diff --git a/analysis/current/analysis-report-hybrid.json b/analysis/current/analysis-report-hybrid.json index b53f25144f9741a071e8549f550598a16de0345e..7e62e3fee4869e908142685a9ee730027ac35f10 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": "20260601T000028Z", - "generated_at": "2026-06-01T00:04:43Z", + "snapshot_id": "20260601T060039Z", + "generated_at": "2026-06-01T06:08:58Z", "evidence_quality": "full", "llm_enrichment": true, "meta_bugs": [ @@ -1426,7 +1426,7 @@ "cluster_id": "cluster-45561-3", "summary": "Cluster of 1 issues and 2 PRs centered on issue #45561.", "status": "open", - "confidence": 0.55, + "confidence": 0.8, "canonical_issue_number": 45561, "canonical_pr_number": 45645, "issue_numbers": [ @@ -1437,7 +1437,9 @@ 45645 ], "evidence_types": [ - "closing_reference" + "closing_reference", + "shared_issue_target", + "soft_similarity" ], "pr_comparisons": [ { @@ -1612,153 +1614,6 @@ } ] }, - { - "cluster_id": "cluster-46023-2", - "summary": "Cluster of 2 related pull requests linked by soft_similarity.", - "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 46023, - "issue_numbers": [], - "pr_numbers": [ - 46023, - 46259 - ], - "evidence_types": [ - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 46023, - "right_pr_number": 46259, - "code_similarity": 0.383, - "size_similarity": 0.247, - "file_overlap": 0.031, - "area_overlap": 0.679, - "patch_similarity": 0.58, - "shared_filenames": [ - "docs/source/en/modeling_rules.md", - "setup.py", - "src/transformers/dependency_versions_table.py", - "src/transformers/models/gemma4/modular_gemma4.py", - "src/transformers/models/kosmos2_5/modeling_kosmos2_5.py", - "utils/rules.toml" - ], - "shared_file_areas": [ - { - "filename": "docs/source/en/modeling_rules.md", - "left_ranges": [ - [ - 228, - 288 - ] - ], - "right_ranges": [ - [ - 228, - 288 - ] - ] - }, - { - "filename": "setup.py", - "left_ranges": [ - [ - 126, - 132 - ] - ], - "right_ranges": [ - [ - 126, - 132 - ] - ] - }, - { - "filename": "src/transformers/dependency_versions_table.py", - "left_ranges": [ - [ - 56, - 62 - ] - ], - "right_ranges": [ - [ - 56, - 62 - ] - ] - }, - { - "filename": "src/transformers/models/gemma4/modular_gemma4.py", - "left_ranges": [ - [ - 558, - 581 - ] - ], - "right_ranges": [ - [ - 1214, - 1220 - ] - ] - }, - { - "filename": "src/transformers/models/kosmos2_5/modeling_kosmos2_5.py", - "left_ranges": [ - [ - 636, - 642 - ], - [ - 970, - 980 - ], - [ - 1402, - 1407 - ], - [ - 1413, - 1422 - ] - ], - "right_ranges": [ - [ - 77, - 83 - ] - ] - }, - { - "filename": "utils/rules.toml", - "left_ranges": [ - [ - 221, - 227 - ], - [ - 249, - 318 - ] - ], - "right_ranges": [ - [ - 221, - 227 - ], - [ - 249, - 318 - ] - ] - } - ] - } - ] - }, { "cluster_id": "cluster-44018-2", "summary": "Cluster of 2 related pull requests linked by soft_similarity.", @@ -2115,7 +1970,7 @@ 45639 ], "target_issue_number": 45561, - "reason": "PRs in cluster-45561-3 are treated as duplicates because they converge on issue #45561 with closing_reference evidence." + "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", @@ -2125,15 +1980,6 @@ ], "target_issue_number": 45647, "reason": "PRs in cluster-45647-3 are treated as duplicates because they converge on issue #45647 with closing_reference, shared_issue_target evidence." - }, - { - "cluster_id": "cluster-46023-2", - "canonical_pr_number": 46023, - "duplicate_pr_numbers": [ - 46259 - ], - "target_issue_number": null, - "reason": "PRs in cluster-46023-2 are treated as duplicates because they share soft_similarity evidence." } ], "best_issue": { diff --git a/analysis/current/analysis-report-hybrid.llm-reviews.json b/analysis/current/analysis-report-hybrid.llm-reviews.json index 4c9c034525e37802361da5fa3895d37b92daf8ad..e5bc9f0e6392900462bfeca791f1a9453b6ebcf9 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": "20260601T000028Z", - "generated_at": "2026-06-01T00:04:43Z", + "snapshot_id": "20260601T060039Z", + "generated_at": "2026-06-01T06:08:58Z", "model": "gpt-5.4-mini?service_tier=flex", "reviews": [ { @@ -10,13 +10,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 1, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:24643", "issue:30277", @@ -80,8 +80,8 @@ "issue:45901|issue:46015", "issue:44509|issue:44512", "issue:35545|issue:36010", - "issue:44918|issue:46165", "issue:44805|issue:45823", + "issue:44918|issue:46165", "issue:43638|issue:44805", "issue:43638|issue:46165", "issue:44805|issue:46294", @@ -114,8 +114,8 @@ "issue:43716|issue:45685", "issue:44918|issue:45161", "issue:43716|issue:44292", - "issue:44292|issue:44918", "issue:45507|issue:45685", + "issue:44292|issue:44918", "issue:43039|issue:44918", "issue:41762|issue:43638", "issue:43827|issue:44512", @@ -129,27 +129,27 @@ "issue:43039|issue:44805", "issue:45800|issue:46304", "issue:43296|issue:46304", - "issue:39290|issue:41720", "issue:41762|issue:43596", + "issue:39290|issue:41720", "issue:43901|issue:44512", - "issue:44589|issue:46304", "issue:45823|issue:46294", + "issue:44589|issue:46304", "issue:45237|issue:45901" ], - "prepared_review_unit_hash": "0b998587bd77bb9943d0978f3fced78f0b8e4663d5802295c37268c7f034d96f", + "prepared_review_unit_hash": "df2da42a21c2ca132200e8d735a737a9d4e84ffd93a0e77528a5294eddbf6dfe", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 67, - "serialized_chars": 47821, + "serialized_chars": 47824, "estimated_input_tokens": 11956, "estimated_eval_tokens": 24168 }, @@ -158,402 +158,402 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose thematic group of unrelated bug reports; only the two BERT + DeepSpeed ZeRO-3 index-out-of-bounds reports look like true duplicates.", - "confidence": 0.58, - "canonical_issue_reason": "issue:43638 is the clearest representative of the only true duplicate subgroup here: BERT + DeepSpeed ZeRO-3 failing with an index-0-out-of-bounds crash during model init/training.", + "summary": "This cluster is mostly noisy and heterogeneous. The only clear duplicate family is the ZeRO-3 \"index 0 is out of bounds for dimension 0 with size 0\" loading failure (43596/43638, with 41762 as a close same-bug report), plus one plausible Qwen3.5 MoE rotary-embedding/offload duplicate (45859/45902). Most other soft links are false positives across different models, APIs, or docs topics.", + "confidence": 0.76, + "canonical_issue_reason": "No single canonical fits the whole mixed set; if one issue must represent the real duplicate family, 43638 is the best pick because it is the clearest ZeRO-3 empty-index loading failure and aligns with the exact duplicate reports around the same bug.", "canonical_pr_reason": null, - "best_issue_reason": "There is no strong global canonical for the whole cluster because the items span many unrelated models, configs, and docs issues; issue:43638 is just the best anchor for the one genuinely duplicated bug family.", + "best_issue_reason": "43638 is the strongest issue to keep: it states the concrete ZeRO-3 index-out-of-bounds failure clearly, is reproducible, and best captures the main duplicate family in this set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44322", "right": "issue:45464", "accept": false, - "reason": "Different failure modes: missing Qwen3.5 config attribute vs streaming chat/completions API error." + "reason": "Different failure surface: config attribute access vs streaming chat/completions inference." }, { "left": "issue:45237", "right": "issue:45685", "accept": false, - "reason": "Unrelated hardware/backend problems: AMD GPU support vs MPS histogram kernel on Int." + "reason": "Unrelated model/runtime bugs: AMD GPU failure vs MPS histogram op limitation." }, { "left": "issue:43716", "right": "issue:45237", "accept": false, - "reason": "Different models and error paths; image preprocessor dtype mismatch is unrelated to AMD GPU execution failure." + "reason": "Different models and code paths; image preprocessor dtype mismatch is not the same as GPT-OSS AMD GPU failure." }, { "left": "issue:43828", "right": "issue:45237", "accept": false, - "reason": "Autocast dtype mismatch on Phi MoE is a different bug than GPT-OSS failing on AMD GPUs." + "reason": "Both are runtime dtype-related only at a high level; the concrete model/code path is different." }, { "left": "issue:45237", "right": "issue:45507", "accept": false, - "reason": "Different model families and symptoms; AMD GPU issue vs invalid method call in GraniteMoEHybrid." + "reason": "Different MoE/model failures; one is AMD GPU support, the other is an invalid method call." }, { "left": "issue:44292", "right": "issue:45237", "accept": false, - "reason": "Qwen-3 NVFP4 runtime error is a different code path from AMD GPU incompatibility." + "reason": "Different model families and different errors; NVFP4 runtime issue is not the same bug as GPT-OSS AMD GPU failure." }, { "left": "issue:30277", "right": "issue:35545", "accept": false, - "reason": "DeepSpeed ZeRO-3 collective mismatch is unrelated to ModernBERT ONNX export failure." + "reason": "DeepSpeed collectives mismatch and ONNX export failure are unrelated." }, { "left": "issue:35545", "right": "issue:42915", "accept": false, - "reason": "ONNX export failure and FineGrainedFP8Config failure are different model/backend issues." + "reason": "ONNX export error vs FP8 config failure are different issues." }, { "left": "issue:35545", "right": "issue:41720", "accept": false, - "reason": "Export-to-ONNX problem is unrelated to Qwen3 auto device mapping cudaErrorAssert." + "reason": "Different problems: ONNX export vs auto device mapping CUDA assert." }, { "left": "issue:45901", "right": "issue:46015", "accept": false, - "reason": "Task crash vs outdated RAG docs; same general area, not the same bug or change." + "reason": "Table QA crash and RAG docs updates are unrelated." }, { "left": "issue:44509", "right": "issue:44512", "accept": false, - "reason": "Both are docs cleanup items, but they target different removed features and would not be one merged fix." + "reason": "Both are v5 docs cleanups, but they target different obsolete references and not one concrete fix." }, { "left": "issue:35545", "right": "issue:36010", "accept": false, - "reason": "ModernBERT ONNX export and GenerationMixin import regression are unrelated." + "reason": "ONNX export error and GenerationMixin import error are unrelated." }, { - "left": "issue:44918", - "right": "issue:46165", + "left": "issue:44805", + "right": "issue:45823", "accept": false, - "reason": "TRL SFT embedding unpacking vs resume-training with trust_remote_code are different training failures." + "reason": "Mask/tensor shape mismatch is a different failure than Gemma4 device-mapping mismatch." }, { - "left": "issue:44805", - "right": "issue:45823", + "left": "issue:44918", + "right": "issue:46165", "accept": false, - "reason": "Mask shape mismatch during indexing is unrelated to Gemma4 PLE device placement mismatch." + "reason": "Different training failures; TRL SFT embedding unpacking is not the same as resume/trust-remote-code issues." }, { "left": "issue:43638", "right": "issue:44805", "accept": false, - "reason": "Both involve indexing errors, but the reported symptoms and contexts differ too much to be confidently the same bug." + "reason": "Both are IndexError/shape-related, but the concrete bug and code path differ." }, { "left": "issue:43638", "right": "issue:46165", "accept": false, - "reason": "ZeRO-3 BERT crash is unrelated to resuming training with trust_remote_code." + "reason": "ZeRO-3 loading failure and resume/trust-remote-code failure are unrelated." }, { "left": "issue:44805", "right": "issue:46294", "accept": false, - "reason": "Mask/index shape mismatch is unrelated to Qwen3.5 video feature/token mismatch." + "reason": "Mask shape mismatch and Qwen3.5 video chat template mismatch are different bugs." }, { "left": "issue:43596", "right": "issue:44805", "accept": false, - "reason": "Similar generic indexing symptom, but different trigger and message; not enough to treat as one bug." + "reason": "Same broad category of tensor errors, but not the same underlying failure." }, { "left": "issue:44805", "right": "issue:45161", "accept": false, - "reason": "Indexing/mask mismatch is unrelated to GPT-OSS tensor-parallel-only failure." + "reason": "Mask shape mismatch is unrelated to GPT-OSS tensor parallelism." }, { "left": "issue:44560", "right": "issue:46165", "accept": false, - "reason": "Video chat-template StopIteration is unrelated to training resume behavior." + "reason": "Video StopIteration and resume/trust-remote-code are different issues." }, { "left": "issue:43596", "right": "issue:46165", "accept": false, - "reason": "ZeRO-3 initialization crash is unrelated to local model resume/trust_remote_code." + "reason": "ZeRO-3 index-out-of-bounds is unrelated to model resume/trust-remote-code." }, { "left": "issue:43572", "right": "issue:46165", "accept": false, - "reason": "Missing pad_token_idx after v5 update is unrelated to training resume failure." + "reason": "StableLm pad_token_idx regression is unrelated to the training resume issue." }, { "left": "issue:34634", "right": "issue:35545", "accept": false, - "reason": "BarkProcessor voice_preset bug and ModernBERT export bug are unrelated." + "reason": "Processor voice_preset bug and ONNX export failure are unrelated." }, { "left": "issue:30277", "right": "issue:43638", "accept": false, - "reason": "Both touch DeepSpeed, but one is collective mismatch in Jamba-v01 and the other is BERT index-out-of-bounds; different bugs." + "reason": "Different DeepSpeed failures: collectives mismatch vs empty-index loading error." }, { "left": "issue:43596", "right": "issue:43638", "accept": true, - "reason": "Near-identical ZeRO-3 BERT index-out-of-bounds reports; same underlying initialization/training bug and plausible single fix." + "reason": "Same ZeRO-3 index-0-out-of-bounds loading bug; the titles differ only in model/context wording." }, { "left": "issue:43975", "right": "issue:44457", "accept": false, - "reason": "Different model-specific tokenizer/ detokenization issue vs LoRA merge/save/reload mismatch." + "reason": "Different tokenization/LoRA issues with no shared concrete failure mode." }, { "left": "issue:43039", "right": "issue:45161", "accept": false, - "reason": "Liger Kernel cross_entropy invocation is unrelated to GPT-OSS tensor-parallel failure." + "reason": "Liger Kernel cross-entropy invocation is unrelated to GPT-OSS tensor parallel failures." }, { "left": "issue:44292", "right": "issue:44457", "accept": false, - "reason": "Qwen-3 NVFP4 runtime error is unrelated to LoRA merge/save behavior." + "reason": "Different model/runtime bugs; NVFP4 execution error is not the same as LoRA merge/save mismatch." }, { "left": "issue:24643", "right": "issue:30277", "accept": false, - "reason": "Different DeepSpeed-related failures: 2-D weight error vs collective mismatch." + "reason": "Both mention DeepSpeed, but one is a 2-D weight error and the other is a collective mismatch." }, { "left": "issue:34928", "right": "issue:44805", "accept": false, - "reason": "Activation-checkpointing/FSDP tensor recomputation bug is unrelated to mask shape mismatch." + "reason": "Activation-checkpointing/FSDP recompute mismatch is not the same as the mask-shape IndexError." }, { "left": "issue:45859", "right": "issue:45902", - "accept": false, - "reason": "CPU offload incompatibility in rotary embedding vs uninitialized memory after materialization are different bugs." + "accept": true, + "reason": "Same Qwen3.5 MoE rotary-embedding/offload bug; one report describes the symptom, the other pinpoints the uninitialized inv_freq materialization path." }, { "left": "issue:43638", "right": "issue:45161", "accept": false, - "reason": "BERT ZeRO-3 index crash is unrelated to GPT-OSS only-TP failure." + "reason": "ZeRO-3 loading failure and GPT-OSS tensor-parallel issue are unrelated." }, { "left": "issue:34928", "right": "issue:36331", "accept": false, - "reason": "FSDP/activation checkpointing shape issue is unrelated to trainer custom compute_loss signature change." + "reason": "FSDP checkpointing mismatch and compute_loss signature change are different issues." }, { "left": "issue:43596", "right": "issue:45161", "accept": false, - "reason": "ZeRO-3 BERT init crash is unrelated to GPT-OSS tensor-parallel-only failure." + "reason": "ZeRO-3 loading error and tensor-parallel failure are unrelated." }, { "left": "issue:43716", "right": "issue:45507", "accept": false, - "reason": "Mistral-3 image dtype mismatch is unrelated to GraniteMoEHybrid invalid method call." + "reason": "Different models and different bugs; dtype mismatch is not the same as invalid method call." }, { "left": "issue:45874", "right": "issue:46294", "accept": false, - "reason": "Different Gemma4 input/embed behavior vs Qwen3.5 video template mismatch." + "reason": "Both are Gemma/Qwen multimodal issues, but the code paths are different (reverse embedding lookup vs video template token mismatch)." }, { "left": "issue:39290", "right": "issue:45081", "accept": false, - "reason": "Gemma3/vLLM config attribute error is unrelated to tokenizer regex patch crash." + "reason": "Different model-specific regressions: Gemma3 config attribute vs Mistral tokenizer regex crash." }, { "left": "issue:45823", "right": "issue:45874", "accept": false, - "reason": "Device-map forward mismatch and expensive reverse embedding lookup are different Gemma4 issues." + "reason": "Both are Gemma4 issues, but they affect different mechanisms and are not the same concrete bug." }, { "left": "issue:43425", "right": "issue:44292", "accept": false, - "reason": "Torch 2.10 compatibility report is unrelated to Qwen-3 NVFP4 runtime failure." + "reason": "Torch version incompatibility and Qwen-3 NVFP4 runtime error are unrelated." }, { "left": "issue:34928", "right": "issue:41720", "accept": false, - "reason": "FSDP checkpointing mismatch is unrelated to Qwen3 auto device mapping assertion failure." + "reason": "Different distributed-training failures with different triggers and error modes." }, { "left": "issue:43828", "right": "issue:45161", "accept": false, - "reason": "Phi autocast dtype mismatch is unrelated to GPT-OSS tensor-parallel-only failure." + "reason": "Autocast dtype mismatch and GPT-OSS tensor-parallel failure are different bugs." + }, + { + "left": "issue:43541", + "right": "issue:43716", + "accept": false, + "reason": "Grouped_mm tracing failure and image-preprocessor dtype mismatch are unrelated." + }, + { + "left": "issue:43828", + "right": "issue:44292", + "accept": false, + "reason": "Both are runtime model errors, but the concrete model and failure mode differ." + }, + { + "left": "issue:43716", + "right": "issue:45685", + "accept": false, + "reason": "Different model/runtime backends and different error types." }, { "left": "issue:44918", "right": "issue:45161", "accept": false, - "reason": "TRL SFT embedding issue is unrelated to GPT-OSS tensor-parallel-only failure." + "reason": "TRL SFT embedding unpacking and GPT-OSS tensor-parallel failures are unrelated." }, { "left": "issue:43716", "right": "issue:44292", "accept": false, - "reason": "Different models and symptoms: Mistral image dtype mismatch vs Qwen NVFP4 runtime error." + "reason": "Different model families and failure mechanisms." }, { - "left": "issue:44292", - "right": "issue:44918", + "left": "issue:45507", + "right": "issue:45685", "accept": false, - "reason": "Qwen NVFP4 runtime error is unrelated to TRL SFT embedding unpacking." + "reason": "Invalid method call is not the same bug as the MPS histogram limitation." }, { - "left": "issue:45507", - "right": "issue:45685", + "left": "issue:44292", + "right": "issue:44918", "accept": false, - "reason": "Invalid method call in GraniteMoEHybrid is unrelated to MPS histogram kernel limitation." + "reason": "Different Qwen/GPT-OSS training/runtime bugs with no shared concrete code path." }, { "left": "issue:43039", "right": "issue:44918", "accept": false, - "reason": "Liger Kernel cross_entropy being called is unrelated to TRL SFT embedding unpacking." + "reason": "Liger Kernel cross-entropy call and TRL embedding unpacking are unrelated." }, { "left": "issue:41762", "right": "issue:43638", - "accept": false, - "reason": "Both are ZeRO-3 load/init failures, but the models and reported symptoms differ enough that they don't look like the same bug." + "accept": true, + "reason": "Same ZeRO-3 index-out-of-bounds loading failure; different model names, same concrete failure mode." }, { "left": "issue:43827", "right": "issue:44512", "accept": false, - "reason": "Both are docs regressions, but they concern different removed features and likely separate cleanup edits." + "reason": "Both are docs regressions from v5 removals, but they point to different obsolete references." }, { "left": "issue:43716", "right": "issue:44560", "accept": false, - "reason": "Mistral image dtype mismatch is unrelated to Qwen3-vl video StopIteration." + "reason": "Different multimodal bugs: dtype mismatch vs video StopIteration." }, { "left": "issue:43828", "right": "issue:45685", "accept": false, - "reason": "Autocast dtype mismatch is unrelated to MPS histogram kernel support." + "reason": "Different runtime issues with no shared underlying bug." }, { "left": "issue:44918", "right": "issue:45685", "accept": false, - "reason": "TRl SFT embedding unpacking is unrelated to MPS histogram kernel support." + "reason": "TRL trainer unpacking and MPS histogram limitations are unrelated." }, { "left": "issue:43039", "right": "issue:43828", "accept": false, - "reason": "Liger Kernel cross_entropy invocation is unrelated to Phi autocast dtype mismatch." + "reason": "Different failures: Liger Kernel cross-entropy vs autocast dtype mismatch." }, { "left": "issue:30277", "right": "issue:36331", "accept": false, - "reason": "DeepSpeed collective mismatch and trainer compute_loss signature mismatch are unrelated." + "reason": "DeepSpeed collectives mismatch and compute_loss signature mismatch are unrelated." }, { "left": "issue:34928", "right": "issue:36010", "accept": false, - "reason": "FSDP activation checkpointing issue is unrelated to GenerationMixin import regression." + "reason": "Activation-checkpointing/FSDP issue and GenerationMixin import failure are unrelated." }, { "left": "issue:45237", "right": "issue:46015", "accept": false, - "reason": "AMD GPU runtime failure is unrelated to outdated RAG docs." + "reason": "GPT-OSS AMD GPU runtime issue and outdated RAG examples are unrelated." }, { "left": "issue:43039", "right": "issue:44805", "accept": false, - "reason": "Liger Kernel cross_entropy call is unrelated to mask shape mismatch." + "reason": "Liger Kernel cross-entropy issue and mask-shape IndexError are unrelated." }, { "left": "issue:45800", "right": "issue:46304", "accept": false, - "reason": "Version incompatibility report and USE_DISTRIBUTED=0 support regression are different compatibility problems." + "reason": "Torch/transformers compatibility problems, but different versions and different breakage." }, { "left": "issue:43296", "right": "issue:46304", "accept": false, - "reason": "PaddleOCR-VL/vLLM load failure is unrelated to pytorch distributed-disabled support breakage." - }, - { - "left": "issue:39290", - "right": "issue:41720", - "accept": false, - "reason": "Gemma3/vLLM config attribute error is unrelated to Qwen3 auto device mapping assertion failure." - }, - { - "left": "issue:41762", - "right": "issue:43596", - "accept": false, - "reason": "Both are ZeRO-3 index-out-of-bounds reports, but they involve different models and likely separate load paths." + "reason": "PaddleOCR-VL/vLLM loading failure and USE_DISTRIBUTED=0 regression are unrelated." }, { - "left": "issue:43901", - "right": "issue:44512", + "left": "issue:45823", + "right": "issue:46294", "accept": false, - "reason": "TextClassificationPipeline docs wording and run-command docs removal are separate documentation issues." + "reason": "Gemma4 device-mapping and Qwen3.5 video chat-template issues are different code paths." }, { "left": "issue:44589", "right": "issue:46304", "accept": false, - "reason": "Float8 storage lookup failure is unrelated to distributed-disabled support breakage." - }, - { - "left": "issue:45823", - "right": "issue:46294", - "accept": false, - "reason": "Gemma4 device-map mismatch and Qwen3.5 video token mismatch are unrelated." + "reason": "Float8 storage lookup error and distributed-disabled compatibility break are unrelated." }, { "left": "issue:45237", "right": "issue:45901", "accept": false, - "reason": "AMD GPU failure is unrelated to table-question-answering crash." + "reason": "GPT-OSS AMD GPU runtime bug and table-question-answering crash are unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet: this is mostly a broad thematic cluster with no single global canonical, and the only clearly duplicate subgroup is the ZeRO-3 BERT index-out-of-bounds pair (issue:43596 / issue:43638). The soft-edge verdicts are generally conservative and avoid overstating similarity." + "feedback": "The summary is well grounded in the packet and stays conservative. The ZeRO-3 index-out-of-bounds family (43596/43638, with 41762 as a near-match) is clearly supported by the titles, and the Qwen3.5 MoE rotary/offload pair (45859/45902) is a plausible same-bug cluster rather than an overclaim. The remaining false links are appropriately treated as heterogeneous across models, APIs, runtime backends, or docs." }, "evaluator_used": true, "retried": false, @@ -566,13 +566,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 2, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:36010", "issue:39290", @@ -666,298 +666,280 @@ "issue:43541|issue:44589", "issue:41093|issue:43596" ], - "prepared_review_unit_hash": "1db38b5924f5634c69bf9409335b40fb936f1b76f9e8015d8514ce9b90f0dd90", + "prepared_review_unit_hash": "1e21385230bc1cfe73c9ee0668f983d7b06af469c67bb25e4d4a4cd93c233e96", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 42, - "serialized_chars": 39559, - "estimated_input_tokens": 9890, - "estimated_eval_tokens": 20036 + "serialized_chars": 39564, + "estimated_input_tokens": 9891, + "estimated_eval_tokens": 20038 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most items are unrelated model-specific regressions or docs/config issues. I only found two clear duplicate pairs: the v5 AutoTokenizer wrong-token-ID regression and the Qwen3.5 video/chat-template regression.", - "confidence": 0.78, - "canonical_issue_reason": "issue:45920 is the best representative of the duplicate tokenizer-regression subgroup: it is broader than the Granite-specific report, remains open, and describes the same v5 wrong-token-ID failure across multiple model families.", + "summary": "Most soft pairs are false positives across unrelated model/runtime bugs. Two pairs look like genuine duplicates: the AutoTokenizer wrong-token-ID regression and the Qwen video-input regression.", + "confidence": 0.82, + "canonical_issue_reason": "issue:45920 is the strongest canonical issue candidate because it describes the broadest and clearest AutoTokenizer v5 regression, spanning multiple model families and subsuming the more specific Granite-only report in issue:45812.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 is the strongest global issue anchor among the accepted duplicates because it captures the widest version-wide tokenizer regression with multiple affected models.", + "best_issue_reason": "issue:45920 is the best overall issue representative: broad impact, explicit regression framing, and the most general statement of the shared bug in this set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44841", "right": "issue:45823", "accept": false, - "reason": "Different model families and different failures: processor loading vs device mismatch during forward." + "reason": "Different failures on different models: Voxtral processor load issue vs Gemma4 device-mismatch during forward." }, { "left": "issue:43039", "right": "issue:44560", "accept": false, - "reason": "Unrelated subsystems: Liger kernel cross-entropy path vs Qwen video StopIteration." + "reason": "Unrelated bugs: Liger Kernel cross-entropy path vs Qwen3-VL video StopIteration." }, { "left": "issue:43975", "right": "issue:46015", "accept": false, - "reason": "Different problems: detokenization bug vs RAG example documentation freshness." + "reason": "Detokenization regression vs unrelated RAG documentation issue." }, { "left": "issue:45812", "right": "issue:45920", "accept": true, - "reason": "Both report the same v5 AutoTokenizer regression producing wrong token IDs across model families; 45920 is the broader tracker." + "reason": "Same underlying AutoTokenizer v5 regression producing wrong token IDs; 45920 is the broader report across more model families." }, { "left": "issue:44863", "right": "issue:46015", "accept": false, - "reason": "Checkpoint-loading bug vs outdated RAG examples; not the same underlying issue." + "reason": "NemotronH checkpoint-loading failure is unrelated to outdated RAG examples." }, { "left": "issue:45020", "right": "issue:46304", "accept": false, - "reason": "Broad remote_code breakage vs a specific USE_DISTRIBUTED=0/pytorch compatibility regression." + "reason": "Both mention version breakage, but one is about remote_code models and the other about USE_DISTRIBUTED=0 support; not the same bug." }, { "left": "issue:43866", "right": "issue:44863", "accept": false, - "reason": "Both are load failures, but for different models and different root causes." + "reason": "Both are load failures, but for different model families and different symptoms; not the same underlying issue." }, { "left": "issue:44292", "right": "issue:46015", "accept": false, - "reason": "Model-loading error vs documentation issue; no shared code-path bug." + "reason": "init_empty_weights/_is_hf_initialized TypeError is unrelated to RAG examples." }, { "left": "issue:43872", "right": "issue:46304", "accept": false, - "reason": "bitsandbytes Int8Params constructor change is unrelated to distributed build support." + "reason": "bitsandbytes Int8Params constructor incompatibility is different from USE_DISTRIBUTED=0 breakage." }, { "left": "issue:39290", "right": "issue:43531", "accept": false, - "reason": "Both mention sliding_window, but one is a missing config attribute in vLLM and the other is a model-specific sliding_window issue." - }, - { - "left": "issue:36010", - "right": "issue:39290", - "accept": false, - "reason": "ImportError for GenerationMixin is unrelated to the Gemma3 sliding_window_pattern problem." + "reason": "Both involve sliding-window concepts, but the affected configs and failure modes differ (Gemma3TextConfig vs Qwen3-MoE)." }, { "left": "issue:43854", "right": "issue:43866", "accept": false, - "reason": "Different models and different load failures; no evidence of the same bug." + "reason": "Different models and different loading problems; no evidence of one shared bug." }, { "left": "issue:41093", "right": "issue:43638", "accept": false, - "reason": "Both are IndexErrors, but the reported shapes and contexts differ; not enough to treat as the same bug." + "reason": "Both are zero3/index errors, but the concrete failures differ: mask shape mismatch vs size-0 index out of bounds." }, { "left": "issue:43792", "right": "issue:45901", "accept": false, - "reason": "Whisper model failure vs table-question-answering crash; unrelated." - }, - { - "left": "issue:43425", - "right": "issue:45070", - "accept": false, - "reason": "Torch version incompatibility is not the same as the PretrainedConfig pydantic-field regression." - }, - { - "left": "issue:43596", - "right": "issue:45800", - "accept": false, - "reason": "Different failures: ZeRO-3/Bert index error vs torch 2.4.1 and transformers 5.8.0 incompatibility." + "reason": "Whisper model load failure is unrelated to table-question-answering crashes." }, { "left": "issue:43975", "right": "issue:45901", "accept": false, - "reason": "Unrelated features and error paths." + "reason": "Detokenization bug and table QA crash are unrelated." }, { "left": "issue:43792", "right": "issue:45070", "accept": false, - "reason": "Different model/runtime bug vs config-model schema break." + "reason": "Whisper runtime issue is unrelated to a PretrainedConfig pydantic regression." }, { "left": "issue:45362", "right": "issue:45464", "accept": false, - "reason": "Both involve Qwen3.5, but the reported failures are different enough that they do not look like one concrete bug." + "reason": "Both are Qwen3.5-related, but one is a transformers chat crash and the other is a streaming chat/completions API failure." }, { "left": "issue:44292", "right": "issue:45901", "accept": false, - "reason": "Qwen-3-8B-NVFP4 loading error is unrelated to table QA crashes." + "reason": "Different subsystems and different errors; no shared bug." }, { "left": "issue:39290", "right": "issue:45823", "accept": false, - "reason": "Gemma3 sliding_window_pattern issue and Gemma4 device mismatch are different bugs." + "reason": "Gemma3 sliding_window_pattern issue is unrelated to Gemma4 device mismatch." }, { "left": "issue:43404", "right": "issue:43866", "accept": false, - "reason": "Weight-tying bug in Mistral3 is unrelated to a corrupted Ovis2 checkpoint." + "reason": "Tied lm_head weights bug vs corrupted checkpoint/loading failure; not the same issue." }, { "left": "issue:45464", "right": "issue:45685", "accept": false, - "reason": "Chat/completions streaming failure vs MPS histogram_mps limitation; different code paths." + "reason": "Qwen3.5 streaming inference failure is unrelated to the MPS histogram Int error." }, { "left": "issue:43541", "right": "issue:45685", "accept": false, - "reason": "Torch dynamo grouped_mm tracing error is unrelated to the MPS histogram issue." + "reason": "Grouped_mm tracing error and MPS histogram error are different code paths and symptoms." }, { "left": "issue:43541", "right": "issue:43828", "accept": false, - "reason": "Both are MoE/dtype-adjacent, but the concrete failures differ and do not look like one bug." + "reason": "Both mention dtype-ish behavior, but one is torch dynamo grouped_mm tracing and the other is an autocast mismatch." }, { "left": "issue:44841", "right": "issue:45084", "accept": false, - "reason": "Processor loading failure and template-node compilation error are unrelated." + "reason": "Processor failure for Voxtral is unrelated to template-node compilation errors." }, { "left": "issue:36010", "right": "issue:39290", "accept": false, - "reason": "Same as above: import error vs config attribute regression." + "reason": "ImportError for GenerationMixin and Gemma3 sliding-window config are unrelated." }, { "left": "issue:45800", "right": "issue:45823", "accept": false, - "reason": "General version incompatibility vs a specific device-map forward mismatch." + "reason": "General torch/transformers version incompatibility is unrelated to Gemma4 device placement." }, { "left": "issue:41093", "right": "issue:41762", "accept": false, - "reason": "Both are IndexErrors, but one is a mask/tensor shape mismatch and the other is a ZeRO-3 zero-size index failure." + "reason": "Both are index errors, but the concrete contexts and failure signatures differ." }, { "left": "issue:43296", "right": "issue:43366", "accept": false, - "reason": "PaddleOCR-VL loading in vLLM and GGUF gpt-oss support are separate issues." + "reason": "Different model-support requests: PaddleOCR-VL in vLLM vs GGUF gpt-oss architecture support." }, { "left": "issue:43572", "right": "issue:45070", "accept": false, - "reason": "Missing StableLmConfig pad_token_idx and PretrainedConfig pydantic break are distinct config regressions." + "reason": "Missing pad_token_idx in StableLmConfig is not the same as a broken PretrainedConfig pydantic field." }, { "left": "issue:43541", "right": "issue:45161", "accept": false, - "reason": "Grouped_mm tracing error vs TP-only GPT-OSS MoE support issue; not the same problem." + "reason": "Different MoE/parallelism issues with no shared concrete code path." }, { "left": "issue:44387", "right": "issue:45005", "accept": false, - "reason": "Memory/OOM regression under int4 quantization vs tied-weights translation model bug." + "reason": "Int4 quantization OOM and tied-weights translation bugs are unrelated." }, { "left": "issue:44805", "right": "issue:44918", "accept": false, - "reason": "Mask-shape IndexError vs SFT trainer embedding-unpacking failure; different bugs." + "reason": "Both are tensor-shape style errors, but the model/trainer contexts differ and the bug signatures do not match closely enough." }, { "left": "issue:43828", "right": "issue:45464", "accept": false, - "reason": "Autocast dtype mismatch in Phi-tiny-MoE is unrelated to Qwen3.5 streaming chat failure." + "reason": "Autocast dtype mismatch for Phi-tiny-MoE is unrelated to Qwen3.5 streaming API failure." }, { "left": "issue:44291", "right": "issue:45070", "accept": false, - "reason": "Unexpected _is_hf_initialized argument and pydantic PretrainedConfig break are different load/config issues." + "reason": "Unexpected _is_hf_initialized argument during init_empty_weights is not the same as the PretrainedConfig pydantic-field regression." }, { "left": "issue:43366", "right": "issue:45464", "accept": false, - "reason": "gpt-oss support and Qwen3.5 chat streaming failure are unrelated." + "reason": "GGUF gpt-oss support request is unrelated to Qwen3.5 streaming inference failure." }, { "left": "issue:43716", "right": "issue:46294", "accept": false, - "reason": "Different multimodal regressions: image-preprocessor dtype mismatch vs video token/feature mismatch." + "reason": "Both are multimodal, but Mistral image-preprocessor dtype mismatch and Qwen3.5 video token/feature mismatch are different bugs." }, { "left": "issue:43531", "right": "issue:45070", "accept": false, - "reason": "Sliding_window issue in Qwen3-MoE is not the same as the PretrainedConfig schema break." + "reason": "Qwen3-MoE sliding_window bug is unrelated to the PretrainedConfig pydantic regression." }, { "left": "issue:45020", "right": "issue:45800", "accept": false, - "reason": "Remote_code breakage and torch 2.4.1 compatibility problems are different versions of different failures." + "reason": "Remote_code breakage and torch 2.4.1/transformers 5.8.0 incompatibility are only loosely related by versioning." }, { "left": "issue:44479", "right": "issue:46294", "accept": true, - "reason": "Both describe the same Qwen3.5 video-input regression where apply_chat_template/video tokens and features no longer align in v5.3.x." + "reason": "Same Qwen video-input regression: both report video-feature/video-token mismatches in the v5.3.x line, with 44479 being the broader umbrella report." }, { "left": "issue:43541", "right": "issue:44589", "accept": false, - "reason": "Torch dynamo grouped_mm tracing error is unrelated to the Float8 storage lookup failure." + "reason": "Grouped_mm tracing failure is unrelated to missing Float8 storage support." }, { "left": "issue:41093", "right": "issue:43596", "accept": false, - "reason": "Different IndexError signatures and contexts; not the same underlying bug." + "reason": "Both are zero3-related index errors, but the concrete failure modes and contexts differ too much to treat as duplicates." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded and conservative overall. The two accepted soft pairs are well supported by the packet: 45812/45920 are the same v5 AutoTokenizer wrong-token-ID regression, and 44479/46294 are the same Qwen3.5 video/apply_chat_template mismatch. The negative verdicts also stay cautious and do not overclaim shared root causes from superficial similarity." + "feedback": "The summary is grounded in the packet and stays conservative. The canonical issue choice (45920) is well supported by the titles, and the two accepted soft pairs (45812/45920 and 44479/46294) look like plausible duplicates based on shared bug descriptions. The remaining verdicts are phrased cautiously and do not overclaim beyond the available evidence." }, "evaluator_used": true, "retried": false, @@ -970,13 +952,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 3, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:36331", "issue:38175", @@ -1061,20 +1043,20 @@ "issue:44910|issue:45859", "issue:42947|issue:44387" ], - "prepared_review_unit_hash": "293895915564a67f82a9e16aff9bf79c5e8cb58299cc160a124c3c358e99e5a8", + "prepared_review_unit_hash": "18ce39af92b151dc6c9b23b49b3468a555a76e3656885efe581f1be7b438ad3c", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37144, + "serialized_chars": 37142, "estimated_input_tokens": 9286, "estimated_eval_tokens": 18828 }, @@ -1083,210 +1065,210 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly a set of unrelated Transformers issues with only one clear duplicate thread: non-persistent buffers being corrupted/junk-filled during `from_pretrained` in v5.", - "confidence": 0.92, - "canonical_issue_reason": "issue:43950 is the clearest canonical issue: it states the regression precisely (`from_pretrained()` corrupts non-persistent buffers) and matches the closest duplicate reports about the same junk-filled buffer behavior.", + "summary": "This cluster is mostly heterogeneous. The only clear duplicate subgroup is the Transformers v5 non-persistent-buffer corruption regression; all other soft pairs look like separate bugs or unrelated model-specific issues.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 43950 is the most explicit and descriptive statement of the non-persistent-buffer corruption regression, making it the best canonical anchor for that duplicate subgroup.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43950 is the best representative issue because it is specific, regression-focused, and central to the only confirmed duplicate pair in this set.", + "best_issue_reason": "43950 best represents the only clearly duplicated bug family in this cluster: it names the regression, the affected API, and the bad buffer behavior directly.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43054", "right": "issue:43994", "accept": false, - "reason": "Both are about SigLIP2, but one is about worse text embeddings and the other about nonsensical pipeline/AutoModel results; the root cause is not clearly the same." + "reason": "Both involve SigLIP2 output quality, but the reported symptoms and contexts differ enough that they do not clearly describe the same concrete bug." }, { "left": "issue:45759", "right": "issue:45902", "accept": false, - "reason": "Both involve Qwen3.5/Qwen 3.6 internals, but one is about `text_config` unwrapping in `from_config` and the other about uninitialized rotary frequencies after materialization; different bugs." + "reason": "One is about AutoModelForCausalLM config unwrapping; the other is about uninitialized rotary embedding memory after materialization. Different code paths and failures." }, { "left": "issue:45542", "right": "issue:45800", "accept": false, - "reason": "Different problems: TensorBoard/TensorFlow backend detection versus a torch/transformers version incompatibility." + "reason": "TensorBoard-without-TensorFlow backend detection is unrelated to the torch/transformers compatibility report." }, { "left": "issue:43716", "right": "issue:44918", "accept": false, - "reason": "Different model families and failure modes: image preprocessor dtype mismatch versus Qwen3.5 embedding unpacking with TRL SFT trainer." + "reason": "Different models and failures: image preprocessor dtype mismatch vs Qwen3.5 embedding unpacking with TRL SFTTrainer." }, { "left": "issue:39186", "right": "issue:44918", "accept": false, - "reason": "FSDP `'weight' must be 2-D` and TRL embedding unpacking are unrelated failures." + "reason": "FSDP weight-dimension failure is not the same bug as Qwen3.5 input-embedding unpacking." }, { "left": "issue:45632", "right": "issue:45698", "accept": false, - "reason": "Both concern loading custom code, but one is a trust_remote_code cache-path collision and the other is loading the wrong custom module after save/load; too different to merge." + "reason": "Both touch custom-module loading, but one is a leaf-directory cache collision and the other is wrong module selection after save_pretrained; not the same defect." }, { "left": "issue:43645", "right": "issue:45399", "accept": false, - "reason": "Notebook custom-model initialization and flash-attn fallback gating are unrelated code paths." + "reason": "Notebook custom-model initialization and flash-attn2 fallback gating are unrelated issues." }, { "left": "issue:43404", "right": "issue:45005", "accept": false, - "reason": "Both mention tied weights, but one is a specific Mistral3 lm_head tying bug while the other is a broader translation-model v5 issue; not the same concrete bug." + "reason": "Both mention tied weights, but these are different models and different failure modes." }, { "left": "issue:40990", "right": "issue:44960", "accept": false, - "reason": "Perplexity on gpt-oss-20b and a GLM5 issue are unrelated." + "reason": "Different models and symptoms; no concrete shared bug is evident." }, { "left": "issue:44530", "right": "issue:45005", "accept": false, - "reason": "PagedAttentionCache/Qwen3.5 crash is a separate cache-kernel issue, not the tied-weights regression." + "reason": "PagedAttentionCache linear_attention crash is unrelated to tied-weight handling." }, { "left": "issue:42915", "right": "issue:45399", "accept": false, - "reason": "FineGrainedFP8Config failure is unrelated to flash-attn fallback checks." + "reason": "FineGrainedFP8 failure on Qwen3Moe is unrelated to the flash-attn2 fallback check ordering bug." }, { "left": "issue:43873", "right": "issue:45005", "accept": false, - "reason": "Quantized offloading behavior and tied-weights issues are different bugs." + "reason": "Quantization/offloading behavior is not the same as tied-weight handling in translation models." }, { "left": "issue:44360", "right": "issue:44512", "accept": false, - "reason": "DSA ReLU discussion and a docs typo are unrelated." + "reason": "DSA indexer behavior is unrelated to the removed run-command documentation issue." }, { "left": "issue:44863", "right": "issue:45005", "accept": false, - "reason": "NemotronH checkpoint loading and tied weights on translation models are different issues." + "reason": "NemotronH checkpoint loading failure is unrelated to translation-model tied weights." }, { "left": "issue:44918", "right": "issue:45507", "accept": false, - "reason": "Qwen3.5 embedding unpacking with TRL and GraniteMoEHybrid invalid method calls are unrelated." + "reason": "Qwen3.5 embedding unpacking and GraniteMoEHybrid invalid-method calls are different problems." }, { "left": "issue:44492", "right": "issue:44512", "accept": false, - "reason": "A cache-strategy typo and a docs command removal note are not the same bug." + "reason": "A cache-strategy typo is not the same as stale documentation about a removed command." }, { "left": "issue:43975", "right": "issue:44292", "accept": false, - "reason": "Incorrect detokenization and Qwen-3-8B-NVFP4 runtime errors are different failures." + "reason": "Detokenization regression and NVFP4 runtime error are different failures." }, { "left": "issue:44292", "right": "issue:45507", "accept": false, - "reason": "Different models and different failure modes." + "reason": "Different models and code paths; no duplicate relationship is apparent." }, { "left": "issue:43950", "right": "issue:44534", "accept": true, - "reason": "Same underlying regression: v5 `from_pretrained` corrupts/fills non-persistent buffers with junk." + "reason": "These are the same non-persistent-buffer corruption regression described with different wording." }, { "left": "issue:39692", "right": "issue:43054", "accept": false, - "reason": "SigLIP2 docs example errors versus general poor SigLIP2 text embeddings; not the same concrete issue." + "reason": "Documentation example errors are not the same as the reported SigLIP2 embedding-quality issue." }, { "left": "issue:43644", "right": "issue:43950", "accept": true, - "reason": "Direct duplicate of the non-persistent buffer corruption/junk-filled buffer regression." + "reason": "Both report the same v5 regression where non-persistent buffers are filled/corrupted with junk." }, { "left": "issue:43828", "right": "issue:44928", "accept": false, - "reason": "Both mention dtype/BF16 behavior, but the actual failures are different model/training paths." + "reason": "Both mention dtype/BF16 issues, but the model, trigger, and failure mode are different." }, { "left": "issue:36331", "right": "issue:41762", "accept": false, - "reason": "CustomTrainer keyword-argument mismatch and DeepSpeed ZeRO-3 load-time index error are unrelated." + "reason": "Trainer signature mismatch and ZeRO-3 load-time index error are unrelated." }, { "left": "issue:43950", "right": "issue:45902", "accept": false, - "reason": "Both touch buffer/materialization behavior, but one is buffer corruption in `from_pretrained` and the other is uninitialized rotary frequency memory after `to_empty`; not the same bug." + "reason": "Both involve buffer/materialization concerns, but 45902 is a model-specific uninitialized-memory bug, not the same regression as 43950." }, { "left": "issue:43872", "right": "issue:44589", "accept": false, - "reason": "Different quantization/storage incompatibilities: bitsandbytes `Int8Params` vs missing Float8 storage." + "reason": "Different low-level quantization/storage incompatibilities." }, { "left": "issue:43866", "right": "issue:44403", "accept": false, - "reason": "Checkpoint corruption and loading noise are unrelated." + "reason": "Corrupted checkpoint and loading noise are unrelated symptoms." }, { "left": "issue:38175", "right": "issue:43054", "accept": false, - "reason": "Both concern SigLIP2, but the reported symptoms differ and no shared concrete bug is evident." + "reason": "Same model family, but the observed problems differ too much to treat as the same underlying bug." }, { "left": "issue:41553", "right": "issue:43883", "accept": false, - "reason": "Bad AutoTokenizer error messaging and missing `all_tied_weights_keys` are unrelated." + "reason": "Tokenizer error messaging and missing tied-weight metadata are unrelated." }, { "left": "issue:43054", "right": "issue:45701", "accept": false, - "reason": "SigLIP2 embedding quality and tokenization changes across transformers versions are different issues." + "reason": "SigLIP2 embedding quality and version-dependent tokenization changes are different issues." }, { "left": "issue:45663", "right": "issue:45902", "accept": false, - "reason": "Gemma4 FSDP2 shared_kv_states bug and Qwen3.5 rotary buffer materialization bug are unrelated." + "reason": "Gemma4 FSDP2 shared_kv_states handling is unrelated to Qwen3_5Moe rotary embedding initialization." }, { "left": "issue:44910", "right": "issue:45859", "accept": false, - "reason": "Qwen3.5 flash-attention position_ids crash and CPU-offload incompatibility in rotary embedding are different code paths." + "reason": "FlashAttention position_ids misinterpretation and CPU-offload incompatibility are different bugs." }, { "left": "issue:42947", "right": "issue:44387", "accept": false, - "reason": "Gradient checkpointing/LoRA behavior and int4 quantization memory growth are different problems." + "reason": "Gradient checkpointing ineffectiveness and int4 quantization memory growth are different issues." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet. The accepted duplicate thread around non-persistent buffer corruption in `from_pretrained` is supported by the issue titles and the soft-pair evidence, and the other soft-edge rejections read conservative rather than overstated. Minor note: the duplicate cluster is really a three-issue thread (`43644`, `43950`, `44534`), not just a pair, but the overall conclusion is still correct." + "feedback": "The summary is grounded in the packet and stays conservative on the soft edges. The accepted duplicate subgroup around issue 43950 is well supported by the titles, and the other pair rejections do not overclaim sameness. No clear evidence of an overstated link." }, "evaluator_used": true, "retried": false, @@ -1299,13 +1281,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 4, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:36331", "issue:41093", @@ -1325,7 +1307,6 @@ "issue:43824", "issue:43827", "issue:43828", - "issue:43854", "issue:43856", "issue:43866", "issue:43872", @@ -1342,7 +1323,6 @@ "issue:44488", "issue:44509", "issue:44661", - "issue:44863", "issue:44960", "issue:44991", "issue:45020", @@ -1353,25 +1333,28 @@ "issue:45399", "issue:45520", "issue:45685", + "issue:45715", "issue:45823", - "issue:45923" + "issue:45923", + "issue:46315" ], "soft_pairs": [ - "issue:41553|issue:43881", "issue:44403|issue:44451", + "issue:41553|issue:43881", "issue:43975|issue:45520", "issue:43054|issue:43493", - "issue:44403|issue:44488", "issue:42947|issue:43856", + "issue:44403|issue:44488", "issue:43643|issue:45020", "issue:41762|issue:42915", "issue:45685|issue:45823", "issue:42491|issue:44368", - "issue:43881|issue:43883", "issue:44991|issue:45399", "issue:43824|issue:43872", + "issue:43881|issue:43883", "issue:45020|issue:45823", "issue:44661|issue:45020", + "issue:45715|issue:46315", "issue:43828|issue:45823", "issue:45081|issue:45399", "issue:43782|issue:44960", @@ -1386,252 +1369,229 @@ "issue:42947|issue:45923", "issue:43940|issue:45313", "issue:36331|issue:41093", - "issue:41553|issue:44991", - "issue:43854|issue:44863", "issue:43257|issue:45305" ], - "prepared_review_unit_hash": "0431bc28b08722a79abfa998fa381e1722cc01cfba4a64123ec7efd8414d380f", + "prepared_review_unit_hash": "32d19a0456792aad16c941d631f377220fcbd59f17d45fad59245dee87ba53ad", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 32, - "serialized_chars": 37434, - "estimated_input_tokens": 9359, - "estimated_eval_tokens": 18974 + "soft_pair_count": 31, + "serialized_chars": 37241, + "estimated_input_tokens": 9311, + "estimated_eval_tokens": 18878 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly heterogeneous closed issues with only a few clear duplicate subclusters. The strongest matches are the v5 docs/pipeline wording issue, the bitsandbytes `_is_hf_initialized` incompatibility, and a likely SigLIP2 implementation discrepancy; the rest are too model- or symptom-specific to merge safely.", - "confidence": 0.78, - "canonical_issue_reason": null, + "summary": "The cluster is mostly a loose set of transformer/model-loading regressions and tokenizer/model-compatibility issues, but the soft edges are generally only thematic, not true duplicates. No PRs are present.", + "confidence": 0.86, + "canonical_issue_reason": "Issue 45020 is the broadest and most central issue here: it covers recent transformers regressions breaking `remote_code`-based model loading, which overlaps the general loading/compatibility theme of the cluster better than the narrower model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43872 is the clearest anchor because it names a concrete, reproducible `_is_hf_initialized` / Int8Params incompatibility and has a very strong duplicate counterpart in 44291.", + "best_issue_reason": "Issue 45020 is the best global issue candidate because it is the most umbrella-like problem statement and could reasonably subsume several of the model-loading/tokenizer regressions, whereas the others are mostly narrow model-specific failures or unrelated training/runtime bugs.