "Qwen3-VL output bugs and PretrainedConfig type-checking are unrelated." + "reason": "Qwen3-VL hallucination/output issues and PretrainedConfig type checking are unrelated." }, { "left": "issue:43525", "right": "issue:45850", "accept": false, - "reason": "A missing pad_token_id on Llama4Config is one specific model config bug, not the broad generic complaint in 45850." + "reason": "A specific missing pad_token_id attribute and a vague 'breaks too many models' complaint are not the same bug." }, { - "left": "issue:41669", - "right": "issue:43673", + "left": "issue:44556", + "right": "issue:45092", "accept": false, - "reason": "Removing import-* usage for import-time performance and a missing GenerationMixin cache during chunked prefill are unrelated." + "reason": "Both mention old-checkpoint incompatibility after v5, but they appear to affect different model/loading paths and are too broad to treat as duplicates." }, { - "left": "issue:44556", - "right": "issue:45092", + "left": "issue:41669", + "right": "issue:43673", "accept": false, - "reason": "Both are about v5 incompatibility, but one is general checkpoint reload after upgrade and the other is a specific remote-code InternVL2/meta-init issue; not clearly the same bug." + "reason": "Import-star performance regression and chunked-prefill cache absence are different issues." }, { "left": "issue:44373", "right": "issue:44961", "accept": false, - "reason": "Wrong docstring for position_ids is unrelated to the placeholder racoon issue." + "reason": "Wrong position_ids docstring and an unrelated issue titled 'racoon' are not duplicates." }, { "left": "issue:43572", "right": "issue:45850", "accept": false, - "reason": "StableLmConfig missing pad_token_idx and the broad 'v5 breaks too many models' complaint are not the same concrete regression." + "reason": "A specific StableLmConfig field regression and a generic 'too many models' complaint do not match." }, { "left": "issue:45362", "right": "issue:45850", "accept": false, - "reason": "Qwen3.5 chat crashes and a generic v5 breakage complaint are too broad to call duplicates." + "reason": "A Qwen3.5 chat crash and a vague broad regression report are not the same concrete bug." }, { "left": "issue:44704", "right": "issue:44908", "accept": false, - "reason": "AutoProcessor kwargs forwarding and inverse_sqrt scheduler kwargs are completely different code paths." + "reason": "AutoProcessor cached_file kwargs forwarding and inverse_sqrt scheduler kwargs are different code paths." }, { "left": "issue:42890", "right": "issue:44908", "accept": false, - "reason": "A flaky integration test and a scheduler kwargs bug are unrelated." + "reason": "Missing set_seed in SAM HQ tests and a scheduler kwargs bug are unrelated." }, { "left": "issue:44246", "right": "issue:44273", "accept": false, - "reason": "Slow import and lazy-loading malfunction are related only at a very high level; the titles do not show the same underlying defect." + "reason": "Both concern import/lazy-loading behavior, but one reports import slowness and the other a functional lazy-loading failure; not the same concrete bug." }, { "left": "issue:43408", "right": "issue:44315", "accept": false, - "reason": "sam3 tracker/model-type warning and Liger Kernel not applied with model_init are unrelated." + "reason": "Model-type mismatch warning and Liger kernel application through model_init are unrelated." }, { "left": "issue:43976", "right": "issue:44393", "accept": false, - "reason": "Python-version support issue and Qwen3-VL bounding-box output bugs are unrelated." + "reason": "Python version support and Qwen3-VL generation errors are unrelated." }, { "left": "issue:43352", "right": "issue:44393", "accept": false, - "reason": "Flash Attention support in Nemotron and Qwen3-VL hallucination/output errors are different issues." + "reason": "Flash Attention support gating and Qwen3-VL output errors are different problems." }, { "left": "issue:43576", "right": "issue:44908", "accept": false, - "reason": "A broken env command and scheduler kwargs handling do not share the same bug." + "reason": "A broken env command and scheduler kwarg handling are unrelated." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Scheduler kwargs and PretrainedConfig type checking do not share the same bug." }, { "left": "issue:44297", "right": "issue:45230", "accept": false, - "reason": "Tokenizer save_pretrained file mismatch and a generic bug report are not the same issue." + "reason": "Tokenizer save_pretrained metadata mismatch and a generic bug report are not duplicates." }, { "left": "issue:44295", "right": "issue:45230", "accept": false, - "reason": "Reading registered position_ids as a buffer and a generic bug report are unrelated." + "reason": "Position_ids buffer access error and a generic bug report are not duplicates." }, { "left": "issue:41669", "right": "issue:44351", "accept": false, - "reason": "Import-* performance cleanup and HybridCache import failure are different problems." + "reason": "Removing import * for performance and missing HybridCache export are different import-related regressions." } ] }, @@ -14491,8 +14548,8 @@ "review_unit_count": 45, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-15354-392", "nodes": [ @@ -14559,8 +14616,8 @@ "issue:41669|issue:44908", "issue:42371|issue:44492", "issue:43576|issue:44273", - "issue:44655|issue:44861", "issue:44360|issue:45468", + "issue:44655|issue:44861", "issue:44492|issue:45850", "issue:44273|issue:44908", "issue:45230|issue:45310", @@ -14598,322 +14655,334 @@ "issue:31515|issue:33453", "issue:43298|issue:44230" ], - "prepared_review_unit_hash": "77388cf968112299073a9ee4618f3332fdd907d3d6722246b268625fcb62e74e", + "prepared_review_unit_hash": "b5b487366e22d8d4461b6b6961f025a8714dedee62d4ecb355c8248bf15ab2fc", "component_budget": { "node_count": 392, "item_count": 392, "soft_pair_count": 1480, - "serialized_chars": 761482, - "estimated_input_tokens": 190371, - "estimated_eval_tokens": 380998 + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 51, - "serialized_chars": 42267, - "estimated_input_tokens": 10567, - "estimated_eval_tokens": 21390 + "serialized_chars": 42264, + "estimated_input_tokens": 10566, + "estimated_eval_tokens": 21388 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated Transformer issues were grouped together. Several are broad v5 regression/loading complaints, but the pairwise evidence is too weak to treat any soft edge as a true duplicate.", - "confidence": 0.83, - "canonical_issue_reason": "issue:45278 is the best representative of the broad v5 regression/load-failure theme: it is specific enough to be actionable, but still general enough to cover the nearby model-loading complaints in this cluster.", + "summary": "This is a loose cluster of mostly unrelated Transformers v5 regression reports. The strongest common thread is broad model-loading / compatibility breakage after version upgrades, but the soft-linked pairs are generally too different to treat as duplicates.", + "confidence": 0.64, + "canonical_issue_reason": "issue:45278 is the best umbrella issue: it concretely reports widespread breakage after upgrading to v5.5.0, which is broader and more representative than the more specific or more vague reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the most central and reusable issue here; the others are either narrower sub-bugs, documentation issues, or too vague to serve as a stable canonical.", + "best_issue_reason": "issue:45278 is the most suitable global anchor for this cluster because it captures the recurring upgrade-regression theme without being as vague as 'too many models break'.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43673", "right": "issue:44492", "accept": false, - "reason": "Cache-related titles, but one is a generation cache bug and the other is a typo/documentation issue." + "reason": "Different bugs: missing cache during chunked prefill vs a typo in cache strategy docs/code." }, { "left": "issue:41669", "right": "issue:45341", "accept": false, - "reason": "Unrelated subjects: import-star cleanup versus a testing_utils bug." + "reason": "Unrelated: import-star performance issue vs a small testing_utils bug." }, { "left": "issue:45278", "right": "issue:45850", "accept": false, - "reason": "Both are broad v5 complaints, but not the same concrete bug or code path." + "reason": "Same broad complaint about v5 breakage, but not the same concrete bug or fixable change." }, { "left": "issue:44393", "right": "issue:45478", "accept": false, - "reason": "Different models and failures: Qwen3-VL bbox output versus Qwen3.5 MoE from_pretrained loading." + "reason": "Different model families and failures: Qwen3-VL 2D bbox issue vs Qwen3.5 MoE from_pretrained error." }, { "left": "issue:43874", "right": "issue:45850", "accept": false, - "reason": "Specific GLM46V image-processor AttributeError versus a generic many-models breakage report." + "reason": "Specific missing-method multimodal token bug vs a generic meta complaint about v5 breaking models." }, { "left": "issue:41669", "right": "issue:44492", "accept": false, - "reason": "Cleanup/import-performance issue versus cache-strategy typo; not the same bug." + "reason": "Import-star cleanup/performance issue is unrelated to cache-strategy typo." }, { "left": "issue:44315", "right": "issue:44829", "accept": false, - "reason": "Both mention training, but the code paths are different: model_init/Liger Kernel versus flash_attention_3 degenerate training." + "reason": "Different code paths: Liger kernel application vs flash_attention_3 training degeneration." }, { "left": "issue:43827", "right": "issue:44908", "accept": false, - "reason": "Documentation after pipeline removals versus a scheduler kwargs bug." + "reason": "Docs referencing removed pipeline APIs is unrelated to inverse_sqrt scheduler kwargs handling." }, { "left": "issue:44077", "right": "issue:44861", "accept": false, - "reason": "Different failures in different utilities; no shared underlying bug." + "reason": "Different failures in different helpers; no shared underlying bug." }, { "left": "issue:31795", "right": "issue:33453", "accept": false, - "reason": "Documentation confusion versus tokenizer loading regression." + "reason": "Docs confusion vs a tokenizer loading regression." }, { "left": "issue:41669", "right": "issue:44908", "accept": false, - "reason": "Import-star cleanup and inverse_sqrt scheduler kwargs are unrelated." + "reason": "Performance/import cleanup is unrelated to scheduler kwargs propagation." }, { "left": "issue:42371", "right": "issue:44492", "accept": false, - "reason": "TF32 API migration versus a cache-strategy typo." + "reason": "TF32 API migration issue vs cache-strategy typo." }, { "left": "issue:43576", "right": "issue:44273", "accept": false, - "reason": "Broken env command and lazy-loading malfunction are separate issues." + "reason": "Broken env command and lazy-loading failure are separate problems." }, { - "left": "issue:44655", - "right": "issue:44861", + "left": "issue:44360", + "right": "issue:45468", "accept": false, - "reason": "save_pretrained pipeline export issue versus tied-weights-key crash." + "reason": "Different model implementations and unrelated bugs." }, { - "left": "issue:44360", - "right": "issue:45468", + "left": "issue:44655", + "right": "issue:44861", "accept": false, - "reason": "Different model/component bugs: DSA indexer ReLU versus Gemma4Audio positional encoding." + "reason": "Saving Pipeline objects is a different path from tied-weight-key attribute handling." }, { "left": "issue:44492", "right": "issue:45850", "accept": false, - "reason": "A typo issue does not match a generic v5 breakage report." + "reason": "A typo in cache strategies is not the same as a broad v5 breakage report." }, { "left": "issue:44273", "right": "issue:44908", "accept": false, - "reason": "Lazy loading and scheduler kwargs are unrelated subsystems." + "reason": "Lazy loading failure vs scheduler kwargs bug." }, { "left": "issue:45230", "right": "issue:45310", "accept": false, - "reason": "Generic bug report versus a specific Qwen3.5 MoE from_pretrained error; not enough to merge." + "reason": "Generic bug report vs a specific Qwen3.5 MoE from_pretrained error." }, { "left": "issue:43519", "right": "issue:44485", "accept": false, - "reason": "Timestamp calculation in Qwen3VL is unrelated to GLM-5 RoPE implementation." + "reason": "Timestamp calculation bug vs RoPE implementation issue." }, { "left": "issue:41669", "right": "issue:43704", "accept": false, - "reason": "Import cleanup versus VRAM leak; no common code-path." + "reason": "Import-star usage cleanup is unrelated to a VRAM leak." }, { "left": "issue:43519", "right": "issue:44908", "accept": false, - "reason": "Different components: processor timestamp math versus scheduler behavior." + "reason": "Timestamp calculation bug vs scheduler keyword handling." }, { "left": "issue:31795", "right": "issue:34689", "accept": false, - "reason": "Documentation ambiguity versus a model-loading regression." + "reason": "Documentation confusion vs a specific model-loading regression." }, { "left": "issue:31795", "right": "issue:37428", "accept": false, - "reason": "Doc issue versus import error from flash attention utils." + "reason": "Docs issue vs an ImportError from a removed symbol." }, { "left": "issue:44315", "right": "issue:45092", "accept": false, - "reason": "Both touch model initialization, but one is about Liger Kernel application and the other about remote-code/meta init incompatibility." + "reason": "Both touch model initialization, but the underlying failures are different." }, { "left": "issue:43704", "right": "issue:44485", "accept": false, - "reason": "VRAM leak and RoPE implementation are unrelated bugs." + "reason": "VRAM leak and RoPE implementation are unrelated." }, { "left": "issue:44485", "right": "issue:45468", "accept": false, - "reason": "Different model implementations and failure modes." + "reason": "Different positional/rope implementations for different models." }, { "left": "issue:42371", "right": "issue:43704", "accept": false, - "reason": "TF32 API guidance and VRAM leakage do not share a code path." + "reason": "TF32 settings migration vs VRAM leak." }, { "left": "issue:43502", "right": "issue:43519", "accept": false, - "reason": "Local-files-only network leak versus tokenizer/processor timestamp bug." + "reason": "Local-files-only network leakage vs Qwen3VL timestamp math." }, { "left": "issue:43519", "right": "issue:44623", "accept": false, - "reason": "Timestamp calculation bug versus missing files on save_pretrained." + "reason": "Different processor bugs: timestamp math vs save_pretrained missing files." }, { "left": "issue:43519", "right": "issue:43704", "accept": false, - "reason": "Processor timestamp bug versus multi-thread VRAM leak." + "reason": "Timestamp bug vs VRAM leak." }, { "left": "issue:44263", "right": "issue:45468", "accept": false, - "reason": "torch.split return values in a GLM indexer and Gemma audio positional encoding are different issues." + "reason": "Different indexer and positional-encoding issues." }, { "left": "issue:43606", "right": "issue:45468", "accept": false, - "reason": "CPU offload device mismatch is unrelated to Gemma audio positional encoding." + "reason": "CPU offload device mismatch vs Gemma audio positional encoding." }, { "left": "issue:44261", "right": "issue:45468", "accept": false, - "reason": "MLA layernorm epsilon precision issue versus a Gemma audio encoding bug." + "reason": "RMS norm epsilon precision issue vs unrelated Gemma audio encoding bug." }, { "left": "issue:43704", "right": "issue:45341", "accept": false, - "reason": "VRAM leak and a testing_utils bug are unrelated." + "reason": "VRAM leak vs a testing_utils bug." }, { "left": "issue:42913", "right": "issue:43066", "accept": false, - "reason": "Both are tokenizer-v5 complaints, but one is a broad behavior regression and the other is a specific decoder-type error; too different to merge." + "reason": "Both are tokenizer-related, but they describe different concrete regressions." }, { "left": "issue:44908", "right": "issue:44998", "accept": false, - "reason": "Scheduler bug versus an unrelated placeholder issue." + "reason": "Scheduler bug vs an unrelated placeholder issue." }, { "left": "issue:44861", "right": "issue:45290", "accept": false, - "reason": "Tied-weights key handling and chat-template tool-call crashing are separate code paths." + "reason": "Different failure modes in different code paths." + }, + { + "left": "issue:43704", + "right": "issue:44998", + "accept": false, + "reason": "VRAM leak vs unrelated placeholder issue." }, { - "left": "issue:43704", + "left": "issue:41669", "right": "issue:44998", "accept": false, - "reason": "VRAM leak report versus unrelated placeholder issue." + "reason": "Import cleanup performance issue vs unrelated placeholder issue." }, { - "left": "issue:44230", - "right": "issue:45310", + "left": "issue:43525", + "right": "issue:44861", + "accept": false, + "reason": "Missing config attribute vs tied-weight-key attribute crash." + }, + { + "left": "issue:43572", + "right": "issue:44861", "accept": false, - "reason": "Feature/support request for fp8 infer support versus a from_pretrained loading error." + "reason": "StableLm pad_token_idx regression vs tied-weight-key handling." }, { "left": "issue:31515", "right": "issue:38617", "accept": false, - "reason": "Slow checkpoint loading and an import error are unrelated." + "reason": "Slow checkpoint loading vs an ImportError for a removed config utility." }, { "left": "issue:44360", "right": "issue:44998", "accept": false, - "reason": "DSA indexer bug versus unrelated placeholder issue." + "reason": "DSA indexer bug vs unrelated placeholder issue." }, { "left": "issue:44230", "right": "issue:44623", "accept": false, - "reason": "FP8 support request versus processor save_pretrained missing files." + "reason": "FP8 support request/bug vs processor save_pretrained missing files." }, { "left": "issue:30333", "right": "issue:31515", "accept": false, - "reason": "MLFlow training-job reporting and checkpoint-loading speed are different problems." + "reason": "MLflow job-reporting issue vs slow from_pretrained loading." }, { "left": "issue:31515", "right": "issue:36296", "accept": false, - "reason": "Checkpoint loading slowdown versus tensor-parallel training bug." + "reason": "Checkpoint loading slowdown vs tensor-parallel training bug." }, { "left": "issue:31515", "right": "issue:31795", "accept": false, - "reason": "Slow loading versus documentation confusion; not the same issue." + "reason": "Performance regression vs documentation confusion." }, { "left": "issue:44230", "right": "issue:45310", "accept": false, - "reason": "FP8 support request and Qwen3.5 MoE from_pretrained error are different underlying changes." + "reason": "FP8 support issue vs Qwen3.5 MoE from_pretrained error; related model family but different bugs." }, { "left": "issue:31515", "right": "issue:33453", "accept": false, - "reason": "Loading speed regression versus tokenizer loading regression." + "reason": "Slow checkpoint loading vs tokenizer loading regression." }, { "left": "issue:43298", "right": "issue:44230", "accept": false, - "reason": "Continuous batching audio-model support and fp8 inference support are unrelated features." + "reason": "Continuous batching audio-model support vs FP8 inference support." } ] }, @@ -14932,8 +15001,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-26350-2", "nodes": [ @@ -14943,12 +15012,12 @@ "soft_pairs": [ "issue:26350|issue:33260" ], - "prepared_review_unit_hash": "07cce9baa57ceac172565ceae10b960646de03b3e915759bd7f47d56b17f6535", + "prepared_review_unit_hash": "042c13160c230077e5b466365afd7261046ad29da3a0dfa6022a334d5544917f", "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 }, @@ -14956,7 +15025,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1589, + "serialized_chars": 1590, "estimated_input_tokens": 398, "estimated_eval_tokens": 1052 }, @@ -14965,18 +15034,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two unrelated community feature requests: one for Flash Attention 2 architecture support and one for GGUF support across more architectures. They share only a broad \u201cexpand architecture support\u201d theme, not the same underlying change.", - "confidence": 0.98, + "summary": "These are two different community contribution tracking issues for unrelated features (Flash Attention 2 vs GGUF support). The soft similarity is superficial and should not be treated as a duplicate.", + "confidence": 0.96, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue #26350 is the older and more discussed request, so it is the better anchor if one issue must represent the cluster; however, it is not a duplicate of #33260.", + "best_issue_reason": "issue:26350 is the stronger representative of the cluster because it has significantly higher discussion activity and is the more established tracking issue, even though it is not a duplicate of issue:33260.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:26350", "right": "issue:33260", "accept": false, - "reason": "Different feature requests for different subsystems: Flash Attention 2 support vs GGUF support. They do not describe the same bug or the same code change, so they should not be merged as duplicates." + "reason": "Different underlying requests: one is about Flash Attention 2 architecture support, the other is about GGUF architecture support. They share a template and scope style, not the same bug or change." } ] }, @@ -14995,8 +15064,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-27088-4", "nodes": [ @@ -15013,12 +15082,12 @@ "issue:43834|issue:44365", "issue:44365|issue:44370" ], - "prepared_review_unit_hash": "da07ef3e1f3e2a5f6d0e74ea80aa590d5d2811aaac92dd5b417fa6b9fc839e84", + "prepared_review_unit_hash": "70efd57aa823bc8b35dbfe143ecae77709451c4243c4a4a348e3037b2ecd7c5b", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4454, + "serialized_chars": 4453, "estimated_input_tokens": 1114, "estimated_eval_tokens": 2484 }, @@ -15026,7 +15095,7 @@ "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4454, + "serialized_chars": 4453, "estimated_input_tokens": 1114, "estimated_eval_tokens": 2484 }, @@ -15035,48 +15104,48 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are issue-template variants for translating docs into different languages; only one is a concrete Turkish instance, while the others are generic placeholders. They are similar in wording but not the same underlying issue.", - "confidence": 0.93, - "canonical_issue_reason": "issue:27088 is the only concrete, language-specific report ([i18n-TR] Turkish), so it is the most informative representative issue.", + "summary": "These are all i18n translation-request issues with highly similar templated wording, but they do not appear to describe the same concrete translation task. The Turkish request is specific, while the others are generic placeholder issues for potentially different languages.", + "confidence": 0.96, + "canonical_issue_reason": "issue:27088 is the only concrete, language-specific request ([i18n-TR] Turkish); the others are templated placeholder issues and are not clearly the same underlying request.", "canonical_pr_reason": null, - "best_issue_reason": "issue:27088 is the best global issue because it is specific and actionable, whereas the others are generic template entries without a distinct concrete language target.", + "best_issue_reason": "issue:27088 is the best representative because it is specific and actionable, whereas the others are generic template variants with no concrete language identified.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:27088", "right": "issue:44370", "accept": false, - "reason": "Different issue instances: Turkish translation vs. a generic placeholder-language template; not the same concrete task." + "reason": "Same translation-template wording, but no evidence they request the same language; one is Turkish and the other is an unspecified placeholder." }, { "left": "issue:27088", "right": "issue:44365", "accept": false, - "reason": "Different issue instances: Turkish translation vs. a generic placeholder-language template; not the same concrete task." + "reason": "Superficially similar i18n request, but the concrete Turkish request is distinct from the generic placeholder template." }, { "left": "issue:27088", "right": "issue:43834", "accept": false, - "reason": "Different issue instances: Turkish translation vs. a generic placeholder-language template; not the same concrete task." + "reason": "Not the same underlying request: Turkish is explicit, while the other issue is a generic language placeholder." }, { "left": "issue:43834", "right": "issue:44370", "accept": false, - "reason": "Both are generic template-style issues with no concrete shared language target shown; too ambiguous to treat as the same underlying issue." + "reason": "Very similar template text, but the placeholders do not establish the same concrete language or translation task." }, { "left": "issue:43834", "right": "issue:44365", "accept": false, - "reason": "Both are generic template-style issues with no concrete shared language target shown; too ambiguous to treat as the same underlying issue." + "reason": "These look like separate language-specific translation requests sharing a template, not duplicates of one concrete issue." }, { "left": "issue:44365", "right": "issue:44370", "accept": false, - "reason": "Both are generic template-style issues with no concrete shared language target shown; too ambiguous to treat as the same underlying issue." + "reason": "Identical template structure is not enough to merge them; the actual target language is unknown and may differ." } ] }, @@ -15095,8 +15164,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-36817-15", "nodes": [ @@ -15128,8 +15197,8 @@ "issue:37015|issue:43837", "issue:37015|issue:44121", "issue:37015|issue:43909", - "issue:37015|issue:42503", "issue:36879|issue:37015", + "issue:37015|issue:42503", "issue:41115|issue:43837", "issue:41115|issue:44121", "issue:41115|issue:42503", @@ -15153,244 +15222,244 @@ "issue:37015|issue:45027", "issue:43837|issue:45177" ], - "prepared_review_unit_hash": "425ebecca23ffca7bea528964385504261a9688191e42e55982b42baa67b5521", + "prepared_review_unit_hash": "f557324d9d0ef7b013b18dbf4ce8c9c20ecb4db0f03ab72b04040ff0a764a899", "component_budget": { "node_count": 15, "item_count": 15, "soft_pair_count": 35, - "serialized_chars": 20700, - "estimated_input_tokens": 5175, - "estimated_eval_tokens": 10606 + "serialized_chars": 20707, + "estimated_input_tokens": 5177, + "estimated_eval_tokens": 10610 }, "budget": { "node_count": 15, "item_count": 15, "soft_pair_count": 35, - "serialized_chars": 20700, - "estimated_input_tokens": 5175, - "estimated_eval_tokens": 10606 + "serialized_chars": 20707, + "estimated_input_tokens": 5177, + "estimated_eval_tokens": 10610 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates; they are separate model-request issues spanning different models and modalities (BERT variants, detection, audio/TTS/ASR, OCR, vision, and a sparse transformer).", + "summary": "This cluster is a broad false-positive: the issues are all model-add/support requests, but they target different concrete models and often different modalities (BERT-like, detection, OCR, speech, audio, vision). They should not be treated as duplicates.", "confidence": 0.96, - "canonical_issue_reason": "No single canonical duplicate exists here; the cluster is a set of distinct feature requests. If one must be chosen as the broadest representative, issue 42738 is the closest umbrella item.", + "canonical_issue_reason": "If forced to pick one representative, issue #42738 is the broadest umbrella-style request and best matches the shared 'model support' theme, though it is not a true duplicate of the rest.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 42738 ('BERT-like models with RoPE') is the broadest and most reusable representative, but it still only fits part of the cluster and does not cover the audio/vision-specific requests.", + "best_issue_reason": "Issue #42738 is the least specific and most general among the set, so it is the closest single representative of the cluster's theme; however, the cluster is too heterogeneous for a strong canonical issue.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36817", "right": "issue:41211", "accept": false, - "reason": "Different requested models (EuroBert vs DEIMv2); same general area of model support, but not the same change." + "reason": "Different model families and domains; only the generic 'add model' framing overlaps." }, { "left": "issue:36817", "right": "issue:43671", "accept": false, - "reason": "EuroBert config support is unrelated to Qwen3-TTS support; different modality and code path." + "reason": "EuroBert config support vs Qwen3 text-to-speech support are unrelated code paths." }, { "left": "issue:36817", "right": "issue:36879", "accept": false, - "reason": "EuroBert and RF-DETR are different models with different integration needs." + "reason": "BERT config addition vs RF-DETR model support; different concrete models and subsystems." }, { "left": "issue:36879", "right": "issue:41211", "accept": false, - "reason": "RF-DETR detection support and DEIMv2 are separate model additions." + "reason": "RF-DETR object detection and DEIMv2 are different models; not the same change." }, { "left": "issue:36879", "right": "issue:43909", "accept": false, - "reason": "RF-DETR is a detection model; LFM2.5 Audio is an audio model request." + "reason": "Object detection request vs audio model request; different integration targets." }, { "left": "issue:36879", "right": "issue:44121", "accept": false, - "reason": "Different model families and implementation targets; not the same bug/change." + "reason": "Detector support vs sparse transformer architecture request; unrelated code paths." }, { "left": "issue:36879", "right": "issue:43837", "accept": false, - "reason": "RF-DETR and Qwen3-ASR are different model requests with different runtime paths." + "reason": "RF-DETR and Qwen3-ASR are different model types and tasks." }, { "left": "issue:36879", "right": "issue:43671", "accept": false, - "reason": "RF-DETR vs Qwen3-TTS are distinct model integrations, not duplicates." + "reason": "Detector support vs text-to-speech support; not the same underlying bug/change." }, { "left": "issue:37015", "right": "issue:43837", "accept": false, - "reason": "NeoBERT and Qwen3-ASR are different model additions and would not merge into one PR." + "reason": "NeoBERT support and Qwen3-ASR support are different model integrations." }, { "left": "issue:37015", "right": "issue:44121", "accept": false, - "reason": "NeoBERT and the sparse transformer request are unrelated model architectures." + "reason": "Distinct model families: NeoBERT vs weight-sparse transformer." }, { "left": "issue:37015", "right": "issue:43909", "accept": false, - "reason": "NeoBERT (text model) and LFM2.5 Audio are different modalities and code paths." + "reason": "NeoBERT request is unrelated to the LFM2.5 audio model request." }, { - "left": "issue:37015", - "right": "issue:42503", + "left": "issue:36879", + "right": "issue:37015", "accept": false, - "reason": "NeoBERT and ModernVBERT are separate model requests, not the same underlying change." + "reason": "RF-DETR detection support and NeoBERT support are different concrete additions." }, { - "left": "issue:36879", - "right": "issue:37015", + "left": "issue:37015", + "right": "issue:42503", "accept": false, - "reason": "RF-DETR and NeoBERT are unrelated model support requests." + "reason": "Both are model requests, but for different BERT-like families with different implementation needs." }, { "left": "issue:41115", "right": "issue:43837", "accept": false, - "reason": "MiniCPM3 architecture support and Qwen3-ASR support are different model integrations." + "reason": "MiniCPM3 architecture support and Qwen3-ASR support are unrelated changes." }, { "left": "issue:41115", "right": "issue:44121", "accept": false, - "reason": "MiniCPM3 architecture and the sparse transformer request are distinct code additions." + "reason": "MiniCPM3 model architecture addition vs sparse transformer support; no shared code path." }, { "left": "issue:41115", "right": "issue:42503", "accept": false, - "reason": "MiniCPM3 and ModernVBERT are separate model families." + "reason": "Different model families and different concrete integration work." }, { "left": "issue:41115", "right": "issue:45027", "accept": false, - "reason": "MiniCPM3 architecture support and Voxtral TTS support are different modalities and code paths." + "reason": "MiniCPM3 architecture request vs Voxtral TTS support are different modalities." }, { "left": "issue:41211", "right": "issue:43671", "accept": false, - "reason": "DEIMv2 and Qwen3-TTS are different model requests with no shared concrete fix." + "reason": "DEIMv2 and Qwen3-TTS are unrelated model requests." }, { "left": "issue:42503", "right": "issue:43837", "accept": false, - "reason": "ModernVBERT models and Qwen3-ASR are different model support requests." + "reason": "ModernVBERT and Qwen3-ASR target different model families and tasks." }, { "left": "issue:42503", "right": "issue:43671", "accept": false, - "reason": "ModernVBERT and Qwen3-TTS are distinct model additions." + "reason": "ModernVBERT models and Qwen3 TTS are different integrations, not duplicates." }, { "left": "issue:42503", "right": "issue:42738", "accept": false, - "reason": "ModernVBERT models are specific additions; 'BERT-like models with RoPE' is only a broad category, not the same request." + "reason": "A specific model-family request vs a generic BERT-like-with-RoPE proposal; only thematic overlap." }, { "left": "issue:42738", "right": "issue:45027", "accept": false, - "reason": "BERT-like RoPE models and Voxtral TTS are different model families and implementation paths." + "reason": "BERT-like models with RoPE and Voxtral TTS are unrelated model classes." }, { "left": "issue:43671", "right": "issue:43837", "accept": false, - "reason": "Qwen3-TTS and Qwen3-ASR are related by naming only; they are separate tasks and code paths." + "reason": "Both are speech-related, but TTS and ASR are different model support paths." }, { "left": "issue:43671", "right": "issue:45027", "accept": false, - "reason": "Qwen3-TTS and Voxtral TTS are different concrete model-support requests." + "reason": "Different speech model families; not enough to treat as the same change." }, { "left": "issue:43671", "right": "issue:45177", "accept": false, - "reason": "TTS support and DeepSeek-OCR2 are unrelated modalities." + "reason": "Text-to-speech support vs DeepSeek-OCR2 support are different modalities." }, { "left": "issue:43671", "right": "issue:44121", "accept": false, - "reason": "Qwen3-TTS and the sparse transformer request are different underlying changes." + "reason": "Speech model support and sparse transformer support are unrelated." }, { "left": "issue:43837", "right": "issue:45322", "accept": false, - "reason": "Qwen3-ASR support and EUPE encoder support are different model integrations." + "reason": "Qwen3-ASR and EUPE address different model types and tasks." }, { "left": "issue:43837", "right": "issue:43909", "accept": false, - "reason": "ASR and audio generation model support are distinct requests, not one bug/change." + "reason": "ASR and audio model requests are related only at a high level, not the same concrete integration." }, { "left": "issue:43909", "right": "issue:44205", "accept": false, - "reason": "LFM2.5 Audio and SAM3-LiteText are different model families and modalities." + "reason": "Audio model support vs SAM3-LiteText support are different changes." }, { "left": "issue:43671", "right": "issue:43909", "accept": false, - "reason": "Qwen3-TTS and LFM2.5 Audio are separate audio-model requests." + "reason": "TTS and audio model support are distinct model integrations." }, { "left": "issue:44121", "right": "issue:45322", "accept": false, - "reason": "Sparse transformer support and EUPE support are different architectures." + "reason": "Weight-sparse transformer support and EUPE support are different architectural requests." }, { "left": "issue:43837", "right": "issue:44121", "accept": false, - "reason": "Qwen3-ASR and the sparse transformer request do not target the same code path." + "reason": "ASR support and sparse transformer support are unrelated code paths." }, { "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 speech-model integrations." }, { "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 families." }, { "left": "issue:43837", "right": "issue:45177", "accept": false, - "reason": "ASR support and OCR2 support are different modalities and not duplicates." + "reason": "ASR support vs OCR2 support; different modalities and implementation work." } ] }, @@ -15409,8 +15478,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-37934-6", "nodes": [ @@ -15431,30 +15500,30 @@ "issue:43668|issue:44031", "issue:43668|issue:43813" ], - "prepared_review_unit_hash": "7d746c0b3b1d7c314490fa1a3a5666b7717f8c809480ad90a20bf5b8ea1285d9", + "prepared_review_unit_hash": "997778c65039c2e4781f8743225cd174657d41c0387daddd0097f5f3eccdc4c7", "component_budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 8, - "serialized_chars": 6293, - "estimated_input_tokens": 1574, - "estimated_eval_tokens": 3404 + "serialized_chars": 6292, + "estimated_input_tokens": 1573, + "estimated_eval_tokens": 3402 }, "budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 8, - "serialized_chars": 6293, - "estimated_input_tokens": 1574, - "estimated_eval_tokens": 3404 + "serialized_chars": 6292, + "estimated_input_tokens": 1573, + "estimated_eval_tokens": 3402 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The cluster is not a true duplicate set: the issues concern unrelated areas (normalization, Mask2Former loss handling, tokenizer warnings, a config type hint, and a typo in a quantization symbol). The only commonality is superficial wording, not the same underlying bug.", - "confidence": 0.91, + "summary": "The items are not one duplicate cluster; they span unrelated bugs in norms, Mask2Former loss handling, config typing, a typo, and tokenizer warnings. No soft pair looks like the same underlying issue/change.", + "confidence": 0.95, "canonical_issue_reason": null, "canonical_pr_reason": null, "best_issue_reason": null, @@ -15464,49 +15533,49 @@ "left": "issue:37934", "right": "issue:43668", "accept": false, - "reason": "Both mention normalization-related terms, but one is about whether Llama4TextL2Norm is RMS norm and the other is a ModernBERTConfig type hint; different files, models, and bug types." + "reason": "Both mention normalization details, but one is about whether Llama4TextL2Norm should be RMSNorm and the other is a type-hint problem in ModernBERTConfig; different code paths and fixes." }, { "left": "issue:42936", "right": "issue:43240", "accept": false, - "reason": "Mask2Former ignore_value handling and kwargs not being forwarded to a loss function are distinct implementation issues in different code paths." + "reason": "Both touch training/loss plumbing, but one is about Mask2Former's ignore_value being unused and the other is about kwargs not reaching the loss function; not the same bug." }, { "left": "issue:42936", "right": "issue:44031", "accept": false, - "reason": "These are unrelated: one is a model-specific ignore_value bug, the other is a tokenizer regex warning after a version change." + "reason": "Mask2Former loss handling vs tokenizer regex warnings are unrelated subsystems and changes." }, { "left": "issue:42936", "right": "issue:43813", "accept": false, - "reason": "A Mask2Former loss/config bug and a typo in a quantization attribute are not the same underlying defect." + "reason": "Mask2Former ignore_value usage and a typo in orig_conversion.quantization_operation are unrelated." }, { "left": "issue:42936", "right": "issue:43668", "accept": false, - "reason": "Different models and different problem classes: unused ignore_value versus an incorrect type hint." + "reason": "Mask2Former loss behavior and a ModernBERTConfig type hint issue are unrelated." }, { "left": "issue:43240", "right": "issue:43668", "accept": false, - "reason": "Forwarding kwargs to a loss function and correcting a config type hint are unrelated fixes." + "reason": "Passing kwargs to a loss function and correcting a config type hint are different problems." }, { "left": "issue:43668", "right": "issue:44031", "accept": false, - "reason": "A ModernBERTConfig annotation issue does not match a tokenizer regex warning regression." + "reason": "ModernBERTConfig norm_eps typing and tokenizer regex warnings are unrelated." }, { "left": "issue:43668", "right": "issue:43813", "accept": false, - "reason": "These only share the fact that both are small correctness issues; one is a type-hint typo, the other is a symbol-name typo in quantization code." + "reason": "A config type-hint error and a field-name typo in conversion code are not the same underlying issue." } ] }, @@ -15525,8 +15594,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-41084-2", "nodes": [ @@ -15536,12 +15605,12 @@ "soft_pairs": [ "issue:41084|issue:43037" ], - "prepared_review_unit_hash": "d285e80d4f39570046f2f340b8ad8458e2a86dfea4ef0aad543c19edf64ffbb0", + "prepared_review_unit_hash": "fc68dbe90644d4dc1c1a0459868931acecac490e7cc26931b7f76888ec100deb", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1479, + "serialized_chars": 1480, "estimated_input_tokens": 370, "estimated_eval_tokens": 996 }, @@ -15549,7 +15618,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1479, + "serialized_chars": 1480, "estimated_input_tokens": 370, "estimated_eval_tokens": 996 }, @@ -15558,18 +15627,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are not duplicates: one is a generic block-decoding request, while the other is a specific request for DeepSeek v3.2 support. There is no shared underlying bug or change here.", - "confidence": 0.97, + "summary": "These two issues are unrelated: one is about block decoding, the other is a DeepSeek v3.2 support request. They should not be clustered as duplicates.", + "confidence": 0.98, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Issue #43037 is the more specific and user-facing request, but it is not a duplicate of #41084.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41084", "right": "issue:43037", "accept": false, - "reason": "Different topics and scopes: block decoding vs model/version support. The titles and available metadata do not indicate the same underlying issue." + "reason": "Different underlying topics with no shared code-path or concrete bug/change: block decoding vs model support for DeepSeek v3.2." } ] }, @@ -15588,8 +15657,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43598-2", "nodes": [ @@ -15599,12 +15668,12 @@ "soft_pairs": [ "issue:43598|issue:43600" ], - "prepared_review_unit_hash": "f52b10dfe50c835a1857ed7ee0257f8076630d6e3b10dfec219389e7e0e92e54", + "prepared_review_unit_hash": "755a27ca33e0c46cf0c3aafb6bd5c4392550366e38bbc2642a179a6763834cb8", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1584, + "serialized_chars": 1583, "estimated_input_tokens": 396, "estimated_eval_tokens": 1048 }, @@ -15612,7 +15681,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1584, + "serialized_chars": 1583, "estimated_input_tokens": 396, "estimated_eval_tokens": 1048 }, @@ -15621,18 +15690,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both issues concern Trainer behavior with custom trainers, but they target different code paths: one is about `remove_unused_column` customizability, the other about FLOPS estimation. They are related in theme, not the same bug.", - "confidence": 0.81, - "canonical_issue_reason": "#43598 is the better canonical issue because it describes the broader Trainer customizability problem more directly and is more likely to serve as the umbrella for related follow-ups.", + "summary": "These are related in that both mention custom trainers and Trainer behavior, but they describe different problems: one is about `remove_unused_column` customizability, the other about FLOPS estimation. Not a duplicate pair.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 43598 is the cleaner representative of the cluster because it is a more concrete Trainer customization concern.", "canonical_pr_reason": null, - "best_issue_reason": "#43598 is the best representative issue for the cluster since it frames the general Trainer customization concern more clearly; #43600 is a narrower instrumentation-specific concern.", + "best_issue_reason": "Issue 43598 is the better standalone issue to keep as the cluster representative: it describes a specific behavior in `Trainer` and is easier to action than the broader FLOPS-estimation request.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43598", "right": "issue:43600", "accept": false, - "reason": "Both mention custom trainers in Trainer, but `remove_unused_column` behavior and FLOPS estimation are different concrete problems and would not be merged as the same underlying issue." + "reason": "Different underlying bugs: one concerns `remove_unused_column` handling, the other FLOPS counting with custom trainers. Shared Trainer/custom trainer context is too broad to treat as the same issue." } ] }, @@ -15651,8 +15720,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43684-2", "nodes": [ @@ -15662,7 +15731,7 @@ "soft_pairs": [ "issue:43684|issue:44405" ], - "prepared_review_unit_hash": "9a9403447ca689c684572748acf97cde2a3631de63d67c81f68089e8774c9db1", + "prepared_review_unit_hash": "111108449c22b443adb6fb6853720c22eda422fdeb4ec4ac7e7ed0c6ee362052", "component_budget": { "node_count": 2, "item_count": 2, @@ -15684,18 +15753,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related auto-model feature requests, but they target different model families and different capabilities: Qwen3-Omni registration for AutoModel/AutoModelForConditionalGeneration versus Qwen3.5 AutoModelForSequenceClassification support. They do not look like duplicates.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "These are two separate feature requests for different Qwen model families and different AutoModel heads. They are related only by a broad theme of model registration, not by the same underlying bug or change.", + "confidence": 0.94, + "canonical_issue_reason": "Issue #43684 is the broader, more general request about registering Qwen3-Omni with AutoModel and AutoModelForConditionalGeneration, so it is the better anchor if this cluster is used as a loose topical grouping.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Issue #43684 is the most representative issue in the pair because it describes a broader registration gap, while #44405 is a distinct request for AutoModelForSequenceClassification support on Qwen3.5.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43684", "right": "issue:44405", "accept": false, - "reason": "Different underlying changes: one is generic registration support for Qwen3-Omni, the other is task-specific sequence classification support for Qwen3.5/Qwen3_5Config. Similar area, but not the same bug or mergeable change." + "reason": "Different model families (Qwen3-Omni vs Qwen3.5) and different target heads (conditional generation vs sequence classification); not the same concrete fix." } ] }, @@ -15714,8 +15783,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43835-4", "nodes": [ @@ -15732,77 +15801,80 @@ "issue:43845|issue:43859", "issue:43846|issue:43859" ], - "prepared_review_unit_hash": "40795cb1c81b0677122066b8db925131904d7b8f7e10699348d9fd8f7ff7ef61", + "prepared_review_unit_hash": "5bc786159cff1dc22217e4a82e3f9732d407c4e8f8a2f54e4ef89453ecd7b958", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4265, - "estimated_input_tokens": 1067, - "estimated_eval_tokens": 2390 + "serialized_chars": 4263, + "estimated_input_tokens": 1066, + "estimated_eval_tokens": 2388 }, "budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4265, - "estimated_input_tokens": 1067, - "estimated_eval_tokens": 2390 + "serialized_chars": 4263, + "estimated_input_tokens": 1066, + "estimated_eval_tokens": 2388 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is dominated by near-empty, generic issue titles with no actionable details. Three issues are effectively identical in form (\u201chuggingface \u201d), while one (\u201cFett\u201d) is unrelated. I would consolidate the huggingface reports under the earliest one, but the evidence is too thin to treat any pair as a true duplicate with confidence.", - "confidence": 0.63, - "canonical_issue_reason": "Issue 43845 is the earliest of the nearly identical \u2018huggingface \u2019 reports and is the most reasonable representative for the cluster.", + "summary": "Three closed issues share the same placeholder title \"huggingface \" and are likely duplicates; the \"Fett\" issue looks unrelated. No PRs are present.", + "confidence": 0.78, + "canonical_issue_reason": "Issue 43845 is the earliest of the near-identical \"huggingface \" reports, so it is the best representative for that duplicate cluster.", "canonical_pr_reason": null, - "best_issue_reason": "43845 is the best issue anchor because it is earliest among the repeated \u2018huggingface \u2019 submissions and matches the cluster\u2019s dominant pattern.", + "best_issue_reason": "43845 is the strongest single issue to keep because it is earliest, has the same title as the other likely duplicates, and there is no evidence that 43835 is related.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43835", "right": "issue:43845", "accept": false, - "reason": "Only weak text similarity; \u2018Fett\u2019 and \u2018huggingface\u2019 do not show the same underlying bug or change." + "reason": "Different titles with no supporting body or target overlap; \"Fett\" does not appear to be the same report as \"huggingface \"." }, { "left": "issue:43835", "right": "issue:43846", "accept": false, - "reason": "Only weak text similarity; no concrete shared problem is described." + "reason": "No evidence these refer to the same underlying issue; the titles are unrelated." }, { "left": "issue:43835", "right": "issue:43859", "accept": false, - "reason": "Only weak text similarity; unrelated-looking titles with no substantive issue details." + "reason": "No indication of the same bug/change; \"Fett\" is distinct from the repeated \"huggingface \" reports." }, { "left": "issue:43845", "right": "issue:43846", - "accept": false, - "reason": "Very similar placeholders, but the bodies are too sparse to verify they are the same underlying issue." + "accept": true, + "reason": "Identical title, very close creation times, and no contrary evidence; these look like duplicate reports of the same issue." }, { "left": "issue:43845", "right": "issue:43859", - "accept": false, - "reason": "Same broad title, but no concrete bug description or shared target to justify merging." + "accept": true, + "reason": "Same title and clear thematic overlap with the other \"huggingface \" issues; likely the same underlying report." }, { "left": "issue:43846", "right": "issue:43859", - "accept": false, - "reason": "Same broad title, but insufficient evidence of one concrete shared defect." + "accept": true, + "reason": "Identical title and close cluster context indicate the same duplicate issue." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: three issues share the exact same title, the earliest one is correctly chosen as the canonical representative, and the unrelated 'Fett' issue is kept separate. The soft-edge verdicts are conservative overall, and the note that no PRs are present is supported by the packet." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -15814,8 +15886,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-44371-5", "nodes": [ @@ -15836,7 +15908,7 @@ "issue:44572|issue:44937", "issue:44573|issue:44937" ], - "prepared_review_unit_hash": "dc176460818cee6d63b5360de7889a2d1934e945d8a7aaae0d5d3966be4c376a", + "prepared_review_unit_hash": "f212e9edab6f94a37e4fcf459fa2c61d7fe84c5c72c43de9d403e057e57aa075", "component_budget": { "node_count": 5, "item_count": 5, @@ -15858,73 +15930,76 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All items appear to be unrelated low-information/spam issues; none has enough concrete evidence to be treated as a duplicate of another. No PRs are present.", - "confidence": 0.86, - "canonical_issue_reason": "Issue 44418 is the best representative only because its title is slightly more descriptive than the spam/placeholder titles; it still lacks enough substance to justify a true canonical duplicate target.", + "summary": "Three issues are exact title matches for \"\" and look like the same spam/duplicate report; the documentation and Google LLC titles do not match that pattern.", + "confidence": 0.91, + "canonical_issue_reason": "issue:44371 is the earliest exact \"\" entry and is the cleanest representative of the repeated spam cluster.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44418 is the least spam-like and most human-readable issue title in the set, so it is the best single representative for triage purposes.", + "best_issue_reason": "issue:44371 best represents the cluster because it is an exact match for the repeated \"\" issues and appears first chronologically.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44371", "right": "issue:44572", - "accept": false, - "reason": "Both are placeholder spam titles with no body detail, targets, filenames, or other evidence of the same underlying issue." + "accept": true, + "reason": "Same exact title \"\", same low-information spam pattern, so likely duplicate reports of the same artifact." }, { "left": "issue:44371", "right": "issue:44573", - "accept": false, - "reason": "Identical generic titles alone are not enough; there is no concrete evidence they describe the same bug or change." + "accept": true, + "reason": "Same exact title \"\" and same spam-like placeholder content." }, { "left": "issue:44371", "right": "issue:44418", "accept": false, - "reason": "A spam placeholder and a documentation-suggestion title do not indicate the same underlying report." + "reason": "Different title and different issue type; documentation enhancement is not the same as the spam placeholder issue." }, { "left": "issue:44371", "right": "issue:44937", "accept": false, - "reason": "The titles are unrelated and there is no shared technical context to justify duplicate treatment." + "reason": "\"Check out \\\"Google LLC\\\"\" is a different spam/phishing-style title, not the same exact issue as \"\"." }, { "left": "issue:44418", "right": "issue:44572", "accept": false, - "reason": "A documentation enhancement request and a spam placeholder are not the same issue." + "reason": "Documentation enhancement suggestion does not match the exact spam placeholder issue." }, { "left": "issue:44418", "right": "issue:44573", "accept": false, - "reason": "No concrete overlap beyond low-information metadata; this is not enough to accept a duplicate edge." + "reason": "Different title and different underlying request; not the same issue." }, { "left": "issue:44572", "right": "issue:44573", - "accept": false, - "reason": "Matching spam-like titles do not establish the same underlying bug or change without additional context." + "accept": true, + "reason": "Exact title match \"\" with no other distinguishing information, so they appear to be duplicates." }, { "left": "issue:44572", "right": "issue:44937", "accept": false, - "reason": "These titles are unrelated and there is no shared code-path or explicit linkage." + "reason": "Different spam wording; no evidence they report the same underlying issue." }, { "left": "issue:44573", "right": "issue:44937", "accept": false, - "reason": "The records have no overlapping technical details; they should not be merged as duplicates." + "reason": "Different titles and no shared concrete problem beyond generic spam-like wording." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: three items share the exact title \"\" and the other two clearly do not. The soft-edge accepts are conservative enough given the exact title matches and matching low-information metadata, and the rejects avoid overclaiming on the non- issues." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -15936,8 +16011,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45145-2", "nodes": [ @@ -15947,12 +16022,12 @@ "soft_pairs": [ "issue:45145|issue:45151" ], - "prepared_review_unit_hash": "cf494af436be5616134f7b713cf8520a40d6e98ce87ce7f849d38300cf5fd9ce", + "prepared_review_unit_hash": "45871a8dfbb2df43a6ce827b7155e31a70bb22f7eecdbfb9434606a3e1ed5de2", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1645, + "serialized_chars": 1646, "estimated_input_tokens": 412, "estimated_eval_tokens": 1080 }, @@ -15960,7 +16035,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1645, + "serialized_chars": 1646, "estimated_input_tokens": 412, "estimated_eval_tokens": 1080 }, @@ -15969,28 +16044,28 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The packet supports that issues 45145 and 45151 are highly similar, with the same title and a soft-similarity match. It does not establish that they are exact duplicates or that their bodies are identical. Issue 45145 is the slightly better canonical choice because it has an inbound reference, but the evidence is still limited.", + "summary": "Likely duplicate issue reports with identical titles and very similar body sizes; #45145 is the better canonical representative because it appeared first and already has an inbound reference.", "confidence": 0.84, - "canonical_issue_reason": "45145 is the better canonical issue only by a small margin: it has the same title as 45151 and one inbound reference, which makes it a slightly stronger anchor for the cluster. The evidence does not prove exact duplication.", + "canonical_issue_reason": "#45145 is the earliest copy and already has an inbound reference, so it is the best canonical issue among these likely duplicates.", "canonical_pr_reason": null, - "best_issue_reason": "45145 is the best global issue candidate because it is the most connected of the pair and has the same title as 45151. However, this is a soft cluster, so the choice is based on linkage strength rather than a confirmed duplicate relationship.", + "best_issue_reason": "#45145 is the best issue representative because it was created first and is already referenced by another issue.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45145", "right": "issue:45151", - "accept": false, - "reason": "This is only a soft-similarity match with deterministic_accept=false. The titles match, but the packet does not provide enough evidence to treat them as the same underlying issue or an exact duplicate." + "accept": true, + "reason": "Identical titles and very similar body lengths strongly suggest the same underlying report, though the bodies themselves are not shown." } ] }, "evaluator_result": { "accept": false, - "feedback": "The summary overstates the evidence. The packet only supports that the two issues share the same title and have a soft-similarity match; it does not prove they are exact duplicates or have the same body text. The soft-edge accept is also too strong given the evidence: the pair is only a soft-similarity match and deterministic_accept is false. The inbound reference on 45145 can justify it as a slightly better canonical candidate, but not a strong or exact duplicate claim." + "feedback": "The packet supports a strong similarity signal (identical titles, close body lengths, one inbound reference on #45145), but it does not directly show the issue bodies or prove the reports are the same. The writeup is a bit too definitive with phrases like \"near-exact duplicates\" and \"effectively the same issue report.\" A more conservative phrasing such as \"likely duplicate\" would be better grounded." }, "evaluator_used": true, "retried": true, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -16056,8 +16131,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45672-2", "nodes": [ @@ -16067,40 +16142,40 @@ "soft_pairs": [ "issue:45672|issue:45674" ], - "prepared_review_unit_hash": "b043f18b803f39253da7363028f9c77ae154fb10f33764ceb312c669bdc73c1f", + "prepared_review_unit_hash": "85a29aca2afae36fb3e827ddf01fd1cce50d95e4b10b94102bb8a0251fbd53ed", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1668, - "estimated_input_tokens": 417, - "estimated_eval_tokens": 1090 + "serialized_chars": 1669, + "estimated_input_tokens": 418, + "estimated_eval_tokens": 1092 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1668, - "estimated_input_tokens": 417, - "estimated_eval_tokens": 1090 + "serialized_chars": 1669, + "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 related only at a very broad level (4-bit inference / bitsandbytes), but they describe different bugs, different failure modes, and likely different fixes. They should not be merged as duplicates.", - "confidence": 0.94, - "canonical_issue_reason": "Issue 45674 is the more general and actionable report: it names the configuration trigger (`llm_int8_skip_modules`) and the resulting regression in 4-bit inference, making it a better representative issue for this cluster.", + "summary": "These are related only at a high level (both 4-bit inference failures), but they describe different bugs and failure modes, so they should not be deduplicated.", + "confidence": 0.96, + "canonical_issue_reason": "Issue 45674 is the clearer representative because it states a specific configuration regression in BitsAndBytesConfig with an explicit failure mode and broader actionable scope.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45674 is the best overall representative because it describes a concrete, reusable configuration bug rather than a model-specific symptom, and the repro/failure mode is clearer.", + "best_issue_reason": "Issue 45674 is the better standalone issue to keep as the canonical reference for this cluster since it describes a concrete, reproducible code-path problem rather than a model-specific symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45672", "right": "issue:45674", "accept": false, - "reason": "Do not dedupe: 45672 is a Gemma4/audio-module `torch.finfo()` TypeError on uint8 weights, while 45674 is a BitsAndBytesConfig regression where `llm_int8_skip_modules` clears the default `lm_head` exclusion and triggers an AssertionError. Same broad area, different code paths and fixes." + "reason": "Different underlying bugs: 45672 is a Gemma4/audio-module torch.finfo TypeError on uint8 weights, while 45674 is an llm_int8_skip_modules / lm_head exclusion AssertionError in 4-bit inference. Same broad area, but not the same fixable code-path." } ] }, @@ -16119,8 +16194,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45762-2", "nodes": [ @@ -16130,29 +16205,29 @@ "soft_pairs": [ "issue:45762|issue:45763" ], - "prepared_review_unit_hash": "52b48b2ce0b96c9bf2b5c2bd05522b0a03c3757bc988decb84346ac41de10946", + "prepared_review_unit_hash": "62b653a03f02e89e38f8b1ac4a30309d040ed04368ae7665f31337cfa82dc629", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1476, - "estimated_input_tokens": 369, - "estimated_eval_tokens": 994 + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1476, - "estimated_input_tokens": 369, - "estimated_eval_tokens": 994 + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues do not look like duplicates of the same underlying bug or request; they are just two separate closed issues with different personal-name titles and dates and no supporting body text.", + "summary": "The two closed issues are just separate personal-name entries with different names/dates and no shared bug, feature, or code context, so this is not a duplicate cluster.", "confidence": 0.98, "canonical_issue_reason": null, "canonical_pr_reason": null, @@ -16163,7 +16238,7 @@ "left": "issue:45762", "right": "issue:45763", "accept": false, - "reason": "Different titles/names and dates, no body, no shared bug description or explicit linkage; this does not indicate the same underlying issue." + "reason": "Same broad surname pattern only; no body, targets, filenames, or discussion linking them to the same underlying issue." } ] }, @@ -16182,8 +16257,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45764-2", "nodes": [ @@ -16193,7 +16268,7 @@ "soft_pairs": [ "issue:45764|issue:45765" ], - "prepared_review_unit_hash": "3d886d2092974221bd5effb3ded5b764e5381609f7336d09af99a1fb6494aa52", + "prepared_review_unit_hash": "d2a692d14b0f6a220670e9ef414cc2c352fe070e58015763ee8a18fb98b62cc4", "component_budget": { "node_count": 2, "item_count": 2, @@ -16215,18 +16290,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two closed issues appear unrelated: both are bare personal-name/date titles with no body or shared technical context, so they should not be clustered as duplicates.", + "summary": "These are two unrelated, empty closed issues with personal-name/date titles and no substantive content. They do not appear to describe the same bug or request, so they should not be clustered as duplicates.", "confidence": 0.98, - "canonical_issue_reason": null, + "canonical_issue_reason": "No canonical issue: neither item contains actionable problem description, and the titles are unrelated personal-name/date strings rather than the same report.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Neither issue is a good cluster anchor because both are effectively empty/spam-like and provide no meaningful defect or feature context.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45764", "right": "issue:45765", "accept": false, - "reason": "The titles refer to different people and different dates, with no shared bug report, code path, or issue context." + "reason": "Reject: the titles reference different people and different dates, and there is no body content or shared technical context indicating the same underlying issue." } ] }, @@ -16285,7 +16360,7 @@ "pull_request:43823|pull_request:45795", "pull_request:43251|pull_request:43254" ], - "prepared_review_unit_hash": "89c73c8ba9e08bb158f968e0590ca6071f3427f941b3b30a742a310d6c241f26", + "prepared_review_unit_hash": "73a5e5c302f1fa1634d1bc3111598f4913e698dbd6cb4de6353f8f2121035ab5", "component_budget": { "node_count": 13, "item_count": 13, @@ -16307,135 +16382,135 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Mostly unrelated PRs with one real duplicate-like pair in `fixed_cross_entropy`. The loss-helper pair (43251/43254) is the only strong merge candidate; the rest are different models, docs, tests, or subsystems.", - "confidence": 0.94, + "summary": "This cluster is mostly a grab bag of unrelated PRs across model additions, docs, and refactors. The only true duplicate pair is the two fixed_cross_entropy changes (43251/43254), which implement the same kwargs pass-through to nn.functional.cross_entropy.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43251 is a reasonable canonical representative for the loss-helper pair because it makes the same `fixed_cross_entropy` change in `src/transformers/loss/loss_utils.py` as PR 43254 and both target issue 43240. The packet does not clearly justify a stronger preference beyond that.", + "canonical_pr_reason": "PR 43251 is the better canonical choice for the duplicate loss fix: it covers the same fixed_cross_entropy change as 43254, is a bit more complete, and explicitly links the tracking issue 43240.", "best_issue_reason": null, - "best_pr_reason": "PR 43251 is a suitable global representative only for the one duplicate-like fix in this cluster: it and PR 43254 patch the same helper with the same supported kwargs in the same file, so either could stand in for that pair.", + "best_pr_reason": "PR 43251 is the best representative of the only real duplicate pair in this cluster; the rest are unrelated, and 43251 is the more complete version of the shared cross-entropy kwargs fix.", "soft_edge_verdicts": [ { "left": "pull_request:39895", "right": "pull_request:43424", "accept": false, - "reason": "VideoPrism integration vs an ExecuTorch dynamic-shape test; different changes and no shared concrete bug." + "reason": "Different changes: VideoPrism model addition vs an executorch export test." }, { "left": "pull_request:41855", "right": "pull_request:43424", "accept": false, - "reason": "Mistral tokenizer helper additions vs ExecuTorch export testing; unrelated code paths." + "reason": "Unrelated areas: Mistral tokenizer methods vs executorch dynamic-shape testing." }, { "left": "pull_request:41855", "right": "pull_request:43823", "accept": false, - "reason": "Tokenizer method fixes vs adding MobileLLM; same broad area only, not the same fix." + "reason": "Different workstreams: tokenizer API additions vs adding MobileLLM." }, { "left": "pull_request:42668", "right": "pull_request:43823", "accept": false, - "reason": "Processor/auto-registration work for different models; not the same underlying change." + "reason": "Processor auto-loading changes are not the same underlying change as adding MobileLLM." }, { "left": "pull_request:43251", "right": "pull_request:43823", "accept": false, - "reason": "Loss helper kwargs fix vs new model implementation; no common code-path problem." + "reason": "Cross-entropy kwargs fix and MobileLLM model addition are unrelated." }, { "left": "pull_request:43251", "right": "pull_request:44827", "accept": false, - "reason": "Cross-entropy helper fix vs Mistral4 model/test changes; different bugs." + "reason": "Loss helper kwargs fix is unrelated to Mistral4 test fixes and model wiring." }, { "left": "pull_request:43424", "right": "pull_request:44827", "accept": false, - "reason": "ExecuTorch dynamic-shape test vs Mistral4 fixes; same general testing theme only." + "reason": "Executorch export tests and Mistral4 test fixes address different code paths." }, { "left": "pull_request:43424", "right": "pull_request:43823", "accept": false, - "reason": "ExecuTorch export test vs MobileLLM addition; unrelated." + "reason": "Different artifacts: executorch export test vs new MobileLLM model implementation." }, { "left": "pull_request:43823", "right": "pull_request:45655", "accept": false, - "reason": "New model addition vs `cls.config` resolution fix; different subsystems." + "reason": "New model addition is unrelated to config resolution order in modeling_utils." }, { "left": "pull_request:43823", "right": "pull_request:43995", "accept": false, - "reason": "MobileLLM implementation vs Falcon refactor; not duplicates." + "reason": "MobileLLM implementation and Falcon output-interface refactor are different changes." }, { "left": "pull_request:43995", "right": "pull_request:44827", "accept": false, - "reason": "Falcon output-collection refactor vs Mistral4 test/model work; different models and goals." + "reason": "Falcon refactor and Mistral4 test fixes do not target the same bug." }, { "left": "pull_request:43995", "right": "pull_request:45213", "accept": false, - "reason": "Falcon refactor vs a model-creation skill PR; completely different artifact types." + "reason": "Model refactor vs model-creation skill content are not the same underlying change." }, { "left": "pull_request:44827", "right": "pull_request:45170", "accept": false, - "reason": "Mistral4 fixes vs CLIP-like `layrnorm` renames; different underlying issues." + "reason": "Mistral4 test fixes and layrnorm->layernorm renaming are different fixes." }, { "left": "pull_request:44827", "right": "pull_request:45795", "accept": false, - "reason": "Mistral4 work vs audio/video processor docs; not the same change." + "reason": "Mistral4 test fixes are unrelated to docs for audio/video processors." }, { "left": "pull_request:44827", "right": "pull_request:45655", "accept": false, - "reason": "Model-specific fixes vs core config-resolution ordering; no shared code-path bug." + "reason": "Different problems: Mistral4 tests vs config resolution order bug." }, { "left": "pull_request:44827", "right": "pull_request:45213", "accept": false, - "reason": "Model-specific fixes vs repo skill/dataset PR; unrelated." + "reason": "Test-fix PR and model-creation skill PR are unrelated." }, { "left": "pull_request:43823", "right": "pull_request:45213", "accept": false, - "reason": "New model implementation vs draft model-creation skill/docs; same domain only." + "reason": "Adding MobileLLM is not the same underlying change as adding a new-model skill doc." }, { "left": "pull_request:43823", "right": "pull_request:45795", "accept": false, - "reason": "Model implementation vs docs for audio/video processors; related topic, not a duplicate." + "reason": "MobileLLM model addition and docs for processors are different topics." }, { "left": "pull_request:43251", "right": "pull_request:43254", "accept": true, - "reason": "Both PRs change `src/transformers/loss/loss_utils.py` in the same `fixed_cross_entropy` helper, add the same kwargs (`weight`, `label_smoothing`), and target the same issue 43240. They are the same underlying fix and could be merged into one PR." + "reason": "Both PRs patch the same helper, fixed_cross_entropy, to forward the same supported kwargs to nn.functional.cross_entropy; they are mergeable into one change." } ] }, "evaluator_result": { - "accept": false, - "feedback": "Reject: the canonical PR rationale overstates the packet. PR 43254 also has explicit issue target 43240, so it does not \"lack the issue linkage.\" The claim that 43251 is broader/cleaner is also not directly grounded in the packet. The accepted duplicate pair itself looks mergeable, but the summary should be more conservative." + "accept": true, + "feedback": "The analysis is grounded in the packet and stays conservative. The only accepted soft edge, 43251/43254, is well supported by shared file/issue target and the same fixed_cross_entropy kwargs fix, and the other pairs are reasonably rejected as unrelated." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -16459,12 +16534,12 @@ "soft_pairs": [ "pull_request:40546|pull_request:42781" ], - "prepared_review_unit_hash": "582e4f4f7ae86a8f5aa999749fca767e875d77206a05f6c275d78d4defc91332", + "prepared_review_unit_hash": "186b72b8074839b415d5d697acda9b17f5abdefadee07ba752039a23e06dfa21", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 7226, + "serialized_chars": 7225, "estimated_input_tokens": 1807, "estimated_eval_tokens": 3870 }, @@ -16472,7 +16547,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 7226, + "serialized_chars": 7225, "estimated_input_tokens": 1807, "estimated_eval_tokens": 3870 }, @@ -16481,18 +16556,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs are about VibeVoice, but they target different deliverables: one adds the base VibeVoice model integration, while the other adds VibeVoice Realtime and an acoustic tokenizer with additional generation/runtime changes. They overlap on shared model scaffolding, but not on the same concrete change.", - "confidence": 0.93, + "summary": "Both pull requests are about VibeVoice, but they are not duplicates: one is the initial VibeVoice integration and the other adds VibeVoice Realtime plus an acoustic tokenizer and related generation/model changes.", + "confidence": 0.87, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #40546 is the better canonical PR because it is the foundational VibeVoice implementation and broader base integration; the other PR builds a separate realtime variant on top of it.", + "canonical_pr_reason": "PR 40546 is the broader original VibeVoice implementation and appears to be the primary integration PR; PR 42781 is a later, more specialized follow-up for the realtime variant.", "best_issue_reason": null, - "best_pr_reason": "PR #40546 is the best representative of the cluster because it covers the core model addition and is the more central, non-draft implementation.", + "best_pr_reason": "PR 40546 is the best representative of the cluster because it introduces the core VibeVoice support that the other PR builds on.", "soft_edge_verdicts": [ { "left": "pull_request:40546", "right": "pull_request:42781", "accept": false, - "reason": "Reject: they share the VibeVoice family and some files, but one is the base model integration and the other is a realtime/acoustic-tokenizer extension with different code paths and scope. They do not look like the same mergeable fix/change." + "reason": "They share the VibeVoice subsystem, but 42781 adds a distinct realtime/acoustic-tokenizer feature set rather than fixing or implementing the same concrete code path as 40546. These are related follow-up changes, not one mergeable duplicate PR." } ] }, @@ -16522,12 +16597,12 @@ "soft_pairs": [ "pull_request:40755|pull_request:41224" ], - "prepared_review_unit_hash": "288f99592cd9551e0c534578064187a69297a2b7027bcd7d01c25f029b5f9df3", + "prepared_review_unit_hash": "9c294abaa62b509f6b7053e1b10763b86dd7da870fb1d14a0563baa59f52460f", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4676, + "serialized_chars": 4674, "estimated_input_tokens": 1169, "estimated_eval_tokens": 2594 }, @@ -16535,7 +16610,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4676, + "serialized_chars": 4674, "estimated_input_tokens": 1169, "estimated_eval_tokens": 2594 }, @@ -16544,18 +16619,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two pull requests are unrelated: one adds TimesFM covariate forecasting, the other adds DINOv3 image classification support. They touch different models, code paths, and documentation, so this is not a duplicate cluster.", - "confidence": 0.99, + "summary": "The two pull requests are about different features in different models: TimesFM covariate forecasting vs. DINOv3 image-classification support. They share only a loose \u201cmodel feature addition\u201d theme, not the same code-path or bug, so they are not duplicates.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "No true canonical PR for this cluster because the PRs target unrelated models and independent changes. If forced to pick a representative, PR 40755 is broader and more feature-complete, but it is not a duplicate of PR 41224.", "best_issue_reason": null, - "best_pr_reason": "PR #40755 is the broader, more substantial feature addition in this cluster, but it is not a duplicate of #41224 and should not be treated as a canonical match.", + "best_pr_reason": "PR 40755 is the stronger standalone representative of the cluster only because it has the larger implementation footprint and includes docs/tests for a concrete feature addition. However, it should not be merged with PR 41224 as a duplicate.", "soft_edge_verdicts": [ { "left": "pull_request:40755", "right": "pull_request:41224", "accept": false, - "reason": "Different models and features (TimesFM covariates vs. DINOv3 image classification); no shared underlying bug or mergeable change." + "reason": "Different models and different functionality: TimesFM covariate forecasting vs. DINOv3 image classification. No shared code-path or same underlying bug/change." } ] }, @@ -16650,7 +16725,7 @@ "pull_request:41797|pull_request:45186", "pull_request:45186|pull_request:45630" ], - "prepared_review_unit_hash": "742e11c30f1d1eb9c7d3b835a1b501e3ea127e469d11f7e9abdad35eda54cf1f", + "prepared_review_unit_hash": "51307f7b5670046e28b0a785f8298bdea7ec719829f47b5014f713358b1433a5", "component_budget": { "node_count": 3, "item_count": 3, @@ -16672,24 +16747,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three separate model-addition PRs (DeepSeek OCR, Isaac, Kimi2-6). They share generic Transformers boilerplate and registry/docs updates, but not the same underlying model or code-path change.", - "confidence": 0.97, + "summary": "All three items are separate model-addition PRs for different architectures (DeepSeek OCR, Isaac, Kimi2-6). They share boilerplate integration files, but not the same underlying change, so they should not be clustered as duplicates.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the cluster is heterogeneous, with each PR introducing a different model integration rather than duplicating the same change.", + "canonical_pr_reason": "PR #45630 is the strongest cluster representative only in the sense that it is the most mature and comprehensive of the three, with the highest review activity and the broadest set of integration updates.", "best_issue_reason": null, - "best_pr_reason": "PR 45630 is the most developed integration by discussion/review activity and scope, so it is the best representative of the cluster, but it is still not a duplicate of the others.", + "best_pr_reason": "PR #45630 is the best single representative if one must be chosen: it is a fully fleshed-out model addition with extensive integration coverage and substantial review attention.", "soft_edge_verdicts": [ { "left": "pull_request:41797", "right": "pull_request:45186", "accept": false, - "reason": "Different model integrations (DeepSeek OCR vs Isaac). The overlap is limited to standard new-model scaffolding in docs, auto mappings, and utility updates, not the same bug or feature." + "reason": "Different model families and different code paths: DeepSeek OCR vs Isaac. Shared auto-registration/docs scaffolding is boilerplate, not evidence of the same change." }, { "left": "pull_request:45186", "right": "pull_request:45630", "accept": false, - "reason": "Different model integrations (Isaac vs Kimi2-6). Shared registry/docs churn is expected for new models and does not indicate a common underlying code-path change." + "reason": "These add different models (Isaac vs Kimi2-6) with distinct configurations, processors, and tests. Similar framework integration files do not make them mergeable into one PR." } ] }, @@ -16719,7 +16794,7 @@ "soft_pairs": [ "pull_request:42975|pull_request:42976" ], - "prepared_review_unit_hash": "a3fc123a430a9441552d609af6d627312dbe0c7aba616f0b8d026cc87d346c2a", + "prepared_review_unit_hash": "497a04fc187db1ffa0866539390949573186db0b32fbf7d3d5ea9490d24b1fac", "component_budget": { "node_count": 2, "item_count": 2, @@ -16741,18 +16816,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two workflow-maintenance PRs that both upgrade GitHub Actions, but they target different actions and different goals. They are related by theme, not the same underlying change, so they should not be deduplicated.", - "confidence": 0.95, + "summary": "Two separate workflow-maintenance PRs: one updates multiple GitHub Actions for Node 24 compatibility, the other upgrades Docker build actions to v6. They overlap in workflow files but address different dependencies and problems.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "No duplicate canonical PR: the two PRs address different action upgrades and are not the same mergeable change.", + "canonical_pr_reason": "PR 42975 is the better canonical representative because it has the clearest, broader purpose (Node 24 compatibility) and covers a wider set of GitHub Actions updates.", "best_issue_reason": null, - "best_pr_reason": "PR 42975 is the better representative of the cluster because it is the broader workflow-actions compatibility update (checkout/cache/upload-artifact) and explicitly targets Node 24 compatibility.", + "best_pr_reason": "PR 42975 is the strongest representative for the cluster since its title and diff clearly describe the main class of change: repository-wide GitHub Actions version upgrades for compatibility.", "soft_edge_verdicts": [ { "left": "pull_request:42975", "right": "pull_request:42976", "accept": false, - "reason": "Both modify GitHub Actions workflows, but 42975 upgrades checkout/cache/upload-artifact for Node 24 compatibility while 42976 bumps docker/build-push-action to v6. Different actions, different concrete change, not the same underlying bug/fix." + "reason": "These are not the same underlying change: 42975 updates checkout/cache/upload-artifact across workflows for Node 24 compatibility, while 42976 only bumps docker/build-push-action to v6 in docker-image workflows. Overlapping files alone is not enough to treat them as duplicates." } ] }, @@ -16795,12 +16870,12 @@ "pull_request:44601|pull_request:45838", "pull_request:45490|pull_request:45838" ], - "prepared_review_unit_hash": "ced02bef74514f2010a2792937efbb35b97d750f38f48d45a37a624086090049", + "prepared_review_unit_hash": "85c28a5c33abc32549e25199ab3f67c2b6394446237b38dc44512b0736ed7bb7", "component_budget": { "node_count": 7, "item_count": 7, "soft_pair_count": 9, - "serialized_chars": 23514, + "serialized_chars": 23515, "estimated_input_tokens": 5879, "estimated_eval_tokens": 12014 }, @@ -16808,7 +16883,7 @@ "node_count": 7, "item_count": 7, "soft_pair_count": 9, - "serialized_chars": 23514, + "serialized_chars": 23515, "estimated_input_tokens": 5879, "estimated_eval_tokens": 12014 }, @@ -16817,66 +16892,66 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are mostly unrelated model additions or infrastructure work; the high soft-similarity scores come from shared Hugging Face boilerplate, docs, and auto-registration files rather than the same underlying change. I found no true duplicate PR pair in this cluster.", - "confidence": 0.97, + "summary": "This cluster is heterogeneous: several unrelated model-addition PRs plus one distributed pipeline-parallel PR. They share repository plumbing patterns (docs, auto mappings) but do not look like duplicates of the same bug or change.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR for dedupe: the cluster spans separate changes for ViT NEPA, Molmo, Molmo2, LFM2 audio, distributed pipeline-parallel loading, CTSM, and PP-OCRv6, with only template/docs overlap.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "No PR is a strong canonical duplicate target here; if a representative must be chosen, 43451 is the most substantial standalone model-addition PR, but it is not a duplicate of the others.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:42978", "right": "pull_request:43448", "accept": false, - "reason": "Different model additions (ViT NEPA vs Molmo); overlap is only in docs and auto-mapping boilerplate." + "reason": "Different model integrations (ViT NEPA vs Molmo); shared docs/auto-registry edits are generic model-onboarding boilerplate, not the same change." }, { "left": "pull_request:42978", "right": "pull_request:43451", "accept": false, - "reason": "Different model families (ViT NEPA vs Molmo2) and distinct implementation paths; shared files are generic registration/docs." + "reason": "Different model additions (ViT NEPA vs Molmo2) with different files, configs, and tests; not one underlying bug or feature." }, { "left": "pull_request:42978", "right": "pull_request:45490", "accept": false, - "reason": "ViT NEPA model addition and CTSM time-series model addition are unrelated changes." + "reason": "ViT NEPA and CTSM are distinct model-support PRs; overlap is only in shared registration/docs scaffolding." }, { "left": "pull_request:43448", "right": "pull_request:43451", "accept": false, - "reason": "Molmo and Molmo2 are related by name but are separate model releases with different code paths, not the same bug/change." + "reason": "Molmo and Molmo2 are related by naming but are separate model families with different codepaths and artifacts; not mergeable as one PR." }, { "left": "pull_request:43448", "right": "pull_request:45490", "accept": false, - "reason": "Molmo vs CTSM are unrelated model additions; only broad docs/auto files overlap." + "reason": "Molmo vs CTSM are unrelated model additions; same broad subsystem, but not the same concrete change." }, { "left": "pull_request:43451", "right": "pull_request:45490", "accept": false, - "reason": "Molmo2 and CTSM touch different subsystems and model implementations, so they are not duplicates." + "reason": "Molmo2 and CTSM are different model introductions with no shared bug or implementation target." }, { "left": "pull_request:43973", "right": "pull_request:45838", "accept": false, - "reason": "Lfm2 audio support and PP-OCRv6 text recognition support are unrelated model families with different code paths." + "reason": "Lfm2 Audio and PP-OCRv6 are unrelated model-support PRs; both touch docs/auto wiring, but that is not enough to treat them as duplicates." }, { "left": "pull_request:44601", "right": "pull_request:45838", "accept": false, - "reason": "Distributed pipeline-parallel loading infrastructure is unrelated to PP-OCRv6 model support; shared similarity is incidental." + "reason": "Pipeline-parallel core loading support is a distributed-systems change, while PP-OCRv6 is a model support PR; they do not fix the same code-path problem." }, { "left": "pull_request:45490", "right": "pull_request:45838", "accept": false, - "reason": "CTSM time-series model support and PP-OCRv6 OCR model support are unrelated changes." + "reason": "CTSM and PP-OCRv6 are different model additions; shared repo plumbing does not indicate the same underlying change." } ] }, @@ -16909,7 +16984,7 @@ "pull_request:43291|pull_request:44815", "pull_request:43665|pull_request:44815" ], - "prepared_review_unit_hash": "cf40429a6a401b024316c5456f4429663061c9d1a3210e5e7a54e1a166b2f7fd", + "prepared_review_unit_hash": "29b0f7d250b5dce5d94af3fd00f266b959de8dcb9e623bd2930654d4e8ad0ca4", "component_budget": { "node_count": 3, "item_count": 3, @@ -16931,30 +17006,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All three are pull requests, but they do not look like the same underlying change. PR 44815 is the only substantive bug-fix PR; 43291 is a Whisper tokenizer/test change, and 43665 is just placeholder test edits. No pair is close enough to merge as a duplicate.", - "confidence": 0.68, + "summary": "The three pull requests are not duplicates of one another. Two are mostly test-only changes, while the third is a substantive dequantization/loading fix in a different subsystem. No soft edge should be merged.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44815 is the strongest canonical candidate because it contains the only real implementation fix, spanning model loading and FP8 dequantization logic. The other two are test-focused and do not match its code path.", + "canonical_pr_reason": "pull_request:44815 is the most substantive and coherent PR: it fixes a concrete dequant/loading path and includes the corresponding implementation and test updates. The other two are narrow test edits without the same underlying code change.", "best_issue_reason": null, - "best_pr_reason": "PR 44815 is the best representative of the cluster: it has the clearest concrete bug fix, the broadest real code impact, and the most substantive diff. The others look like ancillary or unrelated test churn.", + "best_pr_reason": "pull_request:44815 is the best standalone PR because it addresses an actual code-path defect rather than only altering tests, and its scope is internally consistent across the touched files.", "soft_edge_verdicts": [ { "left": "pull_request:43291", "right": "pull_request:43665", "accept": false, - "reason": "Both touch tests, but 43291 also changes Whisper tokenizer behavior while 43665 is only placeholder assertions in unrelated model tests; they are not the same bug or change." + "reason": "Both are PRs with test changes, but they target different models and different behavior; there is no shared underlying bug or fix path." }, { "left": "pull_request:43291", "right": "pull_request:44815", "accept": false, - "reason": "43291 is a Whisper decoding/test fix, whereas 44815 is a dequantization/model-loading fix in a different subsystem; no shared concrete code-path problem." + "reason": "Whisper tokenization/test cleanup and FP8 dequant/loading fixes are unrelated changes in different subsystems." }, { "left": "pull_request:43665", "right": "pull_request:44815", "accept": false, - "reason": "43665 is clearly synthetic test churn, not a real implementation fix, so it is not the same underlying change as 44815." + "reason": "The CLIP/ViT test-only edits do not match the FP8 dequantization fix; they are not the same bug or mergeable as one PR." } ] }, @@ -16984,7 +17059,7 @@ "soft_pairs": [ "pull_request:43488|pull_request:43532" ], - "prepared_review_unit_hash": "8fe2f04d1e47126838a5f89e17d0e57c6ce649a512df17a8da4dcfb024707677", + "prepared_review_unit_hash": "5bee5ec7d4b18827eb0a26ce16231d4bbe814ea80d470732b83bb249052830ce", "component_budget": { "node_count": 2, "item_count": 2, @@ -17006,18 +17081,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates: PR #43488 is a synthetic repo-bot/checking change around setup/versioning and a Vit test, while PR #43532 rewires CI/workflow logic and touches several files. They only share a test filename, not the same underlying fix.", - "confidence": 0.96, + "summary": "The two pull requests are not duplicates: one is a synthetic bot-check/version-bump PR, while the other is a larger CI/testing workflow change with unrelated file edits.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #43532 is the better representative of the cluster because it is the more substantial, real change; PR #43488 is a placeholder-style bot check and does not match the same code path or change intent.", + "canonical_pr_reason": "PR 43532 is the better canonical representative because it contains the substantive workflow/testing changes; PR 43488 is a narrow repository-bot/test trigger PR and is not a good representative for the cluster.", "best_issue_reason": null, - "best_pr_reason": "PR #43532 is the most suitable overall candidate: it has broader, coherent changes across CI and helper code, whereas #43488 is a small synthetic setup/test tweak.", + "best_pr_reason": "PR 43532 is the most meaningful PR in the set, with broader and more realistic code changes across CI and test utilities. PR 43488 is clearly a non-merge bot-check change.", "soft_edge_verdicts": [ { "left": "pull_request:43488", "right": "pull_request:43532", "accept": false, - "reason": "Reject: the overlap is only a shared Vit test file. #43488 changes packaging/versioning and a test, while #43532 changes workflow triggers and related CI utilities; they do not fix the same concrete bug or belong in one PR." + "reason": "They do not appear to fix the same bug or implement the same change. 43488 is a version/dependency tweak for bot checking, while 43532 changes CI workflow and testing utility code plus related test behavior." } ] }, @@ -17047,12 +17122,12 @@ "soft_pairs": [ "pull_request:43612|pull_request:43613" ], - "prepared_review_unit_hash": "7de38e7685f087524a82468bcff6beeae8336e22406c4b289d9dbbc74dbfac70", + "prepared_review_unit_hash": "61504846732f3eaccb2ac117b573fc3f190a7473362c23696e8772008dcd10af", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6082, + "serialized_chars": 6083, "estimated_input_tokens": 1521, "estimated_eval_tokens": 3298 }, @@ -17060,7 +17135,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6082, + "serialized_chars": 6083, "estimated_input_tokens": 1521, "estimated_eval_tokens": 3298 }, @@ -17069,18 +17144,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two PRs are structurally similar pipeline additions, but they implement different features: Promptable Concept Segmentation vs Promptable Visual Segmentation. They share documentation/metadata scaffolding, yet the model docs, pipeline modules, and tests target different code paths and use different underlying model families.", + "summary": "These are two related but distinct PRs: both introduce new segmentation pipelines and share some registration/docs scaffolding, but they target different tasks, model families, and code paths.", "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "No canonical PR: 43612 adds Promptable Concept Segmentation, while 43613 adds Promptable Visual Segmentation; the overlap is mostly shared pipeline/documentation plumbing, not the same underlying change.", "best_issue_reason": null, - "best_pr_reason": null, + "best_pr_reason": "No single best PR for a duplicate cluster here; they are separate feature additions rather than one concrete bug fix or mergeable duplicate change.", "soft_edge_verdicts": [ { "left": "pull_request:43612", "right": "pull_request:43613", "accept": false, - "reason": "Reject: these are not the same underlying bug or change. They add different pipelines with different task semantics and different model/test coverage, so they would not plausibly be merged into one PR." + "reason": "Reject: both are new pipeline additions, but they implement different segmentation tasks with different model/test targets, so they are not the same underlying change and should not be merged as duplicates." } ] }, @@ -17110,7 +17185,7 @@ "soft_pairs": [ "pull_request:43651|pull_request:43663" ], - "prepared_review_unit_hash": "f062008cef8e56e673881cb4fb6961fc4641be5f95147c5ce3f00211e7b9ee6f", + "prepared_review_unit_hash": "9c35be50e5843627e3c42b5f0553aeba729f0636e8483de637ed7b81ec403ab2", "component_budget": { "node_count": 2, "item_count": 2, @@ -17132,18 +17207,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two PRs both touch `trainer.py` and add extension hooks for custom trainers, but they solve different problems: gradient-accumulation loss scaling vs. column filtering based on the model signature. They should not be deduplicated.", - "confidence": 0.98, + "summary": "Two open PRs in `trainer.py`, but they address different Trainer extension points: gradient-accumulation loss scaling vs. signature-column filtering for custom trainers/PEFT. They share a file and broad subsystem, but not the same underlying change.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the changes target different trainer behaviors and are not the same underlying fix.", + "canonical_pr_reason": "No single canonical PR: these are separate trainer customizations with different code paths and behavior changes.", "best_issue_reason": null, - "best_pr_reason": "Neither PR is a good global representative for a duplicate cluster because they are unrelated feature changes, not two versions of the same fix.", + "best_pr_reason": "PR #43651 is the more concrete correctness fix, since it changes actual loss-scaling behavior; PR #43663 is a separate extensibility refactor.", "soft_edge_verdicts": [ { "left": "pull_request:43651", "right": "pull_request:43663", "accept": false, - "reason": "Both modify `trainer.py`, but one adds `_loss_is_scaled_for_ga` for loss normalization and the other adds `_get_signature_columns` for dataset column filtering; they address different code paths and are not mergeable as a single duplicate PR." + "reason": "Different features and code paths: #43651 controls gradient-accumulation loss scaling, while #43663 adds an override hook for signature-column filtering. They are not the same bug/change and would not plausibly merge into one PR." } ] }, @@ -17175,52 +17250,52 @@ "pull_request:43656|pull_request:43842", "pull_request:43836|pull_request:43842" ], - "prepared_review_unit_hash": "f53ec09360f9bbba87f17d58c9446b6ebc6bb5503fa1c4f143337f326455a17f", + "prepared_review_unit_hash": "74962995867f55bdf88a3863ad7fb3b7e19e38d3cc68d86af2132354c7ff75d3", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 5993, - "estimated_input_tokens": 1499, - "estimated_eval_tokens": 3254 + "serialized_chars": 5992, + "estimated_input_tokens": 1498, + "estimated_eval_tokens": 3252 }, "budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 5993, - "estimated_input_tokens": 1499, - "estimated_eval_tokens": 3254 + "serialized_chars": 5992, + "estimated_input_tokens": 1498, + "estimated_eval_tokens": 3252 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "43842 is the cleanest representative of the serve.py TypeAdapter NameError fix. 43836 is a plausible duplicate/near-duplicate because it targets the same file and the same issue area, but 43656 is broader and less directly grounded as the same fix.", - "confidence": 0.78, + "summary": "PR #43842 is the cleanest representative of the TypeAdapter NameError fix in `src/transformers/cli/serve.py`. PR #43836 is plausibly the same underlying issue and shares issue target #43824, but PR #43656 is only loosely similar and is tied to a different target (#43576), so it should not be merged into the same duplicate cluster with the others.", + "confidence": 0.74, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43842 is the most targeted patch for the TypeAdapter NameError in `src/transformers/cli/serve.py`, with a minimal diff and a direct issue link.", + "canonical_pr_reason": "PR #43842 is the most direct and narrowly scoped fix for the TypeAdapter NameError, with an explicit link to issue #43824 and no extra refactoring beyond the bug fix.", "best_issue_reason": null, - "best_pr_reason": "PR 43842 is the best cluster representative because it fixes the concrete TypeAdapter annotation problem with the least extra surface area.", + "best_pr_reason": "PR #43842 is the best overall representative because it addresses the concrete runtime typing failure with the smallest focused change and the clearest issue linkage.", "soft_edge_verdicts": [ { "left": "pull_request:43656", "right": "pull_request:43842", "accept": false, - "reason": "Both touch `serve.py`, but 43656\u2019s `__future__`/annotation rewrite is broader and its explicit issue target differs, so the evidence is not strong enough to call it the same concrete fix as 43842." + "reason": "These only share broad similarity in `serve.py`; they point to different issue targets (#43576 vs #43824), so the packet does not support treating them as the same bugfix or one mergeable PR." }, { "left": "pull_request:43836", "right": "pull_request:43842", "accept": true, - "reason": "These are directly aligned on the same `serve.py` TypeAdapter NameError path, and 43836\u2019s broader annotation cleanup still appears to be addressing the same runtime typing issue." + "reason": "Both target the same `serve.py` TypeAdapter NameError and share issue target #43824, so they plausibly fix the same underlying code-path problem despite different implementation styles." } ] }, "evaluator_result": { "accept": false, - "feedback": "The summary overstates what the packet supports. All three PRs touch the same file, but the evidence does not establish that they are all the same TypeAdapter NameError fix: 43656\u2019s preview shows a broader `__future__`/annotation rewrite and it has a different explicit issue target, so calling it a close variant of 43842 is too strong. The 43836\u219443842 pair is better supported, but the analyst still presents the cluster as one uniform bugfix without enough grounding. Please narrow the claims to what is directly evidenced and be more conservative on the soft-edge pair for 43656." + "feedback": "The writeup overstates the evidence for PR #43656 vs #43842. Those two only share soft similarity and even point to different explicit issue targets (43576 vs 43824), so the packet does not firmly support treating them as the same bugfix or as cleanly mergeable into one PR. #43836 vs #43842 is better grounded because they share issue target 43824, but the summary still collapses all three PRs into one fix more strongly than the packet justifies." }, "evaluator_used": true, "retried": true, @@ -17409,214 +17484,232 @@ "pull_request:44044|pull_request:44129", "pull_request:44044|pull_request:44154" ], - "prepared_review_unit_hash": "394c9d1e321da9e348237be25e57d6ad2fb10665530a1ce559509a9fb7313438", + "prepared_review_unit_hash": "b51f864156a54e40a217adeb8426dd54ef126b08d3e0e9fa90c1b3ebca12470d", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 78, - "serialized_chars": 84904, - "estimated_input_tokens": 21226, - "estimated_eval_tokens": 42708 + "serialized_chars": 84896, + "estimated_input_tokens": 21224, + "estimated_eval_tokens": 42704 }, "budget": { "node_count": 27, "item_count": 27, "soft_pair_count": 72, - "serialized_chars": 75099, - "estimated_input_tokens": 18775, - "estimated_eval_tokens": 37806 + "serialized_chars": 75093, + "estimated_input_tokens": 18774, + "estimated_eval_tokens": 37804 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mainly a shared output-tracing refactor stream across many models. The common umbrella is issue #43979, but only a few pairs are true duplicates at the PR level: same model/file and same concrete output-capture change.", - "confidence": 0.89, - "canonical_issue_reason": "Shared tracker #43979 is the umbrella issue for the standardized output-tracing migration across these PRs.", - "canonical_pr_reason": "PR 44066 is the best representative merge anchor: it explicitly targets #43979, makes a concrete GPT-J/CodeGen output-tracing refactor, and has the clearest direct duplicate overlap with 44722 and 44085.", - "best_issue_reason": "Issue #43979 is the global umbrella for this refactor line and is the natural canonical issue for deduping the related PRs.", - "best_pr_reason": "PR 44066 is the strongest single PR representative because it is explicit, concrete, and closest to the duplicated GPT-J follow-ups.", + "summary": "This cluster is a broad standardized output-tracing refactor umbrella across many model families. The true duplicates are the same-model GPT-J/GPT-Neo PR pairs; most other links are just shared refactor theme. I revised the PR anchor to #44722, which is the clearest GPT-J/CodeGen representative in the packet.", + "confidence": 0.81, + "canonical_issue_reason": "Issue #43979 is the shared umbrella tracker for the standardized output-tracing refactor across these PRs.", + "canonical_pr_reason": "PR #44722 is the strongest representative PR because it explicitly targets #43979 and covers both GPT-J and CodeGen, matching the central code-path change in this cluster.", + "best_issue_reason": "No standalone issue artifact is present; #43979 is the best issue-level anchor because the true duplicate PRs all point back to it.", + "best_pr_reason": "PR #44722 is the clearest anchor for the shared GPT-J/CodeGen refactor: it is explicitly linked to #43979 and spans the same concrete files as the other accepted GPT-J/CodeGen PRs.", "soft_edge_verdicts": [ { "left": "pull_request:43996", "right": "pull_request:44085", "accept": false, - "reason": "Different models/files; both are output-tracing refactors, but not the same code path." + "reason": "Shared tracker only; CVT/FNet and GPT-J are different concrete code paths." }, { "left": "pull_request:43996", "right": "pull_request:44044", "accept": false, - "reason": "Different model families and different concrete implementations." + "reason": "Different model families; the overlap is only the generic output-tracing theme." }, { "left": "pull_request:44066", "right": "pull_request:44085", "accept": true, - "reason": "Same GPT-J output-tracing code path in the same file; 44066 is a broader version of the same fix." + "reason": "Same GPT-J output-tracing refactor in the same file; mergeable into one PR." }, { "left": "pull_request:44007", "right": "pull_request:44072", "accept": false, - "reason": "Different backbones and different files; only the refactor theme is shared." + "reason": "RegNet/ResNet-style refactor vs EfficientNet; not the same underlying change." }, { "left": "pull_request:44072", "right": "pull_request:44722", "accept": false, - "reason": "EfficientNet versus GPT-J/CodeGen; unrelated concrete changes." + "reason": "EfficientNet and GPT-J/CodeGen are unrelated concrete code paths." }, { "left": "pull_request:44066", "right": "pull_request:44072", "accept": false, - "reason": "Different models and different code paths." + "reason": "GPT-J/CodeGen vs EfficientNet; only the broad refactor pattern overlaps." }, { "left": "pull_request:44013", "right": "pull_request:44044", "accept": false, - "reason": "MobileNetV2 and DeBERTa-v2 are unrelated implementations." + "reason": "MobileNetV2 and DeBERTa are separate model implementations." }, { "left": "pull_request:44066", "right": "pull_request:44086", "accept": false, - "reason": "Different model code paths; same thematic refactor only." + "reason": "GPT-J/CodeGen and MGP-STR are different code paths." }, { "left": "pull_request:44066", "right": "pull_request:44071", "accept": false, - "reason": "GPT-J/CodeGen versus MPT; not the same bug or change." + "reason": "GPT-J/CodeGen and MPT are different model implementations." }, { "left": "pull_request:44018", "right": "pull_request:44068", "accept": true, - "reason": "Both are GPT-Neo output-tracing refactors on the same concrete code path." + "reason": "Both are GPT-Neo output-tracing refactors against the same file set." }, { "left": "pull_request:44066", "right": "pull_request:44068", "accept": false, - "reason": "GPT-J/CodeGen and GPT-Neo are separate implementations." + "reason": "GPT-J/CodeGen vs GPT-Neo; same refactor theme only." }, { "left": "pull_request:44018", "right": "pull_request:44066", "accept": false, - "reason": "GPT-Neo versus GPT-J/CodeGen; same pattern, different code path." + "reason": "GPT-Neo and GPT-J/CodeGen are different model code paths." }, { "left": "pull_request:44019", "right": "pull_request:44722", "accept": false, - "reason": "ResNet and GPT-J/CodeGen are unrelated model implementations." + "reason": "ResNet vs GPT-J/CodeGen; no shared concrete bug or fix." }, { "left": "pull_request:44086", "right": "pull_request:44722", "accept": false, - "reason": "MGP-STR and GPT-J/CodeGen are different concrete changes." + "reason": "MGP-STR and GPT-J/CodeGen are unrelated implementations." }, { "left": "pull_request:44019", "right": "pull_request:44085", "accept": false, - "reason": "ResNet versus GPT-J; only the refactor style matches." + "reason": "ResNet and GPT-J are different model families." }, { "left": "pull_request:44018", "right": "pull_request:44019", "accept": false, - "reason": "GPT-Neo and ResNet are different models with different output paths." + "reason": "GPT-Neo and ResNet are unrelated concrete changes." }, { "left": "pull_request:44019", "right": "pull_request:44071", "accept": false, - "reason": "Different model code paths." + "reason": "ResNet vs MPT; only the output-tracing theme matches." }, { "left": "pull_request:44068", "right": "pull_request:44116", "accept": false, - "reason": "GPT-Neo versus Flaubert; unrelated implementations." + "reason": "GPT-Neo and Flaubert/XLM are unrelated model code paths." }, { "left": "pull_request:44068", "right": "pull_request:44722", "accept": false, - "reason": "GPT-Neo and GPT-J/CodeGen are different concrete refactors." + "reason": "GPT-Neo vs GPT-J/CodeGen; same migration pattern, different code." }, { "left": "pull_request:44018", "right": "pull_request:44116", "accept": false, - "reason": "GPT-Neo and Flaubert are unrelated code paths." + "reason": "GPT-Neo and Flaubert/XLM are different implementations." }, { "left": "pull_request:44071", + "right": "pull_request:44722", + "accept": false, + "reason": "MPT and GPT-J/CodeGen are different model code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo and MGP-STR are separate model implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo and GPT-J are different model families." + }, + { + "left": "pull_request:44068", "right": "pull_request:44086", "accept": false, - "reason": "MPT and MGP-STR are different model implementations." + "reason": "GPT-Neo and MGP-STR are unrelated concrete code paths." }, { "left": "pull_request:44071", - "right": "pull_request:44722", + "right": "pull_request:44085", "accept": false, - "reason": "MPT versus GPT-J/CodeGen; not mergeable as one PR." + "reason": "MPT and GPT-J are different models." }, { "left": "pull_request:44059", "right": "pull_request:44086", "accept": false, - "reason": "GPT2 and MGP-STR are different concrete changes." + "reason": "GPT-2 and MGP-STR are unrelated implementations." }, { "left": "pull_request:44085", "right": "pull_request:44086", "accept": false, - "reason": "GPT-J and MGP-STR are unrelated code paths." + "reason": "GPT-J and MGP-STR are different concrete code paths." }, { "left": "pull_request:44059", "right": "pull_request:44722", "accept": false, - "reason": "GPT2 versus GPT-J/CodeGen; same theme, different implementation." + "reason": "GPT-2 vs GPT-J/CodeGen; only the generic refactor theme overlaps." }, { "left": "pull_request:44098", "right": "pull_request:44101", "accept": false, - "reason": "ViLT and Flaubert/XLM are different model families." + "reason": "ViLT and Flaubert/XLM are unrelated model code paths." }, { "left": "pull_request:44018", "right": "pull_request:44071", "accept": false, - "reason": "GPT-Neo and MPT are unrelated concrete refactors." + "reason": "GPT-Neo and MPT are different model implementations." }, { "left": "pull_request:44017", "right": "pull_request:44161", "accept": false, - "reason": "SegFormer and LongT5 are unrelated model implementations." + "reason": "Segformer and LongT5 are different model families." }, { "left": "pull_request:44017", "right": "pull_request:44722", "accept": false, - "reason": "SegFormer versus GPT-J/CodeGen; not the same bug or change." + "reason": "Segformer and GPT-J/CodeGen are unrelated concrete code paths." }, { "left": "pull_request:44017", "right": "pull_request:44073", "accept": false, - "reason": "SegFormer and VisualBERT are different code paths." + "reason": "Segformer and VisualBert are separate implementations." }, { "left": "pull_request:44068", @@ -17628,226 +17721,196 @@ "left": "pull_request:44071", "right": "pull_request:44086", "accept": false, - "reason": "MPT and MGP-STR are unrelated implementations." + "reason": "MPT and MGP-STR are different code paths." }, { - "left": "pull_request:44059", + "left": "pull_request:43997", + "right": "pull_request:44073", + "accept": false, + "reason": "RegNet and VisualBert are unrelated model implementations." + }, + { + "left": "pull_request:44017", "right": "pull_request:44071", "accept": false, - "reason": "GPT2 and MPT are different code paths." + "reason": "Segformer and MPT are different concrete changes." }, { "left": "pull_request:44073", "right": "pull_request:44101", "accept": false, - "reason": "VisualBERT and Flaubert/XLM are different models." + "reason": "VisualBert and Flaubert/XLM are unrelated code paths." }, { "left": "pull_request:44056", "right": "pull_request:44101", "accept": false, - "reason": "MPNet and Flaubert/XLM are different implementations." + "reason": "MPNet and Flaubert/XLM are different model implementations." }, { "left": "pull_request:44059", "right": "pull_request:44068", "accept": false, - "reason": "GPT2 and GPT-Neo are separate model code paths." + "reason": "GPT-2 and GPT-Neo are different model families." }, { - "left": "pull_request:44018", - "right": "pull_request:44059", + "left": "pull_request:44059", + "right": "pull_request:44071", "accept": false, - "reason": "GPT-Neo versus GPT2; same refactor theme only." + "reason": "GPT-2 and MPT are unrelated concrete code paths." }, { "left": "pull_request:44101", "right": "pull_request:44161", "accept": false, - "reason": "Flaubert/XLM and LongT5 are unrelated changes." + "reason": "Flaubert/XLM and LongT5 are different model families." }, { "left": "pull_request:43997", "right": "pull_request:44017", "accept": false, - "reason": "RegNet and SegFormer are different model families." + "reason": "RegNet and Segformer are different models." }, { "left": "pull_request:43997", "right": "pull_request:44056", "accept": false, - "reason": "RegNet and MPNet are unrelated." + "reason": "RegNet and MPNet are unrelated implementations." }, { "left": "pull_request:44071", "right": "pull_request:44076", "accept": false, - "reason": "MPT and ImageGPT are different concrete implementations." + "reason": "MPT and ImageGPT are different concrete code paths." }, { "left": "pull_request:44074", "right": "pull_request:44101", "accept": false, - "reason": "TextNet and Flaubert/XLM are unrelated model code paths." + "reason": "TextNet vs Flaubert/XLM; only the broad refactor theme matches." }, { "left": "pull_request:44056", "right": "pull_request:44161", "accept": false, - "reason": "MPNet and LongT5 are different implementations." + "reason": "MPNet and LongT5 are different model implementations." }, { "left": "pull_request:43997", "right": "pull_request:44018", "accept": false, - "reason": "RegNet and GPT-Neo are unrelated model code paths." + "reason": "RegNet and GPT-Neo are unrelated concrete changes." }, { "left": "pull_request:43997", "right": "pull_request:44161", "accept": false, - "reason": "RegNet and LongT5 are different models." + "reason": "RegNet and LongT5 are different code paths." }, { "left": "pull_request:44074", "right": "pull_request:44161", "accept": false, - "reason": "TextNet and LongT5 are unrelated implementations." + "reason": "TextNet and LongT5 are different model families." }, { "left": "pull_request:44018", - "right": "pull_request:44161", + "right": "pull_request:44059", "accept": false, - "reason": "GPT-Neo and LongT5 are different model code paths." + "reason": "GPT-Neo and GPT-2 are different model implementations." }, { "left": "pull_request:43997", "right": "pull_request:44071", "accept": false, - "reason": "RegNet and MPT are unrelated." - }, - { - "left": "pull_request:44010", - "right": "pull_request:44071", - "accept": false, - "reason": "SqueezeBERT and MPT are different models." - }, - { - "left": "pull_request:44010", - "right": "pull_request:44073", - "accept": false, - "reason": "SqueezeBERT and VisualBERT are different concrete changes." - }, - { - "left": "pull_request:44018", - "right": "pull_request:44129", - "accept": false, - "reason": "GPT-Neo and SpeechT5 are unrelated model implementations." - }, - { - "left": "pull_request:44074", - "right": "pull_request:44161", - "accept": false, - "reason": "TextNet and LongT5 are different code paths." + "reason": "RegNet and MPT are unrelated model code paths." }, { "left": "pull_request:43997", "right": "pull_request:44074", "accept": false, - "reason": "RegNet and TextNet are unrelated models." + "reason": "RegNet and TextNet are different models." }, { - "left": "pull_request:43997", - "right": "pull_request:44129", + "left": "pull_request:44010", + "right": "pull_request:44071", "accept": false, - "reason": "RegNet and SpeechT5 are different implementations." + "reason": "SqueezeBert and MPT are different code paths." }, { - "left": "pull_request:44071", - "right": "pull_request:44129", + "left": "pull_request:44010", + "right": "pull_request:44073", "accept": false, - "reason": "MPT and SpeechT5 are unrelated." + "reason": "SqueezeBert and VisualBert are unrelated implementations." }, { "left": "pull_request:44010", "right": "pull_request:44018", "accept": false, - "reason": "SqueezeBERT and GPT-Neo are different model code paths." - }, - { - "left": "pull_request:43996", - "right": "pull_request:44161", - "accept": false, - "reason": "FNet/CVT and LongT5 are unrelated." - }, - { - "left": "pull_request:43996", - "right": "pull_request:44017", - "accept": false, - "reason": "FNet/CVT and SegFormer are different model families." + "reason": "SqueezeBert vs GPT-Neo; same refactor theme only." }, { "left": "pull_request:44010", "right": "pull_request:44044", "accept": false, - "reason": "SqueezeBERT and DeBERTa-v2 are unrelated." + "reason": "SqueezeBert and DeBERTa are different model code paths." }, { "left": "pull_request:44044", "right": "pull_request:44074", "accept": false, - "reason": "DeBERTa-v2 and TextNet are different concrete changes." + "reason": "DeBERTa and TextNet are unrelated models." }, { "left": "pull_request:44066", "right": "pull_request:44722", "accept": true, - "reason": "Same GPT-J/CodeGen output-tracing refactor; 44722 is a near-duplicate follow-up." + "reason": "Same GPT-J/CodeGen output-tracing refactor; these are effectively duplicate PRs." }, { "left": "pull_request:44085", "right": "pull_request:44722", "accept": true, - "reason": "Both target the same GPT-J output-tracing code path and could be merged into one PR." + "reason": "Both touch the same GPT-J output-tracing code path; 44722 just includes an extra CodeGen slice." }, { "left": "pull_request:44085", "right": "pull_request:44129", "accept": false, - "reason": "GPT-J and SpeechT5 are different implementations." + "reason": "GPT-J and SpeechT5 are different model families." }, { "left": "pull_request:43996", "right": "pull_request:44007", "accept": false, - "reason": "FNet/CVT and RegNet/ResNet/RT-DETR are unrelated." + "reason": "CVT/FNet and RegNet/ResNet/RT-DETR are different implementations." }, { "left": "pull_request:43996", "right": "pull_request:44129", "accept": false, - "reason": "FNet/CVT and SpeechT5 are different model code paths." + "reason": "Vision backbone refactor vs SpeechT5; not the same change." }, { "left": "pull_request:44044", "right": "pull_request:44129", "accept": false, - "reason": "DeBERTa-v2 and SpeechT5 are unrelated implementations." + "reason": "DeBERTa and SpeechT5 are unrelated concrete code paths." }, { "left": "pull_request:44044", "right": "pull_request:44154", "accept": false, - "reason": "DeBERTa-v2 and ViTS are different concrete changes." + "reason": "DeBERTa and Vits are different models; same theme only." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The cluster summary is grounded in the packet: #43979 is the shared issue target across the refactor stream, and the representative PR choice is supported by the GPT-J/CodeGen overlap. The soft-edge verdicts are conservative overall, and the true-accept pairs are limited to same-model/code-path cases that do look mergeable into a single bugfix PR." + "accept": false, + "feedback": "The cluster summary is mostly conservative, but the PR anchoring overreaches. In particular, the claim that PR #44066 is broader/more central than later GPT-J-only follow-ups is not supported by the packet: PR #44722 also touches both `codegen` and `gptj`, is explicitly linked to #43979, and looks at least as central. That makes the canonical PR reasoning insufficiently grounded. The true duplicate/mergeable pairs you marked are plausible, but the canonical PR justification should be revised to avoid overstating evidence." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -17882,70 +17945,70 @@ "pull_request:43997|pull_request:44010", "pull_request:44010|pull_request:44129" ], - "prepared_review_unit_hash": "445c8ef2736f689f9e0062f64ab50a2fc6932c0e9fe647dae3963e1c0e91db7f", + "prepared_review_unit_hash": "4b81b849860b1ba480643613dfc0b5209e578a2815cd6d3b2c4ee6c8d995e3b5", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 78, - "serialized_chars": 84904, - "estimated_input_tokens": 21226, - "estimated_eval_tokens": 42708 + "serialized_chars": 84896, + "estimated_input_tokens": 21224, + "estimated_eval_tokens": 42704 }, "budget": { "node_count": 8, "item_count": 8, "soft_pair_count": 6, - "serialized_chars": 17126, - "estimated_input_tokens": 4282, - "estimated_eval_tokens": 8820 + "serialized_chars": 17124, + "estimated_input_tokens": 4281, + "estimated_eval_tokens": 8818 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These PRs are all separate per-model migrations to standardized output-capturing/tracing patterns. They share a common refactor theme, but they do not appear to be the same concrete fix or mergeable into one PR.", - "confidence": 0.94, + "summary": "These PRs are all model-specific output-tracing refactors, but they touch different code paths and should not be deduplicated across models. They look like a family of similar housekeeping changes rather than one shared bug fix.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "pull_request:44129 is the strongest representative: it has an explicit issue target, a concrete model-specific refactor, and the most clearly scoped change among the set.", + "canonical_pr_reason": "PR 44056 is the clearest representative of the cluster: it has a focused, well-described capture_outputs refactor and adds the model-specific _can_record_outputs wiring in one place.", "best_issue_reason": null, - "best_pr_reason": "pull_request:44129 is the best overall PR because it is well-scoped, explicitly linked to an issue, and clearly documents a concrete output-tracing migration for one code path.", + "best_pr_reason": "PR 44056 is the best overall representative because its title and diff most directly match the common standardized-output-tracing theme without being overly narrow or tangled with unrelated changes.", "soft_edge_verdicts": [ { "left": "pull_request:44024", "right": "pull_request:44076", "accept": false, - "reason": "Both are standardized output-tracing refactors, but for different models (FocalNet vs ImageGPT) and different code paths; not the same bug/change." + "reason": "Same broad refactor theme, but different models and different concrete code paths (FocalNet vs ImageGPT). Not the same underlying change." }, { "left": "pull_request:44073", "right": "pull_request:44074", "accept": false, - "reason": "VisualBert and TextNet are unrelated model implementations; these are parallel refactors, not one shared fix." + "reason": "Both modernize output handling, but VisualBert and TextNet are separate implementations with different behavior changes. Too broad to treat as duplicates." }, { "left": "pull_request:44056", "right": "pull_request:44073", "accept": false, - "reason": "MPNet and VisualBert have different forward paths and output-recording setup; same refactor theme only, not duplicate work." + "reason": "Both are standardized output tracing refactors, but they target different model internals (MPNet vs VisualBert) and are not mergeable as one PR." }, { "left": "pull_request:44056", "right": "pull_request:44074", "accept": false, - "reason": "MPNet and TextNet are separate models with distinct output-capture changes; not the same underlying code-path problem." + "reason": "Different models and different output-shape/API adjustments; similar pattern, but not the same concrete fix." }, { "left": "pull_request:43997", "right": "pull_request:44010", "accept": false, - "reason": "RegNet and SqueezeBert are unrelated model modules; these PRs do not fix the same concrete issue." + "reason": "RegNet and SqueezeBert are unrelated model code paths. Shared decorator imports are not enough to make these duplicates." }, { "left": "pull_request:44010", "right": "pull_request:44129", "accept": false, - "reason": "SqueezeBert and SpeechT5 both touch output-tracing, but they affect different model internals and are not plausibly one merged PR." + "reason": "SqueezeBert vs SpeechT5 are different model implementations, and only SpeechT5 is explicitly tied to issue 43979. Not a duplicate pair." } ] }, @@ -18025,12 +18088,12 @@ "pull_request:44028|pull_request:44030", "pull_request:44029|pull_request:44030" ], - "prepared_review_unit_hash": "28c4cef5238fb6480721688abb6d4913ff90fd6e0deaaabfea8559a681ab8e12", + "prepared_review_unit_hash": "d8301b178ca7552b2203f761c9b5dfdd10fe089db1d51d63d9b9a4d9820d61f6", "component_budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 40, - "serialized_chars": 40179, + "serialized_chars": 40177, "estimated_input_tokens": 10045, "estimated_eval_tokens": 20346 }, @@ -18038,7 +18101,7 @@ "node_count": 13, "item_count": 13, "soft_pair_count": 40, - "serialized_chars": 40179, + "serialized_chars": 40177, "estimated_input_tokens": 10045, "estimated_eval_tokens": 20346 }, @@ -18047,252 +18110,246 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs share a naming pattern and a broad \u201coutput tracing\u201d theme, but they are model-specific refactors in different code paths. I do not see true duplicate PRs here.", - "confidence": 0.95, + "summary": "These are model-specific output-tracing refactors across unrelated architectures, not duplicate fixes for one shared bug or change.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "No single true duplicate PR stands out; 44025 is the best representative anchor because it\u2019s a clear output-tracing refactor across two related model files, but it is still not the same change as the others.", + "canonical_pr_reason": "44003 is the best single representative PR: it has the broadest scope in this set, touching two closely related Mamba-family files with a substantial tracing refactor.", "best_issue_reason": null, - "best_pr_reason": "44025 is the strongest overall representative of the cluster\u2019s theme, with a broader two-file refactor footprint and central similarity to several others, even though it should not be treated as a duplicate of them.", + "best_pr_reason": "44003 is the strongest representative of the pattern because it is broader than the other PRs and contains the clearest, most substantial code-path change.", "soft_edge_verdicts": [ { "left": "pull_request:43998", "right": "pull_request:44002", "accept": false, - "reason": "Different models and code paths; same refactor theme only, not the same bug/change." + "reason": "Same refactor theme, but different model implementations and code paths; not the same concrete bug/change." }, { "left": "pull_request:43998", "right": "pull_request:44000", "accept": false, - "reason": "Separate model implementations; similarity is stylistic, not a shared concrete fix." + "reason": "Different model families and files; only the refactor style matches, not the underlying change." }, { "left": "pull_request:43998", "right": "pull_request:44001", "accept": false, - "reason": "Different architecture and file; not the same underlying change." + "reason": "Different architectures and forward paths; these are separate model-local refactors." }, { "left": "pull_request:43998", "right": "pull_request:44003", "accept": false, - "reason": "Unrelated model code paths; only the refactor pattern matches." + "reason": "Timm backbone vs Mamba/FalconMamba are unrelated code paths; not one mergeable fix." }, { "left": "pull_request:43998", "right": "pull_request:44029", "accept": false, - "reason": "Different model family; no evidence of one merged fix." + "reason": "Different architectures and output-tracing implementations; same theme, not the same change." }, { "left": "pull_request:43999", "right": "pull_request:44001", "accept": false, - "reason": "Different models; shared output-tracing cleanup is too broad to count as duplicate." + "reason": "MobileNetV1 and UnivNet are unrelated models; these are separate refactors, not duplicates." }, { "left": "pull_request:43999", "right": "pull_request:44002", "accept": false, - "reason": "Separate model code paths with no concrete overlap beyond refactor style." + "reason": "Different model code paths; the shared wording does not indicate one underlying bug." }, { "left": "pull_request:43999", "right": "pull_request:44025", "accept": false, - "reason": "Different model families; not the same code-path problem." + "reason": "MobileNetV1 vs Depth Anything are unrelated implementations; not a single concrete fix." }, { "left": "pull_request:43999", "right": "pull_request:44026", "accept": false, - "reason": "Distinct encoder-decoder vs backbone model changes; not duplicate work." + "reason": "Different model families and files; only a similar tracing cleanup pattern." }, { "left": "pull_request:43999", "right": "pull_request:44027", "accept": false, - "reason": "Same generic tracing refactor, but different subsystem and behavior." + "reason": "MobileNetV1 and Speech-Encoder-Decoder are unrelated code paths, so this is not a duplicate." }, { "left": "pull_request:44000", "right": "pull_request:44001", "accept": false, - "reason": "Different models; too generic to be a duplicate." + "reason": "Vision-Text Dual Encoder and UnivNet are separate model implementations; same theme, different changes." }, { "left": "pull_request:44000", "right": "pull_request:44002", "accept": false, - "reason": "Separate code paths and model classes; only the theme overlaps." + "reason": "Different model families and forward logic; not the same bug or PR-worthy change." }, { "left": "pull_request:44000", "right": "pull_request:44003", "accept": false, - "reason": "No shared concrete bug or implementation target." + "reason": "Vision-Text Dual Encoder vs Mamba/FalconMamba are unrelated code paths." }, { "left": "pull_request:44000", "right": "pull_request:44025", "accept": false, - "reason": "Different model families; same refactor motif is not enough." + "reason": "These touch unrelated models; the common output-tracing wording is too broad to merge." }, { "left": "pull_request:44001", "right": "pull_request:44002", "accept": false, - "reason": "Different architectures; not plausibly one PR." + "reason": "UnivNet and UperNet are unrelated model paths; no shared concrete fix." }, { "left": "pull_request:44001", "right": "pull_request:44003", "accept": false, - "reason": "No shared failing code-path, just similar cleanup." + "reason": "Different architectures and files; not one underlying code-path problem." }, { "left": "pull_request:44001", "right": "pull_request:44025", "accept": false, - "reason": "Distinct model-specific output handling; not the same change." + "reason": "UnivNet vs Depth Anything are separate model-local refactors, not duplicates." }, { "left": "pull_request:44002", "right": "pull_request:44029", "accept": false, - "reason": "Different model implementations and output semantics; not duplicate." + "reason": "UperNet and RWKV are unrelated code paths; only a generic refactor theme is shared." }, { "left": "pull_request:44002", "right": "pull_request:44003", "accept": false, - "reason": "Same refactor category, but different concrete fixes." + "reason": "UperNet and Mamba/FalconMamba are different models with different tracing logic." }, { "left": "pull_request:44003", "right": "pull_request:44028", "accept": false, - "reason": "Mamba vs SuperPoint are unrelated code paths; no duplicate evidence." + "reason": "Mamba/FalconMamba and SuperPoint are unrelated architectures; not the same fix." }, { "left": "pull_request:44003", "right": "pull_request:44004", "accept": false, - "reason": "Different model internals and output shapes; not the same fix." + "reason": "Mamba/FalconMamba vs CodeGen are different code paths; same pattern, not same change." }, { "left": "pull_request:43998", "right": "pull_request:44004", "accept": false, - "reason": "Backbone output tracing vs CodeGen cache/output tracing are different changes." + "reason": "Timm backbone and CodeGen are unrelated model implementations." }, { "left": "pull_request:44001", "right": "pull_request:44004", "accept": false, - "reason": "UnivNet and CodeGen are unrelated model code paths." + "reason": "UnivNet and CodeGen are unrelated code paths; not a single bug fix." }, { "left": "pull_request:44002", "right": "pull_request:44004", "accept": false, - "reason": "UperNet segmentation and CodeGen generation logic are not the same bug." - }, - { - "left": "pull_request:44000", - "right": "pull_request:44004", - "accept": false, - "reason": "Different models and output contracts; only broad refactor similarity." + "reason": "UperNet and CodeGen are separate model-specific refactors." }, { "left": "pull_request:44025", "right": "pull_request:44026", "accept": false, - "reason": "Depth Anything vs Vision Encoder-Decoder are separate model changes." + "reason": "Depth Anything and Vision-Encoder-Decoder are different model families; not one mergeable change." }, { "left": "pull_request:44025", "right": "pull_request:44027", "accept": false, - "reason": "Different encoder-decoder stacks; not one concrete fix." + "reason": "Depth Anything vs Speech-Encoder-Decoder are unrelated implementations." }, { "left": "pull_request:44025", "right": "pull_request:44028", "accept": false, - "reason": "Depth estimation vs keypoint model; unrelated code paths." + "reason": "Different model code paths; the shared tracing refactor theme is too broad." }, { "left": "pull_request:44025", "right": "pull_request:44029", "accept": false, - "reason": "Different model families; same tracing refactor pattern only." + "reason": "Depth Anything and RWKV are unrelated models and would not be merged as one PR." }, { "left": "pull_request:44025", "right": "pull_request:44030", "accept": false, - "reason": "Depth Anything and DPR have distinct implementations and behavior." + "reason": "Different architectures and forward logic; not the same underlying change." }, { "left": "pull_request:44026", "right": "pull_request:44029", "accept": false, - "reason": "Vision encoder-decoder and RWKV are unrelated model code paths." + "reason": "Vision-Encoder-Decoder and RWKV are unrelated code paths." }, { "left": "pull_request:44026", "right": "pull_request:44027", "accept": false, - "reason": "Different encoder-decoder models; too broad to merge as duplicates." + "reason": "Vision-Encoder-Decoder and Speech-Encoder-Decoder are similar wrappers, but still separate model-specific changes and not one duplicate fix." }, { "left": "pull_request:44026", "right": "pull_request:44028", "accept": false, - "reason": "Distinct model architectures and outputs; not duplicate work." + "reason": "Different model families and files; not the same concrete bug/change." }, { "left": "pull_request:44026", "right": "pull_request:44030", "accept": false, - "reason": "Different model classes and code paths; only the refactor theme overlaps." + "reason": "Vision-Encoder-Decoder and DPR are unrelated implementations." }, { "left": "pull_request:44027", "right": "pull_request:44028", "accept": false, - "reason": "Speech encoder-decoder vs SuperPoint are not the same underlying change." + "reason": "Speech-Encoder-Decoder and SuperPoint are unrelated code paths." }, { "left": "pull_request:44027", "right": "pull_request:44029", "accept": false, - "reason": "Unrelated model internals; shared tracing refactor is insufficient." + "reason": "Speech-Encoder-Decoder and RWKV are unrelated models; no shared underlying fix." }, { "left": "pull_request:44027", "right": "pull_request:44030", "accept": false, - "reason": "Different model families and behavior; not a duplicate." + "reason": "Speech-Encoder-Decoder and DPR are separate model-local refactors." }, { "left": "pull_request:44028", "right": "pull_request:44029", "accept": false, - "reason": "SuperPoint and RWKV are separate subsystems with no shared concrete fix." + "reason": "SuperPoint and RWKV are unrelated architectures; same theme only." }, { "left": "pull_request:44028", "right": "pull_request:44030", "accept": false, - "reason": "Different model implementations; no same code-path issue." + "reason": "SuperPoint and DPR are different models with different code paths." }, { "left": "pull_request:44029", "right": "pull_request:44030", "accept": false, - "reason": "RWKV and DPR touch different architectures and output handling." + "reason": "RWKV and DPR are unrelated implementations; not a duplicate PR pair." } ] }, @@ -18325,52 +18382,52 @@ "pull_request:44662|pull_request:45075", "pull_request:45075|pull_request:45401" ], - "prepared_review_unit_hash": "1582ed2fc980c4e74af4b0db6df84c61d389c68623c8905a582892cfb7c903d4", + "prepared_review_unit_hash": "935bde6d7171664e4aaae23d32e57fe7f588b447534b3c7dc4fc45677024f4e1", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 10912, - "estimated_input_tokens": 2728, - "estimated_eval_tokens": 5712 + "serialized_chars": 10914, + "estimated_input_tokens": 2729, + "estimated_eval_tokens": 5714 }, "budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 10912, - "estimated_input_tokens": 2728, - "estimated_eval_tokens": 5712 + "serialized_chars": 10914, + "estimated_input_tokens": 2729, + "estimated_eval_tokens": 5714 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three separate model-addition PRs (PenguinVL, DeepSeek-OCR-2, and Voxtral TTS) that share the same Transformers boilerplate but do not fix or implement the same underlying change.", - "confidence": 0.96, + "summary": "These are three separate model-addition PRs for different model families. They overlap on registry/docs plumbing, but each introduces a distinct implementation and test suite, so they are not duplicates.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: each PR introduces a different model integration with distinct configs, processors, model code, and tests.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "None stands out as a duplicate representative; the closest match is not a real duplicate because the PRs target different models and code paths.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:44662", "right": "pull_request:45401", "accept": false, - "reason": "Different models and different implementations (PenguinVL vs Voxtral TTS); shared auto/docs files are just standard model-registration boilerplate, not the same fix." + "reason": "Different models and code paths: PenguinVL vs Voxtral TTS. Shared auto-registry/docs files are just integration plumbing, not the same underlying change." }, { "left": "pull_request:44662", "right": "pull_request:45075", "accept": false, - "reason": "Both add new model support, but for different models (PenguinVL vs DeepSeek-OCR-2) with separate architecture, conversion, and test code." + "reason": "Different model implementations: PenguinVL vs DeepSeek-OCR-2. Similar framework registration changes do not make these the same PR." }, { "left": "pull_request:45075", "right": "pull_request:45401", "accept": false, - "reason": "Distinct model additions with different target problems and unique source trees; not mergeable as one PR for the same underlying change." + "reason": "Distinct model families and feature sets: DeepSeek-OCR-2 vs Voxtral TTS. Overlap is limited to common model-registration files." } ] }, @@ -18400,7 +18457,7 @@ "soft_pairs": [ "pull_request:44979|pull_request:45363" ], - "prepared_review_unit_hash": "db0324771111ae121e3998f1e5cbb1f1eac6293592181e1ad89e5f4517728939", + "prepared_review_unit_hash": "13f14121ee4f5bc5a4a8abeac7d87ffb77157e3133b9410372981e2e1ad30441", "component_budget": { "node_count": 2, "item_count": 2, @@ -18422,18 +18479,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs are about fusion-related APIs in Transformers, but they target different layers: #44979 adds a generic module-fusion API, while #45363 implements n-to-1 kernel fusion via `KernelConfig` and related integration changes. They are related, but not the same underlying change.", - "confidence": 0.89, + "summary": "Two related but distinct open PRs around fusion infrastructure: one introduces a generic Module Fusion API, while the other adds KernelConfig-based n-to-1 kernel fusion integration and an example. They share a theme but not the same concrete change.", + "confidence": 0.88, "canonical_issue_reason": null, - "canonical_pr_reason": "#45363 is the more complete, user-facing fusion feature: it wires `KernelConfig` through model loading/integrations and includes an end-to-end example. #44979 is a separate foundational module-fusion API addition, not the same concrete change.", + "canonical_pr_reason": "PR #44979 is the cleaner representative of the shared theme because it introduces the core Module Fusion API abstraction that the other PR appears to build upon.", "best_issue_reason": null, - "best_pr_reason": "#45363 is the better representative of the cluster because it contains the broader concrete implementation path for kernel fusion, touching core loading/integration code rather than only introducing a standalone API surface.", + "best_pr_reason": "PR #44979 is the best cluster representative overall since it defines the foundational API and is the broadest, most reusable change in the set.", "soft_edge_verdicts": [ { "left": "pull_request:44979", "right": "pull_request:45363", "accept": false, - "reason": "Related fusion work, but not the same bug or change. #44979 adds a generic module-fusion API in a new file, while #45363 implements kernel fusion through `KernelConfig` and integration changes. They would not plausibly be merged as one PR without collapsing distinct features." + "reason": "Related fusion work, but not the same underlying bug or mergeable change: #44979 adds a generic Module Fusion API, while #45363 wires a specific KernelConfig-based n-to-1 fusion path plus example/integration changes. The scopes and code paths differ too much to treat as duplicates." } ] }, @@ -18463,12 +18520,12 @@ "soft_pairs": [ "pull_request:45086|pull_request:45317" ], - "prepared_review_unit_hash": "f75659896d3f7df0091b894c3bd8554367c49f17aa6da9486b53e9524e366814", + "prepared_review_unit_hash": "cf5211ea4d561ac766becc199b0640b55ad7bf2b8f3c1da1eb2a5958e6066ed7", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4414, + "serialized_chars": 4413, "estimated_input_tokens": 1104, "estimated_eval_tokens": 2464 }, @@ -18476,7 +18533,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4414, + "serialized_chars": 4413, "estimated_input_tokens": 1104, "estimated_eval_tokens": 2464 }, @@ -18485,18 +18542,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both pull requests fix the same AttributeError in `_patch_mistral_regex` by switching from `tokenizer.backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer` in `tokenization_utils_tokenizers.py`; 45317 is the more complete version because it also adds tests.", + "summary": "Both pull requests fix the same concrete Mistral regex patch bug in `tokenization_utils_tokenizers.py`: they replace `backend_tokenizer.pre_tokenizer` with `tokenizer.pre_tokenizer` to avoid an `AttributeError`. PR 45317 is the more complete version because it includes tests.", "confidence": 0.99, "canonical_issue_reason": null, - "canonical_pr_reason": "45317 is the better canonical PR because it implements the same bug fix as 45086 and includes accompanying test coverage, making it the more complete representative of the change.", + "canonical_pr_reason": "PR 45317 is the best canonical PR because it addresses the same underlying bug as 45086 but adds test coverage and appears to be the more complete fix.", "best_issue_reason": null, - "best_pr_reason": "45317 is the best PR candidate because it addresses the same concrete Mistral pre-tokenizer AttributeError and adds validation tests, so it is the strongest standalone fix in the cluster.", + "best_pr_reason": "PR 45317 is the strongest standalone candidate: same fix, same target issue, plus tests validating `fix_mistral_regex=True`.", "soft_edge_verdicts": [ { "left": "pull_request:45086", "right": "pull_request:45317", "accept": true, - "reason": "Same underlying code-path bug in `_patch_mistral_regex`, same file, same explicit issue target, and the diffs overlap on the same fix; 45317 just adds tests." + "reason": "Same code-path bug fix in `_patch_mistral_regex` with identical core change; 45317 is an expanded version of 45086 rather than a different PR." } ] }, @@ -18526,7 +18583,7 @@ "soft_pairs": [ "pull_request:45115|pull_request:45569" ], - "prepared_review_unit_hash": "e6750721eb8783fdf126499ec017eeb37c80aed37bd362e5012beeec4196f714", + "prepared_review_unit_hash": "8175693e52fb9b8141930d3c38cbd4f14ba637a712bead4162024409aaffdd6b", "component_budget": { "node_count": 2, "item_count": 2, @@ -18548,18 +18605,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two PRs overlap on NemotronH codepaths, but they are not the same change: #45115 is a narrower inheritance/conversion-mapping refactor toward GraniteMoeHybrid compatibility, while #45569 is a broader model-family split introducing NemotronHDense/NemotronHSparse plus docs and auto-registration updates.", - "confidence": 0.92, + "summary": "Two open PRs overlap on NemotronH internals, but they address different changes: one is a conversion/mapping and inheritance refactor, the other is a broader model split into NemotronHDense/NemotronHSparse with docs/tests. They are related but not duplicates.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "#45569 is the more complete and user-facing PR: it defines the new NemotronH dense/sparse variants, updates docs, and wires the model into auto/config/model registration. It better represents the main architectural change in this cluster.", + "canonical_pr_reason": "PR #45569 is the broader, more representative NemotronH change: it introduces the dense/sparse model split, updates docs, auto-mappings, and tests, so it better captures the main architectural direction.", "best_issue_reason": null, - "best_pr_reason": "#45569 is the best PR because it covers the full NemotronH refactor/splitting effort rather than only a conversion-mapping/inheritance adjustment.", + "best_pr_reason": "PR #45569 is the best global representative because it encompasses the larger user-visible model refactor and package structure changes, whereas #45115 is a narrower conversion-mapping/inheritance cleanup.", "soft_edge_verdicts": [ { "left": "pull_request:45115", "right": "pull_request:45569", "accept": false, - "reason": "They share some NemotronH files, but the underlying changes differ: #45115 focuses on checkpoint conversion naming and GraniteMoeHybrid-style inheritance, whereas #45569 introduces separate dense/sparse model variants and documentation/auto-map updates. They are related, not duplicates." + "reason": "They both touch NemotronH, but the concrete changes differ: #45115 focuses on checkpoint conversion and inheritance/weight-name mapping, while #45569 splits the architecture into dense/sparse variants and updates docs/tests. Not the same underlying bug or mergeable change." } ] }, @@ -18589,7 +18646,7 @@ "soft_pairs": [ "pull_request:45332|pull_request:45333" ], - "prepared_review_unit_hash": "c252bdbcbd96340711e2188c1c881d12f27b95eb3952546a91bf31e460b86297", + "prepared_review_unit_hash": "3ab322e3f27d0e4187329735f60ed3bda4f58c3326b97e870b75603d30039e2f", "component_budget": { "node_count": 2, "item_count": 2, @@ -18611,18 +18668,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open PRs propose related heterogeneous configuration/modeling work, but they are not duplicates: one adds per-layer config support, the other adds broader heterogeneous model support that builds on it. The larger PR is the better cluster representative.", - "confidence": 0.89, + "summary": "These two PRs are related but not duplicates: one adds heterogeneous per-layer configuration support, while the other adds broader heterogeneous model/runtime support built on top of that groundwork.", + "confidence": 0.9, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45332 is the broader, more central change: it covers per-layer config plus the modeling/cache/masking updates that use it. PR #45333 is narrower and largely foundational/config-only.", + "canonical_pr_reason": "PR #45332 is the broader umbrella change, spanning config, cache, masking, modeling, and tests, so it is the best representative of the cluster.", "best_issue_reason": null, - "best_pr_reason": "PR #45332 is the best representative of the cluster because it subsumes the shared heterogeneity groundwork and extends into the main runtime/modeling path, making it the more complete canonical PR.", + "best_pr_reason": "PR #45332 is the most suitable global representative because it covers the end-to-end heterogeneous model support rather than only the config layer.", "soft_edge_verdicts": [ { "left": "pull_request:45332", "right": "pull_request:45333", "accept": false, - "reason": "Related feature work, but not the same concrete change or bug fix. #45333 focuses on per-layer configuration support, while #45332 adds broader heterogeneous model/runtime support that consumes that config. They are complementary rather than duplicate PRs." + "reason": "They share implementation infrastructure, but #45333 is config-only while #45332 adds model/runtime behavior; they are related pieces of one feature, not the same concrete change or bug fix." } ] }, @@ -18652,7 +18709,7 @@ "soft_pairs": [ "pull_request:45415|pull_request:45425" ], - "prepared_review_unit_hash": "16137db7fa18870577d38ab41de5da181f80b0decfeaa00210acd97424fae61b", + "prepared_review_unit_hash": "0746c8e17bfcbb5c5415d646323bdb671022af20e07d24dbacddad2087999fa9", "component_budget": { "node_count": 2, "item_count": 2, @@ -18674,18 +18731,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two typing-focused PRs overlap in a few files, but they address different scopes: one is a broad type-checking sweep across many transformers modules, and the other adds specific typing protocols/helpers for modeling and PEFT-related code paths.", - "confidence": 0.82, + "summary": "Both PRs are typing-related, but they are not the same change: #45415 is a broad repository-wide type-checking sweep, while #45425 adds specific typing aliases/protocols to support modeling_utils and PEFT-related code paths.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45425 is the more focused and concrete change: it adds shared typing aliases/protocols and updates a small set of code paths to use them. PR #45415 is a broader repository-wide typing cleanup/configuration pass, so it is less representative as the cluster anchor.", + "canonical_pr_reason": "#45425 is the more focused, self-contained typing change; #45415 is a broad sweep across many unrelated files plus type-checker config, so it is not a duplicate of the targeted typing-support PR.", "best_issue_reason": null, - "best_pr_reason": "PR #45425 is the best representative for this cluster because it centers on a specific typing abstraction change rather than a wide, multi-file type-checking sweep. PR #45415 looks like a larger housekeeping effort with many unrelated edits, making it a weaker canonical representative.", + "best_pr_reason": "#45425 is the best representative of this cluster because it has one concrete typing support theme and a clear scope; #45415 is too heterogeneous to serve as the canonical change.", "soft_edge_verdicts": [ { "left": "pull_request:45415", "right": "pull_request:45425", "accept": false, - "reason": "Do not merge: the overlap is only superficial typing-related cleanup. #45415 is a broad type-checking pass across many files and pyproject settings, while #45425 introduces specific typing protocols and helpers for modeling/PEFT code paths. They do not appear to fix the same concrete bug or change." + "reason": "Reject: shared typing files and helpers do not mean the same underlying bug/change. #45415 is a wide type-checking pass across many modules and pyproject settings, while #45425 adds specific typing aliases/protocols for modeling_utils/PEFT and related utilities. They could be merged independently." } ] }, @@ -18720,12 +18777,12 @@ "pull_request:45817|pull_request:45843", "pull_request:45836|pull_request:45843" ], - "prepared_review_unit_hash": "6a96123a7dd219e95f907ba16cbb534a26673e315d9e8a31962abce02616020b", + "prepared_review_unit_hash": "3bf54099e12545853acaaf6c3ec07013099b340b5d01086e471b3110266559d7", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 4, - "serialized_chars": 10388, + "serialized_chars": 10387, "estimated_input_tokens": 2597, "estimated_eval_tokens": 5450 }, @@ -18733,7 +18790,7 @@ "node_count": 4, "item_count": 4, "soft_pair_count": 4, - "serialized_chars": 10388, + "serialized_chars": 10387, "estimated_input_tokens": 2597, "estimated_eval_tokens": 5450 }, @@ -18742,36 +18799,36 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated model-parallel/device-placement fixes across different model families, plus one unrelated test-skip PR. They are thematically similar but not the same underlying change.", - "confidence": 0.91, + "summary": "These are related by theme (model-parallel robustness) but they are not duplicate PRs: each targets a different model or a test-only workaround. None should be merged together.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "pull_request:45487 is the broadest and clearest representative of the shared model-parallel/device-placement theme, with concrete code changes rather than only test adjustments.", + "canonical_pr_reason": "PR 45843 is the best representative of the cluster because it contains a concrete model-parallel code fix in the actual modeling path, whereas 45836 is test-only and 45487/45817 are separate model-specific fixes.", "best_issue_reason": null, - "best_pr_reason": "pull_request:45487 best captures the cluster\u2019s main pattern of fixing model-parallel placement issues in model code; the others are either narrower, model-specific variants or unrelated test skips.", + "best_pr_reason": "PR 45843 is the strongest standalone PR to keep as the representative fix: it addresses a real device-placement / no_split_modules propagation bug in Deimv2, with focused code changes and tests.", "soft_edge_verdicts": [ { "left": "pull_request:45487", "right": "pull_request:45843", "accept": false, - "reason": "Both address model-parallel placement, but they fix different model families and different code paths; not the same concrete bug or mergeable as one PR." + "reason": "Same broad model-parallel theme, but different models and different failure paths; not the same underlying bug or change." }, { "left": "pull_request:45487", "right": "pull_request:45836", "accept": false, - "reason": "45487 fixes runtime device-placement/model-parallel behavior, while 45836 only skips failing tests for a cache incompatibility; unrelated underlying change." + "reason": "45487 fixes runtime model-parallel device placement; 45836 only skips failing tests for QuantizedCache incompatibility." }, { "left": "pull_request:45817", "right": "pull_request:45843", "accept": false, - "reason": "Similar high-level theme, but Gemma4 and Deimv2 have distinct implementations and fixes; not the same underlying bug." + "reason": "Both are model-parallel fixes, but Gemma4 and Deimv2 have distinct code paths and patch mechanics; not mergeable as one PR." }, { "left": "pull_request:45836", "right": "pull_request:45843", "accept": false, - "reason": "Test-only skip for MiniCPM/Qwen cache incompatibility versus a Deimv2 model-parallel code fix; clearly different issues." + "reason": "Test skip for MiniCPM-V/Qwen cache incompatibility is unrelated to Deimv2's model-parallel code fix." } ] }, @@ -18801,7 +18858,7 @@ "soft_pairs": [ "pull_request:45548|pull_request:45662" ], - "prepared_review_unit_hash": "70c5056f3f907c8ae594b89749e73943c9328affc7a2cee66a9e02228aa970c6", + "prepared_review_unit_hash": "0412edd54149163229ba62e7516257b6a315206c23787e6d303d1d159b325b8d", "component_budget": { "node_count": 2, "item_count": 2, @@ -18823,18 +18880,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two PRs both touch expert-parallel loading, but they address different bugs and code paths: one fixes EP + ZeRO-3/accelerate launch config handling, the other fixes EP + FSDP2 expert tensors being overwritten by rank-0 broadcast.", - "confidence": 0.94, + "summary": "Both PRs are about expert-parallel (EP) model loading, but they target different failure modes and code paths: PR #45548 focuses on DeepSpeed ZeRO-3 / accelerate launch config handling, while PR #45662 fixes FSDP2/rank-0 broadcast overwriting expert weights during load. They are related but not duplicates.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "No clear canonical duplicate here; PR 45662 is the more self-contained fix for a specific overwrite bug in distributed loading, while 45548 targets a separate ZeRO-3/config serialization issue.", + "canonical_pr_reason": "PR #45662 is the better canonical PR for the cluster because it addresses the more concrete loading corruption bug and includes the broader set of code-path fixes around expert parameter sharding and assignment.", "best_issue_reason": null, - "best_pr_reason": "PR 45662 is the stronger representative of a concrete, narrowly scoped bug fix with multiple code-path changes, but it should not be treated as a duplicate of 45548.", + "best_pr_reason": "PR #45662 is the strongest representative overall: it fixes a specific, user-visible correctness issue in EP weight loading and touches the relevant loading/integration paths directly.", "soft_edge_verdicts": [ { "left": "pull_request:45548", "right": "pull_request:45662", "accept": false, - "reason": "Shared EP/distributed-loading theme is too broad; 45548 fixes accelerate/ZeRO-3 config handling and `distributed_config` serialization, while 45662 fixes FSDP2 rank-0 broadcast overwriting experts via `post_shard_wrap` and related loading paths." + "reason": "Different underlying bugs: #45548 is about EP + DeepSpeed ZeRO-3 loading via accelerate launch and config serialization/has_ep detection, while #45662 is about EP + FSDP2 where expert weights are silently overwritten by rank-0 broadcast. Shared subsystem is not enough to treat them as the same change." } ] }, @@ -18864,40 +18921,40 @@ "soft_pairs": [ "pull_request:45639|pull_request:45645" ], - "prepared_review_unit_hash": "7bdffe171c433d63389d9812273698bf553b4f96b3aa95218a39f09d39882fc2", + "prepared_review_unit_hash": "76ab4bed7ab21edeac39cf454fcb8f795104dc0b4ba4f9a5b10487b0fdfd58c4", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4589, - "estimated_input_tokens": 1148, - "estimated_eval_tokens": 2552 + "serialized_chars": 4588, + "estimated_input_tokens": 1147, + "estimated_eval_tokens": 2550 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4589, - "estimated_input_tokens": 1148, - "estimated_eval_tokens": 2552 + "serialized_chars": 4588, + "estimated_input_tokens": 1147, + "estimated_eval_tokens": 2550 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs target the same xdist-related captured_info logging problem; PR 45645 appears to subsume PR 45639 with the same testing_utils fix plus the CI artifact handling changes.", - "confidence": 0.89, + "summary": "Both PRs target the same xdist collision in captured_info logging; #45645 is the broader follow-up that includes the core fix from #45639 plus CI artifact handling updates.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45645 is the broader, more complete fix for the captured_info xdist collision issue: it includes the testing_utils change plus workflow/notification updates to handle multiple captured_info files in CI.", + "canonical_pr_reason": "#45645 is the more complete implementation of the shared captured_info xdist fix: it includes the worker-safe output-file change and also updates CI consumption of those logs.", "best_issue_reason": null, - "best_pr_reason": "PR 45645 is the best representative because it addresses the underlying bug end-to-end, while 45639 is a narrower subset limited to making the debug log write path xdist-safe.", + "best_pr_reason": "#45645 is the best single PR to keep because it covers the core bug fix and the necessary downstream workflow changes, whereas #45639 is only the smaller subset.", "soft_edge_verdicts": [ { "left": "pull_request:45639", "right": "pull_request:45645", "accept": true, - "reason": "Same underlying xdist/captured_info collision problem, same issue target, and 45645 appears to extend rather than diverge from 45639; these could plausibly be merged into one PR." + "reason": "Same underlying bug/change: making captured_info logging xdist-safe. #45645 extends the same fix path rather than introducing a separate feature." } ] }, diff --git a/analysis/current/manifest.json b/analysis/current/manifest.json index 40cbeec7407eac6061ca3583f7fe177051867ddc..a2a5df2cde11f291ea7dde46eed47c0fd5818240 100644 --- a/analysis/current/manifest.json +++ b/analysis/current/manifest.json @@ -1,8 +1,8 @@ { - "analysis_id": "hybrid-model-20260510t060019z", + "analysis_id": "hybrid-model-20260510t120026z", "archived_artifacts": { - "hybrid": "snapshots/20260510T060019Z/analysis-runs/hybrid-model-20260510t060019z/analysis-report-hybrid.json", - "hybrid_reviews": "snapshots/20260510T060019Z/analysis-runs/hybrid-model-20260510t060019z/analysis-report-hybrid.llm-reviews.json" + "hybrid": "snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/analysis-report-hybrid.llm-reviews.json" }, "artifacts": { "hybrid": "analysis/current/analysis-report-hybrid.json", @@ -10,9 +10,9 @@ }, "channel": "canonical", "model": null, - "published_at": "2026-05-10T06:03:19Z", + "published_at": "2026-05-10T12:08:02Z", "repo": "huggingface/transformers", "schema_version": 1, - "snapshot_id": "20260510T060019Z", + "snapshot_id": "20260510T120026Z", "variant": "hybrid" } diff --git a/snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/analysis-report-hybrid.json b/snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/analysis-report-hybrid.json new file mode 100644 index 0000000000000000000000000000000000000000..99211081567f6240017b8ed8528b7014e1fd37bb --- /dev/null +++ b/snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/analysis-report-hybrid.json @@ -0,0 +1,2076 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260510T120026Z", + "generated_at": "2026-05-10T12:08:00Z", + "evidence_quality": "full", + "llm_enrichment": true, + "meta_bugs": [ + { + "cluster_id": "cluster-43979-11", + "summary": "Cluster of 1 issues and 10 PRs centered on issue #43979.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 43979, + "canonical_pr_number": 44007, + "issue_numbers": [ + 43979 + ], + "pr_numbers": [ + 43996, + 44007, + 44013, + 44044, + 44066, + 44072, + 44085, + 44129, + 44154, + 44722 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 43996, + "right_pr_number": 44007, + "code_similarity": 0.179, + "size_similarity": 0.576, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.429, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44013, + "code_similarity": 0.122, + "size_similarity": 0.318, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.392, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44044, + "code_similarity": 0.245, + "size_similarity": 0.864, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.479, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44066, + "code_similarity": 0.225, + "size_similarity": 0.818, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.408, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44072, + "code_similarity": 0.14, + "size_similarity": 0.303, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.528, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44085, + "code_similarity": 0.216, + "size_similarity": 0.783, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.398, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44129, + "code_similarity": 0.163, + "size_similarity": 0.643, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.229, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44154, + "code_similarity": 0.153, + "size_similarity": 0.535, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.31, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44722, + "code_similarity": 0.225, + "size_similarity": 0.848, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.368, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44013, + "code_similarity": 0.19, + "size_similarity": 0.553, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.531, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44044, + "code_similarity": 0.186, + "size_similarity": 0.667, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.354, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44066, + "code_similarity": 0.188, + "size_similarity": 0.704, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.315, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44072, + "code_similarity": 0.212, + "size_similarity": 0.526, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.708, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44085, + "code_similarity": 0.195, + "size_similarity": 0.735, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.318, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44129, + "code_similarity": 0.103, + "size_similarity": 0.37, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.191, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44154, + "code_similarity": 0.238, + "size_similarity": 0.93, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.344, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44722, + "code_similarity": 0.178, + "size_similarity": 0.679, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.28, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44044, + "code_similarity": 0.126, + "size_similarity": 0.368, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.351, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44066, + "code_similarity": 0.127, + "size_similarity": 0.389, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.325, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44072, + "code_similarity": 0.29, + "size_similarity": 0.952, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.667, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44085, + "code_similarity": 0.131, + "size_similarity": 0.406, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.329, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44129, + "code_similarity": 0.07, + "size_similarity": 0.205, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.192, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44154, + "code_similarity": 0.177, + "size_similarity": 0.594, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.389, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44722, + "code_similarity": 0.118, + "size_similarity": 0.375, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.287, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44066, + "code_similarity": 0.25, + "size_similarity": 0.947, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.404, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44072, + "code_similarity": 0.136, + "size_similarity": 0.351, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.442, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44085, + "code_similarity": 0.24, + "size_similarity": 0.906, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.394, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44129, + "code_similarity": 0.147, + "size_similarity": 0.555, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.243, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44154, + "code_similarity": 0.17, + "size_similarity": 0.62, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.306, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44722, + "code_similarity": 0.257, + "size_similarity": 0.982, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.402, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44072, + "code_similarity": 0.133, + "size_similarity": 0.37, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.393, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44085, + "code_similarity": 0.763, + "size_similarity": 0.957, + "file_overlap": 0.5, + "area_overlap": 0.825, + "patch_similarity": 0.887, + "shared_filenames": [ + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 41 + ], + [ + 174, + 181 + ], + [ + 250, + 257 + ], + [ + 398, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 466 + ], + [ + 471, + 488 + ], + [ + 518, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 606 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ], + "right_ranges": [ + [ + 33, + 40 + ], + [ + 173, + 180 + ], + [ + 249, + 256 + ], + [ + 397, + 405 + ], + [ + 408, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 465 + ], + [ + 470, + 489 + ], + [ + 517, + 539 + ], + [ + 553, + 559 + ], + [ + 565, + 574 + ], + [ + 579, + 597 + ], + [ + 600, + 611 + ], + [ + 633, + 639 + ], + [ + 645, + 652 + ], + [ + 657, + 673 + ], + [ + 716, + 728 + ], + [ + 737, + 743 + ], + [ + 748, + 771 + ], + [ + 789, + 794 + ] + ] + } + ] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44129, + "code_similarity": 0.145, + "size_similarity": 0.526, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.263, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44154, + "code_similarity": 0.174, + "size_similarity": 0.654, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.286, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44722, + "code_similarity": 0.907, + "size_similarity": 0.964, + "file_overlap": 1.0, + "area_overlap": 0.808, + "patch_similarity": 0.874, + "shared_filenames": [ + "src/transformers/models/codegen/modeling_codegen.py", + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/codegen/modeling_codegen.py", + "left_ranges": [ + [ + 245, + 266 + ] + ], + "right_ranges": [ + [ + 228, + 234 + ] + ] + }, + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 41 + ], + [ + 174, + 181 + ], + [ + 250, + 257 + ], + [ + 398, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 466 + ], + [ + 471, + 488 + ], + [ + 518, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 606 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ], + "right_ranges": [ + [ + 33, + 42 + ], + [ + 175, + 182 + ], + [ + 251, + 258 + ], + [ + 399, + 420 + ], + [ + 426, + 435 + ], + [ + 460, + 467 + ], + [ + 472, + 489 + ], + [ + 519, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 612 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 663 + ], + [ + 666, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ] + } + ] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44085, + "code_similarity": 0.137, + "size_similarity": 0.387, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.398, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44129, + "code_similarity": 0.074, + "size_similarity": 0.195, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.231, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44154, + "code_similarity": 0.175, + "size_similarity": 0.566, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.414, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44722, + "code_similarity": 0.124, + "size_similarity": 0.357, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.347, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44129, + "code_similarity": 0.141, + "size_similarity": 0.503, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.272, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44154, + "code_similarity": 0.18, + "size_similarity": 0.684, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.289, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44722, + "code_similarity": 0.728, + "size_similarity": 0.923, + "file_overlap": 0.5, + "area_overlap": 0.791, + "patch_similarity": 0.78, + "shared_filenames": [ + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 40 + ], + [ + 173, + 180 + ], + [ + 249, + 256 + ], + [ + 397, + 405 + ], + [ + 408, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 465 + ], + [ + 470, + 489 + ], + [ + 517, + 539 + ], + [ + 553, + 559 + ], + [ + 565, + 574 + ], + [ + 579, + 597 + ], + [ + 600, + 611 + ], + [ + 633, + 639 + ], + [ + 645, + 652 + ], + [ + 657, + 673 + ], + [ + 716, + 728 + ], + [ + 737, + 743 + ], + [ + 748, + 771 + ], + [ + 789, + 794 + ] + ], + "right_ranges": [ + [ + 33, + 42 + ], + [ + 175, + 182 + ], + [ + 251, + 258 + ], + [ + 399, + 420 + ], + [ + 426, + 435 + ], + [ + 460, + 467 + ], + [ + 472, + 489 + ], + [ + 519, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 612 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 663 + ], + [ + 666, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ] + } + ] + }, + { + "left_pr_number": 44129, + "right_pr_number": 44154, + "code_similarity": 0.099, + "size_similarity": 0.344, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.199, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44129, + "right_pr_number": 44722, + "code_similarity": 0.146, + "size_similarity": 0.545, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.247, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44154, + "right_pr_number": 44722, + "code_similarity": 0.164, + "size_similarity": 0.631, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.255, + "shared_filenames": [], + "shared_file_areas": [] + } + ] + }, + { + "cluster_id": "cluster-41211-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #41211.", + "status": "open", + "confidence": 0.55, + "canonical_issue_number": 41211, + "canonical_pr_number": 44339, + "issue_numbers": [ + 41211 + ], + "pr_numbers": [ + 41356, + 44339 + ], + "evidence_types": [ + "closing_reference" + ], + "pr_comparisons": [ + { + "left_pr_number": 41356, + "right_pr_number": 44339, + "code_similarity": 0.155, + "size_similarity": 0.094, + "file_overlap": 0.273, + "area_overlap": 0.078, + "patch_similarity": 0.183, + "shared_filenames": [ + "docs/source/en/model_doc/deimv2.md", + "src/transformers/models/deimv2/__init__.py", + "src/transformers/models/deimv2/configuration_deimv2.py", + "src/transformers/models/deimv2/modeling_deimv2.py", + "tests/models/deimv2/__init__.py", + "tests/models/deimv2/test_modeling_deimv2.py" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/model_doc/deimv2.md", + "left_ranges": [ + [ + 1, + 132 + ] + ], + "right_ranges": [ + [ + 1, + 65 + ] + ] + }, + { + "filename": "src/transformers/models/deimv2/__init__.py", + "left_ranges": [ + [ + 1, + 15 + ] + ], + "right_ranges": [ + [ + 1, + 29 + ] + ] + }, + { + "filename": "src/transformers/models/deimv2/configuration_deimv2.py", + "left_ranges": [ + [ + 1, + 74 + ] + ], + "right_ranges": [ + [ + 1, + 266 + ] + ] + }, + { + "filename": "tests/models/deimv2/test_modeling_deimv2.py", + "left_ranges": [ + [ + 1, + 15 + ] + ], + "right_ranges": [ + [ + 1, + 1734 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-43240-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #43240.