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:41553", - "right": "issue:43881", + "left": "issue:44403", + "right": "issue:44451", "accept": false, - "reason": "Different models and failure modes; one is Voxtral tokenizer error messaging, the other is a GLM-4v load failure." + "reason": "Both concern loading behavior, but one is about warning/noise during loading while the other is a specific model load failure; not the same underlying bug." }, { - "left": "issue:44403", - "right": "issue:44451", + "left": "issue:41553", + "right": "issue:43881", "accept": false, - "reason": "Generic loading noise vs a specific ScandiBERT load failure; not the same underlying bug." + "reason": "Different models and different failure modes: bad AutoTokenizer error for Voxtral vs a model loading failure for glm-4v-9b." }, { "left": "issue:43975", "right": "issue:45520", "accept": false, - "reason": "Tokenizer detokenization regression vs Python 3.13 flash_attn import-key error; unrelated code paths." + "reason": "One is a tokenizer detokenization regression, the other is a missing `flash_attn` import/KeyError on Python 3.13; unrelated code paths." }, { "left": "issue:43054", "right": "issue:43493", - "accept": true, - "reason": "Both point at the same SigLIP2 text-embedding discrepancy/implementation mismatch relative to expected behavior." - }, - { - "left": "issue:44403", - "right": "issue:44488", "accept": false, - "reason": "Both mention loading, but the concrete issues differ; this is too broad to treat as a duplicate." + "reason": "Both mention SigLIP2, but one is a performance/quality complaint and the other is a correctness discrepancy versus JAX; not a duplicate bug." }, { "left": "issue:42947", "right": "issue:43856", "accept": false, - "reason": "LoRA gradient checkpointing behavior is unrelated to Qwen3 MoE training memory usage." + "reason": "Both are training-related, but one is gradient checkpointing with PEFT LoRA and the other is memory usage in Qwen3 MoE training; different problems." + }, + { + "left": "issue:44403", + "right": "issue:44488", + "accept": false, + "reason": "Both are load-related, but they report different models and different symptoms; this looks like a family of regressions, not one bug." }, { "left": "issue:43643", "right": "issue:45020", "accept": false, - "reason": "One is missing fields in AutoConfig with trust_remote_code, the other is a broader remote_code regression; same area but not the same bug." + "reason": "`trust_remote_code` returning missing fields is a specific config-object bug; 45020 is a broader loading regression across remote-code models." }, { "left": "issue:41762", "right": "issue:42915", "accept": false, - "reason": "Different model families and different loading backends; no concrete shared failure path." + "reason": "Different failures in different contexts: ZeRO-3 load-time IndexError for Gemma3 vs FineGrainedFP8Config failure for Qwen3 MoE." }, { "left": "issue:45685", "right": "issue:45823", "accept": false, - "reason": "MPS histogram kernel limitation is unrelated to a Gemma4 device-mismatch forward bug." + "reason": "Both involve device/backend issues, but one is an MPS histogram implementation error in MoE and the other is a device mismatch in Gemma4 forward." }, { "left": "issue:42491", "right": "issue:44368", "accept": false, - "reason": "Both involve LoRA/training, but one is a hf4.x\u2192hf5.x Qwen3 MoE incompatibility and the other is a tie_word_embeddings warning." - }, - { - "left": "issue:43881", - "right": "issue:43883", - "accept": false, - "reason": "Different models and different errors; Molmo tie-weights attribute issue does not match a GLM-4v load failure." + "reason": "One is a Qwen3 MoE LoRA compatibility break between versions; the other is a warning about `tie_word_embeddings` during fine-tuning. Different root causes." }, { "left": "issue:44991", "right": "issue:45399", "accept": false, - "reason": "Tokenizer loading for est-roberta is unrelated to flash-attn fallback gating." + "reason": "Tokenizer loading for a specific model is not the same as FA2 fallback being blocked by other checks; only loosely related through model loading." }, { "left": "issue:43824", "right": "issue:43872", "accept": false, - "reason": "ImportError for Qwen2.5-VL is unrelated to the bitsandbytes `_is_hf_initialized` constructor issue." + "reason": "Different incompatibilities: missing Qwen2_5_VL import vs bitsandbytes `Int8Params` keyword mismatch." + }, + { + "left": "issue:43881", + "right": "issue:43883", + "accept": false, + "reason": "Both are loading failures, but one is model loading and the other is missing `all_tied_weights_keys` on a different model class; no shared concrete bug." }, { "left": "issue:45020", "right": "issue:45823", "accept": false, - "reason": "Broad remote_code breakage does not match a specific Gemma4 PLE device placement bug." + "reason": "`remote_code` loading regressions are not the same as a Gemma4 device_map runtime mismatch." }, { "left": "issue:44661", "right": "issue:45020", "accept": false, - "reason": "add-new-model-like / tokenizer mapping logic is a different path from remote_code loading regressions." + "reason": "`add-new-model-like` failing in `TOKENIZER_MAPPING_NAMES` is a tooling/extensibility bug, while 45020 is a runtime remote-code loading regression." + }, + { + "left": "issue:45715", + "right": "issue:46315", + "accept": false, + "reason": "Both involve tokenizer speed, but one is `convert_ids_to_tokens` rebuilding special-id sets; the other is slower `convert_tokens_to_ids` with many added tokens. Different hot paths." }, { "left": "issue:43828", "right": "issue:45823", "accept": false, - "reason": "Autocast dtype mismatch in Phi-tiny-MoE is not the same as Gemma4 device mismatch." + "reason": "Autocast dtype mismatch in Phi MoE and Gemma4 device mismatch are different forward-pass bugs." }, { "left": "issue:45081", "right": "issue:45399", "accept": false, - "reason": "Mistral tokenizer regex patch crash is unrelated to flash-attn2 fallback checks." + "reason": "Tokenizer regex patch crash and FA2 fallback gating are unrelated code paths." }, { "left": "issue:43782", "right": "issue:44960", "accept": false, - "reason": "Qwen3VL weight_only loading error and GLM5 are unrelated model-specific failures." - }, - { - "left": "issue:43872", - "right": "issue:44291", - "accept": true, - "reason": "Same underlying `_is_hf_initialized` incompatibility during model init/loading with bitsandbytes/init_empty_weights." - }, - { - "left": "issue:43257", - "right": "issue:43866", - "accept": false, - "reason": "Qwen3 MoE weight conversion under accelerate/deepspeed is unrelated to an Ovis2 checkpoint corruption report." - }, - { - "left": "issue:42915", - "right": "issue:43872", - "accept": false, - "reason": "FineGrainedFP8Config failure and bitsandbytes `_is_hf_initialized` incompatibility are different bugs." - }, - { - "left": "issue:42947", - "right": "issue:45923", - "accept": false, - "reason": "Gradient checkpointing behavior is unrelated to a missing supports_gradient_checkpointing flag." - }, - { - "left": "issue:43940", - "right": "issue:45313", - "accept": false, - "reason": "Both are ZeRO-3 weight-loading failures, but the model variants and missing-weight patterns are not specific enough to assert the same root cause." - }, - { - "left": "issue:36331", - "right": "issue:41093", - "accept": false, - "reason": "Unexpected compute_loss kwarg error and tensor mask shape mismatch are unrelated." - }, - { - "left": "issue:41553", - "right": "issue:44991", - "accept": false, - "reason": "Different tokenizer/model loading issues with no shared concrete failure path." - }, - { - "left": "issue:43854", - "right": "issue:44863", - "accept": false, - "reason": "GLM-4.7-Flash unit-test load failure is unrelated to NemotronH checkpoint loading." + "reason": "Qwen3VL `weight_only=True` loading error and generic GLM5 issue do not point to the same underlying bug." }, { "left": "issue:45020", "right": "issue:45356", "accept": false, - "reason": "Remote_code regressions and Kimi-K2.5 tokenizer codec/fix_mistral_regex problems are different issues." + "reason": "Remote-code loading regressions and Kimi-K2.5 tokenizer codec handling are different issues; the latter only mentions a misleading warning as a secondary symptom." }, { "left": "issue:43742", "right": "issue:45399", "accept": false, - "reason": "MobileLLM key error and flash-attn fallback gating are unrelated." + "reason": "Loading facebook/MobileLLM-125M failing is unrelated to the FA2 fallback gating issue." }, { "left": "issue:43404", "right": "issue:43950", "accept": false, - "reason": "Tied lm_head weights in Mistral3 is unrelated to non-persistent buffer corruption in from_pretrained." + "reason": "Untied `lm_head` weights in Mistral3 and silent corruption of non-persistent buffers are different model state bugs." }, { "left": "issue:43827", "right": "issue:44509", - "accept": true, - "reason": "Both report the same v5 docs regression: stale references to removed pipeline tasks." + "accept": false, + "reason": "Both are docs about removed pipeline tasks, but they concern different documentation locations/symptoms and are not the same concrete issue." }, { "left": "issue:43792", "right": "issue:44220", "accept": false, - "reason": "Whisper cannot run is too broad to safely equate with a specific _torch_extract_fbank_features issue." + "reason": "Whisper load failure and `_torch_extract_fbank_features()` issue are not the same bug; only both live in audio/model loading." }, { "left": "issue:43872", "right": "issue:44291", - "accept": true, - "reason": "Same `_is_hf_initialized` incompatibility manifesting through different loading entry points." + "accept": false, + "reason": "Both mention `_is_hf_initialized`, but one is a bitsandbytes incompatibility and the other is `init_empty_weights`/unexpected kwarg during loading; related but not the same concrete failure." }, { "left": "issue:43257", "right": "issue:43866", "accept": false, - "reason": "Different models and different failure classes; no shared root cause." + "reason": "Qwen3 MoE weight conversion with accelerate/deepspeed and corrupted Ovis2 checkpoint are distinct issues." }, { "left": "issue:42915", "right": "issue:43872", "accept": false, - "reason": "Qwen3 MoE FP8 training failure is not the same as the bitsandbytes/init_empty_weights constructor regression." + "reason": "FineGrainedFP8Config failure and bitsandbytes `_is_hf_initialized` incompatibility are different model-loading/runtime incompatibilities." + }, + { + "left": "issue:42947", + "right": "issue:45923", + "accept": false, + "reason": "Gradient checkpointing ineffective with LoRA and missing `supports_gradient_checkpointing` flag are related by topic, but not the same bug." + }, + { + "left": "issue:43940", + "right": "issue:45313", + "accept": false, + "reason": "Both involve DeepSpeed ZeRO-3 loading failures for Qwen variants, but they affect different model families and likely different missing-weight paths." + }, + { + "left": "issue:36331", + "right": "issue:41093", + "accept": false, + "reason": "`compute_loss()` signature mismatch and mask shape IndexError are completely unrelated training errors." + }, + { + "left": "issue:43257", + "right": "issue:45305", + "accept": false, + "reason": "One is Qwen3 MoE weight conversion with accelerate/deepspeed; the other is gradient averaging with GAS under DeepSpeed and `model_accepts_loss_kwargs=True`. Different code paths." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative about weaker links. The accepted soft-edge pairs are well supported by the issue titles/body themes, and the rejected pairs are cautious rather than overstated. No mergeability or same-bugfix claim appears inflated." - }, - "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 }, @@ -1640,13 +1600,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 5, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:24643", "issue:30064", @@ -1655,7 +1615,6 @@ "issue:39692", "issue:41093", "issue:41553", - "issue:42491", "issue:42617", "issue:42915", "issue:43317", @@ -1667,9 +1626,9 @@ "issue:43638", "issue:43645", "issue:43646", - "issue:43720", "issue:43824", "issue:43828", + "issue:43854", "issue:43950", "issue:43957", "issue:43994", @@ -1687,6 +1646,7 @@ "issue:44910", "issue:44918", "issue:44928", + "issue:44991", "issue:45030", "issue:45325", "issue:45584", @@ -1698,11 +1658,13 @@ "issue:45902" ], "soft_pairs": [ + "issue:43854|issue:44863", "issue:44928|issue:45588", + "issue:41553|issue:44991", "issue:41093|issue:44805", "issue:44479|issue:44560", - "issue:41553|issue:44488", "issue:44530|issue:44863", + "issue:41553|issue:44488", "issue:24643|issue:30064", "issue:45325|issue:45902", "issue:43366|issue:43828", @@ -1728,235 +1690,240 @@ "issue:43596|issue:44661", "issue:44661|issue:44805", "issue:43317|issue:44756", - "issue:38175|issue:39692", - "issue:42491|issue:43720" + "issue:38175|issue:39692" ], - "prepared_review_unit_hash": "a0420541163aee96969bd7ccb5e627e301c2e99e713f13480fb45f92ca8648f7", + "prepared_review_unit_hash": "466adf75844cdee42297c10caed381906fb4b9cb2d07445312b0f9b60abe3eff", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 32, - "serialized_chars": 37486, - "estimated_input_tokens": 9372, - "estimated_eval_tokens": 19000 + "soft_pair_count": 33, + "serialized_chars": 37723, + "estimated_input_tokens": 9431, + "estimated_eval_tokens": 19118 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: the soft pairs span different models, symptoms, and code paths. Only the two Transformers 5.0 custom model initialization issues look like the same underlying bug.", - "confidence": 0.93, - "canonical_issue_reason": "issue:43646 is the broader, cleaner description of the same Transformers 5.0 custom-model initialization regression; 43645 is the notebook-specific variant.", + "summary": "This cluster is mostly heterogeneous and does not look like one true duplicate set. The only convincing soft duplicate is the Transformers 5.0 custom model initialization regression pair (43645/43646).", + "confidence": 0.83, + "canonical_issue_reason": "issue:43646 is the broader report of the Transformers 5.0 custom model initialization regression; issue:43645 appears to be the same bug reproduced in a Jupyter notebook context.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43646 is the best representative because it captures the core regression without the extra Jupyter-specific framing.", + "best_issue_reason": "issue:43646 is the clearest representative of the only real duplicate-like problem in the set, and it states the underlying regression more generally than the notebook-specific variant.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43854", + "right": "issue:44863", + "accept": false, + "reason": "Different models and failure modes: GLM-4.7 loading in tests vs NemotronH checkpoint loading." + }, { "left": "issue:44928", "right": "issue:45588", "accept": false, - "reason": "Both mention Qwen3.5/3D position_ids, but one is a flash-attention illegal-access crash and the other is SDPA/BF16 training instability; different concrete bugs and likely different fixes." + "reason": "Both are Qwen-related runtime bugs, but they hit different code paths: RLHF/3D position_ids vs flash_attention s_aux handling." + }, + { + "left": "issue:41553", + "right": "issue:44991", + "accept": false, + "reason": "Both involve tokenizer loading, but the reported models and symptoms differ; error-message quality vs outright tokenizer load failure." }, { "left": "issue:41093", "right": "issue:44805", "accept": false, - "reason": "Same generic IndexError wording, but no evidence they share the same failing code path or model; too little to treat as a duplicate." + "reason": "The errors look similarly worded, but there is no evidence they come from the same underlying off-by-one bug or code path." }, { "left": "issue:44479", "right": "issue:44560", "accept": false, - "reason": "Both are video-related Qwen regressions, but one is a broad v5.3.0 input regression and the other is a StopIteration in Qwen3-vl-embedding; not clearly the same bug." + "reason": "Both concern video inputs, but one is a v5.3.0 regression across Qwen VL models and the other is a Qwen3-vl-embedding StopIteration issue." }, { - "left": "issue:41553", - "right": "issue:44488", + "left": "issue:44530", + "right": "issue:44863", "accept": false, - "reason": "Both concern loading/tokenizer errors, but they involve different models and different failure modes; not the same underlying issue." + "reason": "PagedAttentionCache linear_attention crash and NemotronH checkpoint loading are unrelated." }, { - "left": "issue:44530", - "right": "issue:44863", + "left": "issue:41553", + "right": "issue:44488", "accept": false, - "reason": "Different model families and different loader/cache failures; no shared concrete bug." + "reason": "Different models and different failures: Voxtral tokenizer error messaging vs sleng-bert loading failure." }, { "left": "issue:24643", "right": "issue:30064", "accept": false, - "reason": "Completely different domains: DeepSpeed training weight shape error vs void segmentation map image-processor failure." + "reason": "Completely unrelated: DeepSpeed training weight shape error vs image processor void-segmentation-map handling." }, { "left": "issue:45325", "right": "issue:45902", "accept": false, - "reason": "Both are low-level model internals, but one is Qwen2.5-VL rope scaling and the other is Qwen3.5-MoE uninitialized memory after meta materialization; unrelated fixes." + "reason": "Both involve positional embeddings/indices, but they affect different models and distinct bugs." }, { "left": "issue:43366", "right": "issue:43828", "accept": false, - "reason": "GGUF/gpt-oss support and a Phi-tiny-MoE autocast dtype mismatch are unrelated." + "reason": "Unrelated: GGUF gpt-oss support vs Phi MoE autocast dtype mismatch." }, { "left": "issue:45663", "right": "issue:45823", "accept": false, - "reason": "Same Gemma4 family, but one is an FSDP2 shared_kv_states KeyError and the other is a device_map forward mismatch; different code paths." + "reason": "Same model family, but different concrete bugs: FSDP2 shared_kv_states rebuild vs device_map auto device mismatch." }, { "left": "issue:43366", "right": "issue:43531", "accept": false, - "reason": "No overlap beyond broad Qwen/GGUF context; unrelated issues." + "reason": "Different issues entirely: GGUF architecture support vs Qwen3-MoE sliding_window behavior." }, { "left": "issue:43950", "right": "issue:45834", "accept": false, - "reason": "Non-persistent buffer loading regression vs Kosmos2.5 OCR index error; unrelated." + "reason": "Non-persistent buffer regression and Kosmos2.5 long-OCR index error are unrelated." }, { "left": "issue:43550", "right": "issue:44910", "accept": false, - "reason": "Different models and different attention paths; one is Bamba SDPA/compile, the other is Qwen3.5 Flash Attention with 3D position_ids." + "reason": "Different models and failure modes: torch.compile/SDPA on Bamba vs Flash Attention illegal memory access on Qwen3.5." }, { "left": "issue:43645", "right": "issue:43646", "accept": true, - "reason": "These describe the same Transformers 5.0 regression breaking custom model initialization; the notebook mention in 43645 is just a usage context, not a distinct bug." + "reason": "Same underlying Transformers 5.0 custom model initialization regression; 43645 is a Jupyter notebook-specific reproduction of the broader 43646 report." }, { "left": "issue:44560", "right": "issue:45685", "accept": false, - "reason": "Video inference StopIteration and an MPS histogram op-not-implemented error are unrelated." + "reason": "Video StopIteration in Qwen3-vl-embedding is unrelated to an MPS histogram kernel limitation in moe." }, { "left": "issue:43366", "right": "issue:44589", "accept": false, - "reason": "GGUF model support and a Float8 storage lookup error are different problems." + "reason": "GGUF gpt-oss support and Float8 storage lookup failure are unrelated bugs." }, { "left": "issue:42915", "right": "issue:43957", "accept": false, - "reason": "Qwen3Moe FP8 config failure vs meta-device loading issues; different failure modes and likely different fixes." + "reason": "Different regressions: Qwen3Moe FineGrainedFP8Config handling vs meta-device model loading on Transformers 5.x." }, { "left": "issue:44560", "right": "issue:44918", "accept": false, - "reason": "Both involve Qwen3.5/video-ish workflows, but one is a video StopIteration and the other is embedding unpacking with TRL SFT; not the same bug." + "reason": "Both mention Qwen3.5/video-ish inputs, but one is a video StopIteration and the other is an SFT trainer embedding unpacking failure." }, { "left": "issue:43824", "right": "issue:43957", "accept": false, - "reason": "Missing class import and meta-device loading regression are unrelated." + "reason": "Importing a missing class and meta-device loading failures are different problems." }, { "left": "issue:42617", "right": "issue:43366", "accept": false, - "reason": "3d_parallel.py runtime failure and GGUF gpt-oss support are unrelated." + "reason": "3d_parallel.py execution failure and GGUF gpt-oss support do not share a concrete bug." }, { "left": "issue:36010", "right": "issue:42915", "accept": false, - "reason": "GenerationMixin import error vs Qwen3Moe FP8 failure; unrelated." + "reason": "GenerationMixin import error and Qwen3Moe FP8 failure are unrelated." }, { "left": "issue:44661", "right": "issue:45823", "accept": false, - "reason": "Tokenizer mapping metadata bug and Gemma4 device mismatch are unrelated." + "reason": "Tokenizer mapping add-new-model-like logic and Gemma4 PLE device mismatch are different issues." }, { "left": "issue:43632", "right": "issue:45834", "accept": false, - "reason": "_is_hf_initialized flag regression and Kosmos2.5 OCR index error are different bugs." + "reason": "Transformers v5 _is_hf_initialized regression is unrelated to Kosmos2.5 OCR indexing." }, { "left": "issue:38175", "right": "issue:43994", "accept": false, - "reason": "Both mention SigLIP2, but one is zero probabilities and the other is a model/pipeline nonsense-results report; not enough evidence of the same root cause." + "reason": "Same model family, but one is unexpected zero probabilities and the other is a docs/pipeline misuse report with nonsensical outputs." }, { "left": "issue:44155", "right": "issue:45584", "accept": false, - "reason": "AudioFlamingo3 batched embedding/token leak vs Whisper empty-transcription generation failure; different models and mechanisms." + "reason": "AudioFlamingo3 track leakage and Whisper empty-transcription handling are unrelated." }, { "left": "issue:44560", "right": "issue:44805", "accept": false, - "reason": "StopIteration in Qwen3-vl-embedding video input vs a generic mask/tensor shape mismatch; not the same failure." + "reason": "Both are index-ish failures, but the models and code paths differ; no evidence of a shared root cause." }, { "left": "issue:44877", "right": "issue:45030", "accept": false, - "reason": "Strict config loading for granite_speech and tiny-random glm4v config validation are separate config issues." + "reason": "Strict config loading for granite_speech and tiny-random glm4v config validation are different config regressions." }, { "left": "issue:43638", "right": "issue:44661", "accept": false, - "reason": "DeepSpeed zero3/BertModel indexing error and tokenizer-mapping-related model addition failure are unrelated." + "reason": "DeepSpeed zero3 Bert initialization and TOKENIZER_MAPPING_NAMES add-new-model-like behavior are unrelated." }, { "left": "issue:43596", "right": "issue:44661", "accept": false, - "reason": "Both touch initialization/loading, but one is a zero3 Bert index error and the other is add-new-model-like metadata handling; different bugs." + "reason": "Different bugs: zero3 init with BertModel vs tokenizer-mapping-related add-new-model-like failure." }, { "left": "issue:44661", "right": "issue:44805", "accept": false, - "reason": "Tokenizer mapping metadata bug vs generic index-mask mismatch; unrelated." + "reason": "Tokenizer-mapping logic and mask-shape IndexError are unrelated." }, { "left": "issue:43317", "right": "issue:44756", "accept": false, - "reason": "device_map/offload loading failure and Strix Halo mmap OOM are different memory-management issues." + "reason": "device_map auto dequantized load failure and Strix Halo mmap OOM are different issues." }, { "left": "issue:38175", "right": "issue:39692", "accept": false, - "reason": "Both involve SigLIP2, but 39692 is a docs/example problem with model-processor mismatch and quantization failure, not the same as zero probabilities." - }, - { - "left": "issue:42491", - "right": "issue:43720", - "accept": false, - "reason": "Qwen3-MoE LoRA compatibility across versions and BitNet packed-weight loading are unrelated." + "reason": "SigLIP2 zero-probabilities bug and SigLIP2 docs example errors are related only by model family, not the same concrete failure." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary stays grounded in the packet and is appropriately conservative. The only accepted pair, issue:43645 / issue:43646, is well supported by the titles and descriptions as the same Transformers 5.0 custom-model initialization regression, with 43645 appearing as a notebook-specific variant. The other pairwise rejections are cautious and do not overclaim shared root causes." + "feedback": "The summary is well grounded in the packet and stays conservative. The only accepted soft duplicate, issue:43645/issue:43646, is supported by the titles and the similarity signal. The other pairwise rejections are phrased cautiously and do not overclaim shared root causes." }, "evaluator_used": true, "retried": false, @@ -1969,13 +1936,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 6, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:36010", "issue:39404", @@ -1991,6 +1958,7 @@ "issue:43644", "issue:43645", "issue:43708", + "issue:43720", "issue:43854", "issue:43856", "issue:43873", @@ -2026,6 +1994,7 @@ "issue:45859" ], "soft_pairs": [ + "issue:42491|issue:43720", "issue:44960|issue:44977", "issue:41720|issue:42915", "issue:45313|issue:45357", @@ -2034,253 +2003,253 @@ "issue:43708|issue:44075", "issue:43644|issue:44534", "issue:42915|issue:44756", - "issue:45488|issue:45812", "issue:39692|issue:43994", + "issue:45488|issue:45812", "issue:43854|issue:43931", "issue:43257|issue:43854", "issue:43856|issue:43873", "issue:42491|issue:43931", - "issue:43644|issue:43645", "issue:43856|issue:44387", + "issue:43644|issue:43645", "issue:42491|issue:45292", "issue:44910|issue:45588", "issue:44488|issue:44991", - "issue:45362|issue:45406", "issue:44164|issue:44291", + "issue:45362|issue:45406", "issue:45276|issue:45335", "issue:43404|issue:43454", "issue:36010|issue:39404", "issue:39692|issue:44960", - "issue:43425|issue:43957", "issue:42947|issue:45704", + "issue:43425|issue:43957", "issue:44402|issue:44488", "issue:45663|issue:45759", "issue:45663|issue:45859", "issue:43873|issue:44387" ], - "prepared_review_unit_hash": "5127f3558a92c098a6e315a39557e1ee247013631ce56d92d1bc8ce4fbf89f46", + "prepared_review_unit_hash": "aa4b0e9d112cc083e61c6254ed7e5762155bd95ad979e4ae50c92124658384d8", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 31, - "serialized_chars": 36443, - "estimated_input_tokens": 9111, - "estimated_eval_tokens": 18478 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37349, + "estimated_input_tokens": 9338, + "estimated_eval_tokens": 18932 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous. The only strong duplicate pair is 43644/44534 about non-persistent buffers being filled with junk. The Gemma/T5Gemma resize_token_embeddings reports are similar in theme but are not a true duplicate pair here because they affect different model families and different submodules.", - "confidence": 0.79, - "canonical_issue_reason": "Issue 43644 is the best canonical issue because it is the clearest, broader regression report and it has the strongest confirmed duplicate match in 44534.", + "summary": "This set is mostly a heterogeneous collection of unrelated Transformers issues, with one clear duplicate pair around the v5 non-persistent buffer regression. Most other soft edges are only loosely similar by subsystem or model family, not the same concrete bug.", + "confidence": 0.82, + "canonical_issue_reason": "issue:43644 is the strongest canonical representative because it has the clearest near-verbatim duplicate in issue:44534 and directly names the underlying Transformers v5 buffer initialization regression.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43644 is the best representative issue for the cluster: it captures the only clearly duplicated bug and is more general than the other model-specific reports.", + "best_issue_reason": "issue:43644 is the best issue candidate overall because it is the most clearly supported by a duplicate-style twin and describes the bug in the broadest, most reusable way.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "Different bugs: qwen3_moe LoRA incompatibility across HF versions vs BitNet packed-weight handling during accelerate loading." + }, { "left": "issue:44960", "right": "issue:44977", "accept": false, - "reason": "Same broad area (generation) but different models and different failure modes." + "reason": "Different models and failure modes: GLM5 vs Qwen3.5 flash-attention generation behavior." }, { "left": "issue:41720", "right": "issue:42915", "accept": false, - "reason": "CUDA auto-device-map assert and LoRA version-compatibility are unrelated bugs." + "reason": "Different code paths: device-mapping cuda assert on Qwen3 vs FineGrainedFP8Config failure on Qwen3Moe." }, { "left": "issue:45313", "right": "issue:45357", "accept": false, - "reason": "Both are Qwen3.5 reports, but one is ZeRO-3 loading and the other is save_pretrained key serialization." + "reason": "Different regressions: DeepSpeed ZeRO-3 loading of language_model weights vs save_pretrained writing wrong visual encoder keys." }, { "left": "issue:44451", "right": "issue:44488", "accept": false, - "reason": "Both are load failures, but they concern different models and no shared concrete code path is evident." + "reason": "Both are load failures for different BERT models, but the titles do not indicate the same underlying cause." }, { "left": "issue:42491", "right": "issue:45276", "accept": false, - "reason": "Different bug classes: qwen3_moe hf4.x/hf5.x incompatibility vs Gemma4 resize_token_embeddings propagation." + "reason": "Unrelated issues: qwen3_moe LoRA on hf4.x vs Gemma4 resize_token_embeddings propagation." }, { "left": "issue:43708", "right": "issue:44075", "accept": false, - "reason": "Trainer checkpoint math and SGD argument handling are unrelated." + "reason": "Trainer checkpoint math and optimizer argument usage are separate problems." }, { "left": "issue:43644", "right": "issue:44534", "accept": true, - "reason": "Near-identical reports of the same non-persistent-buffer junk regression in Transformers 5." + "reason": "Near-identical regression about non-persistent buffers being filled with junk in Transformers v5." }, { "left": "issue:42915", "right": "issue:44756", "accept": false, - "reason": "Qwen3 MoE loading/version issue is unrelated to Strix Halo mmap/OOM behavior." + "reason": "Different failures: Qwen3Moe FP8 config vs mmap-related OOM on Strix Halo." }, { - "left": "issue:45488", - "right": "issue:45812", + "left": "issue:39692", + "right": "issue:43994", "accept": false, - "reason": "Both involve tokenization, but they hit different tokenizer logic and different model families." + "reason": "Both involve SigLIP2, but one is a docs/example mismatch and quantization failure while the other is bad inference output." }, { - "left": "issue:39692", - "right": "issue:43994", + "left": "issue:45488", + "right": "issue:45812", "accept": false, - "reason": "SigLIP2 docs/example errors are not the same as a checkpoint-specific AutoModel/pipeline output regression." + "reason": "Both are tokenizer regressions, but they affect different model families and different tokenization behaviors." }, { "left": "issue:43854", "right": "issue:43931", "accept": false, - "reason": "Different loading failures: unit-test load correctness vs weight-shape mismatch." + "reason": "Different models and distinct loading errors: GLM-4.7-Flash unit-test loading vs Qwen3-VL shape mismatch." }, { "left": "issue:43257", "right": "issue:43854", "accept": false, - "reason": "Qwen3 MoE weight conversion under accelerate/deepspeed is unrelated to GLM-4.7-Flash loading tests." + "reason": "Different bugs: Qwen3 MoE weight conversion with accelerate/deepspeed vs GLM-4.7-Flash loading in tests." }, { "left": "issue:43856", "right": "issue:43873", "accept": false, - "reason": "Both mention memory, but one is training memory usage and the other is quantization/offloading behavior." + "reason": "Both are memory-related, but one is Qwen3 MoE training inefficiency and the other is quantization offloading behavior." }, { "left": "issue:42491", "right": "issue:43931", "accept": false, - "reason": "LoRA hf-version incompatibility is unrelated to Qwen3-VL weight-shape mismatch." + "reason": "Different model families and distinct failures: qwen3_moe LoRA portability vs Qwen3-VL weight shape mismatch." }, { - "left": "issue:43644", - "right": "issue:43645", + "left": "issue:43856", + "right": "issue:44387", "accept": false, - "reason": "Different regressions: buffer initialization junk vs custom model initialization in notebooks." + "reason": "Both mention memory, but the causes differ: MoE training inefficiency vs int4 quantization reserved-memory/OOM regression." }, { - "left": "issue:43856", - "right": "issue:44387", + "left": "issue:43644", + "right": "issue:43645", "accept": false, - "reason": "Related to memory pressure, but not the same concrete bug or fix." + "reason": "Related Transformer 5 regressions, but one is buffer initialization and the other is custom-model initialization in Jupyter." }, { "left": "issue:42491", "right": "issue:45292", "accept": false, - "reason": "LoRA/model-version compatibility is unrelated to the resize_token_embeddings output_embeddings bug." + "reason": "Different issues: qwen3_moe LoRA compatibility vs resize_token_embeddings not updating output embeddings." }, { "left": "issue:44910", "right": "issue:45588", "accept": false, - "reason": "Different flash-attention crashes with different causes and affected paths." + "reason": "Both are flash-attention crashes, but the concrete failure modes differ: illegal memory access from packed-sequence handling vs AttributeError on s_aux=None." }, { "left": "issue:44488", "right": "issue:44991", "accept": false, - "reason": "Tokenizer/load failures for different models; no shared root cause is clear enough." + "reason": "Different tokenizer/loading bugs affecting different models." }, { - "left": "issue:45362", - "right": "issue:45406", + "left": "issue:44164", + "right": "issue:44291", "accept": false, - "reason": "Qwen3.5 chat crash and Gemma4Processor _tokenizer error are separate code paths." + "reason": "Different mechanics: extra_state save/load handling vs init_empty_weights argument compatibility." }, { - "left": "issue:44164", - "right": "issue:44291", + "left": "issue:45362", + "right": "issue:45406", "accept": false, - "reason": "Extra_state save/load handling is distinct from init_empty_weights argument incompatibility." + "reason": "Separate crashes in different serving/chat paths and different model classes." }, { "left": "issue:45276", "right": "issue:45335", "accept": false, - "reason": "Both are resize_token_embeddings regressions, but they affect different model families and different submodules, so this is not a true duplicate." + "reason": "Both concern resize_token_embeddings, but the affected model structures differ and the titles do not show the same concrete bug." }, { "left": "issue:43404", "right": "issue:43454", "accept": false, - "reason": "Both mention lm_head tying, but the model classes and concrete failure paths differ." + "reason": "Both are lm_head weight-tying issues, but they affect different architectures and are not clearly the same code-path bug." }, { "left": "issue:36010", "right": "issue:39404", "accept": false, - "reason": "ImportError for GenerationMixin and Whisper return_language pipeline behavior are unrelated." + "reason": "Completely different problems: import failure for GenerationMixin vs Whisper pipeline return_language regression." }, { "left": "issue:39692", "right": "issue:44960", "accept": false, - "reason": "Different models and different symptom types; not the same bug." + "reason": "Different issues and model families; only superficial similarity in being model-related." }, { - "left": "issue:43425", - "right": "issue:43957", + "left": "issue:42947", + "right": "issue:45704", "accept": false, - "reason": "Torch version compatibility is unrelated to meta-device loading regressions." + "reason": "Unrelated regressions: LoRA gradient checkpointing behavior vs apex RMSNorm memory leak." }, { - "left": "issue:42947", - "right": "issue:45704", + "left": "issue:43425", + "right": "issue:43957", "accept": false, - "reason": "Gradient checkpointing with LoRA is unrelated to the T5 apex RMSNorm memory leak." + "reason": "Torch version compatibility issue vs meta-device loading regression; different root causes." }, { "left": "issue:44402", "right": "issue:44488", "accept": false, - "reason": "Tokenizer vocab-size mismatch is different from a model-loading failure for sleng-bert." + "reason": "Token-vocab mismatch in one model vs inability to load a different model." }, { "left": "issue:45663", "right": "issue:45759", "accept": false, - "reason": "Gemma4 FSDP2 attention-state rebuild and Qwen3.5/3.6 config unwrapping are unrelated." + "reason": "Different bugs in different model families: FSDP2 shared_kv_states rebuild vs composite-config unwrapping." }, { "left": "issue:45663", "right": "issue:45859", "accept": false, - "reason": "Different failures in different models: Gemma4 attention state vs Qwen3_5Moe CPU offload." - }, - { - "left": "issue:43873", - "right": "issue:44387", - "accept": false, - "reason": "Quantization offloading and int4 reserved-memory OOM are related only broadly, not as one duplicate bug." + "reason": "Different issues: Gemma-4 FSDP2 KeyError vs Qwen3_5MoeTextRotaryEmbedding CPU-offload incompatibility." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The cluster-level summary is mostly grounded, and the 43644/44534 duplicate call is well supported by the titles. However, the analyst overstates the Gemma/T5Gemma pair: 45276 and 45335 are similar resize_token_embeddings regressions, but they affect different model families and different submodules, so the packet does not justify treating them as a true duplicate with high confidence. The soft-edge verdicts should stay more conservative." + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The only accepted soft edge is well supported by the near-identical 43644/44534 titles, and the rejected pairs are generally justified by distinct model families or failure modes." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -2290,13 +2259,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 7, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:39404", "issue:41720", @@ -2316,7 +2285,6 @@ "issue:43957", "issue:44155", "issue:44164", - "issue:44265", "issue:44291", "issue:44292", "issue:44387", @@ -2327,10 +2295,8 @@ "issue:44589", "issue:44683", "issue:44756", - "issue:44805", "issue:44863", "issue:44877", - "issue:44898", "issue:44912", "issue:44928", "issue:44960", @@ -2344,8 +2310,10 @@ "issue:45684", "issue:45698", "issue:45701", + "issue:45812", "issue:45823", - "issue:46050" + "issue:46050", + "issue:46315" ], "soft_pairs": [ "issue:44155|issue:45823", @@ -2359,12 +2327,13 @@ "issue:43854|issue:45216", "issue:43381|issue:44928", "issue:44164|issue:44756", - "issue:45588|issue:45663", "issue:45684|issue:45698", "issue:41720|issue:44155", "issue:44479|issue:44877", + "issue:45588|issue:45663", "issue:43957|issue:44756", "issue:43646|issue:43708", + "issue:45812|issue:46315", "issue:45292|issue:45335", "issue:43645|issue:44164", "issue:44402|issue:44451", @@ -2378,237 +2347,234 @@ "issue:43425|issue:45823", "issue:44530|issue:45084", "issue:43425|issue:43716", - "issue:43906|issue:44530", - "issue:44265|issue:44898", - "issue:39404|issue:44912", - "issue:44479|issue:44805" + "issue:43906|issue:44530" ], - "prepared_review_unit_hash": "e88e86c42f1168a91e3dc0df819d6c6715b48096eb644fc5cfff203398e1045c", + "prepared_review_unit_hash": "27f3d8c709d93eafe18e6775a5c2630326634fe35af3910febe78e03ca8e08c8", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 34, - "serialized_chars": 37909, - "estimated_input_tokens": 9478, - "estimated_eval_tokens": 19212 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36690, + "estimated_input_tokens": 9173, + "estimated_eval_tokens": 18602 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most pairs are unrelated regressions across different subsystems. The only clear duplicate here is the Gemma4/resize_token_embeddings reports, where one issue is a broader version of the other.", - "confidence": 0.82, - "canonical_issue_reason": "issue:45276 is the most complete description of the resize_token_embeddings regression and subsumes the narrower 45292 report.", + "summary": "This cluster is loosely related: it mixes several distinct Transformers v5 regressions (loading/saving, tokenization, quantization, checkpointing, and a small resize_token_embeddings subfamily). Most soft links are false positives; only the resize-token-embeddings reports look like the same underlying bug family.", + "confidence": 0.34, + "canonical_issue_reason": "Issue 45701 is the broadest and most central open regression report here, with multiple inbound references and the widest umbrella over v5 behavior changes, even though the cluster is not a single exact duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45276 is the best representative issue: it is specific, reproducible, and covers the broader symptom set in the only clear duplicate pair.", + "best_issue_reason": "Issue 45701 is the best overall representative because it is the most general, active, and referenced issue among these mixed regression reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44155", "right": "issue:45823", "accept": false, - "reason": "Different failures: audio batched inference token leakage vs Gemma4 device-map forward mismatch." + "reason": "Different bugs: batched AudioFlamingo3 leakage between tracks vs Gemma4 PLE device mismatch in forward." }, { "left": "issue:44291", "right": "issue:44756", "accept": false, - "reason": "Unrelated code paths: init_empty_weights argument handling vs mmap-related OOM behavior." + "reason": "Unexpected _is_hf_initialized arg during init_empty_weights is unrelated to Strix Halo mmap-related OOM." }, { "left": "issue:44292", "right": "issue:44912", "accept": false, - "reason": "Both involve 4-bit quantization, but they are different models and failures (NVFP4 load error vs MXFP4 fallback to bf16)." + "reason": "Both are quantization/loading regressions, but the concrete failures and model paths differ (NVFP4 runtime error vs MXFP4 fallback to bf16)." }, { "left": "issue:44960", "right": "issue:46050", "accept": false, - "reason": "GLM5 model-loading issue vs sinq save/load corruption; same broad area, different bug." + "reason": "GLM5 loading failure and sinq save/load corruption are different underlying bugs." }, { "left": "issue:43335", "right": "issue:45663", "accept": false, - "reason": "Sparse-layer config creation bug and Gemma4 FSDP2 attention-state bug are distinct." + "reason": "Sparse-layer creation in SwitchTransformers is unrelated to Gemma4 FSDP2 shared_kv_states rebuilding." }, { "left": "issue:44387", "right": "issue:44756", "accept": false, - "reason": "Different OOM causes: int4 quantization reserved-memory growth vs mmap on Strix Halo." + "reason": "Int4 CUDA memory regression and Strix Halo mmap OOM are both memory issues, but not the same code-path bug." }, { "left": "issue:42947", "right": "issue:43381", "accept": false, - "reason": "Both mention gradient checkpointing, but one is LoRA effectiveness in training and the other is an eval-mode restriction." + "reason": "Gradient checkpointing ineffective with PEFT LoRA is distinct from the eval-mode restriction on checkpointing." }, { "left": "issue:45276", "right": "issue:45292", "accept": true, - "reason": "Same underlying resize_token_embeddings bug; 45292 is the narrower output_embeddings symptom of the broader failure reported in 45276." + "reason": "Same resize_token_embeddings bug family: both say resized embeddings are not propagated, with 45276 adding a Gemma4-specific per-layer/output-embedding manifestation." }, { "left": "issue:43854", "right": "issue:45216", "accept": false, - "reason": "Different regressions: GLM-4.7-Flash loading failure vs Qwen3.5 save_pretrained checkpoint corruption." + "reason": "Model-loading failure in unit tests and Qwen3.5 save_pretrained regression are different issues." }, { "left": "issue:43381", "right": "issue:44928", "accept": false, - "reason": "Gradient-checkpointing eval-mode issue and RLHF NaN/SDPA-fallback issue are unrelated." + "reason": "Eval-mode checkpointing prohibition is unrelated to the RLHF NaN/SDPA fallback gradient explosion." }, { "left": "issue:44164", "right": "issue:44756", "accept": false, - "reason": "save/from_pretrained extra_state handling is unrelated to mmap/OOM on Strix Halo." - }, - { - "left": "issue:45588", - "right": "issue:45663", - "accept": false, - "reason": "Flash-attention sinkless-model crash vs Gemma4 shared_kv_states/FSDP2 KeyError are different code paths." + "reason": "save/from_pretrained extra_state handling is a serialization bug, not an mmap/OOM memory issue." }, { "left": "issue:45684", "right": "issue:45698", "accept": false, - "reason": "Custom-model save permission propagation and wrong-module loading are different serialization bugs." + "reason": "Read-only permission propagation during save_pretrained and wrong custom module loading after save_pretrained are different failure modes." }, { "left": "issue:41720", "right": "issue:44155", "accept": false, - "reason": "Qwen3 device-mapping CUDA assert and AudioFlamingo3 batch leakage are unrelated." + "reason": "CUDA assert on Qwen3 auto device mapping is unrelated to AudioFlamingo3 token/embedding leakage between tracks." }, { "left": "issue:44479", "right": "issue:44877", "accept": false, - "reason": "Video-input regression for Qwen models vs strict config preventing granite_speech loading." + "reason": "Video-input regression across Qwen models and strict granite_speech config loading are not the same bug." + }, + { + "left": "issue:45588", + "right": "issue:45663", + "accept": false, + "reason": "flash_attention sink-less crash and Gemma4 FSDP2 KeyError are different code paths." }, { "left": "issue:43957", "right": "issue:44756", "accept": false, - "reason": "Meta-device loading regressions and Strix Halo mmap OOM are distinct issues." + "reason": "Meta-device loading breakage across some models is unrelated to Strix Halo mmap OOM." }, { "left": "issue:43646", "right": "issue:43708", "accept": false, - "reason": "Custom model initialization breakage and Trainer max_steps resume miscalculation are unrelated." + "reason": "Custom model initialization in v5 and resume_from_checkpoint max_steps calculation are separate regressions." + }, + { + "left": "issue:45812", + "right": "issue:46315", + "accept": false, + "reason": "Both involve tokenization changes, but one is wrong token IDs for Granite models and the other is a slow convert_tokens_to_ids regression; not the same bug." }, { "left": "issue:45292", "right": "issue:45335", - "accept": false, - "reason": "Similar API area, but different manifestations across different model classes; not enough evidence for the same bug." + "accept": true, + "reason": "Same underlying resize_token_embeddings propagation bug: 45335 is the t5gemma-specific case of output_embeddings not being updated." }, { "left": "issue:43645", "right": "issue:44164", "accept": false, - "reason": "Notebook/custom-model init regression is unrelated to extra_state save/load handling." + "reason": "Notebook custom-model initialization breakage is unrelated to extra_state save/from_pretrained handling." }, { "left": "issue:44402", "right": "issue:44451", "accept": false, - "reason": "Tokenizer vocab mismatch for special tokens and a ScandiBERT load failure are not the same concrete bug." + "reason": "Tokenizer vocab-size mismatch on one model and inability to load ScandiBERT are different load-time issues." }, { "left": "issue:43828", "right": "issue:43957", "accept": false, - "reason": "Autocast dtype mismatch and meta-device loading failure are different problems." + "reason": "Autocast dtype mismatch on Phi-tiny-MoE and meta-device loading failures are distinct model/runtime bugs." }, { "left": "issue:43425", "right": "issue:43828", "accept": false, - "reason": "Torch version incompatibility and autocast dtype mismatch are separate failure modes." + "reason": "Torch 2.10 incompatibility is a version-compatibility issue, not the same as the Phi autocast dtype mismatch." }, { "left": "issue:44589", "right": "issue:44756", "accept": false, - "reason": "Float8 storage lookup error is unrelated to mmap/OOM on Strix Halo." + "reason": "Float8 storage lookup failure and Strix Halo OOM are different runtime problems." }, { "left": "issue:43257", "right": "issue:44863", "accept": false, - "reason": "Qwen3 MOE conversion/loading issue and NemotronH checkpoint loading failure are different model-specific bugs." + "reason": "Qwen3 MOE weights not converted under accelerate+deepspeed is unrelated to NemotronH checkpoint loading." }, { "left": "issue:44683", "right": "issue:45823", "accept": false, - "reason": "Compiled flex_attention breakage on torch>=2.9 is unrelated to Gemma4 device-mapping mismatch." + "reason": "Compiled flex_attention failing on torch>=2.9 is unrelated to Gemma4 device mismatch during forward." }, { "left": "issue:39404", "right": "issue:45701", "accept": false, - "reason": "Whisper pipeline return_language regression and tokenizer-version drift are different issues." + "reason": "Whisper pipeline return_language regression and generic tokenizer-version regression are not the same bug." }, { "left": "issue:43611", "right": "issue:43645", "accept": false, - "reason": "base_model_prefix loading breakage is distinct from the notebook custom-model initialization regression." + "reason": "base_model_prefix loading regression and notebook custom model initialization regression are distinct issues." }, { "left": "issue:43425", - "right": "issue:43716", - "accept": false, - "reason": "Torch 2.10 incompatibility and Mistral-3 dtype mismatch do not share the same underlying bug." - }, - { - "left": "issue:43906", - "right": "issue:44530", + "right": "issue:45823", "accept": false, - "reason": "Isolated reproduction of an old issue and Qwen3.5 PagedAttentionCache group-type crash are not the same bug." + "reason": "Torch 2.10 incompatibility is a broad version issue, not the Gemma4 PLE device-mismatch bug." }, { - "left": "issue:44265", - "right": "issue:44898", + "left": "issue:44530", + "right": "issue:45084", "accept": false, - "reason": "torch.export/torch_compilable_check failure is unrelated to Perceiver resizing/interpolation behavior." + "reason": "PagedAttentionCache linear_attention crash and non-template-node compile error are different failures." }, { - "left": "issue:39404", - "right": "issue:44912", + "left": "issue:43425", + "right": "issue:43716", "accept": false, - "reason": "Whisper pipeline return_language regression and MXFP4 quantization loading behavior are unrelated." + "reason": "Torch 2.10 incompatibility and Mistral-3 image-preprocessor/model dtype mismatch are unrelated." }, { - "left": "issue:44479", - "right": "issue:44805", + "left": "issue:43906", + "right": "issue:44530", "accept": false, - "reason": "Video-input regression and mask-shape IndexError are different failures." + "reason": "An isolated reproduction of another issue is not the same bug as the Qwen3.5 PagedAttentionCache crash." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary stays grounded in the packet and the soft-edge calls are conservative. The only accepted pair, issue:45276 vs issue:45292, is plausibly the same resize_token_embeddings regression with one report being a narrower symptom of the broader Gemma4 issue. No other pair is overstated." + "feedback": "The summary stays grounded in the packet: 45701 is a reasonable broad representative given its open state, multiple inbound references, and generic tokenization-regression framing. The soft-edge calls are conservative overall, and the only accepted cluster is the resize_token_embeddings subfamily, which is supported by the issue titles. No PR-mergeability concerns apply here." }, "evaluator_used": true, "retried": false, @@ -2621,13 +2587,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 8, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:32090", "issue:34567", @@ -2647,7 +2613,6 @@ "issue:43632", "issue:43643", "issue:43644", - "issue:43646", "issue:43716", "issue:43856", "issue:43901", @@ -2655,7 +2620,7 @@ "issue:43994", "issue:44079", "issue:44164", - "issue:44451", + "issue:44265", "issue:44479", "issue:44488", "issue:44509", @@ -2666,7 +2631,10 @@ "issue:44683", "issue:44737", "issue:44756", + "issue:44805", "issue:44811", + "issue:44898", + "issue:44912", "issue:44918", "issue:44987", "issue:44991", @@ -2674,249 +2642,255 @@ "issue:45084", "issue:45584", "issue:45759", - "issue:45859", - "issue:46304" + "issue:45859" ], "soft_pairs": [ + "issue:44265|issue:44898", + "issue:39404|issue:44912", + "issue:44479|issue:44805", "issue:43901|issue:44509", "issue:44589|issue:45084", "issue:43425|issue:44918", - "issue:36032|issue:43452", "issue:44164|issue:45759", + "issue:36032|issue:43452", "issue:43299|issue:43931", "issue:43596|issue:44683", - "issue:44991|issue:45081", "issue:43994|issue:44079", - "issue:44671|issue:44991", "issue:43856|issue:44756", + "issue:44671|issue:44991", + "issue:44991|issue:45081", "issue:39404|issue:44671", "issue:44737|issue:45859", "issue:44164|issue:44479", "issue:44488|issue:45081", "issue:43582|issue:43632", - "issue:44661|issue:44683", "issue:43611|issue:43644", + "issue:44661|issue:44683", "issue:40990|issue:43421", "issue:32090|issue:39404", "issue:43716|issue:45084", "issue:44488|issue:44987", "issue:43643|issue:44530", - "issue:36032|issue:42915", "issue:34567|issue:40444", + "issue:36032|issue:42915", "issue:44811|issue:45584", - "issue:44530|issue:46304", - "issue:43541|issue:45084", - "issue:44451|issue:44987", - "issue:43644|issue:43646" + "issue:42915|issue:44530", + "issue:43541|issue:45084" ], - "prepared_review_unit_hash": "0d966ea9026f716fdcd339c6b642fecfdbc6c7496d3cbcc535686041fe7b8a86", + "prepared_review_unit_hash": "f5318ceea4b8a1c74b81cfefd260fed8508972aaf2364a1dbdb9260f5da68cc2", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 30, - "serialized_chars": 36034, - "estimated_input_tokens": 9009, - "estimated_eval_tokens": 18274 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37062, + "estimated_input_tokens": 9266, + "estimated_eval_tokens": 18788 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is thematically similar but not duplicate: it mixes many unrelated Transformers v5 regressions across tokenizers, model loading, training, multimodal paths, docs, and internal runtime behavior.", - "confidence": 0.96, - "canonical_issue_reason": null, + "summary": "The cluster is a loose collection of unrelated Transformers regressions, mostly around v5 loading/tokenizer/model behavior. None of the soft pairs look like the same concrete bug or change, so all soft edges are rejected.", + "confidence": 0.37, + "canonical_issue_reason": "Issue 43611 is the broadest and most representative v5 regression in this set: a generic model-loading breakage. The other issues are narrower, model-specific, tokenizer-specific, or unrelated performance/runtime failures.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a safe global representative; the nearest broad regression issue is still too specific to model initialization and does not cover the rest of the cluster.", + "best_issue_reason": "43611 is the strongest global issue candidate because it is phrased broadly and matches the dominant theme of v5 compatibility breakages without being overly narrow to one model family.