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 43240, + "canonical_pr_number": 43251, + "issue_numbers": [ + 43240 + ], + "pr_numbers": [ + 43251, + 43254 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 43251, + "right_pr_number": 43254, + "code_similarity": 0.794, + "size_similarity": 0.64, + "file_overlap": 1.0, + "area_overlap": 0.667, + "patch_similarity": 0.882, + "shared_filenames": [ + "src/transformers/loss/loss_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/loss/loss_utils.py", + "left_ranges": [ + [ + 30, + 50 + ] + ], + "right_ranges": [ + [ + 30, + 43 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-41115-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #41115.", + "status": "open", + "confidence": 0.55, + "canonical_issue_number": 41115, + "canonical_pr_number": 45613, + "issue_numbers": [ + 41115 + ], + "pr_numbers": [ + 41116, + 45613 + ], + "evidence_types": [ + "closing_reference" + ], + "pr_comparisons": [ + { + "left_pr_number": 41116, + "right_pr_number": 45613, + "code_similarity": 0.805, + "size_similarity": 0.885, + "file_overlap": 0.75, + "area_overlap": 0.854, + "patch_similarity": 0.693, + "shared_filenames": [ + "docs/source/en/_toctree.yml", + "docs/source/en/model_doc/minicpm3.md", + "src/transformers/models/auto/modeling_auto.py", + "src/transformers/models/minicpm3/__init__.py", + "src/transformers/models/minicpm3/configuration_minicpm3.py", + "src/transformers/models/minicpm3/modeling_minicpm3.py", + "src/transformers/models/minicpm3/modular_minicpm3.py", + "tests/models/minicpm3/__init__.py", + "tests/models/minicpm3/test_modeling_minicpm3.py" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/_toctree.yml", + "left_ranges": [ + [ + 713, + 720 + ] + ], + "right_ranges": [ + [ + 713, + 720 + ] + ] + }, + { + "filename": "docs/source/en/model_doc/minicpm3.md", + "left_ranges": [ + [ + 1, + 67 + ] + ], + "right_ranges": [ + [ + 1, + 45 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 285, + 291 + ], + [ + 700, + 706 + ], + [ + 1298, + 1304 + ] + ], + "right_ranges": [ + [ + 699, + 705 + ], + [ + 1301, + 1307 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/__init__.py", + "left_ranges": [ + [ + 1, + 29 + ] + ], + "right_ranges": [ + [ + 1, + 29 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/configuration_minicpm3.py", + "left_ranges": [ + [ + 1, + 141 + ] + ], + "right_ranges": [ + [ + 1, + 126 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/modeling_minicpm3.py", + "left_ranges": [ + [ + 1, + 544 + ] + ], + "right_ranges": [ + [ + 1, + 522 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/modular_minicpm3.py", + "left_ranges": [ + [ + 1, + 444 + ] + ], + "right_ranges": [ + [ + 1, + 342 + ] + ] + }, + { + "filename": "tests/models/minicpm3/test_modeling_minicpm3.py", + "left_ranges": [ + [ + 1, + 109 + ] + ], + "right_ranges": [ + [ + 1, + 136 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-45647-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45647.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 45647, + "canonical_pr_number": 45738, + "issue_numbers": [ + 45647 + ], + "pr_numbers": [ + 45737, + 45738 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 45737, + "right_pr_number": 45738, + "code_similarity": 0.526, + "size_similarity": 0.11, + "file_overlap": 0.5, + "area_overlap": 1.0, + "patch_similarity": 0.024, + "shared_filenames": [ + "src/transformers/models/musicgen_melody/modeling_musicgen_melody.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/musicgen_melody/modeling_musicgen_melody.py", + "left_ranges": [ + [ + 491, + 497 + ] + ], + "right_ranges": [ + [ + 491, + 497 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-43824-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #43824.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 43824, + "canonical_pr_number": 43836, + "issue_numbers": [ + 43824 + ], + "pr_numbers": [ + 43836, + 43842 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 43836, + "right_pr_number": 43842, + "code_similarity": 0.332, + "size_similarity": 0.088, + "file_overlap": 1.0, + "area_overlap": 0.033, + "patch_similarity": 0.017, + "shared_filenames": [ + "src/transformers/cli/serve.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/cli/serve.py", + "left_ranges": [ + [ + 11, + 18 + ], + [ + 359, + 429 + ], + [ + 584, + 590 + ], + [ + 1892, + 1910 + ], + [ + 1917, + 1923 + ] + ], + "right_ranges": [ + [ + 54, + 61 + ], + [ + 587, + 593 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-45081-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45081.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 45081, + "canonical_pr_number": 45317, + "issue_numbers": [ + 45081 + ], + "pr_numbers": [ + 45086, + 45317 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 45086, + "right_pr_number": 45317, + "code_similarity": 0.589, + "size_similarity": 0.34, + "file_overlap": 0.5, + "area_overlap": 1.0, + "patch_similarity": 0.136, + "shared_filenames": [ + "src/transformers/tokenization_utils_tokenizers.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/tokenization_utils_tokenizers.py", + "left_ranges": [ + [ + 1360, + 1370 + ], + [ + 1374, + 1380 + ] + ], + "right_ranges": [ + [ + 1360, + 1370 + ], + [ + 1374, + 1380 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-45561-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45561.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 45561, + "canonical_pr_number": 45645, + "issue_numbers": [ + 45561 + ], + "pr_numbers": [ + 45639, + 45645 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 45639, + "right_pr_number": 45645, + "code_similarity": 0.532, + "size_similarity": 0.619, + "file_overlap": 0.5, + "area_overlap": 0.543, + "patch_similarity": 0.451, + "shared_filenames": [ + "src/transformers/testing_utils.py", + "tests/utils/test_testing_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/testing_utils.py", + "left_ranges": [ + [ + 3529, + 3536 + ], + [ + 3753, + 3787 + ] + ], + "right_ranges": [ + [ + 3525, + 3558 + ], + [ + 3782, + 3788 + ] + ] + }, + { + "filename": "tests/utils/test_testing_utils.py", + "left_ranges": [ + [ + 1, + 86 + ] + ], + "right_ranges": [ + [ + 1, + 114 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-43698-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #43698.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 43698, + "canonical_pr_number": 43779, + "issue_numbers": [ + 43698 + ], + "pr_numbers": [ + 43779, + 43816 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 43779, + "right_pr_number": 43816, + "code_similarity": 0.418, + "size_similarity": 0.538, + "file_overlap": 1.0, + "area_overlap": 0.02, + "patch_similarity": 0.02, + "shared_filenames": [ + "src/transformers/integrations/integration_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/integrations/integration_utils.py", + "left_ranges": [ + [ + 2227, + 2241 + ], + [ + 2303, + 2309 + ] + ], + "right_ranges": [ + [ + 2278, + 2291 + ], + [ + 2309, + 2322 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-44018-2", + "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "status": "open", + "confidence": 0.5, + "canonical_issue_number": null, + "canonical_pr_number": 44068, + "issue_numbers": [], + "pr_numbers": [ + 44018, + 44068 + ], + "evidence_types": [ + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 44018, + "right_pr_number": 44068, + "code_similarity": 0.766, + "size_similarity": 0.939, + "file_overlap": 1.0, + "area_overlap": 0.425, + "patch_similarity": 0.866, + "shared_filenames": [ + "src/transformers/models/gpt_neo/modeling_gpt_neo.py", + "tests/models/gpt_neo/test_modeling_gpt_neo.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/gpt_neo/modeling_gpt_neo.py", + "left_ranges": [ + [ + 26, + 46 + ], + [ + 138, + 143 + ], + [ + 182, + 187 + ], + [ + 283, + 295 + ], + [ + 327, + 341 + ], + [ + 348, + 354 + ], + [ + 360, + 369 + ], + [ + 399, + 406 + ], + [ + 411, + 419 + ], + [ + 428, + 433 + ], + [ + 472, + 492 + ], + [ + 507, + 513 + ], + [ + 519, + 528 + ], + [ + 541, + 559 + ], + [ + 562, + 567 + ], + [ + 595, + 601 + ], + [ + 607, + 614 + ], + [ + 627, + 643 + ], + [ + 685, + 690 + ], + [ + 708, + 714 + ], + [ + 720, + 727 + ], + [ + 740, + 757 + ], + [ + 761, + 766 + ], + [ + 780, + 786 + ], + [ + 791, + 798 + ], + [ + 807, + 822 + ], + [ + 840, + 845 + ] + ], + "right_ranges": [ + [ + 26, + 31 + ], + [ + 34, + 44 + ], + [ + 136, + 141 + ], + [ + 180, + 185 + ], + [ + 281, + 293 + ], + [ + 325, + 339 + ], + [ + 346, + 352 + ], + [ + 358, + 367 + ], + [ + 397, + 404 + ], + [ + 409, + 417 + ], + [ + 426, + 434 + ], + [ + 463, + 483 + ], + [ + 498, + 504 + ], + [ + 510, + 519 + ], + [ + 532, + 550 + ], + [ + 553, + 564 + ], + [ + 586, + 592 + ], + [ + 598, + 605 + ], + [ + 618, + 634 + ], + [ + 676, + 688 + ], + [ + 699, + 705 + ], + [ + 711, + 718 + ], + [ + 731, + 748 + ], + [ + 752, + 762 + ], + [ + 771, + 777 + ], + [ + 782, + 789 + ], + [ + 798, + 813 + ], + [ + 831, + 836 + ] + ] + }, + { + "filename": "tests/models/gpt_neo/test_modeling_gpt_neo.py", + "left_ranges": [ + [ + 458, + 464 + ] + ], + "right_ranges": [ + [ + 458, + 464 + ] + ] + } + ] + } + ] + } + ], + "duplicate_issues": [], + "duplicate_prs": [ + { + "cluster_id": "cluster-41115-3", + "canonical_pr_number": 45613, + "duplicate_pr_numbers": [ + 41116 + ], + "target_issue_number": 41115, + "reason": "PRs in cluster-41115-3 are treated as duplicates because they converge on issue #41115 with closing_reference evidence." + }, + { + "cluster_id": "cluster-41211-3", + "canonical_pr_number": 44339, + "duplicate_pr_numbers": [ + 41356 + ], + "target_issue_number": 41211, + "reason": "PRs in cluster-41211-3 are treated as duplicates because they converge on issue #41211 with closing_reference evidence." + }, + { + "cluster_id": "cluster-43240-3", + "canonical_pr_number": 43251, + "duplicate_pr_numbers": [ + 43254 + ], + "target_issue_number": 43240, + "reason": "PRs in cluster-43240-3 are treated as duplicates because they converge on issue #43240 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-43698-3", + "canonical_pr_number": 43779, + "duplicate_pr_numbers": [ + 43816 + ], + "target_issue_number": 43698, + "reason": "PRs in cluster-43698-3 are treated as duplicates because they converge on issue #43698 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-43824-3", + "canonical_pr_number": 43836, + "duplicate_pr_numbers": [ + 43842 + ], + "target_issue_number": 43824, + "reason": "PRs in cluster-43824-3 are treated as duplicates because they converge on issue #43824 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-43979-11", + "canonical_pr_number": 44007, + "duplicate_pr_numbers": [ + 43996, + 44013, + 44044, + 44066, + 44072, + 44085, + 44129, + 44154, + 44722 + ], + "target_issue_number": 43979, + "reason": "PRs in cluster-43979-11 are treated as duplicates because they converge on issue #43979 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-44018-2", + "canonical_pr_number": 44068, + "duplicate_pr_numbers": [ + 44018 + ], + "target_issue_number": null, + "reason": "PRs in cluster-44018-2 are treated as duplicates because they share soft_similarity evidence." + }, + { + "cluster_id": "cluster-45081-3", + "canonical_pr_number": 45317, + "duplicate_pr_numbers": [ + 45086 + ], + "target_issue_number": 45081, + "reason": "PRs in cluster-45081-3 are treated as duplicates because they converge on issue #45081 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-45561-3", + "canonical_pr_number": 45645, + "duplicate_pr_numbers": [ + 45639 + ], + "target_issue_number": 45561, + "reason": "PRs in cluster-45561-3 are treated as duplicates because they converge on issue #45561 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-45647-3", + "canonical_pr_number": 45738, + "duplicate_pr_numbers": [ + 45737 + ], + "target_issue_number": 45647, + "reason": "PRs in cluster-45647-3 are treated as duplicates because they converge on issue #45647 with closing_reference, shared_issue_target evidence." + } + ], + "best_issue": { + "cluster_id": "cluster-45647-3", + "issue_number": 45647, + "reason": "Issue #45647 is the strongest global issue candidate because it is open, belongs to a cluster with 3 artifacts, and carries 12 discussion comments plus 8 inbound references.", + "score": 162.0 + }, + "best_pr": { + "cluster_id": "cluster-43979-11", + "pr_number": 44007, + "reason": "PR #44007 is the strongest global PR candidate because it is open, belongs to a cluster with 11 artifacts, links to 1 issues, and carries 2 review/discussion events.", + "score": 238.886 + } +} diff --git a/snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/analysis-report-hybrid.llm-reviews.json b/snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/analysis-report-hybrid.llm-reviews.json new file mode 100644 index 0000000000000000000000000000000000000000..b5c6150e84ea28faaa865d6bfa33f9d950698400 --- /dev/null +++ b/snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/analysis-report-hybrid.llm-reviews.json @@ -0,0 +1,19098 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260510T120026Z", + "generated_at": "2026-05-10T12:08:00Z", + "model": "gpt-5.4-mini?service_tier=flex", + "reviews": [ + { + "label": "issue", + "component_index": 1, + "component_count": 16, + "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": "33c647724a046cb82f8c07e404fdcd3910777b05e8d41a790ff5636a7e5d1424", + "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": "These issues are not true duplicates. #16618 is a broad model-loading failure, while #15980 is specifically an authentication/error-message bug for private model downloads, and #13244 is about Tapas tokenization behavior versus TensorFlow code. The only plausible cluster anchor is #16618 by breadth, but it is not the same underlying problem as the others.", + "confidence": 0.96, + "canonical_issue_reason": "#16618 is the broadest and most central-sounding issue in the set, but it still does not actually cover the other reports\u2019 distinct failure modes.", + "canonical_pr_reason": null, + "best_issue_reason": "#16618 is the best representative only because it is the most generic model-loading report; however, it should not be treated as a duplicate of the other two.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:13244", + "right": "issue:16618", + "accept": false, + "reason": "Tapas tokenization mismatch is a different subsystem and symptom than a generic Bert model-loading failure; no clear shared code-path bug." + }, + { + "left": "issue:15980", + "right": "issue:16618", + "accept": false, + "reason": "#15980 is about downloading a private model while logged out and the resulting error message, which is an auth/UX issue, not the same underlying load failure as #16618." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 1, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:24643", + "issue:30277", + "issue:34634", + "issue:34928", + "issue:35545", + "issue:36010", + "issue:36331", + "issue:39290", + "issue:41093", + "issue:41720", + "issue:41762", + "issue:42915", + "issue:43039", + "issue:43404", + "issue:43425", + "issue:43531", + "issue:43541", + "issue:43596", + "issue:43638", + "issue:43716", + "issue:43792", + "issue:43827", + "issue:43828", + "issue:43854", + "issue:43866", + "issue:43901", + "issue:43975", + "issue:44292", + "issue:44322", + "issue:44457", + "issue:44509", + "issue:44512", + "issue:44560", + "issue:44805", + "issue:44841", + "issue:44863", + "issue:44918", + "issue:45070", + "issue:45081", + "issue:45084", + "issue:45161", + "issue:45237", + "issue:45362", + "issue:45464", + "issue:45507", + "issue:45685", + "issue:45800", + "issue:45823" + ], + "soft_pairs": [ + "issue:44322|issue:45464", + "issue:45237|issue:45685", + "issue:43716|issue:45237", + "issue:43828|issue:45237", + "issue:45237|issue:45507", + "issue:44292|issue:45237", + "issue:30277|issue:35545", + "issue:35545|issue:42915", + "issue:35545|issue:41720", + "issue:44509|issue:44512", + "issue:35545|issue:36010", + "issue:44805|issue:45823", + "issue:43638|issue:44805", + "issue:43596|issue:44805", + "issue:44805|issue:45161", + "issue:44805|issue:44918", + "issue:34634|issue:35545", + "issue:30277|issue:43638", + "issue:43596|issue:43638", + "issue:43039|issue:45161", + "issue:43975|issue:44457", + "issue:44292|issue:44457", + "issue:34928|issue:44805", + "issue:24643|issue:30277", + "issue:43638|issue:45161", + "issue:34928|issue:36331", + "issue:43596|issue:45161", + "issue:43716|issue:45507", + "issue:39290|issue:45081", + "issue:43828|issue:45161", + "issue:43425|issue:44292", + "issue:44918|issue:45161", + "issue:34928|issue:41720", + "issue:43716|issue:45685", + "issue:43541|issue:43716", + "issue:43828|issue:44292", + "issue:45507|issue:45685", + "issue:43716|issue:44292", + "issue:44292|issue:44918", + "issue:43039|issue:44918", + "issue:41762|issue:43638", + "issue:43827|issue:44512", + "issue:43716|issue:44560", + "issue:43828|issue:45685", + "issue:44560|issue:45685", + "issue:43039|issue:43828", + "issue:30277|issue:36331", + "issue:34928|issue:36010", + "issue:43039|issue:44805", + "issue:43901|issue:44512", + "issue:41762|issue:43596", + "issue:39290|issue:41720", + "issue:44841|issue:45823", + "issue:43039|issue:44560", + "issue:43866|issue:44863", + "issue:43854|issue:43866", + "issue:39290|issue:43531", + "issue:41093|issue:43638", + "issue:43425|issue:45070", + "issue:43596|issue:45800", + "issue:43792|issue:45070", + "issue:45362|issue:45464", + "issue:43404|issue:43866", + "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:45800|issue:45823", + "issue:41093|issue:41762" + ], + "prepared_review_unit_hash": "86527d6645f7f52158d9fb70f3d41248df1b91bebc892599501571b5c47b9170", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 71, + "serialized_chars": 48867, + "estimated_input_tokens": 12217, + "estimated_eval_tokens": 24690 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is very noisy: most pairs are false positives across unrelated model, runtime, and docs issues. The only clear duplicate family is the DeepSpeed ZeRO-3/Bert index-0-out-of-bounds reports.", + "confidence": 0.61, + "canonical_issue_reason": "Issue 43638 is the clearest representative of the only true duplicate family here: the ZeRO-3/BertModel index-0-out-of-bounds failure. Its wording is broad enough to cover the paired report.", + "canonical_pr_reason": null, + "best_issue_reason": "Best overall representative of the cluster\u2019s only confirmed duplicate subtopic; the rest of the cluster is mostly unrelated or too loosely similar.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44322", + "right": "issue:45464", + "accept": false, + "reason": "Different Qwen3.5 failures: config attribute missing vs streaming chat/completions issue." + }, + { + "left": "issue:45237", + "right": "issue:45685", + "accept": false, + "reason": "Different MoE/runtime paths and different hardware backends; not the same bug." + }, + { + "left": "issue:43716", + "right": "issue:45237", + "accept": false, + "reason": "Mistral-3 dtype mismatch is unrelated to GPT-OSS AMD GPU failure." + }, + { + "left": "issue:43828", + "right": "issue:45237", + "accept": false, + "reason": "Different models and different dtype/device error paths." + }, + { + "left": "issue:45237", + "right": "issue:45507", + "accept": false, + "reason": "GPU runtime failure vs invalid method call in a different MoE model." + }, + { + "left": "issue:44292", + "right": "issue:45237", + "accept": false, + "reason": "Different model families and different NVFP4/GPU failure modes." + }, + { + "left": "issue:30277", + "right": "issue:35545", + "accept": false, + "reason": "DeepSpeed collective mismatch is unrelated to ONNX export." + }, + { + "left": "issue:35545", + "right": "issue:42915", + "accept": false, + "reason": "ONNX export error vs FineGrainedFP8 model failure; unrelated." + }, + { + "left": "issue:35545", + "right": "issue:41720", + "accept": false, + "reason": "Different model/export path vs auto device mapping CUDA assert." + }, + { + "left": "issue:44509", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs regressions, but they cover different removed docs topics." + }, + { + "left": "issue:35545", + "right": "issue:36010", + "accept": false, + "reason": "Export error vs import error for GenerationMixin; unrelated." + }, + { + "left": "issue:44805", + "right": "issue:45823", + "accept": false, + "reason": "Mask/index shape mismatch vs Gemma4 device-mapping mismatch; different bugs." + }, + { + "left": "issue:43638", + "right": "issue:44805", + "accept": false, + "reason": "Both are IndexError-like, but one is ZeRO-3/Bert loading and the other is a mask-shape mismatch." + }, + { + "left": "issue:43596", + "right": "issue:44805", + "accept": false, + "reason": "Different failure mode: ZeRO-3 Bert loading vs mask/tensor shape mismatch." + }, + { + "left": "issue:44805", + "right": "issue:45161", + "accept": false, + "reason": "Different problems: mask shape mismatch vs GPT-OSS tensor-parallel issue." + }, + { + "left": "issue:44805", + "right": "issue:44918", + "accept": false, + "reason": "Different models and different SFT/input-embedding failure paths." + }, + { + "left": "issue:34634", + "right": "issue:35545", + "accept": false, + "reason": "BarkProcessor voice_preset issue is unrelated to ModernBERT ONNX export." + }, + { + "left": "issue:30277", + "right": "issue:43638", + "accept": false, + "reason": "Different DeepSpeed symptoms and different models; not the same bug." + }, + { + "left": "issue:43596", + "right": "issue:43638", + "accept": true, + "reason": "Same ZeRO-3/Bert index-0-out-of-bounds failure with only minor wording differences." + }, + { + "left": "issue:43039", + "right": "issue:45161", + "accept": false, + "reason": "Liger Kernel cross-entropy call issue vs GPT-OSS tensor-parallel failure; unrelated." + }, + { + "left": "issue:43975", + "right": "issue:44457", + "accept": false, + "reason": "Detokenization bug vs LoRA merge/save/load inconsistency; unrelated." + }, + { + "left": "issue:44292", + "right": "issue:44457", + "accept": false, + "reason": "Different model/runtime bug vs LoRA checkpoint consistency issue." + }, + { + "left": "issue:34928", + "right": "issue:44805", + "accept": false, + "reason": "FSDP checkpointing size mismatch is not the same as mask-index mismatch." + }, + { + "left": "issue:24643", + "right": "issue:30277", + "accept": false, + "reason": "Different DeepSpeed errors and different code paths." + }, + { + "left": "issue:43638", + "right": "issue:45161", + "accept": false, + "reason": "ZeRO-3 Bert loading bug is unrelated to GPT-OSS tensor-parallel failure." + }, + { + "left": "issue:34928", + "right": "issue:36331", + "accept": false, + "reason": "Activation-checkpointing size mismatch is unrelated to CustomTrainer signature mismatch." + }, + { + "left": "issue:43596", + "right": "issue:45161", + "accept": false, + "reason": "Different model/load bug vs GPT-OSS TP issue." + }, + { + "left": "issue:43716", + "right": "issue:45507", + "accept": false, + "reason": "Different models and different dtype/method errors." + }, + { + "left": "issue:39290", + "right": "issue:45081", + "accept": false, + "reason": "Gemma3/vLLM config attribute issue is unrelated to Mistral tokenizer regex crash." + }, + { + "left": "issue:43828", + "right": "issue:45161", + "accept": false, + "reason": "Different models and different dtype/TP failures." + }, + { + "left": "issue:44918", + "right": "issue:45161", + "accept": false, + "reason": "TRL embedding unpacking issue is unrelated to GPT-OSS TP failure." + }, + { + "left": "issue:34928", + "right": "issue:41720", + "accept": false, + "reason": "FSDP/activation-checkpointing mismatch is unrelated to CUDA assert on auto device mapping." + }, + { + "left": "issue:43716", + "right": "issue:45685", + "accept": false, + "reason": "Different model-specific dtype issue vs MPS histogram op problem." + }, + { + "left": "issue:43541", + "right": "issue:43716", + "accept": false, + "reason": "Grouped_mm tracing error vs dtype mismatch; not the same bug." + }, + { + "left": "issue:43828", + "right": "issue:44292", + "accept": false, + "reason": "Different MoE models and different dtype/runtime failures." + }, + { + "left": "issue:43039", + "right": "issue:44918", + "accept": false, + "reason": "Cross-entropy/Liger Kernel issue is unrelated to TRL embedding unpacking." + }, + { + "left": "issue:41762", + "right": "issue:43638", + "accept": false, + "reason": "Same broad ZeRO-3 theme, but different model and load path; not clearly the same bug." + }, + { + "left": "issue:43827", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs cleanups for v5, but they reference different removed content." + }, + { + "left": "issue:43716", + "right": "issue:44560", + "accept": false, + "reason": "Different model/runtime issues with no shared code path." + }, + { + "left": "issue:43828", + "right": "issue:45685", + "accept": false, + "reason": "Different models, backends, and error types." + }, + { + "left": "issue:44560", + "right": "issue:45685", + "accept": false, + "reason": "Qwen3-vl StopIteration is unrelated to MPS histogram Int error." + }, + { + "left": "issue:43039", + "right": "issue:43828", + "accept": false, + "reason": "Different model/runtime issues despite both involving execution failures." + }, + { + "left": "issue:30277", + "right": "issue:36331", + "accept": false, + "reason": "DeepSpeed collective mismatch is unrelated to CustomTrainer compute_loss signature." + }, + { + "left": "issue:34928", + "right": "issue:36010", + "accept": false, + "reason": "FSDP size mismatch is unrelated to GenerationMixin import failure." + }, + { + "left": "issue:43039", + "right": "issue:44805", + "accept": false, + "reason": "Liger Kernel cross-entropy call issue is not the same as mask-shape IndexError." + }, + { + "left": "issue:43901", + "right": "issue:44512", + "accept": false, + "reason": "Different docs issues: return_all_scores vs removed run command." + }, + { + "left": "issue:41762", + "right": "issue:43596", + "accept": false, + "reason": "Both are ZeRO-3/load related, but different models and error contexts." + }, + { + "left": "issue:39290", + "right": "issue:41720", + "accept": false, + "reason": "Gemma3/vLLM config bug is unrelated to Qwen3 device-mapping CUDA assert." + }, + { + "left": "issue:44841", + "right": "issue:45823", + "accept": false, + "reason": "Processor failure for Voxtral is unrelated to Gemma4 PLE device mismatch." + }, + { + "left": "issue:43039", + "right": "issue:44560", + "accept": false, + "reason": "Liger Kernel cross-entropy issue is unrelated to Qwen3-vl StopIteration." + }, + { + "left": "issue:43866", + "right": "issue:44863", + "accept": false, + "reason": "Corrupted checkpoint vs implementation unable to load checkpoints; different root causes." + }, + { + "left": "issue:43854", + "right": "issue:43866", + "accept": false, + "reason": "Unit-test load failure for GLM-4.7 is unrelated to Ovis2 checkpoint corruption." + }, + { + "left": "issue:39290", + "right": "issue:43531", + "accept": false, + "reason": "Gemma3/vLLM config attribute issue is unrelated to Qwen3-MoE sliding_window bug." + }, + { + "left": "issue:41093", + "right": "issue:43638", + "accept": false, + "reason": "Both are IndexErrors, but one is a mask/tensor shape mismatch and the other is ZeRO-3 Bert loading." + }, + { + "left": "issue:43425", + "right": "issue:45070", + "accept": false, + "reason": "Torch-version incompatibility vs pydantic PretrainedConfig field regression; unrelated." + }, + { + "left": "issue:43596", + "right": "issue:45800", + "accept": false, + "reason": "ZeRO-3 Bert loading bug is unrelated to torch/transformers version incompatibility." + }, + { + "left": "issue:43792", + "right": "issue:45070", + "accept": false, + "reason": "Whisper run failure is unrelated to the pydantic model-field regression." + }, + { + "left": "issue:45362", + "right": "issue:45464", + "accept": false, + "reason": "Qwen3.5 chat crash vs streaming inference failure; different behavior and likely different bugs." + }, + { + "left": "issue:43404", + "right": "issue:43866", + "accept": false, + "reason": "Mistral3 lm_head tying bug is unrelated to Ovis2 checkpoint corruption." + }, + { + "left": "issue:43828", + "right": "issue:45685", + "accept": false, + "reason": "Different models and backends; not the same issue." + }, + { + "left": "issue:43541", + "right": "issue:45685", + "accept": false, + "reason": "Grouped_mm tracing error is unrelated to MPS histogram Int error." + }, + { + "left": "issue:43541", + "right": "issue:43828", + "accept": false, + "reason": "Different MoE runtime failures and different models." + }, + { + "left": "issue:36010", + "right": "issue:39290", + "accept": false, + "reason": "GenerationMixin import failure is unrelated to Gemma3/vLLM config mismatch." + }, + { + "left": "issue:44841", + "right": "issue:45084", + "accept": false, + "reason": "Processor failure for Voxtral is unrelated to template-node compilation error." + }, + { + "left": "issue:45800", + "right": "issue:45823", + "accept": false, + "reason": "Torch/transformers incompatibility is unrelated to Gemma4 device mismatch." + }, + { + "left": "issue:41093", + "right": "issue:41762", + "accept": false, + "reason": "Both are IndexErrors, but different models and different failure mechanisms." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The only accepted duplicate pair, issue:43596 \u2194 issue:43638, is well supported by the titles and descriptions. The rest of the soft-edge false verdicts are appropriately cautious. Minor caveat: issue:41762 is adjacent to the ZeRO-3 cluster, but treating it as non-duplicate is a defensible conservative call." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 2, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:39186", + "issue:39692", + "issue:40990", + "issue:41093", + "issue:42915", + "issue:43054", + "issue:43296", + "issue:43366", + "issue:43404", + "issue:43531", + "issue:43541", + "issue:43572", + "issue:43596", + "issue:43644", + "issue:43645", + "issue:43716", + "issue:43828", + "issue:43866", + "issue:43873", + "issue:43950", + "issue:43975", + "issue:43994", + "issue:44291", + "issue:44292", + "issue:44360", + "issue:44387", + "issue:44403", + "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:45161", + "issue:45399", + "issue:45464", + "issue:45507", + "issue:45542", + "issue:45632", + "issue:45685", + "issue:45698", + "issue:45800" + ], + "soft_pairs": [ + "issue:43296|issue:43366", + "issue:43572|issue:45070", + "issue:44387|issue:45005", + "issue:43541|issue:45161", + "issue:43828|issue:45464", + "issue:44291|issue:45070", + "issue:43366|issue:45464", + "issue:45020|issue:45800", + "issue:43531|issue:45070", + "issue:43541|issue:44589", + "issue:41093|issue:43596", + "issue:43054|issue:43994", + "issue:45542|issue:45800", + "issue:43716|issue:44918", + "issue:44918|issue:45685", + "issue:45632|issue:45698", + "issue:43404|issue:45005", + "issue:43645|issue:45399", + "issue:40990|issue:44960", + "issue:44530|issue:45005", + "issue:44360|issue:44512", + "issue:42915|issue:45399", + "issue:44863|issue:45005", + "issue:43873|issue:45005", + "issue:39186|issue:44918", + "issue:44492|issue:44512", + "issue:44918|issue:45507", + "issue:43975|issue:44292", + "issue:43950|issue:44534", + "issue:43644|issue:43950", + "issue:44292|issue:45507", + "issue:39692|issue:43054", + "issue:43828|issue:44928", + "issue:43866|issue:44403" + ], + "prepared_review_unit_hash": "0efb76dcc62da0164ea19bdb6b27457dcc36aea720855733fbfb3e1518716a4d", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37472, + "estimated_input_tokens": 9368, + "estimated_eval_tokens": 18992 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a set of loosely related Transformers/v5 regression reports, but only the non-persistent-buffer corruption reports form a true duplicate subcluster.", + "confidence": 0.84, + "canonical_issue_reason": "issue:43950 is the clearest root report for the only exact duplicate subcluster here: non-persistent buffers being corrupted on `from_pretrained()` in v5, with direct matches in 43644 and 44534.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43950 is the best anchor issue because it states the regression most precisely and covers the other matching reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43296", + "right": "issue:43366", + "accept": false, + "reason": "Different problems: vLLM/PaddleOCR-VL loading failure versus GGUF gpt-oss architecture support." + }, + { + "left": "issue:43572", + "right": "issue:45070", + "accept": false, + "reason": "Both are v5 config regressions, but one is missing `pad_token_idx` and the other is a pydantic `PretrainedConfig` field break; not the same bug." + }, + { + "left": "issue:44387", + "right": "issue:45005", + "accept": false, + "reason": "Related to v5 and weights, but one is int4 memory/OOM behavior and the other is tied-weights handling on translation models; different code paths." + }, + { + "left": "issue:43541", + "right": "issue:45161", + "accept": false, + "reason": "Different symptoms and model paths: grouped_mm tracing failure versus GPT-OSS MoE tensor-parallel issue." + }, + { + "left": "issue:43828", + "right": "issue:45464", + "accept": false, + "reason": "One is a torch autocast dtype mismatch, the other is a streaming chat/completions failure on Qwen3.5; not the same bug." + }, + { + "left": "issue:44291", + "right": "issue:45070", + "accept": false, + "reason": "Both are v5 breakages, but init-empty-weights argument handling is distinct from the pydantic config-field regression." + }, + { + "left": "issue:43366", + "right": "issue:45464", + "accept": false, + "reason": "Unrelated model support/report: gpt-oss GGUF versus Qwen3.5 chat streaming failure." + }, + { + "left": "issue:45020", + "right": "issue:45800", + "accept": false, + "reason": "Broad compatibility issues, but one is remote_code regressions and the other is torch 2.4.1 / transformers 5.8.0 incompatibility; not the same concrete change." + }, + { + "left": "issue:43531", + "right": "issue:45070", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior is unrelated to the PretrainedConfig/pydantic regression." + }, + { + "left": "issue:43541", + "right": "issue:44589", + "accept": false, + "reason": "Different runtime failures: torch dynamo grouped_mm tracing versus missing Float8 storage object." + }, + { + "left": "issue:41093", + "right": "issue:43596", + "accept": false, + "reason": "Both are IndexError-style reports, but they concern different tensors, models, and setup conditions." + }, + { + "left": "issue:43054", + "right": "issue:43994", + "accept": false, + "reason": "Same model family, but one is a relative-quality complaint and the other is a loading/pipeline correctness report; root cause is not clearly the same." + }, + { + "left": "issue:45542", + "right": "issue:45800", + "accept": false, + "reason": "TensorBoard-without-TensorFlow backend error is unrelated to a torch/transformers version incompatibility." + }, + { + "left": "issue:43716", + "right": "issue:44918", + "accept": false, + "reason": "Both mention Qwen-like multimodal models, but one is an image-preprocessor dtype mismatch and the other is a TRL SFT embedding unpacking failure." + }, + { + "left": "issue:44918", + "right": "issue:45685", + "accept": false, + "reason": "Completely different failures: TRL embedding unpacking versus an MPS histogram kernel not implemented for int." + }, + { + "left": "issue:45632", + "right": "issue:45698", + "accept": false, + "reason": "Both involve custom code loading, but one is a cache-path collision and the other is loading the wrong custom module after save/load; not clearly the same bug." + }, + { + "left": "issue:43404", + "right": "issue:45005", + "accept": false, + "reason": "Related to tied weights, but the affected model classes and failure modes differ enough that they are not the same concrete issue." + }, + { + "left": "issue:43645", + "right": "issue:45399", + "accept": false, + "reason": "Notebook/custom-model initialization breakage is unrelated to flash-attn2 fallback selection logic." + }, + { + "left": "issue:40990", + "right": "issue:44960", + "accept": false, + "reason": "Different model and symptom: perplexity regression on gpt-oss-20b versus a GLM5 issue." + }, + { + "left": "issue:44530", + "right": "issue:45005", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash is unrelated to tied-weights handling." + }, + { + "left": "issue:44360", + "right": "issue:44512", + "accept": false, + "reason": "DSA indexer/ReLU discussion is unrelated to docs mentioning the removed run command." + }, + { + "left": "issue:42915", + "right": "issue:45399", + "accept": false, + "reason": "Qwen3-MoE FP8 failure and flash-attn2 fallback gating are different bugs." + }, + { + "left": "issue:44863", + "right": "issue:45005", + "accept": false, + "reason": "NemotronH checkpoint loading failure is unrelated to tied-weights regressions." + }, + { + "left": "issue:43873", + "right": "issue:45005", + "accept": false, + "reason": "Quantization/offloading behavior is a different issue from tied-weights handling." + }, + { + "left": "issue:39186", + "right": "issue:44918", + "accept": false, + "reason": "FSDP 2-D weight error and TRL embedding unpacking failure are unrelated." + }, + { + "left": "issue:44492", + "right": "issue:44512", + "accept": false, + "reason": "A typo in cache strategies is not the same as stale docs for the removed transformers run command." + }, + { + "left": "issue:44918", + "right": "issue:45507", + "accept": false, + "reason": "Different model families and failures: TRL embedding unpacking versus GraniteMoEHybrid invalid-method calls." + }, + { + "left": "issue:43975", + "right": "issue:44292", + "accept": false, + "reason": "DeepSeek detokenization bug and Qwen-3 NVFP4 runtime failure are unrelated." + }, + { + "left": "issue:43950", + "right": "issue:44534", + "accept": true, + "reason": "Same underlying regression: v5 corrupts non-persistent buffers on load; 44534 is essentially a reworded duplicate of 43950." + }, + { + "left": "issue:43644", + "right": "issue:43950", + "accept": true, + "reason": "Same bug: both report v5 filling/corrupting non-persistent buffers, with 43950 giving the clearest root-cause description." + }, + { + "left": "issue:44292", + "right": "issue:45507", + "accept": false, + "reason": "Different failures on different models; no shared concrete code-path is evident." + }, + { + "left": "issue:39692", + "right": "issue:43054", + "accept": false, + "reason": "Both concern SigLIP2, but one is a doc example/model-processor mismatch and quantization failure, while the other is a quality/regression report." + }, + { + "left": "issue:43828", + "right": "issue:44928", + "accept": false, + "reason": "Both involve dtype/precision concerns, but the runtime contexts and failure modes are different." + }, + { + "left": "issue:43866", + "right": "issue:44403", + "accept": false, + "reason": "Ovis2 checkpoint corruption and an unrelated loading-noise issue do not look like the same bug." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall. The anchor choice and canonical issue are well supported by the packet, and the only accepted pairings (43950/44534 and 43644/43950) look like the same non-persistent-buffer corruption regression. The negative pair verdicts are appropriately conservative and do not overstate evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 3, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:36331", + "issue:38175", + "issue:41553", + "issue:41628", + "issue:41762", + "issue:42491", + "issue:42915", + "issue:42947", + "issue:43054", + "issue:43404", + "issue:43493", + "issue:43638", + "issue:43643", + "issue:43742", + "issue:43782", + "issue:43792", + "issue:43824", + "issue:43827", + "issue:43828", + "issue:43856", + "issue:43872", + "issue:43881", + "issue:43883", + "issue:43950", + "issue:43975", + "issue:44220", + "issue:44368", + "issue:44387", + "issue:44403", + "issue:44451", + "issue:44488", + "issue:44509", + "issue:44589", + "issue:44661", + "issue:44910", + "issue:44960", + "issue:44991", + "issue:45020", + "issue:45081", + "issue:45237", + "issue:45356", + "issue:45399", + "issue:45520", + "issue:45685", + "issue:45701", + "issue:45800", + "issue:45823", + "issue:45859" + ], + "soft_pairs": [ + "issue:36331|issue:41762", + "issue:41628|issue:45800", + "issue:41553|issue:43883", + "issue:43872|issue:44589", + "issue:38175|issue:43054", + "issue:43054|issue:45701", + "issue:44910|issue:45859", + "issue:41553|issue:43881", + "issue:42947|issue:44387", + "issue:44403|issue:44451", + "issue:43643|issue:45020", + "issue:43975|issue:45520", + "issue:43054|issue:43493", + "issue:43975|issue:45237", + "issue:44403|issue:44488", + "issue:42947|issue:43856", + "issue:43881|issue:43883", + "issue:41762|issue:42915", + "issue:42491|issue:44368", + "issue:45685|issue:45823", + "issue:43975|issue:44451", + "issue:43824|issue:43872", + "issue:45020|issue:45823", + "issue:44991|issue:45399", + "issue:44661|issue:45020", + "issue:43828|issue:45823", + "issue:45020|issue:45356", + "issue:43782|issue:44960", + "issue:45081|issue:45399", + "issue:43827|issue:44509", + "issue:43404|issue:43950", + "issue:43792|issue:44220", + "issue:43742|issue:45399", + "issue:42915|issue:45020", + "issue:43638|issue:45823" + ], + "prepared_review_unit_hash": "cec015b3ff7aed38c924280535f59102455495021ec23cf70845ee5056c18a7b", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38227, + "estimated_input_tokens": 9557, + "estimated_eval_tokens": 19370 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most pairs are distinct bugs across different models, loaders, and runtime paths. The two clear soft duplicates are the SigLIP2 implementation-mismatch pair (43054/43493) and the docs pipeline-removal pair (43827/44509). The earlier SigLIP2 quality-only pairing (38175/43054) is not strong enough to accept from titles alone.", + "confidence": 0.82, + "canonical_issue_reason": "issue:45020 is the best open umbrella for the current remote_code-related regression cluster, but it should be treated as a tracker rather than a duplicate of the more specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the strongest global issue anchor because it is open, broad, and clearly positioned as a tracker for a recurring family of recent compatibility breakages.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36331", + "right": "issue:41762", + "accept": false, + "reason": "Different failures: trainer signature mismatch vs Gemma3 ZeRO-3 loading index error." + }, + { + "left": "issue:41628", + "right": "issue:45800", + "accept": false, + "reason": "Different subsystems: missing AutoImageProcessor import vs torch/transformers version incompatibility." + }, + { + "left": "issue:41553", + "right": "issue:43883", + "accept": false, + "reason": "Different model paths and symptoms: Voxtral tokenizer error handling vs Molmo tied-weights attribute error." + }, + { + "left": "issue:43872", + "right": "issue:44589", + "accept": false, + "reason": "Both are compatibility errors, but the concrete bugs differ: bitsandbytes kwarg mismatch vs float8 storage lookup." + }, + { + "left": "issue:38175", + "right": "issue:43054", + "accept": false, + "reason": "Same model family, but only related quality symptoms are shown; no shared concrete implementation bug is established." + }, + { + "left": "issue:43054", + "right": "issue:45701", + "accept": false, + "reason": "SigLIP2 embedding quality regression is unrelated to a generic tokenization-change report." + }, + { + "left": "issue:44910", + "right": "issue:45859", + "accept": false, + "reason": "Different code paths: Flash Attention packed-sequence crash vs CPU-offload rotary-embedding incompatibility." + }, + { + "left": "issue:41553", + "right": "issue:43881", + "accept": false, + "reason": "Different model/load failures: Voxtral tokenizer error vs glm-4v-9b loading failure." + }, + { + "left": "issue:42947", + "right": "issue:44387", + "accept": false, + "reason": "LoRA gradient-checkpointing effectiveness and int4 OOM are separate training/memory issues." + }, + { + "left": "issue:44403", + "right": "issue:44451", + "accept": false, + "reason": "Generic load-noise report vs a specific model failing to load." + }, + { + "left": "issue:43643", + "right": "issue:45020", + "accept": false, + "reason": "AutoConfig missing-fields under trust_remote_code is a specific bug; 45020 is a broader tracker, not the same defect." + }, + { + "left": "issue:43975", + "right": "issue:45520", + "accept": false, + "reason": "Tokenizer detokenization regression and Python 3.13 flash_attn import failure are unrelated." + }, + { + "left": "issue:43054", + "right": "issue:43493", + "accept": true, + "reason": "Both describe the same SigLIP2 implementation discrepancy; 43493 explicitly frames it as HF vs original JAX mismatch." + }, + { + "left": "issue:43975", + "right": "issue:45237", + "accept": false, + "reason": "DeepSeek tokenization regression and AMD GPU runtime support issue are unrelated." + }, + { + "left": "issue:44403", + "right": "issue:44488", + "accept": false, + "reason": "One is about noisy load logs, the other is a concrete model load failure." + }, + { + "left": "issue:42947", + "right": "issue:43856", + "accept": false, + "reason": "Gradient checkpointing/LoRA and Qwen3 MoE memory inefficiency are different training concerns." + }, + { + "left": "issue:43881", + "right": "issue:43883", + "accept": false, + "reason": "Different models and failures: glm-4v-9b load problem vs Molmo tied-weights attribute error." + }, + { + "left": "issue:41762", + "right": "issue:42915", + "accept": false, + "reason": "Both involve model loading, but Gemma3 ZeRO-3 and Qwen3 MoE FP8 are distinct bugs." + }, + { + "left": "issue:42491", + "right": "issue:44368", + "accept": false, + "reason": "Compatibility regression vs configuration warning; not the same concrete defect." + }, + { + "left": "issue:45685", + "right": "issue:45823", + "accept": false, + "reason": "MPS histogram-not-implemented error and Gemma4 device-mismatch forward error are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:44451", + "accept": false, + "reason": "DeepSeek tokenizer regression and ScandiBERT load failure are not the same bug." + }, + { + "left": "issue:43824", + "right": "issue:43872", + "accept": false, + "reason": "Qwen2.5-VL import failure is unrelated to the bitsandbytes Int8Params incompatibility." + }, + { + "left": "issue:45020", + "right": "issue:45823", + "accept": false, + "reason": "A broad remote_code tracker is too general to equate with a Gemma4 device-map forward mismatch." + }, + { + "left": "issue:44991", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer loading regression and flash-attn2 fallback gating are separate issues." + }, + { + "left": "issue:44661", + "right": "issue:45020", + "accept": false, + "reason": "Registry/add-new-model-like failure is a different bug class than the remote_code regression tracker." + }, + { + "left": "issue:43828", + "right": "issue:45823", + "accept": false, + "reason": "Autocast dtype mismatch in Phi-tiny-MoE is unrelated to Gemma4 device placement." + }, + { + "left": "issue:45020", + "right": "issue:45356", + "accept": false, + "reason": "Kimi-K2.5 tokenizer regression is a specific tokenizer bug, not the same as the broader remote_code tracker." + }, + { + "left": "issue:43782", + "right": "issue:44960", + "accept": false, + "reason": "Qwen3VL weight_only loading error and GLM5 are different model-specific failures." + }, + { + "left": "issue:45081", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer regex crash and flash-attn2 fallback gating are unrelated." + }, + { + "left": "issue:43827", + "right": "issue:44509", + "accept": true, + "reason": "Both are the same docs regression: pipeline tasks removed in v5 but still referenced in documentation." + }, + { + "left": "issue:43404", + "right": "issue:43950", + "accept": false, + "reason": "Mistral3 tied-weight bug and non-persistent buffer corruption are separate loading issues." + }, + { + "left": "issue:43792", + "right": "issue:44220", + "accept": false, + "reason": "Whisper runtime failure is too underspecified here to confirm the same concrete bug as the fbank feature issue." + }, + { + "left": "issue:43742", + "right": "issue:45399", + "accept": false, + "reason": "MobileLLM key error and flash-attn2 fallback blocking are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:45020", + "accept": false, + "reason": "Qwen3 MoE FP8 loading failure is a specific model/config bug, not the same as the broad remote_code tracker." + }, + { + "left": "issue:43638", + "right": "issue:45823", + "accept": false, + "reason": "ZeRO-3/BERT index error and Gemma4 device-mismatch are distinct failures." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The summary is mostly grounded, but it overstates one soft-edge duplicate claim: issue:38175 vs issue:43054 is not clearly supported by the packet. Both are in the SigLIP2 area, but the titles alone only show related model family/symptoms, not the same concrete bug or implementation mismatch. The 43054 vs 43493 pairing is more defensible, and the docs pair 43827 vs 44509 looks solid. Overall, the conservative verdicts are not consistent enough to accept as-is." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 4, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:24643", + "issue:30064", + "issue:36010", + "issue:36331", + "issue:38175", + "issue:41093", + "issue:41553", + "issue:42617", + "issue:42915", + "issue:43257", + "issue:43366", + "issue:43531", + "issue:43550", + "issue:43632", + "issue:43645", + "issue:43646", + "issue:43824", + "issue:43828", + "issue:43854", + "issue:43866", + "issue:43872", + "issue:43940", + "issue:43950", + "issue:43957", + "issue:43994", + "issue:44291", + "issue:44479", + "issue:44488", + "issue:44530", + "issue:44560", + "issue:44589", + "issue:44661", + "issue:44805", + "issue:44863", + "issue:44910", + "issue:44918", + "issue:44928", + "issue:44991", + "issue:45161", + "issue:45305", + "issue:45313", + "issue:45325", + "issue:45538", + "issue:45588", + "issue:45663", + "issue:45823", + "issue:45834" + ], + "soft_pairs": [ + "issue:43257|issue:43866", + "issue:43872|issue:44291", + "issue:43940|issue:45313", + "issue:45161|issue:45823", + "issue:41553|issue:44991", + "issue:42915|issue:43872", + "issue:43854|issue:44863", + "issue:44928|issue:45588", + "issue:36331|issue:41093", + "issue:43257|issue:45305", + "issue:41553|issue:44488", + "issue:43872|issue:45538", + "issue:44530|issue:44863", + "issue:41093|issue:44805", + "issue:44479|issue:44560", + "issue:24643|issue:30064", + "issue:43366|issue:43828", + "issue:45663|issue:45823", + "issue:43366|issue:43531", + "issue:43950|issue:45834", + "issue:43550|issue:44910", + "issue:43645|issue:43646", + "issue:43366|issue:44589", + "issue:43950|issue:45325", + "issue:44560|issue:44918", + "issue:42617|issue:43366", + "issue:43824|issue:43957", + "issue:42915|issue:43957", + "issue:36010|issue:42915", + "issue:44661|issue:45823", + "issue:38175|issue:43994", + "issue:43632|issue:45834" + ], + "prepared_review_unit_hash": "96c7d10b6ea0fe2395abedec970d727ab1bf8d6b26b8332735261b4f73b78f73", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36776, + "estimated_input_tokens": 9194, + "estimated_eval_tokens": 18644 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is mostly heterogeneous issue reports with only two strong duplicate-like pairs: the Transformers 5 custom model init regression, and the `_is_hf_initialized` / `init_empty_weights` regression. I rejected the rest because they describe different model families, different symptoms, or only share a broad subsystem.", + "confidence": 0.72, + "canonical_issue_reason": "No single canonical issue cleanly spans the full set because the items are mostly unrelated; if a representative anchor is needed, issue 44291 is the clearest root-cause report for the `_is_hf_initialized` regression.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44291 is the strongest standalone issue: it states the concrete failure mode, ties it to the v5 regression, and gives the clearest actionable root cause.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43257", + "right": "issue:43866", + "accept": false, + "reason": "Different problems: Qwen3 MoE weight conversion under accelerate+deepspeed vs a corrupted Ovis2 checkpoint." + }, + { + "left": "issue:43872", + "right": "issue:44291", + "accept": true, + "reason": "Same underlying regression around `_is_hf_initialized` / `init_empty_weights` in Transformers 5, with bitsandbytes showing the same incompatibility symptom." + }, + { + "left": "issue:43940", + "right": "issue:45313", + "accept": false, + "reason": "Both involve DeepSpeed ZeRO-3 weight loading, but the models and failure details differ enough that this is not clearly the same bug." + }, + { + "left": "issue:45161", + "right": "issue:45823", + "accept": false, + "reason": "Unrelated model-specific issues: GPT-OSS MoE tensor-parallel behavior vs Gemma4 PLE device placement." + }, + { + "left": "issue:41553", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer-loading failures, but they concern different models and different failure modes." + }, + { + "left": "issue:42915", + "right": "issue:43872", + "accept": false, + "reason": "Different subsystems and symptoms: FineGrainedFP8Config on Qwen3Moe vs the `_is_hf_initialized`/bitsandbytes regression." + }, + { + "left": "issue:43854", + "right": "issue:44863", + "accept": false, + "reason": "Different model-loading bugs for different architectures; only the broad theme of loading failure matches." + }, + { + "left": "issue:44928", + "right": "issue:45588", + "accept": false, + "reason": "Qwen3.5 RLHF/SDPA NaN behavior is unrelated to the flash_attention `s_aux=None` crash." + }, + { + "left": "issue:36331", + "right": "issue:41093", + "accept": false, + "reason": "Different error classes and different code paths: custom trainer API change vs mask/tensor shape mismatch." + }, + { + "left": "issue:43257", + "right": "issue:45305", + "accept": false, + "reason": "Both touch DeepSpeed training, but one is a Qwen MoE weight-conversion problem and the other is a gradient-averaging issue." + }, + { + "left": "issue:41553", + "right": "issue:44488", + "accept": false, + "reason": "Different tokenizer/loading issues for different models; not the same concrete bug." + }, + { + "left": "issue:43872", + "right": "issue:45538", + "accept": false, + "reason": "bitsandbytes/_is_hf_initialized incompatibility is unrelated to CLIPTokenizer's max length default." + }, + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash and NemotronH checkpoint loading are unrelated." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": false, + "reason": "Both are index/shape errors, but they are distinct reports with no evidence of the same root cause." + }, + { + "left": "issue:44479", + "right": "issue:44560", + "accept": false, + "reason": "Both are video-related Qwen issues, but one is a broad v5.3.0 regression and the other is a specific StopIteration in embedding loading." + }, + { + "left": "issue:24643", + "right": "issue:30064", + "accept": false, + "reason": "Training with DeepSpeed and void segmentation-map processing are unrelated." + }, + { + "left": "issue:43366", + "right": "issue:43828", + "accept": false, + "reason": "GPT-OSS architecture support and Phi-tiny-MoE autocast dtype mismatch are separate bugs." + }, + { + "left": "issue:45663", + "right": "issue:45823", + "accept": false, + "reason": "Gemma4 FSDP2 shared_kv_states issue is not the same as the device_map=\"auto\" PLE mismatch." + }, + { + "left": "issue:43366", + "right": "issue:43531", + "accept": false, + "reason": "GPT-OSS support and Qwen3-MoE sliding_window behavior are different issues." + }, + { + "left": "issue:43950", + "right": "issue:45834", + "accept": false, + "reason": "Non-persistent buffer corruption and Kosmos2.5 OCR index error are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:44910", + "accept": false, + "reason": "Both mention attention/compile problems, but Bamba SDPA compile failure is unrelated to Qwen3.5 3D position_ids and flash-attention memory access." + }, + { + "left": "issue:43645", + "right": "issue:43646", + "accept": true, + "reason": "Same Transformers 5 custom-model initialization regression; one report is just a more specific notebook variant." + }, + { + "left": "issue:43366", + "right": "issue:44589", + "accept": false, + "reason": "GPT-OSS support request and Float8 storage import failure are unrelated." + }, + { + "left": "issue:43950", + "right": "issue:45325", + "accept": false, + "reason": "Non-persistent buffer corruption is unrelated to the Qwen2.5-VL rope-index scaling bug." + }, + { + "left": "issue:44560", + "right": "issue:44918", + "accept": false, + "reason": "Both involve Qwen3.5 video/input handling, but the reported failures and likely code paths differ." + }, + { + "left": "issue:42617", + "right": "issue:43366", + "accept": false, + "reason": "3d_parallel execution failure and GPT-OSS architecture support are unrelated." + }, + { + "left": "issue:43824", + "right": "issue:43957", + "accept": false, + "reason": "Missing Qwen2_5_VL export and `torch.device(\"meta\")` loading regressions are different bugs." + }, + { + "left": "issue:42915", + "right": "issue:43957", + "accept": false, + "reason": "Qwen3Moe FP8 failure and meta-device loading regression do not share the same root cause." + }, + { + "left": "issue:36010", + "right": "issue:42915", + "accept": false, + "reason": "GenerationMixin import failure and Qwen3Moe FP8 failure are unrelated." + }, + { + "left": "issue:44661", + "right": "issue:45823", + "accept": false, + "reason": "Tokenizer-mapping registration bug and Gemma4 PLE device mismatch are different issues." + }, + { + "left": "issue:38175", + "right": "issue:43994", + "accept": false, + "reason": "Both mention SigLIP2, but one is zero probabilities and the other nonsensical AutoModel/pipeline outputs; not enough to treat as the same bug." + }, + { + "left": "issue:43632", + "right": "issue:45834", + "accept": false, + "reason": "_is_hf_initialized flag regression and Kosmos2.5 OCR index error are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded and conservative overall. The two accepted soft pairs are plausible duplicate/regression matches, and the rejected pairs are appropriately cautious given the packet. Summary does not overstate the evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 5, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:36010", + "issue:38175", + "issue:39404", + "issue:39692", + "issue:41720", + "issue:42491", + "issue:42915", + "issue:43257", + "issue:43317", + "issue:43596", + "issue:43638", + "issue:43644", + "issue:43645", + "issue:43708", + "issue:43720", + "issue:43854", + "issue:43856", + "issue:43873", + "issue:43931", + "issue:43994", + "issue:44075", + "issue:44155", + "issue:44387", + "issue:44451", + "issue:44488", + "issue:44534", + "issue:44560", + "issue:44661", + "issue:44756", + "issue:44805", + "issue:44877", + "issue:44910", + "issue:44960", + "issue:44977", + "issue:44991", + "issue:45030", + "issue:45276", + "issue:45292", + "issue:45313", + "issue:45335", + "issue:45357", + "issue:45362", + "issue:45406", + "issue:45488", + "issue:45584", + "issue:45588", + "issue:45812" + ], + "soft_pairs": [ + "issue:44155|issue:45584", + "issue:44877|issue:45030", + "issue:44560|issue:44805", + "issue:43638|issue:44661", + "issue:43317|issue:44756", + "issue:43596|issue:44661", + "issue:44661|issue:44805", + "issue:38175|issue:39692", + "issue:43644|issue:44534", + "issue:45313|issue:45357", + "issue:44960|issue:44977", + "issue:45488|issue:45812", + "issue:44451|issue:44488", + "issue:42491|issue:43720", + "issue:43854|issue:43931", + "issue:41720|issue:42915", + "issue:43708|issue:44075", + "issue:43257|issue:43854", + "issue:42491|issue:45276", + "issue:43856|issue:43873", + "issue:42915|issue:44756", + "issue:42491|issue:43931", + "issue:39692|issue:43994", + "issue:43856|issue:44387", + "issue:43644|issue:43645", + "issue:44910|issue:45588", + "issue:45362|issue:45406", + "issue:42491|issue:45292", + "issue:44488|issue:44991", + "issue:36010|issue:39404", + "issue:45276|issue:45335" + ], + "prepared_review_unit_hash": "87f2872331c988ef42b943937ff0372f82f64e6ea96041ad646db5fa51e88d3f", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36407, + "estimated_input_tokens": 9102, + "estimated_eval_tokens": 18460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most items are unrelated bug reports. The clearest duplicate remains the non-persistent-buffer regression pair (43644/44534). The `resize_token_embeddings` issues (45276/45335) are plausibly the same propagation bug, but `43644/43645` should be treated conservatively as non-duplicate without body evidence.", + "confidence": 0.81, + "canonical_issue_reason": "Issue 43644 is the best canonical issue: it is a precise, well-scoped regression report and is an exact duplicate of 44534.", + "canonical_pr_reason": null, + "best_issue_reason": "43644 is the strongest representative issue overall because it has the clearest duplicate evidence and the most specific regression framing.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44155", + "right": "issue:45584", + "accept": false, + "reason": "Different audio/model bugs: batched inference track leakage vs Whisper empty-transcription behavior." + }, + { + "left": "issue:44877", + "right": "issue:45030", + "accept": false, + "reason": "Both involve config loading, but the models and failure modes differ; no clear shared code path." + }, + { + "left": "issue:44560", + "right": "issue:44805", + "accept": false, + "reason": "Unrelated symptoms and models: video StopIteration in Qwen3-VL vs a mask shape IndexError." + }, + { + "left": "issue:43638", + "right": "issue:44661", + "accept": false, + "reason": "Deepspeed/BERT zero3 init bug is unrelated to tokenizer-mapping registration failure." + }, + { + "left": "issue:43317", + "right": "issue:44756", + "accept": false, + "reason": "Both touch offloading/memory, but one is a dequantized load failure and the other is a Strix Halo mmap OOM fix." + }, + { + "left": "issue:43596", + "right": "issue:44661", + "accept": false, + "reason": "Different root causes: zero3/BertModel IndexError vs `add-new-model-like` tokenizer-mapping logic." + }, + { + "left": "issue:44661", + "right": "issue:44805", + "accept": false, + "reason": "Tokenizer-mapping registration problem vs a tensor mask shape mismatch are not the same bug." + }, + { + "left": "issue:38175", + "right": "issue:39692", + "accept": false, + "reason": "Same model family, but one is bad output probabilities and the other is a docs/example mismatch plus quantization failure." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "Near-identical titles describing the same non-persistent-buffer junk regression; clearly the same underlying bug." + }, + { + "left": "issue:45313", + "right": "issue:45357", + "accept": false, + "reason": "Load-time ZeRO-3 weight issue and save_pretrained visual-encoder key regression are different code paths." + }, + { + "left": "issue:44960", + "right": "issue:44977", + "accept": false, + "reason": "Different models and different flash-attention failures; not the same concrete issue." + }, + { + "left": "issue:45488", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different tokenization layers and different model families." + }, + { + "left": "issue:44451", + "right": "issue:44488", + "accept": false, + "reason": "Both are model-loading failures, but the titles point to different repos and no shared concrete bug is evident." + }, + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "Both involve loading/compatibility, but one is Qwen3 MoE LoRA cross-version use and the other is BitNet packed-weight handling." + }, + { + "left": "issue:43854", + "right": "issue:43931", + "accept": false, + "reason": "Different model loading failures with different symptoms and likely different fixes." + }, + { + "left": "issue:41720", + "right": "issue:42915", + "accept": false, + "reason": "Qwen3 auto device mapping CUDA assert is not the same as Qwen3Moe FineGrainedFP8Config failure." + }, + { + "left": "issue:43708", + "right": "issue:44075", + "accept": false, + "reason": "Trainer checkpoint resume step calculation and SGD optimizer args being ignored are unrelated." + }, + { + "left": "issue:43257", + "right": "issue:43854", + "accept": false, + "reason": "Different models and different loading paths; no evidence of a common bug." + }, + { + "left": "issue:42491", + "right": "issue:45276", + "accept": false, + "reason": "LoRA compatibility across versions is unrelated to `resize_token_embeddings` not updating embeddings." + }, + { + "left": "issue:43856", + "right": "issue:43873", + "accept": false, + "reason": "Training memory inefficiency and quantized offload behavior are different problems." + }, + { + "left": "issue:42915", + "right": "issue:44756", + "accept": false, + "reason": "Qwen3Moe FP8 failure on one side and Strix Halo mmap OOM on the other are unrelated." + }, + { + "left": "issue:42491", + "right": "issue:43931", + "accept": false, + "reason": "Different models, different mechanisms, and no shared concrete code-path issue." + }, + { + "left": "issue:39692", + "right": "issue:43994", + "accept": false, + "reason": "Same SigLIP2 family, but one is docs/example misuse plus quantization failure while the other is an inference regression." + }, + { + "left": "issue:43856", + "right": "issue:44387", + "accept": false, + "reason": "Both mention memory, but the causes differ: MoE training efficiency vs increased CUDA reserved memory under int4 quantization." + }, + { + "left": "issue:43644", + "right": "issue:43645", + "accept": false, + "reason": "The titles both mention Transformers 5.0.0 regressions, but one is about non-persistent buffers and the other is about custom models in Jupyter notebooks; the packet does not show the same underlying bug." + }, + { + "left": "issue:44910", + "right": "issue:45588", + "accept": false, + "reason": "Both are flash-attention crashes, but one is a packed-sequence misinterpretation and the other is an `s_aux=None` attribute error." + }, + { + "left": "issue:45362", + "right": "issue:45406", + "accept": false, + "reason": "Different crash points: Qwen3.5 chat generation vs Gemma4Processor missing `_tokenizer`." + }, + { + "left": "issue:44488", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer/model loading failures in v5, but the specific repos and failure modes are too different to call duplicates." + }, + { + "left": "issue:36010", + "right": "issue:39404", + "accept": false, + "reason": "Importing `GenerationMixin` and Whisper `return_language` pipeline behavior are unrelated." + }, + { + "left": "issue:45276", + "right": "issue:45335", + "accept": true, + "reason": "Both report the same `resize_token_embeddings` propagation bug to model-specific embedding/output modules; one fix could plausibly cover both." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The cluster summary is mostly grounded, but one soft-edge acceptance is too strong: `issue:43644` vs `issue:43645` is not supported by the packet. `43644`/`44534` clearly match on the non-persistent-buffer junk regression, but `43645` is titled about custom models in Jupyter notebooks and does not obviously describe the same bug. That pair should stay conservative/rejected unless there is direct evidence in the issue bodies. The `resize_token_embeddings` pair looks plausibly related, but the `43644`/`43645` acceptance overstates the evidence." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 6, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:39404", + "issue:39692", + "issue:41720", + "issue:42947", + "issue:43257", + "issue:43335", + "issue:43381", + "issue:43404", + "issue:43425", + "issue:43454", + "issue:43645", + "issue:43646", + "issue:43708", + "issue:43828", + "issue:43854", + "issue:43873", + "issue:43957", + "issue:44155", + "issue:44164", + "issue:44291", + "issue:44292", + "issue:44387", + "issue:44402", + "issue:44451", + "issue:44479", + "issue:44488", + "issue:44589", + "issue:44683", + "issue:44756", + "issue:44863", + "issue:44877", + "issue:44912", + "issue:44928", + "issue:44960", + "issue:45216", + "issue:45276", + "issue:45292", + "issue:45335", + "issue:45588", + "issue:45663", + "issue:45684", + "issue:45698", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45823", + "issue:45859" + ], + "soft_pairs": [ + "issue:43404|issue:43454", + "issue:44164|issue:44291", + "issue:44402|issue:44488", + "issue:43425|issue:43957", + "issue:39692|issue:44960", + "issue:42947|issue:45704", + "issue:43335|issue:45663", + "issue:43873|issue:44387", + "issue:45663|issue:45759", + "issue:45663|issue:45859", + "issue:44155|issue:45823", + "issue:43854|issue:45216", + "issue:44387|issue:44756", + "issue:44291|issue:44756", + "issue:44292|issue:44912", + "issue:42947|issue:43381", + "issue:45684|issue:45698", + "issue:45276|issue:45292", + "issue:43381|issue:44928", + "issue:44164|issue:44756", + "issue:44479|issue:44877", + "issue:41720|issue:44155", + "issue:43646|issue:43708", + "issue:45588|issue:45663", + "issue:43854|issue:44960", + "issue:43957|issue:44756", + "issue:45292|issue:45335", + "issue:44402|issue:44451", + "issue:43828|issue:43957", + "issue:43645|issue:44164", + "issue:43257|issue:44863", + "issue:43425|issue:43828", + "issue:44589|issue:44756", + "issue:44683|issue:45823", + "issue:39404|issue:45701" + ], + "prepared_review_unit_hash": "0090eedf181c34281bdbe70ecdf4f93a763a26e3850d31d49547d1f25a16289b", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 35, + "serialized_chars": 37609, + "estimated_input_tokens": 9403, + "estimated_eval_tokens": 19062 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous model-loading / quantization / training bugs. Only the resize_token_embeddings issues look like true duplicates; most other soft edges are same subsystem at best, but not the same concrete bug.", + "confidence": 0.93, + "canonical_issue_reason": "issue:45292 is the broadest, most generic representative of the only clear duplicate set (`resize_token_embeddings` not updating output/decoder embeddings).", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45292 best represents the cluster\u2019s one real duplicate thread and is the most general formulation of that bug.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43404", + "right": "issue:43454", + "accept": false, + "reason": "Same symptom class (lm_head tying) but different model families and likely separate fixes." + }, + { + "left": "issue:44164", + "right": "issue:44291", + "accept": false, + "reason": "Both touch loading/saving paths, but one is extra_state handling and the other is init_empty_weights argument compatibility." + }, + { + "left": "issue:44402", + "right": "issue:44488", + "accept": false, + "reason": "Both report load failures for different BERT models, but the titles point to separate model/tokenizer mismatches rather than one bug." + }, + { + "left": "issue:43425", + "right": "issue:43957", + "accept": false, + "reason": "Unrelated failures: Torch version incompatibility vs meta-device loading regression." + }, + { + "left": "issue:39692", + "right": "issue:44960", + "accept": false, + "reason": "Documentation example errors versus a GLM5 loading issue; no shared underlying bug." + }, + { + "left": "issue:42947", + "right": "issue:45704", + "accept": false, + "reason": "Gradient checkpointing/LoRA behavior is unrelated to the T5 apex RMSNorm memory leak." + }, + { + "left": "issue:43335", + "right": "issue:45663", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation bug is unrelated to Gemma4 FSDP shared_kv_states handling." + }, + { + "left": "issue:43873", + "right": "issue:44387", + "accept": false, + "reason": "Both involve quantization/OOM, but one is offloading behavior and the other is reserved-memory growth under int4; different code paths." + }, + { + "left": "issue:45663", + "right": "issue:45759", + "accept": false, + "reason": "Gemma4 FSDP attention state rebuild is unrelated to Qwen multimodal from_config text_config unwrapping." + }, + { + "left": "issue:45663", + "right": "issue:45859", + "accept": false, + "reason": "Different models and failures: Gemma4 FSDP KeyError vs Qwen3_5MoE CPU-offload incompatibility." + }, + { + "left": "issue:44155", + "right": "issue:45823", + "accept": false, + "reason": "AudioFlamingo batched inference token leak is unrelated to Gemma4 device-mapping mismatch." + }, + { + "left": "issue:43854", + "right": "issue:45216", + "accept": false, + "reason": "Different model-loading regressions with no shared concrete failure mode." + }, + { + "left": "issue:44387", + "right": "issue:44756", + "accept": false, + "reason": "OOM-related, but one is int4 quantization memory reservation and the other is mmap on Strix Halo." + }, + { + "left": "issue:44291", + "right": "issue:44756", + "accept": false, + "reason": "init_empty_weights compatibility issue is unrelated to mmap/OOM behavior." + }, + { + "left": "issue:44292", + "right": "issue:44912", + "accept": false, + "reason": "Both are quantization-related, but they involve different quantization formats and different models." + }, + { + "left": "issue:42947", + "right": "issue:43381", + "accept": false, + "reason": "Both mention gradient checkpointing, but one is ineffective LoRA behavior and the other is an eval-mode restriction." + }, + { + "left": "issue:45684", + "right": "issue:45698", + "accept": false, + "reason": "Custom-model save/load issues, but permissions propagation and wrong-module loading are distinct bugs." + }, + { + "left": "issue:45276", + "right": "issue:45292", + "accept": true, + "reason": "Same underlying resize_token_embeddings bug: output embeddings are not updated, with 45276 being a model-specific variant." + }, + { + "left": "issue:43381", + "right": "issue:44928", + "accept": false, + "reason": "Gradient checkpointing eval-mode limitation is unrelated to Qwen3.5 RLHF gradient explosion." + }, + { + "left": "issue:44164", + "right": "issue:44756", + "accept": false, + "reason": "save/from_pretrained extra_state handling is unrelated to Strix Halo mmap OOM." + }, + { + "left": "issue:44479", + "right": "issue:44877", + "accept": false, + "reason": "Video-input regression across Qwen variants versus strict config loading for granite_speech; different problems." + }, + { + "left": "issue:41720", + "right": "issue:44155", + "accept": false, + "reason": "Auto device mapping cudaErrorAssert and cross-track token leak are unrelated." + }, + { + "left": "issue:43646", + "right": "issue:43708", + "accept": false, + "reason": "Custom model initialization breakage is unrelated to Trainer max_steps recomputation on resume." + }, + { + "left": "issue:45588", + "right": "issue:45663", + "accept": false, + "reason": "Flash-attention s_aux=None crash is unrelated to Gemma4 FSDP state rebuilding." + }, + { + "left": "issue:43854", + "right": "issue:44960", + "accept": false, + "reason": "Different GLM-related failures with no evidence they are the same bug." + }, + { + "left": "issue:43957", + "right": "issue:44756", + "accept": false, + "reason": "Meta-device loading regression is unrelated to the Strix Halo mmap workaround." + }, + { + "left": "issue:45292", + "right": "issue:45335", + "accept": true, + "reason": "Same resize_token_embeddings bug: output_embeddings and decoder.embed_tokens are both not refreshed." + }, + { + "left": "issue:44402", + "right": "issue:44451", + "accept": false, + "reason": "Both are model load failures, but the titles indicate distinct model-specific issues." + }, + { + "left": "issue:43828", + "right": "issue:43957", + "accept": false, + "reason": "Autocast dtype mismatch for Phi MoE is unrelated to meta-device loading failures." + }, + { + "left": "issue:43645", + "right": "issue:44164", + "accept": false, + "reason": "Notebook custom-model initialization and extra_state save/load handling are different failures." + }, + { + "left": "issue:43257", + "right": "issue:44863", + "accept": false, + "reason": "Qwen3 MOE deepspeed conversion issue is unrelated to NemotronH checkpoint loading." + }, + { + "left": "issue:43425", + "right": "issue:43828", + "accept": false, + "reason": "Torch version incompatibility is unrelated to Phi-tiny-MoE autocast dtype mismatch." + }, + { + "left": "issue:44589", + "right": "issue:44756", + "accept": false, + "reason": "Float8 storage lookup failure is unrelated to Strix Halo mmap OOM." + }, + { + "left": "issue:44683", + "right": "issue:45823", + "accept": false, + "reason": "Compiled flex_attention on torch>=2.9 is unrelated to Gemma4 PLE device mismatch." + }, + { + "left": "issue:39404", + "right": "issue:45701", + "accept": false, + "reason": "Whisper pipeline return_language regression is unrelated to tokenization changes in a different issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays appropriately conservative. The only clear duplicate cluster is the resize_token_embeddings set (45276/45292/45335), and the other soft-edge rejections are justified by the titles being model- or bug-specific rather than clearly mergeable into one PR." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 7, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:32090", + "issue:36032", + "issue:39404", + "issue:40990", + "issue:42915", + "issue:42947", + "issue:43299", + "issue:43421", + "issue:43425", + "issue:43452", + "issue:43582", + "issue:43596", + "issue:43611", + "issue:43632", + "issue:43644", + "issue:43645", + "issue:43716", + "issue:43856", + "issue:43901", + "issue:43906", + "issue:43931", + "issue:43994", + "issue:44079", + "issue:44164", + "issue:44206", + "issue:44220", + "issue:44265", + "issue:44479", + "issue:44488", + "issue:44509", + "issue:44530", + "issue:44589", + "issue:44661", + "issue:44671", + "issue:44683", + "issue:44737", + "issue:44756", + "issue:44805", + "issue:44863", + "issue:44898", + "issue:44912", + "issue:44918", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45759", + "issue:45823", + "issue:45859" + ], + "soft_pairs": [ + "issue:43906|issue:44530", + "issue:43425|issue:45823", + "issue:43611|issue:43645", + "issue:43425|issue:43716", + "issue:42915|issue:44863", + "issue:44530|issue:45084", + "issue:43901|issue:44509", + "issue:39404|issue:44912", + "issue:44265|issue:44898", + "issue:43299|issue:43931", + "issue:44479|issue:44805", + "issue:43425|issue:44918", + "issue:44589|issue:45084", + "issue:43856|issue:44756", + "issue:40990|issue:42947", + "issue:39404|issue:44671", + "issue:44164|issue:45759", + "issue:43994|issue:44079", + "issue:36032|issue:43452", + "issue:43596|issue:44683", + "issue:44671|issue:44991", + "issue:44991|issue:45081", + "issue:44737|issue:45859", + "issue:32090|issue:39404", + "issue:44164|issue:44479", + "issue:44488|issue:45081", + "issue:44661|issue:44683", + "issue:43582|issue:43632", + "issue:40990|issue:43421", + "issue:43611|issue:43644", + "issue:44206|issue:44220", + "issue:43716|issue:45084" + ], + "prepared_review_unit_hash": "1860ce3a02a5bc3e021b404266db3628f7bbf43aad00c44fcbdd0c99c57769fb", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37316, + "estimated_input_tokens": 9329, + "estimated_eval_tokens": 18914 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a grab bag of unrelated Transformers regressions and docs bugs. Only the audio feature-extraction pair looks plausibly duplicate; otherwise the soft edges are too broad or only share a subsystem. I\u2019d anchor the cluster on the generic v5 model-loading regression around `base_model_prefix`.", + "confidence": 0.46, + "canonical_issue_reason": "issue:43611 is the cleanest representative of the v5 loading regressions: generic, concrete, and closer to the common model-initialization/loading failure mode than the more specialized neighboring reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43611 is the best overall issue to keep as the canonical issue because it is broad enough to represent the cluster\u2019s main v5 model-loading breakage without being overly model-specific.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43906", + "right": "issue:44530", + "accept": false, + "reason": "Both involve Qwen-related behavior, but one is an isolated reproduction of a prior issue and the other is a PagedAttentionCache crash on Qwen3.5; different failures and code paths." + }, + { + "left": "issue:43425", + "right": "issue:45823", + "accept": false, + "reason": "Torch version incompatibility vs a Gemma4 PLE device-mismatch forward bug; unrelated symptoms and fixes." + }, + { + "left": "issue:43611", + "right": "issue:43645", + "accept": false, + "reason": "Both are v5 regressions, but `base_model_prefix` loading failure and custom-model-in-Jupyter initialization are distinct underlying problems." + }, + { + "left": "issue:43425", + "right": "issue:43716", + "accept": false, + "reason": "Torch 2.10 incompatibility is unrelated to the Mistral-3 image preprocessor dtype mismatch." + }, + { + "left": "issue:42915", + "right": "issue:44863", + "accept": false, + "reason": "Different model families and failures: FineGrainedFP8Config on Qwen3Moe vs checkpoint-loading failure in NemotronH." + }, + { + "left": "issue:44530", + "right": "issue:45084", + "accept": false, + "reason": "Qwen3.5 cache crash vs template compilation error; no shared underlying bug." + }, + { + "left": "issue:43901", + "right": "issue:44509", + "accept": false, + "reason": "Both are docs-related, but one is about `return_all_scores` and the other is about removed pipeline task docs; not the same issue." + }, + { + "left": "issue:39404", + "right": "issue:44912", + "accept": false, + "reason": "Whisper pipeline language return regression and gpt-oss quantization/perplexity are unrelated." + }, + { + "left": "issue:44265", + "right": "issue:44898", + "accept": false, + "reason": "`torch.export`/`torch_compilable_check` failure and Perceiver resolution/interpolation failure are different bugs." + }, + { + "left": "issue:43299", + "right": "issue:43931", + "accept": false, + "reason": "Both concern Qwen3-VL loading, but one is a dev-version breakage for MoE models and the other is a weight-shape mismatch on a specific checkpoint; not enough to treat as the same bug." + }, + { + "left": "issue:44479", + "right": "issue:44805", + "accept": false, + "reason": "Video-input regression across Qwen VL variants vs a generic mask-shape IndexError; different failure modes." + }, + { + "left": "issue:43425", + "right": "issue:44918", + "accept": false, + "reason": "Torch compatibility and TRL SFT input-embedding unpacking are unrelated." + }, + { + "left": "issue:44589", + "right": "issue:45084", + "accept": false, + "reason": "Float8 storage lookup failure is not the same as the template-node compilation error." + }, + { + "left": "issue:43856", + "right": "issue:44756", + "accept": false, + "reason": "Both mention memory/OOM, but one is Qwen3 MoE training memory inefficiency and the other is mmap behavior on Strix Halo; different root causes." + }, + { + "left": "issue:40990", + "right": "issue:42947", + "accept": false, + "reason": "Perplexity regression on gpt-oss-20b and gradient-checkpointing ineffectiveness with LoRA are separate issues." + }, + { + "left": "issue:39404", + "right": "issue:44671", + "accept": false, + "reason": "Whisper pipeline language return regression vs CamemBERT masked-LM prediction regression; unrelated model paths." + }, + { + "left": "issue:44164", + "right": "issue:45759", + "accept": false, + "reason": "`extra_state` serialization/loading and `text_config` unwrapping for composite Qwen configs are distinct loader issues." + }, + { + "left": "issue:43994", + "right": "issue:44079", + "accept": false, + "reason": "SigLIP2 bad outputs with AutoModel/pipeline and ModelOutput key assignment when a value was previously None are different bugs." + }, + { + "left": "issue:36032", + "right": "issue:43452", + "accept": false, + "reason": "T5 tokenizer special-token method conflict and gguf-related from_pretrained breakage are unrelated loading failures." + }, + { + "left": "issue:43596", + "right": "issue:44683", + "accept": false, + "reason": "DeepSpeed ZeRO3/BertModel index error and compiled flex_attention failure on newer torch versions do not share the same root cause." + }, + { + "left": "issue:44671", + "right": "issue:44991", + "accept": false, + "reason": "CamemBERT incorrect predictions and tokenizer loading failure for EMBEDDIA/est-roberta are different issues." + }, + { + "left": "issue:44991", + "right": "issue:45081", + "accept": false, + "reason": "Both are tokenizer-loading regressions, but one is model-specific tokenizer loading for EMBEDDIA/est-roberta and the other is Mistral regex patching; not the same bug." + }, + { + "left": "issue:44737", + "right": "issue:45859", + "accept": false, + "reason": "Both involve device placement/offload concerns, but XLNet CPU `arange` work and Qwen3_5Moe rotary embedding CPU-offload incompatibility are separate code paths." + }, + { + "left": "issue:32090", + "right": "issue:39404", + "accept": false, + "reason": "Trainer GPU broadcast TypeError and Whisper `return_language` regression are unrelated." + }, + { + "left": "issue:44164", + "right": "issue:44479", + "accept": false, + "reason": "Serialization/loading of extra state is unrelated to Qwen VL video-input regression." + }, + { + "left": "issue:44488", + "right": "issue:45081", + "accept": false, + "reason": "Loading `cjvt/sleng-bert` and the Mistral regex patch crash are different tokenizer issues." + }, + { + "left": "issue:44661", + "right": "issue:44683", + "accept": false, + "reason": "`add-new-model-like`/mapping-name handling and compiled flex_attention are unrelated." + }, + { + "left": "issue:43582", + "right": "issue:43632", + "accept": false, + "reason": "Apple Silicon allocator warmup TypeError and `_is_hf_initialized` flag breakage are different regressions." + }, + { + "left": "issue:40990", + "right": "issue:43421", + "accept": false, + "reason": "Perplexity regression and runtime special-token post-processor update behavior are not the same bug." + }, + { + "left": "issue:43611", + "right": "issue:43644", + "accept": false, + "reason": "Both are v5 model-loading regressions, but `base_model_prefix` handling and non-persistent buffer initialization are separate underlying defects." + }, + { + "left": "issue:44206", + "right": "issue:44220", + "accept": true, + "reason": "These both point at the same audio feature-extraction path: a regression in `LasrFeatureExtractor`/`_torch_extract_fbank_features` around the same internal function call and crash mode." + }, + { + "left": "issue:43716", + "right": "issue:45084", + "accept": false, + "reason": "Mistral-3 image preprocessor dtype mismatch and template-node compilation errors are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet and stays conservative on the soft edges. The only accepted pair, issue:44206 and issue:44220, looks plausibly like the same audio feature-extraction bug and mergeable into one fix. Anchoring on issue:43611 as the canonical v5 loading regression is also reasonable given the surrounding reports are mostly model-specific or unrelated." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 8, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:34567", + "issue:36032", + "issue:40444", + "issue:42915", + "issue:43262", + "issue:43296", + "issue:43317", + "issue:43322", + "issue:43377", + "issue:43404", + "issue:43452", + "issue:43482", + "issue:43531", + "issue:43540", + "issue:43541", + "issue:43632", + "issue:43643", + "issue:43644", + "issue:43645", + "issue:43646", + "issue:43742", + "issue:43819", + "issue:43883", + "issue:43994", + "issue:44220", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44451", + "issue:44466", + "issue:44488", + "issue:44530", + "issue:44534", + "issue:44610", + "issue:44683", + "issue:44756", + "issue:44811", + "issue:44849", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45127", + "issue:45216", + "issue:45325", + "issue:45357", + "issue:45507", + "issue:45584" + ], + "soft_pairs": [ + "issue:43883|issue:44488", + "issue:44488|issue:44987", + "issue:34567|issue:40444", + "issue:43643|issue:44530", + "issue:44811|issue:45584", + "issue:43644|issue:43646", + "issue:44451|issue:44987", + "issue:43541|issue:45084", + "issue:36032|issue:42915", + "issue:43452|issue:43482", + "issue:42915|issue:44530", + "issue:43317|issue:43482", + "issue:44849|issue:45325", + "issue:43883|issue:44991", + "issue:43404|issue:44466", + "issue:36032|issue:43645", + "issue:45084|issue:45507", + "issue:43262|issue:43377", + "issue:43646|issue:44534", + "issue:44987|issue:44991", + "issue:43994|issue:44610", + "issue:44403|issue:45081", + "issue:43322|issue:43540", + "issue:43377|issue:43819", + "issue:43645|issue:44534", + "issue:42915|issue:45084", + "issue:44451|issue:45081", + "issue:45216|issue:45357", + "issue:44987|issue:45081", + "issue:44220|issue:44683", + "issue:44265|issue:44610", + "issue:43404|issue:45127", + "issue:43296|issue:43531", + "issue:43632|issue:44756", + "issue:42915|issue:44291", + "issue:36032|issue:43742" + ], + "prepared_review_unit_hash": "453af618e3a203eec6efd592b0bf03a1d19af376cdcd30578caa3e95087322f8", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 36, + "serialized_chars": 38525, + "estimated_input_tokens": 9632, + "estimated_eval_tokens": 19520 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most pairs are unrelated model-specific regressions, not true duplicates. The only clear duplicate-like thread is the Qwen3.5 `save_pretrained` serialization bug, where one issue is the broader report and the other is a narrower symptom.", + "confidence": 0.39, + "canonical_issue_reason": "issue:45216 is the broadest description of the Qwen3.5 `save_pretrained` regression; issue:45357 looks like a narrower symptom about incorrect visual encoder keys.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45216 is the best umbrella issue for the one clearly duplicate-like thread in this set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43883", + "right": "issue:44488", + "accept": false, + "reason": "Different bugs: Molmo tied-weights metadata vs loading a specific checkpoint." + }, + { + "left": "issue:44488", + "right": "issue:44987", + "accept": false, + "reason": "Both are load failures, but for different models and different failure modes." + }, + { + "left": "issue:34567", + "right": "issue:40444", + "accept": false, + "reason": "Trainer token counting bug vs Qwen2.5-VL iterable dataset/multi-image training bug." + }, + { + "left": "issue:43643", + "right": "issue:44530", + "accept": false, + "reason": "Missing fields from `AutoConfig` with `trust_remote_code` is unrelated to Qwen3.5 cache group-type errors." + }, + { + "left": "issue:44811", + "right": "issue:45584", + "accept": false, + "reason": "Both involve Whisper, but one is `batch_decode` special-token handling and the other is empty-transcription generation after alignment." + }, + { + "left": "issue:43644", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 regressions, but one is non-persistent buffers and the other is custom model initialization." + }, + { + "left": "issue:44451", + "right": "issue:44987", + "accept": false, + "reason": "Different checkpoints with no shared code-path evidence." + }, + { + "left": "issue:43541", + "right": "issue:45084", + "accept": false, + "reason": "Torch dynamo grouped_mm tracing error is unrelated to template-node compilation." + }, + { + "left": "issue:36032", + "right": "issue:42915", + "accept": false, + "reason": "Tokenizer method conflict vs Qwen3Moe FP8 config loading failure." + }, + { + "left": "issue:43452", + "right": "issue:43482", + "accept": false, + "reason": "Both mention GGUF loading, but one is a general `gguf_file` API regression and the other is a Qwen2.5-GGUF-specific v5 failure." + }, + { + "left": "issue:42915", + "right": "issue:44530", + "accept": false, + "reason": "Qwen3Moe FP8 config issue and Qwen3.5 linear-attention cache crash are different bugs." + }, + { + "left": "issue:43317", + "right": "issue:43482", + "accept": false, + "reason": "Device-map/offload dequantized-model loading issue is not the same as Qwen2.5-GGUF v5 loading failure." + }, + { + "left": "issue:44849", + "right": "issue:45325", + "accept": false, + "reason": "Different Qwen family regressions: hidden-states output vs RoPE position-id scaling." + }, + { + "left": "issue:43883", + "right": "issue:44991", + "accept": false, + "reason": "Different problems: missing tied-weight keys vs tokenizer loading for another model." + }, + { + "left": "issue:43404", + "right": "issue:44466", + "accept": false, + "reason": "Both concern tied weights, but one is a model-specific untied `lm_head` and the other is save serialization differing by device." + }, + { + "left": "issue:36032", + "right": "issue:43645", + "accept": false, + "reason": "Tokenizer attribute conflict is unrelated to custom model initialization in notebooks." + }, + { + "left": "issue:45084", + "right": "issue:45507", + "accept": false, + "reason": "Template-node compilation error is unrelated to GraniteMoEHybrid invalid-method calls." + }, + { + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Audio chat-template sampling-rate bug vs MIMI padding-mask batch inconsistency are different issues." + }, + { + "left": "issue:43646", + "right": "issue:44534", + "accept": false, + "reason": "Custom model initialization and non-persistent buffer junk are separate regressions." + }, + { + "left": "issue:44987", + "right": "issue:44991", + "accept": false, + "reason": "Different model-loading failures with no shared concrete bug." + }, + { + "left": "issue:43994", + "right": "issue:44610", + "accept": false, + "reason": "SigLIP2 nonsense outputs vs OmDet-Turbo image-size mismatch are unrelated." + }, + { + "left": "issue:44403", + "right": "issue:45081", + "accept": false, + "reason": "Generic loading noise is not the same as the Mistral regex patch crash." + }, + { + "left": "issue:43322", + "right": "issue:43540", + "accept": false, + "reason": "Llava Next loading segfault and Qwen3OmniMoe video-input ValueError are different code paths." + }, + { + "left": "issue:43377", + "right": "issue:43819", + "accept": false, + "reason": "Both are model-consistency bugs, but for different models and different mechanisms." + }, + { + "left": "issue:43645", + "right": "issue:44534", + "accept": false, + "reason": "Jupyter custom-model initialization issue is distinct from buffer serialization junk." + }, + { + "left": "issue:42915", + "right": "issue:45084", + "accept": false, + "reason": "Qwen3Moe FP8 config and template compilation are unrelated." + }, + { + "left": "issue:44451", + "right": "issue:45081", + "accept": false, + "reason": "Different models and different failure modes." + }, + { + "left": "issue:45216", + "right": "issue:45357", + "accept": true, + "reason": "Same underlying Qwen3.5 `save_pretrained` serialization regression; the second issue is a narrower symptom about incorrect visual encoder keys." + }, + { + "left": "issue:44987", + "right": "issue:45081", + "accept": false, + "reason": "Loading a specific model vs Mistral tokenizer regex patch crash are unrelated." + }, + { + "left": "issue:44220", + "right": "issue:44683", + "accept": false, + "reason": "Audio feature extraction bug vs flex_attention compile failure are unrelated." + }, + { + "left": "issue:44265", + "right": "issue:44610", + "accept": false, + "reason": "`torch.export`/`torch_compilable_check` failure is unrelated to the OmDet-Turbo processor size mismatch." + }, + { + "left": "issue:43404", + "right": "issue:45127", + "accept": false, + "reason": "Both mention tied embeddings, but one is untied weights in a model class and the other is LoRA merge collapse." + }, + { + "left": "issue:43296", + "right": "issue:43531", + "accept": false, + "reason": "PaddleOCR-VL load failure in vLLM is unrelated to Qwen3-MoE sliding window behavior." + }, + { + "left": "issue:43632", + "right": "issue:44756", + "accept": false, + "reason": "`_is_hf_initialized` flag regression is unrelated to Strix Halo mmap OOM behavior." + }, + { + "left": "issue:42915", + "right": "issue:44291", + "accept": false, + "reason": "Different regressions: Qwen3Moe FP8 config vs generic `init_empty_weights` argument error." + }, + { + "left": "issue:36032", + "right": "issue:43742", + "accept": false, + "reason": "Tokenizer method-name collision is unrelated to a key error loading MobileLLM-125M." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the one accepted soft-edge pair is supported by the issue titles: 45216 is the broader Qwen3.5 `save_pretrained` regression, while 45357 is a narrower symptom about visual encoder keys. The rejected pairs are phrased conservatively and do not overclaim duplication based on the available evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 9, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:33666", + "issue:36032", + "issue:40444", + "issue:40990", + "issue:41553", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43296", + "issue:43317", + "issue:43377", + "issue:43421", + "issue:43425", + "issue:43482", + "issue:43525", + "issue:43531", + "issue:43582", + "issue:43611", + "issue:43632", + "issue:43646", + "issue:43792", + "issue:43818", + "issue:43856", + "issue:44062", + "issue:44079", + "issue:44291", + "issue:44403", + "issue:44462", + "issue:44479", + "issue:44534", + "issue:44568", + "issue:44589", + "issue:44683", + "issue:44811", + "issue:44829", + "issue:44877", + "issue:44987", + "issue:44991", + "issue:45020", + "issue:45081", + "issue:45325", + "issue:45357", + "issue:45479", + "issue:45491", + "issue:45538", + "issue:45542", + "issue:45663", + "issue:45812" + ], + "soft_pairs": [ + "issue:44062|issue:44987", + "issue:42915|issue:44589", + "issue:43611|issue:44534", + "issue:43531|issue:44877", + "issue:41553|issue:43582", + "issue:43317|issue:43856", + "issue:43425|issue:44403", + "issue:44479|issue:45325", + "issue:40990|issue:44811", + "issue:42175|issue:45542", + "issue:43818|issue:45357", + "issue:44462|issue:45812", + "issue:36032|issue:41720", + "issue:44079|issue:45663", + "issue:43421|issue:44568", + "issue:43792|issue:44589", + "issue:43632|issue:44534", + "issue:44403|issue:44991", + "issue:44683|issue:45020", + "issue:43482|issue:45538", + "issue:44829|issue:45479", + "issue:33666|issue:40444", + "issue:44291|issue:44589", + "issue:43482|issue:43525", + "issue:40444|issue:45325", + "issue:43377|issue:45491", + "issue:41720|issue:45081", + "issue:42175|issue:43296", + "issue:43531|issue:44291", + "issue:43611|issue:43646" + ], + "prepared_review_unit_hash": "ce5136055948fe3fb3a5892ba2a69985498d5ed612770c1672f6f579c2fdda02", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36777, + "estimated_input_tokens": 9195, + "estimated_eval_tokens": 18646 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous. The closest-looking pairs are related regressions, but from titles alone they are not confirmed to be the same concrete bug or change, so I would not merge any soft edges.", + "confidence": 0.61, + "canonical_issue_reason": "No single canonical issue stands out: the cluster is mixed, and the nearest matches are only loosely related from titles alone.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45479 is the best standalone representative because it describes a concrete, broadly applicable classification failure more precisely than most other items.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44062", + "right": "issue:44987", + "accept": false, + "reason": "Both are tokenizer/loading regressions, but the titles describe different failure modes and do not establish the same bug." + }, + { + "left": "issue:42915", + "right": "issue:44589", + "accept": false, + "reason": "Both mention FP8-related breakage, but one is a config failure and the other is a storage-object TypeError; not the same issue." + }, + { + "left": "issue:43611", + "right": "issue:44534", + "accept": false, + "reason": "Both are Transformers v5 regressions, but the titles point to different concrete failures and likely different code paths." + }, + { + "left": "issue:43531", + "right": "issue:44877", + "accept": false, + "reason": "Different models and different symptoms; the titles do not show a shared underlying bug." + }, + { + "left": "issue:41553", + "right": "issue:43582", + "accept": false, + "reason": "Unrelated subsystems: tokenizer loading/error messaging versus Apple Silicon allocator warmup." + }, + { + "left": "issue:43317", + "right": "issue:43856", + "accept": false, + "reason": "Both involve memory/device behavior, but one is device-map/offload loading and the other is MoE training memory usage." + }, + { + "left": "issue:43425", + "right": "issue:44403", + "accept": false, + "reason": "Version compatibility and loading-noise complaints are not the same concrete bug." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": false, + "reason": "These are related Qwen VL/video-position regressions, but the titles alone do not confirm the same root cause or code path." + }, + { + "left": "issue:40990", + "right": "issue:44811", + "accept": false, + "reason": "Different models and different behaviors: perplexity evaluation versus Whisper decoding options." + }, + { + "left": "issue:42175", + "right": "issue:45542", + "accept": false, + "reason": "Both touch backend packaging, but the reported breakages are not the same concrete failure." + }, + { + "left": "issue:43818", + "right": "issue:45357", + "accept": false, + "reason": "Different model families and distinct bugs: missing temporal attention versus incorrect saved key mapping." + }, + { + "left": "issue:44462", + "right": "issue:45812", + "accept": false, + "reason": "Both involve AutoTokenizer, but one is repository tokenizer precedence and the other is a Granite token-ID regression." + }, + { + "left": "issue:36032", + "right": "issue:41720", + "accept": false, + "reason": "Tokenizer loading conflict versus CUDA assertion during device mapping; unrelated." + }, + { + "left": "issue:44079", + "right": "issue:45663", + "accept": false, + "reason": "A ModelOutput key assignment bug and a Gemma-4 FSDP2 KeyError are different issues." + }, + { + "left": "issue:43421", + "right": "issue:44568", + "accept": false, + "reason": "Both mention special tokens, but one is runtime post-processor updates and the other is BOS/EOS emission for a specific tokenizer." + }, + { + "left": "issue:43792", + "right": "issue:44589", + "accept": false, + "reason": "A Whisper runtime failure and a missing Float8 storage object are unrelated." + }, + { + "left": "issue:43632", + "right": "issue:44534", + "accept": false, + "reason": "Distinct v5 regressions affecting different initialization/buffer mechanisms." + }, + { + "left": "issue:44403", + "right": "issue:44991", + "accept": false, + "reason": "A noise/logging issue and a tokenizer loading failure are not the same bug." + }, + { + "left": "issue:44683", + "right": "issue:45020", + "accept": false, + "reason": "Compiled flex attention on torch>=2.9 and remote_code loading are separate problems." + }, + { + "left": "issue:43482", + "right": "issue:45538", + "accept": false, + "reason": "GGUF loading under v5 and CLIPTokenizer model_max_length are unrelated." + }, + { + "left": "issue:44829", + "right": "issue:45479", + "accept": false, + "reason": "These are related degenerate-training reports, but the titles do not confirm the same failure path or code change." + }, + { + "left": "issue:33666", + "right": "issue:40444", + "accept": false, + "reason": "Both are Qwen VL training issues, but they affect different workflows and are not the same bug." + }, + { + "left": "issue:44291", + "right": "issue:44589", + "accept": false, + "reason": "Different TypeErrors caused by different v5 changes; no shared code-path evidence." + }, + { + "left": "issue:43482", + "right": "issue:43525", + "accept": false, + "reason": "Different model-loading failures with different symptoms and likely different causes." + }, + { + "left": "issue:40444", + "right": "issue:45325", + "accept": false, + "reason": "Both are Qwen2.5-VL-related, but one is multi-image training and the other is temporal position scaling." + }, + { + "left": "issue:43377", + "right": "issue:45491", + "accept": false, + "reason": "Both involve batching/masking behavior, but they are separate models and separate bugs." + }, + { + "left": "issue:41720", + "right": "issue:45081", + "accept": false, + "reason": "CUDA assertion during device mapping and a Mistral tokenizer regex crash are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:43296", + "accept": false, + "reason": "A packaging/backend install issue and a vLLM model-loading failure are different problems." + }, + { + "left": "issue:43531", + "right": "issue:44291", + "accept": false, + "reason": "Sliding-window Qwen3-MoE behavior and an init_empty_weights argument regression are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:43646", + "accept": false, + "reason": "Both are broad Transformers v5 regressions, but the titles describe different concrete failures and not one mergeable fix." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The cluster-level summary is directionally plausible, but a few claims go beyond what the packet supports. In particular, the two accepted soft pairs are asserted too strongly from titles alone: 44479/45325 is described as the same regression with 45325 as the root cause, and 44829/45479 is called the same zero-loss bug. The packet only gives issue titles, so these should be framed more cautiously as related, possibly overlapping regressions rather than confirmed duplicates or shared code paths." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 10, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:35707", + "issue:39692", + "issue:40444", + "issue:40990", + "issue:41720", + "issue:42915", + "issue:43262", + "issue:43296", + "issue:43317", + "issue:43377", + "issue:43454", + "issue:43482", + "issue:43526", + "issue:43550", + "issue:43572", + "issue:43606", + "issue:43632", + "issue:43644", + "issue:43701", + "issue:43746", + "issue:43792", + "issue:43819", + "issue:43873", + "issue:43881", + "issue:43883", + "issue:43957", + "issue:43994", + "issue:44079", + "issue:44186", + "issue:44265", + "issue:44291", + "issue:44387", + "issue:44466", + "issue:44479", + "issue:44488", + "issue:44534", + "issue:44610", + "issue:44617", + "issue:44877", + "issue:44898", + "issue:45072", + "issue:45325", + "issue:45381", + "issue:45584", + "issue:45759", + "issue:45800", + "issue:45823", + "issue:45834" + ], + "soft_pairs": [ + "issue:44265|issue:45072", + "issue:43632|issue:45072", + "issue:43454|issue:44466", + "issue:43994|issue:44898", + "issue:44079|issue:44534", + "issue:45325|issue:45834", + "issue:43792|issue:43957", + "issue:42915|issue:43482", + "issue:44079|issue:45072", + "issue:40444|issue:45381", + "issue:45325|issue:45381", + "issue:43644|issue:44079", + "issue:44387|issue:44617", + "issue:43881|issue:44488", + "issue:43296|issue:45800", + "issue:44479|issue:45823", + "issue:35707|issue:40444", + "issue:44291|issue:44877", + "issue:43317|issue:43873", + "issue:43606|issue:44265", + "issue:43701|issue:45325", + "issue:44186|issue:44265", + "issue:43606|issue:43819", + "issue:43262|issue:43746", + "issue:41720|issue:45823", + "issue:43526|issue:43883", + "issue:43454|issue:43550", + "issue:43377|issue:44610", + "issue:43377|issue:44898", + "issue:43454|issue:45072", + "issue:40990|issue:45584", + "issue:43572|issue:45759", + "issue:39692|issue:40444" + ], + "prepared_review_unit_hash": "5b3dcbb3bb817f96037c2dd072d87cd443171bfe63d5680526bc0e933dfe659a", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37582, + "estimated_input_tokens": 9396, + "estimated_eval_tokens": 19048 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab-bag of distinct model-loading, multimodal, quantization, and device-mapping regressions. None of the soft edges look like true duplicates, and there are no PRs to consolidate.", + "confidence": 0.62, + "canonical_issue_reason": "issue:43644 is the broadest systemic regression in the set, centered on a Transformers v5 state/buffer behavior break that is representative of the wider v5-related failures here.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43644 is the best umbrella issue candidate because it describes a concrete, cross-cutting v5 regression with broad downstream impact and is likely to be the most reusable reference point.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44265", + "right": "issue:45072", + "accept": false, + "reason": "Different code paths and symptoms: torch.export support for torch_compilable_check versus bfloat16 dtype mismatches during inference." + }, + { + "left": "issue:43632", + "right": "issue:45072", + "accept": false, + "reason": "One is an initialization flag regression; the other is a dtype-mismatch issue under bfloat16 inference. Not the same bug." + }, + { + "left": "issue:43454", + "right": "issue:44466", + "accept": false, + "reason": "Both involve tied weights, but one is a missing weight-tying bug and the other is inconsistent serialization behavior; distinct underlying failures." + }, + { + "left": "issue:43994", + "right": "issue:44898", + "accept": false, + "reason": "Different models and failure modes: SigLIP2 pipeline/AutoModel mismatch versus Perceiver non-default-resolution classification." + }, + { + "left": "issue:44079", + "right": "issue:44534", + "accept": false, + "reason": "ModelOutput key assignment logic is unrelated to non-persistent buffer initialization/serialization." + }, + { + "left": "issue:45325", + "right": "issue:45834", + "accept": false, + "reason": "Both are multimodal indexing bugs, but one is a Qwen2.5-VL temporal RoPE scaling issue and the other is a Kosmos2.5 long-OCR index error." + }, + { + "left": "issue:43792", + "right": "issue:43957", + "accept": false, + "reason": "Different failure classes: Whisper model runtime failure versus meta-device loading breakage across some models." + }, + { + "left": "issue:42915", + "right": "issue:43482", + "accept": false, + "reason": "Different models and mechanisms: FineGrainedFP8Config failure versus GGUF loading failure in v5." + }, + { + "left": "issue:44079", + "right": "issue:45072", + "accept": false, + "reason": "ModelOutput key handling is not the same as dtype mismatches in bf16 inference." + }, + { + "left": "issue:40444", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen2.5-VL related, but one is training with multiple images per prompt and the other is video vision_position_ids; not the same concrete bug." + }, + { + "left": "issue:45325", + "right": "issue:45381", + "accept": false, + "reason": "Closely related family, but still different symptoms: still-image temporal position scaling versus video vision_position_id computation." + }, + { + "left": "issue:43644", + "right": "issue:44079", + "accept": false, + "reason": "Buffer initialization junk and ModelOutput key assignment are unrelated regressions." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Both can end in OOM, but one is a Transformers int4 memory regression and the other is a model-specific video OOM." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "Separate model-loading failures for different models; no shared concrete code-path shown." + }, + { + "left": "issue:43296", + "right": "issue:45800", + "accept": false, + "reason": "Different compatibility problems: vLLM/PaddleOCR-VL loading versus a torch 2.4.1 / Transformers 5.8.0 incompatibility." + }, + { + "left": "issue:44479", + "right": "issue:45823", + "accept": false, + "reason": "Different multimodal/device issues: Qwen VL video regression versus Gemma4 PLE device mismatch in forward." + }, + { + "left": "issue:35707", + "right": "issue:40444", + "accept": false, + "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to Qwen2.5-VL multi-image finetuning failures." + }, + { + "left": "issue:44291", + "right": "issue:44877", + "accept": false, + "reason": "One is an unexpected _is_hf_initialized argument during init_empty_weights; the other is strict config validation blocking granite_speech config loading." + }, + { + "left": "issue:43317", + "right": "issue:43873", + "accept": false, + "reason": "Both mention offloading/quantization, but one is dequantized model loading with auto device maps and the other is a broader offloading complaint; not clearly the same bug." + }, + { + "left": "issue:43606", + "right": "issue:44265", + "accept": false, + "reason": "CPU offload device mismatch and torch.export support for torch_compilable_check are unrelated." + }, + { + "left": "issue:43701", + "right": "issue:45325", + "accept": false, + "reason": "Checkpoint resume key mismatch is unrelated to Qwen2.5-VL RoPE position-id scaling." + }, + { + "left": "issue:44186", + "right": "issue:44265", + "accept": false, + "reason": "Tokenizer crashes on NER padding/truncation are unrelated to torch.export failures on compilable checks." + }, + { + "left": "issue:43606", + "right": "issue:43819", + "accept": false, + "reason": "CPU offload device mismatch is unrelated to DAC.from_latents versus forward-pass mismatch." + }, + { + "left": "issue:43262", + "right": "issue:43746", + "accept": false, + "reason": "Audio chat-template sampling-rate defaults and GraniteSpeech PEFT checkpoint loading are different issues." + }, + { + "left": "issue:41720", + "right": "issue:45823", + "accept": false, + "reason": "Both are device_map=auto issues, but they affect different models and failures; not the same code-path bug." + }, + { + "left": "issue:43526", + "right": "issue:43883", + "accept": false, + "reason": "BEiT label reduction and missing all_tied_weights_keys are unrelated." + }, + { + "left": "issue:43454", + "right": "issue:43550", + "accept": false, + "reason": "Tied-weight loading/serialization versus torch.compile SDPA failure are different bugs." + }, + { + "left": "issue:43377", + "right": "issue:44610", + "accept": false, + "reason": "Missing padding mask support in MIMI encoder is unrelated to OmDet-Turbo processor/model resolution mismatch." + }, + { + "left": "issue:43377", + "right": "issue:44898", + "accept": false, + "reason": "Different models and different failures: MIMI batching discrepancy versus Perceiver resolution handling." + }, + { + "left": "issue:43454", + "right": "issue:45072", + "accept": false, + "reason": "A tied-weights bug is not the same as bf16 dtype mismatch during inference." + }, + { + "left": "issue:40990", + "right": "issue:45584", + "accept": false, + "reason": "High perplexity on WikiText-2 is unrelated to Whisper empty-transcription generation after align_special_tokens." + }, + { + "left": "issue:43572", + "right": "issue:45759", + "accept": false, + "reason": "StableLm pad_token_idx regression and Qwen multimodal config unwrapping are different failures." + }, + { + "left": "issue:39692", + "right": "issue:40444", + "accept": false, + "reason": "SigLIP2 documentation errors are unrelated to Qwen2.5-VL finetuning with multiple images per prompt." + } + ] + }, + "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": 16, + "review_unit_index": 11, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:34634", + "issue:35707", + "issue:36331", + "issue:41720", + "issue:42617", + "issue:42915", + "issue:43296", + "issue:43316", + "issue:43317", + "issue:43404", + "issue:43450", + "issue:43454", + "issue:43526", + "issue:43550", + "issue:43577", + "issue:43606", + "issue:43708", + "issue:43720", + "issue:43749", + "issue:43792", + "issue:43873", + "issue:43881", + "issue:44008", + "issue:44186", + "issue:44387", + "issue:44423", + "issue:44466", + "issue:44479", + "issue:44534", + "issue:44610", + "issue:44683", + "issue:44756", + "issue:44877", + "issue:44898", + "issue:44913", + "issue:44918", + "issue:44977", + "issue:44991", + "issue:45072", + "issue:45081", + "issue:45357", + "issue:45406", + "issue:45632", + "issue:45684", + "issue:45701", + "issue:45704", + "issue:45812", + "issue:45823" + ], + "soft_pairs": [ + "issue:43881|issue:44991", + "issue:44466|issue:45357", + "issue:43792|issue:44683", + "issue:43296|issue:44877", + "issue:43454|issue:43720", + "issue:43873|issue:45701", + "issue:41720|issue:44977", + "issue:44008|issue:45823", + "issue:45632|issue:45684", + "issue:43404|issue:45357", + "issue:34634|issue:35707", + "issue:44423|issue:45406", + "issue:44186|issue:45072", + "issue:43577|issue:43708", + "issue:43873|issue:44756", + "issue:35707|issue:41720", + "issue:44877|issue:44991", + "issue:42617|issue:42915", + "issue:43749|issue:44534", + "issue:43450|issue:45701", + "issue:43550|issue:43720", + "issue:36331|issue:42915", + "issue:44610|issue:44898", + "issue:36331|issue:42617", + "issue:44186|issue:44610", + "issue:44387|issue:45704", + "issue:43526|issue:43881", + "issue:44479|issue:44918", + "issue:42915|issue:43317", + "issue:43316|issue:44913", + "issue:43550|issue:43606", + "issue:45081|issue:45812" + ], + "prepared_review_unit_hash": "787f72bbe02d5f4636fa9028e04a7835af16275847bf28f5c847fd25762d9b80", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37407, + "estimated_input_tokens": 9352, + "estimated_eval_tokens": 18960 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous: many items are model-specific loading/runtime regressions, and the soft pairs share broad themes but not the same concrete bug. I\u2019d use the broad offloading/quantization issue as the cluster representative.", + "confidence": 0.61, + "canonical_issue_reason": "issue:43873 is the most general and central representative of the cluster\u2019s recurring offloading/quantization/device-loading failures, and it has broader linkage than the more model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the best single issue anchor because it is broad, still open, and closest to the repeated loading/offload failure pattern seen across the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "Different models and different failure surfaces: GLM-4V load failure vs tokenizer regression for EMBEDDIA/est-roberta." + }, + { + "left": "issue:44466", + "right": "issue:45357", + "accept": false, + "reason": "Both involve serialization, but one is tied lm_head handling on device-dependent save, the other is incorrect visual encoder keys in Qwen3.5." + }, + { + "left": "issue:43792", + "right": "issue:44683", + "accept": false, + "reason": "Whisper loading failure and compiled flex_attention on torch>=2.9 are unrelated code paths." + }, + { + "left": "issue:43296", + "right": "issue:44877", + "accept": false, + "reason": "PaddleOCR-VL/vLLM load compatibility and strict config loading for granite_speech are distinct issues." + }, + { + "left": "issue:43454", + "right": "issue:43720", + "accept": false, + "reason": "Mistral3 weight tying and BitNet packed-weight unpacking are different loader/model-conversion bugs." + }, + { + "left": "issue:43873", + "right": "issue:45701", + "accept": false, + "reason": "Quantization/offloading behavior and tokenization differences are not the same underlying bug." + }, + { + "left": "issue:41720", + "right": "issue:44977", + "accept": false, + "reason": "Qwen3 auto device-mapping cuda assert is a different failure from Qwen3.5 flash-attention generation regressions." + }, + { + "left": "issue:44008", + "right": "issue:45823", + "accept": false, + "reason": "Gemma3n forward variable collision and Gemma4 PLE device mismatch are separate model-specific bugs." + }, + { + "left": "issue:45632", + "right": "issue:45684", + "accept": false, + "reason": "Cache-path collisions and read-only permission propagation are unrelated save/load issues." + }, + { + "left": "issue:43404", + "right": "issue:45357", + "accept": false, + "reason": "Mistral3 lm_head tying and Qwen3.5 visual encoder key serialization affect different models and mechanisms." + }, + { + "left": "issue:34634", + "right": "issue:35707", + "accept": false, + "reason": "BarkProcessor voice_preset failure and progressive generation with inputs_embeds/past_key_values are unrelated." + }, + { + "left": "issue:44423", + "right": "issue:45406", + "accept": false, + "reason": "Continuous-batching multimodal crash and missing _tokenizer on Gemma4Processor are different serve-time failures." + }, + { + "left": "issue:44186", + "right": "issue:45072", + "accept": false, + "reason": "LayoutLMv2Tokenizer NER/padding crash and bfloat16 dtype mismatch are not the same bug." + }, + { + "left": "issue:43577", + "right": "issue:43708", + "accept": false, + "reason": "Blip2 dtype propagation and Trainer resume/max_steps calculation are different subsystems." + }, + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "Both mention memory/offload concerns, but one is quantization/offloading behavior and the other is a Strix Halo mmap OOM fix." + }, + { + "left": "issue:35707", + "right": "issue:41720", + "accept": false, + "reason": "Input-embedding generation handling is unrelated to Qwen3 device-map cuda asserts." + }, + { + "left": "issue:44877", + "right": "issue:44991", + "accept": false, + "reason": "Strict config loading and tokenizer loading regression are separate problems." + }, + { + "left": "issue:42617", + "right": "issue:42915", + "accept": false, + "reason": "3d_parallel.py startup/runtime failure and Qwen3Moe FineGrainedFP8Config failure are different concrete bugs." + }, + { + "left": "issue:43749", + "right": "issue:44534", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and non-persistent buffer junk are both load/init related but not the same code-path problem." + }, + { + "left": "issue:43450", + "right": "issue:45701", + "accept": false, + "reason": "Video processor batching shape bug and tokenizer version-dependent tokenization are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:43720", + "accept": false, + "reason": "Bamba SDPA torch.compile failure and BitNet packed-weight unpacking are different model/runtime bugs." + }, + { + "left": "issue:36331", + "right": "issue:42915", + "accept": false, + "reason": "CustomTrainer compute_loss signature mismatch and Qwen3Moe FP8 load failure are unrelated." + }, + { + "left": "issue:44610", + "right": "issue:44898", + "accept": false, + "reason": "Both are vision-processing/resolution issues, but they hit different models and different mismatch logic." + }, + { + "left": "issue:36331", + "right": "issue:42617", + "accept": false, + "reason": "Trainer API mismatch and 3d_parallel.py failure are different concrete issues." + }, + { + "left": "issue:44186", + "right": "issue:44610", + "accept": false, + "reason": "Tokenizer crash on NER/batched padding and processor output-size mismatch are different preprocessing bugs." + }, + { + "left": "issue:44387", + "right": "issue:45704", + "accept": false, + "reason": "Int4 quantization OOM and apex FusedRMSNorm memory leak are both memory-related but not the same defect." + }, + { + "left": "issue:43526", + "right": "issue:43881", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug and GLM-4V loading failure are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:44918", + "accept": false, + "reason": "Qwen VL video regression and TRL SFT embedding unpack failure are distinct multimodal/training bugs." + }, + { + "left": "issue:42915", + "right": "issue:43317", + "accept": false, + "reason": "Qwen3Moe FP8 config failure and dequantized model offload loading failure are different failure modes." + }, + { + "left": "issue:43316", + "right": "issue:44913", + "accept": false, + "reason": "Gemma3TextConfig API discrepancy and GPTNeoX rotary_pct reload regression are unrelated config issues." + }, + { + "left": "issue:43550", + "right": "issue:43606", + "accept": false, + "reason": "Bamba SDPA compile failure and Bark CPU-offload device mismatch are different runtime paths." + } + ] + }, + "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": 16, + "review_unit_index": 12, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:30064", + "issue:34567", + "issue:34634", + "issue:35707", + "issue:36010", + "issue:40990", + "issue:41720", + "issue:42617", + "issue:42915", + "issue:43295", + "issue:43322", + "issue:43388", + "issue:43421", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43572", + "issue:43606", + "issue:43653", + "issue:43720", + "issue:43746", + "issue:43782", + "issue:43873", + "issue:43927", + "issue:44186", + "issue:44361", + "issue:44368", + "issue:44451", + "issue:44492", + "issue:44509", + "issue:44514", + "issue:44545", + "issue:44671", + "issue:44743", + "issue:44779", + "issue:44898", + "issue:44991", + "issue:45059", + "issue:45072", + "issue:45161", + "issue:45357", + "issue:45406", + "issue:45464", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45812", + "issue:45859" + ], + "soft_pairs": [ + "issue:41720|issue:43322", + "issue:43295|issue:45701", + "issue:40990|issue:45701", + "issue:41720|issue:45161", + "issue:43531|issue:43572", + "issue:43421|issue:45812", + "issue:44671|issue:44743", + "issue:41720|issue:42617", + "issue:36010|issue:42617", + "issue:43540|issue:43782", + "issue:44186|issue:44361", + "issue:45406|issue:45464", + "issue:30064|issue:34634", + "issue:43421|issue:45701", + "issue:44492|issue:44509", + "issue:44779|issue:45812", + "issue:42915|issue:43322", + "issue:43606|issue:43720", + "issue:43388|issue:45059", + "issue:43550|issue:43653", + "issue:44514|issue:44545", + "issue:43531|issue:45759", + "issue:43606|issue:43653", + "issue:43421|issue:43927", + "issue:34567|issue:35707", + "issue:44368|issue:45357", + "issue:44451|issue:44991", + "issue:43873|issue:45704", + "issue:45759|issue:45859", + "issue:44898|issue:45072", + "issue:41720|issue:43746" + ], + "prepared_review_unit_hash": "523ca39b6cbd7183aecce575cefe8df0ed2b7f1e51f2a89b419e84e09c0e330e", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36984, + "estimated_input_tokens": 9246, + "estimated_eval_tokens": 18748 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a set of unrelated transformer/model bugs. Only one pair is a clear duplicate: the two Qwen2_5_VLProcessor.apply_chat_template padding=False crashes. All other soft pairs differ in model, subsystem, or failure mode.", + "confidence": 0.95, + "canonical_issue_reason": "issue:44545 is the best canonical representative for the only true duplicate pair; it has the same exact bug report as 44514 and slightly more discussion/inbound references.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44545 is the strongest issue to keep as the canonical cluster item because it is the clearest, most engaged report for the duplicated Qwen2_5_VLProcessor.apply_chat_template crash.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41720", + "right": "issue:43322", + "accept": false, + "reason": "Different failures in different models: Qwen3 device-mapping cuda assert vs Llava Next load segfault." + }, + { + "left": "issue:43295", + "right": "issue:45701", + "accept": false, + "reason": "One is a processor/tokenizer API regression; the other is a broad tokenization-version complaint." + }, + { + "left": "issue:40990", + "right": "issue:45701", + "accept": false, + "reason": "Perplexity regression on GPT-OSS is not the same bug as version-dependent tokenization changes." + }, + { + "left": "issue:41720", + "right": "issue:45161", + "accept": false, + "reason": "Different Qwen/MoE issues with unrelated symptoms: cudaErrorAssert vs tensor-parallel not working." + }, + { + "left": "issue:43531", + "right": "issue:43572", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE and missing pad_token_idx in StableLmConfig are unrelated." + }, + { + "left": "issue:43421", + "right": "issue:45812", + "accept": false, + "reason": "Both involve tokenizers, but one is runtime special-token post-processing; the other is Granite token IDs regression." + }, + { + "left": "issue:44671", + "right": "issue:44743", + "accept": false, + "reason": "CamemBERT masked-LM outputs and Qwen3.5 recurrent cache-state resets are different code paths." + }, + { + "left": "issue:41720", + "right": "issue:42617", + "accept": false, + "reason": "Qwen3 auto device mapping crash and a 3d_parallel.py execution failure are unrelated." + }, + { + "left": "issue:36010", + "right": "issue:42617", + "accept": false, + "reason": "ImportError for GenerationMixin is not the same as a parallel script runtime failure." + }, + { + "left": "issue:43540", + "right": "issue:43782", + "accept": false, + "reason": "Video-input processing error and Qwen3VL weight_only load error are different bugs." + }, + { + "left": "issue:44186", + "right": "issue:44361", + "accept": false, + "reason": "Different tokenizer failures on different models and tasks; not the same underlying defect." + }, + { + "left": "issue:45406", + "right": "issue:45464", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer in serve is unrelated to Qwen3.5 streaming inference failure." + }, + { + "left": "issue:30064", + "right": "issue:34634", + "accept": false, + "reason": "Image segmentation map processing and BarkProcessor voice_preset are unrelated." + }, + { + "left": "issue:43421", + "right": "issue:45701", + "accept": false, + "reason": "Runtime post-processor update bug is different from a general tokenization regression report." + }, + { + "left": "issue:44492", + "right": "issue:44509", + "accept": false, + "reason": "A typo in cache strategies and outdated docs for removed pipeline tasks are unrelated." + }, + { + "left": "issue:44779", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different models and likely different fixes." + }, + { + "left": "issue:42915", + "right": "issue:43322", + "accept": false, + "reason": "Qwen3Moe FP8 failure and Llava Next loading segfault are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:43720", + "accept": false, + "reason": "CPU offload device mismatch and BitNet packed-weight unloading are different accelerate/loading bugs." + }, + { + "left": "issue:43388", + "right": "issue:45059", + "accept": false, + "reason": "Metrics label truncation and SAM3 text/bbox weirdness are different problems." + }, + { + "left": "issue:43550", + "right": "issue:43653", + "accept": false, + "reason": "torch.compile/SDPA failure and tokenizer special-token registration are unrelated." + }, + { + "left": "issue:44514", + "right": "issue:44545", + "accept": true, + "reason": "Same exact Qwen2_5_VLProcessor.apply_chat_template padding=False crash; these are clear duplicates." + }, + { + "left": "issue:43531", + "right": "issue:45759", + "accept": false, + "reason": "Qwen3-MoE sliding_window bug and composite-config unwrap behavior are different model/config issues." + }, + { + "left": "issue:43606", + "right": "issue:43653", + "accept": false, + "reason": "CPU offload mismatch and tokenizer mask-token registration are unrelated CI bugs." + }, + { + "left": "issue:43421", + "right": "issue:43927", + "accept": false, + "reason": "Special-token post-processing and DiaConfig token ID persistence are different tokenizer/config defects." + }, + { + "left": "issue:34567", + "right": "issue:35707", + "accept": false, + "reason": "TrainerState token counter not updating and progressive generation with inputs_embeds/past_key_values are unrelated." + }, + { + "left": "issue:44368", + "right": "issue:45357", + "accept": false, + "reason": "Warning about tie_word_embeddings is not the same as incorrect visual encoder keys in save_pretrained." + }, + { + "left": "issue:44451", + "right": "issue:44991", + "accept": false, + "reason": "General model-loading failure for one checkpoint and tokenizer loading regression for another checkpoint are not the same bug." + }, + { + "left": "issue:43873", + "right": "issue:45704", + "accept": false, + "reason": "Quantization offloading behavior and apex FusedRMSNorm memory leak are unrelated." + }, + { + "left": "issue:45759", + "right": "issue:45859", + "accept": false, + "reason": "Composite Qwen config unwrapping and CPU-offload incompatibility in rotary embedding are different issues." + }, + { + "left": "issue:44898", + "right": "issue:45072", + "accept": false, + "reason": "Perceiver non-default resolution failure and dtype mismatches in other models are unrelated." + }, + { + "left": "issue:41720", + "right": "issue:43746", + "accept": false, + "reason": "Qwen3 device-mapping assert and GraniteSpeech PEFT checkpoint loading are different code paths." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The canonical choice for issue:44545 is well grounded: it is effectively the same report as 44514 and has slightly more inbound attention. The soft-edge rejections are conservative and supported by the titles/body metadata; no pair is overstated as a duplicate or same-bug merge. The summary stays within the packet evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 13, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:30064", + "issue:33666", + "issue:34634", + "issue:35707", + "issue:36010", + "issue:39692", + "issue:41720", + "issue:42886", + "issue:42915", + "issue:43335", + "issue:43381", + "issue:43421", + "issue:43525", + "issue:43531", + "issue:43540", + "issue:43596", + "issue:43618", + "issue:43653", + "issue:43697", + "issue:43756", + "issue:43782", + "issue:43844", + "issue:43873", + "issue:43994", + "issue:44008", + "issue:44062", + "issue:44186", + "issue:44220", + "issue:44451", + "issue:44589", + "issue:44610", + "issue:44625", + "issue:44661", + "issue:44671", + "issue:44843", + "issue:44918", + "issue:44936", + "issue:44987", + "issue:45042", + "issue:45081", + "issue:45183", + "issue:45356", + "issue:45381", + "issue:45479", + "issue:45588", + "issue:45704", + "issue:45759", + "issue:45859" + ], + "soft_pairs": [ + "issue:44062|issue:44451", + "issue:44987|issue:45042", + "issue:43531|issue:44589", + "issue:33666|issue:35707", + "issue:43421|issue:43653", + "issue:45588|issue:45759", + "issue:42886|issue:44843", + "issue:43381|issue:43844", + "issue:44918|issue:44936", + "issue:39692|issue:44610", + "issue:43421|issue:44186", + "issue:43525|issue:43531", + "issue:44625|issue:45479", + "issue:45704|issue:45859", + "issue:45081|issue:45356", + "issue:43782|issue:45381", + "issue:34634|issue:41720", + "issue:43994|issue:44671", + "issue:43381|issue:45704", + "issue:39692|issue:43873", + "issue:42915|issue:45759", + "issue:43596|issue:44220", + "issue:43540|issue:44008", + "issue:43618|issue:43697", + "issue:44220|issue:44661", + "issue:30064|issue:36010", + "issue:45183|issue:45479", + "issue:43335|issue:43756" + ], + "prepared_review_unit_hash": "e960158f57910d53a8df1181808acf90b31086782bf4da8ba055c45ce06e878f", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36240, + "estimated_input_tokens": 9060, + "estimated_eval_tokens": 18376 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is highly heterogeneous: tokenizer, multimodal, training, quantization, and docs regressions with no convincing duplicate pairs. The strongest anchor is the Mistral tokenizer regex/offline-loading regression, but it is not a clear duplicate hub for the rest.", + "confidence": 0.91, + "canonical_issue_reason": "Best anchor in this mixed set: explicit tokenizer code-path regression with a concrete crash surface and multiple references.", + "canonical_pr_reason": null, + "best_issue_reason": "Most actionable standalone bug here: precise repro, narrow fix surface, and a clear runtime failure in a core tokenizer-loading path.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44062", + "right": "issue:44451", + "accept": false, + "reason": "Tokenizer constructor error vs model-loading failure; no shared underlying bug." + }, + { + "left": "issue:44987", + "right": "issue:45042", + "accept": false, + "reason": "Different subsystems: specific model load issue vs PIL/torchvision backend dependency regression." + }, + { + "left": "issue:43531", + "right": "issue:44589", + "accept": false, + "reason": "Unrelated Qwen sliding-window config bug vs PyTorch storage-type import/runtime error." + }, + { + "left": "issue:33666", + "right": "issue:35707", + "accept": false, + "reason": "Multi-GPU training problem and generation with inputs_embeds/past_key_values are different failure modes." + }, + { + "left": "issue:43421", + "right": "issue:43653", + "accept": false, + "reason": "Both involve tokenizers, but one is runtime post-processor updates and the other is mask-token registration/empty decode behavior." + }, + { + "left": "issue:45588", + "right": "issue:45759", + "accept": false, + "reason": "Different code paths and symptoms: flash-attention sink handling vs composite Qwen config unwrapping." + }, + { + "left": "issue:42886", + "right": "issue:44843", + "accept": false, + "reason": "Both touch offline loading, but 44843 is a specific mistral-regex model_info call; 42886 is a broader cache-load failure." + }, + { + "left": "issue:43381", + "right": "issue:43844", + "accept": false, + "reason": "Checkpointing-in-eval issue vs ZeRO-3 gradient anomaly; unrelated bugs." + }, + { + "left": "issue:44918", + "right": "issue:44936", + "accept": false, + "reason": "Different trainer issues: SFT embedding unpacking vs train/evaluate lifecycle failure." + }, + { + "left": "issue:39692", + "right": "issue:44610", + "accept": false, + "reason": "Both are processor/model mismatches, but for different models and different root causes." + }, + { + "left": "issue:43421", + "right": "issue:44186", + "accept": false, + "reason": "Different tokenizer bugs: post-processor refresh vs NER padding/truncation crash." + }, + { + "left": "issue:43525", + "right": "issue:43531", + "accept": false, + "reason": "Missing config attribute vs sliding_window behavior; not the same defect." + }, + { + "left": "issue:44625", + "right": "issue:45479", + "accept": false, + "reason": "Shared num_labels wording, but one is config propagation and the other is classification-loss behavior." + }, + { + "left": "issue:45704", + "right": "issue:45859", + "accept": false, + "reason": "Different models and mechanisms: apex RMSNorm leak vs CPU-offload incompatibility in rotary embedding." + }, + { + "left": "issue:45081", + "right": "issue:45356", + "accept": false, + "reason": "Same broad regex-patching area, but one is a tokenizer-load crash and the other is a Kimi codec regression; not clearly one bug." + }, + { + "left": "issue:43782", + "right": "issue:45381", + "accept": false, + "reason": "Different Qwen VL issues: weight-only loading error vs video vision_position_ids mismatch." + }, + { + "left": "issue:34634", + "right": "issue:41720", + "accept": false, + "reason": "Bark voice preset bug and Qwen3 auto-device-map CUDA assert are unrelated." + }, + { + "left": "issue:43994", + "right": "issue:44671", + "accept": false, + "reason": "Different model regressions with no shared code path." + }, + { + "left": "issue:43381", + "right": "issue:45704", + "accept": false, + "reason": "Eval-mode checkpointing bug vs T5 RMSNorm memory leak; unrelated." + }, + { + "left": "issue:39692", + "right": "issue:43873", + "accept": false, + "reason": "Docs/example mismatch vs quantization offloading behavior; not the same underlying issue." + }, + { + "left": "issue:42915", + "right": "issue:45759", + "accept": false, + "reason": "Different Qwen issues: FP8 config failure vs composite-config unwrapping." + }, + { + "left": "issue:43596", + "right": "issue:44220", + "accept": false, + "reason": "Deepspeed/Bert init error vs audio fbank feature extraction; unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44008", + "accept": false, + "reason": "Both are multimodal, but they fail in different modalities and from different root causes." + }, + { + "left": "issue:43618", + "right": "issue:43697", + "accept": false, + "reason": "Different model regressions; only both are output-related at a very high level." + }, + { + "left": "issue:44220", + "right": "issue:44661", + "accept": false, + "reason": "Audio feature extraction bug vs tokenizer-mapping extension bug." + }, + { + "left": "issue:30064", + "right": "issue:36010", + "accept": false, + "reason": "Image segmentation processing bug vs GenerationMixin import regression; unrelated." + }, + { + "left": "issue:45183", + "right": "issue:45479", + "accept": false, + "reason": "Input-validation XOR error vs single-label classification loss bug; different code paths." + }, + { + "left": "issue:43335", + "right": "issue:43756", + "accept": false, + "reason": "Sparse-layer config bug vs RoPE-layer dropping mismatch; not the same issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 14, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:30064", + "issue:36331", + "issue:39692", + "issue:42915", + "issue:43295", + "issue:43322", + "issue:43421", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43577", + "issue:43643", + "issue:43653", + "issue:43716", + "issue:43720", + "issue:43746", + "issue:43782", + "issue:43818", + "issue:43819", + "issue:43825", + "issue:43873", + "issue:43994", + "issue:44008", + "issue:44062", + "issue:44077", + "issue:44186", + "issue:44448", + "issue:44479", + "issue:44485", + "issue:44488", + "issue:44509", + "issue:44561", + "issue:44625", + "issue:44671", + "issue:44811", + "issue:44821", + "issue:44843", + "issue:44877", + "issue:44898", + "issue:44977", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45381", + "issue:45584", + "issue:45685", + "issue:45834" + ], + "soft_pairs": [ + "issue:43720|issue:43819", + "issue:44077|issue:44625", + "issue:43653|issue:43720", + "issue:43716|issue:44008", + "issue:42915|issue:43746", + "issue:43295|issue:45834", + "issue:43746|issue:45381", + "issue:43550|issue:45072", + "issue:43746|issue:43873", + "issue:43550|issue:43819", + "issue:44186|issue:44898", + "issue:44843|issue:45081", + "issue:44062|issue:44488", + "issue:43577|issue:43818", + "issue:43531|issue:43643", + "issue:43643|issue:44877", + "issue:43421|issue:44811", + "issue:43540|issue:45381", + "issue:43653|issue:43819", + "issue:44448|issue:44671", + "issue:43782|issue:44977", + "issue:45584|issue:45834", + "issue:43322|issue:45685", + "issue:43825|issue:43994", + "issue:44821|issue:45084", + "issue:44561|issue:45020", + "issue:44485|issue:44509", + "issue:30064|issue:36331", + "issue:39692|issue:45834", + "issue:44479|issue:45834" + ], + "prepared_review_unit_hash": "5556f93240bfdb776111a24644ea8792bc4371cb0b8413fc464f09618ec9e96e", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36855, + "estimated_input_tokens": 9214, + "estimated_eval_tokens": 18684 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is not a true duplicate set: it mixes many unrelated bugs across tokenizers, multimodal processors, quantization/loading, and version regressions. The soft links are mostly title-level similarity, not the same concrete code-path failure.", + "confidence": 0.95, + "canonical_issue_reason": "No single canonical issue fits well; the items are heterogeneous and do not describe one underlying bug.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the broadest umbrella complaint in the set (recent transformers versions breaking remote_code models), but it still does not subsume the rest as duplicates.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43720", + "right": "issue:43819", + "accept": false, + "reason": "BitNet accelerate-loading failure vs DAC latent/STE mismatch; unrelated model paths." + }, + { + "left": "issue:44077", + "right": "issue:44625", + "accept": false, + "reason": "PatchTSMixer post-init policy vs Qwen3.5 num_labels propagation; different subsystems." + }, + { + "left": "issue:43653", + "right": "issue:43720", + "accept": false, + "reason": "BigBird tokenizer special-token decode bug vs BitNet weight loading bug; not the same issue." + }, + { + "left": "issue:43716", + "right": "issue:44008", + "accept": false, + "reason": "Mistral image dtype mismatch vs Gemma3n audio variable collision; different code paths and modalities." + }, + { + "left": "issue:42915", + "right": "issue:43746", + "accept": false, + "reason": "Qwen3Moe FP8 config failure vs GraniteSpeech PEFT/local checkpoint loading; unrelated failures." + }, + { + "left": "issue:43295", + "right": "issue:45834", + "accept": false, + "reason": "Processor/tokenizer regression vs Kosmos2.5 OCR index error; only broad multimodal overlap." + }, + { + "left": "issue:43746", + "right": "issue:45381", + "accept": false, + "reason": "GraniteSpeech adapter loading vs Qwen2.5-VL video position ids; different models and bugs." + }, + { + "left": "issue:43550", + "right": "issue:45072", + "accept": false, + "reason": "torch.compile SDPA issue vs bf16 dtype mismatch; not the same concrete problem." + }, + { + "left": "issue:43746", + "right": "issue:43873", + "accept": false, + "reason": "PEFT checkpoint loading vs quantization offloading behavior; distinct failure modes." + }, + { + "left": "issue:43550", + "right": "issue:43819", + "accept": false, + "reason": "Bamba SDPA/compile bug vs DAC from_latents mismatch; unrelated." + }, + { + "left": "issue:44186", + "right": "issue:44898", + "accept": false, + "reason": "Tokenizer padding/truncation crash vs image classification/interpolation failure; different components." + }, + { + "left": "issue:44843", + "right": "issue:45081", + "accept": false, + "reason": "Same helper function area, but different bugs: offline model_info call vs backend_tokenizer AttributeError." + }, + { + "left": "issue:44062", + "right": "issue:44488", + "accept": false, + "reason": "AddedToken special-arg conflict vs model loading failure; no concrete duplicate evidence." + }, + { + "left": "issue:43577", + "right": "issue:43818", + "accept": false, + "reason": "Blip2 dtype propagation bug vs Video-LLaVA tower architecture issue; unrelated." + }, + { + "left": "issue:43531", + "right": "issue:43643", + "accept": false, + "reason": "Qwen3-MoE sliding window bug vs trust_remote_code config field loss; distinct areas." + }, + { + "left": "issue:43643", + "right": "issue:44877", + "accept": false, + "reason": "Both involve config loading, but one is missing fields and the other is strict GraniteSpeech config rejection." + }, + { + "left": "issue:43421", + "right": "issue:44811", + "accept": false, + "reason": "Tokenizer backend post-processor sync issue vs Whisper batch_decode skip_special_tokens behavior; not the same bug." + }, + { + "left": "issue:43540", + "right": "issue:45381", + "accept": false, + "reason": "Different video-input regressions in different Qwen model families; not mergeable as one fix." + }, + { + "left": "issue:43653", + "right": "issue:43819", + "accept": false, + "reason": "BigBird special-token registration bug vs DAC latent mismatch; unrelated." + }, + { + "left": "issue:44448", + "right": "issue:44671", + "accept": false, + "reason": "Different models and outputs (Pegasus vs CamemBERT); no shared concrete code-path." + }, + { + "left": "issue:43782", + "right": "issue:44977", + "accept": false, + "reason": "Qwen3VL from_pretrained weight_only error vs Qwen3.5 flash-attention generation issue; distinct failures." + }, + { + "left": "issue:45584", + "right": "issue:45834", + "accept": false, + "reason": "Whisper empty-transcription generation bug vs Kosmos2.5 OCR long-input index error; unrelated." + }, + { + "left": "issue:43322", + "right": "issue:45685", + "accept": false, + "reason": "Llava Next load segfault vs MPS histogram Int limitation; completely different areas." + }, + { + "left": "issue:43825", + "right": "issue:43994", + "accept": false, + "reason": "Pipeline docs/error-message issue vs SigLIP2 bad predictions; not the same bug." + }, + { + "left": "issue:44821", + "right": "issue:45084", + "accept": false, + "reason": "AutoImageProcessor URL loading vs template-node compilation error; unrelated." + }, + { + "left": "issue:44561", + "right": "issue:45020", + "accept": false, + "reason": "Related broad theme around remote_code regressions, but these are not the same concrete breakage." + }, + { + "left": "issue:44485", + "right": "issue:44509", + "accept": false, + "reason": "GLM-5 RoPE implementation discussion vs documentation task-name mismatch; no duplicate overlap." + }, + { + "left": "issue:30064", + "right": "issue:36331", + "accept": false, + "reason": "Image segmentation-map processing bug vs CustomTrainer compute_loss signature change; unrelated." + }, + { + "left": "issue:39692", + "right": "issue:45834", + "accept": false, + "reason": "SigLIP2 docs/model mismatch and quantization failure vs Kosmos2.5 OCR indexing bug; different issues." + }, + { + "left": "issue:44479", + "right": "issue:45834", + "accept": false, + "reason": "Qwen video-input regression vs Kosmos2.5 OCR long-input index error; only generic multimodal similarity." + } + ] + }, + "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": 16, + "review_unit_index": 15, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:34634", + "issue:36010", + "issue:42175", + "issue:43262", + "issue:43329", + "issue:43404", + "issue:43452", + "issue:43526", + "issue:43577", + "issue:43582", + "issue:43606", + "issue:43638", + "issue:43644", + "issue:43650", + "issue:43653", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43746", + "issue:43749", + "issue:43782", + "issue:43819", + "issue:43825", + "issue:43827", + "issue:43906", + "issue:44060", + "issue:44186", + "issue:44368", + "issue:44442", + "issue:44610", + "issue:44671", + "issue:44805", + "issue:44811", + "issue:44857", + "issue:44912", + "issue:44936", + "issue:44991", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45406", + "issue:45459", + "issue:45538", + "issue:45600", + "issue:45701", + "issue:45803" + ], + "soft_pairs": [ + "issue:43825|issue:43827", + "issue:44186|issue:44442", + "issue:43606|issue:45072", + "issue:43577|issue:43582", + "issue:34634|issue:36010", + "issue:44805|issue:44936", + "issue:43452|issue:44991", + "issue:42175|issue:45081", + "issue:44610|issue:45072", + "issue:43720|issue:45072", + "issue:45081|issue:45406", + "issue:43404|issue:43782", + "issue:44060|issue:44368", + "issue:43650|issue:43906", + "issue:44671|issue:45701", + "issue:43582|issue:44857", + "issue:43262|issue:43819", + "issue:43526|issue:43697", + "issue:45459|issue:45803", + "issue:45084|issue:45538", + "issue:43644|issue:43749", + "issue:43329|issue:45600", + "issue:43742|issue:45020", + "issue:43653|issue:44912", + "issue:43638|issue:44936", + "issue:43746|issue:44811" + ], + "prepared_review_unit_hash": "ceb51bba412330ce799f5e96eb586fb67791c88f7b9984d44673c40e03202c3f", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35264, + "estimated_input_tokens": 8816, + "estimated_eval_tokens": 17888 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab bag of distinct Transformers regressions across tokenizers, processors, loading paths, dtype/device issues, and training/runtime errors. I found no true duplicate pair among the soft edges; the closest matches are only topic-level overlaps, not the same underlying bug.", + "confidence": 0.9, + "canonical_issue_reason": "Use issue 45020 as the canonical issue: it is the broadest umbrella report about recent Transformers versions breaking model behavior, and it best represents the general regression theme in this otherwise heterogeneous cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45020 is the best representative because it is open, high-level, and covers the widest slice of version-related breakage; the rest are narrower, model-specific bugs that do not collapse cleanly into one concrete root cause.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43825", + "right": "issue:43827", + "accept": false, + "reason": "Both involve pipeline/translation deprecation fallout, but one is an error message bug and the other is outdated docs; different artifacts and fixes." + }, + { + "left": "issue:44186", + "right": "issue:44442", + "accept": false, + "reason": "Different tokenizer failures on different models and code paths; only broad tokenizer-loading similarity." + }, + { + "left": "issue:43606", + "right": "issue:45072", + "accept": false, + "reason": "Device mismatch in Bark CPU offload vs dtype mismatch in other models; unrelated concrete bugs." + }, + { + "left": "issue:43577", + "right": "issue:43582", + "accept": false, + "reason": "BLIP2 dtype propagation vs Apple Silicon allocator TypeError; no shared underlying defect." + }, + { + "left": "issue:34634", + "right": "issue:36010", + "accept": false, + "reason": "BarkProcessor voice_preset bug is unrelated to a transformers.GenerationMixin import error." + }, + { + "left": "issue:44805", + "right": "issue:44936", + "accept": false, + "reason": "Different failures: mask/index shape mismatch vs trainer lifecycle failure after train()." + }, + { + "left": "issue:43452", + "right": "issue:44991", + "accept": false, + "reason": "Both are loading/tokenizer issues, but the models and root causes differ; not the same bug." + }, + { + "left": "issue:42175", + "right": "issue:45081", + "accept": false, + "reason": "Backend dependency packaging issue vs tokenizer regex patch crash; unrelated." + }, + { + "left": "issue:44610", + "right": "issue:45072", + "accept": false, + "reason": "Processor image size mismatch vs inference dtype mismatch; different code paths." + }, + { + "left": "issue:43720", + "right": "issue:45072", + "accept": false, + "reason": "Packed-weight loading bug in BitNet vs dtype mismatch in inference; not mergeable as one fix." + }, + { + "left": "issue:45081", + "right": "issue:45406", + "accept": false, + "reason": "Both are attribute errors during loading, but one is Mistral tokenizer regex patching and the other is Gemma4 serve processor state." + }, + { + "left": "issue:43404", + "right": "issue:43782", + "accept": false, + "reason": "Mistral lm_head tying bug vs Qwen3VL weight_only load error; different models and mechanisms." + }, + { + "left": "issue:44060", + "right": "issue:44368", + "accept": false, + "reason": "Both mention tied-weights warnings, but the underlying models and behavior are different warning/reporting issues." + }, + { + "left": "issue:43650", + "right": "issue:43906", + "accept": false, + "reason": "Placeholder/irrelevant issue vs isolated reproduction of another issue; no duplication." + }, + { + "left": "issue:44671", + "right": "issue:45701", + "accept": false, + "reason": "CamemBERT masked LM regression vs a general tokenization-change report; only loose version-related similarity." + }, + { + "left": "issue:43582", + "right": "issue:44857", + "accept": false, + "reason": "Allocator warmup TypeError and AMP crash are separate runtime failures." + }, + { + "left": "issue:43262", + "right": "issue:43819", + "accept": false, + "reason": "Audio sampling-rate defaulting bug vs DAC latent reconstruction mismatch; unrelated." + }, + { + "left": "issue:43526", + "right": "issue:43697", + "accept": false, + "reason": "BEiT label reduction bug vs RTDetrV2 output drift; different models and fixes." + }, + { + "left": "issue:45459", + "right": "issue:45803", + "accept": false, + "reason": "Both involve overly broad exception handling, but they are different functions and failure modes." + }, + { + "left": "issue:45084", + "right": "issue:45538", + "accept": false, + "reason": "Template compilation TypeError vs tokenizer max_length constant misuse; not the same bug." + }, + { + "left": "issue:43644", + "right": "issue:43749", + "accept": false, + "reason": "Non-persistent buffer corruption in v5 vs FSDP CPU RAM efficient loading breakage; different subsystems." + }, + { + "left": "issue:43329", + "right": "issue:45600", + "accept": false, + "reason": "Undefined multimodal-token helper bug vs stale auto_mappings references; unrelated maintenance issues." + }, + { + "left": "issue:43742", + "right": "issue:45020", + "accept": false, + "reason": "MobileLLM key error may be version-related, but it is a distinct loading failure from the broad remote_code regression report." + }, + { + "left": "issue:43653", + "right": "issue:44912", + "accept": false, + "reason": "BigBird tokenizer special-token bug vs MXFP4 quantization fallback; no shared root cause." + }, + { + "left": "issue:43638", + "right": "issue:44936", + "accept": false, + "reason": "Deepspeed zero3 indexing error vs trainer evaluate-after-train failure; separate training bugs." + }, + { + "left": "issue:43746", + "right": "issue:44811", + "accept": false, + "reason": "PEFT/local checkpoint loading issue vs Whisper batch_decode skip_special_tokens bug; 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": 2, + "component_count": 16, + "review_unit_index": 16, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:34567", + "issue:34634", + "issue:36010", + "issue:36331", + "issue:38175", + "issue:39692", + "issue:40990", + "issue:42915", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43450", + "issue:43452", + "issue:43454", + "issue:43526", + "issue:43531", + "issue:43550", + "issue:43577", + "issue:43582", + "issue:43606", + "issue:43643", + "issue:43645", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43819", + "issue:43854", + "issue:43927", + "issue:43931", + "issue:44291", + "issue:44479", + "issue:44554", + "issue:44568", + "issue:44589", + "issue:44610", + "issue:44811", + "issue:44849", + "issue:44912", + "issue:44977", + "issue:45042", + "issue:45356", + "issue:45357", + "issue:45488", + "issue:45701", + "issue:45759", + "issue:45823", + "issue:45859" + ], + "soft_pairs": [ + "issue:44589|issue:45042", + "issue:43854|issue:44977", + "issue:38175|issue:44610", + "issue:39692|issue:44912", + "issue:43526|issue:44554", + "issue:36010|issue:43645", + "issue:34634|issue:36331", + "issue:43819|issue:43927", + "issue:43454|issue:43577", + "issue:43257|issue:43329", + "issue:43720|issue:44610", + "issue:43550|issue:44912", + "issue:44568|issue:45488", + "issue:43749|issue:44291", + "issue:43531|issue:45042", + "issue:43262|issue:44811", + "issue:44568|issue:45356", + "issue:36010|issue:36331", + "issue:43643|issue:45759", + "issue:44849|issue:45701", + "issue:43329|issue:43450", + "issue:34567|issue:40990", + "issue:43329|issue:44479", + "issue:43606|issue:45859", + "issue:43257|issue:43931", + "issue:43697|issue:45357", + "issue:45823|issue:45859", + "issue:44291|issue:45042", + "issue:43452|issue:43582", + "issue:42915|issue:43742" + ], + "prepared_review_unit_hash": "db3ecc0d08eadbf788e6b89d0c5c92bc06d6036cfc67a6a90f1178f0f33fcc0c", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36974, + "estimated_input_tokens": 9244, + "estimated_eval_tokens": 18744 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a set of unrelated closed issues from the same repo; only one soft pair looks like the same underlying regression in the video multimodal path.", + "confidence": 0.77, + "canonical_issue_reason": "issue:45701 is the broadest and most reusable issue in the set: it is open, has inbound references, and describes a general tokenization regression that can plausibly anchor related tokenizer behavior reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45701 is the best overall issue candidate because it is broad, still open, and likely to attract/organize follow-up reports around tokenizer behavior changes.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44589", + "right": "issue:45042", + "accept": false, + "reason": "Different failures: one is a storage/backend type error, the other is an image-processor dependency regression." + }, + { + "left": "issue:43854", + "right": "issue:44977", + "accept": false, + "reason": "Different models and code paths: GLM loading/unit-test failure vs Qwen3.5 flash-attention generation behavior." + }, + { + "left": "issue:38175", + "right": "issue:44610", + "accept": false, + "reason": "Unrelated subsystems: SigLIP2 probability output issue vs OmDet-Turbo image size mismatch." + }, + { + "left": "issue:39692", + "right": "issue:44912", + "accept": false, + "reason": "Docs/example errors for SigLIP2 are not the same bug as MXFP4 quantization fallback for git-oss-20b." + }, + { + "left": "issue:43526", + "right": "issue:44554", + "accept": false, + "reason": "Different components: Beit image label reduction vs an MPS attention correctness issue." + }, + { + "left": "issue:36010", + "right": "issue:43645", + "accept": false, + "reason": "Both are import/model-init regressions in v5, but they affect different mechanisms and are not the same code-path bug." + }, + { + "left": "issue:34634", + "right": "issue:36331", + "accept": false, + "reason": "BarkProcessor voice preset handling is unrelated to CustomTrainer.compute_loss signature changes." + }, + { + "left": "issue:43819", + "right": "issue:43927", + "accept": false, + "reason": "Different model behaviors: DAC latent reconstruction mismatch vs DiaConfig losing custom token IDs on save/load." + }, + { + "left": "issue:43454", + "right": "issue:43577", + "accept": false, + "reason": "Different model families and failures: weight tying in AyaVision vs dtype propagation in Blip2." + }, + { + "left": "issue:43257", + "right": "issue:43329", + "accept": false, + "reason": "Qwen3 MoE weight conversion/loading is unrelated to the multimodal token-counting video-branch undefined-variable bug." + }, + { + "left": "issue:43720", + "right": "issue:44610", + "accept": false, + "reason": "Packed-weight loading for BitNet is unrelated to the OmDet-Turbo processor input-size regression." + }, + { + "left": "issue:43550", + "right": "issue:44912", + "accept": false, + "reason": "torch.compile + SDPA failure in Bamba is a different issue than MXFP4 quantization fallback in git-oss-20b." + }, + { + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "Both are tokenizer regressions, but they involve different tokenizers and different failure modes." + }, + { + "left": "issue:43749", + "right": "issue:44291", + "accept": false, + "reason": "Related loading stack, but not the same bug: FSDP CPU RAM efficient loading vs init_empty_weights passing an unexpected argument." + }, + { + "left": "issue:43531", + "right": "issue:45042", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior is unrelated to PIL backend torchvision requirements." + }, + { + "left": "issue:43262", + "right": "issue:44811", + "accept": false, + "reason": "Both are audio processor issues, but one is chat-template sampling rate and the other is batch_decode skip_special_tokens handling." + }, + { + "left": "issue:44568", + "right": "issue:45356", + "accept": false, + "reason": "Different tokenizer regressions affecting different models; not the same underlying bug." + }, + { + "left": "issue:36010", + "right": "issue:36331", + "accept": false, + "reason": "Importing GenerationMixin and CustomTrainer.compute_loss keyword handling are unrelated failures." + }, + { + "left": "issue:43643", + "right": "issue:45759", + "accept": false, + "reason": "AutoConfig remote-code field loss is not the same issue as unwrapping text_config in composite multimodal configs." + }, + { + "left": "issue:44849", + "right": "issue:45701", + "accept": false, + "reason": "Qwen3.5 output_hidden_states behavior is a model forward bug, not the broader tokenization regression." + }, + { + "left": "issue:43329", + "right": "issue:43450", + "accept": false, + "reason": "Both involve video, but one is token counting with undefined vars and the other is batched processor output shape; different bugs." + }, + { + "left": "issue:34567", + "right": "issue:40990", + "accept": false, + "reason": "TrainerState token counters and GPT-OSS perplexity are unrelated." + }, + { + "left": "issue:43329", + "right": "issue:44479", + "accept": true, + "reason": "Both describe the same video-input regression in Qwen VLM paths, centered on the video multimodal code path used by several Qwen variants." + }, + { + "left": "issue:43606", + "right": "issue:45859", + "accept": false, + "reason": "Both mention CPU offload, but they affect different models and different failure points." + }, + { + "left": "issue:43257", + "right": "issue:43931", + "accept": false, + "reason": "Qwen3 MoE loading/conversion and Qwen3-VL weight-shape mismatch are different model-loading problems." + }, + { + "left": "issue:43697", + "right": "issue:45357", + "accept": false, + "reason": "RTDetrV2 output changes and Qwen3.5 save_pretrained key naming are unrelated regressions." + }, + { + "left": "issue:45823", + "right": "issue:45859", + "accept": false, + "reason": "Different offload/device-mapping failures in different models; not the same concrete bug." + }, + { + "left": "issue:44291", + "right": "issue:45042", + "accept": false, + "reason": "Unexpected init argument during empty-weight loading is unrelated to the torchvision dependency regression." + }, + { + "left": "issue:43452", + "right": "issue:43582", + "accept": false, + "reason": "gguf_file tokenizer/model loading and Apple Silicon allocator warmup are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:43742", + "accept": false, + "reason": "Fine-grained FP8 loading for Qwen3Moe is unrelated to the MobileLLM key error." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the only accepted soft pair (issue:43329 with issue:44479) is a conservative match: both point to the same Qwen video-input regression area and could plausibly be fixed together. The canonical issue choice is also reasonable because issue:45701 is open, broad, and has inbound references. A minor caveat is that the cluster is not entirely \u2018closed,\u2019 since several open issues are present, but that does not undermine the overall conclusion." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 17, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:33357", + "issue:34567", + "issue:36010", + "issue:36331", + "issue:38175", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43257", + "issue:43317", + "issue:43452", + "issue:43493", + "issue:43540", + "issue:43577", + "issue:43643", + "issue:43646", + "issue:43696", + "issue:43742", + "issue:43749", + "issue:43756", + "issue:43824", + "issue:43906", + "issue:43927", + "issue:43994", + "issue:44062", + "issue:44291", + "issue:44479", + "issue:44488", + "issue:44560", + "issue:44737", + "issue:44756", + "issue:44792", + "issue:44821", + "issue:44898", + "issue:44912", + "issue:44964", + "issue:44987", + "issue:44991", + "issue:45072", + "issue:45081", + "issue:45137", + "issue:45375", + "issue:45381", + "issue:45405", + "issue:45538", + "issue:45588", + "issue:45823" + ], + "soft_pairs": [ + "issue:41720|issue:43540", + "issue:43317|issue:45538", + "issue:33357|issue:34567", + "issue:42915|issue:44821", + "issue:44987|issue:45538", + "issue:42915|issue:45538", + "issue:43927|issue:44912", + "issue:44479|issue:45381", + "issue:43257|issue:43756", + "issue:43540|issue:44560", + "issue:43540|issue:44479", + "issue:43257|issue:43696", + "issue:43577|issue:45072", + "issue:45137|issue:45405", + "issue:38175|issue:44898", + "issue:44737|issue:45588", + "issue:44062|issue:44991", + "issue:43742|issue:44291", + "issue:43540|issue:45823", + "issue:43906|issue:45381", + "issue:43646|issue:43749", + "issue:44062|issue:45081", + "issue:43749|issue:44756", + "issue:34567|issue:36331", + "issue:36010|issue:42175", + "issue:43493|issue:43994", + "issue:43643|issue:45375", + "issue:42175|issue:43824", + "issue:44291|issue:44821", + "issue:43452|issue:44488", + "issue:44792|issue:44964" + ], + "prepared_review_unit_hash": "499a1faa77392d06b864adeca38057a338435625c404c8fb93dd0034d81112ab", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36473, + "estimated_input_tokens": 9119, + "estimated_eval_tokens": 18494 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is largely heterogeneous and contains several unrelated issue families. The only strong duplicate candidates are the Qwen video-input regression pair and the SigLIP2 correctness pair; the rest are not the same underlying bug/change.", + "confidence": 0.78, + "canonical_issue_reason": "No single issue cleanly represents the cluster: it spans unrelated topics (trainer API, tokenizers, model loading, device_map, multimodal video regressions, dtype issues, etc.).", + "canonical_pr_reason": null, + "best_issue_reason": "None of these issues is a reliable global canonical for the cluster; the set is too mixed for one issue to represent all others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41720", + "right": "issue:43540", + "accept": false, + "reason": "Different Qwen models and failures: CUDA assert during auto device mapping vs video-processing ValueError. Same broad area, not the same bug." + }, + { + "left": "issue:43317", + "right": "issue:45538", + "accept": false, + "reason": "Device-map/offload loading failure and CLIPTokenizer max_length bug are unrelated." + }, + { + "left": "issue:33357", + "right": "issue:34567", + "accept": false, + "reason": "Completely different subsystems: MacOS CLIP bus error vs TrainerState token counting." + }, + { + "left": "issue:42915", + "right": "issue:44821", + "accept": false, + "reason": "Qwen3Moe FP8 loading failure vs AutoImageProcessor URL loading; no shared code path." + }, + { + "left": "issue:44987", + "right": "issue:45538", + "accept": false, + "reason": "Model-loading regression for physical-intelligence/fast vs CLIP tokenizer length issue; unrelated." + }, + { + "left": "issue:42915", + "right": "issue:45538", + "accept": false, + "reason": "FP8 Qwen3Moe load failure is not the same as CLIPTokenizer model_max_length behavior." + }, + { + "left": "issue:43927", + "right": "issue:44912", + "accept": false, + "reason": "DiaConfig save/load token ID loss and git-oss quantization fallback are different bugs." + }, + { + "left": "issue:44479", + "right": "issue:45381", + "accept": true, + "reason": "Both describe the same Qwen2.5-VL / related Qwen video-input regression in v5.3.0, centered on incorrect video position handling." + }, + { + "left": "issue:43257", + "right": "issue:43756", + "accept": false, + "reason": "Qwen3 MoE weight conversion and Smollm3 RoPE-layer dropping are unrelated model-specific issues." + }, + { + "left": "issue:43540", + "right": "issue:44560", + "accept": false, + "reason": "Both are video-input failures, but in different Qwen model lines with different errors; not clearly one bug." + }, + { + "left": "issue:43540", + "right": "issue:44479", + "accept": false, + "reason": "Different model families; 44479 is a broader Qwen2.5-VL/Qwen3-VL regression, while 43540 is Qwen3OmniMoe-specific." + }, + { + "left": "issue:43257", + "right": "issue:43696", + "accept": false, + "reason": "Qwen3 MoE weight conversion vs GPT-oss OOM are unrelated." + }, + { + "left": "issue:43577", + "right": "issue:45072", + "accept": false, + "reason": "Both mention dtype mismatches, but in different models and code paths; not the same concrete bug." + }, + { + "left": "issue:45137", + "right": "issue:45405", + "accept": false, + "reason": "DeepSpeed ZeRO3 deque error and PEFT version bump are unrelated." + }, + { + "left": "issue:38175", + "right": "issue:44898", + "accept": false, + "reason": "Different vision-model correctness problems (SigLIP2 zero probs vs Perceiver non-default resolution failure)." + }, + { + "left": "issue:44737", + "right": "issue:45588", + "accept": false, + "reason": "XLNet CPU arange bug and flash_attention sink-less crash are unrelated." + }, + { + "left": "issue:44062", + "right": "issue:44991", + "accept": false, + "reason": "Tokenizer AddedToken argument error and est-roberta tokenizer loading failure are not the same bug." + }, + { + "left": "issue:43742", + "right": "issue:44291", + "accept": false, + "reason": "Different model-loading failures with different root causes: key error vs unexpected init_empty_weights argument." + }, + { + "left": "issue:43540", + "right": "issue:45823", + "accept": false, + "reason": "Both involve device-related inference issues, but on different models and code paths." + }, + { + "left": "issue:43906", + "right": "issue:45381", + "accept": false, + "reason": "The isolated reproduction points to another issue, but it is not clearly the same as the qwen2.5-vl video-position-id regression." + }, + { + "left": "issue:43646", + "right": "issue:43749", + "accept": false, + "reason": "Custom model initialization breakage and FSDP CPU RAM efficient loading are unrelated." + }, + { + "left": "issue:44062", + "right": "issue:45081", + "accept": false, + "reason": "AddedToken duplication and Mistral regex backend_tokenizer crash are different tokenizer problems." + }, + { + "left": "issue:43749", + "right": "issue:44756", + "accept": false, + "reason": "Both concern loading/memory, but one is an FSDP bug and the other is mmap/OOM on Strix Halo; not the same issue." + }, + { + "left": "issue:34567", + "right": "issue:36331", + "accept": false, + "reason": "Both are Trainer-related, but one is token accounting and the other is compute_loss signature incompatibility." + }, + { + "left": "issue:36010", + "right": "issue:42175", + "accept": false, + "reason": "GenerationMixin import failure and missing TensorFlow backend are unrelated packaging/import issues." + }, + { + "left": "issue:43493", + "right": "issue:43994", + "accept": true, + "reason": "Both are about SigLIP2 producing wrong outputs; one is the concrete HF-vs-JAX discrepancy and the other is the user-visible nonsensical-results symptom." + }, + { + "left": "issue:43643", + "right": "issue:45375", + "accept": false, + "reason": "Both mention missing fields, but one is trust_remote_code AutoConfig serialization and the other is a specific Qwen3_5MoeVisionConfig field-loss bug." + }, + { + "left": "issue:42175", + "right": "issue:43824", + "accept": false, + "reason": "Package backend composition and a missing Qwen2_5_VL import are unrelated." + }, + { + "left": "issue:44291", + "right": "issue:44821", + "accept": false, + "reason": "init_empty_weights/_is_hf_initialized and AutoImageProcessor URL loading are different failures." + }, + { + "left": "issue:43452", + "right": "issue:44488", + "accept": false, + "reason": "gguf_file/from_pretrained breakage and sleng-bert loading failure are not the same underlying bug." + }, + { + "left": "issue:44792", + "right": "issue:44964", + "accept": false, + "reason": "Different multimodal model failures; one is a failed janus test, the other is Phi-4-multimodal-instruct loading." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative: it characterizes the cluster as heterogeneous, and the only accepted soft edges are the two pairs that have clear textual alignment (Qwen video regression and SigLIP2 correctness). The remaining rejections are appropriately cautious and do not overstate duplicate evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 18, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:24643", + "issue:28282", + "issue:29127", + "issue:34689", + "issue:41762", + "issue:42175", + "issue:43295", + "issue:43452", + "issue:43526", + "issue:43531", + "issue:43577", + "issue:43582", + "issue:43643", + "issue:43653", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43756", + "issue:43819", + "issue:44466", + "issue:44484", + "issue:44514", + "issue:44561", + "issue:44568", + "issue:44589", + "issue:44625", + "issue:44704", + "issue:44811", + "issue:44821", + "issue:44843", + "issue:44877", + "issue:44898", + "issue:44936", + "issue:44938", + "issue:44977", + "issue:45005", + "issue:45042", + "issue:45081", + "issue:45216", + "issue:45290", + "issue:45325", + "issue:45356", + "issue:45381", + "issue:45405", + "issue:45479", + "issue:45600", + "issue:45823" + ], + "soft_pairs": [ + "issue:43526|issue:45479", + "issue:28282|issue:34689", + "issue:45042|issue:45081", + "issue:43756|issue:45325", + "issue:43742|issue:44589", + "issue:45405|issue:45600", + "issue:43531|issue:44821", + "issue:43697|issue:45216", + "issue:28282|issue:29127", + "issue:44514|issue:45381", + "issue:44484|issue:44704", + "issue:43582|issue:44977", + "issue:43643|issue:44561", + "issue:44625|issue:44877", + "issue:43720|issue:44898", + "issue:42175|issue:44938", + "issue:43452|issue:43577", + "issue:41762|issue:44936", + "issue:44514|issue:45290", + "issue:43452|issue:44843", + "issue:44466|issue:45005", + "issue:43295|issue:45356", + "issue:43819|issue:44811", + "issue:43653|issue:44568", + "issue:44936|issue:45823", + "issue:24643|issue:29127" + ], + "prepared_review_unit_hash": "27eb51972404d92118598cd891b8b3ddd7c4b3f0c8b693a232f5ccc615f3078d", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35357, + "estimated_input_tokens": 8840, + "estimated_eval_tokens": 17936 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are mostly unrelated Transformers bug reports that only share broad themes like loading, tokenization, config handling, or version regressions. None of the soft pairs looks like the same underlying bug/change, so this cluster should not be deduplicated as one issue set.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43526", + "right": "issue:45479", + "accept": false, + "reason": "Both concern labels/classification, but one is an image-processor bug and the other is a sequence-classification loss issue; different code paths." + }, + { + "left": "issue:28282", + "right": "issue:34689", + "accept": false, + "reason": "Missing PyTorch installation is an environment error; Llama 3.2 Vision loading is a regression in model loading behavior." + }, + { + "left": "issue:45042", + "right": "issue:45081", + "accept": false, + "reason": "One is about PIL backend requiring torchvision; the other is a Mistral regex/tokenizer attribute crash." + }, + { + "left": "issue:43756", + "right": "issue:45325", + "accept": false, + "reason": "Both mention vision/rope ideas, but they affect different models and different positional-index logic." + }, + { + "left": "issue:43742", + "right": "issue:44589", + "accept": false, + "reason": "Different failures: a key error while loading MobileLLM versus a Float8 storage lookup TypeError." + }, + { + "left": "issue:45405", + "right": "issue:45600", + "accept": false, + "reason": "One is a dependency version bump, the other is a stale auto_mappings reference to removed configs." + }, + { + "left": "issue:43531", + "right": "issue:44821", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior and AutoImageProcessor URL loading are unrelated issues." + }, + { + "left": "issue:43697", + "right": "issue:45216", + "accept": false, + "reason": "RTDetrV2 output differences and Qwen3.5 save_pretrained regression are separate bugs." + }, + { + "left": "issue:28282", + "right": "issue:29127", + "accept": false, + "reason": "ImportError for missing PyTorch is not the same as LayoutLMv3 box-validation messaging." + }, + { + "left": "issue:44514", + "right": "issue:45381", + "accept": false, + "reason": "Both involve chat/template or VL inputs, but they fail in different shapes and model-specific position handling." + }, + { + "left": "issue:44484", + "right": "issue:44704", + "accept": false, + "reason": "Save-pretrained shard-size behavior and AutoProcessor kwarg forwarding are unrelated." + }, + { + "left": "issue:43582", + "right": "issue:44977", + "accept": false, + "reason": "Apple Silicon warmup TypeError and Qwen3.5 flash-attention generation are different runtime problems." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "One is missing fields in AutoConfig with trust_remote_code; the other is removal of a helper breaking remote-code models." + }, + { + "left": "issue:44625", + "right": "issue:44877", + "accept": false, + "reason": "Num-label propagation in Qwen3.5 and strict config rejection for granite_speech are not the same bug." + }, + { + "left": "issue:43720", + "right": "issue:44898", + "accept": false, + "reason": "BitNet packed-weight loading and Perceiver non-default-resolution behavior are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:44938", + "accept": false, + "reason": "TensorFlow extra dependency packaging and Python 3.14 load failure are different issues." + }, + { + "left": "issue:43452", + "right": "issue:43577", + "accept": false, + "reason": "GGUF loading failure and Blip2 dtype propagation do not share the same failing code path." + }, + { + "left": "issue:41762", + "right": "issue:44936", + "accept": false, + "reason": "Gemma3 ZeRO-3 loading IndexError and trainer.evaluate() after train are unrelated runtime failures." + }, + { + "left": "issue:44514", + "right": "issue:45290", + "accept": false, + "reason": "Both are apply_chat_template crashes, but one is batched padding=False and the other is assistant tool-call handling; not the same bug." + }, + { + "left": "issue:43452", + "right": "issue:44843", + "accept": false, + "reason": "GGUF tokenizer/model loading and offline mistral regex patching are distinct problems." + }, + { + "left": "issue:44466", + "right": "issue:45005", + "accept": false, + "reason": "Both mention tied weights, but one is a serialization inconsistency and the other is a broader translation-model regression." + }, + { + "left": "issue:43295", + "right": "issue:45356", + "accept": false, + "reason": "Processor.tokenizer regression and Kimi-K2.5 codec/fix_mistral_regex issues are different tokenizer bugs." + }, + { + "left": "issue:43819", + "right": "issue:44811", + "accept": false, + "reason": "DAC latent reconstruction mismatch and Whisper batch_decode skipping special tokens are unrelated." + }, + { + "left": "issue:43653", + "right": "issue:44568", + "accept": false, + "reason": "BigBird special-token registration and mdeberta BOS/EOS handling are separate tokenizer special-token bugs." + }, + { + "left": "issue:44936", + "right": "issue:45823", + "accept": false, + "reason": "Trainer evaluation failure and Gemma4 PLE device mismatch are different layers of the stack." + }, + { + "left": "issue:24643", + "right": "issue:29127", + "accept": false, + "reason": "DeepSpeed weight-shape RuntimeError and LayoutLMv3 box-info error 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": 2, + "component_count": 16, + "review_unit_index": 19, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:28282", + "issue:30064", + "issue:32090", + "issue:33666", + "issue:34567", + "issue:35141", + "issue:36010", + "issue:38175", + "issue:41720", + "issue:43122", + "issue:43295", + "issue:43329", + "issue:43452", + "issue:43504", + "issue:43525", + "issue:43531", + "issue:43550", + "issue:43572", + "issue:43577", + "issue:43653", + "issue:43688", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43756", + "issue:43906", + "issue:43927", + "issue:44361", + "issue:44442", + "issue:44451", + "issue:44464", + "issue:44568", + "issue:44610", + "issue:44743", + "issue:44821", + "issue:44843", + "issue:44849", + "issue:44913", + "issue:44977", + "issue:44991", + "issue:45325", + "issue:45356", + "issue:45357", + "issue:45440", + "issue:45588", + "issue:45812" + ], + "soft_pairs": [ + "issue:30064|issue:33666", + "issue:38175|issue:43295", + "issue:43653|issue:44361", + "issue:44361|issue:44442", + "issue:43577|issue:43697", + "issue:44464|issue:44743", + "issue:44843|issue:44991", + "issue:41720|issue:43906", + "issue:44977|issue:45356", + "issue:43688|issue:45440", + "issue:43531|issue:43742", + "issue:43653|issue:43927", + "issue:43550|issue:43927", + "issue:44464|issue:44977", + "issue:43577|issue:45588", + "issue:43550|issue:44361", + "issue:32090|issue:34567", + "issue:43504|issue:43550", + "issue:43122|issue:43906", + "issue:43504|issue:43720", + "issue:28282|issue:35141", + "issue:32090|issue:33666", + "issue:43504|issue:44610", + "issue:43452|issue:44451", + "issue:44849|issue:45357", + "issue:43742|issue:44821", + "issue:44568|issue:45812", + "issue:33666|issue:36010", + "issue:41720|issue:44464", + "issue:43329|issue:45325", + "issue:43756|issue:44913", + "issue:43525|issue:43572", + "issue:43688|issue:43749" + ], + "prepared_review_unit_hash": "60a0de3a9b98b15a138e0f22a5356a8693905288e89a2aeabebdddfca6049dbb", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 33, + "serialized_chars": 37184, + "estimated_input_tokens": 9296, + "estimated_eval_tokens": 18848 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly a loose semantic group of unrelated Transformers bugs, especially tokenizer/model-loading regressions plus several separate training/inference and vision issues. None of the soft-edge pairs look like true duplicates, so I would not merge any of them.", + "confidence": 0.94, + "canonical_issue_reason": "Issue 45812 is the broadest and most user-impactful recent tokenizer regression in the set: it affects all Granite models and describes a silent v4\u2192v5 ID change, making it a strong representative issue for the cluster\u2019s tokenizer-regression slice.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45812 is the best single representative because it is specific enough to be actionable yet broad enough to capture the most important common theme in the cluster (tokenizer regressions).", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:30064", + "right": "issue:33666", + "accept": false, + "reason": "Different subsystems: image processor void segmentation-map handling vs Qwen2-VL multi-GPU training." + }, + { + "left": "issue:38175", + "right": "issue:43295", + "accept": false, + "reason": "Unrelated bugs: SIGLIP2 zero probabilities vs a regression around processor.tokenizer and image passing." + }, + { + "left": "issue:43653", + "right": "issue:44361", + "accept": false, + "reason": "Different tokenizer failures: BigBird special-token decode issue vs MLukeTokenizer attribute error." + }, + { + "left": "issue:44361", + "right": "issue:44442", + "accept": false, + "reason": "Both are tokenizer-related, but one is an MLukeTokenizer task error and the other is AutoTokenizer failing on a different tokenizer class." + }, + { + "left": "issue:43577", + "right": "issue:43697", + "accept": false, + "reason": "Different model behavior bugs: BLIP2 dtype propagation vs RTDetrV2 output drift in v5." + }, + { + "left": "issue:44464", + "right": "issue:44743", + "accept": false, + "reason": "Different code paths: compiled chunked generation inconsistency vs Qwen3.5 modular recurrent-state reset." + }, + { + "left": "issue:44843", + "right": "issue:44991", + "accept": false, + "reason": "Different tokenizer regressions: offline hub regex patch side effect vs EMBEDDIA/est-roberta loading failure." + }, + { + "left": "issue:41720", + "right": "issue:43906", + "accept": false, + "reason": "Qwen3 auto device mapping CUDA assert is unrelated to an isolated reproduction of another issue." + }, + { + "left": "issue:44977", + "right": "issue:45356", + "accept": false, + "reason": "Both involve Qwen/Kimi generation/tokenization, but the concrete failures are flash-attention generation vs codec/regex tokenizer regression." + }, + { + "left": "issue:43688", + "right": "issue:45440", + "accept": false, + "reason": "Different MoE/model-implementation bugs: auxiliary-loss normalization vs DeepseekV3 native divergence." + }, + { + "left": "issue:43531", + "right": "issue:43742", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to a key error when loading MobileLLM-125M." + }, + { + "left": "issue:43653", + "right": "issue:43927", + "accept": false, + "reason": "BigBird mask-token decode bug and DiaConfig custom token ID loss are separate tokenizer/config issues." + }, + { + "left": "issue:43550", + "right": "issue:43927", + "accept": false, + "reason": "Different failures: Bamba torch.compile SDPA vs DiaConfig save/load token-ID corruption." + }, + { + "left": "issue:44464", + "right": "issue:44977", + "accept": false, + "reason": "Both are generation bugs, but compiled forward inconsistency is not the same concrete issue as Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:43577", + "right": "issue:45588", + "accept": false, + "reason": "BLIP2 dtype loading and flash_attention.py crashing on s_aux=None are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:44361", + "accept": false, + "reason": "No shared bug path: Bamba compile/SDPA issue vs MLukeTokenizer task AttributeError." + }, + { + "left": "issue:32090", + "right": "issue:34567", + "accept": false, + "reason": "Trainer NoneType broadcast error is distinct from TrainerState.num_input_tokens_seen not updating." + }, + { + "left": "issue:43504", + "right": "issue:43550", + "accept": false, + "reason": "Beit pretrained-loading legacy field bug is unrelated to Bamba compile/SDPA behavior." + }, + { + "left": "issue:43122", + "right": "issue:43906", + "accept": false, + "reason": "Tokenizer output differences across versions are not the same as an isolated reproduction of another issue." + }, + { + "left": "issue:43504", + "right": "issue:43720", + "accept": false, + "reason": "Different loading issues: legacy field in Beit vs BitNet packed-weight unpacking during accelerate loading." + }, + { + "left": "issue:28282", + "right": "issue:35141", + "accept": false, + "reason": "AutoModel missing PyTorch import and token-embedding resizing/post_init reinitialization are unrelated." + }, + { + "left": "issue:32090", + "right": "issue:33666", + "accept": false, + "reason": "Trainer broadcast NoneType error is a different failure mode from Qwen2-VL multi-GPU training." + }, + { + "left": "issue:43504", + "right": "issue:44610", + "accept": false, + "reason": "Beit legacy-field loading bug and OmDet-Turbo processor/model image-size mismatch are different problems." + }, + { + "left": "issue:43452", + "right": "issue:44451", + "accept": false, + "reason": "gguf_file breaks specific AutoTokenizer/AutoModel loading paths, while ScandiBERT loading failure is a separate model/tokenizer compatibility issue." + }, + { + "left": "issue:44849", + "right": "issue:45357", + "accept": false, + "reason": "Qwen3.5 hidden-states bug and incorrect visual-encoder keys on save_pretrained are different regressions." + }, + { + "left": "issue:43742", + "right": "issue:44821", + "accept": false, + "reason": "MobileLLM key error on load is unrelated to AutoImageProcessor loading from URL." + }, + { + "left": "issue:44568", + "right": "issue:45812", + "accept": false, + "reason": "mdeberta-v3 special-token addition bug is not the same regression as Granite token-ID changes; different tokenizer families and behaviors." + }, + { + "left": "issue:33666", + "right": "issue:36010", + "accept": false, + "reason": "Qwen2-VL multi-GPU training and GenerationMixin import failure are unrelated." + }, + { + "left": "issue:41720", + "right": "issue:44464", + "accept": false, + "reason": "CUDA assert on auto device mapping is a different failure from compiled-forward generation inconsistency." + }, + { + "left": "issue:43329", + "right": "issue:45325", + "accept": false, + "reason": "Undefined helper/merge_size in video multimodal token counting is distinct from Qwen2.5-VL rope-index scaling for still images." + }, + { + "left": "issue:43756", + "right": "issue:44913", + "accept": false, + "reason": "Smollm3 RoPE layer mismatch and GPTNeoX rotary_pct reload bug are separate configuration issues." + }, + { + "left": "issue:43525", + "right": "issue:43572", + "accept": false, + "reason": "Both are config attribute regressions, but they concern different fields and model families (pad_token_id vs pad_token_idx)." + }, + { + "left": "issue:43688", + "right": "issue:43749", + "accept": false, + "reason": "Aux-loss normalization in MoE models and FSDP_CPU_RAM_EFFICIENT_LOADING are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 20, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:28282", + "issue:32090", + "issue:33357", + "issue:33666", + "issue:36331", + "issue:39692", + "issue:43257", + "issue:43262", + "issue:43278", + "issue:43295", + "issue:43299", + "issue:43334", + "issue:43493", + "issue:43504", + "issue:43540", + "issue:43572", + "issue:43577", + "issue:43606", + "issue:43646", + "issue:43688", + "issue:43720", + "issue:43756", + "issue:43824", + "issue:43825", + "issue:43844", + "issue:43927", + "issue:44291", + "issue:44361", + "issue:44423", + "issue:44466", + "issue:44479", + "issue:44488", + "issue:44554", + "issue:44589", + "issue:44625", + "issue:44734", + "issue:44843", + "issue:44964", + "issue:44977", + "issue:45083", + "issue:45200", + "issue:45229", + "issue:45405", + "issue:45520", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45823" + ], + "soft_pairs": [ + "issue:43493|issue:43825", + "issue:43262|issue:43295", + "issue:43606|issue:43927", + "issue:43720|issue:43927", + "issue:28282|issue:32090", + "issue:28282|issue:33357", + "issue:43577|issue:44554", + "issue:33666|issue:36331", + "issue:39692|issue:43493", + "issue:43720|issue:44361", + "issue:43688|issue:45704", + "issue:43334|issue:43572", + "issue:44964|issue:45405", + "issue:43540|issue:44977", + "issue:44479|issue:44625", + "issue:43493|issue:43844", + "issue:43756|issue:45083", + "issue:43334|issue:45759", + "issue:45200|issue:45229", + "issue:44488|issue:44843", + "issue:43257|issue:43299", + "issue:43824|issue:44589", + "issue:43278|issue:43844", + "issue:43646|issue:44466", + "issue:45229|issue:45823", + "issue:44589|issue:45520", + "issue:43688|issue:45701", + "issue:39692|issue:43295", + "issue:43572|issue:44291", + "issue:33357|issue:33666", + "issue:43504|issue:43606", + "issue:44423|issue:44734", + "issue:43927|issue:44361" + ], + "prepared_review_unit_hash": "a1b233fedade9e9f596428ccc25679fa4f7c7264a9ec36dcc800c2653271f51c", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37975, + "estimated_input_tokens": 9494, + "estimated_eval_tokens": 19244 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is highly heterogeneous: most items are distinct Transformers bugs/regressions across unrelated models, loaders, and runtime paths. I found no true duplicate pairs among the soft edges.", + "confidence": 0.94, + "canonical_issue_reason": "Issue 43646 is the broadest, most representative regression report in the set (\u201cTransformers 5.0.0 breaks custom model initialization\u201d) and is phrased as a concrete framework-level breakage rather than a model-specific symptom.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43646 is the best global exemplar because it describes a clear, high-level compatibility regression with direct impact and a well-scoped failure mode.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43493", + "right": "issue:43825", + "accept": false, + "reason": "Different bugs: SigLIP2 implementation mismatch vs an incorrect pipeline error message about translation tasks." + }, + { + "left": "issue:43262", + "right": "issue:43295", + "accept": false, + "reason": "Unrelated regressions: audio sampling-rate defaulting vs processor.tokenizer / image passthrough breakage." + }, + { + "left": "issue:43606", + "right": "issue:43927", + "accept": false, + "reason": "Different code paths: CPU offload device mismatch in Bark vs DiaConfig losing custom token IDs on save/load." + }, + { + "left": "issue:43720", + "right": "issue:43927", + "accept": false, + "reason": "Different models and failures: BitNet packed-weight loading vs DiaConfig token-id persistence." + }, + { + "left": "issue:28282", + "right": "issue:32090", + "accept": false, + "reason": "ImportError for missing PyTorch is unrelated to Trainer's NoneType broadcast failure." + }, + { + "left": "issue:28282", + "right": "issue:33357", + "accept": false, + "reason": "Different underlying problems: missing PyTorch installation vs MacOS bus error loading a CLIP model." + }, + { + "left": "issue:43577", + "right": "issue:44554", + "accept": false, + "reason": "BLIP2 dtype mismatch is not the same as the MPS attention correctness issue." + }, + { + "left": "issue:33666", + "right": "issue:36331", + "accept": false, + "reason": "Multi-GPU training for Qwen2-VL is unrelated to CustomTrainer.compute_loss signature breakage." + }, + { + "left": "issue:39692", + "right": "issue:43493", + "accept": false, + "reason": "Docs/example errors are separate from a SigLIP2 HF-vs-JAX implementation discrepancy." + }, + { + "left": "issue:43720", + "right": "issue:44361", + "accept": false, + "reason": "BitNet packed-weight loading and MLukeTokenizer task attribute errors are different bugs." + }, + { + "left": "issue:43688", + "right": "issue:45704", + "accept": false, + "reason": "Auxiliary-loss normalization and T5 apex memory leak are unrelated issues." + }, + { + "left": "issue:43334", + "right": "issue:43572", + "accept": false, + "reason": "Both involve missing config fields, but for different models and different attributes (pad_token_id vs pad_token_idx)." + }, + { + "left": "issue:44964", + "right": "issue:45405", + "accept": false, + "reason": "Model loading failure for Phi-4 multimodal is unrelated to a dependency version bump for PEFT." + }, + { + "left": "issue:43540", + "right": "issue:44977", + "accept": false, + "reason": "Video-input processing failure in Qwen3OmniMoe is unrelated to flash-attention generation behavior in Qwen3.5." + }, + { + "left": "issue:44479", + "right": "issue:44625", + "accept": false, + "reason": "Video regression across Qwen variants is different from num_labels propagation in Qwen3.5 configs." + }, + { + "left": "issue:43493", + "right": "issue:43844", + "accept": false, + "reason": "SigLIP2 implementation discrepancy is unrelated to ZeRO-3 gradient growth." + }, + { + "left": "issue:43756", + "right": "issue:45083", + "accept": false, + "reason": "Smollm3 RoPE-layer dropping and a helper length function behaving unexpectedly are different failures." + }, + { + "left": "issue:43334", + "right": "issue:45759", + "accept": false, + "reason": "Both are multimodal config issues, but they affect different fields and initialization paths." + }, + { + "left": "issue:45200", + "right": "issue:45229", + "accept": false, + "reason": "Text-only token-type defaults are unrelated to Gemma 4 multi-GPU CUDA OOM." + }, + { + "left": "issue:44488", + "right": "issue:44843", + "accept": false, + "reason": "Model-loading failure for a specific checkpoint is not the same as an offline-mode tokenizer patch bug." + }, + { + "left": "issue:43257", + "right": "issue:43299", + "accept": false, + "reason": "Qwen3 MoE weight conversion under accelerate/deepspeed is distinct from a general Qwen3VL MoE loading regression." + }, + { + "left": "issue:43824", + "right": "issue:44589", + "accept": false, + "reason": "Missing Qwen2.5-VL import and missing Float8 storage support are unrelated import/runtime issues." + }, + { + "left": "issue:43278", + "right": "issue:43844", + "accept": false, + "reason": "A dtype change in embeddings is not the same bug as exploding gradients under ZeRO-3." + }, + { + "left": "issue:43646", + "right": "issue:44466", + "accept": false, + "reason": "Custom model initialization breakage is different from device-dependent serialization of lm_head.weight." + }, + { + "left": "issue:45229", + "right": "issue:45823", + "accept": false, + "reason": "OOM during multi-GPU inference is not the same as a device mismatch in PLE with device_map=\"auto\"." + }, + { + "left": "issue:44589", + "right": "issue:45520", + "accept": false, + "reason": "Float8 storage lookup failure and Python 3.13 flash_attn KeyError are separate compatibility issues." + }, + { + "left": "issue:43688", + "right": "issue:45701", + "accept": false, + "reason": "Aux-loss normalization and tokenization changes are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:43295", + "accept": false, + "reason": "Documentation example mistakes do not match the processor/tokenizer regression in custom model code." + }, + { + "left": "issue:43572", + "right": "issue:44291", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to init_empty_weights passing an unexpected argument." + }, + { + "left": "issue:33357", + "right": "issue:33666", + "accept": false, + "reason": "MacOS bus error on CLIP and Qwen2-VL multi-GPU training are different problems." + }, + { + "left": "issue:43504", + "right": "issue:43606", + "accept": false, + "reason": "Legacy-field pretrained loading failure is unrelated to a CPU-offload device mismatch." + }, + { + "left": "issue:44423", + "right": "issue:44734", + "accept": false, + "reason": "Continuous-batching multimodal crash and KV-cache continuation indexing bug are distinct serve-path failures." + }, + { + "left": "issue:43927", + "right": "issue:44361", + "accept": false, + "reason": "DiaConfig token-id persistence and MLukeTokenizer task attribute errors do not share the same root cause." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 21, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:15354", + "issue:28218", + "issue:29942", + "issue:30333", + "issue:33357", + "issue:34567", + "issue:38175", + "issue:43232", + "issue:43278", + "issue:43295", + "issue:43504", + "issue:43540", + "issue:43575", + "issue:43606", + "issue:43653", + "issue:43688", + "issue:43701", + "issue:43844", + "issue:43873", + "issue:43994", + "issue:44112", + "issue:44315", + "issue:44464", + "issue:44466", + "issue:44493", + "issue:44521", + "issue:44554", + "issue:44568", + "issue:44610", + "issue:44625", + "issue:44792", + "issue:44849", + "issue:44964", + "issue:45072", + "issue:45084", + "issue:45127", + "issue:45198", + "issue:45200", + "issue:45290", + "issue:45325", + "issue:45356", + "issue:45405", + "issue:45446", + "issue:45520", + "issue:45584", + "issue:45701", + "issue:45759", + "issue:45823" + ], + "soft_pairs": [ + "issue:43504|issue:43653", + "issue:43575|issue:45405", + "issue:44625|issue:45759", + "issue:43606|issue:44112", + "issue:44792|issue:45405", + "issue:43994|issue:44493", + "issue:44521|issue:45290", + "issue:29942|issue:30333", + "issue:44464|issue:45584", + "issue:43844|issue:44964", + "issue:28218|issue:34567", + "issue:45084|issue:45520", + "issue:44554|issue:45446", + "issue:43688|issue:43873", + "issue:44568|issue:45701", + "issue:15354|issue:38175", + "issue:43232|issue:44849", + "issue:43540|issue:45200", + "issue:45200|issue:45823", + "issue:45072|issue:45198", + "issue:44315|issue:44849", + "issue:43278|issue:43701", + "issue:28218|issue:33357", + "issue:44554|issue:45356", + "issue:45200|issue:45325", + "issue:43295|issue:44610", + "issue:44466|issue:45127" + ], + "prepared_review_unit_hash": "847b47c86ca43776114df3781461d17c53d64f6629760b84f5df2960691a68ea", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35873, + "estimated_input_tokens": 8969, + "estimated_eval_tokens": 18194 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a mixed cluster of unrelated issues across tokenization, generation, multimodal preprocessing, training state, quantization, and CI. None of the soft pairs look like the same underlying defect, so I would not collapse any of them as duplicates.", + "confidence": 0.92, + "canonical_issue_reason": "No single canonical issue fits the cluster: the items are heterogeneous and do not describe one shared bug.", + "canonical_pr_reason": null, + "best_issue_reason": "If a representative issue is needed, issue:44521 is the most active and clearly scoped, but it only represents one narrow multimodal chat-template bug and does not subsume the rest.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43504", + "right": "issue:43653", + "accept": false, + "reason": "Legacy field loading for Beit segmentation is unrelated to BigBirdTokenizer special-token registration and empty decode output." + }, + { + "left": "issue:43575", + "right": "issue:45405", + "accept": false, + "reason": "Tensor-parallel OOM on Qwen2-57B-A14B is unrelated to a PEFT minimum-version bump not yet on PyPI." + }, + { + "left": "issue:44625", + "right": "issue:45759", + "accept": false, + "reason": "Qwen3.5 num_labels propagation and Qwen 3.5/3.6 text_config unwrapping are different config bugs with different failure modes." + }, + { + "left": "issue:43606", + "right": "issue:44112", + "accept": false, + "reason": "CPU offload device mismatch in bark-small is unrelated to a stale GraniteSpeech CI test." + }, + { + "left": "issue:44792", + "right": "issue:45405", + "accept": false, + "reason": "Janus image-generation test failures are not the same as a PEFT version availability issue." + }, + { + "left": "issue:43994", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 nonsensical outputs are unrelated to unexpected position-id key handling." + }, + { + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "All-zero assistant masks for multimodal inputs are a different bug from crashing on assistant tool-call messages with no content." + }, + { + "left": "issue:29942", + "right": "issue:30333", + "accept": false, + "reason": "Flash Attention 2 CI failures and MLflow not reporting failed jobs are unrelated subsystems and symptoms." + }, + { + "left": "issue:44464", + "right": "issue:45584", + "accept": false, + "reason": "Chunked generation inconsistency with compiled forward is unrelated to empty-transcription generation after align_special_tokens." + }, + { + "left": "issue:43844", + "right": "issue:44964", + "accept": false, + "reason": "Gradient inflation under ZeRO-3 is not the same as failing to load a multimodal Phi-4 model." + }, + { + "left": "issue:28218", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer extra-space insertion and TrainerState num_input_tokens_seen not updating are unrelated bugs." + }, + { + "left": "issue:45084", + "right": "issue:45520", + "accept": false, + "reason": "Compilation of non-template nodes is unrelated to a flash_attn import KeyError on Python 3.13." + }, + { + "left": "issue:44554", + "right": "issue:45446", + "accept": false, + "reason": "An MPS attention correctness issue is unrelated to an incorrect PyTorch version check for AuxRequest import." + }, + { + "left": "issue:43688", + "right": "issue:43873", + "accept": false, + "reason": "Incorrect auxiliary-loss normalization and offloading/quantization behavior are different code paths and defects." + }, + { + "left": "issue:44568", + "right": "issue:45701", + "accept": false, + "reason": "A specific add_special_tokens regression for mdeberta-v3-base is not the same as a broad report that tokenization changes across versions." + }, + { + "left": "issue:15354", + "right": "issue:38175", + "accept": false, + "reason": "TorchScript generator-expression export failure is unrelated to SigLIP2 producing zero probabilities." + }, + { + "left": "issue:43232", + "right": "issue:44849", + "accept": false, + "reason": "Generation kwargs updates after sync_gpus are unrelated to Qwen3.5 output_hidden_states behavior." + }, + { + "left": "issue:43540", + "right": "issue:45200", + "accept": false, + "reason": "Qwen3OmniMoe video-input processing errors are unrelated to Gemma 4 mm_token_type_ids defaults." + }, + { + "left": "issue:45200", + "right": "issue:45823", + "accept": false, + "reason": "Defaulting mm_token_type_ids to zeros is a different issue from Gemma4 PLE device mismatch during forward." + }, + { + "left": "issue:45072", + "right": "issue:45198", + "accept": false, + "reason": "bfloat16 inference dtype mismatches are unrelated to Wav2Vec2 save_pretrained/tokenization failures." + }, + { + "left": "issue:44315", + "right": "issue:44849", + "accept": false, + "reason": "Liger Kernel not applied with model_init is unrelated to Qwen3.5 output_hidden_states bugs." + }, + { + "left": "issue:43278", + "right": "issue:43701", + "accept": false, + "reason": "Embedding dtype changing between train/eval is unrelated to resume_from_checkpoint key mismatch." + }, + { + "left": "issue:28218", + "right": "issue:33357", + "accept": false, + "reason": "Added-token spacing is not the same defect as a MacOS bus error in a CLIP community model." + }, + { + "left": "issue:44554", + "right": "issue:45356", + "accept": false, + "reason": "MPS attention correctness and Kimi tokenizer codec handling are different issues in different components." + }, + { + "left": "issue:43295", + "right": "issue:44610", + "accept": false, + "reason": "Processor.tokenizer access regression is unrelated to OmDet-Turbo producing 640px inputs instead of 224px." + }, + { + "left": "issue:44466", + "right": "issue:45127", + "accept": false, + "reason": "Device-dependent lm_head serialization is unrelated to LoRA merge collapse with extended vocabulary." + }, + { + "left": "issue:45200", + "right": "issue:45325", + "accept": false, + "reason": "mm_token_type_ids defaults are unrelated to the Qwen2.5-VL rope-index scaling bug." + } + ] + }, + "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": 16, + "review_unit_index": 22, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:37428", + "issue:38175", + "issue:42915", + "issue:43262", + "issue:43278", + "issue:43295", + "issue:43334", + "issue:43425", + "issue:43450", + "issue:43493", + "issue:43525", + "issue:43550", + "issue:43572", + "issue:43575", + "issue:43653", + "issue:43701", + "issue:43819", + "issue:43824", + "issue:43825", + "issue:43827", + "issue:43844", + "issue:43856", + "issue:43873", + "issue:43901", + "issue:43957", + "issue:43975", + "issue:44112", + "issue:44206", + "issue:44360", + "issue:44442", + "issue:44464", + "issue:44514", + "issue:44521", + "issue:44561", + "issue:44610", + "issue:44792", + "issue:44857", + "issue:44898", + "issue:44987", + "issue:45083", + "issue:45127", + "issue:45198", + "issue:45245", + "issue:45412", + "issue:45563", + "issue:45584", + "issue:45593" + ], + "soft_pairs": [ + "issue:44898|issue:45412", + "issue:43493|issue:43701", + "issue:43572|issue:44987", + "issue:37428|issue:43957", + "issue:43572|issue:43957", + "issue:43334|issue:43525", + "issue:43873|issue:44464", + "issue:43295|issue:43450", + "issue:43824|issue:43975", + "issue:43550|issue:43844", + "issue:43295|issue:43873", + "issue:37428|issue:43824", + "issue:38175|issue:45412", + "issue:42915|issue:43824", + "issue:38175|issue:43493", + "issue:44514|issue:44521", + "issue:44442|issue:45198", + "issue:43278|issue:43856", + "issue:43827|issue:43901", + "issue:43653|issue:44112", + "issue:43450|issue:43493", + "issue:43450|issue:43575", + "issue:44792|issue:45563", + "issue:44360|issue:44561", + "issue:45127|issue:45245", + "issue:43819|issue:45083", + "issue:44206|issue:45584", + "issue:43819|issue:44112", + "issue:43262|issue:43450", + "issue:44610|issue:44857", + "issue:37428|issue:43425", + "issue:44857|issue:45593", + "issue:43278|issue:43825" + ], + "prepared_review_unit_hash": "a48961fed2af815973f8dfb216c2e56e301d431718f9320a5a25e077a2ecec27", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 33, + "serialized_chars": 36923, + "estimated_input_tokens": 9231, + "estimated_eval_tokens": 18718 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous; nearly all soft pairs are different bugs or docs/test regressions. The only plausible duplicate-like pair is the shared `pad_token_id` AttributeError regression across Qwen3-VL and Llama4 config loading.", + "confidence": 0.74, + "canonical_issue_reason": "issue:43334 is the clearest representative of the one plausible duplicate theme: config-loading failures caused by missing `pad_token_id` on newer text configs.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43334 is the best cluster representative because it captures the common `pad_token_id` regression pattern and has enough discussion to anchor triage.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44898", + "right": "issue:45412", + "accept": false, + "reason": "Different problems: Perceiver image sizing vs RT-DETR memory reclamation." + }, + { + "left": "issue:43493", + "right": "issue:43701", + "accept": false, + "reason": "SigLIP2 output discrepancy and resume checkpoint key mismatch are unrelated." + }, + { + "left": "issue:43572", + "right": "issue:44987", + "accept": false, + "reason": "Different models and failures: StableLm pad token attr vs loading `physical-intelligence/fast`." + }, + { + "left": "issue:37428", + "right": "issue:43957", + "accept": false, + "reason": "ImportError for flash support is unrelated to meta-device model loading." + }, + { + "left": "issue:43572", + "right": "issue:43957", + "accept": false, + "reason": "No shared code path beyond general loading issues." + }, + { + "left": "issue:43334", + "right": "issue:43525", + "accept": true, + "reason": "Both are the same `pad_token_id` AttributeError regression during model loading; likely one fix." + }, + { + "left": "issue:43873", + "right": "issue:44464", + "accept": false, + "reason": "Quantization/offloading behavior is unrelated to compiled chunked generation outputs." + }, + { + "left": "issue:43295", + "right": "issue:43450", + "accept": false, + "reason": "Different processor bugs: tokenizer/image regression vs batched video shape handling." + }, + { + "left": "issue:43824", + "right": "issue:43975", + "accept": false, + "reason": "ImportError for Qwen2.5-VL is unrelated to DeepSeek detokenization." + }, + { + "left": "issue:43550", + "right": "issue:43844", + "accept": false, + "reason": "torch.compile SDPA failure and DeepSpeed gradient growth are different failures." + }, + { + "left": "issue:43295", + "right": "issue:43873", + "accept": false, + "reason": "Processor regression vs quantization offloading are not the same bug." + }, + { + "left": "issue:37428", + "right": "issue:43824", + "accept": false, + "reason": "Different import failures in different model families." + }, + { + "left": "issue:38175", + "right": "issue:45412", + "accept": false, + "reason": "SigLIP2 zero probabilities and RT-DETR memory leak are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:43824", + "accept": false, + "reason": "Qwen3MoE FP8 failure is unrelated to Qwen2.5-VL import error." + }, + { + "left": "issue:38175", + "right": "issue:43493", + "accept": false, + "reason": "Both are SigLIP2-related, but the reported symptoms are not the same concrete bug." + }, + { + "left": "issue:44514", + "right": "issue:44521", + "accept": false, + "reason": "Same API area, but one is a crash and the other is wrong assistant masks; not enough to treat as one bug." + }, + { + "left": "issue:44442", + "right": "issue:45198", + "accept": false, + "reason": "Different tokenizer/save_pretrained failures in different speech models." + }, + { + "left": "issue:43278", + "right": "issue:43856", + "accept": false, + "reason": "Embedding dtype drift and MoE memory usage are different training issues." + }, + { + "left": "issue:43827", + "right": "issue:43901", + "accept": false, + "reason": "Both are docs-related, but they cover different pipeline/API deprecations." + }, + { + "left": "issue:43653", + "right": "issue:44112", + "accept": false, + "reason": "BigBirdTokenizer special-token registration and GraniteSpeech CI test staleness are unrelated." + }, + { + "left": "issue:43450", + "right": "issue:43493", + "accept": false, + "reason": "Video batch-shape bug is unrelated to SigLIP2 implementation discrepancy." + }, + { + "left": "issue:43450", + "right": "issue:43575", + "accept": false, + "reason": "Video processor shape handling and TP OOM on Qwen2-57B are unrelated." + }, + { + "left": "issue:44792", + "right": "issue:45563", + "accept": false, + "reason": "Janus image-generation test failure and stale paged-generate warning are unrelated." + }, + { + "left": "issue:44360", + "right": "issue:44561", + "accept": false, + "reason": "DSA indexer activation issue and removed helper breaking trust_remote_code are unrelated." + }, + { + "left": "issue:45127", + "right": "issue:45245", + "accept": false, + "reason": "LoRA merge collapse and category cardinality overflow are unrelated." + }, + { + "left": "issue:43819", + "right": "issue:45083", + "accept": false, + "reason": "DAC latent mismatch and qwen3_omni_moe length helper behavior are different code paths." + }, + { + "left": "issue:44206", + "right": "issue:45584", + "accept": false, + "reason": "Feature-extractor arg regression and Whisper empty transcription bug are unrelated." + }, + { + "left": "issue:43819", + "right": "issue:44112", + "accept": false, + "reason": "DAC latent mismatch is unrelated to GraniteSpeech CI device override staleness." + }, + { + "left": "issue:43262", + "right": "issue:43450", + "accept": false, + "reason": "Audio chat-template sampling-rate default bug is unrelated to video processor batching." + }, + { + "left": "issue:44610", + "right": "issue:44857", + "accept": false, + "reason": "Processor output resolution mismatch and AMP loss crash are different problems." + }, + { + "left": "issue:37428", + "right": "issue:43425", + "accept": false, + "reason": "Flash-support import error is unrelated to Torch 2.10 compatibility." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "AMP/CUDA loss crash and missing auxiliary losses are different training bugs." + }, + { + "left": "issue:43278", + "right": "issue:43825", + "accept": false, + "reason": "Embedding dtype regression and pipeline translation-message docs mismatch are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The writeup is grounded in the packet and stays conservative on the soft-edge pairs. The only clearly supported duplicate theme is the shared `pad_token_id` AttributeError during config/model loading (`issue:43334` / `issue:43525`), and the remaining pair rejections are phrased narrowly enough for the evidence available." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 23, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:28218", + "issue:29942", + "issue:32090", + "issue:35141", + "issue:37428", + "issue:38175", + "issue:39692", + "issue:43335", + "issue:43388", + "issue:43504", + "issue:43526", + "issue:43550", + "issue:43606", + "issue:43611", + "issue:43653", + "issue:43720", + "issue:43819", + "issue:43931", + "issue:44112", + "issue:44206", + "issue:44292", + "issue:44351", + "issue:44448", + "issue:44462", + "issue:44493", + "issue:44561", + "issue:44568", + "issue:44743", + "issue:44779", + "issue:44855", + "issue:44857", + "issue:44977", + "issue:45072", + "issue:45083", + "issue:45200", + "issue:45229", + "issue:45290", + "issue:45310", + "issue:45381", + "issue:45459", + "issue:45478", + "issue:45491", + "issue:45520", + "issue:45588", + "issue:45593", + "issue:45712", + "issue:45823", + "issue:45859" + ], + "soft_pairs": [ + "issue:45290|issue:45381", + "issue:43606|issue:45083", + "issue:43335|issue:43550", + "issue:44448|issue:44743", + "issue:37428|issue:44351", + "issue:45200|issue:45588", + "issue:43611|issue:43931", + "issue:39692|issue:43526", + "issue:44462|issue:44561", + "issue:44206|issue:45823", + "issue:44561|issue:44779", + "issue:43388|issue:43526", + "issue:43335|issue:43653", + "issue:43819|issue:44493", + "issue:45310|issue:45478", + "issue:44855|issue:45083", + "issue:45588|issue:45859", + "issue:43550|issue:44112", + "issue:29942|issue:32090", + "issue:43335|issue:43504", + "issue:43335|issue:45083", + "issue:39692|issue:44493", + "issue:44292|issue:45520", + "issue:38175|issue:44493", + "issue:44448|issue:44779", + "issue:28218|issue:35141", + "issue:43335|issue:45072", + "issue:43335|issue:45593", + "issue:43720|issue:44112", + "issue:44977|issue:45229", + "issue:44448|issue:44977", + "issue:43606|issue:44857", + "issue:44568|issue:45491", + "issue:44857|issue:45072", + "issue:45459|issue:45712" + ], + "prepared_review_unit_hash": "ab2b5d43213756b27829843f65cf00cbcf03e43d2aed5b0bcda4835a2d6020fb", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38456, + "estimated_input_tokens": 9614, + "estimated_eval_tokens": 19484 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly false-positive similarity: the only clear duplicate is the repeated Qwen3.5 MoE from_pretrained error. The rest are related only at a broad subsystem level (tokenizers, flash-attn, loading, vision, CI) but describe different bugs.", + "confidence": 0.93, + "canonical_issue_reason": "issue:45310 is the best canonical issue because it is the earlier report of the exact same Qwen3.5 MoE from_pretrained failure later repeated as issue:45478.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the strongest representative because it has an exact duplicate and a concrete, narrowly scoped failure mode with clear reproducibility.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45290", + "right": "issue:45381", + "accept": false, + "reason": "Different bugs: chat template/tool-call handling vs Qwen2.5-VL video position IDs." + }, + { + "left": "issue:43606", + "right": "issue:45083", + "accept": false, + "reason": "Unrelated failures in different models and code paths." + }, + { + "left": "issue:43335", + "right": "issue:43550", + "accept": false, + "reason": "Sparse-layer config bug is unrelated to Bamba torch.compile/SDPA failure." + }, + { + "left": "issue:44448", + "right": "issue:44743", + "accept": false, + "reason": "Different model families and different regressions (output drift vs recurrent state reset)." + }, + { + "left": "issue:37428", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different missing symbols and different modules." + }, + { + "left": "issue:45200", + "right": "issue:45588", + "accept": false, + "reason": "Gemma4 token-type defaults and flash_attention s_aux handling are separate issues." + }, + { + "left": "issue:43611", + "right": "issue:43931", + "accept": false, + "reason": "Both loading errors, but one is a base_model_prefix regression and the other is a shape mismatch." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "Different image-processing problems in different models; not the same underlying bug." + }, + { + "left": "issue:44462", + "right": "issue:44561", + "accept": false, + "reason": "Tokenizer-file selection and removal of is_torch_fx_available are unrelated." + }, + { + "left": "issue:44206", + "right": "issue:45823", + "accept": false, + "reason": "Feature extractor arg validation vs Gemma4 device-map mismatch are not the same code-path." + }, + { + "left": "issue:44561", + "right": "issue:44779", + "accept": false, + "reason": "A missing utility symbol and a Deepseek tokenizer regression are different failures." + }, + { + "left": "issue:43388", + "right": "issue:43526", + "accept": false, + "reason": "Both concern labels, but one is metric gathering and the other is Beit label reduction." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "SwitchTransformers sparse-layer config and BigBird tokenizer special-token registration are unrelated." + }, + { + "left": "issue:43819", + "right": "issue:44493", + "accept": false, + "reason": "DAC latent conversion mismatch and unexpected position-id keys are different bugs." + }, + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Exact same Qwen3.5 MoE from_pretrained error; clear duplicate." + }, + { + "left": "issue:44855", + "right": "issue:45083", + "accept": false, + "reason": "Python 3.13 JIT parsing error is unrelated to qwen3_omni_moe feature-length behavior." + }, + { + "left": "issue:45588", + "right": "issue:45859", + "accept": false, + "reason": "Both touch flash/offload paths, but the concrete failure modes differ." + }, + { + "left": "issue:43550", + "right": "issue:44112", + "accept": false, + "reason": "Bamba SDPA compile failure and GraniteSpeech CI device-override stale test are unrelated." + }, + { + "left": "issue:29942", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention 2 test failures and Trainer _gpu_broadcast_one TypeError are different issues." + }, + { + "left": "issue:43335", + "right": "issue:43504", + "accept": false, + "reason": "Sparse-layer creation bug is unrelated to Beit pretrained preset loading." + }, + { + "left": "issue:43335", + "right": "issue:45083", + "accept": false, + "reason": "Different model families and different helper/config bugs." + }, + { + "left": "issue:39692", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 docs/model mismatch is unrelated to widespread position-id key warnings." + }, + { + "left": "issue:44292", + "right": "issue:45520", + "accept": false, + "reason": "Qwen-3-8B-NVFP4 runtime failure and flash_attn import KeyError are different problems." + }, + { + "left": "issue:38175", + "right": "issue:44493", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 is not the same as position-id key warnings." + }, + { + "left": "issue:44448", + "right": "issue:44779", + "accept": false, + "reason": "Both are regressions across versions, but in different models and failure modes." + }, + { + "left": "issue:28218", + "right": "issue:35141", + "accept": false, + "reason": "Tokenizer spacing bug and embedding reinitialization on resize are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:45072", + "accept": false, + "reason": "Sparse-layer config bug and bfloat16 dtype mismatch are different code paths." + }, + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "SwitchTransformers config issue is unrelated to D-FINE auxiliary-loss behavior." + }, + { + "left": "issue:43720", + "right": "issue:44112", + "accept": false, + "reason": "Packed-weight loading bug and GraniteSpeech CI stale test are unrelated." + }, + { + "left": "issue:44977", + "right": "issue:45229", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug and Gemma4 CUDA OOM are different failures." + }, + { + "left": "issue:44448", + "right": "issue:44977", + "accept": false, + "reason": "Different models and different output-generation regressions." + }, + { + "left": "issue:43606", + "right": "issue:44857", + "accept": false, + "reason": "CPU offload device mismatch and float16 AMP crash are separate issues." + }, + { + "left": "issue:44568", + "right": "issue:45491", + "accept": false, + "reason": "Tokenizer special-token behavior and Gemma3 NaN embeddings are unrelated." + }, + { + "left": "issue:44857", + "right": "issue:45072", + "accept": false, + "reason": "Float16 AMP crash and bfloat16 dtype mismatch are not the same bug." + }, + { + "left": "issue:45459", + "right": "issue:45712", + "accept": false, + "reason": "Tokenizer error masking and leftover dummy classes failing checks are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The judgment is conservative and grounded in the packet. The only clearly supported duplicate is issue:45310/issue:45478, which share the same title and Qwen3.5 MoE from_pretrained failure. The remaining soft-edge rejections stay at the safe side of the evidence and do not overclaim duplication or mergeability." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 24, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:25251", + "issue:29127", + "issue:29942", + "issue:32090", + "issue:34689", + "issue:35141", + "issue:36010", + "issue:37428", + "issue:40444", + "issue:42915", + "issue:42994", + "issue:43065", + "issue:43066", + "issue:43278", + "issue:43381", + "issue:43412", + "issue:43526", + "issue:43550", + "issue:43653", + "issue:43701", + "issue:43873", + "issue:43927", + "issue:43976", + "issue:43994", + "issue:44186", + "issue:44361", + "issue:44448", + "issue:44451", + "issue:44462", + "issue:44466", + "issue:44493", + "issue:44743", + "issue:44857", + "issue:44871", + "issue:44977", + "issue:45083", + "issue:45198", + "issue:45245", + "issue:45372", + "issue:45412", + "issue:45440", + "issue:45459", + "issue:45520", + "issue:45563", + "issue:45588", + "issue:45698", + "issue:45803", + "issue:45823" + ], + "soft_pairs": [ + "issue:29127|issue:32090", + "issue:43927|issue:45698", + "issue:34689|issue:35141", + "issue:45198|issue:45803", + "issue:44977|issue:45563", + "issue:40444|issue:43526", + "issue:43701|issue:45083", + "issue:43550|issue:45083", + "issue:45372|issue:45588", + "issue:43412|issue:45588", + "issue:45245|issue:45563", + "issue:29942|issue:35141", + "issue:43873|issue:44493", + "issue:36010|issue:37428", + "issue:43066|issue:44462", + "issue:45372|issue:45459", + "issue:44857|issue:45588", + "issue:44448|issue:44466", + "issue:43065|issue:45412", + "issue:45588|issue:45823", + "issue:43278|issue:43381", + "issue:25251|issue:32090", + "issue:43653|issue:44743", + "issue:44451|issue:45520", + "issue:43976|issue:45440", + "issue:44186|issue:45198", + "issue:42915|issue:42994", + "issue:43994|issue:45412", + "issue:44361|issue:45198", + "issue:44871|issue:44977" + ], + "prepared_review_unit_hash": "0c711cf7c5b119f011fb4034dd9be5c0e0b579d01792d37d5929e3dc21fc3be9", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36619, + "estimated_input_tokens": 9155, + "estimated_eval_tokens": 18566 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is not a duplicate cluster; the soft links are mostly superficial lexical matches across unrelated bugs. I found no PRs and no convincing canonical duplicate issue.", + "confidence": 0.86, + "canonical_issue_reason": "No single issue is a reliable canonical representative here: the linked items span unrelated code paths (tokenizers, generation, flash-attention, quantization, loading, eval/training behavior) and do not describe the same underlying bug.", + "canonical_pr_reason": null, + "best_issue_reason": "If one issue must be selected as the closest thing to a representative, issue 45588 is the most concrete, reproducible core crash report in a central integration path; however, it is not a duplicate of the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:29127", + "right": "issue:32090", + "accept": false, + "reason": "Different subsystems and failures: LayoutLMv3 box validation/error messaging vs Trainer GPU broadcast rejecting NoneType." + }, + { + "left": "issue:43927", + "right": "issue:45698", + "accept": false, + "reason": "Both involve save/load behavior, but one is about custom token IDs being lost and the other about loading the wrong custom module; different root causes." + }, + { + "left": "issue:34689", + "right": "issue:35141", + "accept": false, + "reason": "Model-loading breakage for a specific vision model is unrelated to embedding reinitialization after resizing token embeddings." + }, + { + "left": "issue:45198", + "right": "issue:45803", + "accept": false, + "reason": "Both touch tokenization/saving, but the reported failures are on different models and different code paths." + }, + { + "left": "issue:44977", + "right": "issue:45563", + "accept": false, + "reason": "Flash-attention generation failure and a stale warning about num_return_sequences are unrelated bugs." + }, + { + "left": "issue:40444", + "right": "issue:43526", + "accept": false, + "reason": "IterableDataset multi-image prompt failure for Qwen2.5-VL is unrelated to BeitImageProcessorFast reduce_labels behavior." + }, + { + "left": "issue:43701", + "right": "issue:45083", + "accept": false, + "reason": "Checkpoint resume key mismatch is not the same bug as qwen3_omni_moe feature-length helper behavior." + }, + { + "left": "issue:43550", + "right": "issue:45083", + "accept": false, + "reason": "torch.compile with SDPA on Bamba-9B-v2 is a different concrete failure than qwen3_omni_moe feature-length calculation." + }, + { + "left": "issue:45372", + "right": "issue:45588", + "accept": false, + "reason": "Gemma 4 processor import failure due to missing ReasoningEffort is unrelated to flash_attention.py crashing on s_aux=None." + }, + { + "left": "issue:43412", + "right": "issue:45588", + "accept": false, + "reason": "Executorch export issues in gemma3n are a different path than sink-less flash-attention integration crashing on s_aux=None." + }, + { + "left": "issue:45245", + "right": "issue:45563", + "accept": false, + "reason": "Category cardinality overflow and a stale generation warning are unrelated." + }, + { + "left": "issue:29942", + "right": "issue:35141", + "accept": false, + "reason": "Flash Attention 2 test failures do not match an embedding reinitialization bug in post_init." + }, + { + "left": "issue:43873", + "right": "issue:44493", + "accept": false, + "reason": "Quantization offloading problems are unrelated to unexpected position-id key warnings." + }, + { + "left": "issue:36010", + "right": "issue:37428", + "accept": false, + "reason": "Two distinct ImportError regressions involving different symbols and modules." + }, + { + "left": "issue:43066", + "right": "issue:44462", + "accept": false, + "reason": "Wrong tokenizer decoder type in v5 is not the same as AutoTokenizer ignoring tokenizer.json from the repo." + }, + { + "left": "issue:45372", + "right": "issue:45459", + "accept": false, + "reason": "One is a missing import dependency for Gemma 4 processing; the other is exception masking when protobuf is absent." + }, + { + "left": "issue:44857", + "right": "issue:45588", + "accept": false, + "reason": "CUDA float16 loss crash in LwDetrImageLoss is unrelated to flash_attention.py's s_aux=None AttributeError." + }, + { + "left": "issue:44448", + "right": "issue:44466", + "accept": false, + "reason": "Different issues around model output/serialization across versions; they do not share the same underlying bug." + }, + { + "left": "issue:43065", + "right": "issue:45412", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder and RT-DETR memory release are separate model lifecycle issues." + }, + { + "left": "issue:45588", + "right": "issue:45823", + "accept": false, + "reason": "Both involve Gemma 4/flash-attention adjacent code, but one is a sink-less flash-attention crash and the other is a device mismatch in PLE forward; not the same bug." + }, + { + "left": "issue:43278", + "right": "issue:43381", + "accept": false, + "reason": "Training/eval dtype drift is not the same as gradient checkpointing being disallowed in eval mode." + }, + { + "left": "issue:25251", + "right": "issue:32090", + "accept": false, + "reason": "Pipeline top_k nesting behavior and Trainer NoneType GPU broadcast are unrelated." + }, + { + "left": "issue:43653", + "right": "issue:44743", + "accept": false, + "reason": "Tokenizer special-token registration causing empty decode is unrelated to recurrent states resetting with cache in qwen3_omni_moe." + }, + { + "left": "issue:44451", + "right": "issue:45520", + "accept": false, + "reason": "ScandiBERT loading failure and Python 3.13 flash_attn import key error are separate environment/model-loading problems." + }, + { + "left": "issue:43976", + "right": "issue:45440", + "accept": false, + "reason": "Python version compatibility and DeepseekV3MoE divergence are different issues." + }, + { + "left": "issue:44186", + "right": "issue:45198", + "accept": false, + "reason": "LayoutLMv2 NER/batched padding crashes are not the same as Wav2Vec2 save_pretrained/tokenization failures." + }, + { + "left": "issue:44361", + "right": "issue:45198", + "accept": false, + "reason": "MLukeTokenizer AttributeError on tasks is unrelated to Wav2Vec2 save/tokenization failure." + }, + { + "left": "issue:44871", + "right": "issue:44977", + "accept": false, + "reason": "EOS token configuration inconsistency is unrelated to flash-attention generation behavior in Qwen3.5." + }, + { + "left": "issue:45083", + "right": "issue:45588", + "accept": false, + "reason": "Feature-extraction length helper behavior in qwen3_omni_moe is unrelated to a flash_attention integration crash." + }, + { + "left": "issue:45588", + "right": "issue:45823", + "accept": false, + "reason": "Different Gemma 4 failures: one is an AttributeError in flash-attention integration, the other is a device placement mismatch during forward." + } + ] + }, + "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": 16, + "review_unit_index": 25, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:29127", + "issue:29942", + "issue:30064", + "issue:32090", + "issue:33357", + "issue:34567", + "issue:36010", + "issue:39692", + "issue:41720", + "issue:43065", + "issue:43278", + "issue:43329", + "issue:43334", + "issue:43381", + "issue:43421", + "issue:43526", + "issue:43531", + "issue:43575", + "issue:43653", + "issue:43701", + "issue:43723", + "issue:43756", + "issue:43761", + "issue:43824", + "issue:43873", + "issue:43994", + "issue:44186", + "issue:44206", + "issue:44265", + "issue:44279", + "issue:44442", + "issue:44479", + "issue:44554", + "issue:44610", + "issue:44734", + "issue:44977", + "issue:45103", + "issue:45198", + "issue:45229", + "issue:45245", + "issue:45372", + "issue:45406", + "issue:45412", + "issue:45440", + "issue:45491", + "issue:45593", + "issue:45834", + "issue:45859" + ], + "soft_pairs": [ + "issue:43421|issue:45491", + "issue:44206|issue:44479", + "issue:44265|issue:45198", + "issue:29942|issue:30064", + "issue:44734|issue:45406", + "issue:43065|issue:43994", + "issue:33357|issue:39692", + "issue:29127|issue:34567", + "issue:44610|issue:45412", + "issue:45103|issue:45372", + "issue:43756|issue:45440", + "issue:43873|issue:45229", + "issue:43723|issue:44554", + "issue:45372|issue:45406", + "issue:43278|issue:43575", + "issue:43761|issue:45859", + "issue:43526|issue:45834", + "issue:43381|issue:43701", + "issue:43653|issue:44442", + "issue:45440|issue:45593", + "issue:43278|issue:43329", + "issue:41720|issue:45229", + "issue:29942|issue:36010", + "issue:32090|issue:33357", + "issue:43334|issue:43531", + "issue:43526|issue:44186", + "issue:44977|issue:45245", + "issue:43824|issue:44279" + ], + "prepared_review_unit_hash": "7aa0c03a362c4c4c2027fc81afdbb50f3d44f6a435f8f890e4ce1dc92e4a2f01", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36197, + "estimated_input_tokens": 9050, + "estimated_eval_tokens": 18356 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is heterogeneous rather than a duplicate cluster. The soft pairs span distinct bugs across tokenizers, vision models, training, serving, and multimodal processors, so none should be merged as the same underlying issue.", + "confidence": 0.84, + "canonical_issue_reason": "Issue 43873 is the closest thing to a cluster anchor: it is open, broadly framed, and has the strongest discussion/inbound-reference footprint. It is still not a duplicate of the others, just the least-specific representative.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43873 is the best representative of the set because it is broad and active, with multiple inbound references, but the cluster does not contain a true duplicate family.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43421", + "right": "issue:45491", + "accept": false, + "reason": "Runtime special-token post-processing bug vs Gemma3 NaN embeddings/sliding-window attention; different code paths and symptoms." + }, + { + "left": "issue:44206", + "right": "issue:44479", + "accept": false, + "reason": "Unsupported `center` arg in a feature extractor vs Qwen VL video-input regression; unrelated subsystems." + }, + { + "left": "issue:44265", + "right": "issue:45198", + "accept": false, + "reason": "`torch.export` / `torch_compilable_check` failure vs Wav2Vec2 save/tokenization failure; not the same bug." + }, + { + "left": "issue:29942", + "right": "issue:30064", + "accept": false, + "reason": "Flash Attention 2 test failures vs void segmentation-map processing; different failure modes." + }, + { + "left": "issue:44734", + "right": "issue:45406", + "accept": false, + "reason": "KV-cache continuation tensor indexing bug in serve vs Gemma4 processor missing `_tokenizer`; unrelated." + }, + { + "left": "issue:43065", + "right": "issue:43994", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder vs SigLIP2 AutoModel/pipeline nonsense outputs; different models and defects." + }, + { + "left": "issue:33357", + "right": "issue:39692", + "accept": false, + "reason": "MacOS bus error on community CLIP model vs SigLIP2 docs/model-processor mismatch and quantization example failure; not one change." + }, + { + "left": "issue:29127", + "right": "issue:34567", + "accept": false, + "reason": "LayoutLMv3 error-message clarity issue vs TrainerState token counter not updating; unrelated features." + }, + { + "left": "issue:44610", + "right": "issue:45412", + "accept": false, + "reason": "OmDet-Turbo image-size mismatch vs RT-DETR memory release issue; distinct bugs." + }, + { + "left": "issue:45103", + "right": "issue:45372", + "accept": false, + "reason": "auto_docstring annotation handling vs Gemma4 processor import dependency failure; different layers entirely." + }, + { + "left": "issue:43756", + "right": "issue:45440", + "accept": false, + "reason": "Smollm3 RoPE-layer mismatch vs DeepSeekV3MoE divergence from remote impl; not the same code-path." + }, + { + "left": "issue:43873", + "right": "issue:45229", + "accept": false, + "reason": "General quantization/offloading behavior vs Gemma4 multi-GPU CUDA OOM; too broad and not the same defect." + }, + { + "left": "issue:43723", + "right": "issue:44554", + "accept": false, + "reason": "AutoTokenizer loading regression vs MPS attention correctness issue; unrelated." + }, + { + "left": "issue:45372", + "right": "issue:45406", + "accept": false, + "reason": "Processor-loading import error vs serve crash from missing `_tokenizer`; different surfaces and fixes." + }, + { + "left": "issue:43278", + "right": "issue:43575", + "accept": false, + "reason": "Training/eval dtype drift vs tensor-parallel OOM on Qwen2-57B; not the same problem." + }, + { + "left": "issue:43761", + "right": "issue:45859", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression vs CPU-offload incompatibility in Qwen3_5Moe rotary embedding; unrelated." + }, + { + "left": "issue:43526", + "right": "issue:45834", + "accept": false, + "reason": "BeitImageProcessorFast `reduce_labels` bug vs Kosmos2.5 long-OCR index error; different processors and failure modes." + }, + { + "left": "issue:43381", + "right": "issue:43701", + "accept": false, + "reason": "Gradient checkpointing in eval mode vs resume-from-checkpoint key mismatch; not the same bug." + }, + { + "left": "issue:43653", + "right": "issue:44442", + "accept": false, + "reason": "BigBirdTokenizer special-token registration/decoding bug vs AutoTokenizer loading FastSpeech2ConformerTokenizer; distinct tokenizer issues." + }, + { + "left": "issue:45440", + "right": "issue:45593", + "accept": false, + "reason": "DeepSeekV3MoE implementation divergence vs D-FINE auxiliary-loss behavior; different model logic." + }, + { + "left": "issue:43278", + "right": "issue:43329", + "accept": false, + "reason": "Embedding dtype mismatch vs undefined variables in multimodal-token counting video branch; unrelated." + }, + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "Qwen3 auto device-mapping assert vs Gemma4 multi-GPU OOM; different models and root causes." + }, + { + "left": "issue:29942", + "right": "issue:36010", + "accept": false, + "reason": "Flash Attention 2 tests vs `GenerationMixin` import error; no shared code path." + }, + { + "left": "issue:32090", + "right": "issue:33357", + "accept": false, + "reason": "Trainer NoneType broadcast error vs MacOS bus error on CLIP model; unrelated runtime failures." + }, + { + "left": "issue:43334", + "right": "issue:43531", + "accept": false, + "reason": "Qwen3-VL missing `pad_token_id` vs Qwen3-MoE sliding-window issue; both Qwen-family but different defects." + }, + { + "left": "issue:43526", + "right": "issue:44186", + "accept": false, + "reason": "Beit fast label-reduction bug vs LayoutLMv2Tokenizer NER/padding crash; different token/image processing logic." + }, + { + "left": "issue:44977", + "right": "issue:45245", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug vs category-count ceiling error; unrelated." + }, + { + "left": "issue:43824", + "right": "issue:44279", + "accept": false, + "reason": "Specific Qwen2_5_VL import failure vs generic dependency issue; too little evidence of 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": 2, + "component_count": 16, + "review_unit_index": 26, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:28218", + "issue:29127", + "issue:32090", + "issue:33290", + "issue:33357", + "issue:34567", + "issue:34689", + "issue:35141", + "issue:38175", + "issue:39692", + "issue:42907", + "issue:42915", + "issue:43262", + "issue:43316", + "issue:43381", + "issue:43441", + "issue:43525", + "issue:43526", + "issue:43575", + "issue:43611", + "issue:43646", + "issue:43701", + "issue:43761", + "issue:43873", + "issue:43931", + "issue:44077", + "issue:44079", + "issue:44188", + "issue:44206", + "issue:44387", + "issue:44442", + "issue:44464", + "issue:44479", + "issue:44743", + "issue:44811", + "issue:44857", + "issue:44898", + "issue:44964", + "issue:44977", + "issue:45072", + "issue:45216", + "issue:45290", + "issue:45440", + "issue:45491", + "issue:45584", + "issue:45588", + "issue:45698", + "issue:45803" + ], + "soft_pairs": [ + "issue:43526|issue:44857", + "issue:43262|issue:44206", + "issue:43441|issue:45588", + "issue:42915|issue:43931", + "issue:44442|issue:45072", + "issue:44464|issue:45491", + "issue:44743|issue:44811", + "issue:43575|issue:43701", + "issue:43646|issue:44077", + "issue:29127|issue:35141", + "issue:43575|issue:44387", + "issue:33357|issue:38175", + "issue:44977|issue:45584", + "issue:44442|issue:45803", + "issue:43316|issue:43525", + "issue:44964|issue:45440", + "issue:44479|issue:45290", + "issue:28218|issue:29127", + "issue:43761|issue:44079", + "issue:42907|issue:45698", + "issue:28218|issue:32090", + "issue:43931|issue:45216", + "issue:33290|issue:34567", + "issue:34689|issue:43611", + "issue:44442|issue:44898", + "issue:39692|issue:45491", + "issue:43761|issue:44188", + "issue:43381|issue:43873" + ], + "prepared_review_unit_hash": "036b88c991f5f8ba7a120b8f0b42fd14aa328fdafcc33624dfaf2d5d4602f421", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36120, + "estimated_input_tokens": 9030, + "estimated_eval_tokens": 18316 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are a heterogeneous mix of unrelated Transformers issues. None of the soft pairs appear to be the same underlying bug/change, so all soft edges should be rejected.", + "confidence": 0.92, + "canonical_issue_reason": "No single issue is clearly canonical for this set: the candidates span unrelated tokenizer, audio, vision, training, quantization, and model-loading problems.", + "canonical_pr_reason": null, + "best_issue_reason": "No global best issue stands out as a representative duplicate target because the cluster is not a single coherent problem area.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43526", + "right": "issue:44857", + "accept": false, + "reason": "Different bugs in different components: BeitImageProcessorFast label reduction vs AMP/CUDA crash in LwDetrImageLoss." + }, + { + "left": "issue:43262", + "right": "issue:44206", + "accept": false, + "reason": "Both are audio-related, but one is a sampling-rate default in apply_chat_template and the other is an unsupported argument passed by LasrFeatureExtractor." + }, + { + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "Both mention flash attention, but they involve different models and different failures: Ministral-3 vs sink-less models with s_aux=None." + }, + { + "left": "issue:42915", + "right": "issue:43931", + "accept": false, + "reason": "Different model/config issues: FineGrainedFP8Config failure for Qwen3Moe vs weight-shape mismatch loading Qwen3-VL-30B." + }, + { + "left": "issue:44442", + "right": "issue:45072", + "accept": false, + "reason": "Unrelated problems: tokenizer loading failure vs dtype mismatches in bfloat16 inference." + }, + { + "left": "issue:44464", + "right": "issue:45491", + "accept": false, + "reason": "Different generation/runtime bugs: compiled-forward chunked generation inconsistency vs NaNs from sliding-window attention on mixed-length batches." + }, + { + "left": "issue:44743", + "right": "issue:44811", + "accept": false, + "reason": "Different code paths and symptoms: Qwen3.5 recurrent state reset vs Whisper batch_decode skip_special_tokens handling." + }, + { + "left": "issue:43575", + "right": "issue:43701", + "accept": false, + "reason": "Different classes of failures: tensor-parallel OOM during model load vs resume_from_checkpoint key mismatch." + }, + { + "left": "issue:43646", + "right": "issue:44077", + "accept": false, + "reason": "Related to initialization, but not the same bug: generic custom model init regression vs patchtsmixer-specific post_init allowance." + }, + { + "left": "issue:29127", + "right": "issue:35141", + "accept": false, + "reason": "Different subsystems: LayoutLMv3 error-message clarity vs embedding reinitialization after resize when tie_word_embedding is False." + }, + { + "left": "issue:43575", + "right": "issue:44387", + "accept": false, + "reason": "Both are OOM reports, but one is TP loading OOM and the other is increased reserved memory under int4 quantization." + }, + { + "left": "issue:33357", + "right": "issue:38175", + "accept": false, + "reason": "Different models and failure modes: MacOS bus error on CLIP vs unexpected zero probabilities on SigLIP2." + }, + { + "left": "issue:44977", + "right": "issue:45584", + "accept": false, + "reason": "Both affect generation, but the causes are different: flash-attention behavior in Qwen3.5 vs empty-transcription handling after align_special_tokens in Whisper." + }, + { + "left": "issue:44442", + "right": "issue:45803", + "accept": false, + "reason": "Tokenizer loading failure vs a bare-except swallowing errors in FuyuBatchFeature; not the same bug." + }, + { + "left": "issue:43316", + "right": "issue:43525", + "accept": false, + "reason": "Both are config/API discrepancies, but they concern different models and different missing attributes." + }, + { + "left": "issue:44964", + "right": "issue:45440", + "accept": false, + "reason": "Different problems: loading Phi-4-multimodal-instruct fails, while DeepseekV3MoE diverges from the remote implementation." + }, + { + "left": "issue:44479", + "right": "issue:45290", + "accept": false, + "reason": "Different regressions: video-input handling for Qwen VL models vs chat-template crashes on assistant tool-call messages." + }, + { + "left": "issue:28218", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer spacing bug vs Trainer GPU broadcast TypeError; unrelated." + }, + { + "left": "issue:43761", + "right": "issue:44079", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression vs ModelOutput key-assignment bug; distinct code paths and symptoms." + }, + { + "left": "issue:42907", + "right": "issue:45698", + "accept": false, + "reason": "Quantization/offloading issue vs save/load custom-module mismatch; not the same underlying bug." + }, + { + "left": "issue:28218", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer spacing bug vs Trainer GPU broadcast TypeError; unrelated." + }, + { + "left": "issue:43931", + "right": "issue:45216", + "accept": false, + "reason": "Loading weight-shape mismatch in Qwen3-VL vs incorrect checkpoint saving in Qwen3.5; different stages of the model lifecycle." + }, + { + "left": "issue:33290", + "right": "issue:34567", + "accept": false, + "reason": "OOM with Adafactor in DeepSpeed vs TrainerState token counter not updating; unrelated." + }, + { + "left": "issue:34689", + "right": "issue:43611", + "accept": false, + "reason": "Both are loading regressions, but they target different versions and different causes." + }, + { + "left": "issue:44442", + "right": "issue:44898", + "accept": false, + "reason": "Tokenizer loading failure vs Perceiver non-default-resolution failure; unrelated." + }, + { + "left": "issue:39692", + "right": "issue:45491", + "accept": false, + "reason": "Documentation/example errors vs a runtime NaN bug; not the same change or bug." + }, + { + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "Different model/runtime issues: CLIPVisionModel hidden_states regression vs attention-kernel divergence under torch.compile." + }, + { + "left": "issue:43381", + "right": "issue:43873", + "accept": false, + "reason": "Gradient checkpointing in eval mode vs quantization offloading behavior; different defects." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 27, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:25251", + "issue:28282", + "issue:29127", + "issue:32090", + "issue:33290", + "issue:33357", + "issue:34689", + "issue:36010", + "issue:41628", + "issue:41950", + "issue:42175", + "issue:43232", + "issue:43295", + "issue:43316", + "issue:43381", + "issue:43388", + "issue:43404", + "issue:43441", + "issue:43475", + "issue:43493", + "issue:43540", + "issue:43550", + "issue:43643", + "issue:43720", + "issue:43742", + "issue:43825", + "issue:43827", + "issue:43856", + "issue:43873", + "issue:44077", + "issue:44162", + "issue:44190", + "issue:44442", + "issue:44479", + "issue:44496", + "issue:44514", + "issue:44568", + "issue:44743", + "issue:44792", + "issue:44855", + "issue:44871", + "issue:44877", + "issue:44913", + "issue:45137", + "issue:45372", + "issue:45479", + "issue:45656", + "issue:45759" + ], + "soft_pairs": [ + "issue:43493|issue:43827", + "issue:43550|issue:44743", + "issue:44568|issue:45479", + "issue:43316|issue:45759", + "issue:41628|issue:45372", + "issue:43720|issue:44442", + "issue:43643|issue:44913", + "issue:32090|issue:33290", + "issue:44190|issue:44496", + "issue:43873|issue:44871", + "issue:34689|issue:36010", + "issue:43381|issue:43856", + "issue:44855|issue:45137", + "issue:44077|issue:44479", + "issue:33290|issue:45656", + "issue:29127|issue:33357", + "issue:43381|issue:43742", + "issue:43540|issue:44514", + "issue:43316|issue:44877", + "issue:42175|issue:43316", + "issue:33290|issue:34689", + "issue:44162|issue:44792", + "issue:43404|issue:43441", + "issue:25251|issue:28282", + "issue:41950|issue:43475", + "issue:43295|issue:43825", + "issue:43232|issue:43388" + ], + "prepared_review_unit_hash": "e536a6e270e6323be3306990329e8df0772ab1a277bb102732135f22307e76d9", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 36089, + "estimated_input_tokens": 9023, + "estimated_eval_tokens": 18302 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is a broad mix of unrelated bug reports across pipelines, config loading, multimodal models, and distributed training. None of the soft pairs look like the same underlying bug/change, so I reject all proposed duplicate links.", + "confidence": 0.91, + "canonical_issue_reason": "issue:41950 is the only open item and is phrased as a concrete, reproducible pipeline regression, making it the most useful anchor in this set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:41950 is the clearest and most actionable issue here: specific failure mode, current open status, and a well-scoped title.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43493", + "right": "issue:43827", + "accept": false, + "reason": "Different problems: SigLIP2 implementation mismatch vs stale docs referencing removed pipelines." + }, + { + "left": "issue:43550", + "right": "issue:44743", + "accept": false, + "reason": "Different model/code paths and failures: torch.compile+SDPA issue vs cache/recurrent-state reset bug." + }, + { + "left": "issue:44568", + "right": "issue:45479", + "accept": false, + "reason": "Tokenizer special-token behavior and sequence-classification loss behavior are unrelated." + }, + { + "left": "issue:43316", + "right": "issue:45759", + "accept": false, + "reason": "Both are config/AutoModel-adjacent, but they concern different models and different failure mechanisms." + }, + { + "left": "issue:41628", + "right": "issue:45372", + "accept": false, + "reason": "Different import errors in different components; no shared code-path or symptom beyond import-time failure." + }, + { + "left": "issue:43720", + "right": "issue:44442", + "accept": false, + "reason": "BitNet weight-loading bug and tokenizer loading bug are unrelated subsystems." + }, + { + "left": "issue:43643", + "right": "issue:44913", + "accept": false, + "reason": "One is trust_remote_code returning incomplete config fields; the other is a GPTNeoX config persistence bug." + }, + { + "left": "issue:32090", + "right": "issue:33290", + "accept": false, + "reason": "Both involve Trainer/DeepSpeed, but one is a NoneType broadcast error and the other is an optimizer OOM." + }, + { + "left": "issue:44190", + "right": "issue:44496", + "accept": false, + "reason": "Dataset loading in a script and unrecognized model config loading are different issues." + }, + { + "left": "issue:43873", + "right": "issue:44871", + "accept": false, + "reason": "Quantization/offloading behavior and Gemma eos_token_id mismatch are unrelated." + }, + { + "left": "issue:34689", + "right": "issue:36010", + "accept": false, + "reason": "Model-loading regression for a specific model family vs a missing GenerationMixin import are different failures." + }, + { + "left": "issue:43381", + "right": "issue:43856", + "accept": false, + "reason": "Gradient checkpointing in eval mode and MoE training memory inefficiency are distinct problems." + }, + { + "left": "issue:44855", + "right": "issue:45137", + "accept": false, + "reason": "Python 3.13 import parsing failure vs DeepSpeed ZeRO3 deque underflow are unrelated." + }, + { + "left": "issue:44077", + "right": "issue:44479", + "accept": false, + "reason": "PatchTSMixer post_init policy and Qwen video-input regression do not share the same bug." + }, + { + "left": "issue:33290", + "right": "issue:45656", + "accept": false, + "reason": "Both mention DeepSpeed/optimizers, but OOM and double optimizer-step are different code-path failures." + }, + { + "left": "issue:29127", + "right": "issue:33357", + "accept": false, + "reason": "LayoutLMv3 error messaging issue and CLIP bus error on macOS are unrelated." + }, + { + "left": "issue:43381", + "right": "issue:43742", + "accept": false, + "reason": "Gradient checkpointing/eval issue is not the same as a MobileLLM key error on load." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Both are video/multimodal, but the failing inputs and processor code paths are different." + }, + { + "left": "issue:43316", + "right": "issue:44877", + "accept": false, + "reason": "Different config-loading bugs for different model families; not the same underlying defect." + }, + { + "left": "issue:42175", + "right": "issue:43316", + "accept": false, + "reason": "Dependency/backend packaging issue vs Gemma config API discrepancy are unrelated." + }, + { + "left": "issue:33290", + "right": "issue:34689", + "accept": false, + "reason": "DeepSpeed optimizer OOM and a model-loading regression are different failure modes." + }, + { + "left": "issue:44162", + "right": "issue:44792", + "accept": false, + "reason": "ESM2 model breakage and a janus test failure are not the same bug." + }, + { + "left": "issue:43404", + "right": "issue:43441", + "accept": false, + "reason": "Weight tying bug in one model class and FlashAttention failure in another are separate issues." + }, + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Pipeline top_k nesting bug is unrelated to missing PyTorch import error." + }, + { + "left": "issue:41950", + "right": "issue:43475", + "accept": false, + "reason": "Pipeline selecting image processors and SAM 3 video output attribute error are different model/pipeline bugs." + }, + { + "left": "issue:43295", + "right": "issue:43825", + "accept": false, + "reason": "Processor/tokenizer regression and pipeline error-message regression are different issues." + }, + { + "left": "issue:43232", + "right": "issue:43388", + "accept": false, + "reason": "Generation kwargs update after sync_gpus and metric gathering last-batch label dropping are unrelated distributed-training bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 28, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:33290", + "issue:35141", + "issue:36331", + "issue:41720", + "issue:42617", + "issue:43065", + "issue:43295", + "issue:43452", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43646", + "issue:43673", + "issue:43701", + "issue:43742", + "issue:43761", + "issue:43827", + "issue:43873", + "issue:43906", + "issue:43931", + "issue:43994", + "issue:44060", + "issue:44077", + "issue:44190", + "issue:44242", + "issue:44464", + "issue:44479", + "issue:44496", + "issue:44534", + "issue:44610", + "issue:44617", + "issue:44734", + "issue:44743", + "issue:44792", + "issue:44855", + "issue:44871", + "issue:44913", + "issue:44936", + "issue:44964", + "issue:44977", + "issue:45127", + "issue:45200", + "issue:45245", + "issue:45290", + "issue:45584", + "issue:45588", + "issue:45600", + "issue:45698" + ], + "soft_pairs": [ + "issue:44871|issue:45584", + "issue:45127|issue:45200", + "issue:43550|issue:43701", + "issue:43606|issue:43701", + "issue:43295|issue:44871", + "issue:43065|issue:44617", + "issue:43761|issue:44479", + "issue:33290|issue:35141", + "issue:33290|issue:36331", + "issue:43452|issue:44496", + "issue:43575|issue:43742", + "issue:43673|issue:44743", + "issue:43646|issue:44913", + "issue:42617|issue:43575", + "issue:44936|issue:45698", + "issue:43761|issue:44242", + "issue:43575|issue:43873", + "issue:44077|issue:44534", + "issue:43931|issue:44060", + "issue:44464|issue:44871", + "issue:44479|issue:44734", + "issue:44617|issue:45584", + "issue:43906|issue:45290", + "issue:44792|issue:44977", + "issue:43065|issue:44610", + "issue:43827|issue:43994", + "issue:44534|issue:44913", + "issue:44190|issue:44964", + "issue:41720|issue:45245", + "issue:44792|issue:45600", + "issue:44855|issue:45588" + ], + "prepared_review_unit_hash": "8add0b8968d643f55cc1a8945bde92962636969ddc51e8c974b4742a4183b4bb", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36968, + "estimated_input_tokens": 9242, + "estimated_eval_tokens": 18740 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is a grab bag of unrelated model-loading, training, generation, and regression reports. None of the soft pairs look like the same underlying bug/change closely enough to merge.", + "confidence": 0.92, + "canonical_issue_reason": "issue:43646 is the broadest, most central umbrella here: it describes a generic custom-model initialization breakage in Transformers 5.0.0, which is more representative than the many narrower, model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 is the best representative issue to keep because it is generic, user-facing, and broad enough to subsume closely related custom-model init/load regressions.