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44265", + "right": "issue:44898", + "accept": false, + "reason": "Both are model/runtime failures, but they involve different APIs and likely different code paths; not the same underlying bug." + }, + { + "left": "issue:39404", + "right": "issue:44912", + "accept": false, + "reason": "Whisper pipeline language return vs MXFP4 quantization fallback are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:44805", + "accept": false, + "reason": "Video-input regression and mask-shape IndexError do not describe the same failure mode." + }, { "left": "issue:43901", "right": "issue:44509", "accept": false, - "reason": "Both are docs-related, but one is about stale return_all_scores wording and the other about removed pipeline task docs; not the same bug." + "reason": "Both are docs-related, but they concern different removed/incorrectly documented pipeline behavior; not one bug." }, { "left": "issue:44589", "right": "issue:45084", "accept": false, - "reason": "Float8 storage lookup failure and template compilation failure are unrelated runtime errors with different code paths." + "reason": "Different TypeErrors in different subsystems; no shared concrete code path." }, { "left": "issue:43425", "right": "issue:44918", "accept": false, - "reason": "Torch version incompatibility and Qwen3.5 embedding unpacking are different failures, not one underlying issue." + "reason": "Torch version incompatibility is a broad environment issue, while the other is a specific SFT trainer embedding unpacking failure." }, { - "left": "issue:36032", - "right": "issue:43452", + "left": "issue:44164", + "right": "issue:45759", "accept": false, - "reason": "Tokenizer special-token name collision and gguf_file loading breakage are distinct tokenizer/loading bugs." + "reason": "Serialization extra_state handling and from_config text_config unwrapping are distinct loading/saving bugs." }, { - "left": "issue:44164", - "right": "issue:45759", + "left": "issue:36032", + "right": "issue:43452", "accept": false, - "reason": "save/from_pretrained extra_state handling and from_config text_config unwrapping affect different serialization paths." + "reason": "Tokenizer special-token method conflict and gguf_file from_pretrained breakage are different tokenizer/model-loading problems." }, { "left": "issue:43299", "right": "issue:43931", "accept": false, - "reason": "Both involve Qwen3-VL loading, but one is a dev-version regression and the other is a weight-shape mismatch; not the same concrete bug." + "reason": "Both involve Qwen3-VL/Qwen3VL loading, but one is a version-specific loading breakage and the other is a weight-shape mismatch on a specific model; not clearly the same bug." }, { "left": "issue:43596", "right": "issue:44683", "accept": false, - "reason": "DeepSpeed zero3 BertModel init and compiled flex_attention on torch>=2.9 are unrelated." + "reason": "DeepSpeed ZeRO-3 init index error and compiled flex_attention torch version failure are unrelated." }, { - "left": "issue:44991", - "right": "issue:45081", + "left": "issue:43994", + "right": "issue:44079", "accept": false, - "reason": "These are different tokenizer/model-loading failures for different models and different internal causes." + "reason": "A nonsensical model/pipeline output report is not the same as a ModelOutput key-assignment bug." }, { - "left": "issue:43994", - "right": "issue:44079", + "left": "issue:43856", + "right": "issue:44756", "accept": false, - "reason": "Siglip2 bad outputs and ModelOutput key assignment with prior None are unrelated behaviors." + "reason": "MoE training memory usage and mmap OOM on Strix Halo are different performance/memory issues." }, { "left": "issue:44671", "right": "issue:44991", "accept": false, - "reason": "CamemBERT masked-LM prediction regression and tokenizer loading failure for a different model are not the same issue." + "reason": "Different model families and different symptoms: incorrect masked LM predictions vs tokenizer loading failure." }, { - "left": "issue:43856", - "right": "issue:44756", + "left": "issue:44991", + "right": "issue:45081", "accept": false, - "reason": "Qwen3 MoE training memory inefficiency and Strix Halo mmap OOM are different memory problems with different fixes." + "reason": "Both are tokenizer-load regressions, but the affected models and failure mechanisms differ." }, { "left": "issue:39404", "right": "issue:44671", "accept": false, - "reason": "Whisper return_language pipeline regression and CamemBERT masked-LM regression are unrelated model-specific bugs." + "reason": "Whisper pipeline behavior and CamemBERT masked-LM predictions are unrelated." }, { "left": "issue:44737", "right": "issue:45859", "accept": false, - "reason": "CPU device placement bug in XLNet relative positions and Qwen3.5 MoE CPU offload incompatibility are not the same code-path problem." + "reason": "CPU device placement bug in XLNet is not the same as Qwen3.5 MoE rotary embedding CPU offload incompatibility." }, { "left": "issue:44164", "right": "issue:44479", "accept": false, - "reason": "Extra_state save/load failure and v5.3 video-input regression are unrelated." + "reason": "save/from_pretrained extra_state handling is unrelated to the Qwen video-input regression." }, { "left": "issue:44488", "right": "issue:45081", "accept": false, - "reason": "Two different tokenizer-loading failures for different models; no shared underlying defect." + "reason": "Two different tokenizer loading issues for different models; no shared concrete fix." }, { "left": "issue:43582", "right": "issue:43632", "accept": false, - "reason": "AppleSilicon allocator warmup TypeError and _is_hf_initialized flag break are separate runtime regressions." + "reason": "Apple Silicon warmup TypeError and `_is_hf_initialized` flag breakage are different regression classes." }, { - "left": "issue:44661", - "right": "issue:44683", + "left": "issue:43611", + "right": "issue:43644", "accept": false, - "reason": "Model-like addition failing inside TOKENIZER_MAPPING_NAMES and compiled flex_attention failures are unrelated." + "reason": "Both are v5 regressions, but one is model loading via base_model_prefix and the other is non-persistent buffer initialization; not the same bug." }, { - "left": "issue:43611", - "right": "issue:43644", + "left": "issue:44661", + "right": "issue:44683", "accept": false, - "reason": "Both are v5 regressions, but one is base_model_prefix loading and the other is non-persistent buffer initialization; not a duplicate." + "reason": "Tokenizer mapping-name handling and flex_attention compilation are unrelated." }, { "left": "issue:40990", "right": "issue:43421", "accept": false, - "reason": "High perplexity on gpt-oss-20b and runtime post-processor updates in TokenizersBackend are unrelated." + "reason": "Perplexity on a specific model and runtime special-token post-processing are not the same issue." }, { "left": "issue:32090", "right": "issue:39404", "accept": false, - "reason": "Trainer GPU broadcast NoneType error and Whisper return_language pipeline regression are different subsystems and bugs." + "reason": "Trainer GPU broadcast NoneType error is unrelated to Whisper return_language behavior." }, { "left": "issue:43716", "right": "issue:45084", "accept": false, - "reason": "Image preprocessor/model dtype mismatch and template compilation failure are unrelated." + "reason": "Image preprocessor/model dtype mismatch and Jinja/template compilation error are unrelated." }, { "left": "issue:44488", "right": "issue:44987", "accept": false, - "reason": "These are distinct model-loading regressions on different models and code paths." + "reason": "Both are model loading failures, but the affected models and symptoms differ." }, { "left": "issue:43643", "right": "issue:44530", "accept": false, - "reason": "AutoConfig trust_remote_code missing fields and PagedAttentionCache invalid group type are unrelated." + "reason": "AutoConfig trust_remote_code missing fields and PagedAttentionCache linear_attention crash are unrelated." }, { - "left": "issue:36032", - "right": "issue:42915", + "left": "issue:34567", + "right": "issue:40444", "accept": false, - "reason": "Tokenizer add_special_tokens conflict and FineGrainedFP8Config Qwen3Moe failure are different bugs." + "reason": "TrainerState token counting and Qwen2.5-VL iterable dataset multi-image finetuning are different bugs." }, { - "left": "issue:34567", - "right": "issue:40444", + "left": "issue:36032", + "right": "issue:42915", "accept": false, - "reason": "TrainerState token counter not updating and Qwen2.5-VL iterable dataset multi-image finetuning failure are unrelated." + "reason": "T5 tokenizer class-method conflict and Qwen3 MoE FineGrainedFP8Config failure are unrelated." }, { "left": "issue:44811", "right": "issue:45584", "accept": false, - "reason": "Both are Whisper issues, but one is batch_decode skip_special_tokens handling and the other is empty-transcription generation after align_special_tokens; not the same bug." + "reason": "Whisper skip_special_tokens handling and empty-transcription generation after align_special_tokens are related by theme but not clearly the same concrete bug." }, { - "left": "issue:44530", - "right": "issue:46304", + "left": "issue:42915", + "right": "issue:44530", "accept": false, - "reason": "Qwen3.5 PagedAttentionCache linear_attention crash and USE_DISTRIBUTED=0 support break are unrelated." + "reason": "Different Qwen-related failures with different triggers: FP8 config vs cache group type error." }, { "left": "issue:43541", "right": "issue:45084", "accept": false, - "reason": "Mixtral grouped_mm tracing failure and template compilation error are different runtime problems." - }, - { - "left": "issue:44451", - "right": "issue:44987", - "accept": false, - "reason": "Loading vesteinn/ScandiBERT and loading physical-intelligence/fast are separate model-loading failures." - }, - { - "left": "issue:43644", - "right": "issue:43646", - "accept": false, - "reason": "Both are v5-related regressions, but one concerns non-persistent buffers and the other custom model initialization; too broad to merge as duplicates." + "reason": "Torch dynamo tracing grouped_mm error and template-node compilation failure are unrelated code paths." } ] }, @@ -2932,13 +2906,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 9, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:36032", "issue:42175", @@ -2952,8 +2926,8 @@ "issue:43452", "issue:43482", "issue:43540", - "issue:43611", "issue:43632", + "issue:43644", "issue:43645", "issue:43646", "issue:43742", @@ -2986,14 +2960,17 @@ "issue:45357", "issue:45507", "issue:45987", - "issue:46178" + "issue:46178", + "issue:46315" ], "soft_pairs": [ + "issue:44451|issue:44987", + "issue:43644|issue:43646", "issue:44206|issue:44220", "issue:44464|issue:45987", "issue:43452|issue:43482", - "issue:43883|issue:44488", "issue:43317|issue:43482", + "issue:43883|issue:44488", "issue:44849|issue:45325", "issue:36032|issue:43645", "issue:45084|issue:45507", @@ -3003,14 +2980,15 @@ "issue:43646|issue:44534", "issue:43089|issue:45987", "issue:43404|issue:44466", - "issue:43883|issue:44991", "issue:43322|issue:43540", "issue:43994|issue:44610", "issue:44403|issue:45081", + "issue:43883|issue:44991", "issue:42915|issue:45084", - "issue:43645|issue:44534", "issue:43377|issue:43819", + "issue:43645|issue:44534", "issue:44265|issue:44610", + "issue:44991|issue:46315", "issue:44451|issue:45081", "issue:44987|issue:45081", "issue:45216|issue:45357", @@ -3019,236 +2997,253 @@ "issue:42915|issue:44291", "issue:43632|issue:44756", "issue:42915|issue:44589", - "issue:44062|issue:44987", "issue:43404|issue:45127", - "issue:43611|issue:44534" + "issue:44062|issue:44987" ], - "prepared_review_unit_hash": "83919fdb0df6aa307b336be5586a9c78a712e6a1bb100f07f083baf58649cfee", + "prepared_review_unit_hash": "9beae67f2838759d00ffa982ca734055f34c3982f694799ef94c9d05f40661cf", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 33, - "serialized_chars": 37081, - "estimated_input_tokens": 9271, - "estimated_eval_tokens": 18798 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38263, + "estimated_input_tokens": 9566, + "estimated_eval_tokens": 19388 }, "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 v5 regressions. Only two soft edges look like true duplicates: the GGUF loading failures and the Qwen3.5 save_pretrained regressions.", - "confidence": 0.67, - "canonical_issue_reason": "issue:43646 is the clearest broad representative: it captures the common v5 regression theme around broken custom-model initialization without being overly model-specific.", + "summary": "This cluster is mostly heterogeneous v5/model-loading regressions that are not true duplicates. The only clear duplicate-looking pair is the Qwen3.5 save_pretrained regression (45216/45357).", + "confidence": 0.86, + "canonical_issue_reason": "Pick issue 45216 as the canonical issue: it is the clearest, broadest description of the underlying Qwen3.5 save_pretrained checkpoint regression, and 45357 reads like a later/finer-grained follow-up on the same bug.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43646 is the best standalone issue in the set because it is concise, broad, and directly describes a core Transformers v5 breakage affecting model initialization.", + "best_issue_reason": "Issue 45216 is the best representative issue in the cluster because it states the regression window and the concrete failure mode most generally, making it the most useful triage target for the duplicate pocket.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44451", + "right": "issue:44987", + "accept": false, + "reason": "Both are load failures on v5, but they name different models and there is no clear shared concrete bug path." + }, + { + "left": "issue:43644", + "right": "issue:43646", + "accept": false, + "reason": "Related to custom-model initialization, but one is about non-persistent buffers and the other about notebook/custom init behavior; too different to merge." + }, { "left": "issue:44206", "right": "issue:44220", "accept": false, - "reason": "Both are audio-feature-extraction bugs, but the titles point to different failure surfaces; not enough evidence they are the same concrete code-path defect." + "reason": "Both are audio-feature bugs, but the titles point to different failing entry points and symptoms." }, { "left": "issue:44464", "right": "issue:45987", "accept": false, - "reason": "Chunked generation with compiled forward is a different problem from StaticCache.get_seq_length returning a tensor." + "reason": "Both touch generation/cache behavior, but one is about chunked generation with compiled forward and the other is a StaticCache length contract issue." }, { "left": "issue:43452", "right": "issue:43482", - "accept": true, - "reason": "Both describe GGUF/model loading breakage in Transformers v5 and plausibly stem from the same loading-path regression." + "accept": false, + "reason": "Both involve GGUF/loading regressions, but one is a broad from_pretrained API break and the other is a model-specific Qwen2.5-GGUF failure." }, { - "left": "issue:43883", - "right": "issue:44488", + "left": "issue:43317", + "right": "issue:43482", "accept": false, - "reason": "One is a missing all_tied_weights_keys attribute in Molmo; the other is a specific model-load failure. Different bugs." + "reason": "Different failure modes: dequantized model device_map offload versus GGUF model loading." }, { - "left": "issue:43317", - "right": "issue:43482", + "left": "issue:43883", + "right": "issue:44488", "accept": false, - "reason": "gpu/cpu offload with dequantized models is unrelated to GGUF loading failures." + "reason": "Different models and different missing attributes/loading symptoms." }, { "left": "issue:44849", "right": "issue:45325", "accept": false, - "reason": "Different Qwen model families and different bugs: hidden-states output vs rope-index scaling." + "reason": "Both are Qwen-family regressions, but they affect different model variants and different code paths." }, { "left": "issue:36032", "right": "issue:43645", "accept": false, - "reason": "T5 tokenizer special-token conflict is unrelated to custom model initialization in notebooks." + "reason": "Tokenizer init conflict versus custom model initialization breakage; not the same bug." }, { "left": "issue:45084", "right": "issue:45507", "accept": false, - "reason": "Different subsystems and failures; no shared concrete code path." + "reason": "Different subsystems and different error types; no shared concrete fix path." }, { "left": "issue:42175", "right": "issue:46178", "accept": false, - "reason": "Both involve backend resolution, but one is an install-extra omission and the other is an unsupported backend mapping entry." + "reason": "Packaging/backend mapping issue versus backend lookup for tensorflow_text; not the same defect." }, { "left": "issue:43262", "right": "issue:43377", "accept": false, - "reason": "Audio chat-template sampling-rate defaulting and MIMI padding-mask batching are different bugs." + "reason": "Audio chat-template sampling-rate bug versus MIMI batching/padding-mask discrepancy." }, { "left": "issue:44987", "right": "issue:44991", "accept": false, - "reason": "Both are model/tokenizer load regressions, but they affect different models and likely different loader paths." + "reason": "Both are v5 load regressions, but they target different models and likely different underlying tokenizer/model problems." }, { "left": "issue:43646", "right": "issue:44534", "accept": false, - "reason": "Custom model initialization and non-persistent buffer serialization are unrelated v5 regressions." + "reason": "Both mention custom-model behavior in v5, but one is initialization and the other is buffer serialization/junk; not the same bug." }, { "left": "issue:43089", "right": "issue:45987", "accept": false, - "reason": "Generation performance overhead is unrelated to cache sequence-length typing." + "reason": "Generation performance overhead is unrelated to the StaticCache return-type issue." }, { "left": "issue:43404", "right": "issue:44466", "accept": false, - "reason": "Both mention tied weights, but one is a missing tie and the other is device-dependent serialization; not the same bug." + "reason": "Weight tying in a multimodal model versus inconsistent lm_head serialization; related theme, different concrete bugs." }, { "left": "issue:43322", "right": "issue:43540", "accept": false, - "reason": "Segfault on Llava Next loading and video-input processing in Qwen3OmniMoe are different issues." + "reason": "Llava Next loading crash and Qwen3OmniMoe video-processing ValueError are unrelated." }, { "left": "issue:43994", "right": "issue:44610", "accept": false, - "reason": "Nonsensical model outputs and wrong processor image size are different problems." + "reason": "Different models and different incorrect-output mechanisms." }, { "left": "issue:44403", "right": "issue:45081", "accept": false, - "reason": "Noise during loading and Mistral regex patch crashes are not the same defect." + "reason": "Noise/logging on load versus a tokenizer regex patch crash; not the same code-path problem." }, { - "left": "issue:42915", - "right": "issue:45084", + "left": "issue:43883", + "right": "issue:44991", "accept": false, - "reason": "Qwen3Moe FP8 config failure is unrelated to template-node compilation errors." + "reason": "Different model families and different loading failures." }, { - "left": "issue:43645", - "right": "issue:44534", + "left": "issue:42915", + "right": "issue:45084", "accept": false, - "reason": "Notebook/custom-model initialization and buffer serialization are different regressions." + "reason": "FineGrainedFP8Config loading failure versus template-node compilation failure; unrelated." }, { "left": "issue:43377", "right": "issue:43819", "accept": false, - "reason": "MIMI batching/padding-mask mismatch and DAC from_latents/STE mismatch are unrelated." + "reason": "Both are model path discrepancies, but they concern different models and different missing-feature causes." + }, + { + "left": "issue:43645", + "right": "issue:44534", + "accept": false, + "reason": "Custom-model notebook init breakage and non-persistent buffer junk are different regression types." }, { "left": "issue:44265", "right": "issue:44610", "accept": false, - "reason": "torch.export with torch_compilable_check is unrelated to OmDet-Turbo processor sizing." + "reason": "torch.export/exportability issue versus processor input-size mismatch; not the same bug." + }, + { + "left": "issue:44991", + "right": "issue:46315", + "accept": false, + "reason": "Tokenizer loading regression versus tokenizer speed regression; different defects." }, { "left": "issue:44451", "right": "issue:45081", "accept": false, - "reason": "Model-load failure for ScandiBERT is unrelated to Mistral regex patching." + "reason": "Model load failure versus Mistral regex patch crash; distinct code paths." }, { "left": "issue:44987", "right": "issue:45081", "accept": false, - "reason": "Different failure modes and different model families." + "reason": "Different affected models and different failure sites." }, { "left": "issue:45216", "right": "issue:45357", "accept": true, - "reason": "Both are Qwen3.5 save_pretrained regressions about incorrect checkpoint serialization, likely the same underlying bug." + "reason": "Same underlying Qwen3.5 save_pretrained checkpoint serialization regression; 45357 is a later manifestation about incorrect visual encoder keys." }, { "left": "issue:44220", "right": "issue:44683", "accept": false, - "reason": "Audio fbank extraction and compiled flex_attention are unrelated." + "reason": "Audio fbank extraction issue versus compiled flex_attention failure on newer torch; unrelated." }, { "left": "issue:36032", "right": "issue:43742", "accept": false, - "reason": "T5 tokenizer load conflict and MobileLLM key error are different loader issues." + "reason": "Tokenizer special-token conflict versus model loading key error; different bugs." }, { "left": "issue:42915", "right": "issue:44291", "accept": false, - "reason": "FP8 config failure and init_empty_weights/_is_hf_initialized TypeError are different bugs." + "reason": "Both involve loading/init regressions, but they are different failure modes and not a single concrete defect." }, { "left": "issue:43632", "right": "issue:44756", "accept": false, - "reason": "Broken _is_hf_initialized flag and mmap/OOM on Strix Halo are unrelated." + "reason": "Different issues: _is_hf_initialized flag breakage versus mmap/OOM on Strix Halo." }, { "left": "issue:42915", "right": "issue:44589", "accept": false, - "reason": "FP8 config failure and Float8 storage deserialization failure are different issues." - }, - { - "left": "issue:44062", - "right": "issue:44987", - "accept": false, - "reason": "AddedToken special-argument duplication is unrelated to model loading regressions." + "reason": "FineGrainedFP8Config failure versus float8 storage deserialization error; unrelated." }, { "left": "issue:43404", "right": "issue:45127", "accept": false, - "reason": "Both touch tied embeddings, but one is a missing tie and the other is LoRA merge collapse with extended vocab; not clearly the same bug." + "reason": "Different tied-weights/model-collapse regressions with different triggers and surfaces." }, { - "left": "issue:43611", - "right": "issue:44534", + "left": "issue:44062", + "right": "issue:44987", "accept": false, - "reason": "base_model_prefix loading breakage is unrelated to non-persistent buffer junk filling." + "reason": "AddedToken special-argument bug versus a model loading regression; not the same issue." } ] }, "evaluator_result": { "accept": true, - "feedback": "The analysis is grounded in the packet and stays mostly conservative. The two accepted soft edges are reasonably supported by their shared GGUF-loading and Qwen3.5 save_pretrained themes, and the remaining rejections do not overclaim sameness based on the available titles. The cluster summary is also appropriately cautious about the set being mostly unrelated regressions." + "feedback": "The cluster summary is grounded in the packet and stays conservative overall. The Qwen3.5 save_pretrained pair (45216/45357) is the only clearly supported duplicate-looking match from the listed soft pairs, and the other rejections are reasonable given only title-level evidence. One mild caveat: the summary is broad and does not attempt exhaustive recall, but it does not overstate the evidence." }, "evaluator_used": true, "retried": false, @@ -3261,13 +3256,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 10, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:33666", "issue:36032", @@ -3286,6 +3281,7 @@ "issue:43525", "issue:43531", "issue:43582", + "issue:43611", "issue:43632", "issue:43792", "issue:43818", @@ -3318,238 +3314,248 @@ "issue:45923" ], "soft_pairs": [ + "issue:43611|issue:44534", "issue:43531|issue:44877", "issue:43296|issue:43531", "issue:36032|issue:41720", "issue:44479|issue:45325", "issue:43425|issue:44403", "issue:40990|issue:44811", + "issue:33666|issue:40444", "issue:43792|issue:44589", "issue:43421|issue:44568", - "issue:33666|issue:40444", "issue:44403|issue:44991", "issue:43632|issue:44534", "issue:44079|issue:45663", - "issue:44683|issue:45020", "issue:43381|issue:45923", + "issue:44683|issue:45020", "issue:44462|issue:45812", "issue:44829|issue:45479", "issue:40444|issue:45325", "issue:43482|issue:45538", - "issue:44291|issue:44589", "issue:43317|issue:43856", - "issue:41553|issue:43582", + "issue:44291|issue:44589", "issue:43482|issue:43525", "issue:43818|issue:45357", "issue:42175|issue:45542", + "issue:41553|issue:43582", "issue:43377|issue:45491", "issue:41720|issue:45081", "issue:43531|issue:44291", "issue:44265|issue:45072" ], - "prepared_review_unit_hash": "ea582e59583f99458a07f163d3a12f7b33b3d4a8c4237beb80491ae1aa546a6f", + "prepared_review_unit_hash": "02a6615e43e4513f11069898ee06a0b741f13c8d480abc4ebb87b90615d8f2aa", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 28, - "serialized_chars": 35723, - "estimated_input_tokens": 8931, - "estimated_eval_tokens": 18118 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36598, + "estimated_input_tokens": 9150, + "estimated_eval_tokens": 18556 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab-bag of unrelated Transformers/model-specific bug reports. The only common thread is broad regression-style loading/training failures, but the soft pairs mostly differ in model family, failure mode, and code path. No PRs are present.", - "confidence": 0.91, - "canonical_issue_reason": "issue:45020 is the broadest regression-style issue here: it frames a recent-version breakage affecting remote_code models, which is the closest umbrella theme among otherwise disparate model/tokenizer bugs.", + "summary": "This set is mostly heterogeneous rather than one tight duplicate cluster. Only two soft edges look plausibly duplicate-level: one around Qwen-VL temporal/rope indexing, and one around AutoTokenizer/tokenizer.json handling.", + "confidence": 0.61, + "canonical_issue_reason": "No single canonical issue: the items split into several unrelated regressions. If one representative must be chosen, issue:45812 is the strongest because it is broad, concrete, and has inbound references.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the best cluster anchor because it is the most generally phrased and most representative of the shared \u2018recent Transformers version regression\u2019 pattern, whereas the others are narrower, model-specific failures.", + "best_issue_reason": "issue:45812 is the best overall representative among the issues because it describes a clear, user-visible regression with broad impact and linked discussion.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43611", + "right": "issue:44534", + "accept": false, + "reason": "Both mention Transformers v5, but the failures are different: model-loading/base_model_prefix vs non-persistent buffer initialization." + }, { "left": "issue:43531", "right": "issue:44877", "accept": false, - "reason": "Both are model-loading/config issues, but one is a Qwen3-MoE sliding_window bug and the other is a Granite Speech config strictness problem; different models and code paths." + "reason": "Different bugs in different models: Qwen3-MoE sliding-window behavior vs Granite speech config loading." }, { "left": "issue:43296", "right": "issue:43531", "accept": false, - "reason": "PaddleOCR-VL load failure and Qwen3-MoE sliding_window failure are unrelated model-specific bugs." + "reason": "Both are Qwen-family issues, but one is PaddleOCR-VL loading in vLLM and the other is Qwen3-MoE sliding-window behavior; not the same bug." }, { "left": "issue:36032", "right": "issue:41720", "accept": false, - "reason": "T5 tokenizer special-token attribute conflict is unrelated to Qwen3 device-mapping cuda assertions." + "reason": "Tokenizer method conflict vs device-mapping cuda assert are unrelated failure modes." }, { "left": "issue:44479", "right": "issue:45325", - "accept": false, - "reason": "Both involve Qwen VL, but one is a broad video-input regression and the other is a specific still-image temporal position-id scaling bug; not the same failure." + "accept": true, + "reason": "Both describe the same Qwen-VL temporal/rope-indexing regression in v5.3.0, affecting visual/video position IDs in closely related code paths." }, { "left": "issue:43425", "right": "issue:44403", "accept": false, - "reason": "Torch 2.10 compatibility and noisy transformer loading are different issues." + "reason": "Torch 2.10 compatibility and noisy loading output are unrelated issues." }, { "left": "issue:40990", "right": "issue:44811", "accept": false, - "reason": "GPT-OSS perplexity on WikiText and Whisper batch_decode skip_special_tokens handling are unrelated." + "reason": "Perplexity quality regression is unrelated to Whisper batch_decode skip_special_tokens handling." + }, + { + "left": "issue:33666", + "right": "issue:40444", + "accept": false, + "reason": "Both are Qwen2-VL training issues, but the concrete failures differ: multi-GPU training vs IterableDataset with multiple images per prompt." }, { "left": "issue:43792", "right": "issue:44589", "accept": false, - "reason": "Whisper-large-v2 runtime failure and missing Float8 storage object are different errors with no shared code path." + "reason": "Whisper runtime failure and missing Float8 storage object are different code-path failures." }, { "left": "issue:43421", "right": "issue:44568", "accept": false, - "reason": "Both concern tokenizer special tokens, but one is a backend post-processor update bug and the other is a tokenizer-specific BOS/EOS regression; too different to merge." - }, - { - "left": "issue:33666", - "right": "issue:40444", - "accept": false, - "reason": "Both mention Qwen VL training, but multi-GPU training and iterable-dataset multi-image prompt failures are distinct problems." + "reason": "Runtime post-processor updates for modified special tokens is not the same as add_special_tokens failing to add BOS/EOS." }, { "left": "issue:44403", "right": "issue:44991", "accept": false, - "reason": "General loading noise vs a specific tokenizer load failure for EMBEDDIA/est-roberta are different symptoms and fixes." + "reason": "Generic loading noise and a tokenizer-loading regression for est-roberta are not the same underlying bug." }, { "left": "issue:43632", "right": "issue:44534", "accept": false, - "reason": "Both are Transformers v5 regressions, but one is about _is_hf_initialized and the other about non-persistent buffers; different underlying bugs." + "reason": "_is_hf_initialized flag breakage and non-persistent buffer junk are separate Transformers v5 regressions." }, { "left": "issue:44079", "right": "issue:45663", "accept": false, - "reason": "ModelOutput key assignment bug and Gemma-4 FSDP2 shared_kv_states KeyError are unrelated." + "reason": "ModelOutput key assignment and Gemma-4 FSDP2 shared_kv_states rebuilding are unrelated." }, { - "left": "issue:44683", - "right": "issue:45020", + "left": "issue:43381", + "right": "issue:45923", "accept": false, - "reason": "Compiled flex_attention on torch>=2.9 and remote_code breakage in recent Transformers are different compatibility issues." + "reason": "Eval-mode gradient checkpointing behavior is distinct from a missing supports_gradient_checkpointing flag on a remote-code model." }, { - "left": "issue:43381", - "right": "issue:45923", + "left": "issue:44683", + "right": "issue:45020", "accept": false, - "reason": "Gradient checkpointing in eval mode and missing supports_gradient_checkpointing flag are related only thematically; they are not the same bug." + "reason": "Compiled flex_attention on torch>=2.9 and remote_code model breakage are different problems." }, { "left": "issue:44462", "right": "issue:45812", - "accept": false, - "reason": "Tokenizer.json being ignored and Granite token IDs being wrong are both tokenizer issues, but they point to different loading/processing failures." + "accept": true, + "reason": "Both point to the same AutoTokenizer/tokenizer.json selection regression, with 45812 being a model-specific manifestation of the broader 44462 bug." }, { "left": "issue:44829", "right": "issue:45479", "accept": false, - "reason": "Both involve degenerate classification behavior, but one is tied to flash_attention_3 and the other to problem_type/num_labels=1 loss logic." + "reason": "Degenerate training with flash_attention_3 and zero-loss single-label classification are different classification-loss bugs." }, { "left": "issue:40444", "right": "issue:45325", "accept": false, - "reason": "Qwen2.5-VL multi-image iterable-dataset training failure and Qwen2.5-VL temporal position-id scaling are different bugs." + "reason": "Qwen2.5-VL iterable-dataset multi-image failure is not the same as the still-image temporal position-id scaling bug." }, { "left": "issue:43482", "right": "issue:45538", "accept": false, - "reason": "GGUF loading failure in Transformers v5 and CLIPTokenizer model_max_length handling are unrelated." - }, - { - "left": "issue:44291", - "right": "issue:44589", - "accept": false, - "reason": "init_empty_weights/_is_hf_initialized TypeError and missing Float8 storage object are distinct runtime issues." + "reason": "GGUF loading failure and CLIPTokenizer max_length behavior are unrelated." }, { "left": "issue:43317", "right": "issue:43856", "accept": false, - "reason": "Device-map offload/load failure and Qwen3 MoE memory inefficiency are different training/inference problems." + "reason": "GPU+CPU offload loading failure and Qwen3 MoE training memory usage are different issues." }, { - "left": "issue:41553", - "right": "issue:43582", + "left": "issue:44291", + "right": "issue:44589", "accept": false, - "reason": "Bad AutoTokenizer error message for Voxtral is unrelated to an Apple Silicon caching_allocator_warmup TypeError." + "reason": "Unexpected _is_hf_initialized argument and missing Float8 storage object are unrelated Transformers v5 breakages." }, { "left": "issue:43482", "right": "issue:43525", "accept": false, - "reason": "Qwen2.5-GGUF loading failure and Llama4Config missing pad_token_id are unrelated model/config issues." + "reason": "Qwen2.5-GGUF loading and Llama4Config missing pad_token_id are different model-specific regressions." }, { "left": "issue:43818", "right": "issue:45357", "accept": false, - "reason": "Video-LLaVA tower/temporal-attention issue and Qwen3.5 save_pretrained key regression are different model families and code paths." + "reason": "Video-LLaVA video_tower/temporal-attention bug and Qwen3.5 save_pretrained key regression do not share the same code path." }, { "left": "issue:42175", "right": "issue:45542", "accept": false, - "reason": "One is missing TensorFlow in a torch extra install, the other is tensorboard-only causing a TF backend error; related packaging area but not the same bug." + "reason": "Both concern TF packaging, but one is missing TensorFlow in the torch extra while the other is a tensorboard-only install causing an undefined backend error." + }, + { + "left": "issue:41553", + "right": "issue:43582", + "accept": false, + "reason": "AutoTokenizer error messaging for Voxtral and Apple Silicon caching_allocator_warmup TypeError are unrelated." }, { "left": "issue:43377", "right": "issue:45491", "accept": false, - "reason": "Both are batching/padding-related model bugs, but they affect different models and mechanisms (padding mask support vs sliding-window all-padding NaNs)." + "reason": "Missing padding-mask support in MIMI encoder and Gemma3 NaN embeddings from sliding-window all-padding windows are distinct bugs." }, { "left": "issue:41720", "right": "issue:45081", "accept": false, - "reason": "Qwen3 auto device mapping cuda assert and Mistral tokenizer regex patch crash are unrelated." + "reason": "Qwen3 auto device mapping cudaErrorAssert and Mistral regex patch crashing are unrelated." }, { "left": "issue:43531", "right": "issue:44291", "accept": false, - "reason": "Sliding_window Qwen3-MoE failure and init_empty_weights/_is_hf_initialized TypeError are different regressions." + "reason": "Qwen3-MoE sliding_window behavior and init_empty_weights/_is_hf_initialized are different regressions." }, { "left": "issue:44265", "right": "issue:45072", "accept": false, - "reason": "torch.export/torch_compilable_check failure and bfloat16 dtype mismatches are unrelated torch compatibility bugs." + "reason": "torch.export with torch_compilable_check and dtype mismatches in bfloat16 inference are separate issues." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall: the cluster is mostly heterogeneous, and the two accepted soft edges are plausibly related duplicate-level regressions based on the titles. The conservative false verdicts are also reasonable. The representative issue choice and summary stay within the packet evidence." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -3558,22 +3564,20 @@ "component_index": 1, "component_count": 18, "review_unit_index": 11, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:35707", "issue:39692", "issue:40444", - "issue:40990", "issue:41720", "issue:42175", "issue:42915", - "issue:43257", "issue:43262", "issue:43296", "issue:43377", @@ -3597,23 +3601,25 @@ "issue:44186", "issue:44265", "issue:44291", + "issue:44387", "issue:44466", "issue:44479", "issue:44488", "issue:44534", "issue:44568", "issue:44610", + "issue:44617", "issue:44877", "issue:44898", "issue:45072", "issue:45325", "issue:45381", - "issue:45584", "issue:45800", "issue:45823", "issue:45834", "issue:45920", - "issue:45923" + "issue:46305", + "issue:46315" ], "soft_pairs": [ "issue:43611|issue:43646", @@ -3631,237 +3637,237 @@ "issue:35707|issue:40444", "issue:43761|issue:44079", "issue:44479|issue:45823", + "issue:44387|issue:44617", "issue:43606|issue:44265", "issue:44291|issue:44877", "issue:43296|issue:45800", - "issue:44568|issue:45920", "issue:43701|issue:45325", + "issue:44568|issue:45920", "issue:43606|issue:43819", "issue:41720|issue:45823", "issue:44186|issue:44265", "issue:43454|issue:43550", "issue:43262|issue:43746", - "issue:43881|issue:44488", + "issue:46305|issue:46315", "issue:39692|issue:40444", + "issue:43881|issue:44488", "issue:43377|issue:44610", - "issue:43454|issue:45072", - "issue:40990|issue:45584", - "issue:43257|issue:45923" + "issue:43454|issue:45072" ], - "prepared_review_unit_hash": "c8e479840f8692cdbb194eba7ae24aa9aac02b19e2e87c03f7b1ae485e8a8452", + "prepared_review_unit_hash": "9f90680cb9dcc4154e73cf6b8cedeaf1b7ed709d9d86e07f0757f60e3e390e0b", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37139, - "estimated_input_tokens": 9285, - "estimated_eval_tokens": 18826 + "serialized_chars": 37162, + "estimated_input_tokens": 9291, + "estimated_eval_tokens": 18838 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose set of unrelated Transformers issues, with a few broad v5 regression themes around model loading, initialization, and tied weights. None of the soft pairs look close enough to merge as true duplicates.", - "confidence": 0.91, - "canonical_issue_reason": "issue:43611 is the broadest and most representative issue in the set: it describes a concrete v5 model-loading regression and overlaps conceptually with several nearby initialization/loading complaints more than the others do.", + "summary": "This cluster is a grab-bag of mostly unrelated Transformers issues, with many superficial similarities around v5 regressions, loading, multimodal inputs, or device/memory errors. None of the soft pairs look like true duplicates of the same concrete bug.", + "confidence": 0.88, + "canonical_issue_reason": "Issue 43646 is the broadest representative of the cluster\u2019s central theme: a Transformers v5 model-initialization regression affecting loading behavior. It is not a duplicate of the others, but it best anchors the shared v5 regression pattern.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43611 is the best single representative issue because it is concrete, well-scoped, and captures one of the recurring regression themes in the cluster without being overly model-specific.", + "best_issue_reason": "Issue 43646 is the most representative issue for the overall cluster because it describes a core v5 initialization breakage, which is the closest common thread among many other regression reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43611", "right": "issue:43646", "accept": false, - "reason": "Both are v5-related loading/init regressions, but one is about base_model_prefix loading and the other about custom model initialization; they are related theme-wise, not the same concrete bug." + "reason": "Both involve v5 loading/init breakage, but one is about `base_model_prefix` and the other about custom model initialization; different failure modes and code paths." }, { "left": "issue:43632", "right": "issue:45072", "accept": false, - "reason": "Different failure modes: _is_hf_initialized/model init regression versus bf16 dtype mismatches in inference." + "reason": "`_is_hf_initialized` flag handling and bfloat16 dtype mismatch are unrelated bugs." }, { "left": "issue:43994", "right": "issue:44898", "accept": false, - "reason": "Both are vision-model correctness issues, but the symptoms and code paths differ: SigLIP2 AutoModel/pipeline output corruption vs Perceiver resolution handling." + "reason": "Different models and different symptoms: SigLIP2 nonsensical outputs vs Perceiver resolution/interpolation failure." }, { "left": "issue:42175", "right": "issue:43296", "accept": false, - "reason": "Completely unrelated: missing TensorFlow backend in an extra install versus PaddleOCR-VL load failure in vLLM." + "reason": "A backend packaging/install issue is unrelated to a PaddleOCR-VL model load failure." }, { "left": "issue:45325", "right": "issue:45834", "accept": false, - "reason": "Different models and bugs: Qwen2.5-VL rope/position-id scaling versus Kosmos2.5 OCR long-input index error." + "reason": "Both mention indexing/positioning, but one is Qwen2.5-VL rope position logic and the other is Kosmos2.5 OCR length handling; not the same bug." }, { "left": "issue:43792", "right": "issue:43957", "accept": false, - "reason": "Whisper inference failure and meta-device loading regressions are not the same underlying bug." + "reason": "Whisper load/runtime failure and meta-device loading regression are different issues." }, { "left": "issue:43454", "right": "issue:44466", "accept": false, - "reason": "Both mention lm_head/tied weights, but one is a broken weight tie causing bad generation while the other is inconsistent serialization depending on device." + "reason": "Both mention tied `lm_head` weights, but one is a model config not tying weights and the other is inconsistent serialization behavior." }, { "left": "issue:44079", "right": "issue:44534", "accept": false, - "reason": "Unrelated issues: ModelOutput key assignment versus non-persistent buffer initialization." + "reason": "ModelOutput key assignment and non-persistent buffer junk are separate internals bugs." }, { "left": "issue:42915", "right": "issue:43482", "accept": false, - "reason": "Different areas: FineGrainedFP8Config support for Qwen3Moe versus GGUF loading failure for Qwen2.5." + "reason": "FineGrainedFP8 failure and GGUF loading failure are unrelated model-loading problems." }, { "left": "issue:40444", "right": "issue:45381", "accept": false, - "reason": "Both involve Qwen2.5-VL, but one is finetuning with multiple images per prompt and the other is video vision_position_ids regression." + "reason": "Both are Qwen2.5-VL related, but one is finetuning with multiple images and the other is video position-id regression; not enough to treat as the same bug." }, { "left": "issue:45325", "right": "issue:45381", "accept": false, - "reason": "Same model family, different bug: still-image temporal position scaling versus video-position-id computation." + "reason": "Both concern Qwen2.5-VL positional indexing, but one targets still-image temporal positions and the other video inputs; related, not the same concrete defect." }, { "left": "issue:43644", "right": "issue:44079", "accept": false, - "reason": "Non-persistent buffer junk initialization is unrelated to ModelOutput key assignment." + "reason": "A junk-buffer serialization regression and a ModelOutput keying bug are different code paths." }, { "left": "issue:35707", "right": "issue:40444", "accept": false, - "reason": "Generation with inputs_embeds/past_key_values is unrelated to Qwen2.5-VL finetuning with multiple images." + "reason": "Progressive generation with `inputs_embeds`/`past_key_values` is unrelated to multimodal finetuning with multiple images." }, { "left": "issue:43761", "right": "issue:44079", "accept": false, - "reason": "CLIPVisionModel hidden_states regression is unrelated to ModelOutput key assignment." + "reason": "CLIPVisionModel hidden_states regression and ModelOutput key assignment are separate issues." }, { "left": "issue:44479", "right": "issue:45823", "accept": false, - "reason": "Different concrete failures: Qwen VL video-input regression versus Gemma4 PLE device mismatch under device_map auto." + "reason": "Video input regression in Qwen family and Gemma4 device mismatch are different bugs." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Both are OOM reports, but one is int4 reserved-memory behavior and the other is a model-specific video OOM." }, { "left": "issue:43606", "right": "issue:44265", "accept": false, - "reason": "CPU-offload device mismatch and torch.export export failures through torch_compilable_check are different problems." + "reason": "CPU offload device mismatch and torch.export export failure with `torch_compilable_check` are unrelated." }, { "left": "issue:44291", "right": "issue:44877", "accept": false, - "reason": "init_empty_weights/_is_hf_initialized argument error is distinct from strict config validation for granite_speech." + "reason": "Unexpected `_is_hf_initialized` argument and strict config rejection are different loading regressions." }, { "left": "issue:43296", "right": "issue:45800", "accept": false, - "reason": "PaddleOCR-VL load failure in vLLM is unrelated to torch/Transformers version incompatibility." + "reason": "A specific model load failure and a broad torch/Transformers compatibility issue are not the same bug." }, { - "left": "issue:44568", - "right": "issue:45920", + "left": "issue:43701", + "right": "issue:45325", "accept": false, - "reason": "Tokenizer special-token regression for mdeberta-v3-base is not the same as broader AutoTokenizer wrong-token-ID regressions across multiple models." + "reason": "Checkpoint key mismatch and Qwen2.5-VL rope-index regression are unrelated." }, { - "left": "issue:43701", - "right": "issue:45325", + "left": "issue:44568", + "right": "issue:45920", "accept": false, - "reason": "Checkpoint resume key mismatch is unrelated to Qwen2.5-VL rope index scaling." + "reason": "Tokenizer special-token omission and wrong token IDs across models are tokenizer regressions, but not the same concrete defect." }, { "left": "issue:43606", "right": "issue:43819", "accept": false, - "reason": "Both mention inference mismatches, but CPU offload device placement and DAC.from_latents vs forward STE mismatch are different bugs." + "reason": "CPU offload device mismatch is unrelated to DAC latents/STE forward mismatch." }, { "left": "issue:41720", "right": "issue:45823", "accept": false, - "reason": "Qwen3 auto device map CUDA assertion and Gemma4 PLE device mismatch are separate model-specific device-placement bugs." + "reason": "Both are device-related, but one is an auto device-mapping CUDA assert and the other is a PLE forward device mismatch." }, { "left": "issue:44186", "right": "issue:44265", "accept": false, - "reason": "Tokenizer NER/padding crash and torch.export with torch_compilable_check are unrelated code paths." + "reason": "Tokenizer crash on padding/truncation and export failures with `torch_compilable_check` are different subsystems." }, { "left": "issue:43454", "right": "issue:43550", "accept": false, - "reason": "AyaVision weight tying/generation failure is unrelated to Bamba-9B-v2 torch.compile SDPA failure." + "reason": "Weight tying in AyaVisionConfig is unrelated to Bamba-9B-v2 torch.compile/SDPA failure." }, { "left": "issue:43262", "right": "issue:43746", "accept": false, - "reason": "Audio chat-template sample-rate defaulting and PEFT checkpoint loading are unrelated." + "reason": "Audio chat-template sampling-rate default and PEFT adapter checkpoint loading are unrelated." }, { - "left": "issue:43881", - "right": "issue:44488", + "left": "issue:46305", + "right": "issue:46315", "accept": false, - "reason": "Two different model-loading failures with no clear shared root cause." + "reason": "Sliding-window mask off-by-one and slow `convert_tokens_to_ids` are separate regressions." }, { "left": "issue:39692", "right": "issue:40444", "accept": false, - "reason": "Documentation example errors for SigLIP2 are not the same bug as Qwen2.5-VL finetuning with multiple images." + "reason": "Documentation/example errors for SigLIP2 are unrelated to Qwen2.5-VL finetuning failures." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "Different models and different load failures; no shared concrete bug." }, { "left": "issue:43377", "right": "issue:44610", "accept": false, - "reason": "MIMI padding-mask batch inconsistency and OmDet-Turbo processor/model resolution mismatch are different bugs." + "reason": "Padding-mask batch discrepancy in MIMI and OmDet-Turbo processor/model size mismatch are unrelated." }, { "left": "issue:43454", "right": "issue:45072", "accept": false, - "reason": "Tied lm_head weights causing bad generation is unrelated to bf16 dtype mismatch issues." - }, - { - "left": "issue:40990", - "right": "issue:45584", - "accept": false, - "reason": "High perplexity on gpt-oss and Whisper empty-transcription generation failure are unrelated." - }, - { - "left": "issue:43257", - "right": "issue:45923", - "accept": false, - "reason": "Qwen3 MOE weight conversion under accelerate+deepspeed and missing supports_gradient_checkpointing on a trust_remote_code model are different issues." + "reason": "lm_head weight tying/serialization and bfloat16 dtype mismatch are different problems." } ] }, @@ -3877,13 +3883,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 12, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:34634", "issue:35707", @@ -3894,6 +3900,7 @@ "issue:42617", "issue:42915", "issue:42947", + "issue:43257", "issue:43317", "issue:43377", "issue:43404", @@ -3903,10 +3910,8 @@ "issue:43550", "issue:43572", "issue:43577", - "issue:43606", "issue:43708", "issue:43720", - "issue:43749", "issue:43873", "issue:43881", "issue:43883", @@ -3915,7 +3920,6 @@ "issue:44387", "issue:44423", "issue:44466", - "issue:44534", "issue:44610", "issue:44756", "issue:44877", @@ -3923,270 +3927,265 @@ "issue:44977", "issue:44991", "issue:45072", + "issue:45325", "issue:45357", "issue:45406", - "issue:45632", - "issue:45684", + "issue:45584", "issue:45701", "issue:45704", "issue:45759", "issue:45823", - "issue:46291" + "issue:45923", + "issue:46291", + "issue:46315" ], "soft_pairs": [ + "issue:40990|issue:45584", + "issue:43257|issue:45923", "issue:43317|issue:43873", "issue:43572|issue:45759", "issue:40990|issue:42947", "issue:43377|issue:44898", "issue:43873|issue:45701", "issue:43454|issue:43720", + "issue:45325|issue:46315", "issue:41720|issue:44977", "issue:44466|issue:45357", "issue:44991|issue:46291", "issue:43526|issue:43883", - "issue:43881|issue:44991", "issue:42617|issue:42915", "issue:35707|issue:41720", "issue:34634|issue:35707", + "issue:43881|issue:44991", "issue:43577|issue:43708", "issue:44008|issue:45823", - "issue:43450|issue:45701", "issue:43404|issue:45357", + "issue:43450|issue:45701", "issue:44186|issue:45072", "issue:44877|issue:44991", "issue:43550|issue:43720", "issue:36331|issue:42915", - "issue:36331|issue:42617", "issue:44423|issue:45406", + "issue:36331|issue:42617", "issue:42175|issue:46291", - "issue:44387|issue:45704", "issue:43873|issue:44756", - "issue:44610|issue:44898", - "issue:43550|issue:43606", - "issue:43749|issue:44534", - "issue:44186|issue:44610", - "issue:45632|issue:45684" + "issue:44387|issue:45704", + "issue:44610|issue:44898" ], - "prepared_review_unit_hash": "33054856126c2f2a82126e9db8c4e3596ca79469e4ef759093c0a01a4a44c0a9", + "prepared_review_unit_hash": "bbd1be4a3cecee02f4664352c8a67aa5e5df8f450c9707ee354c0810e48a35be", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 32, - "serialized_chars": 36883, - "estimated_input_tokens": 9221, - "estimated_eval_tokens": 18698 + "soft_pair_count": 31, + "serialized_chars": 36692, + "estimated_input_tokens": 9173, + "estimated_eval_tokens": 18602 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated Transformer bugs with a few broad thematic overlaps (offloading/loading, tokenization, multimodal processors). I would not merge them as duplicates; only a couple look superficially similar, but the underlying failures differ.", - "confidence": 0.79, - "canonical_issue_reason": "If forced to pick one representative, issue 43873 is the broadest and most connected item in the offloading/quantization area, with multiple soft links and inbound references. It is still only a theme hub, not a true duplicate target.", + "summary": "All soft edges are false positives: the titles often overlap on broad themes like quantization, offloading, multimodal serving, or save/load behavior, but each report points to a different concrete bug and code path.", + "confidence": 0.74, + "canonical_issue_reason": "No single issue is a clear canonical for this set; if forced, issue 43873 is the broadest open report with the most discussion/inbound references, but it is still distinct from most others.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43873 is the best single issue to anchor the cluster because it has the strongest cross-links and the most general failure description among the offloading-related reports.", + "best_issue_reason": "Issue 43873 is the best representative only in a loose sense because it is open, broadly scoped, and has the most cross-references; however, the cluster is too heterogeneous for a true umbrella issue.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:40990", + "right": "issue:45584", + "accept": false, + "reason": "Different symptoms and models: perplexity on gpt-oss-20b vs Whisper empty-transcription generation after align_special_tokens." + }, + { + "left": "issue:43257", + "right": "issue:45923", + "accept": false, + "reason": "One is MoE weight conversion during accelerate+deepspeed loading; the other is a missing gradient-checkpointing flag in a trust_remote_code variant." + }, { "left": "issue:43317", "right": "issue:43873", "accept": false, - "reason": "Both involve offloading/loading, but one is a dequantized-model load failure with device_map=auto and the other is a broader quantization offloading issue; not the same concrete bug." + "reason": "Both mention offloading/quantization, but one is dequantized model loading with gpu+cpu offload while the other is general offloading behavior under quantization." }, { "left": "issue:43572", "right": "issue:45759", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig is unrelated to unwrapping text_config for composite Qwen multimodal configs." + "reason": "Unrelated config issues: missing pad_token_idx in StableLmConfig vs failing to unwrap text_config for composite Qwen multimodal configs." }, { "left": "issue:40990", "right": "issue:42947", "accept": false, - "reason": "High perplexity on a model is not the same bug as gradient checkpointing not being effective with PEFT LoRA." + "reason": "Perplexity regression on a specific model/data pair is not the same bug as PEFT LoRA gradient checkpointing being ineffective." }, { "left": "issue:43377", "right": "issue:44898", "accept": false, - "reason": "MIMI batched-vs-single output mismatch due to padding masks is unrelated to Perceiver image resolution failures." + "reason": "Both involve batched inputs and shape differences, but one is MIMI padding-mask behavior and the other is Perceiver image-resolution handling." }, { "left": "issue:43873", "right": "issue:45701", "accept": false, - "reason": "Quantization/offloading behavior is unrelated to tokenization changes across transformer versions." + "reason": "Offloading/quantization is unrelated to tokenization behavior changes across versions." }, { "left": "issue:43454", "right": "issue:43720", "accept": false, - "reason": "lm_head weight tying in a multimodal model is a different code path from packed-weight unpacking during accelerate loading." + "reason": "One is missing lm_head tying in a multimodal model; the other is packed-weight unpacking during accelerate loading for BitNet." + }, + { + "left": "issue:45325", + "right": "issue:46315", + "accept": false, + "reason": "Qwen2.5-VL rope-index temporal scaling and slow tokenizer conversion performance are unrelated code paths." }, { "left": "issue:41720", "right": "issue:44977", "accept": false, - "reason": "Qwen3 auto device mapping cuda asserts and Qwen3.5 flash-attention generation failures are different failure modes." + "reason": "Both are Qwen-family failures, but one is auto device mapping cuda assert and the other is flash-attention generation correctness." }, { "left": "issue:44466", "right": "issue:45357", "accept": false, - "reason": "Both touch model save/serialization, but one is about tied lm_head weight serialization and the other about visual encoder keys in Qwen3.5 save_pretrained." + "reason": "Both touch save/load serialization, but one is tied lm_head.weight behavior by device and the other is incorrect visual encoder keys in Qwen3.5." }, { "left": "issue:44991", "right": "issue:46291", "accept": false, - "reason": "Tokenizer loading for a specific model is unrelated to kernels v0.15 requiring version/revision metadata." + "reason": "Tokenizer loading for est-roberta and kernels v0.15 version/revision requirements are separate dependency/config issues." }, { "left": "issue:43526", "right": "issue:43883", "accept": false, - "reason": "BeitImageProcessorFast label reduction and Molmo missing all_tied_weights_keys are unrelated." - }, - { - "left": "issue:43881", - "right": "issue:44991", - "accept": false, - "reason": "A model loading failure is not the same as a tokenizer-loading regression for another model." + "reason": "A label-reduction bug in BeitImageProcessorFast is unrelated to a missing all_tied_weights_keys attribute in MolmoForCausalLM." }, { "left": "issue:42617", "right": "issue:42915", "accept": false, - "reason": "Running 3d_parallel.py and Qwen3Moe FineGrainedFP8Config failures are unrelated." + "reason": "3d_parallel.py startup failure and Qwen3Moe FineGrainedFP8Config failure are different model/runtime problems." }, { "left": "issue:35707", "right": "issue:41720", "accept": false, - "reason": "Progressive generation with inputs_embeds/past_key_values is a different generation-path bug from Qwen3 device-mapping cuda asserts." + "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to Qwen3 auto device mapping cuda asserts." }, { "left": "issue:34634", "right": "issue:35707", "accept": false, - "reason": "BarkProcessor voice_preset not working is unrelated to generation with inputs_embeds and past_key_values." + "reason": "BarkProcessor voice_preset is a processor issue, while the other is generation with inputs_embeds and past_key_values." + }, + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "glm-4v-9b loading failure and tokenizer loading failure for est-roberta are different models and failure modes." }, { "left": "issue:43577", "right": "issue:43708", "accept": false, - "reason": "Blip2 dtype handling and Trainer resume_from_checkpoint max_steps calculation are unrelated training/loading bugs." + "reason": "Model dtype propagation on Blip2 is unrelated to Trainer resume_from_checkpoint max_steps calculation." }, { "left": "issue:44008", "right": "issue:45823", "accept": false, - "reason": "Both are Gemma multimodal issues, but one is a variable-name collision in forward and the other is a device mismatch in PLE; not the same bug." + "reason": "Both are Gemma-family forward issues, but one is a variable-name collision in Gemma3nModel.forward and the other is a device mismatch in Gemma4 PLE with device_map=auto." }, { - "left": "issue:43450", - "right": "issue:45701", + "left": "issue:43404", + "right": "issue:45357", "accept": false, - "reason": "Video processor batch-shape bugs do not match tokenization behavior changes across versions." + "reason": "Missing lm_head tying in Mistral3 differs from incorrect visual encoder keys during Qwen3.5 save_pretrained." }, { - "left": "issue:43404", - "right": "issue:45357", + "left": "issue:43450", + "right": "issue:45701", "accept": false, - "reason": "Tied lm_head weight serialization and incorrect visual encoder keys are different model-specific save_pretrained issues." + "reason": "Video processor output shape mismatch is unrelated to tokenizer behavior changes." }, { "left": "issue:44186", "right": "issue:45072", "accept": false, - "reason": "LayoutLMv2 tokenizer batching issues are unrelated to dtype mismatches in SwitchTransformers/TimmWrapperModel." + "reason": "LayoutLMv2 tokenization crashes on NER/batching are unrelated to dtype mismatches in SwitchTransformers and TimmWrapperModel." }, { "left": "issue:44877", "right": "issue:44991", "accept": false, - "reason": "Strict config blocking granite_speech loading is unrelated to tokenizer loading for EMBEDDIA/est-roberta." + "reason": "Strict config blocking granite_speech loading is a config-class issue, not the same as tokenizer loading for est-roberta." }, { "left": "issue:43550", "right": "issue:43720", "accept": false, - "reason": "Bamba torch.compile SDPA failures and BitNet packed-weight unpacking failures are different code paths and root causes." + "reason": "Torch.compile with SDPA on Bamba-9B-v2 and packed-weight unpacking during accelerate loading are different failure paths." }, { "left": "issue:36331", "right": "issue:42915", "accept": false, - "reason": "Trainer compute_loss argument incompatibility is unrelated to Qwen3Moe FineGrainedFP8Config failures." - }, - { - "left": "issue:36331", - "right": "issue:42617", - "accept": false, - "reason": "Unexpected num_items_in_batch in compute_loss has nothing to do with 3d_parallel.py not running." + "reason": "Unexpected num_items_in_batch in CustomTrainer.compute_loss is unrelated to FineGrainedFP8Config failures in Qwen3Moe." }, { "left": "issue:44423", "right": "issue:45406", "accept": false, - "reason": "Continuous-batching multimodal serve crash due to string handling is different from Gemma4Processor missing _tokenizer." + "reason": "Both hit transformers serve, but one crashes on multimodal Qwen3.5 inputs and the other on Gemma4Processor missing _tokenizer." }, { - "left": "issue:42175", - "right": "issue:46291", + "left": "issue:36331", + "right": "issue:42617", "accept": false, - "reason": "TensorFlow backend packaging and kernels version/revision metadata are unrelated." + "reason": "CustomTrainer compute_loss signature mismatch and inability to run 3d_parallel.py are unrelated problems." }, { - "left": "issue:44387", - "right": "issue:45704", + "left": "issue:42175", + "right": "issue:46291", "accept": false, - "reason": "Int4 quantization memory growth/OOM and an apex RMSNorm memory leak are different memory issues with different causes." + "reason": "TensorFlow backend packaging and kernels version/revision requirements are unrelated dependency issues." }, { "left": "issue:43873", "right": "issue:44756", "accept": false, - "reason": "Both mention loading/offloading and memory pressure, but Strix Halo mmap OOM is a platform-specific fix, not the same as quantization offloading behavior." - }, - { - "left": "issue:44610", - "right": "issue:44898", - "accept": false, - "reason": "Processor output size mismatch and Perceiver non-default-resolution failure are both vision preprocessing issues, but they affect different models and mechanisms." - }, - { - "left": "issue:43550", - "right": "issue:43606", - "accept": false, - "reason": "Bamba torch.compile SDPA and Bark-small CPU offload device mismatch are unrelated." - }, - { - "left": "issue:43749", - "right": "issue:44534", - "accept": false, - "reason": "FSDP CPU RAM efficient loading broken is not the same as non-persistent buffers being filled with junk." + "reason": "General quantization offloading bugs are not the same as disabling mmap on Strix Halo to avoid OOM." }, { - "left": "issue:44186", - "right": "issue:44610", + "left": "issue:44387", + "right": "issue:45704", "accept": false, - "reason": "LayoutLMv2 tokenizer NER/batch handling and OmDet processor image-size mismatch are different subsystems." + "reason": "CUDA reserved-memory OOM under int4 quantization is unrelated to T5 using apex.FusedRMSNorm and its memory leak." }, { - "left": "issue:45632", - "right": "issue:45684", + "left": "issue:44610", + "right": "issue:44898", "accept": false, - "reason": "trust_remote_code cache-path collisions and save_pretrained permission propagation are both custom-model save/load topics, but they are distinct bugs." + "reason": "Processor/model resolution mismatch in OmDet-Turbo and Perceiver interpolation behavior are different image-preprocessing bugs." } ] }, @@ -4202,19 +4201,17 @@ "component_index": 1, "component_count": 18, "review_unit_index": 13, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:29159", "issue:30064", - "issue:34567", "issue:34634", - "issue:35707", "issue:36010", "issue:36032", "issue:40990", @@ -4225,7 +4222,6 @@ "issue:43316", "issue:43317", "issue:43322", - "issue:43388", "issue:43421", "issue:43425", "issue:43525", @@ -4237,6 +4233,7 @@ "issue:43606", "issue:43653", "issue:43720", + "issue:43749", "issue:43782", "issue:43881", "issue:44186", @@ -4244,14 +4241,17 @@ "issue:44492", "issue:44509", "issue:44530", + "issue:44534", + "issue:44610", "issue:44671", "issue:44743", "issue:44913", - "issue:45059", "issue:45081", "issue:45161", "issue:45406", "issue:45464", + "issue:45632", + "issue:45684", "issue:45701", "issue:45759", "issue:45812", @@ -4259,9 +4259,13 @@ "issue:46291" ], "soft_pairs": [ + "issue:43749|issue:44534", + "issue:43550|issue:43606", + "issue:44186|issue:44610", "issue:43322|issue:46050", - "issue:43295|issue:45701", + "issue:45632|issue:45684", "issue:42915|issue:43317", + "issue:43295|issue:45701", "issue:41720|issue:43322", "issue:40990|issue:45701", "issue:41720|issue:42617", @@ -4279,232 +4283,242 @@ "issue:43526|issue:43881", "issue:29159|issue:36032", "issue:45406|issue:45464", - "issue:43550|issue:43653", "issue:30064|issue:34634", + "issue:43550|issue:43653", "issue:43606|issue:43720", "issue:43421|issue:45701", "issue:44530|issue:46291", "issue:42915|issue:43322", "issue:43606|issue:43653", "issue:43531|issue:45759", - "issue:44492|issue:44509", - "issue:34567|issue:35707", - "issue:43388|issue:45059" + "issue:44492|issue:44509" ], - "prepared_review_unit_hash": "f446bc5262eb4bf413769d9181efc340b6bce67efbaf8d8971284cfa71c0c19e", + "prepared_review_unit_hash": "a5f121fa25e4ec2b23e304644b0c1a2fb911c19f0cfc3777834470f06c5f95e7", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 31, - "serialized_chars": 36309, - "estimated_input_tokens": 9078, - "estimated_eval_tokens": 18412 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "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": "The set is a loose collection of mostly unrelated issues across tokenizers, processors, model loading, distributed inference, and docs. None of the soft pairs look like the same underlying bug/change, so no duplicate merges are justified.", - "confidence": 0.93, - "canonical_issue_reason": "issue:45812 is the strongest single issue candidate: it is concrete, clearly scoped, and describes a broad v4\u2192v5 regression with multiple references and an obvious failure mode.", + "summary": "The set is mostly a collection of superficially similar but distinct bugs. I found no soft edge that looks like the same concrete issue/change, so all pairwise candidates are rejected. The closest anchor is the tokenizer-regression subgroup, with issue 45701 as the broadest representative and issue 45812 as the most actionable regression report.", + "confidence": 0.58, + "canonical_issue_reason": "issue:45701 is the broadest tokenizer-regression report and is the most plausible umbrella anchor for the tokenizer-related subset; the rest are mostly model- or component-specific failures.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45812 is the most actionable and representative issue here because it names the affected component, the regression nature, and the concrete symptom very clearly.", + "best_issue_reason": "issue:45812 is the clearest high-impact regression report, with a specific silent tokenization break across Granite models and strong triage value.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43749", + "right": "issue:44534", + "accept": false, + "reason": "Different failure modes: FSDP CPU RAM efficient loading vs non-persistent buffer initialization; same training/runtime area, not the same bug." + }, + { + "left": "issue:43550", + "right": "issue:43606", + "accept": false, + "reason": "Both are runtime/model-loading failures, but one is torch.compile+SDPA on Bamba and the other is CPU-offload device mismatch on Bark." + }, + { + "left": "issue:44186", + "right": "issue:44610", + "accept": false, + "reason": "Tokenizer NER/padding crash in LayoutLMv2 is unrelated to OmDet-Turbo processor size mismatch." + }, { "left": "issue:43322", "right": "issue:46050", "accept": false, - "reason": "Different failure classes: model loading segfault for Llava Next vs quantization/save-load breakage for sinq models." + "reason": "Segfault while loading Llava Next is a different bug from quantized-save/load corruption in sinq." }, { - "left": "issue:43295", - "right": "issue:45701", + "left": "issue:45632", + "right": "issue:45684", "accept": false, - "reason": "Both mention tokenization, but one is a processor/tokenizer API regression and the other is a broad version-dependent tokenization change; not the same bug." + "reason": "Both involve save/load workflows, but one is a cache-path collision and the other is file permission propagation." }, { "left": "issue:42915", "right": "issue:43317", "accept": false, - "reason": "Different subsystems and symptoms: Qwen3MoE FP8 failure vs device-map/offload loading failure for a dequantized model." + "reason": "Both touch model loading/quantization, but FineGrainedFP8 failure and dequantized gpu+cpu offload failure are different code paths." + }, + { + "left": "issue:43295", + "right": "issue:45701", + "accept": false, + "reason": "Both mention tokenization regressions, but one is about processor.tokenizer/images in v4.57.5 and the other is a broader tokenization change report." }, { "left": "issue:41720", "right": "issue:43322", "accept": false, - "reason": "Unrelated model-specific failures: auto device mapping assert on Qwen3 vs segfault while loading Llava Next." + "reason": "Different model families and symptoms: cudaErrorAssert in Qwen3 auto device mapping vs segfault on Llava Next load." }, { "left": "issue:40990", "right": "issue:45701", "accept": false, - "reason": "High perplexity is a model/eval quality issue; the other is a tokenizer regression report. No shared concrete code path." + "reason": "High perplexity on one model is not the same as a generic tokenization-version complaint, even if tokenization could be a factor." }, { "left": "issue:41720", "right": "issue:42617", "accept": false, - "reason": "One is a Qwen3 device-mapping assert, the other is a generic script execution problem; no overlap in bug cause." + "reason": "Qwen3 auto device-map assert is unrelated to failing to run 3d_parallel.py." }, { "left": "issue:41720", "right": "issue:45161", "accept": false, - "reason": "Both are distributed-model issues, but they involve different models and different breakpoints (auto device map vs tensor-parallel-only on GPT-OSS MoE)." + "reason": "Both are Qwen3-related, but auto device mapping failure and TP-only MoE failure are distinct issues." }, { "left": "issue:36010", "right": "issue:42617", "accept": false, - "reason": "ImportError for GenerationMixin is unrelated to a failure running 3d_parallel.py." + "reason": "ImportError for GenerationMixin and a 3d_parallel.py runtime failure are unrelated." }, { "left": "issue:43425", "right": "issue:46291", "accept": false, - "reason": "A torch version incompatibility and a kernels import requirement are different dependency problems." + "reason": "Torch 2.10 incompatibility is not the same as kernels requiring version/revision metadata." }, { "left": "issue:45081", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but one is a Mistral regex patch crash and the other is wrong token IDs for Granite; different code paths and fixes." + "reason": "Both are tokenizer regressions, but one is a Mistral regex/backend crash and the other is wrong token IDs for Granite models." }, { "left": "issue:43525", "right": "issue:46291", "accept": false, - "reason": "Missing pad_token_id in a config is unrelated to the kernels version/revision import error." + "reason": "Missing pad_token_id in Llama4Config is unrelated to the kernels import/version requirement." }, { "left": "issue:43316", "right": "issue:44913", "accept": false, - "reason": "Different config issues: Gemma3 text-config API discrepancy vs GPTNeoX rotary_pct persistence bug." + "reason": "Both are config/API discrepancies, but they affect different fields and model configs." }, { "left": "issue:43531", "right": "issue:43572", "accept": false, - "reason": "Qwen3-MoE sliding-window behavior and StableLm missing pad_token_idx are unrelated config/model bugs." + "reason": "Sliding-window behavior in Qwen3-MoE and missing pad_token_idx in StableLmConfig are different regressions." }, { "left": "issue:44671", "right": "issue:44743", "accept": false, - "reason": "CamemBERT masked-LM regression and Qwen modular recurrent-state reset are different model behavior issues." + "reason": "CamemBERT masked-LM prediction errors and Qwen3 modular recurrent-state reset are unrelated model behaviors." }, { "left": "issue:43540", "right": "issue:43782", "accept": false, - "reason": "Both involve Qwen multimodal models, but one is video-input processing and the other is weight_only loading." + "reason": "Video-processing ValueError in Qwen3OmniMoe is not the same as Qwen3VL weight_only loading failure." }, { "left": "issue:43421", "right": "issue:45812", "accept": false, - "reason": "Runtime special-token/post-processor synchronization is a different tokenizer bug than Granite token-ID regression." + "reason": "Runtime special-token post-processor update bug is not the same as Granite-wide wrong token IDs on AutoTokenizer." }, { "left": "issue:44186", "right": "issue:44361", "accept": false, - "reason": "Different tokenizer implementations and failure modes: LayoutLMv2 NER/padding crash vs MLuke task AttributeError." + "reason": "Two tokenizer bugs, but different models and failure surfaces: batched padding/truncation vs task-time AttributeError." }, { "left": "issue:43526", "right": "issue:43881", "accept": false, - "reason": "BEiT image-processor label reduction and glm-4v loading failure do not share an underlying cause." + "reason": "A BEiTImageProcessorFast label reduction bug is unrelated to a glm-4v-9b load failure." }, { "left": "issue:29159", "right": "issue:36032", "accept": false, - "reason": "General fast/slow tokenizer inconsistency is not the same as T5Tokenizer\u2019s add_special_tokens conflict on import/load." + "reason": "Slow/fast tokenizer inconsistency is a different issue from T5Tokenizer's add_special_tokens method conflict." }, { "left": "issue:45406", "right": "issue:45464", "accept": false, - "reason": "Gemma4Processor missing _tokenizer and Qwen3.5 streaming inference crash are distinct API paths and symptoms." + "reason": "Gemma4Processor missing _tokenizer in serve and Qwen3.5 streaming inference failures are separate bugs." }, { - "left": "issue:43550", - "right": "issue:43653", + "left": "issue:30064", + "right": "issue:34634", "accept": false, - "reason": "torch.compile/SDPA failure for Bamba is unrelated to BigBirdTokenizer special-token decode behavior." + "reason": "Image processor void-segmentation handling and BarkProcessor voice_preset are unrelated processor bugs." }, { - "left": "issue:30064", - "right": "issue:34634", + "left": "issue:43550", + "right": "issue:43653", "accept": false, - "reason": "Image segmentation-map handling and Bark voice_preset processing are different processor bugs." + "reason": "torch.compile+SDPA on Bamba and BigBird mask-token registration/decode issues do not share the same underlying code path." }, { "left": "issue:43606", "right": "issue:43720", "accept": false, - "reason": "Both are CI/loading issues, but they affect different models and mechanisms: CPU offload mismatch vs packed-weight unpacking." + "reason": "CPU-offload device mismatch for Bark-small and BitNet packed-weight unpacking during accelerate loading are distinct loading bugs." }, { "left": "issue:43421", "right": "issue:45701", "accept": false, - "reason": "Special-token runtime updates and version-driven tokenization changes are not the same concrete bug." + "reason": "Special-token runtime post-processing and broad tokenization-version changes overlap only loosely; not the same concrete bug." }, { "left": "issue:44530", "right": "issue:46291", "accept": false, - "reason": "PagedAttentionCache group-type crash on Qwen3.5 has no overlap with the kernels import requirement issue." + "reason": "PagedAttentionCache linear_attention crash on Qwen3.5 is unrelated to kernels import metadata requirements." }, { "left": "issue:42915", "right": "issue:43322", "accept": false, - "reason": "Different failures on different architectures/models; no shared code-path evidence." + "reason": "FineGrainedFP8Config failure and Llava Next segfault are different model-loading problems." }, { "left": "issue:43606", "right": "issue:43653", "accept": false, - "reason": "CPU offload device mismatch and token registration/decode failure are unrelated." + "reason": "CPU offload device mismatch is not the same as missing BigBird special-token registration." }, { "left": "issue:43531", "right": "issue:45759", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior and AutoModel config unwrapping are different issues." + "reason": "Qwen3-MoE sliding-window behavior and AutoModelForCausalLM from_config text_config unwrapping are unrelated." }, { "left": "issue:44492", "right": "issue:44509", "accept": false, - "reason": "A typo in cache strategy docs/code and outdated pipeline-task docs are separate documentation problems." - }, - { - "left": "issue:34567", - "right": "issue:35707", - "accept": false, - "reason": "TrainerState token-count tracking and progressive generation with inputs_embeds/past_key_values are unrelated." - }, - { - "left": "issue:43388", - "right": "issue:45059", - "accept": false, - "reason": "Metric-gathering label truncation and SAM3 text/bounding-box behavior are different bugs in different components." + "reason": "A cache-strategy typo and outdated pipeline docs are both doc-ish, but not the same underlying issue." } ] }, @@ -4520,15 +4534,16 @@ "component_index": 1, "component_count": 18, "review_unit_index": 14, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:33666", + "issue:34567", "issue:34634", "issue:35707", "issue:39692", @@ -4537,13 +4552,13 @@ "issue:42886", "issue:43329", "issue:43381", + "issue:43388", "issue:43421", "issue:43525", "issue:43531", "issue:43540", "issue:43653", "issue:43746", - "issue:43782", "issue:43844", "issue:43873", "issue:43927", @@ -4555,7 +4570,6 @@ "issue:44545", "issue:44589", "issue:44610", - "issue:44625", "issue:44843", "issue:44898", "issue:44918", @@ -4563,12 +4577,11 @@ "issue:44987", "issue:44991", "issue:45042", + "issue:45059", "issue:45072", "issue:45081", "issue:45356", "issue:45357", - "issue:45381", - "issue:45479", "issue:45588", "issue:45663", "issue:45704", @@ -4578,13 +4591,15 @@ "issue:46294" ], "soft_pairs": [ + "issue:34567|issue:35707", + "issue:43388|issue:45059", "issue:44514|issue:44545", "issue:43873|issue:45704", "issue:42490|issue:43329", "issue:41720|issue:43746", "issue:43421|issue:43927", - "issue:44898|issue:45072", "issue:33666|issue:35707", + "issue:44898|issue:45072", "issue:45759|issue:45859", "issue:44451|issue:44991", "issue:43531|issue:44589", @@ -4592,244 +4607,242 @@ "issue:44368|issue:45357", "issue:44062|issue:44451", "issue:44987|issue:45042", - "issue:43421|issue:43653", "issue:39692|issue:44610", "issue:45704|issue:45923", - "issue:45663|issue:45923", + "issue:43421|issue:43653", "issue:43381|issue:43844", + "issue:45663|issue:45923", "issue:42886|issue:44843", "issue:44918|issue:44936", - "issue:43421|issue:44186", "issue:45704|issue:45859", "issue:34634|issue:41720", + "issue:43421|issue:44186", "issue:43381|issue:45704", "issue:43525|issue:43531", "issue:45081|issue:45356", - "issue:39692|issue:43873", "issue:43540|issue:46294", - "issue:44625|issue:45479", - "issue:43782|issue:45381" + "issue:39692|issue:43873" ], - "prepared_review_unit_hash": "281cc107725bfb8c50ba977d6a57c9ec0769a4d16a89179a640fa6555bcfe63d", + "prepared_review_unit_hash": "13a649a197437fc631bb71fb9dd08134a58ef351b34c859888cd6ae7d8f48d10", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 31, - "serialized_chars": 37486, - "estimated_input_tokens": 9372, - "estimated_eval_tokens": 19000 + "serialized_chars": 36896, + "estimated_input_tokens": 9224, + "estimated_eval_tokens": 18704 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated issues; only one soft pair is a true duplicate. The Qwen2_5_VLProcessor batched apply_chat_template crash appears twice, while the other candidate pairs share only broad subsystem/topic similarity rather than the same bug.", - "confidence": 0.93, - "canonical_issue_reason": "issue:44545 is the clearest representative of the duplicated Qwen2_5_VLProcessor.apply_chat_template crash, with the same failure mode and slightly better linkage/wording.", + "summary": "This cluster is mostly a grab-bag of unrelated issue reports; the only clear duplicate pair is the Qwen2_5_VLProcessor apply_chat_template crash (#44514/#44545).", + "confidence": 0.92, + "canonical_issue_reason": "#44545 is the cleaner canonical record: same bug as #44514, later report, and slightly more referenced.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44545 is the best issue representative because it is the exact bug report for the only confirmed duplicate pair and is the cleaner, more central report.", + "best_issue_reason": "#44545 is the best representative issue in this set because it is the exact duplicate target for the only strong match and has the clearest, most specific problem statement.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:34567", + "right": "issue:35707", + "accept": false, + "reason": "Different subsystems and failures: TrainerState token counting vs progressive generation with inputs_embeds/past_key_values." + }, + { + "left": "issue:43388", + "right": "issue:45059", + "accept": false, + "reason": "Different code paths: metric gathering/label truncation vs SAM3 text+box prompt behavior." + }, { "left": "issue:44514", "right": "issue:44545", "accept": true, - "reason": "Same bug and same API path: Qwen2_5_VLProcessor.apply_chat_template crashing on batched input when padding=False." + "reason": "Same issue title and same crash in Qwen2_5_VLProcessor.apply_chat_template with batched input and padding=False; clear duplicate." }, { "left": "issue:43873", "right": "issue:45704", "accept": false, - "reason": "Different problems: quantization/offloading behavior vs a T5 RMSNorm memory leak." + "reason": "Unrelated bugs: quantization offloading behavior vs T5/apex FusedRMSNorm memory leak." }, { "left": "issue:42490", "right": "issue:43329", "accept": false, - "reason": "Unrelated bugs: trailing-slash local path import warning vs a multimodal token counting video-branch NameError." + "reason": "Local-path import warning and multimodal token video-branch undefined variables are unrelated." }, { "left": "issue:41720", "right": "issue:43746", "accept": false, - "reason": "Different models and failures: Qwen3 auto device mapping cuda assert vs GraniteSpeech PEFT local checkpoint loading." + "reason": "Different failures: Qwen3 device mapping cuda assert vs PEFT local checkpoint loading." }, { "left": "issue:43421", "right": "issue:43927", "accept": false, - "reason": "Different tokenizer/config issues: runtime special-token post-processor updates vs DiaConfig custom token IDs after save/load." + "reason": "Both token-related, but one is runtime post-processor updating and the other is save/load preserving custom token IDs; not the same bug." }, { - "left": "issue:44898", - "right": "issue:45072", + "left": "issue:33666", + "right": "issue:35707", "accept": false, - "reason": "Different code paths: Perceiver image resolution handling vs dtype mismatches in SwitchTransformers/TimmWrapperModel inference." + "reason": "Multi-GPU training question for Qwen2-VL is not the same bug as progressive generation with inputs_embeds/past_key_values." }, { - "left": "issue:33666", - "right": "issue:35707", + "left": "issue:44898", + "right": "issue:45072", "accept": false, - "reason": "Unrelated training/generation issues for different models and mechanisms." + "reason": "Different models and failure modes: Perceiver interpolation vs dtype mismatch in inference." }, { "left": "issue:45759", "right": "issue:45859", "accept": false, - "reason": "Both involve Qwen 3.5 multimodal configs, but one is from_config text_config unwrapping and the other is CPU-offload rotary embedding compatibility; not the same bug." + "reason": "Composite config unwrapping bug vs CPU-offload incompatibility in rotary embedding; not the same code-path problem." }, { "left": "issue:44451", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer-loading failures, but they affect different models and likely different root causes." + "reason": "Both tokenizer-loading issues, but they affect different models and likely different root causes." }, { "left": "issue:43531", "right": "issue:44589", "accept": false, - "reason": "Different failures: Qwen3-MoE sliding_window behavior vs missing Float8 storage type." + "reason": "Qwen3-MoE sliding_window issue is unrelated to Float8 storage lookup failure." }, { "left": "issue:45588", "right": "issue:45759", "accept": false, - "reason": "Different bugs: flash_attention sink handling vs composite config unwrapping in AutoModelForCausalLM.from_config." + "reason": "Flash-attention sink handling crash and Qwen multimodal config unwrapping are unrelated." }, { "left": "issue:44368", "right": "issue:45357", "accept": false, - "reason": "Same broad Qwen family, but one is a warning about tie_word_embeddings while the other is incorrect visual encoder keys on save_pretrained." + "reason": "A warning about tie_word_embeddings during LoRA fine-tuning is unrelated to saving incorrect visual encoder keys." }, { "left": "issue:44062", "right": "issue:44451", "accept": false, - "reason": "Different tokenizer issues: AddedToken keyword conflict vs loading a specific pretrained model." + "reason": "AddedToken keyword conflict and ScandiBERT load failure are different bugs." }, { "left": "issue:44987", "right": "issue:45042", "accept": false, - "reason": "Different subsystems: loading a specific model vs PIL backend image processor torchvision dependency regression." - }, - { - "left": "issue:43421", - "right": "issue:43653", - "accept": false, - "reason": "Different tokenizer bugs: runtime post-processor update vs BigBirdTokenizer mask token registration/empty decode." + "reason": "Model-loading failure and PIL image processor torchvision dependency are unrelated." }, { "left": "issue:39692", "right": "issue:44610", "accept": false, - "reason": "Different documentation/model mismatch issue vs processor/model input-size mismatch." + "reason": "Doc/example mismatch and processor/model resolution mismatch are different image-processing issues." }, { "left": "issue:45704", "right": "issue:45923", "accept": false, - "reason": "Different bugs: apex RMSNorm leak vs missing supports_gradient_checkpointing flag." + "reason": "Apex RMSNorm memory leak and missing supports_gradient_checkpointing flag are unrelated." }, { - "left": "issue:45663", - "right": "issue:45923", + "left": "issue:43421", + "right": "issue:43653", "accept": false, - "reason": "Different failures: Gemma4 FSDP2 shared_kv_states bug vs missing gradient-checkpointing support flag." + "reason": "Runtime post-processor refresh bug and BigBirdTokenizer special-token registration bug are different tokenizer issues." }, { "left": "issue:43381", "right": "issue:43844", "accept": false, - "reason": "Different training-mode behaviors: checkpointing disallowed in eval vs abnormal gradient growth under ZeRO-3." + "reason": "Eval-mode gradient checkpointing restriction and ZeRO-3 gradient explosion are not the same defect." + }, + { + "left": "issue:45663", + "right": "issue:45923", + "accept": false, + "reason": "Gemma-4 FSDP2 KeyError and missing gradient-checkpointing flag on a trust_remote_code variant are unrelated." }, { "left": "issue:42886", "right": "issue:44843", "accept": false, - "reason": "Both hit HF_HUB_OFFLINE, but one is a cache-load failure and the other is a specific Mistral regex/model_info offline regression." + "reason": "Both concern offline/tokenizer loading, but one is cache lookup failure and the other is an unconditional model_info call in a Mistral regex patch; distinct root causes." }, { "left": "issue:44918", "right": "issue:44936", "accept": false, - "reason": "Different trainer-related issues: Qwen3.5 embedding unpacking vs trainer.evaluate() after trainer.train()." - }, - { - "left": "issue:43421", - "right": "issue:44186", - "accept": false, - "reason": "Different tokenizer bugs affecting different models and failure modes." + "reason": "Different trainer failures: SFT unpacking of Qwen3.5 embeddings vs evaluate() after train()." }, { "left": "issue:45704", "right": "issue:45859", "accept": false, - "reason": "Different bugs: T5 memory leak vs Qwen3.5 MoE rotary embedding CPU-offload incompatibility." + "reason": "Same as above: unrelated T5 RMSNorm leak vs Qwen3_5Moe CPU offload incompatibility." }, { "left": "issue:34634", "right": "issue:41720", "accept": false, - "reason": "Different model families and unrelated failures." + "reason": "BarkProcessor voice preset and Qwen3 auto device mapping are unrelated." + }, + { + "left": "issue:43421", + "right": "issue:44186", + "accept": false, + "reason": "Different tokenizer families and different failure modes: post-processor update vs LayoutLMv2 batching/padding crash." }, { "left": "issue:43381", "right": "issue:45704", "accept": false, - "reason": "Gradient checkpointing eval-mode restriction is unrelated to the T5 RMSNorm memory leak." + "reason": "Gradient checkpointing eval restriction is unrelated to T5 apex memory leak." }, { "left": "issue:43525", "right": "issue:43531", "accept": false, - "reason": "Different config/model issues: missing pad_token_id on Llama4Config vs Qwen3-MoE sliding_window behavior." + "reason": "Missing pad_token_id on Llama4Config is unrelated to Qwen3-MoE sliding_window behavior." }, { "left": "issue:45081", "right": "issue:45356", "accept": false, - "reason": "Related tokenizer area, but different concrete failures: backend_tokenizer attribute crash vs codec handling/incorrect warning regression." - }, - { - "left": "issue:39692", - "right": "issue:43873", - "accept": false, - "reason": "Different problems: SigLIP2 docs/model mismatch vs quantization offloading behavior." + "reason": "Both involve Mistral/Kimi tokenizer handling, but one crashes in _patch_mistral_regex while the other is a codec-handling regression with warning text; not clearly the same bug." }, { "left": "issue:43540", "right": "issue:46294", "accept": false, - "reason": "Both concern Qwen3.5 video, but one is a ValueError in Qwen3OmniMoe video processing and the other is chat-template/video-token mismatch in Qwen3.5-9B." - }, - { - "left": "issue:44625", - "right": "issue:45479", - "accept": false, - "reason": "Different classification/config bugs: num_labels propagation vs degenerate zero loss for single-label classification." + "reason": "Both are Qwen video-input issues, but one is a ValueError in Qwen3OmniMoe and the other is a feature/token mismatch in Qwen3.5-9B; likely different root causes." }, { - "left": "issue:43782", - "right": "issue:45381", + "left": "issue:39692", + "right": "issue:43873", "accept": false, - "reason": "Different multimodal model bugs: weight_only loading failure vs incorrect video vision_position_ids." + "reason": "SigLIP2 docs/example errors are unrelated to quantization offloading behavior." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet: the only clearly duplicated pair is issue:44514/issue:44545, whose titles describe the same Qwen2_5_VLProcessor.apply_chat_template batched-padding crash. The remaining pair judgments stay conservative and do not overclaim duplicates. No pull-request mergeability concerns apply here because these are issue pairs, not PR pairs." + "feedback": "The summary is grounded in the packet, and the only accepted pair (#44514/#44545) is strongly supported by the identical titles plus matching crash description. The other soft-edge rejections are conservative and do not overclaim duplicate status. No PR-pair mergeability concerns apply here." }, "evaluator_used": true, "retried": false, @@ -4842,13 +4855,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 15, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:30064", "issue:36010", @@ -4869,6 +4882,7 @@ "issue:43720", "issue:43746", "issue:43756", + "issue:43782", "issue:43819", "issue:43873", "issue:43950", @@ -4878,7 +4892,6 @@ "issue:44077", "issue:44186", "issue:44220", - "issue:44448", "issue:44488", "issue:44625", "issue:44661", @@ -4900,17 +4913,19 @@ "issue:46050" ], "soft_pairs": [ + "issue:44625|issue:45479", + "issue:43782|issue:45381", "issue:45356|issue:45907", "issue:43994|issue:44671", + "issue:43746|issue:46050", "issue:43720|issue:43819", "issue:30064|issue:36010", - "issue:43746|issue:46050", - "issue:43596|issue:44220", "issue:44077|issue:44625", + "issue:43596|issue:44220", "issue:43653|issue:43720", "issue:42915|issue:43746", - "issue:44220|issue:44661", "issue:43540|issue:44008", + "issue:44220|issue:44661", "issue:43618|issue:43697", "issue:43550|issue:45072", "issue:43746|issue:43873", @@ -4928,217 +4943,222 @@ "issue:43950|issue:45987", "issue:43643|issue:44877", "issue:43421|issue:44811", - "issue:44448|issue:44671", "issue:43540|issue:45381" ], - "prepared_review_unit_hash": "2d46f29d098b03386e2fb72ecc58f5e86cf4202efac3f675d7bcc9a9e81ed331", + "prepared_review_unit_hash": "60e1449be85798848ae9b983ef463237b7af30b778dbd919d2b741d7984466a2", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 37161, - "estimated_input_tokens": 9291, - "estimated_eval_tokens": 18838 + "soft_pair_count": 31, + "serialized_chars": 37453, + "estimated_input_tokens": 9364, + "estimated_eval_tokens": 18984 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly a grab bag of unrelated Transformers bug reports with only shallow keyword overlap around tokenizers, loading, and model regressions. I do not see any strong duplicate pairs; the soft matches are mostly same-subsysteem but different root causes.", - "confidence": 0.63, - "canonical_issue_reason": "Issue 44843 is the clearest, most concrete regression report with a single identifiable failing code path (`_patch_mistral_regex` / offline hub access), so it\u2019s the best anchor for this heterogeneous cluster.", + "summary": "The set is mostly heterogeneous: the soft-similarity pairs span different models, APIs, and failure modes, so there are no strong duplicate merges here. I would keep the cluster anchored on one concrete, well-documented bug report rather than merging across these pairs.", + "confidence": 0.91, + "canonical_issue_reason": "issue:46032 is the strongest representative in this set: it has a clear, concrete code-path failure, active discussion, and inbound references, making it the best anchor for the cluster. The other issues are largely orthogonal or model-specific.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44843 is the most actionable representative issue: specific symptom, clear repro surface, and it sits in a broader tokenizer-regression theme that appears multiple times in the cluster.", + "best_issue_reason": "issue:46032 is the most actionable and clearly reproduced bug here, with the strongest signal that it represents a real underlying defect worth deduplicating against if a matching report appears.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44625", + "right": "issue:45479", + "accept": false, + "reason": "Both involve labels/classification, but one is a config propagation bug and the other is a loss-function regression across sequence-classification models; different underlying defects." + }, + { + "left": "issue:43782", + "right": "issue:45381", + "accept": false, + "reason": "Different models and different failure modes: a Qwen3VL load/error path versus a video-input vision_position_ids bug in Qwen2.5-VL." + }, { "left": "issue:45356", "right": "issue:45907", "accept": false, - "reason": "Both involve tokenizer-related behavior, but 45356 is a Mistral-regex/codec regression while 45907 is about forwarding `revision` in `list_repo_files`; different bugs." + "reason": "Completely unrelated topics: tokenizer regression for Kimi-K2.5 versus hub API forwarding of revision in list_repo_files." }, { "left": "issue:43994", "right": "issue:44671", "accept": false, - "reason": "Both are model-output quality regressions, but they affect different models and different code paths (`SigLIP2` vs `CamemBERT`)." + "reason": "Both are wrong-output reports, but they concern different models and likely different root causes; too broad to treat as the same bug." + }, + { + "left": "issue:43746", + "right": "issue:46050", + "accept": false, + "reason": "Both are load/save related, but one is PEFT adapter loading and the other is quantized model persistence; different code paths and symptoms." }, { "left": "issue:43720", "right": "issue:43819", "accept": false, - "reason": "One is a BitNet loading/unpacking failure, the other is a DAC latent/STE mismatch; unrelated functionality and failure modes." + "reason": "Different model families and failures: packed-weight loading versus DAC latent conversion / missing STE behavior." }, { "left": "issue:30064", "right": "issue:36010", "accept": false, - "reason": "Image processor segmentation-map handling is unrelated to the `GenerationMixin` import error." + "reason": "No meaningful overlap: image processor segmentation-map handling versus a GenerationMixin import error." }, { - "left": "issue:43746", - "right": "issue:46050", + "left": "issue:44077", + "right": "issue:44625", "accept": false, - "reason": "Both mention loading/serialization, but one is PEFT adapter loading for GraniteSpeech and the other is quantized-model save/load corruption; different underlying problems." + "reason": "One is about patchtsmixer post_init policy, the other about num_labels propagation; unrelated implementation bugs." }, { "left": "issue:43596", "right": "issue:44220", "accept": false, - "reason": "Different failures in different paths: DeepSpeed ZeRO-3 init on BertModel vs `_torch_extract_fbank_features()` audio preprocessing." - }, - { - "left": "issue:44077", - "right": "issue:44625", - "accept": false, - "reason": "`patchtsmixer` post-init policy and Qwen3.5 `num_labels` propagation are unrelated config/model issues." + "reason": "Different subsystems entirely: DeepSpeed/BertModel initialization versus audio feature extraction." }, { "left": "issue:43653", "right": "issue:43720", "accept": false, - "reason": "BigBird special-token decode bug and BitNet packed-weight unpacking bug do not share the same root cause." + "reason": "Tokenizer special-token registration bug versus BitNet packed-weight unpacking; not the same underlying issue." }, { "left": "issue:42915", "right": "issue:43746", "accept": false, - "reason": "Qwen3Moe FP8 config failure is not the same as GraniteSpeech local checkpoint loading with PEFT adapters." + "reason": "Different models and components: Qwen3Moe FP8 configuration versus GraniteSpeech PEFT checkpoint loading." }, { - "left": "issue:44220", - "right": "issue:44661", + "left": "issue:43540", + "right": "issue:44008", "accept": false, - "reason": "Audio feature extraction bug vs tokenizer-mapping-name failure when adding a model-like entry; different subsystems." + "reason": "Both are multimodal, but one is a Qwen3OmniMoe video-processing ValueError and the other is a Gemma3n variable-name collision; not duplicates." }, { - "left": "issue:43540", - "right": "issue:44008", + "left": "issue:44220", + "right": "issue:44661", "accept": false, - "reason": "Video-input processing in Qwen3OmniMoe and Gemma 3n variable-name collision are distinct model-specific bugs." + "reason": "Audio fbank extraction and add-new-model-like / TOKENIZER_MAPPING_NAMES logic are unrelated." }, { "left": "issue:43618", "right": "issue:43697", "accept": false, - "reason": "CLIP attentions field regression and RTDetrV2 output differences are separate model API/behavior issues." + "reason": "One is a CLIPOutput field omission, the other is RTDetrV2 behavioral drift in v5; different bug classes." }, { "left": "issue:43550", "right": "issue:45072", "accept": false, - "reason": "Both mention SDPA/dtype/compile issues, but one is a Bamba compile failure and the other is a dtype mismatch in two different models." + "reason": "Different dtype/compile problems in different models and execution paths; too broad to merge." }, { "left": "issue:43746", "right": "issue:43873", "accept": false, - "reason": "Checkpoint/PEFT loading vs offloading-with-quantization are related only at a broad loading level, not the same bug." + "reason": "Checkpoint loading with PEFT adapters is distinct from offloading/quantization behavior." }, { "left": "issue:43295", "right": "issue:45834", "accept": false, - "reason": "Processor/tokenizer regression and Kosmos2.5 OCR index error are different symptoms and code paths." + "reason": "Processor/tokenizer regression and long OCR index error are different model-specific failures." }, { "left": "issue:43550", "right": "issue:43819", "accept": false, - "reason": "Both are model/runtime bugs, but SDPA compile failure and DAC STE mismatch are unrelated." + "reason": "torch.compile/SDPA failure versus DAC.from_latents mismatch; unrelated code paths." }, { "left": "issue:45183", "right": "issue:45479", "accept": false, - "reason": "Validation error messaging across models is not the same as the single-label-classification loss degeneracy bug." + "reason": "One is input validation/error-message logic, the other is a zero-loss regression; not the same defect." }, { "left": "issue:43388", "right": "issue:46032", "accept": false, - "reason": "Metric-gathering last-batch behavior and Mamba2 cache correctness are different training/runtime issues." + "reason": "Metric-gathering label truncation and Mamba2 cache correctness are unrelated training/runtime bugs." }, { "left": "issue:44186", "right": "issue:44898", "accept": false, - "reason": "Tokenizer padding/truncation crash and Perceiver non-default-resolution failure are separate model-specific bugs." + "reason": "Tokenizer crash on NER padding/truncation versus Perceiver image-classification resolution handling; different subsystems." }, { "left": "issue:43335", "right": "issue:43756", "accept": false, - "reason": "SwitchTransformers sparse-layer construction and Smollm3 RoPE-layer dropping are unrelated architecture/config regressions." + "reason": "SwitchTransformers sparse-layer creation and Smollm3 RoPE-layer retention are separate architecture/config issues." }, { "left": "issue:43716", "right": "issue:44008", "accept": false, - "reason": "Image preprocessor dtype mismatch in Mistral-3 is not the same as Gemma 3n audio tensor name collision." + "reason": "Different failures: dtype mismatch between image preprocessor/model versus an audio_mel_mask attribute collision." }, { "left": "issue:44062", "right": "issue:44488", "accept": false, - "reason": "`AddedToken()` keyword collision and failure to load `cjvt/sleng-bert` are distinct tokenizer/load issues." + "reason": "AddedToken keyword conflict is unrelated to loading a specific tokenizer/model checkpoint." }, { "left": "issue:43653", "right": "issue:43819", "accept": false, - "reason": "Special-token registration bug and DAC latent mismatch do not share a code path or symptom." + "reason": "Tokenizer special-token registration and DAC latent/forward mismatch are different domains and code paths." }, { "left": "issue:43531", "right": "issue:43643", "accept": false, - "reason": "Qwen3-MoE sliding-window behavior and strict `trust_remote_code` config loading are unrelated." + "reason": "Qwen3-MoE sliding_window behavior and trust_remote_code AutoConfig field loss are unrelated." }, { "left": "issue:44843", "right": "issue:45081", "accept": false, - "reason": "Both touch Mistral-regex handling, but 44843 is offline hub access from unconditional `model_info()` while 45081 is an AttributeError on tokenizer internals; close but not the same bug." + "reason": "Both mention _patch_mistral_regex, but the failures are different: unconditional model_info offline breakage versus backend_tokenizer AttributeError when fix_mistral_regex=True." }, { "left": "issue:43950", "right": "issue:45987", "accept": false, - "reason": "Non-persistent buffer corruption in `from_pretrained()` is unrelated to `StaticCache.get_seq_length()` returning the wrong type." + "reason": "Non-persistent buffer corruption on from_pretrained is unrelated to StaticCache.get_seq_length returning the wrong type." }, { "left": "issue:43643", "right": "issue:44877", "accept": false, - "reason": "`trust_remote_code` missing fields and strict config rejection for granite_speech are different config-loading failures." + "reason": "One is missing fields with trust_remote_code; the other is strict config blocking granite_speech loading. Similar area, different bug." }, { "left": "issue:43421", "right": "issue:44811", "accept": false, - "reason": "Runtime special-token post-processor update and Whisper `batch_decode()` ignoring `skip_special_tokens` are related to tokenization but not the same defect." - }, - { - "left": "issue:44448", - "right": "issue:44671", - "accept": false, - "reason": "Different model families and different regression surfaces: Pegasus generation output vs CamemBERT masked-LM predictions." + "reason": "Tokenizers post-processor refresh and Whisper batch_decode skip_special_tokens handling are different tokenizer/processor issues." }, { "left": "issue:43540", "right": "issue:45381", "accept": false, - "reason": "Qwen3OmniMoe video-processing error and Qwen2.5-VL video `vision_position_ids` issue are both multimodal, but the failing paths are distinct." + "reason": "Different Qwen video/multimodal bugs with distinct symptoms and likely different fixes." } ] }, @@ -5154,13 +5174,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 16, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:28218", "issue:29159", @@ -5185,6 +5205,7 @@ "issue:43994", "issue:44186", "issue:44442", + "issue:44448", "issue:44479", "issue:44485", "issue:44509", @@ -5211,231 +5232,238 @@ "issue:46209" ], "soft_pairs": [ + "issue:44448|issue:44671", "issue:30064|issue:36331", "issue:43782|issue:44977", "issue:43577|issue:43818", "issue:39692|issue:45834", "issue:45584|issue:45834", - "issue:44821|issue:45084", "issue:43322|issue:45685", + "issue:44821|issue:45084", "issue:43606|issue:45072", "issue:44479|issue:45834", "issue:43825|issue:43994", "issue:44186|issue:44442", "issue:34634|issue:36010", "issue:44485|issue:44509", - "issue:28218|issue:29159", - "issue:44561|issue:45020", "issue:45987|issue:46192", + "issue:44561|issue:45020", + "issue:28218|issue:29159", "issue:43577|issue:43582", - "issue:44610|issue:45072", "issue:44805|issue:44936", + "issue:44610|issue:45072", "issue:43720|issue:45072", "issue:43452|issue:44991", "issue:43825|issue:43827", - "issue:43404|issue:43782", "issue:43582|issue:44857", + "issue:43404|issue:43782", "issue:45081|issue:45406", "issue:43906|issue:46209", - "issue:44671|issue:45701", - "issue:42175|issue:45081" + "issue:42175|issue:45081", + "issue:44671|issue:45701" ], - "prepared_review_unit_hash": "406fb5e0e8454fed6a9c649241042ac7580872901fa1fe3f9d671b48af9a4054", + "prepared_review_unit_hash": "cc678a99f1829732f5cfe0268613c9f128518e202721d48486ebd7e55fa85e8c", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 28, - "serialized_chars": 35532, - "estimated_input_tokens": 8883, - "estimated_eval_tokens": 18022 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36406, + "estimated_input_tokens": 9102, + "estimated_eval_tokens": 18460 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly heterogeneous. The only recurring theme is tokenizer/version-regression reports, but none of the soft pairs clearly describe the same concrete bug or change, so I rejected all candidate merges.", - "confidence": 0.91, - "canonical_issue_reason": "issue:45701 is the broadest live tokenizer-regression report and has the strongest cross-references; it is the best anchor for the small tokenizer-related theme, even though the overall item set is not a true duplicate cluster.", + "summary": "These items do not form a single duplicate family; they span many unrelated areas (tokenizers, vision, model loading, trainer, docs, hardware-specific bugs). The closest overlap is the v5 pipeline-removal/docs subgroup, but even there the artifacts are not the same concrete bug or change.", + "confidence": 0.97, + "canonical_issue_reason": "Issue #43825 is the closest thing to a representative anchor because it has the clearest v5 pipeline-removal angle and some inbound traffic, but there is no true cluster-wide canonical issue here.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45701 is the most generally representative issue in this collection because it captures the broad tokenizer-version regression theme and is open with multiple inbound references.", + "best_issue_reason": "Issue #43825 is the best available issue representative for the small v5 pipeline/doc overlap, though it is not a strong canonical choice for the overall set.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44448", + "right": "issue:44671", + "accept": false, + "reason": "Both are model-output complaints, but they involve different models and different failure modes; not the same underlying bug." + }, { "left": "issue:30064", "right": "issue:36331", "accept": false, - "reason": "Image processing void-segmentation handling and a custom trainer signature error are unrelated code paths." + "reason": "Image segmentation-map processing and trainer custom loss kwargs are unrelated code paths." }, { "left": "issue:43782", "right": "issue:44977", "accept": false, - "reason": "Qwen3VL weight_only loading and Qwen3.5 flash-attention generation are different failures." + "reason": "One is a weight_only loading error for Qwen3VL, the other is a flash-attention generation issue for Qwen3.5; different bugs." }, { "left": "issue:43577", "right": "issue:43818", "accept": false, - "reason": "BLIP2 dtype casting and Video-LLaVA missing temporal attention are distinct model bugs." + "reason": "BLIP2 dtype propagation and Video-LLaVA temporal-attention/weight-sharing are unrelated." }, { "left": "issue:39692", "right": "issue:45834", "accept": false, - "reason": "SigLIP2 doc example errors and Kosmos2.5 OCR index errors are unrelated." + "reason": "SigLIP2 docs errors and Kosmos2.5 OCR index error are different problems." }, { "left": "issue:45584", "right": "issue:45834", "accept": false, - "reason": "Whisper empty-transcription generation and Kosmos2.5 OCR indexing are different problems." + "reason": "Whisper empty transcription failure after special-token alignment is unrelated to Kosmos2.5 long-OCR indexing." }, { - "left": "issue:44821", - "right": "issue:45084", + "left": "issue:43322", + "right": "issue:45685", "accept": false, - "reason": "AutoImageProcessor-from-URL loading and template compilation errors do not share a concrete failure mode." + "reason": "Llava Next segfault and MPS histogram int support error are different models and backends." }, { - "left": "issue:43322", - "right": "issue:45685", + "left": "issue:44821", + "right": "issue:45084", "accept": false, - "reason": "Llava Next segfaults and an MPS histogram integer error are separate backend/model issues." + "reason": "AutoImageProcessor URL loading and template-node compilation are unrelated failures." }, { "left": "issue:43606", "right": "issue:45072", "accept": false, - "reason": "CPU offload device mismatch and bfloat16 dtype mismatches are not the same bug." + "reason": "CPU offload device mismatch in bark-small and bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel are not the same bug." }, { "left": "issue:44479", "right": "issue:45834", "accept": false, - "reason": "Qwen video regression and Kosmos2.5 OCR index errors are unrelated." + "reason": "Video-input regression for Qwen VL models and Kosmos2.5 OCR indexing are unrelated." }, { "left": "issue:43825", "right": "issue:43994", "accept": false, - "reason": "A pipeline v5 error-message wording issue and a SigLIP2 model/pipeline result issue are different changes." + "reason": "v5 pipeline error messaging and SigLIP2 nonsensical outputs are different issues." }, { "left": "issue:44186", "right": "issue:44442", "accept": false, - "reason": "LayoutLMv2 tokenizer crashes and AutoTokenizer loading FastSpeech2ConformerTokenizer are unrelated tokenizer paths." + "reason": "LayoutLMv2 tokenizer crashes on NER/batched padding and AutoTokenizer failing to load FastSpeech2ConformerTokenizer are distinct tokenizer bugs." }, { "left": "issue:34634", "right": "issue:36010", "accept": false, - "reason": "BarkProcessor voice_preset handling and GenerationMixin import removal affect different subsystems." + "reason": "BarkProcessor voice_preset handling and GenerationMixin import failure are unrelated." }, { "left": "issue:44485", "right": "issue:44509", "accept": false, - "reason": "GLM-5 RoPE implementation discussion and v5 pipeline-task doc cleanup are not the same bug." + "reason": "GLM-5 RoPE implementation discussion and docs mentioning removed pipeline tasks are not the same change." }, { - "left": "issue:28218", - "right": "issue:29159", + "left": "issue:45987", + "right": "issue:46192", "accept": false, - "reason": "Both are tokenizer-related, but one is a specific added-token spacing bug and the other is a broader slow/fast inconsistency report." + "reason": "StaticCache.get_seq_length contract mismatch and return_tensors being ignored are separate API issues." }, { "left": "issue:44561", "right": "issue:45020", "accept": false, - "reason": "Removal of is_torch_fx_available and remote_code model breakage are different API breakages." + "reason": "Removal of is_torch_fx_available breaking trust_remote_code and generic remote_code breakages are related only broadly, not the same concrete bug." }, { - "left": "issue:45987", - "right": "issue:46192", + "left": "issue:28218", + "right": "issue:29159", "accept": false, - "reason": "StaticCache sequence-length return type and return_tensors being ignored are unrelated behavior changes." + "reason": "Tokenizer spacing after added tokens and slow/fast tokenizer inconsistency overlap conceptually, but the concrete behaviors and code paths are not clearly the same." }, { "left": "issue:43577", "right": "issue:43582", "accept": false, - "reason": "BLIP2 dtype propagation and Apple Silicon allocator warmup TypeError are unrelated." - }, - { - "left": "issue:44610", - "right": "issue:45072", - "accept": false, - "reason": "OmDet-Turbo processor size mismatch and dtype mismatch issues are different model/precision problems." + "reason": "BLIP2 dtype propagation and AppleSilicon warmup TypeError are unrelated." }, { "left": "issue:44805", "right": "issue:44936", "accept": false, - "reason": "Mask shape IndexError and trainer evaluate-after-train failure are distinct error scenarios." + "reason": "Mask/index shape mismatch and trainer.evaluate() after train() are different runtime failures." + }, + { + "left": "issue:44610", + "right": "issue:45072", + "accept": false, + "reason": "OmDet-Turbo input-size mismatch and dtype mismatch in other models are unrelated." }, { "left": "issue:43720", "right": "issue:45072", "accept": false, - "reason": "BitNet packed-weight unpacking and dtype mismatch inference bugs do not share the same code path." + "reason": "BitNet packed-weight unpacking during accelerate loading and bfloat16 dtype mismatch are different code-path problems." }, { "left": "issue:43452", "right": "issue:44991", "accept": false, - "reason": "gguf_file loading breakage and a model-specific tokenizer load failure in v5 are not the same underlying bug." + "reason": "gguf_file loading breaks and tokenizer loading for EMBEDDIA/est-roberta are different tokenizer/model-loading issues." }, { "left": "issue:43825", "right": "issue:43827", "accept": false, - "reason": "Both mention removed pipeline tasks, but one is an error message issue and the other is stale documentation." + "reason": "Both mention v5 pipeline removals, but one is an error-message bug and the other is a docs reference cleanup; not the same underlying bug or change." }, { - "left": "issue:43404", - "right": "issue:43782", + "left": "issue:43582", + "right": "issue:44857", "accept": false, - "reason": "Mistral3 lm_head tying and Qwen3VL weight_only loading are different model-loading defects." + "reason": "AppleSilicon warmup TypeError and LwDetrImageLoss float16 AMP crash are unrelated." }, { - "left": "issue:43582", - "right": "issue:44857", + "left": "issue:43404", + "right": "issue:43782", "accept": false, - "reason": "Apple Silicon allocator warmup and float16 AMP CUDA loss crashes are unrelated runtime issues." + "reason": "Mistral3 lm_head weight tying and Qwen3VL weight_only loading are different model-loading defects." }, { "left": "issue:45081", "right": "issue:45406", "accept": false, - "reason": "Mistral regex patching and Gemma4Processor missing _tokenizer are separate tokenizer/processor bugs." + "reason": "Mistral regex patch crash and Gemma4Processor missing _tokenizer are unrelated tokenizer/processor bugs." }, { "left": "issue:43906", "right": "issue:46209", "accept": false, - "reason": "A reproduction of an earlier issue and an FP8 inference failure on B200 are not the same change." + "reason": "An isolated reproduction of a prior issue and an FP8 B200 inference failure are not the same bug." }, { - "left": "issue:44671", - "right": "issue:45701", + "left": "issue:42175", + "right": "issue:45081", "accept": false, - "reason": "CamemBERT masked-LM prediction regressions and general tokenization changes are different failures." + "reason": "Missing TensorFlow backend in torch install and Mistral tokenizer regex crash are unrelated." }, { - "left": "issue:42175", - "right": "issue:45081", + "left": "issue:44671", + "right": "issue:45701", "accept": false, - "reason": "Missing TensorFlow in the torch extra and Mistral tokenizer regex patching are unrelated." + "reason": "CamemBERT masked LM predictions in v5 and version-dependent tokenization changes are too broad and not the same concrete defect." } ] }, @@ -5451,13 +5479,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 17, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:34634", "issue:36010", @@ -5473,6 +5501,7 @@ "issue:43550", "issue:43577", "issue:43638", + "issue:43643", "issue:43644", "issue:43645", "issue:43650", @@ -5489,11 +5518,9 @@ "issue:44060", "issue:44291", "issue:44368", - "issue:44387", "issue:44554", "issue:44589", "issue:44610", - "issue:44617", "issue:44811", "issue:44912", "issue:44936", @@ -5505,249 +5532,250 @@ "issue:45459", "issue:45538", "issue:45600", + "issue:45759", "issue:45803", "issue:46285" ], "soft_pairs": [ "issue:43262|issue:43819", - "issue:43653|issue:44912", "issue:45003|issue:46285", "issue:43526|issue:43697", + "issue:43653|issue:44912", "issue:44060|issue:44368", "issue:43650|issue:43906", "issue:43742|issue:45020", "issue:45084|issue:45538", "issue:43329|issue:45600", "issue:43746|issue:44811", - "issue:44589|issue:45042", "issue:43644|issue:43749", + "issue:44589|issue:45042", "issue:43638|issue:44936", "issue:43550|issue:44912", "issue:43819|issue:43927", - "issue:44387|issue:44617", "issue:34634|issue:36331", "issue:43854|issue:44977", "issue:39692|issue:44912", - "issue:36010|issue:43645", "issue:38175|issue:44610", + "issue:36010|issue:43645", "issue:43720|issue:44610", "issue:45459|issue:45803", - "issue:43526|issue:44554", "issue:43454|issue:43577", + "issue:43526|issue:44554", "issue:36010|issue:36331", "issue:43262|issue:44811", "issue:43749|issue:44291", "issue:43531|issue:45042", - "issue:43257|issue:43329" + "issue:43257|issue:43329", + "issue:43643|issue:45759" ], - "prepared_review_unit_hash": "4c9c60b7b92bb8906755388c0500953e639f521c0e26f92e828352333c6a52b0", + "prepared_review_unit_hash": "515588f8868a93dd25def96dd2acdb501da8f8e76a40d1add9fe9fae24d78bcb", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36924, - "estimated_input_tokens": 9231, - "estimated_eval_tokens": 18718 + "serialized_chars": 37019, + "estimated_input_tokens": 9255, + "estimated_eval_tokens": 18766 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly unrelated model- or component-specific regressions; none of the soft pairs look like the same underlying bug or change, so no merges are justified.", - "confidence": 0.92, - "canonical_issue_reason": "issue:45020 is the broadest and most representative regression report in the set: it captures a general class of recent Transformers breakages affecting remote_code models, unlike the more narrowly scoped model/component-specific issues.", + "summary": "This cluster is mostly a heterogeneous mix of unrelated bugs; none of the proposed soft pairs looks like the same concrete underlying issue, so all soft edges should be rejected.", + "confidence": 0.94, + "canonical_issue_reason": "Issue 45020 is the broadest and most representative loading-regression report in the set, but the cluster is too mixed to treat any pair as a true duplicate group.