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44871", + "right": "issue:45584", + "accept": false, + "reason": "Different bugs: Gemma eos token config mismatch vs Whisper empty-transcription generation failure." + }, + { + "left": "issue:45127", + "right": "issue:45200", + "accept": false, + "reason": "Both involve model behavior around tokens/embeddings, but one is LoRA merging collapse and the other is Gemma 4 multimodal token-type defaults." + }, + { + "left": "issue:43550", + "right": "issue:43701", + "accept": false, + "reason": "Unrelated failures: torch.compile+SDPA on Bamba versus resume_from_checkpoint key mismatch." + }, + { + "left": "issue:43606", + "right": "issue:43701", + "accept": false, + "reason": "CPU offload device mismatch is a different problem from checkpoint key mapping." + }, + { + "left": "issue:43295", + "right": "issue:44871", + "accept": false, + "reason": "One is custom model/processor API regression; the other is eos_token_id config inconsistency." + }, + { + "left": "issue:43065", + "right": "issue:44617", + "accept": false, + "reason": "Same family only at a high level (Sam3), but one is a dummy Conv2d concern and the other is a CUDA OOM in video inference." + }, + { + "left": "issue:43761", + "right": "issue:44479", + "accept": false, + "reason": "Different model paths and symptoms: CLIP hidden_states omission versus video-input regression in Qwen VL models." + }, + { + "left": "issue:33290", + "right": "issue:35141", + "accept": false, + "reason": "OOM in DeepSpeed Adafactor is unrelated to token-embedding reinitialization after resize." + }, + { + "left": "issue:33290", + "right": "issue:36331", + "accept": false, + "reason": "Optimizer memory blow-up is not the same as a Trainer compute_loss signature TypeError." + }, + { + "left": "issue:43452", + "right": "issue:44496", + "accept": false, + "reason": "Tokenizer/model GGUF loading breakage is unrelated to an unrecognized custom model config error." + }, + { + "left": "issue:43575", + "right": "issue:43742", + "accept": false, + "reason": "Both are loading failures, but one is TP memory pressure on a large model and the other is a key error on a small model." + }, + { + "left": "issue:43673", + "right": "issue:44743", + "accept": false, + "reason": "Both mention cache/state handling, but one is missing GenerationMixin cache in chunked prefill and the other is Qwen3.5 recurrent state reset." + }, + { + "left": "issue:43646", + "right": "issue:44913", + "accept": false, + "reason": "Generic custom-model initialization breakage is not the same as GPTNeoX rotary_pct not persisting on reload." + }, + { + "left": "issue:42617", + "right": "issue:43575", + "accept": false, + "reason": "3d_parallel.py execution failure is not the same underlying issue as TP loading OOM." + }, + { + "left": "issue:44936", + "right": "issue:45698", + "accept": false, + "reason": "Trainer evaluate-after-train failure differs from wrong custom module being loaded after save_pretrained." + }, + { + "left": "issue:43761", + "right": "issue:44242", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and load-balancing loss omission are separate code paths." + }, + { + "left": "issue:43575", + "right": "issue:43873", + "accept": false, + "reason": "OOM under tensor parallelism is different from offloading/quantization not behaving as expected." + }, + { + "left": "issue:44077", + "right": "issue:44534", + "accept": false, + "reason": "PatchTSMixer post_init allowance and non-persistent buffer junk are distinct initialization regressions." + }, + { + "left": "issue:43931", + "right": "issue:44060", + "accept": false, + "reason": "Weight-shape mismatch while loading Qwen3-VL is unrelated to a bad tied-weights warning in Qwen3-Next." + }, + { + "left": "issue:44464", + "right": "issue:44871", + "accept": false, + "reason": "Chunked generation inconsistency with compiled forward is unrelated to Gemma eos token config." + }, + { + "left": "issue:44479", + "right": "issue:44734", + "accept": false, + "reason": "Video input regression in Qwen VL models is different from KV-cache continuation indexing in serve /v1/responses." + }, + { + "left": "issue:44617", + "right": "issue:45584", + "accept": false, + "reason": "CUDA OOM in Sam3Video and Whisper empty-transcription failure are unrelated runtime issues." + }, + { + "left": "issue:43906", + "right": "issue:45290", + "accept": false, + "reason": "Isolated reproduction of issue 38071 is too vague and not clearly the same as tool-call chat template crashes." + }, + { + "left": "issue:44792", + "right": "issue:44977", + "accept": false, + "reason": "Janus image-generation test failure and Qwen3.5 flash-attention generation bug are different models and failure modes." + }, + { + "left": "issue:43065", + "right": "issue:44610", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder does not match OmDet-Turbo processor output-size mismatch." + }, + { + "left": "issue:43827", + "right": "issue:43994", + "accept": false, + "reason": "Docs still mentioning pipeline() is not the same as nonsensical outputs from AutoModel/pipeline on SigLIP2." + }, + { + "left": "issue:44534", + "right": "issue:44913", + "accept": false, + "reason": "Non-persistent buffer initialization junk and GPTNeoX rotary_pct reload mismatch are unrelated." + }, + { + "left": "issue:44190", + "right": "issue:44964", + "accept": false, + "reason": "Local dataset loading in a no-trainer script is a different bug from loading Phi-4 multimodal-instruct." + }, + { + "left": "issue:41720", + "right": "issue:45245", + "accept": false, + "reason": "Qwen3 auto-device-map cudaErrorAssert is unrelated to the 2^24 category limit." + }, + { + "left": "issue:44792", + "right": "issue:45600", + "accept": false, + "reason": "Janus generate-images test failure is unrelated to stale auto_mappings entries for removed Sam3LiteText configs." + }, + { + "left": "issue:44855", + "right": "issue:45588", + "accept": false, + "reason": "Python 3.13 TorchScript parsing failure in DebertaV2Model is unrelated to flash_attention s_aux=None crashes." + } + ] + }, + "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": 16, + "review_unit_index": 29, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:36246", + "issue:38175", + "issue:41628", + "issue:43065", + "issue:43116", + "issue:43295", + "issue:43335", + "issue:43404", + "issue:43479", + "issue:43540", + "issue:43550", + "issue:43644", + "issue:43650", + "issue:43976", + "issue:44077", + "issue:44112", + "issue:44186", + "issue:44190", + "issue:44220", + "issue:44464", + "issue:44466", + "issue:44496", + "issue:44521", + "issue:44561", + "issue:44683", + "issue:44734", + "issue:44737", + "issue:44855", + "issue:44877", + "issue:44913", + "issue:44933", + "issue:44938", + "issue:45127", + "issue:45216", + "issue:45245", + "issue:45276", + "issue:45357", + "issue:45362", + "issue:45375", + "issue:45381", + "issue:45446", + "issue:45542", + "issue:45563", + "issue:45584", + "issue:45600", + "issue:45632", + "issue:45803" + ], + "soft_pairs": [ + "issue:44737|issue:45803", + "issue:44521|issue:45381", + "issue:45127|issue:45357", + "issue:43335|issue:44855", + "issue:44938|issue:45542", + "issue:43976|issue:44855", + "issue:43550|issue:44190", + "issue:43644|issue:44913", + "issue:44077|issue:44877", + "issue:44496|issue:45803", + "issue:43540|issue:44734", + "issue:45276|issue:45584", + "issue:44112|issue:45600", + "issue:43650|issue:44220", + "issue:43335|issue:44190", + "issue:43116|issue:44190", + "issue:43295|issue:45276", + "issue:45245|issue:45584", + "issue:43479|issue:45375", + "issue:36246|issue:41628", + "issue:44933|issue:45362", + "issue:44186|issue:44734", + "issue:43404|issue:45216", + "issue:44464|issue:45563", + "issue:44683|issue:45446", + "issue:38175|issue:43065", + "issue:44734|issue:45632", + "issue:44561|issue:45632", + "issue:44466|issue:44913" + ], + "prepared_review_unit_hash": "5222244dc94fbace13dcf65f79e2713a592886ca484687a11a521c21985ce478", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35963, + "estimated_input_tokens": 8991, + "estimated_eval_tokens": 18238 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly noisy and heterogeneous: the soft-pair candidates do not appear to share the same underlying bug or change, so I reject all of them. The strongest issue in the set is the multimodal `apply_chat_template` bug because it is the most active, well-scoped, and well-referenced report.", + "confidence": 0.87, + "canonical_issue_reason": "issue:44521 is the most actionable and best-supported report in the set, with the highest discussion/inbound activity and a clear reproducible multimodal bug.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44521 is the best representative issue here because it is concrete, broadly useful, and has the strongest evidence trail among otherwise unrelated reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44737", + "right": "issue:45803", + "accept": false, + "reason": "Different subsystems and failure modes: XLNet CPU placement bug vs Fuyu tensor conversion exception handling." + }, + { + "left": "issue:44521", + "right": "issue:45381", + "accept": false, + "reason": "Both involve multimodal inputs, but one is assistant mask generation and the other is video position IDs in Qwen2.5-VL; root causes do not match." + }, + { + "left": "issue:45127", + "right": "issue:45357", + "accept": false, + "reason": "Distinct regression areas: LoRA merge/tied embeddings vs incorrect visual-encoder keys in `save_pretrained`." + }, + { + "left": "issue:43335", + "right": "issue:44855", + "accept": false, + "reason": "Unrelated models and bugs: SwitchTransformers config layer creation vs DebertaV2 import parsing on Python 3.13." + }, + { + "left": "issue:44938", + "right": "issue:45542", + "accept": false, + "reason": "Both are environment/load issues, but one is Python 3.14 import failure and the other is a missing TensorFlow backend fallback; not the same bug." + }, + { + "left": "issue:43976", + "right": "issue:44855", + "accept": false, + "reason": "Different Python-version regressions with unrelated mechanisms and code paths." + }, + { + "left": "issue:43550", + "right": "issue:44190", + "accept": false, + "reason": "Completely different domains: torch.compile SDPA model failure vs dataset loading in an example script." + }, + { + "left": "issue:43644", + "right": "issue:44913", + "accept": false, + "reason": "Both touch model state/config, but one is non-persistent buffer junk in v5 and the other is a GPTNeoX config reload issue." + }, + { + "left": "issue:44077", + "right": "issue:44877", + "accept": false, + "reason": "Both mention config handling, but one is about forbidding optional `post_init` and the other is strict config loading for a specific model." + }, + { + "left": "issue:44496", + "right": "issue:45803", + "accept": false, + "reason": "Unrelated: model-type/config loading error vs a bare `except` swallowing errors in Fuyu tensor conversion." + }, + { + "left": "issue:43540", + "right": "issue:44734", + "accept": false, + "reason": "Different code paths: Qwen3OmniMoe video preprocessing vs serve KV-cache continuation indexing." + }, + { + "left": "issue:45276", + "right": "issue:45584", + "accept": false, + "reason": "Different bugs in different models: Gemma4 embedding resize propagation vs Whisper empty transcription handling." + }, + { + "left": "issue:44112", + "right": "issue:45600", + "accept": false, + "reason": "Both are CI-related, but one is a stale test override and the other is stale autogenerated mappings; not the same root cause." + }, + { + "left": "issue:43650", + "right": "issue:44220", + "accept": false, + "reason": "No shared underlying bug; the titles are unrelated and provide no common code path." + }, + { + "left": "issue:43335", + "right": "issue:44190", + "accept": false, + "reason": "SwitchTransformers config initialization and example dataset loading are unrelated." + }, + { + "left": "issue:43116", + "right": "issue:44190", + "accept": false, + "reason": "Different example-script failures: multi-label classification output bug vs local dataset loading." + }, + { + "left": "issue:43295", + "right": "issue:45276", + "accept": false, + "reason": "Different multimodal/model-management regressions; no shared code path or fixable overlap." + }, + { + "left": "issue:45245", + "right": "issue:45584", + "accept": false, + "reason": "Completely unrelated error conditions: category cardinality limit vs Whisper transcription edge case." + }, + { + "left": "issue:43479", + "right": "issue:45375", + "accept": false, + "reason": "Both are config-schema issues, but they affect different model families and different missing-field behavior." + }, + { + "left": "issue:36246", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but they refer to different symbols and likely different export/registry problems." + }, + { + "left": "issue:44933", + "right": "issue:45362", + "accept": false, + "reason": "A missing import in `image_utils` is not the same as a chat crash in Qwen3.5." + }, + { + "left": "issue:44186", + "right": "issue:44734", + "accept": false, + "reason": "Tokenizer padding/truncation crash vs server KV-cache continuation indexing; distinct bugs." + }, + { + "left": "issue:43404", + "right": "issue:45216", + "accept": false, + "reason": "Both involve model weights, but one is tied `lm_head` in Mistral3 and the other is Qwen3.5 checkpoint saving." + }, + { + "left": "issue:44464", + "right": "issue:45563", + "accept": false, + "reason": "Generation inconsistency under compilation vs a stale warning about `num_return_sequences`; not the same issue." + }, + { + "left": "issue:44683", + "right": "issue:45446", + "accept": false, + "reason": "Both are in flex_attention, but one is a compiled-kernel failure on newer torch and the other is a version-check/import bug for AuxRequest." + }, + { + "left": "issue:38175", + "right": "issue:43065", + "accept": false, + "reason": "Different models and failure modes: zero probabilities in SigLIP2 vs a dummy Conv2d in Sam3 pixel decoder." + }, + { + "left": "issue:44734", + "right": "issue:45632", + "accept": false, + "reason": "Serve KV-cache indexing and trust_remote_code cache path collisions are unrelated." + }, + { + "left": "issue:44561", + "right": "issue:45632", + "accept": false, + "reason": "Different regression classes: removed helper breaking trust_remote_code vs cache-path collision for local models." + }, + { + "left": "issue:44466", + "right": "issue:44913", + "accept": false, + "reason": "Serialization of tied weights depending on device vs GPTNeoX config reload behavior; 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": 2, + "component_count": 16, + "review_unit_index": 30, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:36246", + "issue:38617", + "issue:41950", + "issue:42175", + "issue:42907", + "issue:43295", + "issue:43299", + "issue:43381", + "issue:43388", + "issue:43441", + "issue:43526", + "issue:43531", + "issue:43643", + "issue:43673", + "issue:43688", + "issue:43827", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44062", + "issue:44075", + "issue:44112", + "issue:44220", + "issue:44242", + "issue:44246", + "issue:44279", + "issue:44336", + "issue:44351", + "issue:44368", + "issue:44496", + "issue:44964", + "issue:45003", + "issue:45020", + "issue:45042", + "issue:45200", + "issue:45216", + "issue:45278", + "issue:45335", + "issue:45357", + "issue:45405", + "issue:45406", + "issue:45563", + "issue:45584", + "issue:45593", + "issue:45600", + "issue:45632", + "issue:45759" + ], + "soft_pairs": [ + "issue:44246|issue:45003", + "issue:36246|issue:38617", + "issue:43643|issue:44496", + "issue:45406|issue:45632", + "issue:45563|issue:45584", + "issue:41950|issue:44016", + "issue:44964|issue:45600", + "issue:43388|issue:43688", + "issue:42907|issue:43441", + "issue:44112|issue:44964", + "issue:43295|issue:45335", + "issue:43854|issue:44964", + "issue:44016|issue:44220", + "issue:43673|issue:44336", + "issue:44496|issue:45042", + "issue:43299|issue:45216", + "issue:43531|issue:45600", + "issue:43526|issue:43901", + "issue:42175|issue:44279", + "issue:45335|issue:45584", + "issue:44075|issue:44368", + "issue:43381|issue:45593", + "issue:43299|issue:45357", + "issue:43901|issue:44496", + "issue:44496|issue:45759", + "issue:44351|issue:45278", + "issue:44016|issue:44062", + "issue:44242|issue:45200", + "issue:43388|issue:43827", + "issue:41950|issue:45020", + "issue:43976|issue:45405" + ], + "prepared_review_unit_hash": "26caec4c4b1c2a0d3ce2f4f1f8c327e109c9fd30d7f9e5f97cee67eb504c3d71", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37045, + "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 mostly heterogeneous, but it has one plausible duplicate axis around post-upgrade import/export regressions in Transformers v5. Most other pairs are different bugs in different code paths and should stay separate.", + "confidence": 0.58, + "canonical_issue_reason": "issue:45278 is the broadest representative report in the set: it describes widespread import errors after upgrading to 5.5.0 and can plausibly subsume concrete missing-symbol cases like issue:44351.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the best umbrella issue because it captures the shared upgrade/regression theme with the widest scope and open discussion, making it the most useful canonical target for related import failures.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44246", + "right": "issue:45003", + "accept": false, + "reason": "Both are import-related, but one is a performance complaint about importing Transformers and the other is a specific sys.modules access bug; different symptoms and likely different fixes." + }, + { + "left": "issue:36246", + "right": "issue:38617", + "accept": false, + "reason": "Both are ImportError reports, but they concern different missing symbols from different modules, so they are not the same underlying bug." + }, + { + "left": "issue:43643", + "right": "issue:44496", + "accept": false, + "reason": "Both involve config/model-loading failures, but one is about trust_remote_code returning incomplete configs and the other is an unrecognized model/config issue; different root causes." + }, + { + "left": "issue:45406", + "right": "issue:45632", + "accept": false, + "reason": "Both mention trust_remote_code behavior, but one is a missing _tokenizer attribute on Gemma4Processor and the other is a cache-path collision for local models; unrelated bugs." + }, + { + "left": "issue:45563", + "right": "issue:45584", + "accept": false, + "reason": "Both are generation-related, but a stale warning on paged generate and a Whisper empty-transcription failure after token alignment are different code paths." + }, + { + "left": "issue:41950", + "right": "issue:44016", + "accept": false, + "reason": "A video-classification pipeline bug and a notebook syntax error are clearly unrelated." + }, + { + "left": "issue:44964", + "right": "issue:45600", + "accept": false, + "reason": "Both are model-loading/CI issues, but they affect different models and different failure modes; no shared concrete bug." + }, + { + "left": "issue:43388", + "right": "issue:43688", + "accept": false, + "reason": "These concern different training utilities: one drops labels in gather_for_metrics, the other misnormalizes auxiliary loss." + }, + { + "left": "issue:42907", + "right": "issue:43441", + "accept": false, + "reason": "Both are about Ministral/Devstral families, but saving dequantized weights and FlashAttention model loading are distinct problems." + }, + { + "left": "issue:44112", + "right": "issue:44964", + "accept": false, + "reason": "A stale GraniteSpeech CI test and a Phi-4 multimodal loading failure are unrelated." + }, + { + "left": "issue:43295", + "right": "issue:45335", + "accept": false, + "reason": "Both are regressions, but one is about processor.tokenizer/image handling in custom multimodal code and the other is resize_token_embeddings not updating decoder embeddings." + }, + { + "left": "issue:43854", + "right": "issue:44964", + "accept": false, + "reason": "Both are model-load failures, but they involve different models and different root causes." + }, + { + "left": "issue:44016", + "right": "issue:44220", + "accept": false, + "reason": "A notebook syntax error and an audio feature-extraction bug are unrelated." + }, + { + "left": "issue:43673", + "right": "issue:44336", + "accept": false, + "reason": "Chunked-prefill cache handling and ANSI coloring in loading_report are unrelated." + }, + { + "left": "issue:44496", + "right": "issue:45042", + "accept": false, + "reason": "An unrecognized OLMo model config and a PIL/torchvision dependency regression are different issues." + }, + { + "left": "issue:43299", + "right": "issue:45216", + "accept": false, + "reason": "Both involve Qwen3.5/Qwen3VL regressions, but one is a loading failure for MoE models and the other is an incorrect save_pretrained checkpoint layout; different code paths." + }, + { + "left": "issue:43531", + "right": "issue:45600", + "accept": false, + "reason": "Sliding_window behavior for Qwen3-MoE and removed auto_mappings references for Sam3LiteText are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:43901", + "accept": false, + "reason": "One is an image-processor label reduction bug, the other is a docs/behavior mismatch in text classification; not the same bug." + }, + { + "left": "issue:42175", + "right": "issue:44279", + "accept": false, + "reason": "A missing TensorFlow backend in the torch extra and a generic dependency issue are too vague and not the same concrete failure." + }, + { + "left": "issue:45335", + "right": "issue:45584", + "accept": false, + "reason": "These are different functional areas: embedding resize behavior vs Whisper generation after token alignment." + }, + { + "left": "issue:44075", + "right": "issue:44368", + "accept": false, + "reason": "Optimizer argument handling and a warning about tie_word_embeddings in Qwen fine-tuning are unrelated." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Gradient checkpointing in eval mode and missing auxiliary losses in D-FINE are different training bugs." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Both are Qwen3.5/Qwen3VL regressions, but the reported failures are different: loading vs incorrect save_pretrained output." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "TextClassificationPipeline docs/behavior and OLMo model loading have no shared underlying bug." + }, + { + "left": "issue:44351", + "right": "issue:45278", + "accept": true, + "reason": "These look like the same upgrade-induced import/export regression: 45278 is a broad report of many import errors after moving to 5.5.0, and 44351 is a concrete missing-symbol instance (HybridCache)." + }, + { + "left": "issue:44016", + "right": "issue:44062", + "accept": false, + "reason": "Notebook syntax and AddedToken keyword-argument errors are unrelated." + }, + { + "left": "issue:44242", + "right": "issue:45200", + "accept": false, + "reason": "Load-balancing loss gating and Gemma 4 mm_token_type_ids defaults are different model/training issues." + }, + { + "left": "issue:43388", + "right": "issue:43827", + "accept": false, + "reason": "Metric label truncation and outdated docs references are unrelated." + }, + { + "left": "issue:41950", + "right": "issue:45020", + "accept": false, + "reason": "A pipeline image-processor bug and broad remote_code breakage are both version regressions, but not the same concrete code-path problem." + }, + { + "left": "issue:43976", + "right": "issue:45405", + "accept": false, + "reason": "Python version compatibility and an unreleased peft version bump are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is grounded in the packet: it correctly treats the set as mostly heterogeneous while identifying a narrow import-regression axis centered on issue:45278. The one accepted soft pair, issue:44351 and issue:45278, is a conservative match on an upgrade-induced import failure theme. The other pairwise rejections are also suitably cautious and do not overclaim shared root causes." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 31, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:32090", + "issue:35141", + "issue:38175", + "issue:39692", + "issue:41950", + "issue:42994", + "issue:43065", + "issue:43295", + "issue:43299", + "issue:43317", + "issue:43441", + "issue:43450", + "issue:43479", + "issue:43525", + "issue:43577", + "issue:43650", + "issue:43723", + "issue:43824", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:44162", + "issue:44261", + "issue:44263", + "issue:44279", + "issue:44297", + "issue:44462", + "issue:44466", + "issue:44521", + "issue:44617", + "issue:44623", + "issue:44683", + "issue:44779", + "issue:44898", + "issue:44933", + "issue:44987", + "issue:45081", + "issue:45216", + "issue:45276", + "issue:45290", + "issue:45325", + "issue:45397", + "issue:45405", + "issue:45542", + "issue:45600", + "issue:45712", + "issue:45823", + "issue:45834" + ], + "soft_pairs": [ + "issue:43824|issue:44933", + "issue:43577|issue:45216", + "issue:43479|issue:43525", + "issue:41950|issue:45823", + "issue:43299|issue:43854", + "issue:44683|issue:45542", + "issue:44279|issue:45405", + "issue:38175|issue:44617", + "issue:39692|issue:43901", + "issue:32090|issue:35141", + "issue:43295|issue:44779", + "issue:44462|issue:45081", + "issue:44297|issue:44462", + "issue:41950|issue:43650", + "issue:43976|issue:44279", + "issue:43976|issue:44987", + "issue:45600|issue:45712", + "issue:44462|issue:44779", + "issue:44521|issue:45325", + "issue:43441|issue:45834", + "issue:43450|issue:44279", + "issue:42994|issue:43317", + "issue:44261|issue:44263", + "issue:43065|issue:44898", + "issue:44466|issue:45276", + "issue:43650|issue:45290", + "issue:43441|issue:44683", + "issue:43723|issue:44162", + "issue:44623|issue:45397" + ], + "prepared_review_unit_hash": "e91c99e1aadaa4b401e935290d8eb83732e108927e85e6ff954de4d07f8b55e7", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36323, + "estimated_input_tokens": 9081, + "estimated_eval_tokens": 18418 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose collection of Transformers regression reports around tokenizers/processors, multimodal loading, and save/load behavior, but the soft pairs generally describe different bugs rather than true duplicates.", + "confidence": 0.66, + "canonical_issue_reason": "issue:44462 is the most representative tokenizer-loading regression here: it is concrete, user-facing, and sits near several other tokenizer-related reports without being overly narrow.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44462 best captures the cluster\u2019s strongest recurring theme\u2014v5 tokenizer loading/selection regressions\u2014and has the clearest chance of serving as the umbrella issue for related tokenizer bugs.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43824", + "right": "issue:44933", + "accept": false, + "reason": "Both are import/load problems in the multimodal/image stack, but one is a missing Qwen2.5-VL symbol and the other is a nonexistent image_utils import; different concrete failures." + }, + { + "left": "issue:43577", + "right": "issue:45216", + "accept": false, + "reason": "Both involve save/load serialization, but BLIP2 dtype handling and Qwen3.5 checkpoint corruption are different bugs in different code paths." + }, + { + "left": "issue:43479", + "right": "issue:43525", + "accept": false, + "reason": "One is default multimodal config initialization from None, the other is a missing pad_token_id attribute; related configuration surface, not the same defect." + }, + { + "left": "issue:41950", + "right": "issue:45823", + "accept": false, + "reason": "Video-classification processor lookup and Gemma4 PLE device mismatch are unrelated failures despite both being pipeline/model issues." + }, + { + "left": "issue:43299", + "right": "issue:43854", + "accept": false, + "reason": "Both are model-loading failures, but they concern different models and different breakpoints." + }, + { + "left": "issue:44683", + "right": "issue:45542", + "accept": false, + "reason": "Compiled flex_attention on torch>=2.9 and tensorboard/TensorFlow backend detection are unrelated." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "A vague dependency issue and a specific PEFT version bump problem are not the same bug." + }, + { + "left": "issue:38175", + "right": "issue:44617", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 and CUDA OOM in Sam3Video are clearly different problems." + }, + { + "left": "issue:39692", + "right": "issue:43901", + "accept": false, + "reason": "SigLIP2 doc example errors and TextClassificationPipeline docs mismatch are documentation issues in different areas, not duplicates." + }, + { + "left": "issue:32090", + "right": "issue:35141", + "accept": false, + "reason": "Trainer GPU broadcast of NoneType and embedding reinitialization after resize_token_embeddings are distinct model/training bugs." + }, + { + "left": "issue:43295", + "right": "issue:44779", + "accept": false, + "reason": "A regression in custom model code using processor.tokenizer/images and Deepseek tokenizer output corruption are different tokenizer-facing failures." + }, + { + "left": "issue:44462", + "right": "issue:45081", + "accept": false, + "reason": "Ignoring tokenizer.json during AutoTokenizer loading is not the same as Mistral regex patch crashing on backend_tokenizer access." + }, + { + "left": "issue:44297", + "right": "issue:44462", + "accept": false, + "reason": "Tokenizer class metadata written by save_pretrained and repository tokenizer.json selection are related serialization/loading topics but separate bugs." + }, + { + "left": "issue:41950", + "right": "issue:43650", + "accept": false, + "reason": "Video-classification processor lookup and a nonspecific 'ADD THE DATA' issue are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Python version support for Transformers 5.1.0 and a dependency issue are not the same underlying defect." + }, + { + "left": "issue:43976", + "right": "issue:44987", + "accept": false, + "reason": "Python compatibility and loading physical-intelligence/fast are different issues." + }, + { + "left": "issue:45600", + "right": "issue:45712", + "accept": false, + "reason": "Broken auto_mappings references to Sam3LiteText configs and leftover dummy classes are different cleanup/CI problems." + }, + { + "left": "issue:44462", + "right": "issue:44779", + "accept": false, + "reason": "AutoTokenizer ignoring tokenizer.json and Deepseek tokenizer producing wrong results are both tokenizer regressions, but they affect different models and mechanisms." + }, + { + "left": "issue:44521", + "right": "issue:45325", + "accept": false, + "reason": "Assistant masks being all-zero in multimodal chat templates is unrelated to Qwen2.5-VL rope index scaling." + }, + { + "left": "issue:43441", + "right": "issue:45834", + "accept": false, + "reason": "FlashAttention failure in Ministral-3 and Kosmos2.5 OCR long-input index error are different model-specific runtime bugs." + }, + { + "left": "issue:43450", + "right": "issue:44279", + "accept": false, + "reason": "Incorrect batched video processor output shape and a generic dependency issue are unrelated." + }, + { + "left": "issue:42994", + "right": "issue:43317", + "accept": false, + "reason": "Quantized model saving failure and dequantized model loading with device_map=auto are different quantization lifecycle bugs." + }, + { + "left": "issue:44261", + "right": "issue:44263", + "accept": false, + "reason": "Missing rms_norm_eps in MLA q_a_layernorm and torch.split return values in GlmMoeDsaIndexer are distinct math/indexing problems." + }, + { + "left": "issue:43065", + "right": "issue:44898", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder and Perceiver non-default-resolution classification failure are unrelated vision-model bugs." + }, + { + "left": "issue:44466", + "right": "issue:45276", + "accept": false, + "reason": "Inconsistent lm_head.weight serialization and Gemma4 resize_token_embeddings not updating embeddings are both weight/resize issues, but they are different model families and code paths." + }, + { + "left": "issue:43650", + "right": "issue:45290", + "accept": false, + "reason": "A placeholder issue and chat template crashes with tool-call assistant messages are not duplicates." + }, + { + "left": "issue:43441", + "right": "issue:44683", + "accept": false, + "reason": "Ministral FlashAttention failure and compiled flex_attention failing on torch>=2.9 are both attention-related but clearly separate failure modes." + }, + { + "left": "issue:43723", + "right": "issue:44162", + "accept": false, + "reason": "Tokenizer loading regression in v5 and ESM2 broken workflow are unrelated." + }, + { + "left": "issue:44623", + "right": "issue:45397", + "accept": false, + "reason": "Missing processor.save_pretrained files and Gemma4 zero3 from_pretrained are different save/load issues." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 32, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:16998", + "issue:22355", + "issue:41628", + "issue:41950", + "issue:42491", + "issue:43122", + "issue:43262", + "issue:43299", + "issue:43352", + "issue:43450", + "issue:43525", + "issue:43643", + "issue:43650", + "issue:43653", + "issue:43673", + "issue:43761", + "issue:43784", + "issue:43881", + "issue:43901", + "issue:43976", + "issue:43994", + "issue:44016", + "issue:44188", + "issue:44220", + "issue:44242", + "issue:44246", + "issue:44291", + "issue:44295", + "issue:44297", + "issue:44336", + "issue:44361", + "issue:44373", + "issue:44496", + "issue:44521", + "issue:44589", + "issue:44617", + "issue:44623", + "issue:44671", + "issue:44749", + "issue:44779", + "issue:44987", + "issue:45250", + "issue:45362", + "issue:45397", + "issue:45542", + "issue:45593", + "issue:45823" + ], + "soft_pairs": [ + "issue:44623|issue:44749", + "issue:43653|issue:45593", + "issue:44671|issue:44779", + "issue:43784|issue:45542", + "issue:44016|issue:45823", + "issue:44361|issue:45593", + "issue:44987|issue:45542", + "issue:43901|issue:43994", + "issue:43122|issue:43650", + "issue:44016|issue:44291", + "issue:43262|issue:44521", + "issue:43673|issue:43761", + "issue:44188|issue:44336", + "issue:43525|issue:43881", + "issue:44297|issue:45397", + "issue:43450|issue:44617", + "issue:43525|issue:43643", + "issue:42491|issue:43299", + "issue:44589|issue:45362", + "issue:41628|issue:43976", + "issue:43901|issue:44246", + "issue:44336|issue:44373", + "issue:44242|issue:45593", + "issue:16998|issue:22355", + "issue:43352|issue:45250", + "issue:44295|issue:45397", + "issue:44246|issue:44496", + "issue:41950|issue:44220" + ], + "prepared_review_unit_hash": "d0520b907a9db231559c86ca2e314ba0215eb2d8ea5319e55ab533e30ddd6a50", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35153, + "estimated_input_tokens": 8789, + "estimated_eval_tokens": 17834 + }, + "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 across tokenization, model loading, processors, docs, and performance. None of the proposed soft pairs look like the same underlying bug/change; they are at best loosely similar by subsystem or version.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:44521 is the strongest standalone report: it describes a concrete, reproducible multimodal chat-template regression with broader user impact than the more niche or vague reports here.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44623", + "right": "issue:44749", + "accept": false, + "reason": "Processor save/load files regression vs a tokenization slowdown report; different symptoms and code paths." + }, + { + "left": "issue:43653", + "right": "issue:45593", + "accept": false, + "reason": "BigBirdTokenizer special-token registration bug vs D-FINE auxiliary-loss behavior; unrelated models and failures." + }, + { + "left": "issue:44671", + "right": "issue:44779", + "accept": false, + "reason": "Different tokenizer/model correctness bugs affecting different model families; not the same underlying defect." + }, + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "Import-time NameError in sentence-transformers integration vs TensorFlow backend detection issue; unrelated." + }, + { + "left": "issue:44016", + "right": "issue:45823", + "accept": false, + "reason": "Notebook syntax error vs Gemma4 device-mismatch runtime bug; completely different problem types." + }, + { + "left": "issue:44361", + "right": "issue:45593", + "accept": false, + "reason": "MLukeTokenizer task AttributeError vs D-FINE auxiliary-loss omission; unrelated subsystems." + }, + { + "left": "issue:44987", + "right": "issue:45542", + "accept": false, + "reason": "Model loading failure for a specific checkpoint vs backend-import error; no shared code-path evidence." + }, + { + "left": "issue:43901", + "right": "issue:43994", + "accept": false, + "reason": "Docs mismatch about return_all_scores vs a model inference quality bug; not the same issue." + }, + { + "left": "issue:43122", + "right": "issue:43650", + "accept": false, + "reason": "Tokenizer output changed across versions vs a nonsensical placeholder issue title; no technical overlap." + }, + { + "left": "issue:44016", + "right": "issue:44291", + "accept": false, + "reason": "Notebook syntax error vs init_empty_weights argument mismatch; unrelated." + }, + { + "left": "issue:43262", + "right": "issue:44521", + "accept": false, + "reason": "Both touch chat/processor behavior, but one is audio sampling-rate defaulting and the other is multimodal assistant-mask generation; different bugs." + }, + { + "left": "issue:43673", + "right": "issue:43761", + "accept": false, + "reason": "Generation cache/chunked_prefill regression vs CLIPVision hidden_states handling; unrelated models and code paths." + }, + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "torch.compile attention-kernel divergence vs ANSI escaping in loading_report; unrelated." + }, + { + "left": "issue:43525", + "right": "issue:43881", + "accept": false, + "reason": "Missing config attribute on Llama4Config vs GLM-4V loading failure; different model-specific loading issues." + }, + { + "left": "issue:44297", + "right": "issue:45397", + "accept": false, + "reason": "Tokenizer save_pretrained metadata mismatch vs Gemma-4 zero3 from_pretrained failure; unrelated." + }, + { + "left": "issue:43450", + "right": "issue:44617", + "accept": false, + "reason": "Incorrect batched video-processor shape vs CUDA OOM; one is a shape bug, the other is resource exhaustion." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": false, + "reason": "Missing pad_token_id on one config vs missing fields when trust_remote_code=True; both are config-related but not the same bug." + }, + { + "left": "issue:42491", + "right": "issue:43299", + "accept": false, + "reason": "Both involve Qwen3 MoE loading, but one is a LoRA compatibility break and the other is base-model loading in v5; too different to merge." + }, + { + "left": "issue:44589", + "right": "issue:45362", + "accept": false, + "reason": "Float8 storage deserialization error vs Qwen3.5 chat crash; no shared failure mode." + }, + { + "left": "issue:41628", + "right": "issue:43976", + "accept": false, + "reason": "Missing AutoImageProcessor export vs Python-version compatibility regression; unrelated." + }, + { + "left": "issue:43901", + "right": "issue:44246", + "accept": false, + "reason": "Documentation mismatch vs slow import performance; different issue classes." + }, + { + "left": "issue:44336", + "right": "issue:44373", + "accept": false, + "reason": "ANSI codes in loading_report vs wrong position_ids docstring; unrelated docs/utility issues." + }, + { + "left": "issue:44242", + "right": "issue:45593", + "accept": false, + "reason": "Load-balancing loss gating bug vs D-FINE auxiliary-loss omission; both are loss-related but not the same architecture or code path." + }, + { + "left": "issue:16998", + "right": "issue:22355", + "accept": false, + "reason": "Model_max_length question for DeBERTa vs missing transformers.onnx module import; unrelated." + }, + { + "left": "issue:43352", + "right": "issue:45250", + "accept": false, + "reason": "Specific model not supporting Flash Attention 2.0 yet vs a generic Flash Attention 2.0 issue; not the same concrete bug." + }, + { + "left": "issue:44295", + "right": "issue:45397", + "accept": false, + "reason": "position_ids buffer access error vs Gemma-4 zero3 loading failure; unrelated." + }, + { + "left": "issue:44246", + "right": "issue:44496", + "accept": false, + "reason": "Import performance complaint vs unrecognized model/config missing model_type; different failure modes." + }, + { + "left": "issue:41950", + "right": "issue:44220", + "accept": false, + "reason": "Video-classification pipeline using image processors vs _torch_extract_fbank_features() bug; both are pipeline/helper issues but not the same defect." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 33, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:36246", + "issue:42757", + "issue:43065", + "issue:43232", + "issue:43257", + "issue:43262", + "issue:43408", + "issue:43475", + "issue:43526", + "issue:43540", + "issue:43596", + "issue:43673", + "issue:43824", + "issue:43901", + "issue:43976", + "issue:44162", + "issue:44242", + "issue:44246", + "issue:44295", + "issue:44336", + "issue:44351", + "issue:44466", + "issue:44485", + "issue:44496", + "issue:44561", + "issue:44749", + "issue:44779", + "issue:44857", + "issue:44871", + "issue:44877", + "issue:44964", + "issue:44987", + "issue:44995", + "issue:45003", + "issue:45137", + "issue:45200", + "issue:45292", + "issue:45335", + "issue:45362", + "issue:45375", + "issue:45459", + "issue:45561", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45701", + "issue:45759" + ], + "soft_pairs": [ + "issue:36246|issue:42757", + "issue:36246|issue:44351", + "issue:44336|issue:44485", + "issue:43673|issue:44561", + "issue:43824|issue:43976", + "issue:43976|issue:44964", + "issue:44246|issue:44871", + "issue:44242|issue:44857", + "issue:45362|issue:45375", + "issue:44162|issue:45200", + "issue:44295|issue:44749", + "issue:44779|issue:45292", + "issue:44877|issue:45375", + "issue:45292|issue:45584", + "issue:45292|issue:45701", + "issue:43901|issue:45003", + "issue:45561|issue:45563", + "issue:44987|issue:45459", + "issue:36246|issue:43596", + "issue:43232|issue:44336", + "issue:43526|issue:45003", + "issue:45375|issue:45759", + "issue:44496|issue:45003", + "issue:43065|issue:43408", + "issue:43257|issue:45137", + "issue:44995|issue:45200", + "issue:43262|issue:44246", + "issue:43475|issue:43540", + "issue:44466|issue:45335", + "issue:44995|issue:45588" + ], + "prepared_review_unit_hash": "c6642a9653d9ee3307a10ba8b671e78efa09ee5e5fe99163974bc31e10133f7d", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36195, + "estimated_input_tokens": 9049, + "estimated_eval_tokens": 18354 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the paired issues share at most superficial wording (imports, generation, config, embeddings, or model-family names) but describe different failures and code paths. I would not merge any of the soft pairs as duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "No single issue is a true duplicate hub for this cluster; if one must be selected, issue 45292 is the closest generic core-library bug because it concerns a reusable modeling utility regression and has the broadest overlap with nearby embedding-related reports.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45292 is the best representative issue only in a loose sense: it is a concrete framework-level bug with a reusable API impact, whereas most other items are model-specific or unrelated. Still, the cluster overall is too mixed for a strong canonical choice.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36246", + "right": "issue:42757", + "accept": false, + "reason": "Both are ImportError reports, but for different missing symbols in different packages; unrelated failures." + }, + { + "left": "issue:36246", + "right": "issue:44351", + "accept": false, + "reason": "Qwen2.5-VL image processor import failure is unrelated to missing HybridCache export." + }, + { + "left": "issue:44336", + "right": "issue:44485", + "accept": false, + "reason": "ANSI color output in loading_report and GLM-5 RoPE implementation are unrelated subsystems and bugs." + }, + { + "left": "issue:43673", + "right": "issue:44561", + "accept": false, + "reason": "Chunked-prefill cache handling and removal of is_torch_fx_available are different generation/runtime issues." + }, + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "Missing Qwen2_5_VLForConditionalGeneration export is unrelated to Python version compatibility." + }, + { + "left": "issue:43976", + "right": "issue:44964", + "accept": false, + "reason": "Python 3.9/3.10 support mismatch is unrelated to loading Phi-4 multimodal." + }, + { + "left": "issue:44246", + "right": "issue:44871", + "accept": false, + "reason": "Slow imports and inconsistent eos_token_id configuration are different problems." + }, + { + "left": "issue:44242", + "right": "issue:44857", + "accept": false, + "reason": "MoE load-balancing loss logic and LwDetrImageLoss AMP crash do not share the same code path." + }, + { + "left": "issue:45362", + "right": "issue:45375", + "accept": false, + "reason": "Both mention Qwen3.5, but one is a chat crash and the other is a config field being dropped; not the same bug." + }, + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "ESM2 breakage and Gemma-4 mm_token_type_ids defaulting are unrelated model-specific issues." + }, + { + "left": "issue:44295", + "right": "issue:44749", + "accept": false, + "reason": "Position_ids buffer access error and preprocessing slowdown after upgrade are different failures." + }, + { + "left": "issue:44779", + "right": "issue:45292", + "accept": false, + "reason": "Tokenizer output changes and resize_token_embeddings not updating output embeddings are distinct regressions." + }, + { + "left": "issue:44877", + "right": "issue:45375", + "accept": false, + "reason": "Strict config rejection for granite_speech is unrelated to the Qwen3_5MoeVisionConfig field-drop bug." + }, + { + "left": "issue:45292", + "right": "issue:45584", + "accept": false, + "reason": "Embedding resize propagation and Whisper empty-transcription generation failure are not the same underlying bug." + }, + { + "left": "issue:45292", + "right": "issue:45701", + "accept": false, + "reason": "resize_token_embeddings behavior and general tokenization changes across versions are not the same code-path problem." + }, + { + "left": "issue:43901", + "right": "issue:45003", + "accept": false, + "reason": "Pipeline docs mismatch and unsafe sys.modules access are unrelated." + }, + { + "left": "issue:45561", + "right": "issue:45563", + "accept": false, + "reason": "xdist race on captured_info.txt and stale warning in paged generate are separate testing vs generation issues." + }, + { + "left": "issue:44987", + "right": "issue:45459", + "accept": false, + "reason": "Model loading failure with transformers>=5.1.0 and tokenizer error masking without protobuf are different bugs." + }, + { + "left": "issue:36246", + "right": "issue:43596", + "accept": false, + "reason": "Qwen2.5-VL import error is unrelated to the DeepSpeed ZeRO3 BertModel index error." + }, + { + "left": "issue:43232", + "right": "issue:44336", + "accept": false, + "reason": "Generation kwargs update after sync_gpus and ANSI codes in loading_report are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:45003", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug is unrelated to sys.modules access in modeling_utils." + }, + { + "left": "issue:45375", + "right": "issue:45759", + "accept": false, + "reason": "Both are about Qwen multimodal configs, but one drops a new field under strict config and the other fails to unwrap text_config from composite configs." + }, + { + "left": "issue:44496", + "right": "issue:45003", + "accept": false, + "reason": "Unrecognized model/config loading and unsafe sys.modules access are different load-time issues." + }, + { + "left": "issue:43065", + "right": "issue:43408", + "accept": false, + "reason": "Dummy conv in Sam3PixelDecoder and a sam3_video/sam3_tracker warning are not the same defect." + }, + { + "left": "issue:43257", + "right": "issue:45137", + "accept": false, + "reason": "Qwen3 MOE weight conversion under accelerate+deepspeed and ZeRO3 deque underflow are different DeepSpeed problems." + }, + { + "left": "issue:44995", + "right": "issue:45200", + "accept": false, + "reason": "GlmMoeDsa stale indexer cache crash and Gemma-4 token-id defaults are unrelated." + }, + { + "left": "issue:43262", + "right": "issue:44246", + "accept": false, + "reason": "Audio chat-template sampling rate behavior and import slowness are unrelated." + }, + { + "left": "issue:43475", + "right": "issue:43540", + "accept": false, + "reason": "Missing fpn_position_embeddings in SAM3 video output and Qwen3OmniMoe video-input ValueError are different bugs." + }, + { + "left": "issue:44466", + "right": "issue:45335", + "accept": false, + "reason": "Inconsistent serialization of tied weights and resize_token_embeddings not reaching decoder.embed_tokens are different weight-handling issues." + }, + { + "left": "issue:44995", + "right": "issue:45588", + "accept": false, + "reason": "Stale indexer cache in GlmMoeDsa and flash_attention AttributeError on s_aux=None are different runtime paths." + } + ] + }, + "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": 16, + "review_unit_index": 34, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:28282", + "issue:30990", + "issue:41628", + "issue:42371", + "issue:42886", + "issue:42994", + "issue:43232", + "issue:43452", + "issue:43475", + "issue:43482", + "issue:43525", + "issue:43643", + "issue:43646", + "issue:43653", + "issue:43673", + "issue:43827", + "issue:43931", + "issue:44038", + "issue:44060", + "issue:44077", + "issue:44162", + "issue:44297", + "issue:44361", + "issue:44368", + "issue:44623", + "issue:44734", + "issue:44749", + "issue:44869", + "issue:44933", + "issue:44977", + "issue:45003", + "issue:45081", + "issue:45092", + "issue:45103", + "issue:45137", + "issue:45229", + "issue:45230", + "issue:45310", + "issue:45356", + "issue:45375", + "issue:45381", + "issue:45397", + "issue:45459", + "issue:45478", + "issue:45561", + "issue:45563", + "issue:45600", + "issue:45656" + ], + "soft_pairs": [ + "issue:42886|issue:45229", + "issue:43653|issue:44162", + "issue:44368|issue:45563", + "issue:44368|issue:44977", + "issue:45397|issue:45656", + "issue:44060|issue:44077", + "issue:45003|issue:45561", + "issue:43673|issue:44734", + "issue:45137|issue:45600", + "issue:43931|issue:44368", + "issue:44933|issue:45459", + "issue:44869|issue:45356", + "issue:44297|issue:44623", + "issue:43646|issue:44162", + "issue:44361|issue:44749", + "issue:43232|issue:43673", + "issue:43673|issue:43827", + "issue:43452|issue:45003", + "issue:44623|issue:45656", + "issue:28282|issue:30990", + "issue:42994|issue:43482", + "issue:44749|issue:45310", + "issue:44749|issue:45656", + "issue:42371|issue:45092", + "issue:43475|issue:43643", + "issue:43475|issue:45381", + "issue:45230|issue:45397", + "issue:44038|issue:44623", + "issue:41628|issue:44933", + "issue:45081|issue:45103", + "issue:45375|issue:45478", + "issue:42994|issue:43525" + ], + "prepared_review_unit_hash": "852c1418e25fa469c9f2c60fe559407a6afeb135920adbe61dd15aa30fe64976", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37218, + "estimated_input_tokens": 9305, + "estimated_eval_tokens": 18866 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: most items are distinct Transformers regressions, model-specific loading bugs, or docs/tests issues rather than duplicates. No soft edge looks like the same underlying bug/change.", + "confidence": 0.71, + "canonical_issue_reason": "issue:43646 is the broadest cross-cutting regression here: Transformers 5.0 breaking custom model initialization. It is the best anchor among otherwise unrelated, model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 is the most representative and general issue in the set, with the widest apparent impact on downstream custom models.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42886", + "right": "issue:45229", + "accept": false, + "reason": "Offline tokenizer cache failure and Gemma4 multi-GPU OOM are unrelated bugs." + }, + { + "left": "issue:43653", + "right": "issue:44162", + "accept": false, + "reason": "BigBirdTokenizer special-token decode bug vs. an ESM2 breakage; different code paths and failures." + }, + { + "left": "issue:44368", + "right": "issue:45563", + "accept": false, + "reason": "Both mention warnings, but one is a Qwen3.5 LoRA tie-weights warning and the other is a stale generate warning for num_return_sequences." + }, + { + "left": "issue:44368", + "right": "issue:44977", + "accept": false, + "reason": "Qwen3.5 tie-weights warning and flash-attention generation failure are distinct issues." + }, + { + "left": "issue:45397", + "right": "issue:45656", + "accept": false, + "reason": "Gemma-4 ZeRO3 from_pretrained failure and a deepspeed double-step bug are not the same bug." + }, + { + "left": "issue:44060", + "right": "issue:44077", + "accept": false, + "reason": "Qwen3-Next tied-weights warning and patchtsmixer post_init compatibility are unrelated." + }, + { + "left": "issue:45003", + "right": "issue:45561", + "accept": false, + "reason": "Unsafe sys.modules access and pytest-xdist race on captured_info.txt are different failures." + }, + { + "left": "issue:43673", + "right": "issue:44734", + "accept": false, + "reason": "Both are generation/cache-related, but one is a chunked_prefill cache regression and the other is a serve KV-cache indexing crash." + }, + { + "left": "issue:45137", + "right": "issue:45600", + "accept": false, + "reason": "DeepSpeed ZeRO3 deque pop error and a stale auto_mappings reference are unrelated." + }, + { + "left": "issue:43931", + "right": "issue:44368", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch and Qwen3.5 tie-word-embeddings warning are different bugs." + }, + { + "left": "issue:44933", + "right": "issue:45459", + "accept": false, + "reason": "Missing image_utils import and tokenizer error masking when protobuf is absent are separate issues." + }, + { + "left": "issue:44869", + "right": "issue:45356", + "accept": false, + "reason": "Whisper word-timestamp decode crash and Kimi-K2.5 tokenizer codec regression are not the same underlying bug." + }, + { + "left": "issue:44297", + "right": "issue:44623", + "accept": false, + "reason": "Tokenizer save metadata mismatch and processor save missing files affect different save_pretrained paths." + }, + { + "left": "issue:43646", + "right": "issue:44162", + "accept": false, + "reason": "Generic custom model initialization breakage is broader, but ESM2 is a separate model-specific regression." + }, + { + "left": "issue:44361", + "right": "issue:44749", + "accept": false, + "reason": "MLukeTokenizer AttributeError and a multilingual text slowdown regression are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:43673", + "accept": false, + "reason": "Different generation internals: sync_gpus update bug vs. chunked_prefill cache regression." + }, + { + "left": "issue:43673", + "right": "issue:43827", + "accept": false, + "reason": "Generation cache regression and docs still referencing removed pipeline() API are unrelated." + }, + { + "left": "issue:43452", + "right": "issue:45003", + "accept": false, + "reason": "GGUF loading failures and unsafe sys.modules access are different code paths." + }, + { + "left": "issue:44623", + "right": "issue:45656", + "accept": false, + "reason": "Processor save regression and deepspeed double optimizer step are unrelated." + }, + { + "left": "issue:28282", + "right": "issue:30990", + "accept": false, + "reason": "Missing PyTorch import error and Sentence Transformers loading hang are not the same bug." + }, + { + "left": "issue:42994", + "right": "issue:43482", + "accept": false, + "reason": "Quantized model saving failure and GGUF loading failure are separate quantization-related issues." + }, + { + "left": "issue:44749", + "right": "issue:45310", + "accept": false, + "reason": "A slowdown regression after upgrade and a Qwen3.5 Moe from_pretrained error are different failures." + }, + { + "left": "issue:44749", + "right": "issue:45656", + "accept": false, + "reason": "Performance regression in data filtering and deepspeed optimizer stepping are unrelated." + }, + { + "left": "issue:42371", + "right": "issue:45092", + "accept": false, + "reason": "TF32 settings guidance and remote-code meta-initialization incompatibility are unrelated." + }, + { + "left": "issue:43475", + "right": "issue:43643", + "accept": false, + "reason": "SAM3 output attribute error and missing fields in trust_remote_code AutoConfig are different bugs." + }, + { + "left": "issue:43475", + "right": "issue:45381", + "accept": false, + "reason": "SAM3 vision encoder missing attribute and qwen2.5-vl video position IDs are different model-specific issues." + }, + { + "left": "issue:45230", + "right": "issue:45397", + "accept": false, + "reason": "Generic bug report placeholder and Gemma-4 ZeRO3 from_pretrained failure are not duplicates." + }, + { + "left": "issue:44038", + "right": "issue:44623", + "accept": false, + "reason": "Qwen3-VL-Moe v5 breakage and processor.save_pretrained missing files are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:44933", + "accept": false, + "reason": "AutoImageProcessor import failure and a missing image_utils import are similar only superficially; they are different symbols and failure points." + }, + { + "left": "issue:45081", + "right": "issue:45103", + "accept": false, + "reason": "Mistral regex/tokenizer backend crash and auto-docstring future-annotations crash are unrelated." + }, + { + "left": "issue:45375", + "right": "issue:45478", + "accept": false, + "reason": "Missing config field in Qwen3_5MoeVisionConfig and Qwen3.5 Moe from_pretrained error are not the same concrete bug." + }, + { + "left": "issue:42994", + "right": "issue:43525", + "accept": false, + "reason": "Quantized save failure and Llama4Config missing pad_token_id 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": 2, + "component_count": 16, + "review_unit_index": 35, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "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:43475", + "issue:43704", + "issue:43723", + "issue:43784", + "issue:43824", + "issue:43874", + "issue:43901", + "issue:44038", + "issue:44060", + "issue:44188", + "issue:44222", + "issue:44242", + "issue:44351", + "issue:44373", + "issue:44496", + "issue:44568", + "issue:44623", + "issue:44661", + "issue:44734", + "issue:44792", + "issue:44964", + "issue:45092", + "issue:45103", + "issue:45125", + "issue:45137", + "issue:45200", + "issue:45310", + "issue:45356", + "issue:45375", + "issue:45561", + "issue:45600", + "issue:45712", + "issue:45812" + ], + "soft_pairs": [ + "issue:42673|issue:43704", + "issue:43475|issue:44661", + "issue:45310|issue:45375", + "issue:41628|issue:44351", + "issue:42831|issue:44222", + "issue:42222|issue:45200", + "issue:44222|issue:45310", + "issue:42994|issue:43824", + "issue:42175|issue:45712", + "issue:36010|issue:41628", + "issue:43377|issue:44242", + "issue:44038|issue:45310", + "issue:44496|issue:45561", + "issue:44964|issue:45137", + "issue:45561|issue:45600", + "issue:43901|issue:45561", + "issue:43232|issue:44792", + "issue:42898|issue:44568", + "issue:30990|issue:35141", + "issue:43232|issue:44188", + "issue:42898|issue:45812", + "issue:44188|issue:44373", + "issue:43784|issue:43824", + "issue:44623|issue:45310", + "issue:43278|issue:45137", + "issue:44734|issue:45103", + "issue:30333|issue:33290", + "issue:43723|issue:44568", + "issue:42898|issue:45356", + "issue:44060|issue:45125", + "issue:43874|issue:45092" + ], + "prepared_review_unit_hash": "cd2428df64605287eabe893b99733196faf47804324479cd473e957e01530f0c", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37036, + "estimated_input_tokens": 9259, + "estimated_eval_tokens": 18774 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly a loose collection of unrelated Transformers issues; the only clear duplicate pair is the two identical Qwen3ForCausalLM VRAM-leak reports. Everything else looks like adjacent but distinct bugs, regressions, or docs/test issues.", + "confidence": 0.88, + "canonical_issue_reason": "issue:42673 is the best canonical issue because it is the older of the only exact duplicate pair and the titles/body indicate the same concrete bug as issue:43704.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:42673 is the best overall representative because it is the clearest duplicate target in the cluster and anchors the only strong same-bug pairing.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42673", + "right": "issue:43704", + "accept": true, + "reason": "Exact same title and same reported VRAM leak in Qwen3ForCausalLM with multiple dataloader threads; these are the same underlying bug." + }, + { + "left": "issue:43475", + "right": "issue:44661", + "accept": false, + "reason": "Different failures: one is a Sam3VisionEncoderOutput attribute error, the other is an auto-mapping/config handling failure for tokenizer mapping names." + }, + { + "left": "issue:45310", + "right": "issue:45375", + "accept": false, + "reason": "Both involve Qwen3.5 MoE, but one is a from_pretrained error while the other is a missing config field dropped by strict loading; not the same code-path." + }, + { + "left": "issue:41628", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different symbols and likely different export paths; not the same missing-import bug." + }, + { + "left": "issue:42831", + "right": "issue:44222", + "accept": false, + "reason": "Accuracy with FineGrainedFP8 and FP8 save_pretrained for moe are different problems affecting different code paths." + }, + { + "left": "issue:42222", + "right": "issue:45200", + "accept": false, + "reason": "VitPose breakage and Gemma 4 mm_token_type_ids defaults are unrelated model-specific bugs." + }, + { + "left": "issue:44222", + "right": "issue:45310", + "accept": false, + "reason": "Shared MoE/FP8 context is too broad; one is save_pretrained behavior, the other is a from_pretrained error in a different path." + }, + { + "left": "issue:42994", + "right": "issue:43824", + "accept": false, + "reason": "Quantized model saving failure and missing Qwen2_5_VLForConditionalGeneration import are unrelated issues." + }, + { + "left": "issue:42175", + "right": "issue:45712", + "accept": false, + "reason": "TensorFlow backend packaging and leftover dummy classes are different maintenance/packaging problems." + }, + { + "left": "issue:36010", + "right": "issue:41628", + "accept": false, + "reason": "Both are import-name errors, but for different classes from different modules; not the same missing export." + }, + { + "left": "issue:43377", + "right": "issue:44242", + "accept": false, + "reason": "MIMI padding-mask batching bug and MoE load-balancing loss omission are unrelated model behaviors." + }, + { + "left": "issue:44038", + "right": "issue:45310", + "accept": false, + "reason": "Different Qwen model families and different failure modes; too broad to be the same underlying bug." + }, + { + "left": "issue:44496", + "right": "issue:45561", + "accept": false, + "reason": "Unrecognized model loading and xdist captured-info race are unrelated." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Phi-4 multimodal load failure and DeepSpeed ZeRO3 deque error are unrelated." + }, + { + "left": "issue:45561", + "right": "issue:45600", + "accept": false, + "reason": "A pytest-xdist race on captured_info.txt is not the same as stale auto_mappings references to removed configs." + }, + { + "left": "issue:43901", + "right": "issue:45561", + "accept": false, + "reason": "Documentation mismatch for return_all_scores and xdist worker file race are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:44792", + "accept": false, + "reason": "Generation KV/sync_gpus handling and Janus image-generation test failure are different issues." + }, + { + "left": "issue:42898", + "right": "issue:44568", + "accept": false, + "reason": "Both concern tokenizer behavior in v5, but one is cleanup spacing semantics while the other is BOS/EOS insertion for a specific tokenizer." + }, + { + "left": "issue:30990", + "right": "issue:35141", + "accept": false, + "reason": "Sentence Transformers loading stall and token embedding resize reinitialization are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:44188", + "accept": false, + "reason": "Generation kwargs update after sync_gpus and attention-kernel divergence under torch.compile are different code paths." + }, + { + "left": "issue:42898", + "right": "issue:45812", + "accept": false, + "reason": "Different tokenizer regressions affecting different behaviors and models." + }, + { + "left": "issue:44188", + "right": "issue:44373", + "accept": false, + "reason": "Attention kernel branching bug and docstring typo/mismatch for position_ids are unrelated." + }, + { + "left": "issue:43784", + "right": "issue:43824", + "accept": false, + "reason": "Sentence-transformers import NameError and missing Qwen2.5-VL class import are different import failures." + }, + { + "left": "issue:44623", + "right": "issue:45310", + "accept": false, + "reason": "Processor save_pretrained file emission and Qwen3.5 MoE from_pretrained error are separate failures." + }, + { + "left": "issue:43278", + "right": "issue:45137", + "accept": false, + "reason": "Embedding dtype drift in evaluate and DeepSpeed ZeRO3 deque underflow are unrelated." + }, + { + "left": "issue:44734", + "right": "issue:45103", + "accept": false, + "reason": "KV cache continuation crash in serve and auto_docstring crash with future annotations are different subsystems." + }, + { + "left": "issue:30333", + "right": "issue:33290", + "accept": false, + "reason": "MLflow reporting on failed training jobs and OOM with adafactor in deepspeed are unrelated." + }, + { + "left": "issue:43723", + "right": "issue:44568", + "accept": false, + "reason": "General tokenizer loading issue and a specific add_special_tokens/BOS-EOS regression are not the same bug." + }, + { + "left": "issue:42898", + "right": "issue:45356", + "accept": false, + "reason": "Different tokenizer regressions: cleanup spacing behavior versus codec handling and warning text for Kimi-K2.5." + }, + { + "left": "issue:44060", + "right": "issue:45125", + "accept": false, + "reason": "Incorrect tied-weights warning and missing tensor-parallel plan are unrelated Qwen3.5/Moe issues." + }, + { + "left": "issue:43874", + "right": "issue:45092", + "accept": false, + "reason": "Missing image-patch counting method and meta-init incompatibility for InternVL2 are different multimodal loading problems." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The writeup is grounded in the packet: the only clearly supported duplicate pair is issue:42673 \u2194 issue:43704, which do have identical titles and matching VRAM-leak descriptions. The remaining pair judgments stay conservative and do not overclaim equivalence based on broad thematic similarity alone. " + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 36, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:28282", + "issue:30333", + "issue:34689", + "issue:35141", + "issue:36331", + "issue:41628", + "issue:42175", + "issue:42757", + "issue:42831", + "issue:42898", + "issue:43232", + "issue:43479", + "issue:43644", + "issue:43723", + "issue:43873", + "issue:43976", + "issue:44079", + "issue:44188", + "issue:44206", + "issue:44222", + "issue:44261", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44360", + "issue:44373", + "issue:44448", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44514", + "issue:44623", + "issue:44779", + "issue:44869", + "issue:44871", + "issue:44933", + "issue:44945", + "issue:44964", + "issue:44977", + "issue:45092", + "issue:45103", + "issue:45310", + "issue:45357", + "issue:45405", + "issue:45406", + "issue:45656", + "issue:45712", + "issue:45834" + ], + "soft_pairs": [ + "issue:42898|issue:44779", + "issue:45405|issue:45712", + "issue:42898|issue:44448", + "issue:44188|issue:44964", + "issue:44295|issue:45656", + "issue:43479|issue:44079", + "issue:44933|issue:45712", + "issue:42175|issue:44933", + "issue:44222|issue:44297", + "issue:43976|issue:44933", + "issue:43479|issue:44869", + "issue:44297|issue:45656", + "issue:44206|issue:45103", + "issue:44222|issue:44623", + "issue:42831|issue:44623", + "issue:41628|issue:42757", + "issue:30333|issue:36331", + "issue:30333|issue:34689", + "issue:43232|issue:45834", + "issue:43723|issue:44977", + "issue:44222|issue:44484", + "issue:44297|issue:45310", + "issue:43873|issue:44945", + "issue:28282|issue:30333", + "issue:30333|issue:35141", + "issue:44263|issue:45092", + "issue:44263|issue:44360", + "issue:44483|issue:45406", + "issue:44373|issue:44485", + "issue:44514|issue:45103", + "issue:44295|issue:44623", + "issue:43479|issue:43644", + "issue:44261|issue:45092", + "issue:43723|issue:44871", + "issue:45092|issue:45357" + ], + "prepared_review_unit_hash": "34e8d5637beb7c0e9b9c38ec411cf49564020d061a571554407fa99967276cbb", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38020, + "estimated_input_tokens": 9505, + "estimated_eval_tokens": 19266 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose topical grouping of unrelated Transformers issues (tokenizers, save/load, model loading, training, backend imports, docs, and assorted regressions). None of the soft pairs look like the same underlying defect, so I reject all merges.", + "confidence": 0.82, + "canonical_issue_reason": "issue:44297 is the most concrete, self-contained bug report in the set (tokenizer.save_pretrained writes a mismatched tokenizer_class), so it is a reasonable anchor if one issue must represent the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44297 is the clearest actionable report here: specific failure mode, specific API, and easier to triage than the broader version-wide or import-related regressions.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42898", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer-related, but one is a v5 cleanup-space behavior change and the other is a Deepseek tokenization regression; different code paths and symptoms." + }, + { + "left": "issue:45405", + "right": "issue:45712", + "accept": false, + "reason": "Unrelated: a dependency version bump vs leftover dummy classes leaking into dir() and repo checks." + }, + { + "left": "issue:42898", + "right": "issue:44448", + "accept": false, + "reason": "One is a tokenizer cleanup-space behavior change; the other is a model output drift report for Pegasus. Same broad area, not the same bug." + }, + { + "left": "issue:44188", + "right": "issue:44964", + "accept": false, + "reason": "Attention-kernel branching under torch.compile vs a Phi-4 multimodal model-loading failure are different defects." + }, + { + "left": "issue:44295", + "right": "issue:45656", + "accept": false, + "reason": "A position_ids buffer-read issue is unrelated to deepspeed calling optimizer.step() twice." + }, + { + "left": "issue:43479", + "right": "issue:44079", + "accept": false, + "reason": "Config initialization bug for Phi4MultimodalConfig vs ModelOutput key assignment bug; distinct objects and failure modes." + }, + { + "left": "issue:44933", + "right": "issue:45712", + "accept": false, + "reason": "Import from image_utils vs leftover dummy class leakage; different modules and errors." + }, + { + "left": "issue:42175", + "right": "issue:44933", + "accept": false, + "reason": "Packaging/backend omission with pip install .[torch] vs a missing/nonexistent image_utils import." + }, + { + "left": "issue:44222", + "right": "issue:44297", + "accept": false, + "reason": "Both mention save_pretrained, but one is FP8 MoE persistence and the other is tokenizer config metadata; not the same bug." + }, + { + "left": "issue:43976", + "right": "issue:44933", + "accept": false, + "reason": "Python 3.9/3.10 compatibility regression vs an image_utils import failure; unrelated." + }, + { + "left": "issue:43479", + "right": "issue:44869", + "accept": false, + "reason": "Config default initialization vs Whisper word-timestamp decode crash; different subsystems." + }, + { + "left": "issue:44297", + "right": "issue:45656", + "accept": false, + "reason": "Tokenizer save metadata mismatch vs deepspeed optimizer double-step; no shared code path." + }, + { + "left": "issue:44206", + "right": "issue:45103", + "accept": false, + "reason": "Feature extractor center-arg regression vs auto_docstring annotation crash; unrelated areas." + }, + { + "left": "issue:44222", + "right": "issue:44623", + "accept": false, + "reason": "FP8 save_pretrained on MoE models vs processor.save_pretrained missing files; same broad save API, different concrete defects." + }, + { + "left": "issue:42831", + "right": "issue:44623", + "accept": false, + "reason": "FP8 accuracy problem vs missing processor files on save; not the same failure." + }, + { + "left": "issue:41628", + "right": "issue:42757", + "accept": false, + "reason": "Different missing imports: AutoImageProcessor vs is_offline_mode." + }, + { + "left": "issue:30333", + "right": "issue:36331", + "accept": false, + "reason": "MLflow reporting after failed training jobs vs a Trainer compute_loss signature mismatch." + }, + { + "left": "issue:30333", + "right": "issue:34689", + "accept": false, + "reason": "Training-job reporting/MLflow issue vs Llama 3.2 vision model-loading regression." + }, + { + "left": "issue:43232", + "right": "issue:45834", + "accept": false, + "reason": "Generation kwarg handling after sync_gpus vs long OCR input index error; different models and code paths." + }, + { + "left": "issue:43723", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer loading regression in v5 vs Qwen3.5 generation issues with flash-attention." + }, + { + "left": "issue:44222", + "right": "issue:44484", + "accept": false, + "reason": "FP8 save_pretrained bug vs a question about max_shard_size in save_pretrained; not the same defect." + }, + { + "left": "issue:44297", + "right": "issue:45310", + "accept": false, + "reason": "Tokenizer save_pretrained metadata mismatch vs Qwen3.5 MoE from_pretrained regression; different layers of the stack." + }, + { + "left": "issue:43873", + "right": "issue:44945", + "accept": false, + "reason": "Quantized offloading behavior vs incorrect pipeline-parallel output; related broad area, but distinct bugs." + }, + { + "left": "issue:28282", + "right": "issue:30333", + "accept": false, + "reason": "PyTorch-missing import error vs MLflow job-reporting problem." + }, + { + "left": "issue:30333", + "right": "issue:35141", + "accept": false, + "reason": "MLflow/reporting issue vs embedding reinitialization after resizing token embeddings." + }, + { + "left": "issue:44263", + "right": "issue:45092", + "accept": false, + "reason": "torch.split return-value bug in a DSA indexer vs InternVL2 meta-init incompatibility; unrelated." + }, + { + "left": "issue:44263", + "right": "issue:44360", + "accept": false, + "reason": "Both touch the DSA indexer, but one is a torch.split shape/return issue and the other is a missing ReLU; same subsystem, different bugs." + }, + { + "left": "issue:44483", + "right": "issue:45406", + "accept": false, + "reason": "Chat/completions API rejection vs Gemma4Processor missing _tokenizer; distinct failures." + }, + { + "left": "issue:44373", + "right": "issue:44485", + "accept": false, + "reason": "Wrong docstring for position_ids vs GLM-5 RoPE implementation; one is documentation, the other is model math." + }, + { + "left": "issue:44514", + "right": "issue:45103", + "accept": false, + "reason": "Batched chat-template crash vs auto_docstring annotation crash; unrelated functionality." + }, + { + "left": "issue:44295", + "right": "issue:44623", + "accept": false, + "reason": "Reading a registered buffer issue vs processor.save_pretrained missing files; different APIs and symptoms." + }, + { + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Phi4Multimodal config defaulting bug vs non-persistent buffers being filled with junk in Transformers 5.0.0." + }, + { + "left": "issue:44261", + "right": "issue:45092", + "accept": false, + "reason": "MLA q_a_layernorm precision/config bug vs old InternVL2 remote-code meta-init incompatibility; not the same underlying defect." + }, + { + "left": "issue:43723", + "right": "issue:44871", + "accept": false, + "reason": "Tokenizer loading failure in v5 vs Gemma-3 eos_token_id configuration mismatch." + }, + { + "left": "issue:45092", + "right": "issue:45357", + "accept": false, + "reason": "InternVL2 meta-init compatibility bug vs Qwen3.5 save_pretrained writing incorrect visual encoder keys." + } + ] + }, + "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": 16, + "review_unit_index": 37, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:22355", + "issue:28282", + "issue:30990", + "issue:33357", + "issue:38175", + "issue:39401", + "issue:41628", + "issue:42371", + "issue:42757", + "issue:42831", + "issue:42886", + "issue:42907", + "issue:42994", + "issue:43066", + "issue:43122", + "issue:43232", + "issue:43408", + "issue:43475", + "issue:43479", + "issue:43576", + "issue:43723", + "issue:43756", + "issue:43784", + "issue:43824", + "issue:43937", + "issue:44038", + "issue:44060", + "issue:44117", + "issue:44261", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44373", + "issue:44442", + "issue:44514", + "issue:44568", + "issue:44704", + "issue:44779", + "issue:44913", + "issue:44945", + "issue:45245", + "issue:45278", + "issue:45290", + "issue:45310", + "issue:45593", + "issue:45600", + "issue:45753" + ], + "soft_pairs": [ + "issue:30990|issue:38175", + "issue:41628|issue:43824", + "issue:39401|issue:44568", + "issue:43479|issue:44913", + "issue:22355|issue:28282", + "issue:42831|issue:45310", + "issue:42886|issue:42994", + "issue:44263|issue:44373", + "issue:44945|issue:45753", + "issue:42757|issue:45278", + "issue:42831|issue:44038", + "issue:43576|issue:43784", + "issue:42886|issue:43066", + "issue:43232|issue:45593", + "issue:42371|issue:44263", + "issue:43756|issue:44060", + "issue:44117|issue:44704", + "issue:30990|issue:33357", + "issue:42907|issue:44945", + "issue:44704|issue:45290", + "issue:43122|issue:44514", + "issue:44295|issue:44297", + "issue:41628|issue:45278", + "issue:43723|issue:44779", + "issue:43723|issue:44442", + "issue:43784|issue:45600", + "issue:42907|issue:43937", + "issue:43232|issue:45245", + "issue:44261|issue:44373", + "issue:43408|issue:43475" + ], + "prepared_review_unit_hash": "ff07c3af84ba9bd3a9fd71850e6018ee92f5b36966b4109f03d7dfbc1e3cf667", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 35655, + "estimated_input_tokens": 8914, + "estimated_eval_tokens": 18084 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose collection of mostly unrelated bug reports, with some superficial similarity around imports/tokenizers/model-specific regressions but no single duplicate family.", + "confidence": 0.95, + "canonical_issue_reason": "No clear canonical issue: the items span many distinct subsystems and failure modes, so they do not form one duplicate cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "If one issue must be highlighted, issue 45278 is the broadest umbrella for upgrade-related import errors, but it still does not represent the whole mixed cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:30990", + "right": "issue:38175", + "accept": false, + "reason": "Both are model-loading / inference complaints, but they affect different models and different symptoms." + }, + { + "left": "issue:41628", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but for different missing symbols and likely different code paths." + }, + { + "left": "issue:39401", + "right": "issue:44568", + "accept": false, + "reason": "Both are tokenizer issues, but one is offset mapping and the other is special-token insertion behavior." + }, + { + "left": "issue:43479", + "right": "issue:44913", + "accept": false, + "reason": "Different config persistence bugs in different model classes; not the same defect." + }, + { + "left": "issue:22355", + "right": "issue:28282", + "accept": false, + "reason": "One is a missing transformers.onnx module, the other is an AutoModel PyTorch dependency error." + }, + { + "left": "issue:42831", + "right": "issue:45310", + "accept": false, + "reason": "One is an FP8 accuracy issue, the other is a Qwen3.5 MoE from_pretrained failure." + }, + { + "left": "issue:42886", + "right": "issue:42994", + "accept": false, + "reason": "Offline tokenizer cache loading and quantized model saving are unrelated code paths." + }, + { + "left": "issue:44263", + "right": "issue:44373", + "accept": false, + "reason": "A tensor-splitting bug and a docstring issue are not the same underlying problem." + }, + { + "left": "issue:44945", + "right": "issue:45753", + "accept": false, + "reason": "Pipeline-parallel output corruption and a Qwen3.5 image loop are different model/runtime failures." + }, + { + "left": "issue:42757", + "right": "issue:45278", + "accept": false, + "reason": "The first is one missing import from huggingface_hub; the second is a broad report of many import errors, but not the same concrete bug." + }, + { + "left": "issue:42831", + "right": "issue:44038", + "accept": false, + "reason": "Different model- and version-specific issues with no shared failing code path." + }, + { + "left": "issue:43576", + "right": "issue:43784", + "accept": false, + "reason": "Broken CLI env command and a NameError during sentence-transformers import are different failures." + }, + { + "left": "issue:42886", + "right": "issue:43066", + "accept": false, + "reason": "Offline cache loading and tokenizer decoder type mismatch are unrelated tokenizer/config bugs." + }, + { + "left": "issue:43232", + "right": "issue:45593", + "accept": false, + "reason": "Generation kwargs handling and D-FINE auxiliary loss behavior are different subsystems." + }, + { + "left": "issue:42371", + "right": "issue:44263", + "accept": false, + "reason": "TF32 configuration guidance and torch.split return values are unrelated." + }, + { + "left": "issue:43756", + "right": "issue:44060", + "accept": false, + "reason": "Different model architecture/config bugs; same broad family only." + }, + { + "left": "issue:44117", + "right": "issue:44704", + "accept": false, + "reason": "Tokenizer mapping fallback and AutoProcessor cached_file kwargs are different loader issues." + }, + { + "left": "issue:30990", + "right": "issue:33357", + "accept": false, + "reason": "Both mention loading/inference, but one is a Sentence Transformers hang and the other is a macOS bus error with a CLIP model." + }, + { + "left": "issue:42907", + "right": "issue:44945", + "accept": false, + "reason": "Quantized model saving and pipeline-parallel output errors are unrelated." + }, + { + "left": "issue:44704", + "right": "issue:45290", + "accept": false, + "reason": "AutoProcessor kwargs forwarding and chat template tool-call crashing are distinct behaviors." + }, + { + "left": "issue:43122", + "right": "issue:44514", + "accept": false, + "reason": "Tokenizer output changes and processor chat-template batching failures do not indicate the same bug." + }, + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "Reading a registered buffer and saving tokenizer metadata are different serialization issues." + }, + { + "left": "issue:41628", + "right": "issue:45278", + "accept": false, + "reason": "One specific missing import is not enough to treat it as the same defect as a broad multi-import regression report." + }, + { + "left": "issue:43723", + "right": "issue:44779", + "accept": false, + "reason": "Both involve tokenizer loading/behavior, but they target different failure modes and likely different tokenizer implementations." + }, + { + "left": "issue:43723", + "right": "issue:44442", + "accept": false, + "reason": "AutoTokenizer loading failure and a specific FastSpeech2ConformerTokenizer failure are not the same issue." + }, + { + "left": "issue:43784", + "right": "issue:45600", + "accept": false, + "reason": "Sentence-transformers import NameError and stale auto_mappings references break for different reasons." + }, + { + "left": "issue:42907", + "right": "issue:43937", + "accept": false, + "reason": "Model saving and invalid GenerationConfig validation are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:45245", + "accept": false, + "reason": "Generation sync behavior and category-count limits are different code paths." + }, + { + "left": "issue:44261", + "right": "issue:44373", + "accept": false, + "reason": "A precision/config bug and a wrong docstring are not duplicates." + }, + { + "left": "issue:43408", + "right": "issue:43475", + "accept": false, + "reason": "Both touch SAM 3 video, but one is a model-type warning and the other is a missing attribute crash; not clearly 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": 2, + "component_count": 16, + "review_unit_index": 38, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:22355", + "issue:28218", + "issue:29127", + "issue:30990", + "issue:36296", + "issue:39401", + "issue:42222", + "issue:42371", + "issue:42757", + "issue:42831", + "issue:42907", + "issue:43066", + "issue:43116", + "issue:43122", + "issue:43441", + "issue:43452", + "issue:43493", + "issue:43784", + "issue:43824", + "issue:43825", + "issue:43874", + "issue:43937", + "issue:43976", + "issue:43992", + "issue:44038", + "issue:44075", + "issue:44222", + "issue:44295", + "issue:44297", + "issue:44351", + "issue:44360", + "issue:44485", + "issue:44488", + "issue:44521", + "issue:44625", + "issue:44704", + "issue:44779", + "issue:44829", + "issue:44936", + "issue:44964", + "issue:44977", + "issue:45083", + "issue:45200", + "issue:45216", + "issue:45446", + "issue:45478", + "issue:45753" + ], + "soft_pairs": [ + "issue:44360|issue:44485", + "issue:43122|issue:44521", + "issue:44625|issue:44829", + "issue:43116|issue:45200", + "issue:29127|issue:30990", + "issue:42907|issue:44964", + "issue:43122|issue:44779", + "issue:44038|issue:44297", + "issue:42907|issue:43452", + "issue:39401|issue:44488", + "issue:44038|issue:44295", + "issue:43824|issue:44351", + "issue:36296|issue:43116", + "issue:44038|issue:44222", + "issue:39401|issue:44521", + "issue:42757|issue:44351", + "issue:45216|issue:45478", + "issue:43066|issue:44779", + "issue:22355|issue:28218", + "issue:42907|issue:44222", + "issue:43122|issue:44977", + "issue:43992|issue:44704", + "issue:43784|issue:45446", + "issue:43937|issue:44485", + "issue:42222|issue:43441", + "issue:42371|issue:43874", + "issue:43976|issue:45446", + "issue:43493|issue:45753", + "issue:44075|issue:44936", + "issue:42831|issue:45753", + "issue:43116|issue:43825", + "issue:44038|issue:45478", + "issue:44222|issue:45478", + "issue:43116|issue:43493", + "issue:45083|issue:45446" + ], + "prepared_review_unit_hash": "deee57f29e32d9952ddadc204d11b6ac04eca99e944ccf1262d4287023215377", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 35, + "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": "Mostly false-positive similarity matches: the items span unrelated tokenizer, model-loading, training, export, and config bugs, so there is no cohesive duplicate cluster.