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the strongest standalone issue candidate because it is upstream-facing, broadly scoped, and more representative of the cluster than the highly specific bugs around single models, processors, or trainers.", + "best_issue_reason": "45020 is the best cluster anchor because it is the most general current regression report; the others are narrow, model-specific edge cases rather than duplicates of it.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43262", "right": "issue:43819", "accept": false, - "reason": "Audio chat-template sampling-rate defaults and DAC latent reconstruction are unrelated code paths and failure modes." - }, - { - "left": "issue:43653", - "right": "issue:44912", - "accept": false, - "reason": "Tokenizer special-token registration and MXFP4 quantization loading are different bugs with no shared underlying fix." + "reason": "Both are audio-related, but one is chat-template sampling-rate handling and the other is a DAC latent/STE correctness bug; different code paths." }, { "left": "issue:45003", "right": "issue:46285", "accept": false, - "reason": "Unsafe sys.modules access and CLIP text model structure inconsistency are separate modeling-utils vs architecture issues." + "reason": "Both involve modeling internals, but one is unsafe sys.modules access and the other is CLIP text-model structure changes; not the same bug." }, { "left": "issue:43526", "right": "issue:43697", "accept": false, - "reason": "Beit label reduction and RTDetr output drift are different processor/model problems; they would not be merged into one fix." + "reason": "BEiT label reduction and RTDetrV2 output drift are separate vision-model issues with no shared concrete failure mode." + }, + { + "left": "issue:43653", + "right": "issue:44912", + "accept": false, + "reason": "Tokenizer special-token registration is unrelated to MXFP4 quantization loading behavior." }, { "left": "issue:44060", "right": "issue:44368", "accept": false, - "reason": "Incorrect weight tying in Qwen3-Next and a tied-embeddings warning during LoRA fine-tuning are not the same concrete bug." + "reason": "Both mention tied-weights warnings, but they concern different models and different underlying warnings, so they are not duplicates." }, { "left": "issue:43650", "right": "issue:43906", "accept": false, - "reason": "A placeholder 'ADD THE DATA' issue and an isolated reproduction of another bug are not duplicates." + "reason": "A vague placeholder issue is unrelated to an isolated reproduction of another bug." }, { "left": "issue:43742", "right": "issue:45020", "accept": false, - "reason": "Loading a specific MobileLLM checkpoint failing with KeyError is distinct from the broad remote_code regression report." + "reason": "A specific MobileLLM loading KeyError is not the same concrete issue as the broader remote_code regression." }, { "left": "issue:45084", "right": "issue:45538", "accept": false, - "reason": "Template compilation errors and CLIPTokenizer max-length semantics are unrelated issues." + "reason": "These are unrelated: template-node compilation versus tokenizer max length defaults." }, { "left": "issue:43329", "right": "issue:45600", "accept": false, - "reason": "A multimodal token counting video-branch bug and stale auto_mappings references are different maintenance problems." + "reason": "One is a missing video-token-count implementation path; the other is stale auto-mapping config references." }, { "left": "issue:43746", "right": "issue:44811", "accept": false, - "reason": "PEFT checkpoint loading for GraniteSpeech and Whisper batch_decode skip_special_tokens behavior are unrelated." + "reason": "PEFT local-checkpoint loading and Whisper batch decoding are unrelated code paths." }, { - "left": "issue:44589", - "right": "issue:45042", + "left": "issue:43644", + "right": "issue:43749", "accept": false, - "reason": "Missing Float8 storage lookup and unnecessary torchvision dependency in PIL image processors affect different subsystems." + "reason": "Non-persistent buffer junk and FSDP CPU RAM-efficient loading are both loading-related, but not the same bug or fix." }, { - "left": "issue:43644", - "right": "issue:43749", + "left": "issue:44589", + "right": "issue:45042", "accept": false, - "reason": "Non-persistent buffer initialization corruption and FSDP CPU RAM efficient loading are both loading-related but not the same code-path bug." + "reason": "Float8 storage lookup failure is unrelated to PIL image processor torchvision requirements." }, { "left": "issue:43638", "right": "issue:44936", "accept": false, - "reason": "Zero3 empty-index failure on non-pretrained Bert and evaluate-after-train behavior are separate trainer/distributed-training issues." + "reason": "Zero3/non-pretrained-Bert index errors and train-then-evaluate failure are different trainer/runtime problems." }, { "left": "issue:43550", "right": "issue:44912", "accept": false, - "reason": "torch.compile+SDPA failure in Bamba and MXFP4 fallback during load for git-oss are different model/runtime regressions." + "reason": "torch.compile/SDPA failures are unrelated to MXFP4 quantization fallback behavior." }, { "left": "issue:43819", "right": "issue:43927", "accept": false, - "reason": "DAC latent mismatch and DiaConfig token-ID persistence loss are unrelated implementation bugs." - }, - { - "left": "issue:44387", - "right": "issue:44617", - "accept": false, - "reason": "General int4 quantization memory regression and Sam3Video OOM are not the same concrete defect." + "reason": "DAC latent reconstruction mismatch and DiaConfig token-ID persistence are distinct model bugs." }, { "left": "issue:34634", "right": "issue:36331", "accept": false, - "reason": "BarkProcessor voice_preset and CustomTrainer.compute_loss signature mismatch are unrelated API issues." + "reason": "BarkProcessor voice_preset and CustomTrainer.compute_loss signature mismatch are unrelated." }, { "left": "issue:43854", "right": "issue:44977", "accept": false, - "reason": "Model loading test failure for GLM-4.7-Flash and flash-attention generation issues in Qwen3.5 are different bugs." + "reason": "These are different model-specific generation/loading regressions, not one shared code-path bug." }, { "left": "issue:39692", "right": "issue:44912", "accept": false, - "reason": "SigLIP2 doc/example errors and MXFP4 load fallback are distinct; one is docs/model mismatch, the other quantization loading." + "reason": "SigLIP2 doc/example errors are unrelated to git-oss MXFP4 quantization loading." }, { - "left": "issue:36010", - "right": "issue:43645", + "left": "issue:38175", + "right": "issue:44610", "accept": false, - "reason": "GenerationMixin import breakage and Jupyter custom model initialization breakage are separate import/runtime issues." + "reason": "Unexpected zero probabilities and OmDet-Turbo resolution mismatch are separate vision-model issues." }, { - "left": "issue:38175", - "right": "issue:44610", + "left": "issue:36010", + "right": "issue:43645", "accept": false, - "reason": "Unexpected zero probabilities in SigLIP2 and OmDet-Turbo processor size mismatch are different model/processor regressions." + "reason": "GenerationMixin import breakage is not the same as the Jupyter custom-model initialization regression." }, { "left": "issue:43720", "right": "issue:44610", "accept": false, - "reason": "BitNet packed-weight unpacking during accelerate loading and OmDet-Turbo input-size mismatch are unrelated." + "reason": "BitNet packed-weight loading and OmDet-Turbo image-size mismatch are unrelated." }, { "left": "issue:45459", "right": "issue:45803", "accept": false, - "reason": "Tokenizer protobuf error masking and FuyuBatchFeature bare except swallowing interrupts are both error-handling smells but not the same fix or code path." + "reason": "Both involve broad exception handling, but they target different components and different failure modes." }, { - "left": "issue:43526", - "right": "issue:44554", + "left": "issue:43454", + "right": "issue:43577", "accept": false, - "reason": "Beit label reduction and MPS attention head-dimension correctness are unrelated bugs." + "reason": "Weight tying garbage-generation and Blip2 dtype handling are different model-loading bugs." }, { - "left": "issue:43454", - "right": "issue:43577", + "left": "issue:43526", + "right": "issue:44554", "accept": false, - "reason": "AyaVision weight tying and BLIP2 dtype propagation are distinct model initialization issues." + "reason": "BEiT label reduction and MPS attention correctness are unrelated issues." }, { "left": "issue:36010", "right": "issue:36331", "accept": false, - "reason": "An import error for GenerationMixin and a custom trainer loss-signature error are unrelated failures." + "reason": "Importing GenerationMixin and compute_loss keyword-argument incompatibility are separate API breakages." }, { "left": "issue:43262", "right": "issue:44811", "accept": false, - "reason": "Audio apply_chat_template sampling-rate defaulting and Whisper batch_decode skip_special_tokens behavior are separate processor bugs." + "reason": "Audio chat-template sampling rate and Whisper batch_decode skip_special_tokens are different processor behaviors." }, { "left": "issue:43749", "right": "issue:44291", "accept": false, - "reason": "FSDP CPU RAM loading and init_empty_weights _is_hf_initialized argument mismatch are different loading mechanisms and fixes." + "reason": "FSDP CPU RAM-efficient loading and init_empty_weights _is_hf_initialized are related to low-memory loading, but not the same concrete bug." }, { "left": "issue:43531", "right": "issue:45042", "accept": false, - "reason": "Qwen3-MoE sliding_window and PIL backend torchvision dependency are unrelated regressions." + "reason": "Qwen3-MoE sliding-window behavior is unrelated to PIL backend torchvision dependency requirements." }, { "left": "issue:43257", "right": "issue:43329", "accept": false, - "reason": "Qwen3 MOE weight conversion under accelerate+deepspeed and multimodal token counting video-branch undefined names are unrelated." + "reason": "MoE weight conversion with accelerate/deepspeed and multimodal token counting video-path bugs are unrelated." + }, + { + "left": "issue:43643", + "right": "issue:45759", + "accept": false, + "reason": "Both touch config/model loading, but one is missing fields from trust_remote_code and the other is text_config unwrapping for composite Qwen configs; not the same fix." } ] }, @@ -5763,13 +5791,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 18, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:24643", "issue:29159", @@ -5789,7 +5817,6 @@ "issue:43577", "issue:43582", "issue:43606", - "issue:43643", "issue:43696", "issue:43697", "issue:43742", @@ -5800,6 +5827,7 @@ "issue:44479", "issue:44560", "issue:44568", + "issue:44737", "issue:44821", "issue:44849", "issue:44898", @@ -5815,13 +5843,12 @@ "issue:45405", "issue:45488", "issue:45538", + "issue:45588", "issue:45701", - "issue:45759", "issue:45859", "issue:45907" ], "soft_pairs": [ - "issue:43643|issue:45759", "issue:42490|issue:43257", "issue:43329|issue:43450", "issue:43606|issue:45859", @@ -5829,180 +5856,175 @@ "issue:44568|issue:45356", "issue:29159|issue:34567", "issue:43329|issue:44479", - "issue:44849|issue:45701", "issue:42915|issue:43742", - "issue:44291|issue:45042", "issue:43257|issue:43931", + "issue:44849|issue:45701", + "issue:44291|issue:45042", "issue:43452|issue:43582", "issue:43927|issue:44912", "issue:43697|issue:45357", "issue:44568|issue:45488", - "issue:33357|issue:34567", - "issue:42915|issue:44821", "issue:41720|issue:43540", + "issue:42915|issue:44821", + "issue:33357|issue:34567", "issue:42915|issue:45538", "issue:44987|issue:45538", "issue:44479|issue:45381", "issue:43540|issue:44479", "issue:43317|issue:45538", "issue:43540|issue:44560", - "issue:44987|issue:45907", "issue:45137|issue:45405", + "issue:44987|issue:45907", "issue:24643|issue:29159", "issue:43742|issue:44291", "issue:43577|issue:45072", - "issue:44062|issue:44991", "issue:38175|issue:44898", - "issue:43257|issue:43696" + "issue:44062|issue:44991", + "issue:43257|issue:43696", + "issue:44737|issue:45588" ], - "prepared_review_unit_hash": "7863ce51a072931d6ac40d058aec43ecb68585198b5a3d073426d05f61484e21", + "prepared_review_unit_hash": "4e58f44ef3ce4f591246d3add0b5af86ebb988e17545e1beb8bb3cfd80cfcbd3", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37725, - "estimated_input_tokens": 9432, - "estimated_eval_tokens": 19120 + "serialized_chars": 37720, + "estimated_input_tokens": 9430, + "estimated_eval_tokens": 19116 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly heterogeneous across tokenizer, loading, quantization, device-map, and training bugs. The only plausible subcluster is the Qwen video regressions, but the evidence is only soft title similarity; there are no explicit duplicate links or shared-target signals to confirm a merged duplicate set.", - "confidence": 0.67, - "canonical_issue_reason": "issue:44479 is the best available issue representative for the small Qwen video-regression subcluster because it broadly names the shared video-input failure mode. That said, the packet does not show explicit duplicate links or shared-target evidence, so this is only a cautious representative choice, not a confirmed duplicate hub.", + "summary": "These issues are mostly unrelated; the soft-similarity links do not rise to duplicate-level matches. I\u2019d keep one concrete tokenizer regression as the best representative issue, but no PR is present.", + "confidence": 0.89, + "canonical_issue_reason": "issue:45488 is the clearest standalone report: a specific tokenizer regression with concrete impact on DeepSeek V3/R1-style models. It is more actionable than the vaguer tokenization reports in the set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44479 is the strongest overall issue pick because it is the broadest, most central report among the only closely related video issues. The rest of the cluster is too mixed, and the related Qwen entries remain only tentatively similar.", + "best_issue_reason": "issue:45488 is the best issue to anchor this cluster because it is specific, reproducible, and describes a real user-facing regression rather than a broad symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43643", - "right": "issue:45759", - "accept": false, - "reason": "Both involve config/model loading, but they describe different failures and code paths." - }, { "left": "issue:42490", "right": "issue:43257", "accept": false, - "reason": "Trailing-slash local-path import warning is unrelated to Qwen3 MOE weight conversion." + "reason": "Both involve loading, but one is a path/import warning and the other is Qwen3 MoE weight conversion with accelerate+deepspeed; different failures and code paths." }, { "left": "issue:43329", "right": "issue:43450", "accept": false, - "reason": "Both are video-related, but one is a missing helper in multimodal token counting and the other is a batched video processor shape bug." + "reason": "Both are video/multimodal related, but one is an undefined-helper bug in video token counting and the other is a batched-shape bug in video processors; not the same defect." }, { "left": "issue:43606", "right": "issue:45859", "accept": false, - "reason": "Both mention CPU offload, but they are different model-specific incompatibilities." + "reason": "Both mention CPU offload, but they affect different models and different modules; no evidence they are the same concrete bug." }, { "left": "issue:34567", "right": "issue:40990", "accept": false, - "reason": "TrainerState token accounting and GPT-oss perplexity are unrelated failures." + "reason": "TrainerState token counting not updating is unrelated to GPT-OSS perplexity behavior." }, { "left": "issue:44568", "right": "issue:45356", "accept": false, - "reason": "These are separate tokenizer regressions affecting different models and behaviors." + "reason": "Both are tokenizer regressions, but they describe different model-specific issues and different tokenizer behaviors." }, { "left": "issue:29159", "right": "issue:34567", "accept": false, - "reason": "Slow/fast tokenizer inconsistency is unrelated to TrainerState token counting." + "reason": "Tokenizer consistency versus TrainerState token counting are unrelated areas." }, { "left": "issue:43329", "right": "issue:44479", "accept": false, - "reason": "They are both in the Qwen video area, but the concrete symptoms differ and the packet does not prove the same bug." + "reason": "The latter is a broad video-input regression across Qwen VL families; the former is a specific missing-undefined-helper bug. They may be in the same area, but not clearly the same underlying defect." }, { - "left": "issue:44849", - "right": "issue:45701", + "left": "issue:42915", + "right": "issue:43742", "accept": false, - "reason": "One is a specific Qwen3.5 output bug; the other is a broad tokenization regression report." + "reason": "Different model-loading errors: FineGrainedFP8Config for Qwen3 MoE versus a key error on MobileLLM-125M." }, { - "left": "issue:42915", - "right": "issue:43742", + "left": "issue:43257", + "right": "issue:43931", "accept": false, - "reason": "Qwen3Moe FineGrainedFP8Config failure and MobileLLM-125M key error are different loading problems." + "reason": "Both are model loading problems, but one is MOE weight conversion with deepspeed while the other is a Qwen3-VL shape mismatch; different root causes." }, { - "left": "issue:44291", - "right": "issue:45042", + "left": "issue:44849", + "right": "issue:45701", "accept": false, - "reason": "init_empty_weights/_is_hf_initialized and PIL-backend torchvision requirement are unrelated." + "reason": "Qwen3.5 hidden-state output bug is unrelated to the generic tokenization-change report." }, { - "left": "issue:43257", - "right": "issue:43931", + "left": "issue:44291", + "right": "issue:45042", "accept": false, - "reason": "Qwen3 MOE weight conversion and Qwen3-VL weight-shape mismatch are different loading bugs." + "reason": "Unexpected init_empty_weights argument and PIL image-processor torchvision dependency are unrelated." }, { "left": "issue:43452", "right": "issue:43582", "accept": false, - "reason": "gguf_file loading breakage and AppleSilicon TypeError are unrelated." + "reason": "GGUF loading failures and Apple Silicon caching_allocator_warmup TypeError are unrelated." }, { "left": "issue:43927", "right": "issue:44912", "accept": false, - "reason": "DiaConfig token-ID loss and MXFP4 quantization fallback are different bugs." + "reason": "DiaConfig save/load token IDs and MXFP4 quantization fallback are different bugs." }, { "left": "issue:43697", "right": "issue:45357", "accept": false, - "reason": "Different model families and different symptoms: RTDetrV2 output drift versus Qwen3.5 save_pretrained key handling." + "reason": "RTDetrV2 output changes and Qwen3.5 save_pretrained visual-key corruption are different code paths." }, { "left": "issue:44568", "right": "issue:45488", "accept": false, - "reason": "Both are tokenizer-related, but they affect different models and failure modes." + "reason": "Both are tokenizer regressions, but one is about add_special_tokens/BOS-EOS behavior and the other is LlamaTokenizer overriding tokenizer.json pre-tokenizer settings; different concrete bugs." }, { - "left": "issue:33357", - "right": "issue:34567", + "left": "issue:41720", + "right": "issue:43540", "accept": false, - "reason": "MacOS bus error with CLIP and TrainerState token counting are unrelated." + "reason": "Auto device-mapping CUDA assert and Qwen3OmniMoe video ValueError are unrelated." }, { "left": "issue:42915", "right": "issue:44821", "accept": false, - "reason": "Qwen3Moe FP8 config failure and AutoImageProcessor URL loading are unrelated." + "reason": "Qwen3 MoE FP8 loading and AutoImageProcessor-from-URL are different failures." }, { - "left": "issue:41720", - "right": "issue:43540", + "left": "issue:33357", + "right": "issue:34567", "accept": false, - "reason": "Auto device mapping CUDA assert and Qwen3OmniMoe video ValueError are distinct failures." + "reason": "MacOS bus error with CLIP and TrainerState token counting are unrelated." }, { "left": "issue:42915", "right": "issue:45538", "accept": false, - "reason": "Qwen3Moe FP8 config failure and CLIPTokenizer model_max_length bug are unrelated." + "reason": "Qwen3 MoE FP8 load failure is unrelated to CLIPTokenizer's max-length bug." }, { "left": "issue:44987", @@ -6014,70 +6036,85 @@ "left": "issue:44479", "right": "issue:45381", "accept": false, - "reason": "Both are Qwen video regressions, but 45381 looks like a specific vision_position_ids symptom and the packet does not confirm the same underlying code-path bug." + "reason": "Both concern video input in Qwen VL models, but they report different regressions and symptoms; not enough to treat as the same bug." }, { "left": "issue:43540", "right": "issue:44479", "accept": false, - "reason": "Both involve video, but the Qwen3OmniMoe StopIteration report is not clearly the same regression as 44479." + "reason": "Both are video-related, but one is a Qwen3OmniMoe processing error and the other is a broader v5.3.0 regression across multiple VL families; different concrete problems." }, { - "left": "issue:44987", - "right": "issue:45907", + "left": "issue:43317", + "right": "issue:45538", + "accept": false, + "reason": "Dequantized model offload failure and CLIPTokenizer max-length bug are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44560", "accept": false, - "reason": "Model loading failure and list_repo_files revision forwarding are unrelated." + "reason": "Both involve video inputs, but they affect different models and describe different failures (ValueError vs StopIteration)." }, { "left": "issue:45137", "right": "issue:45405", "accept": false, - "reason": "DeepSpeed ZeRO3 deque error and MIN_PEFT_VERSION packaging issue are unrelated." + "reason": "DeepSpeed ZeRO3 deque error and a PEFT version bump are unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45907", + "accept": false, + "reason": "Model loading regression and list_repo_files not forwarding revision are unrelated." }, { "left": "issue:24643", "right": "issue:29159", "accept": false, - "reason": "DeepSpeed weight-dimension error and tokenizer inconsistency are unrelated." + "reason": "DeepSpeed training weight shape error and tokenizer slow/fast inconsistency are unrelated." }, { "left": "issue:43742", "right": "issue:44291", "accept": false, - "reason": "MobileLLM-125M key error and unexpected _is_hf_initialized are different loading failures." + "reason": "MobileLLM key error and init_empty_weights argument handling are different loading bugs." }, { "left": "issue:43577", "right": "issue:45072", "accept": false, - "reason": "Both mention dtype, but the underlying bugs differ." + "reason": "Both mention dtype issues, but they affect different architectures and code paths; not the same bug." }, { - "left": "issue:44062", - "right": "issue:44991", + "left": "issue:38175", + "right": "issue:44898", "accept": false, - "reason": "AddedToken special-argument duplication and EMBEDDIA tokenizer loading are different tokenizer bugs." + "reason": "SIGLIP zero probabilities and Perceiver non-default-resolution failure are different vision-model issues." }, { - "left": "issue:38175", - "right": "issue:44898", + "left": "issue:44062", + "right": "issue:44991", "accept": false, - "reason": "Unexpected zero probabilities in SigLIP2 and Perceiver non-default-resolution failure are unrelated image-model issues." + "reason": "AddedToken keyword collision and est-roberta tokenizer loading failure are not the same defect." }, { "left": "issue:43257", "right": "issue:43696", "accept": false, - "reason": "Qwen3 MOE weight conversion and GPT-oss CUDA OOM are unrelated." + "reason": "Qwen3 MoE weight conversion and GPT-OSS out-of-memory are unrelated." + }, + { + "left": "issue:44737", + "right": "issue:45588", + "accept": false, + "reason": "XLNet CPU placement bug and flash_attention s_aux=None crash are unrelated." } ] }, - "evaluator_result": { - "accept": false, - "feedback": "The overall clustering judgment is mostly conservative, but the canonical/best-issue reasoning overstates the evidence. In the packet there are no explicit duplicate links or shared-target evidence, so describing issue:44479 as a \u201cconfirmed duplicate-worthy subset\u201d and citing a \u201cdirect duplicate-style link\u201d is too strong. The Qwen video issues do look plausibly related by titles, but the packet only supports a tentative soft similarity, not confirmation. A more cautious summary would be acceptable." - }, - "evaluator_used": true, - "retried": true, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null @@ -6087,13 +6124,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 19, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:28282", "issue:29127", @@ -6108,6 +6145,7 @@ "issue:43493", "issue:43526", "issue:43531", + "issue:43582", "issue:43643", "issue:43646", "issue:43697", @@ -6122,13 +6160,13 @@ "issue:44488", "issue:44589", "issue:44625", - "issue:44737", "issue:44756", "issue:44792", "issue:44821", "issue:44877", "issue:44913", "issue:44964", + "issue:44977", "issue:45042", "issue:45081", "issue:45216", @@ -6138,27 +6176,25 @@ "issue:45381", "issue:45405", "issue:45479", - "issue:45588", "issue:45600", "issue:45920", "issue:45987", "issue:46121" ], "soft_pairs": [ - "issue:44737|issue:45588", "issue:43404|issue:45987", "issue:34567|issue:36331", - "issue:44062|issue:45081", "issue:43646|issue:43749", + "issue:44062|issue:45081", "issue:44913|issue:46121", - "issue:43493|issue:43994", "issue:43906|issue:45381", + "issue:43493|issue:43994", "issue:43749|issue:44756", "issue:43257|issue:43756", "issue:44792|issue:44964", "issue:44291|issue:44821", - "issue:36010|issue:42175", "issue:28282|issue:34689", + "issue:36010|issue:42175", "issue:42175|issue:43824", "issue:45042|issue:45081", "issue:45357|issue:45920", @@ -6171,208 +6207,209 @@ "issue:43756|issue:45325", "issue:43531|issue:44821", "issue:44625|issue:44877", - "issue:43697|issue:45216" + "issue:43697|issue:45216", + "issue:43582|issue:44977" ], - "prepared_review_unit_hash": "d3a084ed0677bf37bb7fed8cbbb4dc0551a772033067fd8c0d2154c0e3808efe", + "prepared_review_unit_hash": "686b7a32efa7d16ad8e92f75474cc1478af555c4641c4f9e5540a0659eb49b95", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 36146, - "estimated_input_tokens": 9037, - "estimated_eval_tokens": 18330 + "serialized_chars": 36070, + "estimated_input_tokens": 9018, + "estimated_eval_tokens": 18292 }, "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 bugs and regressions, with one plausible duplicate SigLIP2 pair. No PRs are present.", - "confidence": 0.41, - "canonical_issue_reason": "issue:45920 is the broadest, most representative open report here: a v5 regression in AutoTokenizer producing wrong token IDs across multiple model families, with clear user impact and ongoing relevance.", + "summary": "This cluster is mostly a grab bag of unrelated Transformers issues across loading, configs, trainers, tokenizers, and multimodal models. The SigLIP2 pair is only adjacent/possibly related, not a confirmed duplicate; the rest should remain separate.", + "confidence": 0.74, + "canonical_issue_reason": "No single canonical issue cleanly represents the cluster because the items are largely unrelated. If one must be chosen, issue 43493 is the broadest technical report and the most plausible anchor, but it does not unify the whole set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 has the strongest cross-cutting scope and is the best single issue to anchor this otherwise heterogeneous cluster.", + "best_issue_reason": "Issue 43493 is the best representative issue for the cluster only in the limited sense that it is the most detailed substantive report among the adjacent SigLIP2 items; it should not be read as a true cluster-wide duplicate anchor.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44737", - "right": "issue:45588", - "accept": false, - "reason": "Different subsystems and symptoms: XLNet positional encoding device placement vs flash-attention sink handling." - }, { "left": "issue:43404", "right": "issue:45987", "accept": false, - "reason": "Unrelated bugs: Mistral3 weight tying vs StaticCache sequence-length return type." + "reason": "Different bugs: Mistral lm_head tying vs StaticCache sequence-length return type. Same area, not the same defect." }, { "left": "issue:34567", "right": "issue:36331", "accept": false, - "reason": "Both involve Trainer APIs, but one is a stale counter and the other is a custom loss signature change; not the same bug." + "reason": "Both touch trainer behavior, but one is token counting and the other is a new compute_loss kwarg. Different code paths." }, { - "left": "issue:44062", - "right": "issue:45081", + "left": "issue:43646", + "right": "issue:43749", "accept": false, - "reason": "Distinct tokenizer failures with different code paths: AddedToken kwargs conflict vs Mistral regex patching." + "reason": "One is a Transformers 5 custom model init regression; the other is FSDP CPU RAM efficient loading. Too different to merge." }, { - "left": "issue:43646", - "right": "issue:43749", + "left": "issue:44062", + "right": "issue:45081", "accept": false, - "reason": "Different failures: config field loss under trust_remote_code vs FSDP CPU RAM efficient loading." + "reason": "Different tokenizer-related failures: AddedToken argument handling vs mistral regex patching/backend access." }, { "left": "issue:44913", "right": "issue:46121", "accept": false, - "reason": "Both touch RoPE/config handling, but one is config reload persistence and the other is a validation TypeError on JSON-loaded lists." - }, - { - "left": "issue:43493", - "right": "issue:43994", - "accept": true, - "reason": "Both point to the same SigLIP2 correctness problem: the HF implementation diverges from expected behavior, and the model/pipeline yields bad outputs for the same family/checkpoint." + "reason": "Both involve RoPE config/validation, but one is silent reload persistence and the other is a TypeError in rope dict conversion. Not the same bug." }, { "left": "issue:43906", "right": "issue:45381", "accept": false, - "reason": "Different model families and symptoms; one is an isolated reproduction of an older issue, the other is a Qwen2.5-VL video position-id bug." + "reason": "Isolated reproduction of an old issue versus a Qwen2.5-VL video position-id regression. Different models and symptoms." + }, + { + "left": "issue:43493", + "right": "issue:43994", + "accept": false, + "reason": "These are SigLIP2-adjacent but not clearly the same bug: one reports a JAX/HF implementation discrepancy, the other nonsensical AutoModel/pipeline outputs. Related, but unconfirmed duplicate." }, { "left": "issue:43749", "right": "issue:44756", "accept": false, - "reason": "Both concern loading/performance, but one is FSDP RAM-efficient loading and the other is mmap/OOM on Strix Halo." + "reason": "Both concern memory/loading, but one is an FSDP loading regression and the other is a Strix Halo mmap OOM workaround. Different fixes." }, { "left": "issue:43257", "right": "issue:43756", "accept": false, - "reason": "Different Qwen3-MoE vs Smollm3 model-specific bugs with no clear shared code path." + "reason": "Qwen3 MoE weight conversion with accelerate/deepspeed is unrelated to Smollm3 RoPE layer dropping." }, { "left": "issue:44792", "right": "issue:44964", "accept": false, - "reason": "Separate multimodal model failures affecting different checkpoints and code paths." + "reason": "Different multimodal model issues: Janus image-generation test failure vs Phi-4 multimodal model loading failure." }, { "left": "issue:44291", "right": "issue:44821", "accept": false, - "reason": "Different loading bugs: init_empty_weights / _is_hf_initialized vs loading AutoImageProcessor from URL." + "reason": "init_empty_weights/_is_hf_initialized argument handling is unrelated to AutoImageProcessor URL loading." }, { - "left": "issue:36010", - "right": "issue:42175", + "left": "issue:28282", + "right": "issue:34689", "accept": false, - "reason": "Unrelated issues: GenerationMixin import path vs missing TensorFlow backend in a torch install extra." + "reason": "PyTorch-not-found ImportError versus a specific Llama 3.2 vision loading regression. Not the same defect." }, { - "left": "issue:28282", - "right": "issue:34689", + "left": "issue:36010", + "right": "issue:42175", "accept": false, - "reason": "Generic missing-PyTorch ImportError vs a specific Llama 3.2 Vision model-loading regression." + "reason": "GenerationMixin import failure and missing TensorFlow backend in a torch install are unrelated packaging/import problems." }, { "left": "issue:42175", "right": "issue:43824", "accept": false, - "reason": "Different missing-dependency/import problems: backend packaging vs missing Qwen2.5-VL class export." + "reason": "Missing TensorFlow backend versus missing Qwen2_5_VL symbol. Different import/export issues." }, { "left": "issue:45042", "right": "issue:45081", "accept": false, - "reason": "Different bugs: PIL backend now requiring torchvision vs Mistral regex patch crashing on tokenizer internals." + "reason": "PIL image processors wrongly requiring torchvision is unrelated to Mistral regex patch crashes." }, { "left": "issue:45357", "right": "issue:45920", "accept": false, - "reason": "Separate regressions: incorrect visual encoder keys on save_pretrained vs wrong token IDs on tokenization." + "reason": "Qwen3.5 save_pretrained checkpoint key regression is unrelated to AutoTokenizer producing wrong token IDs across multiple models." }, { "left": "issue:43452", "right": "issue:44488", "accept": false, - "reason": "Both are loading-related, but one is a gguf_file/from_pretrained breakage and the other is a model-specific load failure; not enough evidence of the same bug." + "reason": "gguf_file loading breakage is a different root cause from the cjvt/sleng-bert load failure." }, { "left": "issue:43742", "right": "issue:44589", "accept": false, - "reason": "Different model load failures: key error for MobileLLM-125M vs missing Float8 storage object." + "reason": "MobileLLM-125M key error and missing Float8 storage are distinct loading/runtime failures." }, { "left": "issue:45405", "right": "issue:45600", "accept": false, - "reason": "One is a dependency version bump issue; the other is stale auto_mappings CI breakage." + "reason": "PEFT version bump and auto_mappings referencing removed configs are unrelated maintenance issues." }, { "left": "issue:28282", "right": "issue:29127", "accept": false, - "reason": "Different problems: PyTorch-not-found ImportError vs LayoutLMv3 box-content validation clarity." + "reason": "PyTorch missing ImportError and LayoutLMv3 box validation messaging are unrelated." }, { "left": "issue:43643", "right": "issue:45375", "accept": false, - "reason": "Both involve config field handling, but one is missing fields from trust_remote_code and the other is a specific Qwen3_5MoeVisionConfig field dropped by strict config parsing." + "reason": "trust_remote_code missing fields is a general config-return issue; Qwen3_5MoeVisionConfig missing deepstack_visual_indexes is a specific schema omission." }, { "left": "issue:43526", "right": "issue:45479", "accept": false, - "reason": "Different label/classification issues affecting different components and failure modes." + "reason": "BeitImageProcessorFast label reduction bug and single-label classification zero-loss regression are unrelated." }, { "left": "issue:43756", "right": "issue:45325", "accept": false, - "reason": "Different Qwen/Smollm3 RoPE issues with distinct symptoms and code paths." + "reason": "Both involve RoPE, but they affect different models and different calculations (layer dropping vs position-id scaling)." }, { "left": "issue:43531", "right": "issue:44821", "accept": false, - "reason": "Different domains entirely: Qwen3-MoE sliding_window behavior vs image processor URL loading." + "reason": "Qwen3-MoE sliding_window behavior is unrelated to AutoImageProcessor URL loading." }, { "left": "issue:44625", "right": "issue:44877", "accept": false, - "reason": "Separate config propagation bugs affecting different models and strictness logic." + "reason": "Num-label propagation bug and strict config blocking granite_speech loading are different configuration issues." }, { "left": "issue:43697", "right": "issue:45216", "accept": false, - "reason": "Different regressions: RTDetrV2 output changes in v5 vs Qwen3.5 save_pretrained checkpoint corruption." + "reason": "RTDetrV2 output differences in v5 and Qwen3.5 save_pretrained checkpoint corruption are separate regressions." + }, + { + "left": "issue:43582", + "right": "issue:44977", + "accept": false, + "reason": "Apple Silicon caching_allocator_warmup TypeError and Qwen3.5 flash-attention generation failure are unrelated." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet: the cluster is indeed heterogeneous, issue:45920 is the broadest open report, and the only accepted soft pair (issue:43493 / issue:43994) is plausibly the same SigLIP2 bug family. The other pairwise rejections are phrased conservatively and do not overclaim evidence. No PR pairs are present." + "accept": false, + "feedback": "The summary is mostly grounded, but it overstates one soft-edge pair: issue 43493 vs 43994 is marked deterministic_accept=false in the packet, so calling it the same underlying bug is too strong without more evidence. Please keep that verdict conservative (e.g. 'possibly related' or 'adjacent but unconfirmed') and avoid elevating the pair to accepted." }, "evaluator_used": true, - "retried": false, - "accepted_nontrivial_soft_edge": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -6381,13 +6418,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 20, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:24643", "issue:29127", @@ -6401,8 +6438,8 @@ "issue:43295", "issue:43452", "issue:43531", + "issue:43550", "issue:43577", - "issue:43582", "issue:43653", "issue:43697", "issue:43720", @@ -6440,237 +6477,251 @@ ], "soft_pairs": [ "issue:43720|issue:44898", - "issue:43582|issue:44977", "issue:44514|issue:45381", "issue:24643|issue:29127", "issue:41762|issue:44936", "issue:43452|issue:43577", - "issue:43452|issue:44843", "issue:44466|issue:45005", "issue:43906|issue:46202", "issue:45372|issue:46170", + "issue:43452|issue:44843", "issue:43819|issue:44811", "issue:43295|issue:45356", "issue:30064|issue:33666", "issue:44936|issue:45823", + "issue:44484|issue:44704", "issue:43653|issue:44361", "issue:44361|issue:44442", - "issue:44484|issue:44704", "issue:38175|issue:43295", "issue:44514|issue:45290", "issue:45542|issue:46178", "issue:43653|issue:44568", "issue:43577|issue:43697", + "issue:44493|issue:46050", "issue:29159|issue:33357", "issue:44843|issue:44991", - "issue:44493|issue:46050", "issue:44464|issue:44743", "issue:43653|issue:43927", + "issue:41720|issue:43906", "issue:43531|issue:43742", - "issue:41720|issue:43906" + "issue:43550|issue:43927", + "issue:44464|issue:44977", + "issue:43550|issue:44361" ], - "prepared_review_unit_hash": "44834958a2c1d5146b30be2d98f62cc6d4e720d3b0bc0260bc45f27e05923dec", + "prepared_review_unit_hash": "f7c42ba2f36d82ed2d43516914f8b848bbb4656ab991e142654151986afa6179", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36646, - "estimated_input_tokens": 9162, - "estimated_eval_tokens": 18580 + "soft_pair_count": 31, + "serialized_chars": 37237, + "estimated_input_tokens": 9310, + "estimated_eval_tokens": 18876 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is highly heterogeneous: the soft pairs mostly share only broad themes like loading, tokenization, or generation, but not the same concrete bug path. I would not merge any of the suggested soft edges.", - "confidence": 0.35, - "canonical_issue_reason": "issue:43452 is the broadest concrete regression here: it breaks both AutoTokenizer and AutoModelForCausalLM loading via gguf_file, making it the best anchor if a canonical issue must be chosen.", + "summary": "This cluster is mostly heterogeneous: the issues share broad Transformers/tokenizer/model-loading themes, but the soft pairs describe different concrete bugs and code paths. I would not merge any of the proposed soft edges; no PRs are present.", + "confidence": 0.74, + "canonical_issue_reason": "If a single issue must stand in for the cluster, issue:44843 is the best representative because it is a broad framework-level loading regression (AutoTokenizer/_patch_mistral_regex/offline mode) rather than a model-specific edge case.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43452 is the most representative standalone loader regression in this set; it is broader and more central than the narrower model-specific reports.", + "best_issue_reason": "Issue:44843 is the most suitable global representative because it concerns a central loading path with wider impact across models and environments, while most other issues are narrower model-specific failures.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43720", "right": "issue:44898", "accept": false, - "reason": "Different models and failures: BitNet packed-weight loading vs Perceiver image-size interpolation." - }, - { - "left": "issue:43582", - "right": "issue:44977", - "accept": false, - "reason": "Unrelated code paths: Apple Silicon allocator warmup TypeError vs Qwen3.5 flash-attention generation." + "reason": "Different failures in different code paths: BitNet accelerate loading/unpacking vs Perceiver image classification with interpolation." }, { "left": "issue:44514", "right": "issue:45381", "accept": false, - "reason": "Both are Qwen2.5-VL, but one is batched chat templating with padding=False and the other is video vision_position_ids; different bugs." + "reason": "Both are Qwen2.5-VL related, but one is chat-template batching/padding and the other is video vision_position_ids; not the same bug." }, { "left": "issue:24643", "right": "issue:29127", "accept": false, - "reason": "DeepSpeed weight-shape error is unrelated to LayoutLMv3 box-content validation and messaging." + "reason": "Unrelated errors: DeepSpeed training weight shape vs LayoutLMv3 box-content validation/message clarity." }, { "left": "issue:41762", "right": "issue:44936", "accept": false, - "reason": "Gemma3 ZeRO-3 load crash is not the same as trainer.evaluate() failing after train()." + "reason": "Different paths and symptoms: Gemma3 ZeRO-3 loading failure vs trainer.evaluate() breaking after train()." }, { "left": "issue:43452", "right": "issue:43577", "accept": false, - "reason": "Loader failure with gguf_file is unrelated to Blip2 dtype remaining float32." + "reason": "Tokenizer/AutoModel gguf_file loading breakage is unrelated to Blip2 dtype remaining float32 on load." }, { - "left": "issue:43452", - "right": "issue:44843", + "left": "issue:44466", + "right": "issue:45005", "accept": false, - "reason": "Both concern model loading, but gguf_file breakage and offline model_info() calls in tokenizer patching are different bugs." - }, - { - "left": "issue:44466", - "right": "issue:45005", - "accept": false, - "reason": "Both mention tied weights, but one is device-dependent serialization of lm_head.weight and the other is a broader translation-model tied-weights issue." + "reason": "Both mention tied weights, but one is device-dependent serialization and the other is translation-model behavior; too broad to treat as the same bug." }, { "left": "issue:43906", "right": "issue:46202", "accept": false, - "reason": "A reproduction of another issue is not the same as the Gemma pipeline roundtrip failure on '\u2581'." + "reason": "Different reproductions: issue 38071 replay vs Gemma pipeline roundtrip failure on a specific token." }, { "left": "issue:45372", "right": "issue:46170", "accept": false, - "reason": "Gemma 4 processor import failure from mistral_common is unrelated to Windows/ROCm torch.distributed import/type errors." + "reason": "Unrelated import/dependency failures: mistral_common ReasoningEffort for Gemma 4 vs Windows/ROCm torch.distributed issues." + }, + { + "left": "issue:43452", + "right": "issue:44843", + "accept": false, + "reason": "Both affect model loading, but gguf_file handling and offline model_info() calls are distinct failure modes." }, { "left": "issue:43819", "right": "issue:44811", "accept": false, - "reason": "DAC latent/forward mismatch is unrelated to Whisper batch_decode ignoring skip_special_tokens." + "reason": "DAC latents/STE mismatch is unrelated to Whisper batch_decode skipping special tokens." }, { "left": "issue:43295", "right": "issue:45356", "accept": false, - "reason": "Different regressions: processor.tokenizer access/images passthrough vs Kimi-K2.5 codec handling and warning behavior." + "reason": "Different regressions in processor/tokenizer behavior: processor.tokenizer access vs Kimi-K2.5 codec/regex warning handling." }, { "left": "issue:30064", "right": "issue:33666", "accept": false, - "reason": "Void segmentation-map processing and Qwen2-VL multi-GPU training are unrelated." + "reason": "Void segmentation maps in image processing vs Qwen2-VL multi-GPU training are unrelated." }, { "left": "issue:44936", "right": "issue:45823", "accept": false, - "reason": "Trainer lifecycle failure is not the same as Gemma4 device_map='auto' forward mismatch." + "reason": "Trainer state after training vs Gemma4 PLE device mismatch during forward are different bugs." + }, + { + "left": "issue:44484", + "right": "issue:44704", + "accept": false, + "reason": "Shard-size configuration question is unrelated to AutoProcessor kwargs not reaching cached_file." }, { "left": "issue:43653", "right": "issue:44361", "accept": false, - "reason": "Different tokenizer bugs affecting different models and code paths." + "reason": "BigBirdTokenizer special-token registration bug is unrelated to MLukeTokenizer attribute errors." }, { "left": "issue:44361", "right": "issue:44442", "accept": false, - "reason": "MLukeTokenizer task AttributeError is unrelated to AutoTokenizer failing to load FastSpeech2ConformerTokenizer." - }, - { - "left": "issue:44484", - "right": "issue:44704", - "accept": false, - "reason": "save_pretrained shard-size behavior is unrelated to AutoProcessor kwargs not reaching cached_file." + "reason": "Different tokenizer failures: MLukeTokenizer task handling vs AutoTokenizer loading FastSpeech2ConformerTokenizer." }, { "left": "issue:38175", "right": "issue:43295", "accept": false, - "reason": "Unexpected zero probabilities in SigLIP2 is unrelated to the processor/tokenizer regression." + "reason": "Siglip2 zero probabilities are unrelated to the processor.tokenizer regression in v4.57.5." }, { "left": "issue:44514", "right": "issue:45290", "accept": false, - "reason": "Both hit apply_chat_template, but one is batched padding=False and the other is assistant tool-call messages with no content." + "reason": "Same API surface, but different crash triggers: batched padding=False vs assistant tool-call messages without content." }, { "left": "issue:45542", "right": "issue:46178", "accept": false, - "reason": "Both are backend-installation related, but they fail in different backend-mapping scenarios and code paths." + "reason": "Different backend registration problems: missing TensorFlow backend vs offending tensorflow_text backend mapping." }, { "left": "issue:43653", "right": "issue:44568", "accept": false, - "reason": "BigBird special-token registration and mdeberta BOS/EOS insertion are different tokenizer bugs." + "reason": "Different tokenizer bugs affecting different models and token types." }, { "left": "issue:43577", "right": "issue:43697", "accept": false, - "reason": "Blip2 dtype persistence and RTDetrV2 output differences are unrelated model-specific regressions." + "reason": "Blip2 dtype persistence and RTDetrV2 output drift are unrelated." + }, + { + "left": "issue:44493", + "right": "issue:46050", + "accept": false, + "reason": "Position-id key warnings and sinq quantization save/load breakage are different serialization/compatibility issues." }, { "left": "issue:29159", "right": "issue:33357", "accept": false, - "reason": "Slow/fast tokenizer inconsistency is unrelated to the MacOS CLIP bus error." + "reason": "Slow-vs-fast tokenizer inconsistency is unrelated to the MacOS CLIP bus error." }, { "left": "issue:44843", "right": "issue:44991", "accept": false, - "reason": "Offline model_info() misuse in tokenizer patching is not the same as the est-roberta tokenizer loading failure." - }, - { - "left": "issue:44493", - "right": "issue:46050", - "accept": false, - "reason": "Position-id warnings are unrelated to quantized-model save/load breakage." + "reason": "Both are tokenizer-loading regressions, but one is offline model_info() patching and the other is a specific model tokenizer load failure." }, { "left": "issue:44464", "right": "issue:44743", "accept": false, - "reason": "Chunked generation with compiled forward and recurrent-state reset with cache are different generation bugs." + "reason": "Chunked generation with compiled forward is a different problem from recurrent states being reset with cache and seq_len>1." }, { "left": "issue:43653", "right": "issue:43927", "accept": false, - "reason": "BigBird special-token registration is unrelated to DiaConfig losing custom token IDs on save/load." + "reason": "BigBirdTokenizer special token registration and DiaConfig custom token ID persistence are unrelated." + }, + { + "left": "issue:41720", + "right": "issue:43906", + "accept": false, + "reason": "Qwen3 auto device mapping cudaErrorAssert is unrelated to an isolated reproduction of issue 38071." }, { "left": "issue:43531", "right": "issue:43742", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior is unrelated to MobileLLM-125M key errors during load." + "reason": "Qwen3-MoE sliding_window behavior and MobileLLM key errors are different code paths." }, { - "left": "issue:41720", - "right": "issue:43906", + "left": "issue:43550", + "right": "issue:43927", + "accept": false, + "reason": "torch.compile SDPA failure in Bamba-9B-v2 is unrelated to DiaConfig save/load token ID loss." + }, + { + "left": "issue:44464", + "right": "issue:44977", + "accept": false, + "reason": "Different generation regressions: compiled-forward chunking vs flash-attention behavior in Qwen3.5." + }, + { + "left": "issue:43550", + "right": "issue:44361", "accept": false, - "reason": "Qwen3 device-mapping cudaErrorAssert is unrelated to an isolated reproduction of another issue." + "reason": "Compilation/SDPA failure and tokenizer AttributeError are not the same underlying bug." } ] }, @@ -6686,13 +6737,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 21, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:28282", "issue:32090", @@ -6743,261 +6794,234 @@ "issue:46209" ], "soft_pairs": [ - "issue:43550|issue:43927", - "issue:44464|issue:44977", - "issue:43550|issue:44361", "issue:44977|issue:46209", "issue:43643|issue:44561", "issue:32090|issue:34567", "issue:43577|issue:45588", "issue:28282|issue:35141", - "issue:43504|issue:43550", + "issue:45491|issue:46002", "issue:32090|issue:33666", + "issue:43504|issue:43550", "issue:43504|issue:43720", - "issue:45491|issue:46002", "issue:44466|issue:46050", "issue:43504|issue:44610", "issue:41720|issue:44464", "issue:33666|issue:36010", - "issue:43742|issue:44821", "issue:43688|issue:45440", - "issue:44849|issue:46209", + "issue:43742|issue:44821", "issue:45823|issue:46050", "issue:28282|issue:32090", "issue:28282|issue:33357", "issue:43262|issue:43295", "issue:43720|issue:43927", - "issue:43452|issue:44451", + "issue:44849|issue:46209", "issue:43606|issue:43927", "issue:43994|issue:46192", + "issue:43452|issue:44451", "issue:33666|issue:36331", "issue:43756|issue:44913", "issue:43720|issue:44361", "issue:43525|issue:43572", - "issue:44849|issue:45357", "issue:43493|issue:43825", + "issue:44849|issue:45357", "issue:39692|issue:43493" ], - "prepared_review_unit_hash": "3d68c933e896489d49e2ea134bca4a930ea286446b7da563a2f9ae55acfdf011", + "prepared_review_unit_hash": "847c0d6dc4362ab8e392a7078022c42ee3bec7f7679e3105330f40260afe4b68", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 34, - "serialized_chars": 37451, - "estimated_input_tokens": 9363, - "estimated_eval_tokens": 18982 + "soft_pair_count": 31, + "serialized_chars": 36554, + "estimated_input_tokens": 9139, + "estimated_eval_tokens": 18534 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a grab bag of mostly unrelated transformer issues; the soft-similarity links look like false positives. The broadest and most canonical issue is the v5 regression breaking `trust_remote_code` models.", - "confidence": 0.71, - "canonical_issue_reason": "issue:44561 is the best representative because it describes a broad, version-level regression with a clear root cause and impact on multiple remote-code models.", + "summary": "This cluster is mostly a grab-bag of unrelated issue reports that only share superficial regression/loading/model keywords. I don\u2019t see any soft pair that looks like the same underlying bug or change, and there are no PRs in the set.", + "confidence": 0.96, + "canonical_issue_reason": "Issue 44561 is the broadest, clearest representative of the cluster\u2019s overall v5/trust_remote_code regression theme, but it is still not a duplicate hub for the other reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44561 has the clearest cross-cutting regression scope and is likely to be the most useful anchor for related reports.", + "best_issue_reason": "44561 is the most general and actionable issue in the set, with a concrete compatibility regression and a clearer scope than the other mostly model-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43550", - "right": "issue:43927", - "accept": false, - "reason": "Different bugs: one is torch.compile/SDPA failure in Bamba, the other is config persistence losing token IDs." - }, - { - "left": "issue:44464", - "right": "issue:44977", - "accept": false, - "reason": "Both are generation failures, but they involve different models and different failure modes (compiled forward vs flash-attention)." - }, - { - "left": "issue:43550", - "right": "issue:44361", - "accept": false, - "reason": "Unrelated failures: torch.compile SDPA vs tokenizer task AttributeError." - }, { "left": "issue:44977", "right": "issue:46209", "accept": false, - "reason": "Different root causes and subsystems: flash-attention generation vs FP8 inference on B200." + "reason": "Different bugs: Qwen3.5 flash-attention generation failure vs FP8 inference failure on B200." }, { "left": "issue:43643", "right": "issue:44561", "accept": false, - "reason": "Both affect remote-code workflows, but one is missing fields in AutoConfig output while the other is a v5 API removal breakage." + "reason": "Both touch trust_remote_code, but one is missing config fields and the other is removal of a helper breaking models; not the same code-path bug." }, { "left": "issue:32090", "right": "issue:34567", "accept": false, - "reason": "Trainer broadcast type error and TrainerState token counter bug are unrelated." + "reason": "Unrelated Trainer distributed broadcast TypeError vs TrainerState token-count tracking bug." }, { "left": "issue:43577", "right": "issue:45588", "accept": false, - "reason": "Blip2 dtype loading and flash_attention.py s_aux=None crash are different code paths." + "reason": "Different subsystems: Blip2 dtype propagation vs flash_attention handling of s_aux=None." }, { "left": "issue:28282", "right": "issue:35141", "accept": false, - "reason": "ImportError for missing PyTorch and embedding reinitialization during post_init are not the same bug." + "reason": "Missing PyTorch import and embedding reinitialization are unrelated failures." }, { - "left": "issue:43504", - "right": "issue:43550", + "left": "issue:45491", + "right": "issue:46002", "accept": false, - "reason": "Legacy-field loading bug in BEiT is unrelated to Bamba torch.compile/SDPA failure." + "reason": "Both mention NaNs, but one is Gemma3 sliding-window attention and the other is classification-head init/checkpointing." }, { "left": "issue:32090", "right": "issue:33666", "accept": false, - "reason": "Trainer broadcast error and multi-GPU training request are not the same issue." + "reason": "Distributed broadcast type error is not the same as multi-GPU training behavior for Qwen2-VL." }, { "left": "issue:43504", - "right": "issue:43720", + "right": "issue:43550", "accept": false, - "reason": "Different model-loading bugs: legacy preset field vs packed-weights unpacking during accelerate loading." + "reason": "One is a legacy-field load bug in Beit; the other is a torch.compile/SDPA failure in Bamba." }, { - "left": "issue:45491", - "right": "issue:46002", + "left": "issue:43504", + "right": "issue:43720", "accept": false, - "reason": "Both mention NaNs, but one is Gemma3 sliding-window attention and the other is classification head init/checkpoint behavior." + "reason": "Both are loading issues, but the concrete causes differ: legacy preset field vs unpacking packed BitNet weights." }, { "left": "issue:44466", "right": "issue:46050", "accept": false, - "reason": "Both involve save/load behavior, but one is tied-weight serialization and the other is quantized sinq persistence." + "reason": "Both concern save/load behavior, but one is tied-weight serialization and the other is sinq quantization persistence." }, { "left": "issue:43504", "right": "issue:44610", "accept": false, - "reason": "BEiT legacy field loading and OmDet-Turbo processor size mismatch are unrelated." + "reason": "Model preset loading bug vs processor/model input-size mismatch; different layers of the stack." }, { "left": "issue:41720", "right": "issue:44464", "accept": false, - "reason": "Qwen3 device-mapping assert and chunked generation with compiled forward are different concrete failures." + "reason": "CUDA assert under auto device mapping is unrelated to inconsistent chunked generation with compiled forward." }, { "left": "issue:33666", "right": "issue:36010", "accept": false, - "reason": "Multi-GPU training and GenerationMixin import breakage are unrelated." - }, - { - "left": "issue:43742", - "right": "issue:44821", - "accept": false, - "reason": "KeyError on MobileLLM loading vs AutoImageProcessor URL loading are different components and bugs." + "reason": "Multi-GPU training issue and GenerationMixin import failure are different problems." }, { "left": "issue:43688", "right": "issue:45440", "accept": false, - "reason": "Both concern model correctness, but they are different architectures and different implementation mismatches." + "reason": "Both involve MoE models, but one is auxiliary-loss normalization and the other is a native-vs-remote implementation divergence." }, { - "left": "issue:44849", - "right": "issue:46209", + "left": "issue:43742", + "right": "issue:44821", "accept": false, - "reason": "Qwen3.5 hidden-states bug and FP8 inference on B200 are unrelated." + "reason": "Key error loading MobileLLM-125M is unrelated to loading AutoImageProcessor from a URL." }, { "left": "issue:45823", "right": "issue:46050", "accept": false, - "reason": "Device-mapped Gemma4 PLE mismatch and quantized save/load breakage are different problems." + "reason": "Device mismatch with device_map=\"auto\" is not the same as quantized save/load breakage." }, { "left": "issue:28282", "right": "issue:32090", "accept": false, - "reason": "Missing PyTorch import error and Trainer GPU broadcast type error are unrelated." + "reason": "ImportError for PyTorch and GPU broadcast TypeError are unrelated." }, { "left": "issue:28282", "right": "issue:33357", "accept": false, - "reason": "ImportError for missing PyTorch and Mac bus error with CLIP are not the same underlying bug." + "reason": "PyTorch import missing vs Mac bus error in CLIP are different failure modes and code paths." }, { "left": "issue:43262", "right": "issue:43295", "accept": false, - "reason": "Both are regressions, but one is audio chat-template sampling-rate handling and the other is processor.tokenizer/image passthrough breakage." + "reason": "Audio chat-template sampling-rate default bug and processor.tokenizer/image-passing regression are separate issues." }, { "left": "issue:43720", "right": "issue:43927", "accept": false, - "reason": "Packed-weight loading failure and lost custom token IDs are separate loading bugs." + "reason": "BitNet loading with packed weights and DiaConfig custom-token-ID loss are unrelated save/load bugs." }, { - "left": "issue:43452", - "right": "issue:44451", + "left": "issue:44849", + "right": "issue:46209", "accept": false, - "reason": "gguf_file tokenizer/model loading and ScandiBERT load failure do not match on root cause." + "reason": "Qwen3.5 hidden-states bug is unrelated to FP8 inference failure on B200." }, { "left": "issue:43606", "right": "issue:43927", "accept": false, - "reason": "CPU offload device mismatch is unrelated to config save/load token ID loss." + "reason": "CPU-offload device mismatch and config token-ID loss do not share the same underlying bug." }, { "left": "issue:43994", "right": "issue:46192", "accept": false, - "reason": "SigLIP2 inference quality bug and return_tensors being ignored are different issues." + "reason": "SigLIP2 inference correctness issues are unrelated to return_tensors being ignored when text_kwargs is passed." + }, + { + "left": "issue:43452", + "right": "issue:44451", + "accept": false, + "reason": "GGUF-related tokenizer/model loading breakage is not the same as ScandiBERT loading failure." }, { "left": "issue:33666", "right": "issue:36331", "accept": false, - "reason": "Multi-GPU training request and unexpected Trainer compute_loss kwarg are distinct Trainer issues." + "reason": "Qwen2-VL multi-GPU training and CustomTrainer.compute_loss signature mismatch are different Trainer problems." }, { "left": "issue:43756", "right": "issue:44913", "accept": false, - "reason": "Smollm3 RoPE layer mismatch and GPTNeoX rotary_pct reload regression are similar in theme but not the same bug." + "reason": "Smollm3 RoPE-layer behavior and GPTNeoXConfig reload of rotary_pct are related only thematically, not as the same bug." }, { "left": "issue:43720", "right": "issue:44361", "accept": false, - "reason": "Packed-weight unpacking failure and MLukeTokenizer task AttributeError are unrelated." + "reason": "BitNet model loading and MLukeTokenizer task AttributeError are unrelated." }, { "left": "issue:43525", "right": "issue:43572", "accept": false, - "reason": "Both involve missing pad-token config fields after v5, but they affect different models/config attributes and are not one concrete bug." - }, - { - "left": "issue:44849", - "right": "issue:45357", - "accept": false, - "reason": "Hidden_states bug and incorrect visual-encoder key serialization are different Qwen3.5 issues." + "reason": "Similar symptom class, but different configs and missing fields: Llama4 pad_token_id vs StableLm pad_token_idx." }, { "left": "issue:43493", @@ -7009,7 +7033,7 @@ "left": "issue:39692", "right": "issue:43493", "accept": false, - "reason": "Doc example errors for SigLIP2 and JAX-vs-HF implementation discrepancy are separate issues." + "reason": "Documentation-example errors are not the same underlying bug as the SigLIP2 implementation discrepancy." } ] }, @@ -7025,13 +7049,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 22, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:33357", "issue:33666", @@ -7085,14 +7109,14 @@ "soft_pairs": [ "issue:44964|issue:45405", "issue:44479|issue:44625", - "issue:43688|issue:43749", "issue:43577|issue:44554", + "issue:43688|issue:43749", "issue:44568|issue:45812", "issue:43540|issue:44977", "issue:45200|issue:45229", "issue:44488|issue:44843", - "issue:43688|issue:45704", "issue:39692|issue:43295", + "issue:43688|issue:45704", "issue:44493|issue:46285", "issue:43756|issue:45083", "issue:43493|issue:43844", @@ -7101,227 +7125,227 @@ "issue:45229|issue:45823", "issue:33357|issue:33666", "issue:43334|issue:45759", - "issue:43927|issue:44361", "issue:43824|issue:44589", - "issue:43504|issue:43606", + "issue:43927|issue:44361", "issue:43278|issue:43844", + "issue:43504|issue:43606", "issue:44589|issue:45520", "issue:44625|issue:45759", "issue:43257|issue:43299", - "issue:43688|issue:45701", "issue:43572|issue:44291", "issue:43575|issue:45405", + "issue:43688|issue:45701", "issue:43606|issue:44112", "issue:43504|issue:43653" ], - "prepared_review_unit_hash": "1f891236216bc87b152740ad13992372c8691da729a147150612226dc7cc0798", + "prepared_review_unit_hash": "23323ac74f7c20ee6c7c4ef873357eb1972a8ee9d2e77effe6d23bad4ffe9ca4", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36896, - "estimated_input_tokens": 9224, - "estimated_eval_tokens": 18704 + "serialized_chars": 36905, + "estimated_input_tokens": 9227, + "estimated_eval_tokens": 18710 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The pairs are mostly adjacent regressions in similar subsystems, but none look like true duplicates: they differ in the concrete failure mode or code path. I\u2019d anchor the set on the Granite tokenizer regression issue as the cleanest concrete bug report.", - "confidence": 0.68, - "canonical_issue_reason": "issue:45812 is the cleanest representative concrete regression here: it is user-facing, reproducible, and narrowly describes a wrong-tokenization failure with a clear scope.", + "summary": "This cluster is not a duplicate set; it mixes many unrelated model-loading, tokenizer, config, dtype, and training/runtime regressions across different model families. None of the soft edges look like the same underlying bug/change, so all should be rejected.", + "confidence": 0.94, + "canonical_issue_reason": "If a single issue must represent the cluster, issue 43646 is the broadest umbrella ('Transformers 5.0.0 breaks custom model initialization'), but it still does not actually duplicate most of the others.