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue fits well; the cluster is heterogeneous and the soft links appear to connect unrelated bugs rather than one underlying defect.", + "canonical_pr_reason": null, + "best_issue_reason": "If forced to pick a representative, issue:43122 is the broadest tokenizer-regression report, but it still does not subsume the rest of the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "Unrelated DSA/indexer discussion versus GLM-5 RoPE implementation." + }, + { + "left": "issue:43122", + "right": "issue:44521", + "accept": false, + "reason": "Tokenizer regression and multimodal assistant-mask bug are different code paths." + }, + { + "left": "issue:44625", + "right": "issue:44829", + "accept": false, + "reason": "num_labels propagation issue is unrelated to flash_attention_3 training degeneration." + }, + { + "left": "issue:43116", + "right": "issue:45200", + "accept": false, + "reason": "Example-script classification output bug is unrelated to mm_token_type_ids defaults." + }, + { + "left": "issue:29127", + "right": "issue:30990", + "accept": false, + "reason": "LayoutLMv3 error messaging and Sentence Transformers loading hang are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:44964", + "accept": false, + "reason": "Save-dequantized failure and Phi-4 multimodal load failure are different problems." + }, + { + "left": "issue:43122", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenization regressions, but on different models with different symptoms and code paths." + }, + { + "left": "issue:44038", + "right": "issue:44297", + "accept": false, + "reason": "Qwen3-VL-Moe loading failure is unrelated to tokenizer_class metadata saved by save_pretrained." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Dequantized-save bug and gguf_file/pretrained loading bug are not the same defect." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "offset_mapping bug and model loading failure are unrelated." + }, + { + "left": "issue:44038", + "right": "issue:44295", + "accept": false, + "reason": "Qwen3-VL-Moe loading issue and position_ids buffer access error are unrelated." + }, + { + "left": "issue:43824", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different symbols and different causes." + }, + { + "left": "issue:36296", + "right": "issue:43116", + "accept": false, + "reason": "Tensor-parallel training bug is unrelated to the example script's empty multi-label results." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "Model loading failure and FP8 save_pretrained bug affect different code paths." + }, + { + "left": "issue:39401", + "right": "issue:44521", + "accept": false, + "reason": "Tokenizer offset_mapping and multimodal assistant_masks are unrelated tokenizer features." + }, + { + "left": "issue:42757", + "right": "issue:44351", + "accept": false, + "reason": "Different import failures from different packages/symbols; not the same bug." + }, + { + "left": "issue:45216", + "right": "issue:45478", + "accept": false, + "reason": "Same model family, but save_pretrained checkpoint corruption and from_pretrained failure are distinct bugs." + }, + { + "left": "issue:43066", + "right": "issue:44779", + "accept": false, + "reason": "Wrong decoder type and Deepseek tokenization regression are different issues." + }, + { + "left": "issue:22355", + "right": "issue:28218", + "accept": false, + "reason": "Missing transformers.onnx module and extra tokenizer space are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:44222", + "accept": false, + "reason": "Both touch saving, but they are model-specific quantization/save bugs with different failure modes." + }, + { + "left": "issue:43122", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer regression and Qwen3.5 flash-attention generation bug are unrelated." + }, + { + "left": "issue:43992", + "right": "issue:44704", + "accept": false, + "reason": "Missing embed_tokens.weight and cached_file kwargs forwarding are unrelated." + }, + { + "left": "issue:43784", + "right": "issue:45446", + "accept": false, + "reason": "sentence-transformers import NameError and flex_attention version check are unrelated." + }, + { + "left": "issue:43937", + "right": "issue:44485", + "accept": false, + "reason": "Invalid GenerationConfig and RoPE implementation are unrelated." + }, + { + "left": "issue:42222", + "right": "issue:43441", + "accept": false, + "reason": "Vitpose model breakage and Ministral FlashAttention failure are unrelated." + }, + { + "left": "issue:42371", + "right": "issue:43874", + "accept": false, + "reason": "TF32 settings warning and missing image-patch helper are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:45446", + "accept": false, + "reason": "Python version compatibility issue and flex_attention version check are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:45753", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and Qwen3_5 infinite-loop bug are unrelated." + }, + { + "left": "issue:44075", + "right": "issue:44936", + "accept": false, + "reason": "Optimizer argument handling and trainer.evaluate-after-train are different trainer bugs." + }, + { + "left": "issue:42831", + "right": "issue:45753", + "accept": false, + "reason": "FineGrainedFP8 accuracy issue and image-specific infinite loop are unrelated." + }, + { + "left": "issue:43116", + "right": "issue:43825", + "accept": false, + "reason": "Example-script classification bug and pipeline translation error message are unrelated." + }, + { + "left": "issue:44038", + "right": "issue:45478", + "accept": false, + "reason": "Related model area, but Qwen3-VL-Moe load failure and Qwen3.5 MoE from_pretrained error are not the same concrete bug." + }, + { + "left": "issue:44222", + "right": "issue:45478", + "accept": false, + "reason": "FP8 save_pretrained bug and Qwen3.5 MoE from_pretrained failure are distinct save/load problems." + }, + { + "left": "issue:43116", + "right": "issue:43493", + "accept": false, + "reason": "Classification example bug and SigLIP2 discrepancy are unrelated." + }, + { + "left": "issue:45083", + "right": "issue:45446", + "accept": false, + "reason": "qwen3_omni_moe feature-length helper bug and flex_attention import/version check 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": 2, + "component_count": 16, + "review_unit_index": 39, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:22355", + "issue:29127", + "issue:33453", + "issue:38617", + "issue:39401", + "issue:41628", + "issue:42175", + "issue:42222", + "issue:42673", + "issue:43012", + "issue:43064", + "issue:43066", + "issue:43122", + "issue:43381", + "issue:43408", + "issue:43421", + "issue:43450", + "issue:43577", + "issue:43673", + "issue:43742", + "issue:43827", + "issue:43937", + "issue:43986", + "issue:44075", + "issue:44117", + "issue:44291", + "issue:44410", + "issue:44483", + "issue:44484", + "issue:44496", + "issue:44779", + "issue:44829", + "issue:44945", + "issue:44964", + "issue:44977", + "issue:45071", + "issue:45072", + "issue:45081", + "issue:45216", + "issue:45278", + "issue:45310", + "issue:45405", + "issue:45446", + "issue:45464", + "issue:45542", + "issue:45563", + "issue:45588", + "issue:45656" + ], + "soft_pairs": [ + "issue:43064|issue:43673", + "issue:43937|issue:45563", + "issue:38617|issue:45278", + "issue:39401|issue:44779", + "issue:33453|issue:39401", + "issue:22355|issue:29127", + "issue:44117|issue:45588", + "issue:43827|issue:45446", + "issue:43381|issue:44945", + "issue:43937|issue:44977", + "issue:45405|issue:45446", + "issue:43450|issue:43986", + "issue:42673|issue:45072", + "issue:44117|issue:45081", + "issue:38617|issue:41628", + "issue:39401|issue:43421", + "issue:44410|issue:45310", + "issue:44829|issue:45216", + "issue:44291|issue:45071", + "issue:44484|issue:45310", + "issue:42222|issue:45542", + "issue:43408|issue:43742", + "issue:43937|issue:44496", + "issue:42175|issue:42222", + "issue:44075|issue:45656", + "issue:44483|issue:45464", + "issue:44964|issue:45278", + "issue:43012|issue:43577", + "issue:43066|issue:43122" + ], + "prepared_review_unit_hash": "d4d76187492f6c77e4f04646167697b1e8c80d3ff4c2300fd4339d5566f22b38", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36332, + "estimated_input_tokens": 9083, + "estimated_eval_tokens": 18422 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These issues are mostly unrelated individual regressions; none of the soft edges look like true duplicates or mergeable PR-equivalent changes.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43064", + "right": "issue:43673", + "accept": false, + "reason": "Different failures: distributed optimizer state corruption vs missing GenerationMixin cache during chunked prefill." + }, + { + "left": "issue:43937", + "right": "issue:45563", + "accept": false, + "reason": "Both mention generation, but one is invalid GenerationConfig and the other is a stale warning for num_return_sequences." + }, + { + "left": "issue:38617", + "right": "issue:45278", + "accept": false, + "reason": "Same broad import-error theme, but different missing symbols and likely different regressions." + }, + { + "left": "issue:39401", + "right": "issue:44779", + "accept": false, + "reason": "Tokenizer-related, but Qwen3 wrong offset mapping and Deepseek incorrect tokenization are different bugs." + }, + { + "left": "issue:33453", + "right": "issue:39401", + "accept": false, + "reason": "Both are tokenizer regressions, but one is generic loading regression and the other is a model-specific offset_mapping bug." + }, + { + "left": "issue:22355", + "right": "issue:29127", + "accept": false, + "reason": "Completely different problems: missing transformers.onnx module vs LayoutLMv3 error-message clarity." + }, + { + "left": "issue:44117", + "right": "issue:45588", + "accept": false, + "reason": "Unrelated code paths: tokenizer mapping lookup vs flash_attention crash on s_aux=None." + }, + { + "left": "issue:43827", + "right": "issue:45446", + "accept": false, + "reason": "Docs reference cleanup and a PyTorch version check bug are not the same underlying issue." + }, + { + "left": "issue:43381", + "right": "issue:44945", + "accept": false, + "reason": "Gradient checkpointing eval-mode behavior is unrelated to pipeline-parallelism output corruption." + }, + { + "left": "issue:43937", + "right": "issue:44977", + "accept": false, + "reason": "Different concrete bugs: invalid generation config vs Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:45405", + "right": "issue:45446", + "accept": false, + "reason": "Packaging/version pin issue versus an import-version check bug; not the same fix." + }, + { + "left": "issue:43450", + "right": "issue:43986", + "accept": false, + "reason": "Batched video shape bug and AutoProcessor crash without torchvision are different failure modes." + }, + { + "left": "issue:42673", + "right": "issue:45072", + "accept": false, + "reason": "VRAM leak in Qwen3ForCausalLM threads is unrelated to dtype mismatches in other models." + }, + { + "left": "issue:44117", + "right": "issue:45081", + "accept": false, + "reason": "TOKENIZER_MAPPING_NAMES None handling is unrelated to the Mistral regex patch crash." + }, + { + "left": "issue:38617", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but for different exported names and likely unrelated causes." + }, + { + "left": "issue:39401", + "right": "issue:43421", + "accept": false, + "reason": "Wrong offset mapping and stale post-processor updates after special-token edits are different tokenizer bugs." + }, + { + "left": "issue:44410", + "right": "issue:45310", + "accept": false, + "reason": "Different Qwen-family issues: missing projections in qwen3next vs from_pretrained failure in Qwen3.5 MoE." + }, + { + "left": "issue:44829", + "right": "issue:45216", + "accept": false, + "reason": "Flash-attention training degeneration is unrelated to incorrect checkpoint saving in Qwen3.5." + }, + { + "left": "issue:44291", + "right": "issue:45071", + "accept": false, + "reason": "init_empty_weights argument mismatch and PretrainedConfig type-checking breakage are separate v5 regressions." + }, + { + "left": "issue:44484", + "right": "issue:45310", + "accept": false, + "reason": "A save_pretrained shard-size question is unrelated to the Qwen3.5 MoE loading error." + }, + { + "left": "issue:42222", + "right": "issue:45542", + "accept": false, + "reason": "Different subsystems entirely: VitPose model breakage vs missing TensorFlow backend when only tensorboard is installed." + }, + { + "left": "issue:43408", + "right": "issue:43742", + "accept": false, + "reason": "SAM3 model-type warning and MobileLLM load key error are not the same bug." + }, + { + "left": "issue:44075", + "right": "issue:45656", + "accept": false, + "reason": "SGD args ignored and optimizer.step() called twice under deepspeed are different optimizer bugs." + }, + { + "left": "issue:44483", + "right": "issue:45464", + "accept": false, + "reason": "General chat/completions API failure is not the same as the Qwen3.5 streaming-specific failure." + }, + { + "left": "issue:44964", + "right": "issue:45278", + "accept": false, + "reason": "Model-loading failure for Phi-4 multimodal is unrelated to the broad import-error report." + }, + { + "left": "issue:43012", + "right": "issue:43577", + "accept": false, + "reason": "One is a PyTorch warning during bf16 compilation; the other is incorrect dtype state when loading BLIP2." + }, + { + "left": "issue:43066", + "right": "issue:43122", + "accept": false, + "reason": "Both are tokenizer-v5 regressions, but decoder-type mismatch and changed tokenization output 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": 2, + "component_count": 16, + "review_unit_index": 40, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:38617", + "issue:41628", + "issue:42222", + "issue:42371", + "issue:42673", + "issue:42757", + "issue:42831", + "issue:43012", + "issue:43064", + "issue:43097", + "issue:43335", + "issue:43408", + "issue:43441", + "issue:43452", + "issue:43874", + "issue:43906", + "issue:43931", + "issue:43976", + "issue:43986", + "issue:44038", + "issue:44062", + "issue:44117", + "issue:44220", + "issue:44261", + "issue:44263", + "issue:44315", + "issue:44351", + "issue:44368", + "issue:44410", + "issue:44484", + "issue:44485", + "issue:44493", + "issue:44704", + "issue:44829", + "issue:44945", + "issue:44964", + "issue:44977", + "issue:44987", + "issue:45071", + "issue:45103", + "issue:45200", + "issue:45245", + "issue:45341", + "issue:45375", + "issue:45459", + "issue:45478", + "issue:45542", + "issue:45588" + ], + "soft_pairs": [ + "issue:42371|issue:44261", + "issue:38617|issue:44220", + "issue:44410|issue:45478", + "issue:43874|issue:44261", + "issue:44484|issue:45478", + "issue:43097|issue:44493", + "issue:44263|issue:44485", + "issue:42222|issue:45459", + "issue:43012|issue:43408", + "issue:43441|issue:44829", + "issue:41628|issue:45071", + "issue:42673|issue:44977", + "issue:45071|issue:45542", + "issue:42371|issue:42831", + "issue:42673|issue:43906", + "issue:43931|issue:44038", + "issue:44117|issue:44987", + "issue:43097|issue:43452", + "issue:43335|issue:45071", + "issue:43976|issue:44945", + "issue:43408|issue:44368", + "issue:43874|issue:44263", + "issue:43986|issue:44484", + "issue:44829|issue:45245", + "issue:38617|issue:42757", + "issue:43064|issue:44315", + "issue:44410|issue:45375", + "issue:38617|issue:44351", + "issue:43097|issue:45588", + "issue:44062|issue:44117", + "issue:44038|issue:44410", + "issue:42673|issue:45200", + "issue:44964|issue:45071", + "issue:44704|issue:45103", + "issue:44261|issue:44485", + "issue:43931|issue:44410", + "issue:44964|issue:45341" + ], + "prepared_review_unit_hash": "e0e09981cbd70bb07b71ecfed1ed921744b12c27e3b1d6beec78ebc61d22e8a7", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 37, + "serialized_chars": 38752, + "estimated_input_tokens": 9688, + "estimated_eval_tokens": 19632 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a heterogeneous cluster of unrelated Transformers bug reports; none of the soft pairs look like the same underlying defect, so no duplicate merge is warranted.", + "confidence": 0.71, + "canonical_issue_reason": "No single canonical issue fits this cluster: it mixes unrelated import errors, model-specific regressions, config warnings, and training/runtime bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44704 is the cleanest representative because it is a concrete core API bug with a clear failure mode and broader potential impact.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42371", + "right": "issue:44261", + "accept": false, + "reason": "TF32 settings guidance is unrelated to the MLA rms_norm_eps precision bug." + }, + { + "left": "issue:38617", + "right": "issue:44220", + "accept": false, + "reason": "Different import/pathology: a missing config symbol vs an audio feature-extraction issue." + }, + { + "left": "issue:44410", + "right": "issue:45478", + "accept": false, + "reason": "Different Qwen3 model families and different failure modes." + }, + { + "left": "issue:43874", + "right": "issue:44261", + "accept": false, + "reason": "Image patch-count method bug is unrelated to the MLA precision/config issue." + }, + { + "left": "issue:44484", + "right": "issue:45478", + "accept": false, + "reason": "Save-pretrained shard sizing is unrelated to Qwen3.5 MoE loading." + }, + { + "left": "issue:43097", + "right": "issue:44493", + "accept": false, + "reason": "Removed tie_embeddings API warning is not the same as unexpected position-id keys." + }, + { + "left": "issue:44263", + "right": "issue:44485", + "accept": false, + "reason": "Both are GLM-related, but they describe different code paths and bugs." + }, + { + "left": "issue:42222", + "right": "issue:45459", + "accept": false, + "reason": "VitPose breakage is unrelated to tokenizer protobuf error masking." + }, + { + "left": "issue:43012", + "right": "issue:43408", + "accept": false, + "reason": "bfloat16 compile warning and sam3 model-type warning are different issues." + }, + { + "left": "issue:43441", + "right": "issue:44829", + "accept": false, + "reason": "Both mention FlashAttention, but one is inference failure and the other is degenerate training." + }, + { + "left": "issue:41628", + "right": "issue:45071", + "accept": false, + "reason": "Missing AutoImageProcessor import is unrelated to PretrainedConfig type checking." + }, + { + "left": "issue:42673", + "right": "issue:44977", + "accept": false, + "reason": "VRAM leak in dataloader threads is not the same as flash-attention generation failure." + }, + { + "left": "issue:45071", + "right": "issue:45542", + "accept": false, + "reason": "Type-checking regression is unrelated to the TensorFlow backend import error." + }, + { + "left": "issue:42371", + "right": "issue:42831", + "accept": false, + "reason": "TF32 settings control is unrelated to FineGrainedFP8 accuracy." + }, + { + "left": "issue:42673", + "right": "issue:43906", + "accept": false, + "reason": "These are different Qwen3-related reports with no shared concrete failure path." + }, + { + "left": "issue:43931", + "right": "issue:44038", + "accept": false, + "reason": "Different model variants and different loading failures." + }, + { + "left": "issue:44117", + "right": "issue:44987", + "accept": false, + "reason": "Tokenizer mapping returning None is not clearly the same as that model-loading failure." + }, + { + "left": "issue:43097", + "right": "issue:43452", + "accept": false, + "reason": "API removal warning and gguf_file loading breakage are different bugs." + }, + { + "left": "issue:43335", + "right": "issue:45071", + "accept": false, + "reason": "Sparse-layer config creation bug is unrelated to type-checking behavior." + }, + { + "left": "issue:43976", + "right": "issue:44945", + "accept": false, + "reason": "Python-version compatibility and pipeline-parallel output corruption are unrelated." + }, + { + "left": "issue:43408", + "right": "issue:44368", + "accept": false, + "reason": "Different warnings in different models; not the same underlying bug." + }, + { + "left": "issue:43874", + "right": "issue:44263", + "accept": false, + "reason": "Both mention GLM, but one is a missing processor method and the other is a torch.split handling issue." + }, + { + "left": "issue:43986", + "right": "issue:44484", + "accept": false, + "reason": "AutoProcessor torchvision crash is unrelated to save_pretrained sharding." + }, + { + "left": "issue:44829", + "right": "issue:45245", + "accept": false, + "reason": "FlashAttention training degeneration is unrelated to category-cardinality limits." + }, + { + "left": "issue:38617", + "right": "issue:42757", + "accept": false, + "reason": "Different missing imports from different modules/symbols." + }, + { + "left": "issue:43064", + "right": "issue:44315", + "accept": false, + "reason": "Wrong optimizer state distribution is not the same as Liger Kernel not being applied." + }, + { + "left": "issue:44410", + "right": "issue:45375", + "accept": false, + "reason": "Missing projections in qwen3next is unrelated to a silently dropped config field." + }, + { + "left": "issue:38617", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols and code paths." + }, + { + "left": "issue:43097", + "right": "issue:45588", + "accept": false, + "reason": "Config API removal and flash_attention.py s_aux=None crash are different bugs." + }, + { + "left": "issue:44062", + "right": "issue:44117", + "accept": false, + "reason": "Token AddedToken kwarg duplication is unrelated to tokenizer mapping None handling." + }, + { + "left": "issue:44038", + "right": "issue:44410", + "accept": false, + "reason": "Different Qwen3-family model issues with different symptoms." + }, + { + "left": "issue:42673", + "right": "issue:45200", + "accept": false, + "reason": "VRAM leak in multi-thread loading is unrelated to Gemma 4 mm_token_type_ids defaults." + }, + { + "left": "issue:44964", + "right": "issue:45071", + "accept": false, + "reason": "Phi-4 multimodal load failure is unrelated to a type-checking regression." + }, + { + "left": "issue:44704", + "right": "issue:45103", + "accept": false, + "reason": "cached_file kwargs forwarding and auto_docstring annotation crashes are unrelated." + }, + { + "left": "issue:44261", + "right": "issue:44485", + "accept": false, + "reason": "MLA layernorm epsilon precision and GLM-5 RoPE implementation are different bugs." + }, + { + "left": "issue:43931", + "right": "issue:44410", + "accept": false, + "reason": "Qwen3-VL shape mismatch and qwen3next missing projections are separate model issues." + }, + { + "left": "issue:44964", + "right": "issue:45341", + "accept": false, + "reason": "Phi-4 multimodal loading is unrelated to a testing_utils bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 41, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:36683", + "issue:38617", + "issue:42617", + "issue:42757", + "issue:42886", + "issue:42907", + "issue:42994", + "issue:43012", + "issue:43064", + "issue:43066", + "issue:43097", + "issue:43208", + "issue:43299", + "issue:43381", + "issue:43475", + "issue:43502", + "issue:43526", + "issue:43550", + "issue:43618", + "issue:43761", + "issue:43824", + "issue:43825", + "issue:43867", + "issue:43873", + "issue:43901", + "issue:43906", + "issue:43986", + "issue:43992", + "issue:44038", + "issue:44117", + "issue:44230", + "issue:44368", + "issue:44380", + "issue:44483", + "issue:44484", + "issue:44514", + "issue:44655", + "issue:44683", + "issue:45216", + "issue:45290", + "issue:45341", + "issue:45372", + "issue:45381", + "issue:45561", + "issue:45588", + "issue:45753", + "issue:45859" + ], + "soft_pairs": [ + "issue:43502|issue:43992", + "issue:45341|issue:45859", + "issue:43986|issue:44117", + "issue:43097|issue:43873", + "issue:43066|issue:43906", + "issue:43986|issue:45290", + "issue:45341|issue:45561", + "issue:43986|issue:45372", + "issue:43097|issue:43901", + "issue:42617|issue:45341", + "issue:43986|issue:45341", + "issue:43825|issue:44655", + "issue:45341|issue:45588", + "issue:43992|issue:44484", + "issue:43012|issue:43550", + "issue:43208|issue:43550", + "issue:43012|issue:44368", + "issue:42886|issue:43381", + "issue:42907|issue:45753", + "issue:42907|issue:44655", + "issue:43208|issue:43526", + "issue:44483|issue:44514", + "issue:43012|issue:44683", + "issue:42757|issue:43824", + "issue:38617|issue:43502", + "issue:44483|issue:45290", + "issue:44038|issue:44230", + "issue:43618|issue:43761", + "issue:43064|issue:44380", + "issue:43299|issue:43867", + "issue:44483|issue:45381", + "issue:43901|issue:44655", + "issue:36683|issue:43475", + "issue:43867|issue:45216", + "issue:42994|issue:44655" + ], + "prepared_review_unit_hash": "f958b5cdfbb50b31faa9c1d4ce91eab86b9575f3707d780adf56079b2aeb15f2", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 35, + "serialized_chars": 37419, + "estimated_input_tokens": 9355, + "estimated_eval_tokens": 18966 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No soft pair looks like a safe duplicate merge; the set is mostly unrelated regression reports across different models and subsystems.", + "confidence": 0.68, + "canonical_issue_reason": "No safe canonical issue emerges here: the candidate pairs are mostly unrelated, so choosing one as the duplicate anchor would be arbitrary.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the strongest standalone anchor among these issues: it is open, broad in scope (quantization/offloading), and has the most surrounding discussion/references.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43502", + "right": "issue:43992", + "accept": false, + "reason": "Different bugs: local_files_only network leakage vs a missing embed_tokens weight during UMT5 loading." + }, + { + "left": "issue:45341", + "right": "issue:45859", + "accept": false, + "reason": "Unrelated subsystems: a testing_utils bug vs Qwen3_5MoeTextRotaryEmbedding CPU offload compatibility." + }, + { + "left": "issue:43986", + "right": "issue:44117", + "accept": false, + "reason": "Both are loading-related, but one is a missing torchvision dependency path and the other is a tokenizer mapping None-handling bug." + }, + { + "left": "issue:43097", + "right": "issue:43873", + "accept": false, + "reason": "Different problems: a removed API warning vs quantization/offloading behavior." + }, + { + "left": "issue:43066", + "right": "issue:43906", + "accept": false, + "reason": "No clear match: tokenizer decoder type regression vs an isolated reproduction of a different issue." + }, + { + "left": "issue:43986", + "right": "issue:45290", + "accept": false, + "reason": "Different areas: AutoProcessor video-model import failure vs chat template crashes on tool-call assistant messages." + }, + { + "left": "issue:45341", + "right": "issue:45561", + "accept": false, + "reason": "Shared file name is not enough; one is a generic testing_utils bug and the other is a specific xdist race on captured_info.txt." + }, + { + "left": "issue:43986", + "right": "issue:45372", + "accept": false, + "reason": "Different import failures from different dependencies and models." + }, + { + "left": "issue:43097", + "right": "issue:43901", + "accept": false, + "reason": "Completely unrelated: removed tie_embeddings API vs stale pipeline docs about return_all_scores." + }, + { + "left": "issue:42617", + "right": "issue:45341", + "accept": false, + "reason": "Different failures: 3d_parallel.py execution vs a testing_utils bug." + }, + { + "left": "issue:43986", + "right": "issue:45341", + "accept": false, + "reason": "Different domains: video-model loading/import path vs testing infrastructure." + }, + { + "left": "issue:43825", + "right": "issue:44655", + "accept": false, + "reason": "Different surface areas: pipeline task messaging vs saving Pipeline objects." + }, + { + "left": "issue:45341", + "right": "issue:45588", + "accept": false, + "reason": "Testing-utils race bug vs a flash_attention AttributeError in sink-less models." + }, + { + "left": "issue:43992", + "right": "issue:44484", + "accept": false, + "reason": "Different save/load concerns: missing embed weight on load vs max_shard_size default behavior." + }, + { + "left": "issue:43012", + "right": "issue:43550", + "accept": false, + "reason": "One is a warning about bfloat16 compilation precision; the other is an actual torch.compile SDPA failure on Bamba." + }, + { + "left": "issue:43208", + "right": "issue:43550", + "accept": false, + "reason": "xLSTM training bugs are unrelated to a Bamba SDPA compile regression." + }, + { + "left": "issue:43012", + "right": "issue:44368", + "accept": false, + "reason": "Both mention warnings, but they are about different configs and different underlying behaviors." + }, + { + "left": "issue:42886", + "right": "issue:43381", + "accept": false, + "reason": "Offline cache loading and gradient checkpointing in eval mode are separate bugs." + }, + { + "left": "issue:42907", + "right": "issue:45753", + "accept": false, + "reason": "Saving dequantized models and an infinite-generation loop are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:44655", + "accept": false, + "reason": "Both involve saving, but one is quantized model serialization and the other is Pipeline save_pretrained support." + }, + { + "left": "issue:43208", + "right": "issue:43526", + "accept": false, + "reason": "Different components: xLSTM training bugs vs BeitImageProcessorFast reduce_labels handling." + }, + { + "left": "issue:44483", + "right": "issue:44514", + "accept": false, + "reason": "Different chat-related failures: OpenAI-style endpoint rejection vs batched apply_chat_template crash." + }, + { + "left": "issue:43012", + "right": "issue:44683", + "accept": false, + "reason": "A warning about bfloat16 precision is not the same bug as flex_attention compilation failure on newer torch." + }, + { + "left": "issue:42757", + "right": "issue:43824", + "accept": false, + "reason": "Both are ImportErrors, but for different missing symbols and different code paths." + }, + { + "left": "issue:38617", + "right": "issue:43502", + "accept": false, + "reason": "Unrelated imports: layer_type_validation removal vs local_files_only still making API calls." + }, + { + "left": "issue:44483", + "right": "issue:45290", + "accept": false, + "reason": "Different chat-stack regressions: request validation vs chat template handling of tool-call messages." + }, + { + "left": "issue:44038", + "right": "issue:44230", + "accept": false, + "reason": "Both mention Qwen3-VL/MoE, but one is a v5 bug report and the other is fp8 inference support." + }, + { + "left": "issue:43618", + "right": "issue:43761", + "accept": false, + "reason": "Both are CLIP regressions, but one is missing attentions and the other is hidden_states=None; not the same concrete failure." + }, + { + "left": "issue:43064", + "right": "issue:44380", + "accept": false, + "reason": "Trainer/FSDP optimizer-state corruption is unrelated to GPT2 attention scaling being ignored under SDPA/FlashAttention." + }, + { + "left": "issue:43299", + "right": "issue:43867", + "accept": false, + "reason": "Different model-loading bugs: Qwen3VL-Moe load failure vs state_dict sorting load error." + }, + { + "left": "issue:44483", + "right": "issue:45381", + "accept": false, + "reason": "OpenAI chat-completions request handling is unrelated to qwen2.5-vl vision_position_ids for video input." + }, + { + "left": "issue:43901", + "right": "issue:44655", + "accept": false, + "reason": "Docs drift about return_all_scores is unrelated to Pipeline save_pretrained failures." + }, + { + "left": "issue:36683", + "right": "issue:43475", + "accept": false, + "reason": "Different missing-attribute errors on different model classes: Gemma3Config vocab_size vs Sam3VisionEncoderOutput fpn_position_embeddings." + }, + { + "left": "issue:43867", + "right": "issue:45216", + "accept": false, + "reason": "Both involve checkpoints, but one is a load-time state_dict ordering error and the other is a save_pretrained regression." + }, + { + "left": "issue:42994", + "right": "issue:44655", + "accept": false, + "reason": "Both are save-related, but one is quantized model saving and the other is Pipeline object serialization." + } + ] + }, + "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": 16, + "review_unit_index": 42, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:36010", + "issue:36296", + "issue:36683", + "issue:41669", + "issue:42757", + "issue:42890", + "issue:43064", + "issue:43066", + "issue:43097", + "issue:43299", + "issue:43352", + "issue:43441", + "issue:43525", + "issue:43575", + "issue:43618", + "issue:43756", + "issue:43867", + "issue:43881", + "issue:43906", + "issue:43937", + "issue:43976", + "issue:43992", + "issue:44077", + "issue:44117", + "issue:44220", + "issue:44242", + "issue:44246", + "issue:44263", + "issue:44355", + "issue:44360", + "issue:44380", + "issue:44484", + "issue:44485", + "issue:44492", + "issue:44655", + "issue:44855", + "issue:44938", + "issue:45216", + "issue:45310", + "issue:45357", + "issue:45375", + "issue:45381", + "issue:45440", + "issue:45478", + "issue:45479", + "issue:45561", + "issue:45753" + ], + "soft_pairs": [ + "issue:43352|issue:43937", + "issue:36010|issue:42757", + "issue:43066|issue:45381", + "issue:43756|issue:44485", + "issue:44220|issue:44655", + "issue:44263|issue:45440", + "issue:36683|issue:43525", + "issue:43618|issue:45216", + "issue:43906|issue:45753", + "issue:36683|issue:44242", + "issue:43299|issue:45753", + "issue:36683|issue:43881", + "issue:43352|issue:43976", + "issue:43441|issue:44380", + "issue:43992|issue:45310", + "issue:43352|issue:44380", + "issue:43867|issue:45375", + "issue:44380|issue:45479", + "issue:41669|issue:44246", + "issue:43352|issue:45478", + "issue:43618|issue:44380", + "issue:44355|issue:44855", + "issue:43064|issue:44492", + "issue:36683|issue:44117", + "issue:42890|issue:45561", + "issue:43097|issue:43352", + "issue:43867|issue:45357", + "issue:43976|issue:44938", + "issue:43992|issue:45478", + "issue:44360|issue:45440", + "issue:43064|issue:44655", + "issue:36296|issue:43575", + "issue:44077|issue:44380", + "issue:43867|issue:44484" + ], + "prepared_review_unit_hash": "fef0bb11ca8d90e301e94cad4a8a138fb3319207779285c5dd52bd6f9a8653cc", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 34, + "serialized_chars": 36869, + "estimated_input_tokens": 9218, + "estimated_eval_tokens": 18692 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous: most items are distinct Transformers regressions spanning different models, versions, and subsystems. The soft-similarity edges are mostly false positives rather than true duplicates.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue cleanly subsumes this set; the items split into unrelated model-specific loading, config, attention, and tooling regressions. If forced to pick a representative, issue 43097 is the broadest v5 regression, but it does not duplicate the others.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43097 is the best representative issue only because it is a broad, user-facing v5 breakage with clear impact. It is still not a good canonical for the whole cluster, since the rest are mostly separate bugs.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43352", + "right": "issue:43937", + "accept": false, + "reason": "Different failures: FlashAttention support gap for Nemotron vs invalid GenerationConfig for GLM-5." + }, + { + "left": "issue:36010", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but for different symbols from different packages and unrelated code paths." + }, + { + "left": "issue:43066", + "right": "issue:45381", + "accept": false, + "reason": "Wrong tokenizer decoder type vs Qwen2.5-VL video position ids; different model and bug." + }, + { + "left": "issue:43756", + "right": "issue:44485", + "accept": false, + "reason": "Smollm3 RoPE layer mismatch vs GLM-5 RoPE implementation discussion; distinct model behaviors." + }, + { + "left": "issue:44220", + "right": "issue:44655", + "accept": false, + "reason": "Audio feature extraction issue vs pipeline save_pretrained failure; unrelated subsystems." + }, + { + "left": "issue:44263", + "right": "issue:45440", + "accept": false, + "reason": "DSA torch.split behavior vs DeepseekV3MoE divergence; not the same bug." + }, + { + "left": "issue:36683", + "right": "issue:43525", + "accept": false, + "reason": "Gemma3 missing vocab_size vs Llama4 missing pad_token_id; same symptom class, different model/config fields." + }, + { + "left": "issue:43618", + "right": "issue:45216", + "accept": false, + "reason": "CLIPOutput attentions omission vs Qwen3.5 save_pretrained regression; unrelated fixes." + }, + { + "left": "issue:43906", + "right": "issue:45753", + "accept": false, + "reason": "Isolated reproduction of a prior issue vs Qwen3_5 infinite loop on a specific image; not duplicates." + }, + { + "left": "issue:36683", + "right": "issue:44242", + "accept": false, + "reason": "Gemma3 config attribute error vs load balancing loss omission in MoE routing; different code paths." + }, + { + "left": "issue:43299", + "right": "issue:45753", + "accept": false, + "reason": "Qwen3VL MoE model loading breakage vs Qwen3_5 image-specific infinite loop; separate problems." + }, + { + "left": "issue:36683", + "right": "issue:43881", + "accept": false, + "reason": "Gemma3 config attribute error vs GLM-4V-9B loading failure; different models and failures." + }, + { + "left": "issue:43352", + "right": "issue:43976", + "accept": false, + "reason": "FlashAttention support bug vs Python version compatibility issue; unrelated." + }, + { + "left": "issue:43441", + "right": "issue:44380", + "accept": false, + "reason": "Ministral FlashAttention failure vs GPT2 attention scaling ignored under SDPA/FlashAttention; different models and fixes." + }, + { + "left": "issue:43992", + "right": "issue:45310", + "accept": false, + "reason": "Missing embed_tokens.weight in UMT5Encoder vs Qwen3.5 Moe from_pretrained error; distinct loading bugs." + }, + { + "left": "issue:43352", + "right": "issue:44380", + "accept": false, + "reason": "Both involve attention backends, but one is unsupported model/backend, the other ignores a config scaling parameter." + }, + { + "left": "issue:43867", + "right": "issue:45375", + "accept": false, + "reason": "State_dict ordering load error vs missing Qwen3_5MoeVisionConfig field; different failure mechanisms." + }, + { + "left": "issue:44380", + "right": "issue:45479", + "accept": false, + "reason": "Attention scaling backend bug vs degenerate zero-loss classification bug; unrelated." + }, + { + "left": "issue:41669", + "right": "issue:44246", + "accept": false, + "reason": "Import-time slowdown from import * usage vs intermittent import slowdown; same broad symptom, but not the same concrete bug." + }, + { + "left": "issue:43352", + "right": "issue:45478", + "accept": false, + "reason": "Different Qwen/Nemotron style issue vs Qwen3.5 Moe from_pretrained regression; unrelated model-specific breakages." + }, + { + "left": "issue:43618", + "right": "issue:44380", + "accept": false, + "reason": "CLIP attentions field regression vs GPT2 attention scaling ignored; different affected code and behavior." + }, + { + "left": "issue:44355", + "right": "issue:44855", + "accept": false, + "reason": "Compiled Python files import errors vs Python 3.13 IndentationError in DebertaV2Model; distinct root causes." + }, + { + "left": "issue:43064", + "right": "issue:44492", + "accept": false, + "reason": "FSDP2 optimizer state corruption vs a typo in cache strategies; unrelated." + }, + { + "left": "issue:36683", + "right": "issue:44117", + "accept": false, + "reason": "Model config missing vocab_size vs tokenizer mapping sometimes returning None; different layers of the stack." + }, + { + "left": "issue:42890", + "right": "issue:45561", + "accept": false, + "reason": "Flaky seed-dependent integration tests vs xdist race on captured_info.txt; both test issues but not the same bug." + }, + { + "left": "issue:43097", + "right": "issue:43352", + "accept": false, + "reason": "Broad removed-API regression vs model-specific FlashAttention support issue; not one underlying change." + }, + { + "left": "issue:43867", + "right": "issue:45357", + "accept": false, + "reason": "Generic state_dict loading error vs Qwen3.5 visual encoder key regression; different failure path." + }, + { + "left": "issue:43976", + "right": "issue:44938", + "accept": false, + "reason": "Python 3.9 compatibility regression vs Python 3.14 load failure; version issues only, not same bug." + }, + { + "left": "issue:43992", + "right": "issue:45478", + "accept": false, + "reason": "UMT5Encoder missing weight vs Qwen3.5 Moe from_pretrained error; unrelated loading bugs." + }, + { + "left": "issue:44360", + "right": "issue:45440", + "accept": false, + "reason": "Indexer missing ReLU vs DeepseekV3MoE divergence; different algorithmic issues." + }, + { + "left": "issue:43064", + "right": "issue:44655", + "accept": false, + "reason": "Optimizer-state corruption in training vs pipeline save_pretrained failure; unrelated subsystems." + }, + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Tensor-parallel training bug vs Qwen2-57B-A14B-Instruct OOM on load; different symptoms and likely causes." + }, + { + "left": "issue:44077", + "right": "issue:44380", + "accept": false, + "reason": "Optional post_init policy change vs GPT2 attention scaling backend bug; unrelated." + }, + { + "left": "issue:43867", + "right": "issue:44484", + "accept": false, + "reason": "State_dict loading error vs save_pretrained shard size question; 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": 2, + "component_count": 16, + "review_unit_index": 43, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:28282", + "issue:30333", + "issue:30990", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:36296", + "issue:42548", + "issue:42757", + "issue:42831", + "issue:42890", + "issue:43010", + "issue:43502", + "issue:43576", + "issue:43618", + "issue:43723", + "issue:43827", + "issue:43874", + "issue:43901", + "issue:44062", + "issue:44075", + "issue:44112", + "issue:44222", + "issue:44230", + "issue:44315", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44368", + "issue:44387", + "issue:44485", + "issue:44556", + "issue:44655", + "issue:44749", + "issue:44857", + "issue:44938", + "issue:44961", + "issue:44964", + "issue:44977", + "issue:45071", + "issue:45125", + "issue:45278", + "issue:45341", + "issue:45375", + "issue:45478", + "issue:45588", + "issue:45600", + "issue:45850" + ], + "soft_pairs": [ + "issue:28282|issue:33453", + "issue:43618|issue:44857", + "issue:43618|issue:44977", + "issue:43618|issue:45588", + "issue:43901|issue:44938", + "issue:44655|issue:44938", + "issue:42890|issue:44964", + "issue:42890|issue:44112", + "issue:42890|issue:45600", + "issue:43502|issue:44062", + "issue:43010|issue:44075", + "issue:42890|issue:45341", + "issue:42548|issue:44351", + "issue:44222|issue:45125", + "issue:43502|issue:44485", + "issue:44355|issue:44749", + "issue:45125|issue:45478", + "issue:30990|issue:31795", + "issue:45125|issue:45375", + "issue:44387|issue:44556", + "issue:30333|issue:33453", + "issue:43010|issue:45278", + "issue:43723|issue:44556", + "issue:43502|issue:44222", + "issue:43010|issue:44368", + "issue:44315|issue:44360", + "issue:43010|issue:45850", + "issue:43502|issue:44961", + "issue:42831|issue:43874", + "issue:30333|issue:36296", + "issue:44355|issue:45071", + "issue:44222|issue:44230", + "issue:44355|issue:44655", + "issue:43827|issue:44961", + "issue:33453|issue:34689", + "issue:42757|issue:43576" + ], + "prepared_review_unit_hash": "a36944e34c21615e3afea26aca77962ef1353149f7aa24154a38ffae30e94db3", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 36, + "serialized_chars": 37954, + "estimated_input_tokens": 9489, + "estimated_eval_tokens": 19234 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose collection of Transformers v5-era regression/import/load issues, but the proposed soft pairs generally point to different concrete bugs. I would use the broad import-regression report as the representative issue, and reject all soft edges as duplicates.", + "confidence": 0.83, + "canonical_issue_reason": "issue:45278 is the broadest and most representative of the cluster\u2019s recurring theme: breakages after upgrading to v5, especially import/loading regressions.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 best captures the dominant, user-facing problem pattern in this set and is broad enough to serve as the cluster anchor.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:28282", + "right": "issue:33453", + "accept": false, + "reason": "Different failure classes: missing PyTorch dependency vs tokenizer loading regression." + }, + { + "left": "issue:43618", + "right": "issue:44857", + "accept": false, + "reason": "Unrelated symptoms and code paths; CLIP output attribute handling is not the same as LwDetr AMP crash." + }, + { + "left": "issue:43618", + "right": "issue:44977", + "accept": false, + "reason": "Same broad model family only; one is CLIPOutput attentions, the other is Qwen3.5 flash-attention generation." + }, + { + "left": "issue:43618", + "right": "issue:45588", + "accept": false, + "reason": "Both touch attention code, but one is missing attentions assignment and the other is an s_aux=None crash in flash_attention.py." + }, + { + "left": "issue:43901", + "right": "issue:44938", + "accept": false, + "reason": "Documentation mismatch vs Python 3.14 import/load failure; not the same bug." + }, + { + "left": "issue:44655", + "right": "issue:44938", + "accept": false, + "reason": "Saving pipeline objects and loading transformers on Python 3.14 are different issues." + }, + { + "left": "issue:42890", + "right": "issue:44964", + "accept": false, + "reason": "Test seeding flakiness is unrelated to a multimodal model load failure." + }, + { + "left": "issue:42890", + "right": "issue:44112", + "accept": false, + "reason": "Both are CI/test problems, but they affect different tests and different causes." + }, + { + "left": "issue:42890", + "right": "issue:45600", + "accept": false, + "reason": "Test seed flakiness is not the same as stale auto_mappings config references breaking CI." + }, + { + "left": "issue:43502", + "right": "issue:44062", + "accept": false, + "reason": "local_files_only network leakage and AddedToken keyword collision are unrelated." + }, + { + "left": "issue:43010", + "right": "issue:44075", + "accept": false, + "reason": "Cache no_grad decoration and SGD args not being used are different code paths." + }, + { + "left": "issue:42890", + "right": "issue:45341", + "accept": false, + "reason": "A missing set_seed in tests is not the same as a separate testing_utils bug." + }, + { + "left": "issue:42548", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different symbols and likely different export fixes." + }, + { + "left": "issue:44222", + "right": "issue:45125", + "accept": false, + "reason": "FP8 save_pretrained bug and missing tensor-parallel plan are distinct model/serialization problems." + }, + { + "left": "issue:43502", + "right": "issue:44485", + "accept": false, + "reason": "local_files_only leakage is unrelated to GLM-5 RoPE implementation." + }, + { + "left": "issue:44355", + "right": "issue:44749", + "accept": false, + "reason": "Compiled Python file errors and slowed filtering after upgrade are different regressions." + }, + { + "left": "issue:45125", + "right": "issue:45478", + "accept": false, + "reason": "Both involve Qwen3.5 MoE, but one is a tensor-parallel plan issue and the other is a broader from_pretrained failure." + }, + { + "left": "issue:30990", + "right": "issue:31795", + "accept": false, + "reason": "Sentence Transformers loading hang and forward() docs confusion are unrelated." + }, + { + "left": "issue:45125", + "right": "issue:45375", + "accept": false, + "reason": "Missing tensor-parallel plan and missing config field are different Qwen3.5 MoE defects." + }, + { + "left": "issue:44387", + "right": "issue:44556", + "accept": false, + "reason": "Int4 CUDA memory/OOM behavior is not the same as checkpoint reload failure after upgrade." + }, + { + "left": "issue:30333", + "right": "issue:33453", + "accept": false, + "reason": "MLflow reporting after failed training and tokenizer loading regression are unrelated." + }, + { + "left": "issue:43010", + "right": "issue:45278", + "accept": false, + "reason": "A cache update no_grad request is not the same as broad import errors after v5.5.0." + }, + { + "left": "issue:43723", + "right": "issue:44556", + "accept": false, + "reason": "Tokenizer loading regression and checkpoint reload regression may rhyme, but they are not the same concrete bug." + }, + { + "left": "issue:43502", + "right": "issue:44222", + "accept": false, + "reason": "local_files_only network calls and FP8 save_pretrained for MoE are unrelated." + }, + { + "left": "issue:43010", + "right": "issue:44368", + "accept": false, + "reason": "No_grad on cache updates and a tied-embeddings warning in LoRA fine-tuning are different issues." + }, + { + "left": "issue:44315", + "right": "issue:44360", + "accept": false, + "reason": "Liger Kernel application and a DSA indexer ReLU discussion have no shared bug path." + }, + { + "left": "issue:43010", + "right": "issue:45850", + "accept": false, + "reason": "A specific cache mutation bug is not the same as a vague report that v5 breaks too many models." + }, + { + "left": "issue:43502", + "right": "issue:44961", + "accept": false, + "reason": "Network requests with local_files_only and an unrelated placeholder issue are not duplicates." + }, + { + "left": "issue:42831", + "right": "issue:43874", + "accept": false, + "reason": "FineGrainedFP8 accuracy and GLM46V image patch counting are separate model bugs." + }, + { + "left": "issue:30333", + "right": "issue:36296", + "accept": false, + "reason": "Failed-job MLflow reporting and tensor-parallel training are unrelated." + }, + { + "left": "issue:44355", + "right": "issue:45071", + "accept": false, + "reason": "Compiled-file errors and PretrainedConfig type-checking breakage are different regressions." + }, + { + "left": "issue:44222", + "right": "issue:44230", + "accept": false, + "reason": "Both are fp8-themed, but one is MoE save_pretrained and the other is fp8 inference support." + }, + { + "left": "issue:44355", + "right": "issue:44655", + "accept": false, + "reason": "Compiled Python file failures and pipeline save_pretrained are unrelated." + }, + { + "left": "issue:43827", + "right": "issue:44961", + "accept": false, + "reason": "Documentation referencing pipeline() and the placeholder racoon issue are not duplicates." + }, + { + "left": "issue:33453", + "right": "issue:34689", + "accept": false, + "reason": "Tokenizer loading regression and Llama 3.2 vision model loading breakage are different model-loading bugs." + }, + { + "left": "issue:42757", + "right": "issue:43576", + "accept": false, + "reason": "Missing is_offline_mode import from huggingface_hub and broken transformers env command are different failures." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 16, + "review_unit_index": 44, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:31795", + "issue:33453", + "issue:36296", + "issue:38617", + "issue:41628", + "issue:41669", + "issue:42548", + "issue:42890", + "issue:43010", + "issue:43299", + "issue:43352", + "issue:43408", + "issue:43525", + "issue:43526", + "issue:43572", + "issue:43576", + "issue:43673", + "issue:43824", + "issue:43901", + "issue:43976", + "issue:44038", + "issue:44246", + "issue:44261", + "issue:44273", + "issue:44295", + "issue:44297", + "issue:44315", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44373", + "issue:44393", + "issue:44556", + "issue:44623", + "issue:44704", + "issue:44908", + "issue:44961", + "issue:45003", + "issue:45071", + "issue:45092", + "issue:45125", + "issue:45230", + "issue:45362", + "issue:45468", + "issue:45542", + "issue:45656", + "issue:45753", + "issue:45850" + ], + "soft_pairs": [ + "issue:43299|issue:44556", + "issue:43010|issue:45656", + "issue:33453|issue:36296", + "issue:43576|issue:43824", + "issue:43576|issue:45071", + "issue:44961|issue:45468", + "issue:31795|issue:38617", + "issue:41628|issue:42548", + "issue:38617|issue:43576", + "issue:43010|issue:43976", + "issue:43576|issue:44351", + "issue:44908|issue:44961", + "issue:36296|issue:43901", + "issue:41628|issue:44355", + "issue:36296|issue:43526", + "issue:44261|issue:44360", + "issue:44623|issue:45230", + "issue:44273|issue:45542", + "issue:44273|issue:45003", + "issue:44393|issue:45753", + "issue:43576|issue:44556", + "issue:44038|issue:45125", + "issue:44393|issue:45071", + "issue:43525|issue:45850", + "issue:44556|issue:45092", + "issue:41669|issue:43673", + "issue:44373|issue:44961", + "issue:43572|issue:45850", + "issue:45362|issue:45850", + "issue:44704|issue:44908", + "issue:42890|issue:44908", + "issue:44246|issue:44273", + "issue:43408|issue:44315", + "issue:43976|issue:44393", + "issue:43352|issue:44393", + "issue:43576|issue:44908", + "issue:44908|issue:45071", + "issue:44297|issue:45230", + "issue:44295|issue:45230", + "issue:41669|issue:44351" + ], + "prepared_review_unit_hash": "dd171e46d16a7c9cd02795734c092479224567693fff408c7645f63e5777f74f", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 40, + "serialized_chars": 38936, + "estimated_input_tokens": 9734, + "estimated_eval_tokens": 19724 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No soft pair looks like a safe duplicate merge on titles alone. Most are broad v5 regressions, import/export failures, or unrelated subsystem bugs that share only a loose theme.", + "confidence": 0.58, + "canonical_issue_reason": "issue:44246 is the best standalone representative here: it is a concrete, actionable import-regression report, while most other candidates are either vague or narrower symptoms of different bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44246 is the strongest overall issue to anchor the cluster because it describes a specific user-visible failure with a plausible shared area (import/lazy loading) and enough detail to be useful for triage.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43299", + "right": "issue:44556", + "accept": false, + "reason": "Both are v5 checkpoint-loading regressions, but they hit different model families and likely different failure paths; too broad to treat as the same bug." + }, + { + "left": "issue:43010", + "right": "issue:45656", + "accept": false, + "reason": "Cache update no_grad decoration and deepspeed double optimizer-step are unrelated code paths." + }, + { + "left": "issue:33453", + "right": "issue:36296", + "accept": false, + "reason": "Tokenizer-loading regression and tensor-parallel training bug are different problems." + }, + { + "left": "issue:43576", + "right": "issue:43824", + "accept": false, + "reason": "Broken env command and missing top-level Qwen VL export are unrelated symptoms." + }, + { + "left": "issue:43576", + "right": "issue:45071", + "accept": false, + "reason": "A broken env command is not the same as PreTrainedConfig type-checking breakage." + }, + { + "left": "issue:44961", + "right": "issue:45468", + "accept": false, + "reason": "No meaningful overlap beyond both being generic issue reports." + }, + { + "left": "issue:31795", + "right": "issue:38617", + "accept": false, + "reason": "Documentation confusion and an ImportError for a missing symbol are not the same bug." + }, + { + "left": "issue:41628", + "right": "issue:42548", + "accept": false, + "reason": "Both are import failures, but for different missing symbols and likely different export fixes." + }, + { + "left": "issue:38617", + "right": "issue:43576", + "accept": false, + "reason": "Missing layer_type_validation import and the broken env command are unrelated." + }, + { + "left": "issue:43010", + "right": "issue:43976", + "accept": false, + "reason": "Decorating cache updates with no_grad and Python-version incompatibility are unrelated." + }, + { + "left": "issue:43576", + "right": "issue:44351", + "accept": false, + "reason": "Both concern imports, but they refer to different public symbols and different regressions." + }, + { + "left": "issue:44908", + "right": "issue:44961", + "accept": false, + "reason": "Scheduler kwargs handling and an unrelated issue titled 'racoon' do not match." + }, + { + "left": "issue:36296", + "right": "issue:43901", + "accept": false, + "reason": "Tensor-parallel training and a pipeline docs mismatch are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:44355", + "accept": false, + "reason": "Top-level import failure and compiled-file execution errors do not point to the same defect." + }, + { + "left": "issue:36296", + "right": "issue:43526", + "accept": false, + "reason": "Tensor parallel training and BeitImageProcessorFast label reduction are unrelated." + }, + { + "left": "issue:44261", + "right": "issue:44360", + "accept": false, + "reason": "A precision/config issue in MLA q_a_layernorm is not the same as a DSA indexer activation omission." + }, + { + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "A missing processor.save_pretrained output bug and an empty 'Bug report' title are not enough to establish sameness." + }, + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Lazy loading failures and a TensorBoard-without-TensorFlow backend error are different import-time problems." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Both touch import/lazy-loading machinery, but the reported failures and likely fixes are different." + }, + { + "left": "issue:44393", + "right": "issue:45753", + "accept": false, + "reason": "Same model family, but different symptoms and repros; not safe to merge as one bug." + }, + { + "left": "issue:43576", + "right": "issue:44556", + "accept": false, + "reason": "Broken env command and checkpoint reload regressions are unrelated." + }, + { + "left": "issue:44038", + "right": "issue:45125", + "accept": false, + "reason": "Qwen3-VL/Moe loading trouble and missing tensor-parallel plan are different concrete issues." + }, + { + "left": "issue:44393", + "right": "issue:45071", + "accept": false, + "reason": "Qwen3-VL hallucination/output issues and PretrainedConfig type checking are unrelated." + }, + { + "left": "issue:43525", + "right": "issue:45850", + "accept": false, + "reason": "A specific missing pad_token_id attribute and a vague 'breaks too many models' complaint are not the same bug." + }, + { + "left": "issue:44556", + "right": "issue:45092", + "accept": false, + "reason": "Both mention old-checkpoint incompatibility after v5, but they appear to affect different model/loading paths and are too broad to treat as duplicates." + }, + { + "left": "issue:41669", + "right": "issue:43673", + "accept": false, + "reason": "Import-star performance regression and chunked-prefill cache absence are different issues." + }, + { + "left": "issue:44373", + "right": "issue:44961", + "accept": false, + "reason": "Wrong position_ids docstring and an unrelated issue titled 'racoon' are not duplicates." + }, + { + "left": "issue:43572", + "right": "issue:45850", + "accept": false, + "reason": "A specific StableLmConfig field regression and a generic 'too many models' complaint do not match." + }, + { + "left": "issue:45362", + "right": "issue:45850", + "accept": false, + "reason": "A Qwen3.5 chat crash and a vague broad regression report are not the same concrete bug." + }, + { + "left": "issue:44704", + "right": "issue:44908", + "accept": false, + "reason": "AutoProcessor cached_file kwargs forwarding and inverse_sqrt scheduler kwargs are different code paths." + }, + { + "left": "issue:42890", + "right": "issue:44908", + "accept": false, + "reason": "Missing set_seed in SAM HQ tests and a scheduler kwargs bug are unrelated." + }, + { + "left": "issue:44246", + "right": "issue:44273", + "accept": false, + "reason": "Both concern import/lazy-loading behavior, but one reports import slowness and the other a functional lazy-loading failure; not the same concrete bug." + }, + { + "left": "issue:43408", + "right": "issue:44315", + "accept": false, + "reason": "Model-type mismatch warning and Liger kernel application through model_init are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44393", + "accept": false, + "reason": "Python version support and Qwen3-VL generation errors are unrelated." + }, + { + "left": "issue:43352", + "right": "issue:44393", + "accept": false, + "reason": "Flash Attention support gating and Qwen3-VL output errors are different problems." + }, + { + "left": "issue:43576", + "right": "issue:44908", + "accept": false, + "reason": "A broken env command and scheduler kwarg handling are unrelated." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Scheduler kwargs and PretrainedConfig type checking do not share the same bug." + }, + { + "left": "issue:44297", + "right": "issue:45230", + "accept": false, + "reason": "Tokenizer save_pretrained metadata mismatch and a generic bug report are not duplicates." + }, + { + "left": "issue:44295", + "right": "issue:45230", + "accept": false, + "reason": "Position_ids buffer access error and a generic bug report are not duplicates." + }, + { + "left": "issue:41669", + "right": "issue:44351", + "accept": false, + "reason": "Removing import * for performance and missing HybridCache export are different import-related regressions." + } + ] + }, + "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": 16, + "review_unit_index": 45, + "review_unit_count": 45, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-15354-392", + "nodes": [ + "issue:30333", + "issue:31515", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:36296", + "issue:37428", + "issue:38617", + "issue:41669", + "issue:42371", + "issue:42913", + "issue:43066", + "issue:43298", + "issue:43502", + "issue:43519", + "issue:43525", + "issue:43572", + "issue:43576", + "issue:43606", + "issue:43673", + "issue:43704", + "issue:43827", + "issue:43874", + "issue:44077", + "issue:44230", + "issue:44261", + "issue:44263", + "issue:44273", + "issue:44315", + "issue:44360", + "issue:44393", + "issue:44485", + "issue:44492", + "issue:44623", + "issue:44655", + "issue:44829", + "issue:44861", + "issue:44908", + "issue:44998", + "issue:45092", + "issue:45230", + "issue:45278", + "issue:45290", + "issue:45310", + "issue:45341", + "issue:45468", + "issue:45478", + "issue:45850" + ], + "soft_pairs": [ + "issue:43673|issue:44492", + "issue:41669|issue:45341", + "issue:45278|issue:45850", + "issue:44393|issue:45478", + "issue:43874|issue:45850", + "issue:41669|issue:44492", + "issue:44315|issue:44829", + "issue:43827|issue:44908", + "issue:44077|issue:44861", + "issue:31795|issue:33453", + "issue:41669|issue:44908", + "issue:42371|issue:44492", + "issue:43576|issue:44273", + "issue:44360|issue:45468", + "issue:44655|issue:44861", + "issue:44492|issue:45850", + "issue:44273|issue:44908", + "issue:45230|issue:45310", + "issue:43519|issue:44485", + "issue:41669|issue:43704", + "issue:43519|issue:44908", + "issue:31795|issue:34689", + "issue:31795|issue:37428", + "issue:44315|issue:45092", + "issue:43704|issue:44485", + "issue:44485|issue:45468", + "issue:42371|issue:43704", + "issue:43502|issue:43519", + "issue:43519|issue:44623", + "issue:43519|issue:43704", + "issue:44263|issue:45468", + "issue:43606|issue:45468", + "issue:44261|issue:45468", + "issue:43704|issue:45341", + "issue:42913|issue:43066", + "issue:44908|issue:44998", + "issue:44861|issue:45290", + "issue:43704|issue:44998", + "issue:41669|issue:44998", + "issue:43525|issue:44861", + "issue:43066|issue:44998", + "issue:43572|issue:44861", + "issue:31515|issue:38617", + "issue:44360|issue:44998", + "issue:44230|issue:44623", + "issue:30333|issue:31515", + "issue:31515|issue:36296", + "issue:31515|issue:31795", + "issue:44230|issue:45310", + "issue:31515|issue:33453", + "issue:43298|issue:44230" + ], + "prepared_review_unit_hash": "b5b487366e22d8d4461b6b6961f025a8714dedee62d4ecb355c8248bf15ab2fc", + "component_budget": { + "node_count": 392, + "item_count": 392, + "soft_pair_count": 1480, + "serialized_chars": 761408, + "estimated_input_tokens": 190352, + "estimated_eval_tokens": 380960 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 51, + "serialized_chars": 42264, + "estimated_input_tokens": 10566, + "estimated_eval_tokens": 21388 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a loose cluster of mostly unrelated Transformers v5 regression reports. The strongest common thread is broad model-loading / compatibility breakage after version upgrades, but the soft-linked pairs are generally too different to treat as duplicates.", + "confidence": 0.64, + "canonical_issue_reason": "issue:45278 is the best umbrella issue: it concretely reports widespread breakage after upgrading to v5.5.0, which is broader and more representative than the more specific or more vague reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the most suitable global anchor for this cluster because it captures the recurring upgrade-regression theme without being as vague as 'too many models break'.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43673", + "right": "issue:44492", + "accept": false, + "reason": "Different bugs: missing cache during chunked prefill vs a typo in cache strategy docs/code." + }, + { + "left": "issue:41669", + "right": "issue:45341", + "accept": false, + "reason": "Unrelated: import-star performance issue vs a small testing_utils bug." + }, + { + "left": "issue:45278", + "right": "issue:45850", + "accept": false, + "reason": "Same broad complaint about v5 breakage, but not the same concrete bug or fixable change." + }, + { + "left": "issue:44393", + "right": "issue:45478", + "accept": false, + "reason": "Different model families and failures: Qwen3-VL 2D bbox issue vs Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:43874", + "right": "issue:45850", + "accept": false, + "reason": "Specific missing-method multimodal token bug vs a generic meta complaint about v5 breaking models." + }, + { + "left": "issue:41669", + "right": "issue:44492", + "accept": false, + "reason": "Import-star cleanup/performance issue is unrelated to cache-strategy typo." + }, + { + "left": "issue:44315", + "right": "issue:44829", + "accept": false, + "reason": "Different code paths: Liger kernel application vs flash_attention_3 training degeneration." + }, + { + "left": "issue:43827", + "right": "issue:44908", + "accept": false, + "reason": "Docs referencing removed pipeline APIs is unrelated to inverse_sqrt scheduler kwargs handling." + }, + { + "left": "issue:44077", + "right": "issue:44861", + "accept": false, + "reason": "Different failures in different helpers; no shared underlying bug." + }, + { + "left": "issue:31795", + "right": "issue:33453", + "accept": false, + "reason": "Docs confusion vs a tokenizer loading regression." + }, + { + "left": "issue:41669", + "right": "issue:44908", + "accept": false, + "reason": "Performance/import cleanup is unrelated to scheduler kwargs propagation." + }, + { + "left": "issue:42371", + "right": "issue:44492", + "accept": false, + "reason": "TF32 API migration issue vs cache-strategy typo." + }, + { + "left": "issue:43576", + "right": "issue:44273", + "accept": false, + "reason": "Broken env command and lazy-loading failure are separate problems." + }, + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "Different model implementations and unrelated bugs." + }, + { + "left": "issue:44655", + "right": "issue:44861", + "accept": false, + "reason": "Saving Pipeline objects is a different path from tied-weight-key attribute handling." + }, + { + "left": "issue:44492", + "right": "issue:45850", + "accept": false, + "reason": "A typo in cache strategies is not the same as a broad v5 breakage report." + }, + { + "left": "issue:44273", + "right": "issue:44908", + "accept": false, + "reason": "Lazy loading failure vs scheduler kwargs bug." + }, + { + "left": "issue:45230", + "right": "issue:45310", + "accept": false, + "reason": "Generic bug report vs a specific Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:43519", + "right": "issue:44485", + "accept": false, + "reason": "Timestamp calculation bug vs RoPE implementation issue." + }, + { + "left": "issue:41669", + "right": "issue:43704", + "accept": false, + "reason": "Import-star usage cleanup is unrelated to a VRAM leak." + }, + { + "left": "issue:43519", + "right": "issue:44908", + "accept": false, + "reason": "Timestamp calculation bug vs scheduler keyword handling." + }, + { + "left": "issue:31795", + "right": "issue:34689", + "accept": false, + "reason": "Documentation confusion vs a specific model-loading regression." + }, + { + "left": "issue:31795", + "right": "issue:37428", + "accept": false, + "reason": "Docs issue vs an ImportError from a removed symbol." + }, + { + "left": "issue:44315", + "right": "issue:45092", + "accept": false, + "reason": "Both touch model initialization, but the underlying failures are different." + }, + { + "left": "issue:43704", + "right": "issue:44485", + "accept": false, + "reason": "VRAM leak and RoPE implementation are unrelated." + }, + { + "left": "issue:44485", + "right": "issue:45468", + "accept": false, + "reason": "Different positional/rope implementations for different models." + }, + { + "left": "issue:42371", + "right": "issue:43704", + "accept": false, + "reason": "TF32 settings migration vs VRAM leak." + }, + { + "left": "issue:43502", + "right": "issue:43519", + "accept": false, + "reason": "Local-files-only network leakage vs Qwen3VL timestamp math." + }, + { + "left": "issue:43519", + "right": "issue:44623", + "accept": false, + "reason": "Different processor bugs: timestamp math vs save_pretrained missing files." + }, + { + "left": "issue:43519", + "right": "issue:43704", + "accept": false, + "reason": "Timestamp bug vs VRAM leak." + }, + { + "left": "issue:44263", + "right": "issue:45468", + "accept": false, + "reason": "Different indexer and positional-encoding issues." + }, + { + "left": "issue:43606", + "right": "issue:45468", + "accept": false, + "reason": "CPU offload device mismatch vs Gemma audio positional encoding." + }, + { + "left": "issue:44261", + "right": "issue:45468", + "accept": false, + "reason": "RMS norm epsilon precision issue vs unrelated Gemma audio encoding bug." + }, + { + "left": "issue:43704", + "right": "issue:45341", + "accept": false, + "reason": "VRAM leak vs a testing_utils bug." + }, + { + "left": "issue:42913", + "right": "issue:43066", + "accept": false, + "reason": "Both are tokenizer-related, but they describe different concrete regressions." + }, + { + "left": "issue:44908", + "right": "issue:44998", + "accept": false, + "reason": "Scheduler bug vs an unrelated placeholder issue." + }, + { + "left": "issue:44861", + "right": "issue:45290", + "accept": false, + "reason": "Different failure modes in different code paths." + }, + { + "left": "issue:43704", + "right": "issue:44998", + "accept": false, + "reason": "VRAM leak vs unrelated placeholder issue." + }, + { + "left": "issue:41669", + "right": "issue:44998", + "accept": false, + "reason": "Import cleanup performance issue vs unrelated placeholder issue." + }, + { + "left": "issue:43525", + "right": "issue:44861", + "accept": false, + "reason": "Missing config attribute vs tied-weight-key attribute crash." + }, + { + "left": "issue:43572", + "right": "issue:44861", + "accept": false, + "reason": "StableLm pad_token_idx regression vs tied-weight-key handling." + }, + { + "left": "issue:31515", + "right": "issue:38617", + "accept": false, + "reason": "Slow checkpoint loading vs an ImportError for a removed config utility." + }, + { + "left": "issue:44360", + "right": "issue:44998", + "accept": false, + "reason": "DSA indexer bug vs unrelated placeholder issue." + }, + { + "left": "issue:44230", + "right": "issue:44623", + "accept": false, + "reason": "FP8 support request/bug vs processor save_pretrained missing files." + }, + { + "left": "issue:30333", + "right": "issue:31515", + "accept": false, + "reason": "MLflow job-reporting issue vs slow from_pretrained loading." + }, + { + "left": "issue:31515", + "right": "issue:36296", + "accept": false, + "reason": "Checkpoint loading slowdown vs tensor-parallel training bug." + }, + { + "left": "issue:31515", + "right": "issue:31795", + "accept": false, + "reason": "Performance regression vs documentation confusion." + }, + { + "left": "issue:44230", + "right": "issue:45310", + "accept": false, + "reason": "FP8 support issue vs Qwen3.5 MoE from_pretrained error; related model family but different bugs." + }, + { + "left": "issue:31515", + "right": "issue:33453", + "accept": false, + "reason": "Slow checkpoint loading vs tokenizer loading regression." + }, + { + "left": "issue:43298", + "right": "issue:44230", + "accept": false, + "reason": "Continuous batching audio-model support vs FP8 inference support." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 3, + "component_count": 16, + "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": "042c13160c230077e5b466365afd7261046ad29da3a0dfa6022a334d5544917f", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1590, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1590, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two different community contribution tracking issues for unrelated features (Flash Attention 2 vs GGUF support). The soft similarity is superficial and should not be treated as a duplicate.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:26350 is the stronger representative of the cluster because it has significantly higher discussion activity and is the more established tracking issue, even though it is not a duplicate of issue:33260.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:26350", + "right": "issue:33260", + "accept": false, + "reason": "Different underlying requests: one is about Flash Attention 2 architecture support, the other is about GGUF architecture support. They share a template and scope style, not 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": 4, + "component_count": 16, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-27088-4", + "nodes": [ + "issue:27088", + "issue:43834", + "issue:44365", + "issue:44370" + ], + "soft_pairs": [ + "issue:27088|issue:44370", + "issue:27088|issue:44365", + "issue:27088|issue:43834", + "issue:43834|issue:44370", + "issue:43834|issue:44365", + "issue:44365|issue:44370" + ], + "prepared_review_unit_hash": "70efd57aa823bc8b35dbfe143ecae77709451c4243c4a4a348e3037b2ecd7c5b", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4453, + "estimated_input_tokens": 1114, + "estimated_eval_tokens": 2484 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4453, + "estimated_input_tokens": 1114, + "estimated_eval_tokens": 2484 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are all i18n translation-request issues with highly similar templated wording, but they do not appear to describe the same concrete translation task. The Turkish request is specific, while the others are generic placeholder issues for potentially different languages.", + "confidence": 0.96, + "canonical_issue_reason": "issue:27088 is the only concrete, language-specific request ([i18n-TR] Turkish); the others are templated placeholder issues and are not clearly the same underlying request.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:27088 is the best representative because it is specific and actionable, whereas the others are generic template variants with no concrete language identified.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:27088", + "right": "issue:44370", + "accept": false, + "reason": "Same translation-template wording, but no evidence they request the same language; one is Turkish and the other is an unspecified placeholder." + }, + { + "left": "issue:27088", + "right": "issue:44365", + "accept": false, + "reason": "Superficially similar i18n request, but the concrete Turkish request is distinct from the generic placeholder template." + }, + { + "left": "issue:27088", + "right": "issue:43834", + "accept": false, + "reason": "Not the same underlying request: Turkish is explicit, while the other issue is a generic language placeholder." + }, + { + "left": "issue:43834", + "right": "issue:44370", + "accept": false, + "reason": "Very similar template text, but the placeholders do not establish the same concrete language or translation task." + }, + { + "left": "issue:43834", + "right": "issue:44365", + "accept": false, + "reason": "These look like separate language-specific translation requests sharing a template, not duplicates of one concrete issue." + }, + { + "left": "issue:44365", + "right": "issue:44370", + "accept": false, + "reason": "Identical template structure is not enough to merge them; the actual target language is unknown and may differ." + } + ] + }, + "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": 16, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-36817-15", + "nodes": [ + "issue:36817", + "issue:36879", + "issue:37015", + "issue:41115", + "issue:41211", + "issue:42503", + "issue:42738", + "issue:43671", + "issue:43837", + "issue:43909", + "issue:44121", + "issue:44205", + "issue:45027", + "issue:45177", + "issue:45322" + ], + "soft_pairs": [ + "issue:36817|issue:41211", + "issue:36817|issue:43671", + "issue:36817|issue:36879", + "issue:36879|issue:41211", + "issue:36879|issue:43909", + "issue:36879|issue:44121", + "issue:36879|issue:43837", + "issue:36879|issue:43671", + "issue:37015|issue:43837", + "issue:37015|issue:44121", + "issue:37015|issue:43909", + "issue:36879|issue:37015", + "issue:37015|issue:42503", + "issue:41115|issue:43837", + "issue:41115|issue:44121", + "issue:41115|issue:42503", + "issue:41115|issue:45027", + "issue:41211|issue:43671", + "issue:42503|issue:43837", + "issue:42503|issue:43671", + "issue:42503|issue:42738", + "issue:42738|issue:45027", + "issue:43671|issue:43837", + "issue:43671|issue:45027", + "issue:43671|issue:45177", + "issue:43671|issue:44121", + "issue:43837|issue:45322", + "issue:43837|issue:43909", + "issue:43909|issue:44205", + "issue:43671|issue:43909", + "issue:44121|issue:45322", + "issue:43837|issue:44121", + "issue:43837|issue:45027", + "issue:37015|issue:45027", + "issue:43837|issue:45177" + ], + "prepared_review_unit_hash": "f557324d9d0ef7b013b18dbf4ce8c9c20ecb4db0f03ab72b04040ff0a764a899", + "component_budget": { + "node_count": 15, + "item_count": 15, + "soft_pair_count": 35, + "serialized_chars": 20707, + "estimated_input_tokens": 5177, + "estimated_eval_tokens": 10610 + }, + "budget": { + "node_count": 15, + "item_count": 15, + "soft_pair_count": 35, + "serialized_chars": 20707, + "estimated_input_tokens": 5177, + "estimated_eval_tokens": 10610 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is a broad false-positive: the issues are all model-add/support requests, but they target different concrete models and often different modalities (BERT-like, detection, OCR, speech, audio, vision). They should not be treated as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": "If forced to pick one representative, issue #42738 is the broadest umbrella-style request and best matches the shared 'model support' theme, though it is not a true duplicate of the rest.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #42738 is the least specific and most general among the set, so it is the closest single representative of the cluster's theme; however, the cluster is too heterogeneous for a strong canonical issue.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36817", + "right": "issue:41211", + "accept": false, + "reason": "Different model families and domains; only the generic 'add model' framing overlaps." + }, + { + "left": "issue:36817", + "right": "issue:43671", + "accept": false, + "reason": "EuroBert config support vs Qwen3 text-to-speech support are unrelated code paths." + }, + { + "left": "issue:36817", + "right": "issue:36879", + "accept": false, + "reason": "BERT config addition vs RF-DETR model support; different concrete models and subsystems." + }, + { + "left": "issue:36879", + "right": "issue:41211", + "accept": false, + "reason": "RF-DETR object detection and DEIMv2 are different models; not the same change." + }, + { + "left": "issue:36879", + "right": "issue:43909", + "accept": false, + "reason": "Object detection request vs audio model request; different integration targets." + }, + { + "left": "issue:36879", + "right": "issue:44121", + "accept": false, + "reason": "Detector support vs sparse transformer architecture request; unrelated code paths." + }, + { + "left": "issue:36879", + "right": "issue:43837", + "accept": false, + "reason": "RF-DETR and Qwen3-ASR are different model types and tasks." + }, + { + "left": "issue:36879", + "right": "issue:43671", + "accept": false, + "reason": "Detector support vs text-to-speech support; not the same underlying bug/change." + }, + { + "left": "issue:37015", + "right": "issue:43837", + "accept": false, + "reason": "NeoBERT support and Qwen3-ASR support are different model integrations." + }, + { + "left": "issue:37015", + "right": "issue:44121", + "accept": false, + "reason": "Distinct model families: NeoBERT vs weight-sparse transformer." + }, + { + "left": "issue:37015", + "right": "issue:43909", + "accept": false, + "reason": "NeoBERT request is unrelated to the LFM2.5 audio model request." + }, + { + "left": "issue:36879", + "right": "issue:37015", + "accept": false, + "reason": "RF-DETR detection support and NeoBERT support are different concrete additions." + }, + { + "left": "issue:37015", + "right": "issue:42503", + "accept": false, + "reason": "Both are model requests, but for different BERT-like families with different implementation needs." + }, + { + "left": "issue:41115", + "right": "issue:43837", + "accept": false, + "reason": "MiniCPM3 architecture support and Qwen3-ASR support are unrelated changes." + }, + { + "left": "issue:41115", + "right": "issue:44121", + "accept": false, + "reason": "MiniCPM3 model architecture addition vs sparse transformer support; no shared code path." + }, + { + "left": "issue:41115", + "right": "issue:42503", + "accept": false, + "reason": "Different model families and different concrete integration work." + }, + { + "left": "issue:41115", + "right": "issue:45027", + "accept": false, + "reason": "MiniCPM3 architecture request vs Voxtral TTS support are different modalities." + }, + { + "left": "issue:41211", + "right": "issue:43671", + "accept": false, + "reason": "DEIMv2 and Qwen3-TTS are unrelated model requests." + }, + { + "left": "issue:42503", + "right": "issue:43837", + "accept": false, + "reason": "ModernVBERT and Qwen3-ASR target different model families and tasks." + }, + { + "left": "issue:42503", + "right": "issue:43671", + "accept": false, + "reason": "ModernVBERT models and Qwen3 TTS are different integrations, not duplicates." + }, + { + "left": "issue:42503", + "right": "issue:42738", + "accept": false, + "reason": "A specific model-family request vs a generic BERT-like-with-RoPE proposal; only thematic overlap." + }, + { + "left": "issue:42738", + "right": "issue:45027", + "accept": false, + "reason": "BERT-like models with RoPE and Voxtral TTS are unrelated model classes." + }, + { + "left": "issue:43671", + "right": "issue:43837", + "accept": false, + "reason": "Both are speech-related, but TTS and ASR are different model support paths." + }, + { + "left": "issue:43671", + "right": "issue:45027", + "accept": false, + "reason": "Different speech model families; not enough to treat as the same change." + }, + { + "left": "issue:43671", + "right": "issue:45177", + "accept": false, + "reason": "Text-to-speech support vs DeepSeek-OCR2 support are different modalities." + }, + { + "left": "issue:43671", + "right": "issue:44121", + "accept": false, + "reason": "Speech model support and sparse transformer support are unrelated." + }, + { + "left": "issue:43837", + "right": "issue:45322", + "accept": false, + "reason": "Qwen3-ASR and EUPE address different model types and tasks." + }, + { + "left": "issue:43837", + "right": "issue:43909", + "accept": false, + "reason": "ASR and audio model requests are related only at a high level, not the same concrete integration." + }, + { + "left": "issue:43909", + "right": "issue:44205", + "accept": false, + "reason": "Audio model support vs SAM3-LiteText support are different changes." + }, + { + "left": "issue:43671", + "right": "issue:43909", + "accept": false, + "reason": "TTS and audio model support are distinct model integrations." + }, + { + "left": "issue:44121", + "right": "issue:45322", + "accept": false, + "reason": "Weight-sparse transformer support and EUPE support are different architectural requests." + }, + { + "left": "issue:43837", + "right": "issue:44121", + "accept": false, + "reason": "ASR support and sparse transformer support are unrelated code paths." + }, + { + "left": "issue:43837", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-ASR and Voxtral TTS are different speech-model integrations." + }, + { + "left": "issue:37015", + "right": "issue:45027", + "accept": false, + "reason": "NeoBERT and Voxtral TTS are unrelated model families." + }, + { + "left": "issue:43837", + "right": "issue:45177", + "accept": false, + "reason": "ASR support vs OCR2 support; different modalities and implementation work." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 6, + "component_count": 16, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-37934-6", + "nodes": [ + "issue:37934", + "issue:42936", + "issue:43240", + "issue:43668", + "issue:43813", + "issue:44031" + ], + "soft_pairs": [ + "issue:37934|issue:43668", + "issue:42936|issue:43240", + "issue:42936|issue:44031", + "issue:42936|issue:43813", + "issue:42936|issue:43668", + "issue:43240|issue:43668", + "issue:43668|issue:44031", + "issue:43668|issue:43813" + ], + "prepared_review_unit_hash": "997778c65039c2e4781f8743225cd174657d41c0387daddd0097f5f3eccdc4c7", + "component_budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 8, + "serialized_chars": 6292, + "estimated_input_tokens": 1573, + "estimated_eval_tokens": 3402 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 8, + "serialized_chars": 6292, + "estimated_input_tokens": 1573, + "estimated_eval_tokens": 3402 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The items are not one duplicate cluster; they span unrelated bugs in norms, Mask2Former loss handling, config typing, a typo, and tokenizer warnings. No soft pair looks like the same underlying issue/change.