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45812 is the strongest anchor because it is specific enough to triage and general enough to represent a real class of tokenizer regressions without being overly broad.", + "best_issue_reason": "Issue 43646 is the most generic version-wide regression in the set and best serves as an approximate umbrella, though the cluster is too heterogeneous for a true canonical duplicate.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44964", "right": "issue:45405", "accept": false, - "reason": "Model-loading regression vs a dependency version bump; different underlying problems." + "reason": "Different problems: Phi-4 multimodal loading regression vs an unreleased PEFT version bump." }, { "left": "issue:44479", "right": "issue:44625", "accept": false, - "reason": "Video-input regression vs num_labels propagation; not the same bug." + "reason": "One is a video-input regression in Qwen VL models; the other is num_labels propagation in Qwen3.5 config plumbing." }, { - "left": "issue:43688", - "right": "issue:43749", + "left": "issue:43577", + "right": "issue:44554", "accept": false, - "reason": "Aux-loss normalization vs FSDP CPU-efficient loading; unrelated failure modes." + "reason": "Blip2 dtype-casting bug is unrelated to the MPS attention correctness issue." }, { - "left": "issue:43577", - "right": "issue:44554", + "left": "issue:43688", + "right": "issue:43749", "accept": false, - "reason": "dtype propagation issue vs MPS attention correctness bug." + "reason": "Aux-loss normalization and FSDP CPU RAM efficient loading are different code paths and failure modes." }, { "left": "issue:44568", "right": "issue:45812", "accept": false, - "reason": "Tokenizer special-token handling for one model vs broad Granite token-ID regression; not the same code-path." + "reason": "Both involve tokenization, but one is BOS/EOS insertion for mdeberta while the other is Granite token ID regression; not the same bug." }, { "left": "issue:43540", "right": "issue:44977", "accept": false, - "reason": "Video processing ValueError vs flash-attention generation regression; different symptoms and fixes." + "reason": "Qwen3Omni video input processing is unrelated to Qwen3.5 flash-attention generation behavior." }, { "left": "issue:45200", "right": "issue:45229", "accept": false, - "reason": "Default token-type IDs vs multi-GPU inference OOM; not duplicate defects." + "reason": "Gemma4 token-type defaults for fine-tuning vs multi-GPU inference OOM are distinct issues." }, { "left": "issue:44488", "right": "issue:44843", "accept": false, - "reason": "Model loading failure vs offline model_info call in tokenizer patch; distinct issues." + "reason": "Model load failure for a specific repo is not the same as the offline model_info call regression." }, { - "left": "issue:43688", - "right": "issue:45704", + "left": "issue:39692", + "right": "issue:43295", "accept": false, - "reason": "Aux-loss normalization bug vs T5 RMSNorm memory leak; unrelated." + "reason": "Documentation/example mistakes are not the same as a runtime processor.tokenizer API regression." }, { - "left": "issue:39692", - "right": "issue:43295", + "left": "issue:43688", + "right": "issue:45704", "accept": false, - "reason": "Example/documentation errors vs processor.tokenizer API regression; related subsystem, different bug." + "reason": "Different model-specific bugs: auxiliary loss normalization vs T5 apex RMSNorm memory leak." }, { "left": "issue:44493", "right": "issue:46285", "accept": false, - "reason": "Unexpected position-id keys vs CLIP class structure change; different concrete problems." + "reason": "Unexpected position-id keys during loading is unrelated to CLIP text-model attribute structure changes." }, { "left": "issue:43756", "right": "issue:45083", "accept": false, - "reason": "RoPE-layer dropping bug vs feature-extract length helper behavior." + "reason": "Smollm3 RoPE-layer dropping and qwen3_omni_moe feature-length helper behavior are unrelated." }, { "left": "issue:43493", "right": "issue:43844", "accept": false, - "reason": "SigLIP2 implementation mismatch vs ZeRO-3 gradient growth; unrelated." + "reason": "HF-vs-JAX SigLIP2 discrepancy is not the same as the ZeRO-3 gradient blow-up bug." }, { "left": "issue:43334", "right": "issue:43572", "accept": false, - "reason": "Missing pad_token_id in Qwen3-VL vs missing pad_token_idx in StableLmConfig; same theme, different bug." + "reason": "Both mention missing token IDs, but they affect different configs/models and different fields." }, { "left": "issue:43646", "right": "issue:44466", "accept": false, - "reason": "Custom model init breakage vs inconsistent lm_head serialization; not the same issue." + "reason": "Custom model initialization breakage is separate from device-dependent lm_head serialization." }, { "left": "issue:45229", "right": "issue:45823", "accept": false, - "reason": "CUDA OOM vs device mismatch during forward; different failure modes." + "reason": "OOM in Gemma4 inference and PLE device mismatch during forward are different failures." }, { "left": "issue:33357", "right": "issue:33666", "accept": false, - "reason": "MacOS CLIP bus error vs Qwen2-VL multi-GPU training; unrelated." + "reason": "MacOS CLIP bus error and Qwen2-VL multi-GPU training are unrelated." }, { "left": "issue:43334", "right": "issue:45759", "accept": false, - "reason": "Missing pad_token_id vs composite-config unwrapping in from_config; different code paths." - }, - { - "left": "issue:43927", - "right": "issue:44361", - "accept": false, - "reason": "Config save/load token ID loss vs tokenizer AttributeError on tasks; different bugs." + "reason": "Missing pad_token_id in Qwen3-VL config is not the same as failing to unwrap text_config for composite multimodal configs." }, { "left": "issue:43824", "right": "issue:44589", "accept": false, - "reason": "ImportError for a model class vs float8 storage lookup failure; unrelated." + "reason": "ImportError for a missing Qwen2.5-VL class and a float8 storage lookup error are unrelated." }, { - "left": "issue:43504", - "right": "issue:43606", + "left": "issue:43927", + "right": "issue:44361", "accept": false, - "reason": "Legacy-field load failure vs CPU offload device mismatch; different model paths." + "reason": "Config save/load losing token IDs is a different bug from MLukeTokenizer AttributeError on tasks." }, { "left": "issue:43278", "right": "issue:43844", "accept": false, - "reason": "Embedding dtype drift vs abnormal gradient growth under ZeRO-3; not duplicates." + "reason": "Embedding dtype mismatch at eval time is unrelated to the ZeRO-3 gradient increase issue." + }, + { + "left": "issue:43504", + "right": "issue:43606", + "accept": false, + "reason": "Legacy config field loading and CPU-offload device mismatch are different classes of bugs." }, { "left": "issue:44589", "right": "issue:45520", "accept": false, - "reason": "Float8 storage deserialization error vs flash_attn KeyError on Python 3.13; different causes." + "reason": "Float8 storage resolution and flash_attn import key errors are different backend/import issues." }, { "left": "issue:44625", "right": "issue:45759", "accept": false, - "reason": "num_labels propagation vs text_config unwrapping; both config-related but not the same bug." + "reason": "Num_labels propagation and text_config unwrapping are related to config plumbing but not the same concrete failure." }, { "left": "issue:43257", "right": "issue:43299", "accept": false, - "reason": "Accelerate/DeepSpeed conversion issue vs Qwen3VL MoE loading regression; different code paths." - }, - { - "left": "issue:43688", - "right": "issue:45701", - "accept": false, - "reason": "Aux-loss normalization vs tokenization changes; unrelated." + "reason": "Accelerate+DeepSpeed weight conversion bug is distinct from the Qwen3VL MOE loading regression." }, { "left": "issue:43572", "right": "issue:44291", "accept": false, - "reason": "Missing config field vs init_empty_weights argument mismatch; different regressions." + "reason": "StableLm pad_token_idx loss and unexpected _is_hf_initialized argument are unrelated v5 regressions." }, { "left": "issue:43575", "right": "issue:45405", "accept": false, - "reason": "Tensor-parallel OOM vs unreleased dependency version bump; not duplicate." + "reason": "Tensor-parallel OOM for Qwen2-57B is unrelated to the PEFT version pin issue." + }, + { + "left": "issue:43688", + "right": "issue:45701", + "accept": false, + "reason": "Auxiliary loss normalization and tokenization changes are entirely different issues." }, { "left": "issue:43606", "right": "issue:44112", "accept": false, - "reason": "Runtime device mismatch vs stale CI test; one is a product bug, the other a test issue." + "reason": "Bark CPU-offload device mismatch and GraniteSpeech stale-device CI test are not the same bug." }, { "left": "issue:43504", "right": "issue:43653", "accept": false, - "reason": "Beit legacy-field load bug vs BigBirdTokenizer special-token decode bug; different models and failures." + "reason": "Beit legacy-field loading failure and BigBirdTokenizer special-token decoding are unrelated." } ] }, @@ -7337,13 +7361,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 23, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:15354", "issue:28218", @@ -7402,8 +7426,8 @@ "issue:43994|issue:44493", "issue:29942|issue:30333", "issue:43844|issue:44964", - "issue:28218|issue:34567", "issue:43688|issue:43873", + "issue:28218|issue:34567", "issue:29159|issue:32090", "issue:44521|issue:45290", "issue:44554|issue:45446", @@ -7417,28 +7441,28 @@ "issue:43232|issue:44849", "issue:44186|issue:46192", "issue:45406|issue:46121", - "issue:28218|issue:33357", "issue:45823|issue:46192", + "issue:28218|issue:33357", "issue:43493|issue:43701", "issue:43295|issue:43450", "issue:43295|issue:44610", "issue:39692|issue:46192", "issue:44554|issue:45356" ], - "prepared_review_unit_hash": "53db50bab84aeef608aa72e5aba0b3f01150d64547d79eca33ca8120fb6bdb7c", + "prepared_review_unit_hash": "2e5d353488406f25c9f83a07ce3d79bcb3b51cc0525213b47bf802c08eb2ced2", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36484, + "serialized_chars": 36481, "estimated_input_tokens": 9121, "estimated_eval_tokens": 18498 }, @@ -7447,186 +7471,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are a loose thematic cluster, not a true duplicate set. Most soft pairs are only related at a broad subsystem level (tokenizers, processors, generation, quantization, serving) and do not describe the same concrete bug or change.", - "confidence": 0.96, - "canonical_issue_reason": "No single canonical issue stands out: the cluster spans many unrelated failures. If one must be chosen as the closest representative, issue:43295 is the broadest regression-style report in core multimodal processor/tokenizer behavior.", + "summary": "This cluster is mostly a set of unrelated Transformers bugs that only share broad keywords like tokenizers, multimodal processors, generation, or serving. I found no soft pair that clearly refers to the same underlying defect, so no merge-worthy duplicates here.", + "confidence": 0.91, + "canonical_issue_reason": "issue:44521 is the best anchor if one must pick a canonical issue: it is active, broadly scoped to multimodal `apply_chat_template`, and has inbound references; however, it does not actually unify the rest of the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "Issue:43295 is the best representative among an otherwise heterogeneous set because it describes a user-facing regression in core processor/tokenizer handling and is broader than the highly specific reports.", + "best_issue_reason": "issue:44521 is the strongest standalone issue in this set because it is clearly described, currently open, and broad enough to represent a real multimodal template regression.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44792", "right": "issue:45405", "accept": false, - "reason": "Janus image-generation test failure and a PEFT version pin issue are unrelated bugs." + "reason": "Completely unrelated: a Janus test failure vs a PEFT version-pin release issue." }, { "left": "issue:44423", "right": "issue:44734", "accept": false, - "reason": "Both are serve-related crashes, but one is continuous batching multimodal input handling and the other is KV-cache continuation indexing in /v1/responses; different code paths." + "reason": "Both are serve crashes, but the endpoints and failure modes differ (`str.to` on multimodal batching vs KV-cache tensor indexing)." }, { "left": "issue:44464", "right": "issue:45584", "accept": false, - "reason": "Chunked generation with compiled forward and Whisper empty-transcription generation after token alignment are different generation bugs." + "reason": "Different subsystems and bugs: compiled-forward chunked generation vs Whisper empty-transcription generation." }, { "left": "issue:45588", "right": "issue:46050", "accept": false, - "reason": "Flash-attention sinkless-model None handling and sinq save/load quantization breakage are unrelated." + "reason": "Flash-attention `s_aux=None` crash is unrelated to sinq save/load quantization breakage." }, { "left": "issue:43994", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 bad outputs and position-id key errors are not the same underlying issue." + "reason": "One is a SigLIP2 model-output bug; the other is a generic key/position-id regression across models." }, { "left": "issue:29942", "right": "issue:30333", "accept": false, - "reason": "Flash Attention 2 test failures and MLFlow job state reporting are unrelated." + "reason": "FlashAttention test failures and MLflow job-finalization issues are unrelated." }, { "left": "issue:43844", "right": "issue:44964", "accept": false, - "reason": "ZeRO-3 gradient growth and Phi-4 multimodal loading are different training/loading problems." + "reason": "ZeRO-3 gradient explosion is unrelated to a Phi-4 multimodal load failure." }, { - "left": "issue:28218", - "right": "issue:34567", + "left": "issue:43688", + "right": "issue:43873", "accept": false, - "reason": "Tokenizer spacing after added tokens and TrainerState token counter updates are unrelated." + "reason": "Aux-loss normalization and quantization/offloading are different bug classes." }, { - "left": "issue:43688", - "right": "issue:43873", + "left": "issue:28218", + "right": "issue:34567", "accept": false, - "reason": "Aux-loss normalization in MoE models and offloading with quantization are different bugs." + "reason": "Tokenizer spacing behavior and TrainerState token counting are unrelated." }, { "left": "issue:29159", "right": "issue:32090", "accept": false, - "reason": "Slow/fast tokenizer inconsistency and NoneType GPU broadcast errors are unrelated." + "reason": "Tokenizer parity issue vs Trainer NoneType broadcast error; no shared code path." }, { "left": "issue:44521", "right": "issue:45290", "accept": false, - "reason": "Both involve apply_chat_template, but one is zero assistant masks for multimodal inputs and the other is a crash on tool-call messages with no content." + "reason": "Both involve chat templates, but one is zero assistant masks for multimodal inputs while the other is a crash on tool-call assistant messages." }, { "left": "issue:44554", "right": "issue:45446", "accept": false, - "reason": "MPS attention correctness and a PyTorch version check in flex_attention are unrelated." + "reason": "MPS attention correctness and a flex_attention PyTorch version check are unrelated." }, { "left": "issue:15354", "right": "issue:38175", "accept": false, - "reason": "TorchScript export failure for ViT and SigLIP2 zero probabilities are unrelated." + "reason": "TorchScript export failure on ViT is unrelated to SigLIP2 zero-probability outputs." }, { "left": "issue:45084", "right": "issue:45520", "accept": false, - "reason": "Template-node compilation TypeError and flash_attn import KeyError on Python 3.13 are unrelated." + "reason": "Template-node compilation error and a Python 3.13 flash_attn import KeyError are distinct." }, { "left": "issue:43540", "right": "issue:45200", "accept": false, - "reason": "Qwen3OmniMoe video processing failure and Gemma4 text-only mm_token_type_ids handling are different multimodal issues." + "reason": "Video-input processing error in Qwen3OmniMoe is unrelated to Gemma4 text-only mm_token_type_ids handling." }, { "left": "issue:43873", "right": "issue:46050", "accept": false, - "reason": "Quantization offloading behavior and sinq save/load corruption are different quantization problems." + "reason": "Both touch quantization/offload, but one is offloading not working and the other is save/load corruption after sinq quantization." }, { "left": "issue:45200", "right": "issue:45823", "accept": false, - "reason": "Gemma4 text-only token-type defaults and device-map forward device mismatch are separate bugs." + "reason": "Gemma4 input-defaulting bug and Gemma4 device-map forward mismatch are different failures." }, { "left": "issue:43278", "right": "issue:43701", "accept": false, - "reason": "BF16-to-FP32 embedding dtype drift and resume-from-checkpoint key mismatch are unrelated." + "reason": "Embedding dtype drift and resume checkpoint key mismatch do not share a concrete bug." }, { "left": "issue:45072", "right": "issue:45198", "accept": false, - "reason": "bfloat16 dtype mismatches and Wav2Vec2 save/tokenization failures are unrelated." + "reason": "Bfloat16 inference dtype mismatches and Wav2Vec2 save/tokenization failure are unrelated." }, { "left": "issue:43232", "right": "issue:44849", "accept": false, - "reason": "Generation kwargs after sync_gpus and Qwen3.5 hidden-states output are different generation bugs." + "reason": "Generation kwarg update after sync_gpus and Qwen3.5 output_hidden_states bug are separate generation issues." }, { "left": "issue:44186", "right": "issue:46192", "accept": false, - "reason": "LayoutLMv2 NER/padding crashes and return_tensors being ignored with text_kwargs are different tokenizer API bugs." + "reason": "LayoutLMv2 NER/padding crash and `return_tensors` being ignored are different tokenizer bugs." }, { "left": "issue:45406", "right": "issue:46121", "accept": false, - "reason": "Gemma4Processor _tokenizer attribute crash and rope-parameter list handling are unrelated." + "reason": "Gemma4 serve processor missing `_tokenizer` is unrelated to rope-parameter JSON list handling." }, { - "left": "issue:28218", - "right": "issue:33357", + "left": "issue:45823", + "right": "issue:46192", "accept": false, - "reason": "Tokenizer spacing behavior and a MacOS bus error on CLIP are unrelated." + "reason": "Device-mapping mismatch in Gemma4 and `return_tensors` precedence are not the same defect." }, { - "left": "issue:45823", - "right": "issue:46192", + "left": "issue:28218", + "right": "issue:33357", "accept": false, - "reason": "Gemma4 device placement and return_tensors handling are unrelated." + "reason": "Added-token spacing and CLIP bus error on MacOS are unrelated." }, { "left": "issue:43493", "right": "issue:43701", "accept": false, - "reason": "SigLIP2 implementation discrepancy and checkpoint resume key mismatch are unrelated." + "reason": "SigLIP2 implementation discrepancy is unrelated to checkpoint key mismatches." }, { "left": "issue:43295", "right": "issue:43450", "accept": false, - "reason": "Processor.tokenizer regression and batched video shape errors are different multimodal processor bugs." + "reason": "Processor/tokenizer regression and batched video shape bug are different multimodal issues." }, { "left": "issue:43295", "right": "issue:44610", "accept": false, - "reason": "Processor/tokenizer regression and OmDet-Turbo image-size mismatch are not the same bug." + "reason": "Custom processor/tokenizer regression and OmDet-Turbo input-size mismatch are unrelated." }, { "left": "issue:39692", "right": "issue:46192", "accept": false, - "reason": "SigLIP2 docs/model mismatch and return_tensors handling are unrelated." + "reason": "SigLIP2 docs/model mismatch and ignored `return_tensors` are separate problems." }, { "left": "issue:44554", "right": "issue:45356", "accept": false, - "reason": "MPS attention correctness and Kimi-K2.5 tokenizer codec regression are unrelated." + "reason": "Attention correctness on MPS and Kimi-K2.5 tokenizer regression are unrelated." } ] }, @@ -7642,13 +7666,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 24, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:37428", "issue:38175", @@ -7702,9 +7726,9 @@ "issue:45381|issue:46209", "issue:43295|issue:43873", "issue:43572|issue:44987", + "issue:44898|issue:45412", "issue:43950|issue:46285", "issue:43572|issue:43957", - "issue:44898|issue:45412", "issue:43550|issue:43844", "issue:44466|issue:45127", "issue:45200|issue:45325", @@ -7718,32 +7742,32 @@ "issue:43526|issue:46192", "issue:38175|issue:43493", "issue:43653|issue:44112", - "issue:44442|issue:45198", "issue:42915|issue:43824", + "issue:44442|issue:45198", "issue:43295|issue:43493", "issue:38175|issue:45412", "issue:43262|issue:43450", "issue:44514|issue:44521", - "issue:45759|issue:46285", "issue:37428|issue:43824", "issue:45520|issue:46170", + "issue:45759|issue:46285", "issue:41720|issue:46209", "issue:44561|issue:46234" ], - "prepared_review_unit_hash": "cff56ff462b9df5133188fc6dd6ed52d3350c4acc7783af1083faadbad1a7f0f", + "prepared_review_unit_hash": "f64c2b3d7b39f048de2e04fe73e63c9cfe16a7cc95ad95b1c0f5b94652cfd4aa", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36452, + "serialized_chars": 36449, "estimated_input_tokens": 9113, "estimated_eval_tokens": 18482 }, @@ -7752,198 +7776,198 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated Transformers regression reports. The only likely duplicate-like pair is the Qwen3-VL / Llama4 missing `pad_token_id` config failure; the rest are separate bugs across different models/subsystems. No PRs are present.", - "confidence": 0.72, - "canonical_issue_reason": "issue:43334 is the best anchor for the only plausible duplicate pair: it matches 43525 in the same `pad_token_id`-missing model-load failure pattern on new config classes.", + "summary": "The cluster is mostly noisy and spans unrelated Transformers regressions. The only soft edge that looks like a true duplicate is the pair of pad_token_id AttributeErrors in different config classes; the rest are distinct bugs affecting different models, modalities, or code paths.", + "confidence": 0.37, + "canonical_issue_reason": "issue:43334 is the clearest representative of the shared regression: a load-time AttributeError caused by missing pad_token_id on model/config objects. It is more general than the Llama4-specific report and best captures the underlying config API breakage.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43334 is the clearest representative of the one near-duplicate theme and is described as a concrete load failure with a specific exception path.", + "best_issue_reason": "issue:43334 is the best global issue because it describes a concrete, reproducible regression in a widely shared config access path and is broad enough to cover the closely related Llama4 report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45381", "right": "issue:46209", "accept": false, - "reason": "Different failures: Qwen2.5-VL vision position ids vs FP8 inference OOM on B200." + "reason": "Different bugs: one is wrong vision_position_ids for Qwen2.5-VL video/still image handling, the other is FP8 inference failure on B200 GPUs." }, { "left": "issue:43295", "right": "issue:43873", "accept": false, - "reason": "Processor/tokenizer regression vs quantization offloading behavior; unrelated code paths." + "reason": "Unrelated symptoms and code paths: processor.tokenizer regression vs quantization/offloading behavior." }, { "left": "issue:43572", "right": "issue:44987", "accept": false, - "reason": "StableLm `pad_token_idx` issue is unrelated to a separate model-loading failure for `physical-intelligence/fast`." + "reason": "Both are load failures, but they point to different root causes and different models; not the same bug." + }, + { + "left": "issue:44898", + "right": "issue:45412", + "accept": false, + "reason": "Perceiver positional-encoding failure and RT-DETR memory release issue are unrelated." }, { "left": "issue:43950", "right": "issue:46285", "accept": false, - "reason": "Non-persistent buffer corruption on load is unrelated to CLIP text model structure inconsistency." + "reason": "Both involve model structure/loading changes in v5, but the specific defects differ: non-persistent buffer corruption vs CLIP text_model structure mismatch." }, { "left": "issue:43572", "right": "issue:43957", "accept": false, - "reason": "Different regressions: missing pad token field vs meta-device loading breakage." - }, - { - "left": "issue:44898", - "right": "issue:45412", - "accept": false, - "reason": "Perceiver interpolation failure and RT-DETR memory release bug are unrelated." + "reason": "Different classes of regressions: missing pad_token_idx in a config vs meta-device loading breakage." }, { "left": "issue:43550", "right": "issue:43844", "accept": false, - "reason": "torch.compile SDPA failure for Bamba and DeepSpeed gradient growth are distinct issues." + "reason": "Bamba torch.compile/SDPA failure is unrelated to DeepSpeed ZeRO-3 gradient growth." }, { "left": "issue:44466", "right": "issue:45127", "accept": false, - "reason": "Serialization of tied weights vs LoRA merge collapse; similar theme but not the same bug." + "reason": "Both mention tied weights/serialization-adjacent behavior, but one is inconsistent lm_head.weight saving and the other is post-LoRA model collapse with extended vocab." }, { "left": "issue:45200", "right": "issue:45325", "accept": false, - "reason": "Different multimodal bugs: missing default mm token ids vs wrong temporal rope scaling." + "reason": "Both are Qwen2.5-VL multimodal issues, but one is missing mm_token_type_ids defaults while the other is incorrect rope/temporal position scaling." }, { "left": "issue:43450", "right": "issue:43575", "accept": false, - "reason": "Video processor batch-shape bug and tensor-parallel OOM are unrelated." + "reason": "Video processor batch-shape bug and Qwen2-57B tensor-parallel OOM are unrelated." }, { "left": "issue:45520", "right": "issue:46178", "accept": false, - "reason": "Both are import/backend errors, but for different optional dependencies and different failure modes." + "reason": "Different import/backend issues: flash_attn lookup on Python 3.13 vs tensorflow_text backend mapping." }, { "left": "issue:43334", "right": "issue:43525", "accept": true, - "reason": "Both report the same `AttributeError`/missing `pad_token_id` regression on new config classes during model loading." + "reason": "Same underlying load-time regression: both configs lack pad_token_id and crash when Transformers code expects that attribute." }, { "left": "issue:37428", "right": "issue:43957", "accept": false, - "reason": "Flash-attention import error and meta-device loading breakage are separate regressions." + "reason": "Both are loading/import issues, but one is a missing helper import and the other is meta-device model loading breakage." }, { "left": "issue:42490", "right": "issue:43696", "accept": false, - "reason": "Trailing-slash import warning is unrelated to GPT-oss OOM." + "reason": "Trailing-slash local path warning and GPT-oss CUDA OOM are unrelated." }, { "left": "issue:43824", "right": "issue:43975", "accept": false, - "reason": "ImportError for `Qwen2_5_VLForConditionalGeneration` is unrelated to detokenization behavior in deepseek-coder." + "reason": "ImportError for Qwen2_5_VL class and DeepSeek detokenization bug are different defects." }, { "left": "issue:43278", "right": "issue:43856", "accept": false, - "reason": "Embedding dtype mismatch and Qwen3 MoE memory inefficiency are different problems." + "reason": "Training/eval dtype drift is unrelated to Qwen3 MoE memory inefficiency." }, { "left": "issue:43526", "right": "issue:46192", "accept": false, - "reason": "Beit `reduce_labels` bug is unrelated to `return_tensors` being ignored with `text_kwargs`." + "reason": "BeitImageProcessorFast label reduction bug and return_tensors being ignored are distinct processor issues." }, { "left": "issue:38175", "right": "issue:43493", "accept": false, - "reason": "Siglip2 zero-probabilities issue is not the same as the JAX/HF implementation discrepancy." + "reason": "SigLIP2 zero-probabilities and HF-vs-JAX discrepancy are related to the same model family but not the same concrete bug." }, { "left": "issue:43653", "right": "issue:44112", "accept": false, - "reason": "Tokenizer special-token registration and GraniteSpeech CI test failure are unrelated." + "reason": "BigBirdTokenizer special-token registration and GraniteSpeech CI test staleness are unrelated." }, { - "left": "issue:44442", - "right": "issue:45198", + "left": "issue:42915", + "right": "issue:43824", "accept": false, - "reason": "AutoTokenizer load failure and Wav2Vec2 save/tokenization failure affect different tokenization paths." + "reason": "Qwen3Moe FP8 failure and Qwen2_5_VL import error are different model-specific issues." }, { - "left": "issue:42915", - "right": "issue:43824", + "left": "issue:44442", + "right": "issue:45198", "accept": false, - "reason": "Qwen3Moe FP8 failure and Qwen2.5-VL import error are unrelated." + "reason": "Tokenizer load failure and Wav2Vec2 save_pretrained/tokenization failure are separate bugs." }, { "left": "issue:43295", "right": "issue:43493", "accept": false, - "reason": "Processor/tokenizer regression and Siglip2 implementation discrepancy are unrelated." + "reason": "Processor.tokenizer regression and SigLIP2 implementation discrepancy do not share the same failing code path." }, { "left": "issue:38175", "right": "issue:45412", "accept": false, - "reason": "Siglip2 output issue is unrelated to RT-DETR memory cleanup." + "reason": "SigLIP2 probability issue and RT-DETR memory cleanup bug are unrelated." }, { "left": "issue:43262", "right": "issue:43450", "accept": false, - "reason": "Audio chat-template sampling-rate default and video batch-shape bug are unrelated." + "reason": "Audio chat-template sampling-rate defaulting and video batch-shape handling are different processor bugs." }, { "left": "issue:44514", "right": "issue:44521", "accept": false, - "reason": "Same model family, but one is a batching crash and the other is wrong assistant masks; not the same concrete bug." - }, - { - "left": "issue:45759", - "right": "issue:46285", - "accept": false, - "reason": "Composite Qwen config unwrapping and CLIP structure inconsistency are different model-loading changes." + "reason": "Both involve Qwen2.5-VL apply_chat_template, but one is a batched-input crash and the other is incorrect assistant_masks output; not clearly the same bug." }, { "left": "issue:37428", "right": "issue:43824", "accept": false, - "reason": "Missing flash-attention helper import and missing Qwen2.5-VL export are unrelated import errors." + "reason": "Different import errors for different symbols and root causes." }, { "left": "issue:45520", "right": "issue:46170", "accept": false, - "reason": "Python 3.13 flash_attn backend lookup and Windows ROCm torch.distributed import failure are separate platform-specific issues." + "reason": "flash_attn import lookup on Python 3.13 and Windows ROCm torch.distributed dependency issues are unrelated." + }, + { + "left": "issue:45759", + "right": "issue:46285", + "accept": false, + "reason": "Composite Qwen multimodal config unwrapping and CLIP text_model structure mismatch are different regressions." }, { "left": "issue:41720", "right": "issue:46209", "accept": false, - "reason": "Qwen3 auto device map assert and FP8 inference OOM are different runtime failures." + "reason": "Qwen3 auto-device-map cudaErrorAssert and FP8 inference failure on B200 are separate hardware/runtime problems." }, { "left": "issue:44561", "right": "issue:46234", "accept": false, - "reason": "Missing `is_torch_fx_available` and missing `is_falcon_mamba_ssm_available` are different import-utils symbols and regressions." + "reason": "Removal of is_torch_fx_available and missing is_falcon_mamba_ssm_available are different API compatibility issues." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The one accepted duplicate pair (issue:43334 / issue:43525) is well supported by the shared missing-`pad_token_id` load failure pattern, and the rejected soft edges are generally reasoned from the issue titles without overclaiming. No PRs are present, so that part is also consistent." + "feedback": "Grounded and conservative overall. The only accepted soft edge is well supported by the titles: both issues are load-time `pad_token_id` AttributeErrors on config objects, so treating them as the same regression is reasonable. The remaining pairings are phrased cautiously as distinct bugs and do not overstate evidence from the packet." }, "evaluator_used": true, "retried": false, @@ -7956,13 +7980,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 25, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:28218", "issue:29942", @@ -8023,270 +8047,277 @@ "issue:44857|issue:45593", "issue:43606|issue:45083", "issue:44792|issue:45563", - "issue:37428|issue:43425", - "issue:45127|issue:45245", "issue:44448|issue:44743", + "issue:37428|issue:43425", "issue:45200|issue:45588", - "issue:43335|issue:43653", + "issue:45127|issue:45245", "issue:43611|issue:43931", "issue:45290|issue:45381", + "issue:43335|issue:43653", "issue:44855|issue:45083", "issue:43550|issue:44112", - "issue:44206|issue:45823", "issue:43819|issue:44493", - "issue:43335|issue:43504", + "issue:44206|issue:45823", "issue:29942|issue:32090", + "issue:43335|issue:43504", "issue:45588|issue:45859", "issue:37428|issue:44351", "issue:43278|issue:43825", "issue:45459|issue:46234", "issue:39692|issue:43526", - "issue:43335|issue:45083", "issue:43388|issue:43526", + "issue:43335|issue:45083", "issue:45310|issue:45478", "issue:39692|issue:44493", "issue:28218|issue:35141", "issue:43720|issue:44112", - "issue:43335|issue:44112" + "issue:43335|issue:44112", + "issue:43335|issue:45593" ], - "prepared_review_unit_hash": "887832fce16cddb7411f7665c4bf498212b0a7e084f1fb30b352111a682354ca", + "prepared_review_unit_hash": "84da5fe9a3f434dff1060f313eb4dc4deffa6ad83056961cf96a7da8a383d3c0", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 34, - "serialized_chars": 38127, - "estimated_input_tokens": 9532, - "estimated_eval_tokens": 19320 + "soft_pair_count": 35, + "serialized_chars": 38417, + "estimated_input_tokens": 9605, + "estimated_eval_tokens": 19466 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous; the only clear duplicate pair is the Qwen3.5 Moe `from_pretrained` error. Everything else looks like separate bugs, regressions, or docs issues in different code paths.", - "confidence": 0.95, - "canonical_issue_reason": "Issue 45310 is the strongest canonical candidate because it matches 45478 on the same Qwen3.5 Moe `from_pretrained` error title, has an inbound reference, and has a slightly longer body with more context.", + "summary": "Most soft pairs are only loosely related by subsystem or release context. The only clear duplicate is the pair of Qwen3.5 MoE `from_pretrained` issues (45310 and 45478).", + "confidence": 0.86, + "canonical_issue_reason": "Issue 45310 is the clearest canonical issue because it shares the exact same title as 45478, appears earlier, and already has an inbound reference; 45478 reads like a later duplicate report of the same Qwen3.5 MoE `from_pretrained` failure.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45310 is the best representative issue in this cluster because it is the confirmed duplicate target for 45478, while also providing slightly more context and an inbound reference.", + "best_issue_reason": "Issue 45310 is the best overall representative for the duplicated Qwen3.5 MoE `from_pretrained` regression because it is the earlier report and already has an inbound reference.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44610", "right": "issue:44857", "accept": false, - "reason": "Different models and failures: OmDet-Turbo processor input-size mismatch vs LwDetrImageLoss AMP crash." + "reason": "Different failures: processor/model input-size mismatch vs AMP float16 crash in a loss module." }, { "left": "issue:43819", "right": "issue:44112", "accept": false, - "reason": "Both are model/runtime bugs, but the DAC STE mismatch and GraniteSpeech stale device override are unrelated code paths." + "reason": "Both are runtime bugs, but they affect different code paths and symptoms." }, { "left": "issue:43827", "right": "issue:43901", "accept": false, - "reason": "Both are docs/pipeline related, but one is about removed `pipeline()` usage and the other about `return_all_scores` docs for text classification." + "reason": "Both involve docs around v5 changes, but they are different pipeline/documentation issues." }, { "left": "issue:43819", "right": "issue:45083", "accept": false, - "reason": "Unrelated subsystems: DAC latent reconstruction vs qwen3_omni_moe feature-length helper behavior." + "reason": "Unrelated components: DAC latent decoding vs a qwen3_omni_moe helper function." }, { "left": "issue:44206", "right": "issue:45584", "accept": false, - "reason": "Both involve audio-ish workflows, but they describe distinct regressions: unsupported `center` arg vs empty transcription after special-token alignment." + "reason": "Different bugs: unsupported `center` argument in feature extraction vs empty-transcription generation failure." }, { "left": "issue:43335", "right": "issue:43550", "accept": false, - "reason": "Different models and failures: SwitchTransformers sparse-layer construction vs Bamba torch.compile SDPA behavior." + "reason": "Different models and failures: SwitchTransformers sparse-layer construction vs Bamba SDPA/torch.compile." }, { "left": "issue:44857", "right": "issue:45593", "accept": false, - "reason": "Different detection/training behaviors: AMP crash in LwDetrImageLoss vs missing auxiliary losses in D-FINE." + "reason": "Both are training-loss related, but one is an AMP crash and the other is missing auxiliary losses." }, { "left": "issue:43606", "right": "issue:45083", "accept": false, - "reason": "CPU offload device mismatch in Bark is unrelated to qwen3_omni_moe feature-length calculations." + "reason": "Different subsystems and root causes: CPU offload device mismatch vs qwen3_omni_moe output-length helper behavior." }, { "left": "issue:44792", "right": "issue:45563", "accept": false, - "reason": "A failing Janus image-generation test is not the same bug as a stale `num_return_sequences` warning in paged generate()." - }, - { - "left": "issue:37428", - "right": "issue:43425", - "accept": false, - "reason": "ImportError for a flash-attention helper and Torch 2.10 incompatibility are different release/runtime issues." + "reason": "A failed image-generation test and a stale warning in generation are unrelated." }, { - "left": "issue:45127", - "right": "issue:45245", + "left": "issue:44448", + "right": "issue:44743", "accept": false, - "reason": "LoRA merge/collapsed outputs and the 2^24 category limit are unrelated model/data constraints." + "reason": "Different models and code paths: Pegasus output regression vs Qwen recurrent-state reset in cache handling." }, { - "left": "issue:44448", - "right": "issue:44743", + "left": "issue:37428", + "right": "issue:43425", "accept": false, - "reason": "Different code paths: Pegasus v4/v5 output drift vs recurrent state reset in modular_qwen3_5." + "reason": "Import error for a missing helper is not the same as general Torch 2.10 incompatibility." }, { "left": "issue:45200", "right": "issue:45588", "accept": false, - "reason": "Gemma4 text-only token-type defaults and flash_attention `s_aux=None` handling are separate bugs." + "reason": "Gemma4 token-type-id issue and a flash-attention `s_aux=None` crash are separate problems." }, { - "left": "issue:43335", - "right": "issue:43653", + "left": "issue:45127", + "right": "issue:45245", "accept": false, - "reason": "SwitchTransformers sparse-layer creation and BigBirdTokenizer special-token registration are unrelated." + "reason": "LoRA/vocab collapse and category-count overflow are unrelated bugs." }, { "left": "issue:43611", "right": "issue:43931", "accept": false, - "reason": "Base-model-prefix loading regression and Qwen3-VL weight-shape mismatch are different model-loading failures." + "reason": "Both are loading issues, but they involve different models and mismatch mechanisms." }, { "left": "issue:45290", "right": "issue:45381", "accept": false, - "reason": "Chat-template tool-call handling and Qwen2.5-VL vision position IDs are separate issues." + "reason": "Chat-template tool-call handling and Qwen2.5-VL video position IDs are distinct failures." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "Different components: SwitchTransformers config vs BigBird tokenizer special-token registration." }, { "left": "issue:44855", "right": "issue:45083", "accept": false, - "reason": "Python 3.13 torch.jit indentation parsing and qwen3_omni_moe feature-length helper behavior are unrelated." + "reason": "Python 3.13 import parsing error is unrelated to qwen3_omni_moe helper behavior." }, { "left": "issue:43550", "right": "issue:44112", "accept": false, - "reason": "Bamba torch.compile SDPA failure is not the same as a stale GraniteSpeech CI device override test." - }, - { - "left": "issue:44206", - "right": "issue:45823", - "accept": false, - "reason": "Both are device/feature regressions, but one is a feature-extractor arg issue and the other is a Gemma4 PLE forward device mismatch." + "reason": "A compile/SDPA model bug and a CI stale-device-test failure are not the same issue." }, { "left": "issue:43819", "right": "issue:44493", "accept": false, - "reason": "DAC latent mismatch and position-id key warnings are unrelated model internals." + "reason": "Different symptoms and code paths: DAC latent mismatch vs position-id key warnings." }, { - "left": "issue:43335", - "right": "issue:43504", + "left": "issue:44206", + "right": "issue:45823", "accept": false, - "reason": "SwitchTransformers sparse-layer bug and Beit pretrained preset loading bug are different model families and failures." + "reason": "Different models and failures: feature-extractor crash vs Gemma4 PLE device mismatch." }, { "left": "issue:29942", "right": "issue:32090", "accept": false, - "reason": "Flash Attention 2 test failures and Trainer `_gpu_broadcast_one` NoneType errors do not describe the same underlying bug." + "reason": "Flash Attention test failures and Trainer `_gpu_broadcast_one` TypeError are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:43504", + "accept": false, + "reason": "Different model families and bugs; only broad ML/model-loading similarity." }, { "left": "issue:45588", "right": "issue:45859", "accept": false, - "reason": "Flash-attention sinkless-model crash and Qwen3.5 Moe CPU-offload incompatibility are different runtime problems." + "reason": "Flash-attention sinkless-model crash and Qwen rotary embedding CPU-offload incompatibility are separate issues." }, { "left": "issue:37428", "right": "issue:44351", "accept": false, - "reason": "Missing `_flash_supports_window_size` import and missing `HybridCache` import are separate API/export regressions." + "reason": "Two import errors, but for different symbols and different causes." }, { "left": "issue:43278", "right": "issue:43825", "accept": false, - "reason": "Embedding dtype drift between train/eval and a pipeline v5 translation message regression are unrelated." + "reason": "Training/eval dtype drift is unrelated to pipeline v5 error messaging." }, { "left": "issue:45459", "right": "issue:46234", "accept": false, - "reason": "Tokenizer protobuf error masking and a missing `is_falcon_mamba_ssm_available` utility are separate import-utils issues." + "reason": "Tokenizer protobuf-error masking and missing import_utils availability helper are distinct API issues." }, { "left": "issue:39692", "right": "issue:43526", "accept": false, - "reason": "SigLIP2 docs errors and BeitImageProcessorFast `reduce_labels` behavior are different concerns." + "reason": "Docs-example errors are unrelated to Beit fast processor label reduction." }, { - "left": "issue:43335", - "right": "issue:45083", + "left": "issue:43388", + "right": "issue:43526", "accept": false, - "reason": "SwitchTransformers sparse-layer creation is unrelated to qwen3_omni_moe feature-length helper behavior." + "reason": "Both affect labels, but one is metric gathering and the other is image processor label reduction." }, { - "left": "issue:43388", - "right": "issue:43526", + "left": "issue:43335", + "right": "issue:45083", "accept": false, - "reason": "Metric gathering label truncation and BeitImageProcessorFast label reduction are not the same bug." + "reason": "No concrete overlap beyond both being internal model helpers." }, { "left": "issue:45310", "right": "issue:45478", "accept": true, - "reason": "Exact same Qwen3.5 Moe `from_pretrained` error on transformers>=5.4.0; 45478 is a duplicate report." + "reason": "Same reported bug: identical Qwen3.5 MoE `from_pretrained` failure title, with 45478 looking like a later duplicate report." }, { "left": "issue:39692", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 doc issues and position-id key warnings are unrelated." + "reason": "Documentation example errors are unrelated to key/position-id loading warnings." }, { "left": "issue:28218", "right": "issue:35141", "accept": false, - "reason": "Tokenizer added-space behavior and embedding reinitialization on resize are separate tokenizer/model-init bugs." + "reason": "Tokenizer spacing bug and embedding reinitialization after resize are different issues." }, { "left": "issue:43720", "right": "issue:44112", "accept": false, - "reason": "BitNet packed-weight loading failure and GraniteSpeech stale device test failure are unrelated." + "reason": "Packed-weight loading in BitNet and a GraniteSpeech stale-device CI test are unrelated." }, { "left": "issue:43335", "right": "issue:44112", "accept": false, - "reason": "SwitchTransformers sparse-layer configuration and GraniteSpeech CI device override do not share the same underlying defect." + "reason": "Different model/config bug versus CI test flakiness; not mergeable as one fix." + }, + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation and D-FINE auxiliary-loss omission are separate model bugs." } ] }, "evaluator_result": { "accept": false, - "feedback": "The duplicate-pair call on 45310/45478 looks correct and the rest of the soft-edge rejections are conservative. However, the justification overreaches when it says 45310 has the \"richest thread\" among the pair: the packet shows 45478 has more discussion activity (4 vs 2), so that comparative claim is not grounded. Tighten the representative/canonical rationale to only the supported facts (same title, inbound reference, slightly longer body) and avoid unsupported thread-quality claims." + "feedback": "The duplicate call on issues 45310/45478 is plausible from the identical titles and ordering, and the soft-edge rejections are generally conservative. However, the canonical_issue_reason overstates the packet evidence by claiming 45310 has the \"more informative title/body\"\u2014the packet only provides titles and body lengths, not the body content or relative informativeness. Please keep the justification limited to what is explicitly supported (identical title, later duplicate-like report, inbound reference on 45310)." }, "evaluator_used": true, "retried": true, @@ -8299,13 +8330,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 26, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:25251", "issue:29127", @@ -8317,7 +8348,6 @@ "issue:40444", "issue:42490", "issue:43278", - "issue:43335", "issue:43381", "issue:43412", "issue:43526", @@ -8325,6 +8355,7 @@ "issue:43575", "issue:43606", "issue:43611", + "issue:43653", "issue:43701", "issue:43746", "issue:43854", @@ -8335,6 +8366,7 @@ "issue:44493", "issue:44561", "issue:44568", + "issue:44743", "issue:44779", "issue:44857", "issue:44977", @@ -8347,7 +8379,6 @@ "issue:45491", "issue:45563", "issue:45588", - "issue:45593", "issue:45698", "issue:45803", "issue:45823", @@ -8358,14 +8389,13 @@ ], "soft_pairs": [ "issue:44448|issue:44779", - "issue:43335|issue:45593", "issue:42490|issue:43854", "issue:43606|issue:44857", "issue:45920|issue:46192", "issue:44360|issue:44561", "issue:38175|issue:44493", - "issue:44857|issue:45072", "issue:44977|issue:45229", + "issue:44857|issue:45072", "issue:44448|issue:44977", "issue:34689|issue:35141", "issue:43701|issue:45083", @@ -8377,239 +8407,240 @@ "issue:42490|issue:43575", "issue:44561|issue:44779", "issue:44977|issue:45563", - "issue:44568|issue:45491", "issue:44857|issue:45588", "issue:43746|issue:46209", - "issue:45372|issue:45588", + "issue:44568|issue:45491", "issue:43412|issue:45588", + "issue:45372|issue:45588", "issue:43278|issue:43381", - "issue:43611|issue:46285", "issue:45198|issue:45803", "issue:45588|issue:45823", + "issue:43611|issue:46285", "issue:44448|issue:44466", + "issue:45245|issue:45563", "issue:25251|issue:32090", - "issue:45245|issue:45563" + "issue:43653|issue:44743" ], - "prepared_review_unit_hash": "2b5a0ae93dd148847c34cbf4db58d4db58676dd16f47d07b895170d00ef58e5e", + "prepared_review_unit_hash": "dcfebee9fe332132c88ca179358d6d98a2439e4ed0bf5806e257ec047ed923b1", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37485, - "estimated_input_tokens": 9372, - "estimated_eval_tokens": 19000 + "serialized_chars": 37506, + "estimated_input_tokens": 9377, + "estimated_eval_tokens": 19010 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated issue set with a few superficial v5/regression and tokenizer/load-save themes; no true duplicate pair stands out.", - "confidence": 0.32, - "canonical_issue_reason": "Issue 44448 is the broadest representative of the v4/v5 behavior-regression theme and is one of the more centrally connected items in the soft-pair graph.", + "summary": "These items are largely unrelated, spanning tokenizer regressions, model-loading/saving issues, device/dtype mismatches, export failures, and model-specific runtime bugs. None of the soft pairs look like the same underlying bug or change, so all soft edges should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": "Issue 45920 is the best representative because it is the broadest, most explicit v5 tokenizer regression umbrella and covers multiple affected models, making it the strongest canonical issue for a tokenizer-bug cluster.", "canonical_pr_reason": null, - "best_issue_reason": "44448 is the best single issue anchor for this cluster because it reflects an observable version-regression in model behavior, while most other issues are narrower component-specific bugs.", + "best_issue_reason": "Issue 45920 is the strongest standalone issue: it is specific enough to be actionable, but broad enough to subsume related tokenizer reports across several models.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44448", "right": "issue:44779", "accept": false, - "reason": "Both mention v4/v5 regressions, but one is Pegasus generation drift and the other is DeepSeek tokenizer IDs; different code paths." - }, - { - "left": "issue:43335", - "right": "issue:45593", - "accept": false, - "reason": "Sparse-layer config bug vs D-FINE auxiliary-loss omission; unrelated models and failures." + "reason": "Both are v5 output regressions, but one is Pegasus output drift and the other is a DeepSeek tokenizer error; different code paths and symptoms." }, { "left": "issue:42490", "right": "issue:43854", "accept": false, - "reason": "Local-path trailing-slash import warning vs GLM-4.7 test load failure; different symptoms." + "reason": "One is a local-path import warning with trailing slash, the other is a model-loading failure for GLM-4.7; not the same defect." }, { "left": "issue:43606", "right": "issue:44857", "accept": false, - "reason": "CPU-offload device mismatch vs FP16 CUDA crash in a loss module; not the same bug." + "reason": "CPU offload device mismatch in bark-small is unrelated to an AMP float16 crash in LwDetrImageLoss." }, { "left": "issue:45920", "right": "issue:46192", "accept": false, - "reason": "Both are tokenization-related, but one is wrong IDs across models and the other is ignored return_tensors kwargs." + "reason": "Tokenizer wrong IDs across multiple models is different from return_tensors being ignored when text_kwargs is passed." }, { "left": "issue:44360", "right": "issue:44561", "accept": false, - "reason": "DSA indexer activation omission vs v5 removal of is_torch_fx_available; no shared failure mode." + "reason": "DSA indexer ReLU discussion has no overlap with the removal of is_torch_fx_available breaking trust_remote_code models." }, { "left": "issue:38175", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 zero probabilities and unexpected position-id keys are unrelated inference issues." + "reason": "Siglip2 zero probabilities and unexpected position-id keys are different model-specific inference bugs." }, { - "left": "issue:44857", - "right": "issue:45072", + "left": "issue:44977", + "right": "issue:45229", "accept": false, - "reason": "Both involve dtype problems, but different models, ops, and failure points." + "reason": "Qwen3.5 flash-attention generation failure and Gemma4 multi-GPU OOM are unrelated failures." }, { - "left": "issue:44977", - "right": "issue:45229", + "left": "issue:44857", + "right": "issue:45072", "accept": false, - "reason": "Flash-attention generation bug vs multi-GPU OOM; distinct runtime issues." + "reason": "Both mention dtype-related problems, but one is an AMP crash in a loss module and the other is a bfloat16 mismatch in different models." }, { "left": "issue:44448", "right": "issue:44977", "accept": false, - "reason": "Both are v5 generation regressions, but Pegasus output drift and Qwen flash-attention failure are different paths." + "reason": "Pegasus v4/v5 output differences and Qwen3.5 flash-attention failures affect different models and code paths." }, { "left": "issue:34689", "right": "issue:35141", "accept": false, - "reason": "Model loading breakage for Llama 3.2 vs embedding reinit after resize; unrelated." + "reason": "Model loading breakage for Llama 3.2 vision is unrelated to output embeddings being reinitialized after resize when weights are untied." }, { "left": "issue:43701", "right": "issue:45083", "accept": false, - "reason": "Checkpoint key mismatch and a feature-length helper bug are unrelated." + "reason": "Checkpoint key mismatch and qwen3_omni_moe feature-length helper behavior are distinct issues." }, { "left": "issue:29127", "right": "issue:32090", "accept": false, - "reason": "LayoutLMv3 error messaging vs Trainer broadcast NoneType; no overlap." + "reason": "LayoutLMv3 error-message clarity and Trainer _gpu_broadcast_one NoneType failure are unrelated." }, { "left": "issue:43550", "right": "issue:45083", "accept": false, - "reason": "torch.compile/SDPA failure in Bamba is separate from qwen3_omni_moe length calculation." + "reason": "Bamba torch.compile/SDPA failure is not the same underlying bug as a qwen3_omni_moe output-length helper issue." }, { "left": "issue:43927", "right": "issue:45698", "accept": false, - "reason": "Both concern save/load, but one loses custom token IDs and the other loads the wrong custom module." + "reason": "Both involve save/load behavior, but one loses custom token IDs and the other loads the wrong custom module; different metadata paths." }, { "left": "issue:40444", "right": "issue:43526", "accept": false, - "reason": "IterableDataset multi-image finetuning failure vs BEiT label reduction bug." + "reason": "IterableDataset with multiple images per prompt and BeitImageProcessorFast reduce_labels are different preprocessing bugs." }, { "left": "issue:29942", "right": "issue:35141", "accept": false, - "reason": "Flash Attention 2 test failures vs output embedding reinit after resize." + "reason": "Flash Attention 2 test failures are unrelated to embedding reinitialization during post_init." }, { "left": "issue:42490", "right": "issue:43575", "accept": false, - "reason": "Trailing-slash import warning vs tp OOM on Qwen2-57B; unrelated." + "reason": "Trailing-slash local-model warning is unrelated to Qwen2-57B TP out-of-memory." }, { "left": "issue:44561", "right": "issue:44779", "accept": false, - "reason": "Trust_remote_code breakage from API removal vs DeepSeek tokenizer regression; different root causes." + "reason": "A missing helper in v5 trust_remote_code support is not the same bug as DeepSeek tokenizer regression." }, { "left": "issue:44977", "right": "issue:45563", "accept": false, - "reason": "Flash-attention generation issue vs stale warning for num_return_sequences." - }, - { - "left": "issue:44568", - "right": "issue:45491", - "accept": false, - "reason": "Special tokens omission in tokenizer vs Gemma3 NaNs from sliding-window attention." + "reason": "Flash-attention generation failure and a stale warning for num_return_sequences are different generate-path issues." }, { "left": "issue:44857", "right": "issue:45588", "accept": false, - "reason": "Different crash sites: LwDetr AMP loss vs flash_attention.py s_aux=None." + "reason": "Float16 AMP crash in LwDetrImageLoss and flash_attention.py crashing on s_aux=None are separate defects." }, { "left": "issue:43746", "right": "issue:46209", "accept": false, - "reason": "PEFT checkpoint loading vs FP8 inference on B200; unrelated." + "reason": "PEFT checkpoint loading for GraniteSpeech and FP8 inference failure on B200 are unrelated." }, { - "left": "issue:45372", - "right": "issue:45588", + "left": "issue:44568", + "right": "issue:45491", "accept": false, - "reason": "Gemma4 processor import failure vs flash_attention.py None aux crash." + "reason": "Tokenizer add_special_tokens behavior and Gemma3 NaN embeddings with sliding-window attention are not the same bug." }, { "left": "issue:43412", "right": "issue:45588", "accept": false, - "reason": "Executorch export guard/erfinv issue vs flash_attention None aux crash." + "reason": "Executorch export guard/support issues are unrelated to flash_attention.py AttributeError on sink-less models." }, { - "left": "issue:43278", - "right": "issue:43381", + "left": "issue:45372", + "right": "issue:45588", "accept": false, - "reason": "Eval/train dtype drift vs gradient checkpointing disallowed in eval; not the same bug." + "reason": "Gemma4 processor import failure and flash_attention.py s_aux=None crash are different failures." }, { - "left": "issue:43611", - "right": "issue:46285", + "left": "issue:43278", + "right": "issue:43381", "accept": false, - "reason": "Base_model_prefix loading regression vs CLIPTextModel structure mismatch; distinct model-loading issues." + "reason": "Embedding dtype drift in eval and gradient checkpointing being unusable in eval mode are distinct." }, { "left": "issue:45198", "right": "issue:45803", "accept": false, - "reason": "Wav2Vec2 save/tokenization failure vs Fuyu bare-except bug." + "reason": "Wav2Vec2 save/tokenization failure and Fuyu bare-except swallowing errors are unrelated." }, { "left": "issue:45588", "right": "issue:45823", "accept": false, - "reason": "flash_attention s_aux=None crash vs Gemma4 device_map mismatch; different components." + "reason": "Flash-attention sink-less crash and Gemma4 PLE device mismatch are different runtime bugs." + }, + { + "left": "issue:43611", + "right": "issue:46285", + "accept": false, + "reason": "base_model_prefix loading breakage and CLIPTextModel structural inconsistency are both v5 model-structure issues but not the same underlying defect." }, { "left": "issue:44448", "right": "issue:44466", "accept": false, - "reason": "One is output drift between v4/v5, the other is lm_head serialization depending on device." + "reason": "Pegasus v4/v5 output mismatch and lm_head.weight serialization varying by device are separate regression classes." + }, + { + "left": "issue:45245", + "right": "issue:45563", + "accept": false, + "reason": "Category-cardinality limit error and stale generate warning are unrelated." }, { "left": "issue:25251", "right": "issue:32090", "accept": false, - "reason": "Top_k pipeline output nesting bug vs GPU broadcast NoneType error." + "reason": "Pipeline top_k nesting change and Trainer _gpu_broadcast_one NoneType error affect different subsystems and code paths." }, { - "left": "issue:45245", - "right": "issue:45563", + "left": "issue:43653", + "right": "issue:44743", "accept": false, - "reason": "Category limit error vs stale generate warning." + "reason": "BigBirdTokenizer special-token decoding issue and qwen3_5 recurrent-state reset bug are unrelated." } ] }, @@ -8625,19 +8656,20 @@ "component_index": 1, "component_count": 18, "review_unit_index": 27, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:10000", "issue:15354", "issue:29127", "issue:29942", "issue:30064", + "issue:32090", "issue:33357", "issue:34567", "issue:36010", @@ -8668,7 +8700,6 @@ "issue:44479", "issue:44610", "issue:44734", - "issue:44743", "issue:44871", "issue:44977", "issue:45198", @@ -8683,238 +8714,238 @@ "issue:46121" ], "soft_pairs": [ - "issue:43653|issue:44743", "issue:36010|issue:37428", - "issue:45459|issue:45712", "issue:43065|issue:45412", - "issue:44186|issue:45198", + "issue:45459|issue:45712", "issue:43611|issue:46121", + "issue:44186|issue:45198", "issue:44361|issue:45198", + "issue:29127|issue:34567", "issue:33357|issue:39692", "issue:44871|issue:44977", - "issue:29127|issue:34567", "issue:45372|issue:45459", "issue:44206|issue:44479", - "issue:29942|issue:30064", "issue:42915|issue:42994", - "issue:43976|issue:45440", "issue:43994|issue:45412", + "issue:29942|issue:30064", + "issue:43976|issue:45440", "issue:43278|issue:43575", "issue:43066|issue:44462", "issue:44265|issue:45198", "issue:10000|issue:15354", "issue:42490|issue:43873", "issue:43873|issue:45229", - "issue:43421|issue:45491", - "issue:43381|issue:43701", - "issue:44734|issue:45406", "issue:43065|issue:43994", + "issue:44734|issue:45406", + "issue:43381|issue:43701", + "issue:43421|issue:45491", "issue:43653|issue:44442", "issue:44610|issue:45412", - "issue:41720|issue:45229" + "issue:41720|issue:45229", + "issue:32090|issue:33357" ], - "prepared_review_unit_hash": "4b1ccfadf8ee22363058be0af29cee1d90a6b90d40a5e442d4558b0bb25a8278", + "prepared_review_unit_hash": "129375939ff4ce0b3e72ee0d2e5ebe18320f88e0a1b45157669aa27c3144aec0", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36437, - "estimated_input_tokens": 9110, - "estimated_eval_tokens": 18476 + "serialized_chars": 36454, + "estimated_input_tokens": 9114, + "estimated_eval_tokens": 18484 }, "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-, tokenizer-, quantization-, serving-, or environment-specific bugs rather than duplicates of one underlying defect. No strong duplicate canonical emerges.", - "confidence": 0.97, - "canonical_issue_reason": "No single issue is a clear canonical target; the cluster spans many unrelated failure modes across different models and subsystems.", + "summary": "This cluster is not a real duplicate set: it mixes unrelated issues across tokenizers, generation, training, serving, quantization, and model-specific regressions. All soft edges look like false positives.", + "confidence": 0.99, + "canonical_issue_reason": "No safe canonical issue: the items span many unrelated code paths and would not collapse into one duplicate cluster.", "canonical_pr_reason": null, - "best_issue_reason": "No issue is a strong global representative. If forced, 43873 is the broadest/general discussion, but it is still not a duplicate hub.", + "best_issue_reason": "No single issue is a good global representative; the cluster is too heterogeneous to use one issue as the canonical target.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43653", - "right": "issue:44743", + "left": "issue:36010", + "right": "issue:37428", "accept": false, - "reason": "Tokenizer special-token registration/empty decode is unrelated to Qwen recurrent-state cache reset behavior." + "reason": "Different import failures in different modules: GenerationMixin vs flash-attention utility." }, { - "left": "issue:36010", - "right": "issue:37428", + "left": "issue:43065", + "right": "issue:45412", "accept": false, - "reason": "These are different import errors in different modules with no shared underlying code-path." + "reason": "Sam3PixelDecoder dummy layer vs RT-DETR memory leak; unrelated bugs." }, { "left": "issue:45459", "right": "issue:45712", "accept": false, - "reason": "Hiding tokenizer errors when protobuf is missing is unrelated to leftover dummy classes leaking into dir(transformers)." + "reason": "Tokenizer error masking with missing protobuf vs leftover dummy classes; no shared bug." }, { - "left": "issue:43065", - "right": "issue:45412", + "left": "issue:43611", + "right": "issue:46121", "accept": false, - "reason": "A dummy Conv2d in Sam3PixelDecoder is unrelated to RT-DETR memory not being released on deletion." + "reason": "base_model_prefix loading regression vs rope validation TypeError; different code paths." }, { "left": "issue:44186", "right": "issue:45198", "accept": false, - "reason": "Both involve tokenization, but they affect different models and different failure modes; not the same bug." + "reason": "LayoutLMv2 NER/padding crash vs Wav2Vec2 save/tokenization failure; different tokenizer behavior." }, { - "left": "issue:43611", - "right": "issue:46121", + "left": "issue:44361", + "right": "issue:45198", "accept": false, - "reason": "base_model_prefix loading breaks are unrelated to rope-validation TypeError on a JSON-loaded list." + "reason": "MLukeTokenizer AttributeError is unrelated to Wav2Vec2 tokenization/save_pretrained issues." }, { - "left": "issue:44361", - "right": "issue:45198", + "left": "issue:29127", + "right": "issue:34567", "accept": false, - "reason": "MLukeTokenizer AttributeError is a different tokenizer issue from Wav2Vec2 save/tokenization failure." + "reason": "LayoutLMv3 error messaging vs TrainerState token counter update; unrelated training/model issues." }, { "left": "issue:33357", "right": "issue:39692", "accept": false, - "reason": "Both mention vision models, but the concrete bugs differ: MacOS bus error vs docs/model-processor mismatch and quantization failure." + "reason": "MacOS CLIP bus error vs SigLIP2 documentation/example errors; different failure modes." }, { "left": "issue:44871", "right": "issue:44977", "accept": false, - "reason": "Gemma eos_token_id config mismatch is unrelated to Qwen3.5 flash-attention generation failure." - }, - { - "left": "issue:29127", - "right": "issue:34567", - "accept": false, - "reason": "LayoutLMv3 error handling for boxes and TrainerState token counting are unrelated subsystems." + "reason": "Gemma-3 eos_token mismatch vs Qwen3.5 flash-attention generation bug; different model/config paths." }, { "left": "issue:45372", "right": "issue:45459", "accept": false, - "reason": "Gemma 4 processor loading blocked by a mistral_common import is not the same as tokenizer error masking without protobuf." + "reason": "Gemma4 processor import dependency issue vs protobuf decode-error handling; unrelated." }, { "left": "issue:44206", "right": "issue:44479", "accept": false, - "reason": "Unsupported feature-extractor argument regression and video-input regression are different code paths and models." + "reason": "LasrFeatureExtractor crash vs video-input regression for Qwen VL models; not the same bug." }, { - "left": "issue:29942", - "right": "issue:30064", + "left": "issue:42915", + "right": "issue:42994", "accept": false, - "reason": "Flash Attention 2 test failures are unrelated to image processor void-segmentation-map handling." + "reason": "Qwen3Moe FP8 config failure vs quantized model saving failure; separate quantization problems." }, { - "left": "issue:42915", - "right": "issue:42994", + "left": "issue:43994", + "right": "issue:45412", "accept": false, - "reason": "FP8 quantization config failure on load and quantized model saving failure are distinct bugs." + "reason": "SigLIP2 bad outputs vs RT-DETR memory leak; unrelated subsystems." }, { - "left": "issue:43976", - "right": "issue:45440", + "left": "issue:29942", + "right": "issue:30064", "accept": false, - "reason": "Python version compatibility regression is unrelated to DeepSeekV3 implementation divergence." + "reason": "Flash Attention 2 test failures vs void segmentation-map processing; different regressions." }, { - "left": "issue:43994", - "right": "issue:45412", + "left": "issue:43976", + "right": "issue:45440", "accept": false, - "reason": "SigLIP2 nonsense outputs and RT-DETR memory leaks are unrelated issues." + "reason": "Python version compatibility issue vs DeepseekV3MoE divergence; unrelated." }, { "left": "issue:43278", "right": "issue:43575", "accept": false, - "reason": "Embedding dtype drift in eval and Qwen2 OOM under tensor parallelism are different problems." + "reason": "Embedding dtype drift vs tensor-parallel OOM; different bugs." }, { "left": "issue:43066", "right": "issue:44462", "accept": false, - "reason": "Wrong tokenizer decoder type in v5 is not the same as AutoTokenizer ignoring repository tokenizer.json." + "reason": "Tokenizer decoder type regression vs AutoTokenizer ignoring tokenizer.json; both tokenizer-related but distinct." }, { "left": "issue:44265", "right": "issue:45198", "accept": false, - "reason": "torch.export failures from torch_compilable_check are unrelated to Wav2Vec2 save/tokenization failures." + "reason": "torch.export/torch_compilable_check failure vs Wav2Vec2 tokenization/save failure; unrelated." }, { "left": "issue:10000", "right": "issue:15354", "accept": false, - "reason": "German DistilBertModel and GeneratorExp/TorchScript export are unrelated model/framework issues." + "reason": "DistilBertModel issue vs GeneratorExp unsupported in torch.jit.script; unrelated model/export bugs." }, { "left": "issue:42490", "right": "issue:43873", "accept": false, - "reason": "A trailing-slash import warning is unrelated to quantization offloading not working." + "reason": "Trailing-slash local model warning vs quantization offloading regression; different behavior." }, { "left": "issue:43873", "right": "issue:45229", "accept": false, - "reason": "Offloading/quantization behavior is not the same as multi-GPU CUDA OOM for Gemma4." + "reason": "Offloading-with-quantization problem vs multi-GPU CUDA OOM; both memory-related but not the same bug." }, { - "left": "issue:43421", - "right": "issue:45491", + "left": "issue:43065", + "right": "issue:43994", "accept": false, - "reason": "Runtime post-processor updates in TokenizersBackend are unrelated to Gemma3 NaN embeddings with mixed-length batching." + "reason": "Dummy Conv2d in Sam3PixelDecoder vs SigLIP2 inference issues; unrelated." }, { - "left": "issue:43381", - "right": "issue:43701", + "left": "issue:44734", + "right": "issue:45406", "accept": false, - "reason": "Eval-mode gradient checkpointing and resume-from-checkpoint key mismatch are different training-flow bugs." + "reason": "KV-cache continuation indexing crash vs Gemma4Processor missing _tokenizer; different code paths." }, { - "left": "issue:44734", - "right": "issue:45406", + "left": "issue:43381", + "right": "issue:43701", "accept": false, - "reason": "KV-cache continuation tensor indexing in serve is unrelated to Gemma4Processor missing _tokenizer." + "reason": "Eval-mode checkpointing restriction vs resume_from_checkpoint key mismatch; unrelated training-state bugs." }, { - "left": "issue:43065", - "right": "issue:43994", + "left": "issue:43421", + "right": "issue:45491", "accept": false, - "reason": "A dummy layer in Sam3PixelDecoder is unrelated to SigLIP2 wrong/nonsensical outputs." + "reason": "Runtime special-token update feature vs Gemma3 NaN embeddings; not the same issue." }, { "left": "issue:43653", "right": "issue:44442", "accept": false, - "reason": "BigBirdTokenizer special-token registration is unrelated to AutoTokenizer failing to load FastSpeech2ConformerTokenizer." + "reason": "BigBirdTokenizer special-token registration vs AutoTokenizer loading FastSpeech2ConformerTokenizer; distinct tokenizer problems." }, { "left": "issue:44610", "right": "issue:45412", "accept": false, - "reason": "Processor image-size mismatch and RT-DETR memory leaks are different bugs." + "reason": "Processor/model size mismatch vs RT-DETR memory leak; unrelated." }, { "left": "issue:41720", "right": "issue:45229", "accept": false, - "reason": "Qwen3 auto device mapping cudaErrorAssert and Gemma4 multi-GPU OOM are different failure modes." + "reason": "Qwen3 auto device-mapping assert vs Gemma4 multi-GPU OOM; different hardware/runtime failures." + }, + { + "left": "issue:32090", + "right": "issue:33357", + "accept": false, + "reason": "Trainer NoneType broadcast error vs MacOS bus error with CLIP; unrelated issues." } ] }, @@ -8930,17 +8961,16 @@ "component_index": 1, "component_count": 18, "review_unit_index": 28, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:29127", "issue:29942", - "issue:32090", "issue:33357", "issue:35141", "issue:36010", @@ -8963,6 +8993,7 @@ "issue:43761", "issue:43824", "issue:43931", + "issue:44062", "issue:44077", "issue:44186", "issue:44206", @@ -8984,10 +9015,10 @@ "issue:45588", "issue:45593", "issue:45834", - "issue:45859" + "issue:45859", + "issue:45907" ], "soft_pairs": [ - "issue:32090|issue:33357", "issue:43756|issue:45440", "issue:43278|issue:43329", "issue:45372|issue:45406", @@ -8999,212 +9030,213 @@ "issue:45103|issue:45372", "issue:43526|issue:44186", "issue:43575|issue:43701", - "issue:43575|issue:44387", - "issue:42490|issue:43381", "issue:44977|issue:45245", "issue:43526|issue:44857", + "issue:43575|issue:44387", + "issue:42490|issue:43381", "issue:43334|issue:43531", "issue:43824|issue:44279", "issue:44442|issue:45072", - "issue:29127|issue:35141", "issue:43646|issue:44077", - "issue:44743|issue:44811", + "issue:29127|issue:35141", "issue:42915|issue:43931", + "issue:44743|issue:44811", "issue:43262|issue:44206", "issue:43452|issue:45588", "issue:44977|issue:45584", - "issue:33357|issue:38175" + "issue:33357|issue:38175", + "issue:44062|issue:45907" ], - "prepared_review_unit_hash": "472958a73cff785d13966d21fb63f26b556bcbc72c8fc144b9c2a05ddcf878a0", + "prepared_review_unit_hash": "d01e2f2fe6f73195c83f191b188be16c4931549874503038603d48aadd4a0573", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 35308, - "estimated_input_tokens": 8827, - "estimated_eval_tokens": 17910 + "serialized_chars": 35862, + "estimated_input_tokens": 8966, + "estimated_eval_tokens": 18188 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: the candidate pairs mostly share only broad subsystem vocabulary, not the same underlying bug. I would not merge any of the soft edges. If a single representative issue is needed, issue 43761 is the cleanest, most concrete regression report in the set.", - "confidence": 0.61, - "canonical_issue_reason": "Issue 43761 is the most self-contained and clearly scoped report here, with a specific regression, concrete expected behavior, and an explicit failure condition/version context.", + "summary": "The items do not form a single duplicate cluster; they span many unrelated bugs and regressions across loading, tokenization, vision, audio, attention, and training paths. All soft-edge matches look like topic-level similarity only, not the same underlying defect.", + "confidence": 0.98, + "canonical_issue_reason": "No clear canonical issue: the set is heterogeneous, and the apparent matches are false positives rather than one underlying bug.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43761 is a strong representative because it describes one precise code-path failure rather than a broad or speculative symptom.", + "best_issue_reason": "No single issue is a strong cluster anchor; at best, issue 43646 is the broadest/generalized regression report, but it still does not represent the rest of the items.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:32090", - "right": "issue:33357", - "accept": false, - "reason": "Different failures: Trainer GPU broadcast type error vs MacOS bus error loading a CLIP model." - }, { "left": "issue:43756", "right": "issue:45440", "accept": false, - "reason": "Both concern model implementation correctness, but they target different models and different bugs." + "reason": "Both mention model internals, but one is about Smollm3 RoPE layer handling and the other is DeepSeekV3MoE implementation divergence; different code paths and fixes." }, { "left": "issue:43278", "right": "issue:43329", "accept": false, - "reason": "Unrelated areas: embedding dtype drift vs multimodal token counting video-branch NameError-style bug." + "reason": "Different bugs: BF16->FP32 embedding dtype shift in eval vs undefined video-token helper/merge_size in multimodal token counting." }, { "left": "issue:45372", "right": "issue:45406", "accept": false, - "reason": "Gemma 4 loading problems, but one is a dependency import failure and the other is a missing processor attribute in serve." + "reason": "Same Gemma 4 area, but one is a mistral_common import/dependency failure and the other is a missing _tokenizer attribute in serve; not the same defect." }, { "left": "issue:43526", "right": "issue:45834", "accept": false, - "reason": "Both are vision/OCR-adjacent, but the concrete failures are different and do not share a code path." + "reason": "BeitImageProcessorFast reduce_labels bug vs Kosmos2.5 long OCR index error; unrelated processor paths." }, { "left": "issue:29942", "right": "issue:36010", "accept": false, - "reason": "Flash Attention 2 test failures are not the same as a GenerationMixin import error." + "reason": "Flash Attention 2 test failures and GenerationMixin import errors are unrelated regression types." }, { "left": "issue:43723", "right": "issue:44554", "accept": false, - "reason": "Tokenizer-loading issue and MPS attention correctness bug are unrelated." + "reason": "Tokenizer loading in v5 vs an MPS attention correctness issue; different subsystems and failure modes." }, { "left": "issue:43761", "right": "issue:45859", "accept": false, - "reason": "CLIP vision hidden-state regression is unrelated to Qwen rotary embedding CPU offload incompatibility." + "reason": "CLIPVisionModel hidden_states regression vs Qwen3_MoeTextRotaryEmbedding CPU-offload incompatibility; not the same code path." }, { "left": "issue:45440", "right": "issue:45593", "accept": false, - "reason": "Different models and different defects: remote implementation divergence vs missing auxiliary losses." + "reason": "DeepSeekV3MoE divergence and D-FINE auxiliary-loss behavior are unrelated model-specific bugs." }, { "left": "issue:45103", "right": "issue:45372", "accept": false, - "reason": "Auto-docstring annotation handling crash is unrelated to the mistral_common ReasoningEffort import error." + "reason": "auto_docstring annotation parsing crash vs Gemma 4 ReasoningEffort import breakage; unrelated." }, { "left": "issue:43526", "right": "issue:44186", "accept": false, - "reason": "Different components and failures: BEiT image processor label reduction vs LayoutLMv2 tokenizer padding/NER crash." + "reason": "Both are tokenizer/image-processing issues, but BEiT reduce_labels and LayoutLMv2 NER/padding crashes are different implementations." }, { "left": "issue:43575", "right": "issue:43701", "accept": false, - "reason": "Tensor-parallel OOM and resume checkpoint key mismatch are unrelated issues." + "reason": "TP loading OOM and resume_from_checkpoint key mismatch are different training/runtime problems." }, { - "left": "issue:43575", - "right": "issue:44387", + "left": "issue:44977", + "right": "issue:45245", "accept": false, - "reason": "Both mention OOM, but one is model-load TP memory pressure and the other is int4 quantization reserved-memory behavior." + "reason": "Qwen3.5 flash-attention generation bug vs category-count limit in another pipeline; no shared underlying bug." }, { - "left": "issue:42490", - "right": "issue:43381", + "left": "issue:43526", + "right": "issue:44857", "accept": false, - "reason": "Local-model slash-path import warning is unrelated to gradient checkpointing in eval mode." + "reason": "BEiT label reduction and LwDetrImageLoss float16 CUDA crash are unrelated." }, { - "left": "issue:44977", - "right": "issue:45245", + "left": "issue:43575", + "right": "issue:44387", "accept": false, - "reason": "Flash-attention generation bug and category-cardinality runtime error are unrelated." + "reason": "Both involve memory pressure, but one is tensor-parallel loading OOM and the other is increased reserved memory under int4 quantization; different root causes." }, { - "left": "issue:43526", - "right": "issue:44857", + "left": "issue:42490", + "right": "issue:43381", "accept": false, - "reason": "Different image-loss/processor bugs with no shared concrete code path." + "reason": "Local-path import warning vs gradient checkpointing in eval mode; unrelated." }, { "left": "issue:43334", "right": "issue:43531", "accept": false, - "reason": "Both involve Qwen3/Qwen3-MoE, but one is a missing pad_token_id on load and the other is a sliding-window runtime issue." + "reason": "Qwen3-VL pad_token_id loading error and Qwen3-MoE sliding_window behavior are different model/configuration bugs." }, { "left": "issue:43824", "right": "issue:44279", "accept": false, - "reason": "ImportError for a specific model class is not the same as a general dependency issue." + "reason": "Qwen2_5_VL import failure vs a generic dependency issue; too broad to be the same concrete bug." }, { "left": "issue:44442", "right": "issue:45072", "accept": false, - "reason": "Tokenizer loading failure and dtype mismatch in inference are unrelated." - }, - { - "left": "issue:29127", - "right": "issue:35141", - "accept": false, - "reason": "Different model subsystems and different symptoms: clearer error messaging vs output embedding reinitialization." + "reason": "Tokenizer load failure and BF16 dtype mismatch in inference are unrelated." }, { "left": "issue:43646", "right": "issue:44077", "accept": false, - "reason": "Transformers 5.0 custom initialization breakage is broader and distinct from patchtsmixer's optional post_init restriction." + "reason": "General Transformers 5 initialization breakage and patchtsmixer optional post_init restriction are related only at a broad API level, not the same fixable defect." }, { - "left": "issue:44743", - "right": "issue:44811", + "left": "issue:29127", + "right": "issue:35141", "accept": false, - "reason": "Qwen recurrent-state caching bug is unrelated to Whisper batch_decode special-token handling." + "reason": "LayoutLMv3 error messaging and token-embedding reinitialization in post_init are unrelated." }, { "left": "issue:42915", "right": "issue:43931", "accept": false, - "reason": "FineGrainedFP8Config failure and Qwen3-VL weight-shape mismatch are different loading problems." + "reason": "Qwen3Moe FineGrainedFP8Config failure and Qwen3-VL weight-shape mismatch are different loading/compatibility issues." + }, + { + "left": "issue:44743", + "right": "issue:44811", + "accept": false, + "reason": "Qwen3_5 recurrent-state cache bug vs Whisper batch_decode skip_special_tokens handling; unrelated model families and code paths." }, { "left": "issue:43262", "right": "issue:44206", "accept": false, - "reason": "Audio chat-template sampling-rate default mismatch is unrelated to an unsupported center argument crash." + "reason": "Audio chat-template sample-rate default bug vs LasrFeatureExtractor unsupported center arg; different processors and fixes." }, { "left": "issue:43452", "right": "issue:45588", "accept": false, - "reason": "Tokenizer/model loading with gguf_file has no concrete overlap with flash_attention s_aux=None handling." + "reason": "gguf_file loading breakage and flash_attention sink-less s_aux=None crash are unrelated." }, { "left": "issue:44977", "right": "issue:45584", "accept": false, - "reason": "Flash-attention generation issue is unrelated to Whisper empty-transcription failure." + "reason": "Qwen3.5 flash-attention generation failure vs Whisper empty-transcription/alignment failure; different models and root causes." }, { "left": "issue:33357", "right": "issue:38175", "accept": false, - "reason": "Both are vision-model symptoms, but one is a MacOS bus error and the other is unexpected zero probabilities." + "reason": "Both are model inference issues, but MacOS bus error on community CLIP and zero probabilities on siglip2-base are not the same underlying bug." + }, + { + "left": "issue:44062", + "right": "issue:45907", + "accept": false, + "reason": "AddedToken special-argument duplication and list_repo_files revision forwarding are unrelated utility/API bugs." } ] }, @@ -9220,13 +9252,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 29, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:28218", "issue:32090", @@ -9243,12 +9275,12 @@ "issue:43550", "issue:43611", "issue:43643", + "issue:43650", "issue:43720", "issue:43761", "issue:43827", "issue:43873", "issue:43931", - "issue:44062", "issue:44077", "issue:44188", "issue:44190", @@ -9272,14 +9304,13 @@ "issue:45698", "issue:45759", "issue:45803", - "issue:45907", + "issue:46202", "issue:46209", "issue:46234" ], "soft_pairs": [ - "issue:44062|issue:45907", - "issue:44464|issue:45491", "issue:28218|issue:32090", + "issue:44464|issue:45491", "issue:33290|issue:34567", "issue:44442|issue:44898", "issue:43381|issue:43873", @@ -9294,227 +9325,222 @@ "issue:44442|issue:45803", "issue:43550|issue:44743", "issue:43493|issue:43827", - "issue:43873|issue:44871", "issue:42907|issue:45698", "issue:42175|issue:46234", - "issue:43761|issue:44188", + "issue:43873|issue:44871", "issue:43643|issue:44913", + "issue:43761|issue:44188", "issue:43316|issue:45759", "issue:45290|issue:46209", - "issue:44855|issue:45137", "issue:33290|issue:34689", "issue:44190|issue:44496", "issue:44568|issue:45479", - "issue:44077|issue:44479" + "issue:44077|issue:44479", + "issue:44855|issue:45137", + "issue:43650|issue:46202" ], - "prepared_review_unit_hash": "3e4d3e965a1bc3d1f47fd0042240d080f1bfd199076b5a3936b8d7fb1cea13ab", + "prepared_review_unit_hash": "791e08a582ef2b611b2bd7469a92c4b6104f140f69d32d6c4d8491c375290867", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 36145, - "estimated_input_tokens": 9037, - "estimated_eval_tokens": 18330 + "serialized_chars": 36053, + "estimated_input_tokens": 9014, + "estimated_eval_tokens": 18284 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated Transformers bug reports that only share broad topical similarity (loading, configs, generation, quantization). I found no soft pair that looks like the same underlying issue, so none should be merged as duplicates.", - "confidence": 0.9, - "canonical_issue_reason": "No clear duplicate cluster exists here. If one issue must represent the set, issue:45698 is the broadest and most active loading/serialization regression report, with multiple inbound refs and high discussion activity.", + "summary": "These items are mostly unrelated Transformers issues that were linked only by loose textual similarity. No soft edge looks like the same underlying bug/change, so none should be merged as duplicates.", + "confidence": 0.91, + "canonical_issue_reason": "If a representative issue is needed, issue 45698 is the broadest and most active loading/saving report here, with multiple references and discussion around from_pretrained/save_pretrained behavior.", "canonical_pr_reason": null, - "best_issue_reason": "Issue:45698 is the best single representative because it is active, well-discussed, and concerns a core save/load path that is plausibly adjacent to several other configuration/loading regressions in the set.", + "best_issue_reason": "Issue 45698 is the strongest standalone triage candidate because it is open, has higher discussion/inbound refs, and concerns a general model-loading regression rather than a model-specific edge case.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44062", - "right": "issue:45907", + "left": "issue:28218", + "right": "issue:32090", "accept": false, - "reason": "Completely different code paths: AddedToken construction vs Hub repo file listing/revision forwarding." + "reason": "Tokenizer added-token spacing is unrelated to Trainer's _gpu_broadcast_one NoneType error." }, { "left": "issue:44464", "right": "issue:45491", "accept": false, - "reason": "Both are generation/model-behavior bugs, but the concrete failures differ: compiled forward inconsistency vs Gemma3 sliding-window NaNs on mixed-length batches." - }, - { - "left": "issue:28218", - "right": "issue:32090", - "accept": false, - "reason": "Tokenizer added-space behavior is unrelated to the Trainer/NoneType GPU broadcast error." + "reason": "Chunked generation with compiled forward is a different failure mode from Gemma3 mixed-length NaN embeddings." }, { "left": "issue:33290", "right": "issue:34567", "accept": false, - "reason": "DeepSpeed/Adafactor OOM and num_input_tokens_seen not updating are unrelated trainer issues." + "reason": "Deepspeed Adafactor OOM and num_input_tokens_seen not updating are distinct trainer/optimizer bugs." }, { "left": "issue:44442", "right": "issue:44898", "accept": false, - "reason": "Tokenizer loading failure and Perceiver interpolation failure are unrelated model-specific bugs." + "reason": "Tokenizer loading failure and Perceiver image-classification resolution failure are unrelated code paths." }, { "left": "issue:43381", "right": "issue:43873", "accept": false, - "reason": "Gradient checkpointing eval-mode restriction and quantization offloading failure are different mechanisms." + "reason": "Gradient checkpointing eval-mode behavior does not match quantization offloading issues." }, { "left": "issue:44964", "right": "issue:45440", "accept": false, - "reason": "Separate multimodal model-loading regression vs DeepSeekV3MoE implementation divergence." + "reason": "Phi-4 multimodal loading regression and DeepseekV3MoE implementation divergence are different model-specific bugs." }, { "left": "issue:43316", "right": "issue:43525", "accept": false, - "reason": "Both are config/API issues, but for different models and different missing fields; not the same bug." + "reason": "Gemma3TextConfig API mismatch is not the same as Llama4Config missing pad_token_id." }, { "left": "issue:34689", "right": "issue:43611", "accept": false, - "reason": "Both involve version-related loading breakages, but one is a specific model loader regression and the other is a base_model_prefix handling change." + "reason": "Llama 3.2 Vision loading failure and base_model_prefix loading regression are separate model-loading problems." }, { "left": "issue:32090", "right": "issue:33290", "accept": false, - "reason": "Trainer broadcast NoneType error and DeepSpeed Adafactor OOM do not share a concrete failure mode." + "reason": "Trainer GPU broadcast type error and Adafactor/deepspeed OOM are different training failures." }, { "left": "issue:43720", "right": "issue:44442", "accept": false, - "reason": "BitNet packed-weight loading and FastSpeech2 tokenizer loading are unrelated." + "reason": "BitNet packed-weight loading bug is unrelated to FastSpeech2ConformerTokenizer auto-loading." }, { "left": "issue:44479", "right": "issue:45290", "accept": false, - "reason": "Video-input regression across Qwen VL models is unrelated to chat-template tokenization crashing on tool-call messages." + "reason": "Video-input regression for Qwen variants is not the same as chat-template crashes on tool-call assistant messages." }, { "left": "issue:43931", "right": "issue:45216", "accept": false, - "reason": "Model shape mismatch on load and incorrect checkpoint saving are separate serialization/load-path problems." + "reason": "Qwen3-VL weight-shape mismatch and Qwen3.5 save_pretrained regression are different load/save failures." }, { "left": "issue:34689", "right": "issue:36010", "accept": false, - "reason": "Both mention loading/import breakage, but one is a model-load regression and the other is a GenerationMixin import path issue." + "reason": "Llama 3.2 Vision loading and GenerationMixin import failure are unrelated import/config issues." }, { "left": "issue:44442", "right": "issue:45803", "accept": false, - "reason": "Tokenizer load failure and Fuyu tensor conversion exception swallowing are unrelated." + "reason": "Tokenizer loading and Fuyu batch tensor conversion error are different subsystems and symptoms." }, { "left": "issue:43550", "right": "issue:44743", "accept": false, - "reason": "torch.compile SDPA failure in Bamba and recurrent-state reset in modular_qwen3_5 are different model/runtime bugs." + "reason": "Bamba torch.compile/SDPA failure is unrelated to Qwen3.5 recurrent-state cache reset." }, { "left": "issue:43493", "right": "issue:43827", "accept": false, - "reason": "SigLIP2 implementation mismatch and docs still mentioning pipeline() are unrelated." - }, - { - "left": "issue:43873", - "right": "issue:44871", - "accept": false, - "reason": "Quantization offloading behavior and Gemma3 eos_token_id mismatch are not the same underlying issue." + "reason": "SigLIP2 implementation divergence is not the same as docs still referencing removed pipeline()." }, { "left": "issue:42907", "right": "issue:45698", "accept": false, - "reason": "Both touch save/load flows, but one is about saving dequantized Ministrals/Devstrals and the other about loading the wrong custom module after save_pretrained; too different to merge." + "reason": "Dequantized save failure for Ministral/Devstral is unrelated to wrong custom-module loading after save_pretrained." }, { "left": "issue:42175", "right": "issue:46234", "accept": false, - "reason": "Backend dependency installation and a missing import_utils helper are unrelated." + "reason": "TensorFlow missing from a torch extra install and a missing import_utils symbol are different packaging/API issues." }, { - "left": "issue:43761", - "right": "issue:44188", + "left": "issue:43873", + "right": "issue:44871", "accept": false, - "reason": "CLIPVisionModel hidden_states regression and attention-kernel divergence under torch.compile are different bugs." + "reason": "Quantization offloading behavior is unrelated to Gemma-3 eos_token_id config inconsistency." }, { "left": "issue:43643", "right": "issue:44913", "accept": false, - "reason": "Missing fields with trust_remote_code and rotary_pct not persisting on reload are distinct config-handling problems." + "reason": "trust_remote_code missing fields and GPTNeoX rotary_pct reload fallback are different config-loading bugs." + }, + { + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "CLIPVision hidden_states regression is unrelated to attention kernel divergence under torch.compile." }, { "left": "issue:43316", "right": "issue:45759", "accept": false, - "reason": "Gemma3TextConfig API discrepancy and Qwen multimodal text_config unwrapping are related only at a very high level; they are not the same concrete defect." + "reason": "Gemma3TextConfig API discrepancy and Qwen multimodal text_config unwrapping are different config-shape issues." }, { "left": "issue:45290", "right": "issue:46209", "accept": false, - "reason": "Chat-template tool-call crash is unrelated to FP8 inference failure on B200." - }, - { - "left": "issue:44855", - "right": "issue:45137", - "accept": false, - "reason": "Python 3.13 Deberta import parsing error and DeepSpeed ZeRO3 deque IndexError are unrelated." + "reason": "Chat template tool-call crash and FP8 inference failure on B200 are unrelated failure modes." }, { "left": "issue:33290", "right": "issue:34689", "accept": false, - "reason": "Adafactor/DeepSpeed OOM and Llama 3.2 loading regression are unrelated." + "reason": "Deepspeed Adafactor OOM and Llama 3.2 Vision loading error are not the same bug." }, { "left": "issue:44190", "right": "issue:44496", "accept": false, - "reason": "Local dataset loading in an example script and unrecognized model type in config.json are different failures." + "reason": "Local dataset loading in a training script and unrecognized model/config.json errors are unrelated." }, { "left": "issue:44568", "right": "issue:45479", "accept": false, - "reason": "Tokenizer add_special_tokens regression and sequence-classification zero-loss bug are unrelated." + "reason": "mDeBERTa special-token regression and sequence-classification zero-loss bug affect different model/task paths." }, { "left": "issue:44077", "right": "issue:44479", "accept": false, - "reason": "PatchTSMixer post_init validation and Qwen VL video-input regression are unrelated." + "reason": "patchtsmixer post_init policy and Qwen video-input regression are unrelated model-specific changes." }, { - "left": "issue:43873", - "right": "issue:44871", + "left": "issue:44855", + "right": "issue:45137", + "accept": false, + "reason": "TorchScript import IndentationError and DeepSpeed ZeRO3 deque error are completely different runtime failures." + }, + { + "left": "issue:43650", + "right": "issue:46202", "accept": false, - "reason": "Quantization offloading behavior and Gemma3 token-ID inconsistency are different issues." + "reason": "Vague data request and Gemma pipeline roundtrip failure on \u2581 are not the same issue." } ] }, @@ -9530,13 +9556,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 30, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:25251", "issue:28282", @@ -9555,12 +9581,13 @@ "issue:43441", "issue:43452", "issue:43475", + "issue:43504", "issue:43540", "issue:43550", + "issue:43572", "issue:43575", "issue:43606", "issue:43646", - "issue:43650", "issue:43701", "issue:43716", "issue:43742", @@ -9580,29 +9607,28 @@ "issue:45200", "issue:45584", "issue:45656", + "issue:45698", "issue:45803", "issue:46050", "issue:46170", - "issue:46178", - "issue:46202" + "issue:46178" ], "soft_pairs": [ - "issue:43650|issue:46202", - "issue:44987|issue:46178", "issue:29127|issue:33357", - "issue:25251|issue:28282", + "issue:44987|issue:46178", "issue:43295|issue:43825", + "issue:25251|issue:28282", "issue:43550|issue:43701", "issue:42617|issue:43575", "issue:43606|issue:43701", + "issue:43540|issue:44514", "issue:33290|issue:35141", "issue:43575|issue:43742", - "issue:43540|issue:44514", "issue:33290|issue:45656", "issue:43316|issue:43716", - "issue:44871|issue:45584", - "issue:44162|issue:44792", "issue:43404|issue:43441", + "issue:44162|issue:44792", + "issue:44871|issue:45584", "issue:43316|issue:44877", "issue:43295|issue:44871", "issue:45803|issue:46170", @@ -9610,201 +9636,209 @@ "issue:41950|issue:43475", "issue:43646|issue:44913", "issue:44871|issue:46050", - "issue:43452|issue:44496", "issue:45127|issue:45200", + "issue:43452|issue:44496", "issue:43761|issue:44479", - "issue:43232|issue:43388" + "issue:43232|issue:43388", + "issue:43504|issue:45698", + "issue:43572|issue:46178" ], - "prepared_review_unit_hash": "68d17b440e64b82f630c97bfa68be107d12449be8bc1803168485d46ea548e43", + "prepared_review_unit_hash": "5920f4656bb8c5b20e6462e925a8c543f3fa35ff8bbd324fd5d59451a2456851", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 27, - "serialized_chars": 35132, - "estimated_input_tokens": 8783, - "estimated_eval_tokens": 17822 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36078, + "estimated_input_tokens": 9020, + "estimated_eval_tokens": 18296 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a grab bag of unrelated Transformers issues; the soft links are mostly title-level similarity only, so none should be merged as duplicates.", - "confidence": 0.92, - "canonical_issue_reason": "If a canonical has to be chosen, issue:44871 is the least-bad anchor because it is concrete, reproducible, and appears in multiple soft-link candidates, but the overall cluster is too heterogeneous to support a strong canonical.", + "summary": "No true duplicate cluster emerges here: the soft edges span unrelated bugs across loading, pipelines, configs, multimodal processors, and training/runtime issues. I would anchor the cluster on the broad from_pretrained/save_pretrained regression, but I accept no soft merges.", + "confidence": 0.88, + "canonical_issue_reason": "issue:45698 is the broadest, most representative user-facing loading/serialization regression in the set, with the most discussion and inbound references.", "canonical_pr_reason": null, - "best_issue_reason": "Issue:44871 is the most actionable representative here: a specific Gemma config mismatch with clear symptoms, unlike the broader or more ambiguous neighboring issues.", + "best_issue_reason": "issue:45698 best represents the cluster because it is a general from_pretrained/save_pretrained correctness bug with wide practical impact.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43650", - "right": "issue:46202", + "left": "issue:29127", + "right": "issue:33357", "accept": false, - "reason": "Unrelated: a placeholder/open-ended issue versus a Gemma pipeline roundtrip bug around a special token." + "reason": "Different bugs: LayoutLMv3 error clarity vs CLIP bus error on macOS." }, { "left": "issue:44987", "right": "issue:46178", "accept": false, - "reason": "Different failure modes and code paths: model loading for one repo/model versus backend registration for tensorflow_text." + "reason": "Different failure modes: model loading regression vs unsupported backend mapping for tensorflow_text." }, { - "left": "issue:29127", - "right": "issue:33357", + "left": "issue:43295", + "right": "issue:43825", "accept": false, - "reason": "Distinct bugs: LayoutLMv3 box-content error versus a MacOS bus error in CLIP community model loading." + "reason": "Different code paths: processor/tokenizer regression vs pipeline translation-task messaging." }, { "left": "issue:25251", "right": "issue:28282", "accept": false, - "reason": "Not the same bug: pipeline top_k output nesting versus missing PyTorch import for AutoModel." - }, - { - "left": "issue:43295", - "right": "issue:43825", - "accept": false, - "reason": "Both are regressions, but one is custom model code accessing processor.tokenizer while the other is a pipeline error-message/translation-support issue." + "reason": "Unrelated: pipeline top_k nesting bug vs missing PyTorch ImportError." }, { "left": "issue:43550", "right": "issue:43701", "accept": false, - "reason": "Completely different problems: torch.compile/SDPA OOM versus resume_from_checkpoint key mismatch." + "reason": "Different problems: torch.compile/SDPA failure vs checkpoint key mismatch." }, { "left": "issue:42617", "right": "issue:43575", "accept": false, - "reason": "Different symptoms and likely different paths: a 3d_parallel.py runtime issue versus tensor-parallel OOM on a specific Qwen model." + "reason": "Different load/runtime issues: 3d_parallel.py failure vs tensor-parallel OOM." }, { "left": "issue:43606", "right": "issue:43701", "accept": false, - "reason": "Device mismatch with CPU offload is unrelated to checkpoint key mismatches." + "reason": "Different bugs: CPU offload device mismatch vs resume_from_checkpoint key mismatch." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Both multimodal, but different paths: Qwen3OmniMoe video input processing vs batched chat template padding crash." }, { "left": "issue:33290", "right": "issue:35141", "accept": false, - "reason": "Different bugs: DeepSpeed Adafactor OOM versus token embedding reinitialization during post_init." + "reason": "Different areas: DeepSpeed/Adafactor OOM vs embedding reinitialization after resize." }, { "left": "issue:43575", "right": "issue:43742", "accept": false, - "reason": "Both are load-time issues, but one is tensor-parallel OOM and the other is a KeyError while loading MobileLLM." - }, - { - "left": "issue:43540", - "right": "issue:44514", - "accept": false, - "reason": "Both involve multimodal/video processing, but the failures are different processor/code-path bugs and not clearly mergeable." + "reason": "Different load failures: huge-model OOM vs missing key for MobileLLM." }, { "left": "issue:33290", "right": "issue:45656", "accept": false, - "reason": "Related to DeepSpeed, but one is an OOM and the other is double optimizer stepping; not the same underlying defect." + "reason": "Both involve DeepSpeed, but one is OOM while the other is duplicate optimizer stepping." }, { "left": "issue:43316", "right": "issue:43716", "accept": false, - "reason": "Different models and different issues: Gemma3TextConfig API discrepancy versus image-preprocessor dtype mismatch in Mistral-3." + "reason": "Different issues: config API discrepancy vs image preprocessor/model dtype mismatch." }, { - "left": "issue:44871", - "right": "issue:45584", + "left": "issue:43404", + "right": "issue:43441", "accept": false, - "reason": "Both are tokenization/generation edge cases, but one is Gemma eos_token_id inconsistency and the other is Whisper empty-transcription handling after align_special_tokens." + "reason": "Different model bugs: lm_head tying vs FlashAttention failure." }, { "left": "issue:44162", "right": "issue:44792", "accept": false, - "reason": "Unrelated: ESM2 breakage versus a Janus image-generation test failure." + "reason": "Unrelated models and symptoms: ESM2 breakage vs janus image-generation test failure." }, { - "left": "issue:43404", - "right": "issue:43441", + "left": "issue:44871", + "right": "issue:45584", "accept": false, - "reason": "Different model-specific regressions: lm_head weight tying versus FlashAttention failure in Ministral-3." + "reason": "Different token/generation bugs: Gemma eos_token config mismatch vs Whisper empty transcription regression." }, { "left": "issue:43316", "right": "issue:44877", "accept": false, - "reason": "Config-related issues, but not the same defect: API mismatch in Gemma3TextConfig versus strict config validation blocking granite_speech." + "reason": "Different config issues: Gemma3TextConfig API mismatch vs granite_speech strict config load." }, { "left": "issue:43295", "right": "issue:44871", "accept": false, - "reason": "Both are regression-style model/processor bugs, but they affect different mechanisms: processor.tokenizer access versus Gemma eos_token_id configuration." + "reason": "Different regressions: custom processor/tokenizer handling vs Gemma eos_token_id inconsistency." }, { "left": "issue:45803", "right": "issue:46170", "accept": false, - "reason": "Unrelated: a bare except swallowing errors versus a Windows ROCm torch.distributed import/dependency problem." + "reason": "Different concerns: swallowed exceptions in Fuyu tensor conversion vs Windows ROCm torch.distributed dependency." }, { "left": "issue:43065", "right": "issue:44617", "accept": false, - "reason": "Same product area (SAM3) but not the same issue: dummy Conv2d placeholder versus CUDA OOM." + "reason": "Different Sam3 issues: dummy Conv2d placeholder vs CUDA OOM." }, { "left": "issue:41950", "right": "issue:43475", "accept": false, - "reason": "Different subsystems and symptoms: video-classification pipeline processor selection versus missing fpn_position_embeddings on a SAM3 vision encoder output." + "reason": "Different multimodal bugs: pipeline processor selection vs missing fpn_position_embeddings attribute." }, { "left": "issue:43646", "right": "issue:44913", "accept": false, - "reason": "Both involve model init/config behavior, but one is a broad Transformers 5 custom-init breakage and the other is a GPTNeoX config field not persisting." + "reason": "Both are init/config-related, but the model-init breakage and rotary_pct reload bug are distinct." }, { "left": "issue:44871", "right": "issue:46050", "accept": false, - "reason": "Both concern save/load or config persistence, but one is Gemma eos_token_id inconsistency and the other is quantized model round-tripping after save_pretrained." + "reason": "Different serialization bugs: eos_token config mismatch vs quantized model save/load breakage." }, { - "left": "issue:43452", - "right": "issue:44496", + "left": "issue:45127", + "right": "issue:45200", "accept": false, - "reason": "Both are model-loading failures, but one is gguf_file handling and the other is a missing model_type in config.json." + "reason": "Different training/model-behavior issues: LoRA merge collapse vs missing mm_token_type_ids default." }, { - "left": "issue:45127", - "right": "issue:45200", + "left": "issue:43452", + "right": "issue:44496", "accept": false, - "reason": "Different feature areas: LoRA merge with extended vocabulary versus Gemma 4 multimodal token type IDs defaulting." + "reason": "Different loading failures: gguf handling vs missing model_type in config.json." }, { "left": "issue:43761", "right": "issue:44479", "accept": false, - "reason": "Different regressions in different models: CLIPVisionModel hidden_states behavior versus Qwen video-input handling." + "reason": "Different regressions in different models: CLIPVision hidden_states vs Qwen VL video input handling." }, { "left": "issue:43232", "right": "issue:43388", "accept": false, - "reason": "Completely different code paths: generation kwargs sync handling versus gather_for_metrics dropping tuple labels." + "reason": "Different utilities and symptoms: generation kwargs update vs metrics gathering dropping labels." + }, + { + "left": "issue:43504", + "right": "issue:45698", + "accept": false, + "reason": "Both are loading-related, but legacy preset-field handling and wrong custom-module loading are different root causes." + }, + { + "left": "issue:43572", + "right": "issue:46178", + "accept": false, + "reason": "Unrelated: missing StableLm pad_token_idx vs unsupported tensorflow_text backend mapping." } ] }, @@ -9820,23 +9854,22 @@ "component_index": 1, "component_count": 18, "review_unit_index": 31, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:35141", "issue:41720", "issue:42490", "issue:43065", + "issue:43116", "issue:43335", - "issue:43504", "issue:43540", "issue:43550", - "issue:43572", "issue:43644", "issue:43673", "issue:43761", @@ -9846,14 +9879,17 @@ "issue:43994", "issue:44060", "issue:44077", + "issue:44112", "issue:44190", "issue:44242", "issue:44462", "issue:44479", + "issue:44521", "issue:44534", "issue:44610", "issue:44617", "issue:44734", + "issue:44737", "issue:44743", "issue:44792", "issue:44855", @@ -9864,21 +9900,18 @@ "issue:45127", "issue:45245", "issue:45357", + "issue:45381", "issue:45542", "issue:45584", "issue:45588", "issue:45600", "issue:45632", - "issue:45698", "issue:45803", "issue:45907", "issue:45987", - "issue:46178", "issue:46234" ], "soft_pairs": [ - "issue:43504|issue:45698", - "issue:43572|issue:46178", "issue:44077|issue:44534", "issue:45542|issue:46234", "issue:43673|issue:44743", @@ -9889,219 +9922,242 @@ "issue:44190|issue:44964", "issue:44792|issue:45600", "issue:44479|issue:44734", - "issue:44855|issue:45588", "issue:43827|issue:43994", + "issue:44855|issue:45588", "issue:41720|issue:45245", "issue:43065|issue:44610", "issue:43931|issue:44060", "issue:45803|issue:45987", "issue:35141|issue:42490", - "issue:43335|issue:44855", "issue:43644|issue:44913", "issue:43976|issue:44855", + "issue:43335|issue:44855", "issue:43550|issue:44190", "issue:45632|issue:46234", "issue:44077|issue:44877", "issue:43540|issue:44734", + "issue:45127|issue:45357", "issue:44462|issue:45907", - "issue:45127|issue:45357" + "issue:44521|issue:45381", + "issue:43335|issue:44190", + "issue:44737|issue:45803", + "issue:44112|issue:45600", + "issue:43116|issue:44190" ], - "prepared_review_unit_hash": "975b58e3002b250fea734bef48467125f5152ce12b479fa9c3bb0134d0365cbe", + "prepared_review_unit_hash": "7b8e91d8e58f17f7729042704219caa70bb7ca6bf5bbd6f295048922775f2c2d", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 28, - "serialized_chars": 35603, - "estimated_input_tokens": 8901, - "estimated_eval_tokens": 18058 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37102, + "estimated_input_tokens": 9276, + "estimated_eval_tokens": 18808 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No soft pair looks like a true duplicate; the items mostly span unrelated bugs across different models, configs, loaders, and runtime paths. The strongest issue candidate is the open, heavily discussed custom-module loading regression.", - "confidence": 0.91, - "canonical_issue_reason": "issue:45698 is the strongest representative issue: it is open, has the most discussion activity, and describes a concrete, user-facing loading regression with clear reproduction potential.", + "summary": "The items are not a duplicate cluster; they span many unrelated bugs/regressions across different models, configs, and utilities. All soft-edge pairs are false positives on broad vocabulary overlap rather than the same underlying change or bug.", + "confidence": 0.98, + "canonical_issue_reason": "No single canonical issue fits this set because the issues are semantically unrelated; there is no clear duplicate-center artifact to anchor the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45698 is the best standalone issue to keep as canonical because it has broad impact, active discussion, and a focused bug report about incorrect module loading after save/load.", + "best_issue_reason": "No global best issue is appropriate for a duplicate cluster here. If one had to be picked as a representative, it would be a self-contained regression like issue:44534, but it does not represent the rest of the set.