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:37934", + "right": "issue:43668", + "accept": false, + "reason": "Both mention normalization details, but one is about whether Llama4TextL2Norm should be RMSNorm and the other is a type-hint problem in ModernBERTConfig; different code paths and fixes." + }, + { + "left": "issue:42936", + "right": "issue:43240", + "accept": false, + "reason": "Both touch training/loss plumbing, but one is about Mask2Former's ignore_value being unused and the other is about kwargs not reaching the loss function; not the same bug." + }, + { + "left": "issue:42936", + "right": "issue:44031", + "accept": false, + "reason": "Mask2Former loss handling vs tokenizer regex warnings are unrelated subsystems and changes." + }, + { + "left": "issue:42936", + "right": "issue:43813", + "accept": false, + "reason": "Mask2Former ignore_value usage and a typo in orig_conversion.quantization_operation are unrelated." + }, + { + "left": "issue:42936", + "right": "issue:43668", + "accept": false, + "reason": "Mask2Former loss behavior and a ModernBERTConfig type hint issue are unrelated." + }, + { + "left": "issue:43240", + "right": "issue:43668", + "accept": false, + "reason": "Passing kwargs to a loss function and correcting a config type hint are different problems." + }, + { + "left": "issue:43668", + "right": "issue:44031", + "accept": false, + "reason": "ModernBERTConfig norm_eps typing and tokenizer regex warnings are unrelated." + }, + { + "left": "issue:43668", + "right": "issue:43813", + "accept": false, + "reason": "A config type-hint error and a field-name typo in conversion code 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": 7, + "component_count": 16, + "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": "fc68dbe90644d4dc1c1a0459868931acecac490e7cc26931b7f76888ec100deb", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1480, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1480, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are unrelated: one is about block decoding, the other is a DeepSeek v3.2 support request. They should not be clustered as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Issue #43037 is the more specific and user-facing request, but it is not a duplicate of #41084.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41084", + "right": "issue:43037", + "accept": false, + "reason": "Different underlying topics with no shared code-path or concrete bug/change: block decoding vs model support for DeepSeek v3.2." + } + ] + }, + "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": 16, + "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": "755a27ca33e0c46cf0c3aafb6bd5c4392550366e38bbc2642a179a6763834cb8", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1583, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1583, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related in that both mention custom trainers and Trainer behavior, but they describe different problems: one is about `remove_unused_column` customizability, the other about FLOPS estimation. Not a duplicate pair.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 43598 is the cleaner representative of the cluster because it is a more concrete Trainer customization concern.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43598 is the better standalone issue to keep as the cluster representative: it describes a specific behavior in `Trainer` and is easier to action than the broader FLOPS-estimation request.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43598", + "right": "issue:43600", + "accept": false, + "reason": "Different underlying bugs: one concerns `remove_unused_column` handling, the other FLOPS counting with custom trainers. Shared Trainer/custom trainer context is too broad to treat as 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": 9, + "component_count": 16, + "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": "111108449c22b443adb6fb6853720c22eda422fdeb4ec4ac7e7ed0c6ee362052", + "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 two separate feature requests for different Qwen model families and different AutoModel heads. They are related only by a broad theme of model registration, not by the same underlying bug or change.", + "confidence": 0.94, + "canonical_issue_reason": "Issue #43684 is the broader, more general request about registering Qwen3-Omni with AutoModel and AutoModelForConditionalGeneration, so it is the better anchor if this cluster is used as a loose topical grouping.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #43684 is the most representative issue in the pair because it describes a broader registration gap, while #44405 is a distinct request for AutoModelForSequenceClassification support on Qwen3.5.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43684", + "right": "issue:44405", + "accept": false, + "reason": "Different model families (Qwen3-Omni vs Qwen3.5) and different target heads (conditional generation vs sequence classification); not the same concrete fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 10, + "component_count": 16, + "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": "5bc786159cff1dc22217e4a82e3f9732d407c4e8f8a2f54e4ef89453ecd7b958", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4263, + "estimated_input_tokens": 1066, + "estimated_eval_tokens": 2388 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4263, + "estimated_input_tokens": 1066, + "estimated_eval_tokens": 2388 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Three closed issues share the same placeholder title \"huggingface \" and are likely duplicates; the \"Fett\" issue looks unrelated. No PRs are present.", + "confidence": 0.78, + "canonical_issue_reason": "Issue 43845 is the earliest of the near-identical \"huggingface \" reports, so it is the best representative for that duplicate cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "43845 is the strongest single issue to keep because it is earliest, has the same title as the other likely duplicates, and there is no evidence that 43835 is related.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43835", + "right": "issue:43845", + "accept": false, + "reason": "Different titles with no supporting body or target overlap; \"Fett\" does not appear to be the same report as \"huggingface \"." + }, + { + "left": "issue:43835", + "right": "issue:43846", + "accept": false, + "reason": "No evidence these refer to the same underlying issue; the titles are unrelated." + }, + { + "left": "issue:43835", + "right": "issue:43859", + "accept": false, + "reason": "No indication of the same bug/change; \"Fett\" is distinct from the repeated \"huggingface \" reports." + }, + { + "left": "issue:43845", + "right": "issue:43846", + "accept": true, + "reason": "Identical title, very close creation times, and no contrary evidence; these look like duplicate reports of the same issue." + }, + { + "left": "issue:43845", + "right": "issue:43859", + "accept": true, + "reason": "Same title and clear thematic overlap with the other \"huggingface \" issues; likely the same underlying report." + }, + { + "left": "issue:43846", + "right": "issue:43859", + "accept": true, + "reason": "Identical title and close cluster context indicate the same duplicate issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: three issues share the exact same title, the earliest one is correctly chosen as the canonical representative, and the unrelated 'Fett' issue is kept separate. The soft-edge verdicts are conservative overall, and the note that no PRs are present is supported by the packet." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 11, + "component_count": 16, + "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": "f212e9edab6f94a37e4fcf459fa2c61d7fe84c5c72c43de9d403e057e57aa075", + "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": "Three issues are exact title matches for \"\" and look like the same spam/duplicate report; the documentation and Google LLC titles do not match that pattern.", + "confidence": 0.91, + "canonical_issue_reason": "issue:44371 is the earliest exact \"\" entry and is the cleanest representative of the repeated spam cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44371 best represents the cluster because it is an exact match for the repeated \"\" issues and appears first chronologically.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44371", + "right": "issue:44572", + "accept": true, + "reason": "Same exact title \"\", same low-information spam pattern, so likely duplicate reports of the same artifact." + }, + { + "left": "issue:44371", + "right": "issue:44573", + "accept": true, + "reason": "Same exact title \"\" and same spam-like placeholder content." + }, + { + "left": "issue:44371", + "right": "issue:44418", + "accept": false, + "reason": "Different title and different issue type; documentation enhancement is not the same as the spam placeholder issue." + }, + { + "left": "issue:44371", + "right": "issue:44937", + "accept": false, + "reason": "\"Check out \\\"Google LLC\\\"\" is a different spam/phishing-style title, not the same exact issue as \"\"." + }, + { + "left": "issue:44418", + "right": "issue:44572", + "accept": false, + "reason": "Documentation enhancement suggestion does not match the exact spam placeholder issue." + }, + { + "left": "issue:44418", + "right": "issue:44573", + "accept": false, + "reason": "Different title and different underlying request; not the same issue." + }, + { + "left": "issue:44572", + "right": "issue:44573", + "accept": true, + "reason": "Exact title match \"\" with no other distinguishing information, so they appear to be duplicates." + }, + { + "left": "issue:44572", + "right": "issue:44937", + "accept": false, + "reason": "Different spam wording; no evidence they report the same underlying issue." + }, + { + "left": "issue:44573", + "right": "issue:44937", + "accept": false, + "reason": "Different titles and no shared concrete problem beyond generic spam-like wording." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: three items share the exact title \"\" and the other two clearly do not. The soft-edge accepts are conservative enough given the exact title matches and matching low-information metadata, and the rejects avoid overclaiming on the non- issues." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 12, + "component_count": 16, + "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": "45871a8dfbb2df43a6ce827b7155e31a70bb22f7eecdbfb9434606a3e1ed5de2", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1646, + "estimated_input_tokens": 412, + "estimated_eval_tokens": 1080 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1646, + "estimated_input_tokens": 412, + "estimated_eval_tokens": 1080 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Likely duplicate issue reports with identical titles and very similar body sizes; #45145 is the better canonical representative because it appeared first and already has an inbound reference.", + "confidence": 0.84, + "canonical_issue_reason": "#45145 is the earliest copy and already has an inbound reference, so it is the best canonical issue among these likely duplicates.", + "canonical_pr_reason": null, + "best_issue_reason": "#45145 is the best issue representative because it was created first and is already referenced by another issue.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45145", + "right": "issue:45151", + "accept": true, + "reason": "Identical titles and very similar body lengths strongly suggest the same underlying report, though the bodies themselves are not shown." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The packet supports a strong similarity signal (identical titles, close body lengths, one inbound reference on #45145), but it does not directly show the issue bodies or prove the reports are the same. The writeup is a bit too definitive with phrases like \"near-exact duplicates\" and \"effectively the same issue report.\" A more conservative phrasing such as \"likely duplicate\" would be better grounded." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 13, + "component_count": 16, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45295-2", + "nodes": [ + "issue:45295", + "issue:45373" + ], + "soft_pairs": [], + "prepared_review_unit_hash": "f5fd95fada2f530ab0999e343c17c16266cde0c7182bb22509972ca44827cd49", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 0, + "serialized_chars": 1286, + "estimated_input_tokens": 322, + "estimated_eval_tokens": 900 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 0, + "serialized_chars": 1286, + "estimated_input_tokens": 322, + "estimated_eval_tokens": 900 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues describe the same missing Gemma 4 sequence-classification support; the later, more specific open issue is the better cluster representative.", + "confidence": 0.94, + "canonical_issue_reason": "issue #45373 is the clearest canonical tracker: it is open, explicitly names the missing class `Gemma4ForSequenceClassification`, and states the module-level gap relative to Gemma 2/3.", + "canonical_pr_reason": null, + "best_issue_reason": "issue #45373 best captures the concrete request and likely remaining work, while #45295 is a shorter earlier duplicate with less detail and is already closed.", + "best_pr_reason": null, + "soft_edge_verdicts": [] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 14, + "component_count": 16, + "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": "85a29aca2afae36fb3e827ddf01fd1cce50d95e4b10b94102bb8a0251fbd53ed", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1669, + "estimated_input_tokens": 418, + "estimated_eval_tokens": 1092 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1669, + "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 related only at a high level (both 4-bit inference failures), but they describe different bugs and failure modes, so they should not be deduplicated.", + "confidence": 0.96, + "canonical_issue_reason": "Issue 45674 is the clearer representative because it states a specific configuration regression in BitsAndBytesConfig with an explicit failure mode and broader actionable scope.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45674 is the better standalone issue to keep as the canonical reference for this cluster since it describes a concrete, reproducible code-path problem rather than a model-specific symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45672", + "right": "issue:45674", + "accept": false, + "reason": "Different underlying bugs: 45672 is a Gemma4/audio-module torch.finfo TypeError on uint8 weights, while 45674 is an llm_int8_skip_modules / lm_head exclusion AssertionError in 4-bit inference. Same broad area, but not the same fixable 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": 15, + "component_count": 16, + "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": "62b653a03f02e89e38f8b1ac4a30309d040ed04368ae7665f31337cfa82dc629", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two closed issues are just separate personal-name entries with different names/dates and no shared bug, feature, or code context, so this is not a duplicate cluster.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45762", + "right": "issue:45763", + "accept": false, + "reason": "Same broad surname pattern only; no body, targets, filenames, or discussion linking them to the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 16, + "component_count": 16, + "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": "d2a692d14b0f6a220670e9ef414cc2c352fe070e58015763ee8a18fb98b62cc4", + "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": "These are two unrelated, empty closed issues with personal-name/date titles and no substantive content. They do not appear to describe the same bug or request, so they should not be clustered as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": "No canonical issue: neither item contains actionable problem description, and the titles are unrelated personal-name/date strings rather than the same report.", + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a good cluster anchor because both are effectively empty/spam-like and provide no meaningful defect or feature context.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45764", + "right": "issue:45765", + "accept": false, + "reason": "Reject: the titles reference different people and different dates, and there is no body content or shared technical context indicating the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 1, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-39895-13", + "nodes": [ + "pull_request:39895", + "pull_request:41855", + "pull_request:42668", + "pull_request:43251", + "pull_request:43254", + "pull_request:43424", + "pull_request:43823", + "pull_request:43995", + "pull_request:44827", + "pull_request:45170", + "pull_request:45213", + "pull_request:45655", + "pull_request:45795" + ], + "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:44827", + "pull_request:43995|pull_request:45213", + "pull_request:44827|pull_request:45170", + "pull_request:44827|pull_request:45795", + "pull_request:44827|pull_request:45655", + "pull_request:44827|pull_request:45213", + "pull_request:43823|pull_request:45213", + "pull_request:43823|pull_request:45795", + "pull_request:43251|pull_request:43254" + ], + "prepared_review_unit_hash": "73a5e5c302f1fa1634d1bc3111598f4913e698dbd6cb4de6353f8f2121035ab5", + "component_budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 19, + "serialized_chars": 35208, + "estimated_input_tokens": 8802, + "estimated_eval_tokens": 17860 + }, + "budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 19, + "serialized_chars": 35208, + "estimated_input_tokens": 8802, + "estimated_eval_tokens": 17860 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly a grab bag of unrelated PRs across model additions, docs, and refactors. The only true duplicate pair is the two fixed_cross_entropy changes (43251/43254), which implement the same kwargs pass-through to nn.functional.cross_entropy.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43251 is the better canonical choice for the duplicate loss fix: it covers the same fixed_cross_entropy change as 43254, is a bit more complete, and explicitly links the tracking issue 43240.", + "best_issue_reason": null, + "best_pr_reason": "PR 43251 is the best representative of the only real duplicate pair in this cluster; the rest are unrelated, and 43251 is the more complete version of the shared cross-entropy kwargs fix.", + "soft_edge_verdicts": [ + { + "left": "pull_request:39895", + "right": "pull_request:43424", + "accept": false, + "reason": "Different changes: VideoPrism model addition vs an executorch export test." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43424", + "accept": false, + "reason": "Unrelated areas: Mistral tokenizer methods vs executorch dynamic-shape testing." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43823", + "accept": false, + "reason": "Different workstreams: tokenizer API additions vs adding MobileLLM." + }, + { + "left": "pull_request:42668", + "right": "pull_request:43823", + "accept": false, + "reason": "Processor auto-loading changes are not the same underlying change as adding MobileLLM." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43823", + "accept": false, + "reason": "Cross-entropy kwargs fix and MobileLLM model addition are unrelated." + }, + { + "left": "pull_request:43251", + "right": "pull_request:44827", + "accept": false, + "reason": "Loss helper kwargs fix is unrelated to Mistral4 test fixes and model wiring." + }, + { + "left": "pull_request:43424", + "right": "pull_request:44827", + "accept": false, + "reason": "Executorch export tests and Mistral4 test fixes address different code paths." + }, + { + "left": "pull_request:43424", + "right": "pull_request:43823", + "accept": false, + "reason": "Different artifacts: executorch export test vs new MobileLLM model implementation." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45655", + "accept": false, + "reason": "New model addition is unrelated to config resolution order in modeling_utils." + }, + { + "left": "pull_request:43823", + "right": "pull_request:43995", + "accept": false, + "reason": "MobileLLM implementation and Falcon output-interface refactor are different changes." + }, + { + "left": "pull_request:43995", + "right": "pull_request:44827", + "accept": false, + "reason": "Falcon refactor and Mistral4 test fixes do not target the same bug." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45213", + "accept": false, + "reason": "Model refactor vs model-creation skill content are not the same underlying change." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45170", + "accept": false, + "reason": "Mistral4 test fixes and layrnorm->layernorm renaming are different fixes." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45795", + "accept": false, + "reason": "Mistral4 test fixes are unrelated to docs for audio/video processors." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45655", + "accept": false, + "reason": "Different problems: Mistral4 tests vs config resolution order bug." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45213", + "accept": false, + "reason": "Test-fix PR and model-creation skill PR are unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45213", + "accept": false, + "reason": "Adding MobileLLM is not the same underlying change as adding a new-model skill doc." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45795", + "accept": false, + "reason": "MobileLLM model addition and docs for processors are different topics." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43254", + "accept": true, + "reason": "Both PRs patch the same helper, fixed_cross_entropy, to forward the same supported kwargs to nn.functional.cross_entropy; they are mergeable into one change." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The analysis is grounded in the packet and stays conservative. The only accepted soft edge, 43251/43254, is well supported by shared file/issue target and the same fixed_cross_entropy kwargs fix, and the other pairs are reasonably rejected as unrelated." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 2, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-40546-2", + "nodes": [ + "pull_request:40546", + "pull_request:42781" + ], + "soft_pairs": [ + "pull_request:40546|pull_request:42781" + ], + "prepared_review_unit_hash": "186b72b8074839b415d5d697acda9b17f5abdefadee07ba752039a23e06dfa21", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7225, + "estimated_input_tokens": 1807, + "estimated_eval_tokens": 3870 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7225, + "estimated_input_tokens": 1807, + "estimated_eval_tokens": 3870 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both pull requests are about VibeVoice, but they are not duplicates: one is the initial VibeVoice integration and the other adds VibeVoice Realtime plus an acoustic tokenizer and related generation/model changes.", + "confidence": 0.87, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 40546 is the broader original VibeVoice implementation and appears to be the primary integration PR; PR 42781 is a later, more specialized follow-up for the realtime variant.", + "best_issue_reason": null, + "best_pr_reason": "PR 40546 is the best representative of the cluster because it introduces the core VibeVoice support that the other PR builds on.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40546", + "right": "pull_request:42781", + "accept": false, + "reason": "They share the VibeVoice subsystem, but 42781 adds a distinct realtime/acoustic-tokenizer feature set rather than fixing or implementing the same concrete code path as 40546. These are related follow-up changes, not one mergeable duplicate PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 3, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-40755-2", + "nodes": [ + "pull_request:40755", + "pull_request:41224" + ], + "soft_pairs": [ + "pull_request:40755|pull_request:41224" + ], + "prepared_review_unit_hash": "9c294abaa62b509f6b7053e1b10763b86dd7da870fb1d14a0563baa59f52460f", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4674, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4674, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two pull requests are about different features in different models: TimesFM covariate forecasting vs. DINOv3 image-classification support. They share only a loose \u201cmodel feature addition\u201d theme, not the same code-path or bug, so they are not duplicates.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "No true canonical PR for this cluster because the PRs target unrelated models and independent changes. If forced to pick a representative, PR 40755 is broader and more feature-complete, but it is not a duplicate of PR 41224.", + "best_issue_reason": null, + "best_pr_reason": "PR 40755 is the stronger standalone representative of the cluster only because it has the larger implementation footprint and includes docs/tests for a concrete feature addition. However, it should not be merged with PR 41224 as a duplicate.", + "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 code-path or 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": 4, + "component_count": 26, + "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": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-41797-3", + "nodes": [ + "pull_request:41797", + "pull_request:45186", + "pull_request:45630" + ], + "soft_pairs": [ + "pull_request:41797|pull_request:45186", + "pull_request:45186|pull_request:45630" + ], + "prepared_review_unit_hash": "51307f7b5670046e28b0a785f8298bdea7ec719829f47b5014f713358b1433a5", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10429, + "estimated_input_tokens": 2608, + "estimated_eval_tokens": 5472 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10429, + "estimated_input_tokens": 2608, + "estimated_eval_tokens": 5472 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "All three items are separate model-addition PRs for different architectures (DeepSeek OCR, Isaac, Kimi2-6). They share boilerplate integration files, but not the same underlying change, so they should not be clustered as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45630 is the strongest cluster representative only in the sense that it is the most mature and comprehensive of the three, with the highest review activity and the broadest set of integration updates.", + "best_issue_reason": null, + "best_pr_reason": "PR #45630 is the best single representative if one must be chosen: it is a fully fleshed-out model addition with extensive integration coverage and substantial review attention.", + "soft_edge_verdicts": [ + { + "left": "pull_request:41797", + "right": "pull_request:45186", + "accept": false, + "reason": "Different model families and different code paths: DeepSeek OCR vs Isaac. Shared auto-registration/docs scaffolding is boilerplate, not evidence of the same change." + }, + { + "left": "pull_request:45186", + "right": "pull_request:45630", + "accept": false, + "reason": "These add different models (Isaac vs Kimi2-6) with distinct configurations, processors, and tests. Similar framework integration files do not make them mergeable into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 6, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-42975-2", + "nodes": [ + "pull_request:42975", + "pull_request:42976" + ], + "soft_pairs": [ + "pull_request:42975|pull_request:42976" + ], + "prepared_review_unit_hash": "497a04fc187db1ffa0866539390949573186db0b32fbf7d3d5ea9490d24b1fac", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5935, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5935, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two separate workflow-maintenance PRs: one updates multiple GitHub Actions for Node 24 compatibility, the other upgrades Docker build actions to v6. They overlap in workflow files but address different dependencies and problems.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 42975 is the better canonical representative because it has the clearest, broader purpose (Node 24 compatibility) and covers a wider set of GitHub Actions updates.", + "best_issue_reason": null, + "best_pr_reason": "PR 42975 is the strongest representative for the cluster since its title and diff clearly describe the main class of change: repository-wide GitHub Actions version upgrades for compatibility.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42975", + "right": "pull_request:42976", + "accept": false, + "reason": "These are not the same underlying change: 42975 updates checkout/cache/upload-artifact across workflows for Node 24 compatibility, while 42976 only bumps docker/build-push-action to v6 in docker-image workflows. Overlapping files alone is not enough to treat them as duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 7, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-42978-7", + "nodes": [ + "pull_request:42978", + "pull_request:43448", + "pull_request:43451", + "pull_request:43973", + "pull_request:44601", + "pull_request:45490", + "pull_request:45838" + ], + "soft_pairs": [ + "pull_request:42978|pull_request:43448", + "pull_request:42978|pull_request:43451", + "pull_request:42978|pull_request:45490", + "pull_request:43448|pull_request:43451", + "pull_request:43448|pull_request:45490", + "pull_request:43451|pull_request:45490", + "pull_request:43973|pull_request:45838", + "pull_request:44601|pull_request:45838", + "pull_request:45490|pull_request:45838" + ], + "prepared_review_unit_hash": "85c28a5c33abc32549e25199ab3f67c2b6394446237b38dc44512b0736ed7bb7", + "component_budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 9, + "serialized_chars": 23515, + "estimated_input_tokens": 5879, + "estimated_eval_tokens": 12014 + }, + "budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 9, + "serialized_chars": 23515, + "estimated_input_tokens": 5879, + "estimated_eval_tokens": 12014 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is heterogeneous: several unrelated model-addition PRs plus one distributed pipeline-parallel PR. They share repository plumbing patterns (docs, auto mappings) but do not look like duplicates of the same bug or change.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:42978", + "right": "pull_request:43448", + "accept": false, + "reason": "Different model integrations (ViT NEPA vs Molmo); shared docs/auto-registry edits are generic model-onboarding boilerplate, not the same change." + }, + { + "left": "pull_request:42978", + "right": "pull_request:43451", + "accept": false, + "reason": "Different model additions (ViT NEPA vs Molmo2) with different files, configs, and tests; not one underlying bug or feature." + }, + { + "left": "pull_request:42978", + "right": "pull_request:45490", + "accept": false, + "reason": "ViT NEPA and CTSM are distinct model-support PRs; overlap is only in shared registration/docs scaffolding." + }, + { + "left": "pull_request:43448", + "right": "pull_request:43451", + "accept": false, + "reason": "Molmo and Molmo2 are related by naming but are separate model families with different codepaths and artifacts; not mergeable as one PR." + }, + { + "left": "pull_request:43448", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo vs CTSM are unrelated model additions; same broad subsystem, but not the same concrete change." + }, + { + "left": "pull_request:43451", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo2 and CTSM are different model introductions with no shared bug or implementation target." + }, + { + "left": "pull_request:43973", + "right": "pull_request:45838", + "accept": false, + "reason": "Lfm2 Audio and PP-OCRv6 are unrelated model-support PRs; both touch docs/auto wiring, but that is not enough to treat them as duplicates." + }, + { + "left": "pull_request:44601", + "right": "pull_request:45838", + "accept": false, + "reason": "Pipeline-parallel core loading support is a distributed-systems change, while PP-OCRv6 is a model support PR; they do not fix the same code-path problem." + }, + { + "left": "pull_request:45490", + "right": "pull_request:45838", + "accept": false, + "reason": "CTSM and PP-OCRv6 are different model additions; shared repo plumbing does not indicate 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": 8, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43291-3", + "nodes": [ + "pull_request:43291", + "pull_request:43665", + "pull_request:44815" + ], + "soft_pairs": [ + "pull_request:43291|pull_request:43665", + "pull_request:43291|pull_request:44815", + "pull_request:43665|pull_request:44815" + ], + "prepared_review_unit_hash": "29b0f7d250b5dce5d94af3fd00f266b959de8dcb9e623bd2930654d4e8ad0ca4", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7093, + "estimated_input_tokens": 1774, + "estimated_eval_tokens": 3804 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7093, + "estimated_input_tokens": 1774, + "estimated_eval_tokens": 3804 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The three pull requests are not duplicates of one another. Two are mostly test-only changes, while the third is a substantive dequantization/loading fix in a different subsystem. No soft edge should be merged.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "pull_request:44815 is the most substantive and coherent PR: it fixes a concrete dequant/loading path and includes the corresponding implementation and test updates. The other two are narrow test edits without the same underlying code change.", + "best_issue_reason": null, + "best_pr_reason": "pull_request:44815 is the best standalone PR because it addresses an actual code-path defect rather than only altering tests, and its scope is internally consistent across the touched files.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43291", + "right": "pull_request:43665", + "accept": false, + "reason": "Both are PRs with test changes, but they target different models and different behavior; there is no shared underlying bug or fix path." + }, + { + "left": "pull_request:43291", + "right": "pull_request:44815", + "accept": false, + "reason": "Whisper tokenization/test cleanup and FP8 dequant/loading fixes are unrelated changes in different subsystems." + }, + { + "left": "pull_request:43665", + "right": "pull_request:44815", + "accept": false, + "reason": "The CLIP/ViT test-only edits do not match the FP8 dequantization fix; they are not the same bug or mergeable as one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 9, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43488-2", + "nodes": [ + "pull_request:43488", + "pull_request:43532" + ], + "soft_pairs": [ + "pull_request:43488|pull_request:43532" + ], + "prepared_review_unit_hash": "5bee5ec7d4b18827eb0a26ce16231d4bbe814ea80d470732b83bb249052830ce", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4563, + "estimated_input_tokens": 1141, + "estimated_eval_tokens": 2538 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4563, + "estimated_input_tokens": 1141, + "estimated_eval_tokens": 2538 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two pull requests are not duplicates: one is a synthetic bot-check/version-bump PR, while the other is a larger CI/testing workflow change with unrelated file edits.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43532 is the better canonical representative because it contains the substantive workflow/testing changes; PR 43488 is a narrow repository-bot/test trigger PR and is not a good representative for the cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR 43532 is the most meaningful PR in the set, with broader and more realistic code changes across CI and test utilities. PR 43488 is clearly a non-merge bot-check change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43488", + "right": "pull_request:43532", + "accept": false, + "reason": "They do not appear to fix the same bug or implement the same change. 43488 is a version/dependency tweak for bot checking, while 43532 changes CI workflow and testing utility code plus related test behavior." + } + ] + }, + "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": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43612-2", + "nodes": [ + "pull_request:43612", + "pull_request:43613" + ], + "soft_pairs": [ + "pull_request:43612|pull_request:43613" + ], + "prepared_review_unit_hash": "61504846732f3eaccb2ac117b573fc3f190a7473362c23696e8772008dcd10af", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6083, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6083, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two related but distinct PRs: both introduce new segmentation pipelines and share some registration/docs scaffolding, but they target different tasks, model families, and code paths.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: 43612 adds Promptable Concept Segmentation, while 43613 adds Promptable Visual Segmentation; the overlap is mostly shared pipeline/documentation plumbing, not the same underlying change.", + "best_issue_reason": null, + "best_pr_reason": "No single best PR for a duplicate cluster here; they are separate feature additions rather than one concrete bug fix or mergeable duplicate change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43612", + "right": "pull_request:43613", + "accept": false, + "reason": "Reject: both are new pipeline additions, but they implement different segmentation tasks with different model/test targets, so they are not the same underlying change and should not be merged as duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 11, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43651-2", + "nodes": [ + "pull_request:43651", + "pull_request:43663" + ], + "soft_pairs": [ + "pull_request:43651|pull_request:43663" + ], + "prepared_review_unit_hash": "9c35be50e5843627e3c42b5f0553aeba729f0636e8483de637ed7b81ec403ab2", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4292, + "estimated_input_tokens": 1073, + "estimated_eval_tokens": 2402 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4292, + "estimated_input_tokens": 1073, + "estimated_eval_tokens": 2402 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs in `trainer.py`, but they address different Trainer extension points: gradient-accumulation loss scaling vs. signature-column filtering for custom trainers/PEFT. They share a file and broad subsystem, but not the same underlying change.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR: these are separate trainer customizations with different code paths and behavior changes.", + "best_issue_reason": null, + "best_pr_reason": "PR #43651 is the more concrete correctness fix, since it changes actual loss-scaling behavior; PR #43663 is a separate extensibility refactor.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43651", + "right": "pull_request:43663", + "accept": false, + "reason": "Different features and code paths: #43651 controls gradient-accumulation loss scaling, while #43663 adds an override hook for signature-column filtering. They are not the same bug/change and would not plausibly merge into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 12, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43656-3", + "nodes": [ + "pull_request:43656", + "pull_request:43836", + "pull_request:43842" + ], + "soft_pairs": [ + "pull_request:43656|pull_request:43842", + "pull_request:43836|pull_request:43842" + ], + "prepared_review_unit_hash": "74962995867f55bdf88a3863ad7fb3b7e19e38d3cc68d86af2132354c7ff75d3", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 5992, + "estimated_input_tokens": 1498, + "estimated_eval_tokens": 3252 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 5992, + "estimated_input_tokens": 1498, + "estimated_eval_tokens": 3252 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "PR #43842 is the cleanest representative of the TypeAdapter NameError fix in `src/transformers/cli/serve.py`. PR #43836 is plausibly the same underlying issue and shares issue target #43824, but PR #43656 is only loosely similar and is tied to a different target (#43576), so it should not be merged into the same duplicate cluster with the others.", + "confidence": 0.74, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #43842 is the most direct and narrowly scoped fix for the TypeAdapter NameError, with an explicit link to issue #43824 and no extra refactoring beyond the bug fix.", + "best_issue_reason": null, + "best_pr_reason": "PR #43842 is the best overall representative because it addresses the concrete runtime typing failure with the smallest focused change and the clearest issue linkage.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43656", + "right": "pull_request:43842", + "accept": false, + "reason": "These only share broad similarity in `serve.py`; they point to different issue targets (#43576 vs #43824), so the packet does not support treating them as the same bugfix or one mergeable PR." + }, + { + "left": "pull_request:43836", + "right": "pull_request:43842", + "accept": true, + "reason": "Both target the same `serve.py` TypeAdapter NameError and share issue target #43824, so they plausibly fix the same underlying code-path problem despite different implementation styles." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The writeup overstates the evidence for PR #43656 vs #43842. Those two only share soft similarity and even point to different explicit issue targets (43576 vs 43824), so the packet does not firmly support treating them as the same bugfix or as cleanly mergeable into one PR. #43836 vs #43842 is better grounded because they share issue target 43824, but the summary still collapses all three PRs into one fix more strongly than the packet justifies." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 13, + "component_count": 26, + "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": 26, + "review_unit_index": 1, + "review_unit_count": 2, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43996-28", + "nodes": [ + "pull_request:43996", + "pull_request:43997", + "pull_request:44007", + "pull_request:44010", + "pull_request:44013", + "pull_request:44017", + "pull_request:44018", + "pull_request:44019", + "pull_request:44044", + "pull_request:44056", + "pull_request:44059", + "pull_request:44066", + "pull_request:44068", + "pull_request:44071", + "pull_request:44072", + "pull_request:44073", + "pull_request:44074", + "pull_request:44076", + "pull_request:44085", + "pull_request:44086", + "pull_request:44098", + "pull_request:44101", + "pull_request:44116", + "pull_request:44129", + "pull_request:44154", + "pull_request:44161", + "pull_request:44722" + ], + "soft_pairs": [ + "pull_request:43996|pull_request:44085", + "pull_request:43996|pull_request:44044", + "pull_request:44066|pull_request:44085", + "pull_request:44007|pull_request:44072", + "pull_request:44072|pull_request:44722", + "pull_request:44066|pull_request:44072", + "pull_request:44013|pull_request:44044", + "pull_request:44066|pull_request:44086", + "pull_request:44066|pull_request:44071", + "pull_request:44018|pull_request:44068", + "pull_request:44066|pull_request:44068", + "pull_request:44018|pull_request:44066", + "pull_request:44019|pull_request:44722", + "pull_request:44086|pull_request:44722", + "pull_request:44019|pull_request:44085", + "pull_request:44018|pull_request:44019", + "pull_request:44019|pull_request:44071", + "pull_request:44068|pull_request:44116", + "pull_request:44068|pull_request:44722", + "pull_request:44018|pull_request:44116", + "pull_request:44071|pull_request:44722", + "pull_request:44018|pull_request:44086", + "pull_request:44018|pull_request:44085", + "pull_request:44068|pull_request:44086", + "pull_request:44071|pull_request:44085", + "pull_request:44059|pull_request:44086", + "pull_request:44085|pull_request:44086", + "pull_request:44059|pull_request:44722", + "pull_request:44098|pull_request:44101", + "pull_request:44018|pull_request:44071", + "pull_request:44017|pull_request:44161", + "pull_request:44017|pull_request:44722", + "pull_request:44017|pull_request:44073", + "pull_request:44068|pull_request:44071", + "pull_request:44071|pull_request:44086", + "pull_request:43997|pull_request:44073", + "pull_request:44017|pull_request:44071", + "pull_request:44073|pull_request:44101", + "pull_request:44059|pull_request:44071", + "pull_request:44073|pull_request:44161", + "pull_request:44056|pull_request:44101", + "pull_request:44059|pull_request:44068", + "pull_request:44018|pull_request:44059", + "pull_request:44101|pull_request:44161", + "pull_request:43997|pull_request:44017", + "pull_request:43997|pull_request:44056", + "pull_request:44071|pull_request:44076", + "pull_request:44074|pull_request:44101", + "pull_request:44056|pull_request:44161", + "pull_request:43997|pull_request:44018", + "pull_request:43997|pull_request:44161", + "pull_request:44074|pull_request:44161", + "pull_request:44018|pull_request:44161", + "pull_request:43997|pull_request:44071", + "pull_request:44010|pull_request:44071", + "pull_request:44010|pull_request:44073", + "pull_request:44018|pull_request:44129", + "pull_request:43997|pull_request:44074", + "pull_request:43997|pull_request:44129", + "pull_request:44071|pull_request:44129", + "pull_request:44010|pull_request:44018", + "pull_request:43996|pull_request:44161", + "pull_request:43996|pull_request:44017", + "pull_request:44010|pull_request:44044", + "pull_request:44044|pull_request:44074", + "pull_request:44066|pull_request:44722", + "pull_request:44085|pull_request:44722", + "pull_request:44085|pull_request:44129", + "pull_request:43996|pull_request:44007", + "pull_request:43996|pull_request:44129", + "pull_request:44044|pull_request:44129", + "pull_request:44044|pull_request:44154" + ], + "prepared_review_unit_hash": "b51f864156a54e40a217adeb8426dd54ef126b08d3e0e9fa90c1b3ebca12470d", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 78, + "serialized_chars": 84896, + "estimated_input_tokens": 21224, + "estimated_eval_tokens": 42704 + }, + "budget": { + "node_count": 27, + "item_count": 27, + "soft_pair_count": 72, + "serialized_chars": 75093, + "estimated_input_tokens": 18774, + "estimated_eval_tokens": 37804 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a broad standardized output-tracing refactor umbrella across many model families. The true duplicates are the same-model GPT-J/GPT-Neo PR pairs; most other links are just shared refactor theme. I revised the PR anchor to #44722, which is the clearest GPT-J/CodeGen representative in the packet.", + "confidence": 0.81, + "canonical_issue_reason": "Issue #43979 is the shared umbrella tracker for the standardized output-tracing refactor across these PRs.", + "canonical_pr_reason": "PR #44722 is the strongest representative PR because it explicitly targets #43979 and covers both GPT-J and CodeGen, matching the central code-path change in this cluster.", + "best_issue_reason": "No standalone issue artifact is present; #43979 is the best issue-level anchor because the true duplicate PRs all point back to it.", + "best_pr_reason": "PR #44722 is the clearest anchor for the shared GPT-J/CodeGen refactor: it is explicitly linked to #43979 and spans the same concrete files as the other accepted GPT-J/CodeGen PRs.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43996", + "right": "pull_request:44085", + "accept": false, + "reason": "Shared tracker only; CVT/FNet and GPT-J are different concrete code paths." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44044", + "accept": false, + "reason": "Different model families; the overlap is only the generic output-tracing theme." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44085", + "accept": true, + "reason": "Same GPT-J output-tracing refactor in the same file; mergeable into one PR." + }, + { + "left": "pull_request:44007", + "right": "pull_request:44072", + "accept": false, + "reason": "RegNet/ResNet-style refactor vs EfficientNet; not the same underlying change." + }, + { + "left": "pull_request:44072", + "right": "pull_request:44722", + "accept": false, + "reason": "EfficientNet and GPT-J/CodeGen are unrelated concrete code paths." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44072", + "accept": false, + "reason": "GPT-J/CodeGen vs EfficientNet; only the broad refactor pattern overlaps." + }, + { + "left": "pull_request:44013", + "right": "pull_request:44044", + "accept": false, + "reason": "MobileNetV2 and DeBERTa are separate model implementations." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J/CodeGen and MGP-STR are different code paths." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-J/CodeGen and MPT are different model implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44068", + "accept": true, + "reason": "Both are GPT-Neo output-tracing refactors against the same file set." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-J/CodeGen vs GPT-Neo; same refactor theme only." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44066", + "accept": false, + "reason": "GPT-Neo and GPT-J/CodeGen are different model code paths." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44722", + "accept": false, + "reason": "ResNet vs GPT-J/CodeGen; no shared concrete bug or fix." + }, + { + "left": "pull_request:44086", + "right": "pull_request:44722", + "accept": false, + "reason": "MGP-STR and GPT-J/CodeGen are unrelated implementations." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44085", + "accept": false, + "reason": "ResNet and GPT-J are different model families." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44019", + "accept": false, + "reason": "GPT-Neo and ResNet are unrelated concrete changes." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44071", + "accept": false, + "reason": "ResNet vs MPT; only the output-tracing theme matches." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert/XLM are unrelated model code paths." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-Neo vs GPT-J/CodeGen; same migration pattern, different code." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert/XLM are different implementations." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44722", + "accept": false, + "reason": "MPT and GPT-J/CodeGen are different model code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo and MGP-STR are separate model implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo and GPT-J are different model families." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo and MGP-STR are unrelated concrete code paths." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44085", + "accept": false, + "reason": "MPT and GPT-J are different models." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-2 and MGP-STR are unrelated implementations." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J and MGP-STR are different concrete code paths." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-2 vs GPT-J/CodeGen; only the generic refactor theme overlaps." + }, + { + "left": "pull_request:44098", + "right": "pull_request:44101", + "accept": false, + "reason": "ViLT and Flaubert/XLM are unrelated model code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo and MPT are different model implementations." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44161", + "accept": false, + "reason": "Segformer and LongT5 are different model families." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44722", + "accept": false, + "reason": "Segformer and GPT-J/CodeGen are unrelated concrete code paths." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44073", + "accept": false, + "reason": "Segformer and VisualBert are separate implementations." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo and MPT are different models." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44086", + "accept": false, + "reason": "MPT and MGP-STR are different code paths." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44073", + "accept": false, + "reason": "RegNet and VisualBert are unrelated model implementations." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44071", + "accept": false, + "reason": "Segformer and MPT are different concrete changes." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44101", + "accept": false, + "reason": "VisualBert and Flaubert/XLM are unrelated code paths." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44101", + "accept": false, + "reason": "MPNet and Flaubert/XLM are different model implementations." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-2 and GPT-Neo are different model families." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-2 and MPT are unrelated concrete code paths." + }, + { + "left": "pull_request:44101", + "right": "pull_request:44161", + "accept": false, + "reason": "Flaubert/XLM and LongT5 are different model families." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44017", + "accept": false, + "reason": "RegNet and Segformer are different models." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44056", + "accept": false, + "reason": "RegNet and MPNet are unrelated implementations." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44076", + "accept": false, + "reason": "MPT and ImageGPT are different concrete code paths." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44101", + "accept": false, + "reason": "TextNet vs Flaubert/XLM; only the broad refactor theme matches." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44161", + "accept": false, + "reason": "MPNet and LongT5 are different model implementations." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44018", + "accept": false, + "reason": "RegNet and GPT-Neo are unrelated concrete changes." + }, + { + "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 different model families." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44059", + "accept": false, + "reason": "GPT-Neo and GPT-2 are different model implementations." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44071", + "accept": false, + "reason": "RegNet and MPT are unrelated model code paths." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44074", + "accept": false, + "reason": "RegNet and TextNet are different models." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44071", + "accept": false, + "reason": "SqueezeBert and MPT are different code paths." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44073", + "accept": false, + "reason": "SqueezeBert and VisualBert are unrelated implementations." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44018", + "accept": false, + "reason": "SqueezeBert vs GPT-Neo; same refactor theme only." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44044", + "accept": false, + "reason": "SqueezeBert and DeBERTa are different model code paths." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44074", + "accept": false, + "reason": "DeBERTa and TextNet are unrelated models." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J/CodeGen output-tracing refactor; these are effectively duplicate PRs." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44722", + "accept": true, + "reason": "Both touch the same GPT-J output-tracing code path; 44722 just includes an extra CodeGen slice." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-J and SpeechT5 are different model families." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44007", + "accept": false, + "reason": "CVT/FNet and RegNet/ResNet/RT-DETR are different implementations." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44129", + "accept": false, + "reason": "Vision backbone refactor vs SpeechT5; not the same change." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44129", + "accept": false, + "reason": "DeBERTa and SpeechT5 are unrelated concrete code paths." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44154", + "accept": false, + "reason": "DeBERTa and Vits are different models; same theme only." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The cluster summary is mostly conservative, but the PR anchoring overreaches. In particular, the claim that PR #44066 is broader/more central than later GPT-J-only follow-ups is not supported by the packet: PR #44722 also touches both `codegen` and `gptj`, is explicitly linked to #43979, and looks at least as central. That makes the canonical PR reasoning insufficiently grounded. The true duplicate/mergeable pairs you marked are plausible, but the canonical PR justification should be revised to avoid overstating evidence." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 14, + "component_count": 26, + "review_unit_index": 2, + "review_unit_count": 2, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43996-28", + "nodes": [ + "pull_request:43997", + "pull_request:44010", + "pull_request:44024", + "pull_request:44056", + "pull_request:44073", + "pull_request:44074", + "pull_request:44076", + "pull_request:44129" + ], + "soft_pairs": [ + "pull_request:44024|pull_request:44076", + "pull_request:44073|pull_request:44074", + "pull_request:44056|pull_request:44073", + "pull_request:44056|pull_request:44074", + "pull_request:43997|pull_request:44010", + "pull_request:44010|pull_request:44129" + ], + "prepared_review_unit_hash": "4b81b849860b1ba480643613dfc0b5209e578a2815cd6d3b2c4ee6c8d995e3b5", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 78, + "serialized_chars": 84896, + "estimated_input_tokens": 21224, + "estimated_eval_tokens": 42704 + }, + "budget": { + "node_count": 8, + "item_count": 8, + "soft_pair_count": 6, + "serialized_chars": 17124, + "estimated_input_tokens": 4281, + "estimated_eval_tokens": 8818 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These PRs are all model-specific output-tracing refactors, but they touch different code paths and should not be deduplicated across models. They look like a family of similar housekeeping changes rather than one shared bug fix.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44056 is the clearest representative of the cluster: it has a focused, well-described capture_outputs refactor and adds the model-specific _can_record_outputs wiring in one place.", + "best_issue_reason": null, + "best_pr_reason": "PR 44056 is the best overall representative because its title and diff most directly match the common standardized-output-tracing theme without being overly narrow or tangled with unrelated changes.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44024", + "right": "pull_request:44076", + "accept": false, + "reason": "Same broad refactor theme, but different models and different concrete code paths (FocalNet vs ImageGPT). Not the same underlying change." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44074", + "accept": false, + "reason": "Both modernize output handling, but VisualBert and TextNet are separate implementations with different behavior changes. Too broad to treat as duplicates." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44073", + "accept": false, + "reason": "Both are standardized output tracing refactors, but they target different model internals (MPNet vs VisualBert) and are not mergeable as one PR." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44074", + "accept": false, + "reason": "Different models and different output-shape/API adjustments; similar pattern, but not the same concrete fix." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44010", + "accept": false, + "reason": "RegNet and SqueezeBert are unrelated model code paths. Shared decorator imports are not enough to make these duplicates." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44129", + "accept": false, + "reason": "SqueezeBert vs SpeechT5 are different model implementations, and only SpeechT5 is explicitly tied to issue 43979. Not a duplicate pair." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 15, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43998-13", + "nodes": [ + "pull_request:43998", + "pull_request:43999", + "pull_request:44000", + "pull_request:44001", + "pull_request:44002", + "pull_request:44003", + "pull_request:44004", + "pull_request:44025", + "pull_request:44026", + "pull_request:44027", + "pull_request:44028", + "pull_request:44029", + "pull_request:44030" + ], + "soft_pairs": [ + "pull_request:43998|pull_request:44002", + "pull_request:43998|pull_request:44000", + "pull_request:43998|pull_request:44001", + "pull_request:43998|pull_request:44003", + "pull_request:43998|pull_request:44029", + "pull_request:43999|pull_request:44001", + "pull_request:43999|pull_request:44002", + "pull_request:43999|pull_request:44025", + "pull_request:43999|pull_request:44026", + "pull_request:43999|pull_request:44027", + "pull_request:44000|pull_request:44001", + "pull_request:44000|pull_request:44002", + "pull_request:44000|pull_request:44003", + "pull_request:44000|pull_request:44025", + "pull_request:44001|pull_request:44002", + "pull_request:44001|pull_request:44003", + "pull_request:44001|pull_request:44025", + "pull_request:44002|pull_request:44029", + "pull_request:44002|pull_request:44003", + "pull_request:44003|pull_request:44028", + "pull_request:44003|pull_request:44004", + "pull_request:43998|pull_request:44004", + "pull_request:44001|pull_request:44004", + "pull_request:44002|pull_request:44004", + "pull_request:44000|pull_request:44004", + "pull_request:44025|pull_request:44026", + "pull_request:44025|pull_request:44027", + "pull_request:44025|pull_request:44028", + "pull_request:44025|pull_request:44029", + "pull_request:44025|pull_request:44030", + "pull_request:44026|pull_request:44029", + "pull_request:44026|pull_request:44027", + "pull_request:44026|pull_request:44028", + "pull_request:44026|pull_request:44030", + "pull_request:44027|pull_request:44028", + "pull_request:44027|pull_request:44029", + "pull_request:44027|pull_request:44030", + "pull_request:44028|pull_request:44029", + "pull_request:44028|pull_request:44030", + "pull_request:44029|pull_request:44030" + ], + "prepared_review_unit_hash": "d8301b178ca7552b2203f761c9b5dfdd10fe089db1d51d63d9b9a4d9820d61f6", + "component_budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40177, + "estimated_input_tokens": 10045, + "estimated_eval_tokens": 20346 + }, + "budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40177, + "estimated_input_tokens": 10045, + "estimated_eval_tokens": 20346 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are model-specific output-tracing refactors across unrelated architectures, not duplicate fixes for one shared bug or change.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "44003 is the best single representative PR: it has the broadest scope in this set, touching two closely related Mamba-family files with a substantial tracing refactor.", + "best_issue_reason": null, + "best_pr_reason": "44003 is the strongest representative of the pattern because it is broader than the other PRs and contains the clearest, most substantial code-path change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43998", + "right": "pull_request:44002", + "accept": false, + "reason": "Same refactor theme, but different model implementations and code paths; not the same concrete bug/change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44000", + "accept": false, + "reason": "Different model families and files; only the refactor style matches, not the underlying change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44001", + "accept": false, + "reason": "Different architectures and forward paths; these are separate model-local refactors." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44003", + "accept": false, + "reason": "Timm backbone vs Mamba/FalconMamba are unrelated code paths; not one mergeable fix." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44029", + "accept": false, + "reason": "Different architectures and output-tracing implementations; same theme, not the same change." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44001", + "accept": false, + "reason": "MobileNetV1 and UnivNet are unrelated models; these are separate refactors, not duplicates." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model code paths; the shared wording does not indicate one underlying bug." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44025", + "accept": false, + "reason": "MobileNetV1 vs Depth Anything are unrelated implementations; not a single concrete fix." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44026", + "accept": false, + "reason": "Different model families and files; only a similar tracing cleanup pattern." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44027", + "accept": false, + "reason": "MobileNetV1 and Speech-Encoder-Decoder are unrelated code paths, so this is not a duplicate." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44001", + "accept": false, + "reason": "Vision-Text Dual Encoder and UnivNet are separate model implementations; same theme, different changes." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model families and forward logic; not the same bug or PR-worthy change." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44003", + "accept": false, + "reason": "Vision-Text Dual Encoder vs Mamba/FalconMamba are unrelated code paths." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44025", + "accept": false, + "reason": "These touch unrelated models; the common output-tracing wording is too broad to merge." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44002", + "accept": false, + "reason": "UnivNet and UperNet are unrelated model paths; no shared concrete fix." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44003", + "accept": false, + "reason": "Different architectures and files; not one underlying code-path problem." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44025", + "accept": false, + "reason": "UnivNet vs Depth Anything are separate model-local refactors, not duplicates." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44029", + "accept": false, + "reason": "UperNet and RWKV are unrelated code paths; only a generic refactor theme is shared." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44003", + "accept": false, + "reason": "UperNet and Mamba/FalconMamba are different models with different tracing logic." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44028", + "accept": false, + "reason": "Mamba/FalconMamba and SuperPoint are unrelated architectures; not the same fix." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44004", + "accept": false, + "reason": "Mamba/FalconMamba vs CodeGen are different code paths; same pattern, not same change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44004", + "accept": false, + "reason": "Timm backbone and CodeGen are unrelated model implementations." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "UnivNet and CodeGen are unrelated code paths; not a single bug fix." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44004", + "accept": false, + "reason": "UperNet and CodeGen are separate model-specific refactors." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44026", + "accept": false, + "reason": "Depth Anything and Vision-Encoder-Decoder are different model families; not one mergeable change." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44027", + "accept": false, + "reason": "Depth Anything vs Speech-Encoder-Decoder are unrelated implementations." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44028", + "accept": false, + "reason": "Different model code paths; the shared tracing refactor theme is too broad." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44029", + "accept": false, + "reason": "Depth Anything and RWKV are unrelated models and would not be merged as one PR." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44030", + "accept": false, + "reason": "Different architectures and forward logic; not the same underlying change." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44029", + "accept": false, + "reason": "Vision-Encoder-Decoder and RWKV are unrelated code paths." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44027", + "accept": false, + "reason": "Vision-Encoder-Decoder and Speech-Encoder-Decoder are similar wrappers, but still separate model-specific changes and not one duplicate fix." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44028", + "accept": false, + "reason": "Different model families and files; not the same concrete bug/change." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44030", + "accept": false, + "reason": "Vision-Encoder-Decoder and DPR are unrelated implementations." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44028", + "accept": false, + "reason": "Speech-Encoder-Decoder and SuperPoint are unrelated code paths." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44029", + "accept": false, + "reason": "Speech-Encoder-Decoder and RWKV are unrelated models; no shared underlying fix." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44030", + "accept": false, + "reason": "Speech-Encoder-Decoder and DPR are separate model-local refactors." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44029", + "accept": false, + "reason": "SuperPoint and RWKV are unrelated architectures; same theme only." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44030", + "accept": false, + "reason": "SuperPoint and DPR are different models with different code paths." + }, + { + "left": "pull_request:44029", + "right": "pull_request:44030", + "accept": false, + "reason": "RWKV and DPR are unrelated implementations; not a duplicate PR pair." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 16, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-44662-3", + "nodes": [ + "pull_request:44662", + "pull_request:45075", + "pull_request:45401" + ], + "soft_pairs": [ + "pull_request:44662|pull_request:45401", + "pull_request:44662|pull_request:45075", + "pull_request:45075|pull_request:45401" + ], + "prepared_review_unit_hash": "935bde6d7171664e4aaae23d32e57fe7f588b447534b3c7dc4fc45677024f4e1", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 10914, + "estimated_input_tokens": 2729, + "estimated_eval_tokens": 5714 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 10914, + "estimated_input_tokens": 2729, + "estimated_eval_tokens": 5714 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are three separate model-addition PRs for different model families. They overlap on registry/docs plumbing, but each introduces a distinct implementation and test suite, so they are not duplicates.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:44662", + "right": "pull_request:45401", + "accept": false, + "reason": "Different models and code paths: PenguinVL vs Voxtral TTS. Shared auto-registry/docs files are just integration plumbing, not the same underlying change." + }, + { + "left": "pull_request:44662", + "right": "pull_request:45075", + "accept": false, + "reason": "Different model implementations: PenguinVL vs DeepSeek-OCR-2. Similar framework registration changes do not make these the same PR." + }, + { + "left": "pull_request:45075", + "right": "pull_request:45401", + "accept": false, + "reason": "Distinct model families and feature sets: DeepSeek-OCR-2 vs Voxtral TTS. Overlap is limited to common model-registration files." + } + ] + }, + "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": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-44979-2", + "nodes": [ + "pull_request:44979", + "pull_request:45363" + ], + "soft_pairs": [ + "pull_request:44979|pull_request:45363" + ], + "prepared_review_unit_hash": "13f14121ee4f5bc5a4a8abeac7d87ffb77157e3133b9410372981e2e1ad30441", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4406, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4406, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two related but distinct open PRs around fusion infrastructure: one introduces a generic Module Fusion API, while the other adds KernelConfig-based n-to-1 kernel fusion integration and an example. They share a theme but not the same concrete change.", + "confidence": 0.88, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #44979 is the cleaner representative of the shared theme because it introduces the core Module Fusion API abstraction that the other PR appears to build upon.", + "best_issue_reason": null, + "best_pr_reason": "PR #44979 is the best cluster representative overall since it defines the foundational API and is the broadest, most reusable change in the set.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44979", + "right": "pull_request:45363", + "accept": false, + "reason": "Related fusion work, but not the same underlying bug or mergeable change: #44979 adds a generic Module Fusion API, while #45363 wires a specific KernelConfig-based n-to-1 fusion path plus example/integration changes. The scopes and code paths differ too much to treat as duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 18, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45086-2", + "nodes": [ + "pull_request:45086", + "pull_request:45317" + ], + "soft_pairs": [ + "pull_request:45086|pull_request:45317" + ], + "prepared_review_unit_hash": "cf5211ea4d561ac766becc199b0640b55ad7bf2b8f3c1da1eb2a5958e6066ed7", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4413, + "estimated_input_tokens": 1104, + "estimated_eval_tokens": 2464 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4413, + "estimated_input_tokens": 1104, + "estimated_eval_tokens": 2464 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both pull requests fix the same concrete Mistral regex patch bug in `tokenization_utils_tokenizers.py`: they replace `backend_tokenizer.pre_tokenizer` with `tokenizer.pre_tokenizer` to avoid an `AttributeError`. PR 45317 is the more complete version because it includes tests.", + "confidence": 0.99, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45317 is the best canonical PR because it addresses the same underlying bug as 45086 but adds test coverage and appears to be the more complete fix.", + "best_issue_reason": null, + "best_pr_reason": "PR 45317 is the strongest standalone candidate: same fix, same target issue, plus tests validating `fix_mistral_regex=True`.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45086", + "right": "pull_request:45317", + "accept": true, + "reason": "Same code-path bug fix in `_patch_mistral_regex` with identical core change; 45317 is an expanded version of 45086 rather than a different 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": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45115-2", + "nodes": [ + "pull_request:45115", + "pull_request:45569" + ], + "soft_pairs": [ + "pull_request:45115|pull_request:45569" + ], + "prepared_review_unit_hash": "8175693e52fb9b8141930d3c38cbd4f14ba637a712bead4162024409aaffdd6b", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6066, + "estimated_input_tokens": 1517, + "estimated_eval_tokens": 3290 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6066, + "estimated_input_tokens": 1517, + "estimated_eval_tokens": 3290 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs overlap on NemotronH internals, but they address different changes: one is a conversion/mapping and inheritance refactor, the other is a broader model split into NemotronHDense/NemotronHSparse with docs/tests. They are related but not duplicates.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45569 is the broader, more representative NemotronH change: it introduces the dense/sparse model split, updates docs, auto-mappings, and tests, so it better captures the main architectural direction.", + "best_issue_reason": null, + "best_pr_reason": "PR #45569 is the best global representative because it encompasses the larger user-visible model refactor and package structure changes, whereas #45115 is a narrower conversion-mapping/inheritance cleanup.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45115", + "right": "pull_request:45569", + "accept": false, + "reason": "They both touch NemotronH, but the concrete changes differ: #45115 focuses on checkpoint conversion and inheritance/weight-name mapping, while #45569 splits the architecture into dense/sparse variants and updates docs/tests. Not the same underlying bug or 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": 20, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45332-2", + "nodes": [ + "pull_request:45332", + "pull_request:45333" + ], + "soft_pairs": [ + "pull_request:45332|pull_request:45333" + ], + "prepared_review_unit_hash": "3ab322e3f27d0e4187329735f60ed3bda4f58c3326b97e870b75603d30039e2f", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5398, + "estimated_input_tokens": 1350, + "estimated_eval_tokens": 2956 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5398, + "estimated_input_tokens": 1350, + "estimated_eval_tokens": 2956 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two PRs are related but not duplicates: one adds heterogeneous per-layer configuration support, while the other adds broader heterogeneous model/runtime support built on top of that groundwork.", + "confidence": 0.9, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45332 is the broader umbrella change, spanning config, cache, masking, modeling, and tests, so it is the best representative of the cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR #45332 is the most suitable global representative because it covers the end-to-end heterogeneous model support rather than only the config layer.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45332", + "right": "pull_request:45333", + "accept": false, + "reason": "They share implementation infrastructure, but #45333 is config-only while #45332 adds model/runtime behavior; they are related pieces of one feature, not the same concrete change or bug 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": 21, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45415-2", + "nodes": [ + "pull_request:45415", + "pull_request:45425" + ], + "soft_pairs": [ + "pull_request:45415|pull_request:45425" + ], + "prepared_review_unit_hash": "0746c8e17bfcbb5c5415d646323bdb671022af20e07d24dbacddad2087999fa9", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5514, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5514, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs are typing-related, but they are not the same change: #45415 is a broad repository-wide type-checking sweep, while #45425 adds specific typing aliases/protocols to support modeling_utils and PEFT-related code paths.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "#45425 is the more focused, self-contained typing change; #45415 is a broad sweep across many unrelated files plus type-checker config, so it is not a duplicate of the targeted typing-support PR.", + "best_issue_reason": null, + "best_pr_reason": "#45425 is the best representative of this cluster because it has one concrete typing support theme and a clear scope; #45415 is too heterogeneous to serve as the canonical change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45415", + "right": "pull_request:45425", + "accept": false, + "reason": "Reject: shared typing files and helpers do not mean the same underlying bug/change. #45415 is a wide type-checking pass across many modules and pyproject settings, while #45425 adds specific typing aliases/protocols for modeling_utils/PEFT and related utilities. They could be merged independently." + } + ] + }, + "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": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45487-4", + "nodes": [ + "pull_request:45487", + "pull_request:45817", + "pull_request:45836", + "pull_request:45843" + ], + "soft_pairs": [ + "pull_request:45487|pull_request:45843", + "pull_request:45487|pull_request:45836", + "pull_request:45817|pull_request:45843", + "pull_request:45836|pull_request:45843" + ], + "prepared_review_unit_hash": "3bf54099e12545853acaaf6c3ec07013099b340b5d01086e471b3110266559d7", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 4, + "serialized_chars": 10387, + "estimated_input_tokens": 2597, + "estimated_eval_tokens": 5450 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 4, + "serialized_chars": 10387, + "estimated_input_tokens": 2597, + "estimated_eval_tokens": 5450 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related by theme (model-parallel robustness) but they are not duplicate PRs: each targets a different model or a test-only workaround. None should be merged together.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45843 is the best representative of the cluster because it contains a concrete model-parallel code fix in the actual modeling path, whereas 45836 is test-only and 45487/45817 are separate model-specific fixes.", + "best_issue_reason": null, + "best_pr_reason": "PR 45843 is the strongest standalone PR to keep as the representative fix: it addresses a real device-placement / no_split_modules propagation bug in Deimv2, with focused code changes and tests.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45487", + "right": "pull_request:45843", + "accept": false, + "reason": "Same broad model-parallel theme, but different models and different failure paths; not the same underlying bug or change." + }, + { + "left": "pull_request:45487", + "right": "pull_request:45836", + "accept": false, + "reason": "45487 fixes runtime model-parallel device placement; 45836 only skips failing tests for QuantizedCache incompatibility." + }, + { + "left": "pull_request:45817", + "right": "pull_request:45843", + "accept": false, + "reason": "Both are model-parallel fixes, but Gemma4 and Deimv2 have distinct code paths and patch mechanics; not mergeable as one PR." + }, + { + "left": "pull_request:45836", + "right": "pull_request:45843", + "accept": false, + "reason": "Test skip for MiniCPM-V/Qwen cache incompatibility is unrelated to Deimv2's model-parallel code 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": 23, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45548-2", + "nodes": [ + "pull_request:45548", + "pull_request:45662" + ], + "soft_pairs": [ + "pull_request:45548|pull_request:45662" + ], + "prepared_review_unit_hash": "0412edd54149163229ba62e7516257b6a315206c23787e6d303d1d159b325b8d", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4505, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4505, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs are about expert-parallel (EP) model loading, but they target different failure modes and code paths: PR #45548 focuses on DeepSpeed ZeRO-3 / accelerate launch config handling, while PR #45662 fixes FSDP2/rank-0 broadcast overwriting expert weights during load. They are related but not duplicates.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45662 is the better canonical PR for the cluster because it addresses the more concrete loading corruption bug and includes the broader set of code-path fixes around expert parameter sharding and assignment.", + "best_issue_reason": null, + "best_pr_reason": "PR #45662 is the strongest representative overall: it fixes a specific, user-visible correctness issue in EP weight loading and touches the relevant loading/integration paths directly.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45548", + "right": "pull_request:45662", + "accept": false, + "reason": "Different underlying bugs: #45548 is about EP + DeepSpeed ZeRO-3 loading via accelerate launch and config serialization/has_ep detection, while #45662 is about EP + FSDP2 where expert weights are silently overwritten by rank-0 broadcast. Shared subsystem is not enough to treat them as the same change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 24, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45639-2", + "nodes": [ + "pull_request:45639", + "pull_request:45645" + ], + "soft_pairs": [ + "pull_request:45639|pull_request:45645" + ], + "prepared_review_unit_hash": "76ab4bed7ab21edeac39cf454fcb8f795104dc0b4ba4f9a5b10487b0fdfd58c4", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4588, + "estimated_input_tokens": 1147, + "estimated_eval_tokens": 2550 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4588, + "estimated_input_tokens": 1147, + "estimated_eval_tokens": 2550 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs target the same xdist collision in captured_info logging; #45645 is the broader follow-up that includes the core fix from #45639 plus CI artifact handling updates.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "#45645 is the more complete implementation of the shared captured_info xdist fix: it includes the worker-safe output-file change and also updates CI consumption of those logs.", + "best_issue_reason": null, + "best_pr_reason": "#45645 is the best single PR to keep because it covers the core bug fix and the necessary downstream workflow changes, whereas #45639 is only the smaller subset.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45639", + "right": "pull_request:45645", + "accept": true, + "reason": "Same underlying bug/change: making captured_info logging xdist-safe. #45645 extends the same fix path rather than introducing a separate feature." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 25, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45683-2", + "nodes": [ + "pull_request:45683", + "pull_request:45734" + ], + "soft_pairs": [ + "pull_request:45683|pull_request:45734" + ], + "prepared_review_unit_hash": "eba4ca0f26b2d3be0c00e316270a9341d31af2208b44dbab31b1d3a8e81e7538", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4469, + "estimated_input_tokens": 1118, + "estimated_eval_tokens": 2492 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4469, + "estimated_input_tokens": 1118, + "estimated_eval_tokens": 2492 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs touch the same quantization helper, but they fix different bugs: one special-cases audio submodules, the other changes how default skip lists are combined with user-supplied skips. They should not be clustered as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "45734 is the better representative PR because it fixes the core skip-module composition logic in `get_modules_to_not_convert`, which is broader and more central than the narrow audio-module special case in 45683.", + "best_issue_reason": null, + "best_pr_reason": "45734 is the strongest standalone PR in the set: it addresses a concrete quantization failure mode in the main code path and its fix is directly about the helper\u2019s behavior rather than a model-specific workaround.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45683", + "right": "pull_request:45734", + "accept": false, + "reason": "Both touch quantization skip logic, but 45683 adds audio-tower/embed_audio exclusions while 45734 changes additive handling of `skip_modules`; these are different underlying bugs and would not naturally merge into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 26, + "component_count": 26, + "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 + } + ] +} diff --git a/snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/manifest.json b/snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..57270d309b7877bdd92578ada2f7c2796ef48f57 --- /dev/null +++ b/snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/manifest.json @@ -0,0 +1,14 @@ +{ + "analysis_id": "hybrid-model-20260510t120026z", + "artifacts": { + "hybrid": "snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "model": null, + "published_at": "2026-05-10T12:08:02Z", + "repo": "huggingface/transformers", + "schema_version": 1, + "snapshot_id": "20260510T120026Z", + "variant": "hybrid" +} diff --git a/snapshots/20260510T120026Z/manifest.json b/snapshots/20260510T120026Z/manifest.json index 36774460c4e2b04e8067041213570a286e4535aa..0c99dd0a63f8b29046d87d642716fe102732686e 100644 --- a/snapshots/20260510T120026Z/manifest.json +++ b/snapshots/20260510T120026Z/manifest.json @@ -32,6 +32,24 @@ "timeline_events": 322 }, "extracted_at": "2026-05-10T12:00:26Z", + "published_analysis": { + "canonical_analysis_id": "hybrid-model-20260510t120026z", + "runs": { + "hybrid-model-20260510t120026z": { + "analysis_id": "hybrid-model-20260510t120026z", + "artifacts": { + "hybrid": "snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "manifest_path": "snapshots/20260510T120026Z/analysis-runs/hybrid-model-20260510t120026z/manifest.json", + "model": null, + "published_at": "2026-05-10T12:08:02Z", + "variant": "hybrid" + } + }, + "schema_version": 1 + }, "repo": "huggingface/transformers", "snapshot_id": "20260510T120026Z", "watermark": {