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43504", - "right": "issue:45698", - "accept": false, - "reason": "Different bugs: one is a legacy-field preset loading failure for BEiT semantic segmentation, the other is wrong custom-module resolution after save/load." - }, - { - "left": "issue:43572", - "right": "issue:46178", - "accept": false, - "reason": "Unrelated regressions: StableLm pad token config metadata vs an import-utils backend registration error for tensorflow_text." - }, { "left": "issue:44077", "right": "issue:44534", "accept": false, - "reason": "Different code paths: a restriction on optional post_init in patchtsmixer versus a v5 buffer initialization corruption issue." + "reason": "Both mention v5 behavior changes, but one is about optional post_init in patchtsmixer and the other about non-persistent buffers being junk; different code paths and failures." }, { "left": "issue:45542", "right": "issue:46234", "accept": false, - "reason": "Not the same problem: TensorBoard/TensorFlow backend setup versus a missing import_utils availability helper." + "reason": "Completely different areas: TensorBoard/TensorFlow backend installation vs a missing import utility for Falcon Mamba SSM." }, { "left": "issue:43673", "right": "issue:44743", "accept": false, - "reason": "Both involve cache behavior, but they target different mechanisms and models: chunked_prefill cache presence vs recurrent state reset in modular_qwen3_5." + "reason": "Both involve cache/state handling, but one is a GenerationMixin chunked-prefill cache bug and the other is Qwen3.5 recurrent state reset logic; not the same defect." }, { "left": "issue:43761", "right": "issue:44242", "accept": false, - "reason": "Different model behaviors: CLIP hidden_states output regression versus missing load-balancing loss when router logits are disabled." + "reason": "Unrelated model forward regression in CLIPVisionModel vs MoE load-balancing loss omission." }, { "left": "issue:44792", "right": "issue:44977", "accept": false, - "reason": "Separate issues: a failing janus image-generation test versus Qwen3.5 flash-attention generation malfunction." + "reason": "Different models and symptoms: a janus image-generation test failure vs Qwen3.5 flash-attention generation issues." }, { "left": "issue:44617", "right": "issue:45584", "accept": false, - "reason": "No shared underlying bug: Sam3Video CUDA OOM is unrelated to Whisper empty-transcription generation failure after token alignment." + "reason": "CUDA OOM in Sam3Video is not the same as Whisper empty-transcription failure after align_special_tokens." }, { "left": "issue:44534", "right": "issue:44913", "accept": false, - "reason": "Different root causes: junk-filled non-persistent buffers versus GPTNeoX rotary_pct not persisting on reload." + "reason": "One is a v5 non-persistent buffer initialization regression; the other is GPTNeoXConfig rotary_pct not persisting on reload." }, { "left": "issue:44190", "right": "issue:44964", "accept": false, - "reason": "Unrelated failures: local dataset loading in a training script versus loading a multimodal model with latest transformers." + "reason": "Dataset loading in an example script is unrelated to loading a specific multimodal model." }, { "left": "issue:44792", "right": "issue:45600", "accept": false, - "reason": "One is a model test failure, the other is a stale auto-mapping reference to removed Sam3LiteText configs breaking CI." + "reason": "A failed janus generation test and stale auto_mappings references to removed configs are different problems." }, { "left": "issue:44479", "right": "issue:44734", "accept": false, - "reason": "Different failure modes and code paths: video-input regression in several Qwen variants versus KV-cache continuation indexing in transformers serve." + "reason": "Different subsystems: video-input regression for Qwen VL models vs serve/KV-cache tensor indexing." }, { - "left": "issue:44855", - "right": "issue:45588", + "left": "issue:43827", + "right": "issue:43994", "accept": false, - "reason": "Unrelated issues: Python 3.13 import parsing failure in DebertaV2Model versus flash_attention sink-less model AttributeError." + "reason": "Docs still referencing pipeline() is not the same as a broken pipeline inference result for SigLIP2." }, { - "left": "issue:43827", - "right": "issue:43994", + "left": "issue:44855", + "right": "issue:45588", "accept": false, - "reason": "Docs still referencing removed pipeline() calls is not the same as incorrect inference results for SigLIP2 AutoModel/pipeline usage." + "reason": "Python 3.13 import parsing error in DebertaV2 is unrelated to flash_attention.py crashing on s_aux=None." }, { "left": "issue:41720", "right": "issue:45245", "accept": false, - "reason": "Different symptoms and subsystems: Qwen3 auto device mapping CUDA assert versus a category-count RuntimeError." + "reason": "Qwen3 auto device mapping CUDA assert and a categories-limit runtime error are unrelated." }, { "left": "issue:43065", "right": "issue:44610", "accept": false, - "reason": "Not duplicates: a dummy Conv2d in Sam3PixelDecoder versus a processor/model input-size mismatch in OmDet-Turbo." + "reason": "A dummy Conv2d in Sam3PixelDecoder is not the same as OmDet-Turbo processor/model input-size mismatch." }, { "left": "issue:43931", "right": "issue:44060", "accept": false, - "reason": "Different regressions: Qwen3-VL weight shape mismatch on load versus an incorrect tied-weights warning involving dt_bias." + "reason": "Weight-shape mismatch while loading Qwen3-VL is unrelated to an incorrect tied-weights warning in Qwen3-Next." }, { "left": "issue:45803", "right": "issue:45987", "accept": false, - "reason": "One concerns exception handling in tensor conversion, the other a return-type contract bug in StaticCache.get_seq_length()." + "reason": "A broad exception swallowing KeyboardInterrupt in FuyuBatchFeature is unrelated to StaticCache.get_seq_length type mismatch." }, { "left": "issue:35141", "right": "issue:42490", "accept": false, - "reason": "Different areas entirely: token embedding reinitialization after resize versus a trailing-slash path import warning." - }, - { - "left": "issue:43335", - "right": "issue:44855", - "accept": false, - "reason": "Unrelated model/config bug versus a Python 3.13 parser/import problem caused by @torch.jit.script comment placement." + "reason": "Token embedding resizing/post_init reinit bug has nothing to do with local model path trailing-slash import warnings." }, { "left": "issue:43644", "right": "issue:44913", "accept": false, - "reason": "Separate v5 regression classes: non-persistent buffer junk versus GPTNeoX config serialization losing rotary_pct." + "reason": "Non-persistent buffer junk initialization and GPTNeoX rotary_pct reload persistence are different regressions." }, { "left": "issue:43976", "right": "issue:44855", "accept": false, - "reason": "No overlap: Python-version compatibility regression versus a DebertaV2 import parsing error." + "reason": "Python-version support breakage is not the same as an IndentationError from a TorchScript-decorated function comment." + }, + { + "left": "issue:43335", + "right": "issue:44855", + "accept": false, + "reason": "SwitchTransformers sparse-layer config bug is unrelated to DebertaV2 import parsing on Python 3.13." }, { "left": "issue:43550", "right": "issue:44190", "accept": false, - "reason": "Different subsystems: SDPA/torch.compile failure in Bamba versus dataset loading in a training script." + "reason": "Bamba torch.compile/SDPA failure and local dataset loading are entirely different code paths." }, { "left": "issue:45632", "right": "issue:46234", "accept": false, - "reason": "Cache-path collision for trust_remote_code local models is unrelated to a missing import_utils availability symbol." + "reason": "A trust_remote_code cache-path collision is not the same as a missing import_utils symbol." }, { "left": "issue:44077", "right": "issue:44877", "accept": false, - "reason": "Different strictness bugs: patchtsmixer post_init policy versus granite_speech config loading under strict config checks." + "reason": "Optional post_init policy for patchtsmixer and strict config rejection of granite_speech are unrelated." }, { "left": "issue:43540", "right": "issue:44734", "accept": false, - "reason": "Different code paths: Qwen3OmniMoe video-input processing error versus transformers serve KV-cache continuation indexing." + "reason": "Video input validation in Qwen3OmniMoe does not match a KV-cache continuation tensor-indexing crash in serve." + }, + { + "left": "issue:45127", + "right": "issue:45357", + "accept": false, + "reason": "LoRA merge with extended vocabulary and incorrect visual-encoder save_pretrained keys are different model-state bugs." }, { "left": "issue:44462", "right": "issue:45907", "accept": false, - "reason": "Not the same bug: tokenizer.json selection logic versus list_repo_files failing to forward revision." + "reason": "AutoTokenizer ignoring tokenizer.json is unrelated to list_repo_files failing to forward revision." }, { - "left": "issue:45127", - "right": "issue:45357", + "left": "issue:44521", + "right": "issue:45381", + "accept": false, + "reason": "All-zero assistant_masks for multimodal inputs and wrong vision_position_ids for qwen2.5-vl video inputs are not the same bug." + }, + { + "left": "issue:43335", + "right": "issue:44190", + "accept": false, + "reason": "A SwitchTransformers config edge case is unrelated to a local dataset loading failure in an example script." + }, + { + "left": "issue:44737", + "right": "issue:45803", + "accept": false, + "reason": "CPU device placement bug in XLNet positional encoding is unrelated to a bare-except error swallowing KeyboardInterrupt." + }, + { + "left": "issue:44112", + "right": "issue:45600", + "accept": false, + "reason": "CI stale test failure and auto_mappings referencing removed configs are different maintenance issues." + }, + { + "left": "issue:43116", + "right": "issue:44190", "accept": false, - "reason": "Distinct issues: LoRA merging with extended vocabulary on tied embeddings versus incorrect visual encoder keys during save_pretrained." + "reason": "Empty multi-label results in an image-classification example are not the same as local dataset loading problems." } ] }, @@ -10117,13 +10173,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 32, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:36246", "issue:38175", @@ -10131,58 +10187,53 @@ "issue:41628", "issue:42907", "issue:43065", - "issue:43116", "issue:43295", - "issue:43335", + "issue:43388", "issue:43404", "issue:43441", "issue:43479", + "issue:43526", "issue:43531", "issue:43643", "issue:43650", + "issue:43688", "issue:43854", + "issue:43901", "issue:44112", "issue:44186", - "issue:44190", "issue:44220", "issue:44246", "issue:44464", "issue:44496", - "issue:44521", "issue:44561", "issue:44683", "issue:44734", - "issue:44737", "issue:44933", "issue:44938", "issue:44964", "issue:45003", + "issue:45042", "issue:45216", "issue:45245", "issue:45276", + "issue:45335", "issue:45362", "issue:45375", - "issue:45381", "issue:45446", "issue:45542", "issue:45563", "issue:45584", + "issue:45588", "issue:45600", "issue:45632", - "issue:45803", "issue:46082", "issue:46121" ], "soft_pairs": [ - "issue:44521|issue:45381", - "issue:43335|issue:44190", - "issue:44737|issue:45803", - "issue:44112|issue:45600", - "issue:43116|issue:44190", "issue:46082|issue:46121", "issue:43479|issue:45375", - "issue:45276|issue:45584", "issue:44938|issue:45542", + "issue:45276|issue:45584", "issue:43295|issue:45276", "issue:45245|issue:45584", "issue:44464|issue:45563", @@ -10199,220 +10250,225 @@ "issue:44964|issue:45600", "issue:44112|issue:44964", "issue:45563|issue:45584", - "issue:44734|issue:45632", "issue:36246|issue:38617", + "issue:44734|issue:45632", "issue:44561|issue:45632", "issue:42907|issue:43441", - "issue:43531|issue:45600" - ], - "prepared_review_unit_hash": "812d09aafe246ff508aaaa37560f0acb15e07d014ede82964cac2d2990ec1913", - "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "issue:43531|issue:45600", + "issue:43526|issue:43901", + "issue:43295|issue:45335", + "issue:43441|issue:45588", + "issue:44496|issue:45042", + "issue:43388|issue:43688" + ], + "prepared_review_unit_hash": "d4c0fba6fb770d0493b52b9fb911d491930ebd85db585f23a365c6e8d1087afc", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36142, - "estimated_input_tokens": 9036, - "estimated_eval_tokens": 18328 + "serialized_chars": 36137, + "estimated_input_tokens": 9035, + "estimated_eval_tokens": 18326 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The candidate pairs are mostly broad Transformers issues that share only subsystem-level similarity, not the same concrete bug. I would not merge any of the soft edges, and there is no clear single canonical issue in this heterogeneous set.", - "confidence": 0.98, - "canonical_issue_reason": null, - "canonical_pr_reason": null, - "best_issue_reason": null, - "best_pr_reason": null, + "summary": "This cluster is heterogeneous: the pairs share only broad lexical similarity (imports, loading, config, generation, embeddings, CI) but not the same concrete bug. No soft-edge pair is safe to merge as a duplicate or same PR/change.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue is a good fit; the items span many unrelated regressions across different subsystems, so choosing one as the representative would be misleading.", + "canonical_pr_reason": "No PRs are present in this cluster.", + "best_issue_reason": "No issue is a strong global representative for duplicate triage. If forced to pick a nearby theme, issue 44964 is the most user-facing model-loading regression, but it still does not unify the rest of the cluster.", + "best_pr_reason": "No PRs are present in this cluster.", "soft_edge_verdicts": [ - { - "left": "issue:44521", - "right": "issue:45381", - "accept": false, - "reason": "Both are multimodal/Qwen-related, but one is about assistant masks in chat templating and the other about video vision position ids; different failure modes and code paths." - }, - { - "left": "issue:43335", - "right": "issue:44190", - "accept": false, - "reason": "Completely unrelated bugs: SwitchTransformers config creation vs an image-classification example script failing to load a local dataset." - }, - { - "left": "issue:44737", - "right": "issue:45803", - "accept": false, - "reason": "XLNet device placement bug and a Fuyu exception-handling bug are unrelated code paths." - }, - { - "left": "issue:44112", - "right": "issue:45600", - "accept": false, - "reason": "Both mention CI, but one is a stale device-override test and the other is broken auto-mapping references to removed configs." - }, - { - "left": "issue:43116", - "right": "issue:44190", - "accept": false, - "reason": "Both touch example scripts, but one is a multi-label classification output bug and the other is a dataset-loading problem in a different script." - }, { "left": "issue:46082", "right": "issue:46121", "accept": false, - "reason": "Different config-validation bugs: head_dim rejection in LlamaConfig vs rope validation type handling." + "reason": "Different bugs: one is LlamaConfig validation around explicit head_dim, the other is rope-validation handling of a JSON-loaded list." }, { "left": "issue:43479", "right": "issue:45375", "accept": false, - "reason": "Both are multimodal config issues, but they involve different model families and different initialization/serialization bugs." + "reason": "Both are multimodal config issues, but they affect different config classes and failure modes; not the same underlying bug." }, { - "left": "issue:45276", - "right": "issue:45584", + "left": "issue:44938", + "right": "issue:45542", "accept": false, - "reason": "Unrelated issues: embedding resize propagation in gemma4 versus Whisper generation on empty transcription." + "reason": "Both mention backend/loading failures, but one is a Python 3.14 import/load issue and the other is a TensorBoard-without-TF backend error." }, { - "left": "issue:44938", - "right": "issue:45542", + "left": "issue:45276", + "right": "issue:45584", "accept": false, - "reason": "Python 3.14 import failure and TensorBoard/TensorFlow backend detection are distinct problems." + "reason": "Completely different code paths: Gemma resize_token_embeddings vs Whisper empty-transcription generation." }, { "left": "issue:43295", "right": "issue:45276", "accept": false, - "reason": "A processor/tokenizer regression is not the same as token-embedding resize not propagating in gemma4." + "reason": "Both involve embedding/tokenizer behavior, but one is a regression in processor/tokenizer access and the other is model embedding resizing; not the same bug." }, { "left": "issue:45245", "right": "issue:45584", "accept": false, - "reason": "Different bugs: category-cardinality runtime error vs Whisper empty-transcription generation failure." + "reason": "Unrelated issues: category-count limits versus Whisper generation on empty transcription." }, { "left": "issue:44464", "right": "issue:45563", "accept": false, - "reason": "Both are generation-related, but one is incorrect outputs from compiled forward and the other is a stale warning about num_return_sequences." + "reason": "Both are generation-related, but one is inconsistent compiled forward outputs and the other is a stale warning in paged generate; different defects." }, { "left": "issue:43650", "right": "issue:44220", "accept": false, - "reason": "The first is a vague data request; the second is a specific feature-extraction bug." + "reason": "The titles do not describe the same problem; one is a vague data request, the other is an fbank feature extraction bug." }, { "left": "issue:44186", "right": "issue:44734", "accept": false, - "reason": "Different crash sites and models: LayoutLMv2Tokenizer padding/NER vs server KV-cache continuation indexing." + "reason": "Different failures in different paths: LayoutLMv2 tokenizer padding/truncation vs transformers serve KV-cache continuation indexing." }, { "left": "issue:36246", "right": "issue:41628", "accept": false, - "reason": "Both are import errors, but for different symbols and different APIs; not the same missing-export bug." + "reason": "Both are import errors, but for different missing symbols in different modules; not the same underlying breakage." }, { "left": "issue:43404", "right": "issue:45216", "accept": false, - "reason": "Weight-tying in Mistral3 and checkpoint saving regression in Qwen3.5 are different model-specific issues." + "reason": "Different bugs: lm_head weight tying in Mistral3 vs Qwen3.5 save_pretrained checkpoint correctness." }, { "left": "issue:44246", "right": "issue:45003", "accept": false, - "reason": "Import-performance regression and unsafe sys.modules access are not the same underlying defect." + "reason": "Both touch import/runtime behavior, but one is slow import performance and the other is unsafe sys.modules access." }, { "left": "issue:44683", "right": "issue:45446", "accept": false, - "reason": "Both involve flex_attention/versioning, but one is a torch>=2.9 compiled failure and the other is an incorrect version check for AuxRequest import." + "reason": "Both involve flex_attention, but one is compiled flex_attention on newer torch and the other is a version-check bug for AuxRequest import; not the same fix." }, { "left": "issue:38175", "right": "issue:43065", "accept": false, - "reason": "Different models and unrelated symptoms: zero probabilities in SigLIP2 vs a dummy Conv2d in Sam3PixelDecoder." + "reason": "Completely different model behaviors: zero probabilities in SigLIP2 vs dummy Conv2d in Sam3PixelDecoder." }, { "left": "issue:44933", "right": "issue:45362", "accept": false, - "reason": "A missing import from image_utils and a Qwen3.5 chat crash are unrelated." + "reason": "One is an image_utils import reference issue, the other is a Qwen3.5-35B chat crash; no shared concrete code-path bug." }, { "left": "issue:43643", "right": "issue:44496", "accept": false, - "reason": "AutoConfig missing fields under trust_remote_code is unrelated to an unrecognized-model config.json error." + "reason": "Both are config/loading related, but one is trust_remote_code field loss and the other is an unrecognized model/config.json model_type error." }, { "left": "issue:43854", "right": "issue:44964", "accept": false, - "reason": "Different model-loading failures for different models; same general area but not the same bug." + "reason": "Both are model loading failures, but they concern different models and different causes; not mergeable as one bug." }, { "left": "issue:44964", "right": "issue:45600", "accept": false, - "reason": "A model-loading regression and broken auto-mapping references are unrelated." + "reason": "Different failure modes: Phi-4 multimodal loading versus stale auto_mappings references to removed Sam3LiteText configs." }, { "left": "issue:44112", "right": "issue:44964", "accept": false, - "reason": "CI test flakiness and a Phi-4-multimodal loading issue are different problems." + "reason": "CI test flakiness in GraniteSpeech is unrelated to loading Phi-4 multimodal with latest transformers." }, { "left": "issue:45563", "right": "issue:45584", "accept": false, - "reason": "A stale generate warning and Whisper empty-transcription failure are not the same code-path issue." + "reason": "Stale warning in paged generate is not the same as Whisper failing on empty transcription after align_special_tokens." }, { - "left": "issue:44734", - "right": "issue:45632", + "left": "issue:36246", + "right": "issue:38617", "accept": false, - "reason": "KV-cache continuation indexing in the server is unrelated to trust_remote_code cache-path collisions." + "reason": "Both are ImportError reports, but they reference different symbols and modules; not the same missing export." }, { - "left": "issue:36246", - "right": "issue:38617", + "left": "issue:44734", + "right": "issue:45632", "accept": false, - "reason": "Different import failures for different names; not a duplicate export regression." + "reason": "Different issues: server KV-cache tensor indexing versus trust_remote_code cache-path collisions." }, { "left": "issue:44561", "right": "issue:45632", "accept": false, - "reason": "Both involve trust_remote_code, but one breaks models due to a removed helper and the other is a local cache-path collision." + "reason": "Both involve trust_remote_code, but one is a removed utility breaking remote-code models and the other is a local cache-path collision; distinct bugs." }, { "left": "issue:42907", "right": "issue:43441", "accept": false, - "reason": "Ministral/Devstral save_pretrained behavior and FlashAttention failures are separate bugs." + "reason": "Ministral/Devstral save-dequantization and FlashAttention failures are unrelated." }, { "left": "issue:43531", "right": "issue:45600", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior and Sam3LiteText auto-mapping cleanup are unrelated." + "reason": "Qwen3-MoE sliding_window behavior is unrelated to Sam3LiteText config references in auto_mappings." + }, + { + "left": "issue:43526", + "right": "issue:43901", + "accept": false, + "reason": "BeitImageProcessorFast label reduction and TextClassificationPipeline docs mismatch are unrelated." + }, + { + "left": "issue:43295", + "right": "issue:45335", + "accept": false, + "reason": "Both mention embedding/tokenizer effects, but they target different models and different embedding tensors." + }, + { + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "Different flash-attention-related problems in different models/code paths; not the same underlying bug." + }, + { + "left": "issue:44496", + "right": "issue:45042", + "accept": false, + "reason": "Unrelated model/config loading error versus image processor torchvision dependency regression." + }, + { + "left": "issue:43388", + "right": "issue:43688", + "accept": false, + "reason": "Different training/evaluation bugs: gather_for_metrics label truncation vs auxiliary-loss normalization." } ] }, @@ -10428,13 +10484,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 33, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:32090", "issue:35141", @@ -10445,18 +10501,17 @@ "issue:43295", "issue:43299", "issue:43381", - "issue:43388", - "issue:43441", "issue:43479", "issue:43525", - "issue:43526", "issue:43577", "issue:43650", "issue:43673", - "issue:43688", "issue:43824", + "issue:43854", "issue:43901", + "issue:43976", "issue:44016", + "issue:44062", "issue:44075", "issue:44220", "issue:44242", @@ -10466,10 +10521,10 @@ "issue:44368", "issue:44496", "issue:44617", + "issue:44683", "issue:44779", "issue:44933", "issue:45020", - "issue:45042", "issue:45200", "issue:45216", "issue:45278", @@ -10477,25 +10532,21 @@ "issue:45357", "issue:45372", "issue:45405", + "issue:45542", "issue:45584", - "issue:45588", "issue:45593", "issue:45759", "issue:45812", + "issue:45823", "issue:46209", "issue:46234" ], "soft_pairs": [ - "issue:43526|issue:43901", - "issue:43295|issue:45335", - "issue:43388|issue:43688", - "issue:44496|issue:45042", - "issue:43441|issue:45588", "issue:43673|issue:44336", "issue:41950|issue:44016", "issue:44075|issue:44368", - "issue:42175|issue:44279", "issue:43299|issue:45216", + "issue:42175|issue:44279", "issue:43381|issue:45593", "issue:43901|issue:44496", "issue:44016|issue:44220", @@ -10507,224 +10558,229 @@ "issue:43479|issue:43525", "issue:44242|issue:45200", "issue:43577|issue:45216", - "issue:38175|issue:44617", "issue:43299|issue:45357", - "issue:44779|issue:45812", - "issue:43295|issue:44779", + "issue:38175|issue:44617", "issue:43824|issue:44933", - "issue:39692|issue:43901", + "issue:44779|issue:45812", "issue:44279|issue:45405", + "issue:39692|issue:43901", "issue:44351|issue:45278", - "issue:41950|issue:45020" + "issue:41950|issue:45020", + "issue:43295|issue:44779", + "issue:44016|issue:44062", + "issue:41950|issue:45823", + "issue:43299|issue:43854", + "issue:43976|issue:45405", + "issue:44683|issue:45542" ], - "prepared_review_unit_hash": "d13d94114697c6c868db115248ff328c6768884b8b85d4c536e0e171d331c768", + "prepared_review_unit_hash": "ace6640e99a7322675b5eae934b072f5f5a0bccb48e55653fe42712d156ad88d", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36767, - "estimated_input_tokens": 9192, - "estimated_eval_tokens": 18640 + "serialized_chars": 36648, + "estimated_input_tokens": 9162, + "estimated_eval_tokens": 18580 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No soft pair looks like a true duplicate: the items span unrelated bugs across tokenizers, processors, configs, imports, training utilities, and docs. A broad version-regression import issue is the closest umbrella representative, but it does not unify the set.", - "confidence": 0.61, - "canonical_issue_reason": "issue:45278 is the broadest, most umbrella-like report here because it summarizes multiple import breakages after a Transformers version upgrade, making it the closest representative among otherwise unrelated issues.", + "summary": "This cluster is a loose mix of recent Transformers regressions, import/load failures, and unrelated model/docs bugs. The soft pairs do not look like true duplicates of the same concrete code-path issue, so none should be merged.", + "confidence": 0.31, + "canonical_issue_reason": "issue:45278 is the broadest and most representative user-facing regression in the set: an upgrade-triggered import breakage affecting many symbols. It fits the dominant recent-version breakage theme better than the narrower model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the most suitable cluster representative because it describes a wide regression class rather than a single narrow symptom.", + "best_issue_reason": "issue:45278 is the best triage anchor because it states a concrete, version-scoped failure mode and has the clearest broad scope among the issues listed.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43526", - "right": "issue:43901", - "accept": false, - "reason": "Different problems: image-processor label reduction bug vs docs still mentioning return_all_scores." - }, - { - "left": "issue:43295", - "right": "issue:45335", - "accept": false, - "reason": "Different code paths: multimodal processor.tokenizer regression vs t5gemma embedding resize not propagating." - }, - { - "left": "issue:43388", - "right": "issue:43688", - "accept": false, - "reason": "Different training bugs: last-batch metric gathering vs auxiliary-loss normalization." - }, - { - "left": "issue:44496", - "right": "issue:45042", - "accept": false, - "reason": "Different bugs: model config recognition failure vs PIL image processor requiring torchvision." - }, - { - "left": "issue:43441", - "right": "issue:45588", - "accept": false, - "reason": "Both mention flash attention, but they hit different failures/models and different code paths." - }, { "left": "issue:43673", "right": "issue:44336", "accept": false, - "reason": "Unrelated: generation cache/chunked prefill vs ANSI formatting in loading_report." + "reason": "One is a GenerationMixin cache regression; the other is ANSI formatting in loading_report. Different subsystems and failure modes." }, { "left": "issue:41950", "right": "issue:44016", "accept": false, - "reason": "Different domains: video-classification pipeline processor lookup vs a notebook syntax error." + "reason": "A pipeline processor-selection bug is unrelated to a notebook syntax error." }, { "left": "issue:44075", "right": "issue:44368", "accept": false, - "reason": "Different issues: SGD optimizer args ignored vs a configuration warning about tie_word_embeddings." + "reason": "SGD optimizer arguments being ignored is unrelated to a tie_word_embeddings warning." }, { - "left": "issue:42175", - "right": "issue:44279", + "left": "issue:43299", + "right": "issue:45216", "accept": false, - "reason": "Different dependency complaints: missing tensorflow in extras vs a generic transformers dependency issue." + "reason": "Qwen3VL model-loading regression and Qwen3.5 save_pretrained checkpoint corruption are different concrete bugs." }, { - "left": "issue:43299", - "right": "issue:45216", + "left": "issue:42175", + "right": "issue:44279", "accept": false, - "reason": "Different regressions: Qwen3VL MoE model loading vs Qwen3.5 save_pretrained checkpoint corruption." + "reason": "A missing TensorFlow backend install and a generic dependency issue are not the same underlying problem." }, { "left": "issue:43381", "right": "issue:45593", "accept": false, - "reason": "Different bugs: gradient checkpointing disallowed in eval vs D-FINE auxiliary losses omitted when denoising is off." + "reason": "Gradient checkpointing in eval mode and D-FINE auxiliary-loss behavior are unrelated." }, { "left": "issue:43901", "right": "issue:44496", "accept": false, - "reason": "Docs-only mismatch vs a model loading/config error; not the same bug." + "reason": "A documentation mismatch in TextClassificationPipeline is unrelated to an unrecognized model/config loading error." }, { "left": "issue:44016", "right": "issue:44220", "accept": false, - "reason": "Notebook syntax error vs audio feature extraction bug; unrelated." + "reason": "Notebook syntax error and _torch_extract_fbank_features() failure do not share the same bug." }, { "left": "issue:44496", "right": "issue:45759", "accept": false, - "reason": "Different model-instantiation failures: generic unrecognized model vs missing text_config unwrapping for composite Qwen multimodal configs." + "reason": "Unrecognized model/config loading and composite multimodal config unwrapping are different code paths." }, { "left": "issue:43650", "right": "issue:46209", "accept": false, - "reason": "No clear relation: placeholder title vs FP8 inference failure on B200." + "reason": "A placeholder request is unrelated to an FP8 inference failure on B200." }, { "left": "issue:45372", "right": "issue:46234", "accept": false, - "reason": "Different missing symbols/import surfaces: ReasoningEffort import error vs falcon-mamba availability helper missing." + "reason": "Gemma processor loading import failure and a missing falcon_mamba utility export are separate import bugs." }, { "left": "issue:45335", "right": "issue:45584", "accept": false, - "reason": "Different bugs: token embedding resize propagation vs Whisper empty-transcription generation failure." + "reason": "Tokenizer/embedding resize behavior is unrelated to Whisper empty-transcription generation after token alignment." }, { "left": "issue:32090", "right": "issue:35141", "accept": false, - "reason": "Both involve None/tie_word_embedding themes, but one is Trainer GPU broadcast of None and the other is post_init reinitialization." + "reason": "Trainer _gpu_broadcast_one NoneType error and embedding reinitialization after resize are different failures." }, { "left": "issue:43479", "right": "issue:43525", "accept": false, - "reason": "Different config problems: default vision/audio init from None vs missing pad_token_id attribute." + "reason": "Default multimodal config initialization and missing pad_token_id are unrelated config issues." }, { "left": "issue:44242", "right": "issue:45200", "accept": false, - "reason": "Different mechanisms: router-logit load balancing loss vs mm_token_type_ids defaulting for text-only fine-tuning." + "reason": "Missing load-balancing loss and mm_token_type_ids defaulting are different modeling/training bugs." }, { "left": "issue:43577", "right": "issue:45216", "accept": false, - "reason": "Different regressions: dtype not applied to Blip2 submodules vs incorrect Qwen3.5 save_pretrained output." + "reason": "Dtype loading for Blip2 and incorrect Qwen3.5 save_pretrained output are unrelated regressions." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Both are Qwen3.5/Qwen3VL-related, but one is loading failure and the other is save_pretrained key serialization; not the same bug." }, { "left": "issue:38175", "right": "issue:44617", "accept": false, - "reason": "Different symptoms and subsystems: unexpected zero probabilities vs CUDA out-of-memory." + "reason": "Zero probabilities in SigLIP2 and CUDA OOM in Sam3Video are completely different issues." }, { - "left": "issue:43299", - "right": "issue:45357", + "left": "issue:43824", + "right": "issue:44933", "accept": false, - "reason": "Same model family but different concrete bugs: loading failure vs save_pretrained writing wrong visual-encoder keys." + "reason": "Both are import errors, but they involve different missing symbols and different modules." }, { "left": "issue:44779", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different model families and likely different fixes." + "reason": "Both are tokenizer regressions, but for different model families and likely different code paths." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "A dependency installation problem is not the same as bumping MIN_PEFT_VERSION to an unreleased version." + }, + { + "left": "issue:39692", + "right": "issue:43901", + "accept": false, + "reason": "Separate documentation problems in different sections with different concrete defects." + }, + { + "left": "issue:44351", + "right": "issue:45278", + "accept": false, + "reason": "A missing HybridCache export may be related to broader import breakage, but the evidence is too weak to treat it as the same concrete bug as the multi-import regression." + }, + { + "left": "issue:41950", + "right": "issue:45020", + "accept": false, + "reason": "A video-classification processor lookup bug is not the same as a broad remote_code compatibility regression." }, { "left": "issue:43295", "right": "issue:44779", "accept": false, - "reason": "Processor/tokenizer-image handling regression vs DeepSeek tokenizer output bug; not the same issue." + "reason": "Processor/tokenizer API regression and Deepseek tokenization errors are different model-specific bugs." }, { - "left": "issue:43824", - "right": "issue:44933", + "left": "issue:44016", + "right": "issue:44062", "accept": false, - "reason": "Different import failures: missing Qwen2.5-VL class vs nonexistent image_utils import." + "reason": "Notebook syntax error and AddedToken constructor argument conflict are unrelated." }, { - "left": "issue:39692", - "right": "issue:43901", + "left": "issue:41950", + "right": "issue:45823", "accept": false, - "reason": "Different docs issues: SigLIP2 example errors vs TextClassificationPipeline return_all_scores documentation drift." + "reason": "Video-classification processor lookup and Gemma4 PLE device mismatch are unrelated." }, { - "left": "issue:44279", - "right": "issue:45405", + "left": "issue:43299", + "right": "issue:43854", "accept": false, - "reason": "Generic dependency problem vs a specific unreleased minimum PEFT version bump." + "reason": "Qwen3VL MoE loading and GLM-4.7-Flash unit-test loading are different model-specific failures." }, { - "left": "issue:44351", - "right": "issue:45278", + "left": "issue:43976", + "right": "issue:45405", "accept": false, - "reason": "Related by imports, but 45278 is a broad report while this is a specific HybridCache export failure; not clearly one bug." + "reason": "Python version compatibility and an unreleased PEFT version requirement are unrelated." }, { - "left": "issue:41950", - "right": "issue:45020", + "left": "issue:44683", + "right": "issue:45542", "accept": false, - "reason": "Different problems: video-classification pipeline processor lookup vs remote_code model breakage." + "reason": "Compiled flex_attention on torch>=2.9 and TensorBoard-without-TF backend errors have no shared underlying bug." } ] }, @@ -10740,42 +10796,42 @@ "component_index": 1, "component_count": 18, "review_unit_index": 34, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ - "issue:41950", "issue:42994", "issue:43065", - "issue:43299", "issue:43317", "issue:43441", "issue:43450", "issue:43525", + "issue:43643", "issue:43650", "issue:43653", "issue:43723", - "issue:43854", + "issue:43784", "issue:43881", "issue:43901", "issue:43976", "issue:43994", "issue:44016", - "issue:44062", "issue:44162", "issue:44246", "issue:44261", "issue:44263", "issue:44279", + "issue:44291", "issue:44297", "issue:44361", "issue:44462", "issue:44466", "issue:44521", + "issue:44617", "issue:44623", "issue:44671", "issue:44683", @@ -10787,7 +10843,6 @@ "issue:45276", "issue:45325", "issue:45397", - "issue:45405", "issue:45542", "issue:45593", "issue:45600", @@ -10798,18 +10853,13 @@ "issue:46132" ], "soft_pairs": [ - "issue:44016|issue:44062", - "issue:41950|issue:45823", - "issue:43976|issue:45405", - "issue:43299|issue:43854", - "issue:44683|issue:45542", - "issue:44462|issue:45081", "issue:43650|issue:46132", + "issue:44462|issue:45081", "issue:43976|issue:44987", "issue:44297|issue:44462", "issue:43976|issue:44279", - "issue:45081|issue:45907", "issue:43450|issue:44279", + "issue:45081|issue:45907", "issue:44521|issue:45325", "issue:44466|issue:45276", "issue:43441|issue:45834", @@ -10828,222 +10878,220 @@ "issue:43901|issue:43994", "issue:44297|issue:45397", "issue:43525|issue:43881", - "issue:43901|issue:44246" + "issue:43450|issue:44617", + "issue:43901|issue:44246", + "issue:43525|issue:43643", + "issue:43784|issue:45542", + "issue:44016|issue:44291" ], - "prepared_review_unit_hash": "8ded950a8648a019489007aedacc52603b58cd7de229e0c5ab9be8d2b6d1ec0f", + "prepared_review_unit_hash": "5620ea6855bfa79c5cab3185d15daaceabdf6310635dc4a615e570d1d39f600a", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 31, - "serialized_chars": 36740, - "estimated_input_tokens": 9185, - "estimated_eval_tokens": 18626 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 35943, + "estimated_input_tokens": 8986, + "estimated_eval_tokens": 18228 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: the issues span unrelated tokenizer, vision, loading, docs, and performance bugs. I found no soft pair that looks like the same underlying change or defect, so all soft edges are rejected.", - "confidence": 0.36, - "canonical_issue_reason": "Issue 41950 is the strongest standalone report in the set: it has the highest discussion activity and inbound references, and is a clear, concrete bug. It is only a representative choice, not evidence that the cluster is a true duplicate set.", + "summary": "This is a heterogeneous cluster of mostly unrelated closed issues across tokenizer loading/saving, model-specific regressions, multimodal preprocessing, config/serialization bugs, and CI/docs noise. None of the soft pairs look like true duplicates of the same concrete bug or change.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue for the full cluster; the only plausible anchor is issue 44462, but the rest of the items span too many unrelated failure modes to treat as duplicates of it.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 41950 is the best overall issue to represent this group because it is active, well-specified, and externally referenced, even though the cluster itself is not cohesive.", + "best_issue_reason": "Issue 44462 is the broadest tokenizer-loading report in the set and the closest thing to a representative issue, but it still does not cover the cluster as a whole.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44016", - "right": "issue:44062", + "left": "issue:43650", + "right": "issue:46132", "accept": false, - "reason": "Notebook syntax error and AddedToken keyword conflict are unrelated failures in different subsystems." + "reason": "Placeholder/low-signal issue vs an AttentionInterface.register behavior change; unrelated artifacts." }, { - "left": "issue:41950", - "right": "issue:45823", + "left": "issue:44462", + "right": "issue:45081", "accept": false, - "reason": "Video-classification processor lookup and Gemma4 device-mapping mismatch are different bugs." + "reason": "Both involve tokenizers, but one is AutoTokenizer ignoring tokenizer.json while the other is a Mistral regex patch crash; different code paths and bugs." }, { "left": "issue:43976", - "right": "issue:45405", + "right": "issue:44987", "accept": false, - "reason": "Python version compatibility and a PEFT version bump are unrelated dependency problems." + "reason": "Python version compatibility problem vs a model loading failure; no concrete shared bug." }, { - "left": "issue:43299", - "right": "issue:43854", + "left": "issue:44297", + "right": "issue:44462", "accept": false, - "reason": "Different model-loading regressions affecting different models and likely different code paths." + "reason": "Tokenizer save_pretrained class metadata mismatch vs AutoTokenizer ignoring tokenizer.json; adjacent area, but distinct failures." }, { - "left": "issue:44683", - "right": "issue:45542", + "left": "issue:43976", + "right": "issue:44279", "accept": false, - "reason": "FlexAttention on torch 2.9 and missing TensorFlow backend for tensorboard are unrelated." + "reason": "Version-support issue vs vague dependency issue; not the same underlying bug." }, { - "left": "issue:44462", - "right": "issue:45081", + "left": "issue:43450", + "right": "issue:44279", "accept": false, - "reason": "Both are tokenizer-loading issues, but they involve different mechanisms and different failure modes." + "reason": "Video processor batch-shape bug vs dependency issue; unrelated." }, { - "left": "issue:43650", - "right": "issue:46132", + "left": "issue:45081", + "right": "issue:45907", "accept": false, - "reason": "Generic placeholder request and AttentionInterface behavior change are not the same bug." - }, - { - "left": "issue:43976", - "right": "issue:44987", - "accept": false, - "reason": "Python compatibility issue and a model-loading failure for a specific repo are unrelated." - }, - { - "left": "issue:44297", - "right": "issue:44462", - "accept": false, - "reason": "Tokenizer save metadata mismatch is distinct from AutoTokenizer ignoring tokenizer.json on load." - }, - { - "left": "issue:43976", - "right": "issue:44279", - "accept": false, - "reason": "Dependency/version support issue and a generic dependency issue are not the same concrete defect." - }, - { - "left": "issue:45081", - "right": "issue:45907", - "accept": false, - "reason": "Mistral tokenizer regex patch crash and list_repo_files revision forwarding are unrelated." - }, - { - "left": "issue:43450", - "right": "issue:44279", - "accept": false, - "reason": "Video processor batch-shape bug and a generic dependency issue do not match." + "reason": "Mistral tokenizer regex crash vs list_repo_files revision forwarding; completely different functionality." }, { "left": "issue:44521", "right": "issue:45325", "accept": false, - "reason": "Assistant mask generation and temporal rope indexing are different multimodal bugs." + "reason": "Both multimodal, but assistant mask generation and temporal rope index scaling are different code paths." }, { "left": "issue:44466", "right": "issue:45276", "accept": false, - "reason": "Serialization of lm_head.weight and resize_token_embeddings propagation are different issues." + "reason": "Both touch embeddings/weights, but serialization of lm_head.weight and resize_token_embeddings updating modules are different bugs." }, { "left": "issue:43441", "right": "issue:45834", "accept": false, - "reason": "FlashAttention failure for Ministral-3 and OCR index error for Kosmos2.5 are unrelated model bugs." + "reason": "FlashAttention loading failure vs OCR long-input index error; unrelated model-specific issues." }, { "left": "issue:42994", "right": "issue:43317", "accept": false, - "reason": "Saving quantized models and loading dequantized models with offload are different code paths and failures." + "reason": "Quantized saving vs dequantized/offload loading; same broad area, but not the same concrete defect." }, { "left": "issue:43653", "right": "issue:45593", "accept": false, - "reason": "Tokenizer special-token registration and D-FINE auxiliary losses are unrelated." + "reason": "Tokenizer special-token registration bug vs D-FINE auxiliary-loss behavior; unrelated." }, { "left": "issue:45600", "right": "issue:45712", "accept": false, - "reason": "Removed config references in auto_mappings and leftover dummy classes are separate CI cleanups." + "reason": "Both are repo hygiene/auto-generated-file issues, but they affect different files and failures." }, { "left": "issue:44623", "right": "issue:44749", "accept": false, - "reason": "Missing processor save files and tokenizer/data filtering slowdown are unrelated." + "reason": "Missing files in processor.save_pretrained vs performance regression during data filtering; unrelated." }, { "left": "issue:44623", "right": "issue:45397", "accept": false, - "reason": "Processor save_pretrained file omission and gemma-4 zero3 loading are unrelated." + "reason": "Processor save bug vs Gemma4 zero3 loading bug; different code paths." }, { "left": "issue:43065", "right": "issue:44898", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder and Perceiver resize/interpolation failure are different vision bugs." + "reason": "Dummy Conv2d in Sam3PixelDecoder vs Perceiver interpolation failure; different models and errors." }, { "left": "issue:44261", "right": "issue:44263", "accept": false, - "reason": "LayerNorm epsilon config issue and torch.split return-value handling are unrelated." + "reason": "Missing rms_norm_eps causing precision drift vs torch.split return-value handling; unrelated math bugs." }, { "left": "issue:44361", "right": "issue:45593", "accept": false, - "reason": "MLukeTokenizer AttributeError and D-FINE denoising-loss behavior are unrelated." + "reason": "MLukeTokenizer AttributeError vs D-FINE auxiliary-loss issue; no shared underlying defect." }, { "left": "issue:44671", "right": "issue:44779", "accept": false, - "reason": "CamemBERT masked-LM regression and Deepseek tokenizer regression are different model-specific bugs." + "reason": "Both are v5 regressions, but one is CamemBERT masked-LM output corruption and the other is Deepseek tokenizer output corruption; not the same bug." }, { "left": "issue:44016", "right": "issue:45823", "accept": false, - "reason": "Notebook syntax error and Gemma4 device mismatch are unrelated." + "reason": "Notebook syntax error vs Gemma4 device mismatch; unrelated." }, { "left": "issue:43441", "right": "issue:44683", "accept": false, - "reason": "Both mention attention/flash attention, but they are distinct failures on different operators and versions." + "reason": "FlashAttention model failure vs compiled flex_attention failure on torch>=2.9; different implementations and causes." }, { "left": "issue:43723", "right": "issue:44162", "accept": false, - "reason": "Tokenizer loading regression and ESM2 model breakage are not the same underlying defect." + "reason": "Tokenizer loading regression vs ESM2 broken; both loading-related, but not the same failure mode." }, { "left": "issue:43901", "right": "issue:43994", "accept": false, - "reason": "Docs mismatch for pipeline output and incorrect model inference results are different issues." + "reason": "Docs mismatch for return_all_scores vs bad model outputs in pipeline; different problem class." }, { "left": "issue:44297", "right": "issue:45397", "accept": false, - "reason": "Tokenizer serialization metadata and gemma-4 zero3 loading are unrelated." + "reason": "Tokenizer serialization metadata bug vs Gemma4 zero3 loading bug; unrelated." }, { "left": "issue:43525", "right": "issue:43881", "accept": false, - "reason": "Missing pad_token_id on Llama4Config and glm-4v-9b loading failure are different model/config bugs." + "reason": "Missing pad_token_id in Llama4Config vs glm-4v-9b loading failure; different model/config issues." + }, + { + "left": "issue:43450", + "right": "issue:44617", + "accept": false, + "reason": "Incorrect batched video processor shape vs CUDA OOM in Sam3Video; not the same bug." }, { "left": "issue:43901", "right": "issue:44246", "accept": false, - "reason": "Pipeline docs mismatch and import-time slowdown are unrelated." + "reason": "Documentation mismatch vs intermittent import slowness; unrelated." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": false, + "reason": "Specific missing config field vs trust_remote_code returning incomplete config fields; related theme, but not the same concrete issue." + }, + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "Sentence-transformers import NameError vs missing TensorFlow backend handling; different optional-dependency paths." + }, + { + "left": "issue:44016", + "right": "issue:44291", + "accept": false, + "reason": "Notebook syntax error vs init_empty_weights TypeError from unexpected argument; unrelated." } ] }, @@ -11059,13 +11107,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 35, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:16998", "issue:22355", @@ -11078,20 +11126,17 @@ "issue:43262", "issue:43299", "issue:43352", - "issue:43450", - "issue:43525", - "issue:43643", + "issue:43526", "issue:43673", "issue:43761", - "issue:43784", "issue:43824", + "issue:43901", "issue:43976", - "issue:44016", + "issue:44162", "issue:44188", "issue:44220", "issue:44242", "issue:44246", - "issue:44291", "issue:44295", "issue:44336", "issue:44351", @@ -11101,33 +11146,31 @@ "issue:44521", "issue:44561", "issue:44589", - "issue:44617", "issue:44749", "issue:44779", "issue:44857", "issue:44871", "issue:44964", + "issue:45003", "issue:45081", + "issue:45200", "issue:45250", "issue:45292", "issue:45362", "issue:45397", - "issue:45542", + "issue:45412", "issue:45593", + "issue:46123", "issue:46132" ], "soft_pairs": [ - "issue:43450|issue:44617", - "issue:43525|issue:43643", "issue:44188|issue:44336", - "issue:43784|issue:45542", - "issue:44016|issue:44291", "issue:43262|issue:44521", - "issue:43673|issue:43761", "issue:41628|issue:43976", "issue:44589|issue:45362", - "issue:44336|issue:44373", + "issue:43673|issue:43761", "issue:16998|issue:22355", + "issue:44336|issue:44373", "issue:42491|issue:43299", "issue:44246|issue:44496", "issue:44295|issue:45397", @@ -11137,134 +11180,115 @@ "issue:43824|issue:43976", "issue:41950|issue:44220", "issue:43976|issue:44964", - "issue:36246|issue:42757", "issue:42886|issue:46132", + "issue:36246|issue:42757", "issue:36246|issue:44351", "issue:44779|issue:45292", "issue:44242|issue:44857", - "issue:43673|issue:44561", "issue:45081|issue:45362", + "issue:43673|issue:44561", "issue:44295|issue:44749", - "issue:43352|issue:45250" + "issue:43352|issue:45250", + "issue:44162|issue:45200", + "issue:43901|issue:45003", + "issue:43526|issue:45003", + "issue:44496|issue:45003", + "issue:45412|issue:46123" ], - "prepared_review_unit_hash": "c1dd519f264f0534cb8b1cb40e0352f0b2e3a0eaf0103e65154765f2ce5faad3", + "prepared_review_unit_hash": "780239c227bd63f1c777810292e1030f9b2a859a733a5cc6a83f8459de6eecd2", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36333, - "estimated_input_tokens": 9084, - "estimated_eval_tokens": 18424 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36013, + "estimated_input_tokens": 9004, + "estimated_eval_tokens": 18264 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose topical grouping of unrelated Transformers issues. The pairs share broad themes like imports, multimodal chat, or Qwen/MoE, but they do not describe the same concrete bug or mergeable fix.", - "confidence": 0.82, - "canonical_issue_reason": "issue:44521 is the strongest representative: it is a concrete, well-scoped multimodal `apply_chat_template` bug with active discussion and inbound references, making it the best anchor among these heterogeneous reports.", + "summary": "This cluster is heterogeneous, with mostly unrelated regression/import/model-loading issues that only loosely share Transformers-v5 breakage themes. None of the soft pairs looks like the same underlying bug or change.", + "confidence": 0.42, + "canonical_issue_reason": "Issue 44561 is the broadest core v5 regression here: a low-level API removal that breaks trust_remote_code models and best matches the cluster\u2019s general breakage/import theme.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 is the clearest and most actionable issue in the set; it has the most direct bug description and the strongest evidence of user impact.", + "best_issue_reason": "Issue 44561 is the best representative issue because it is a concrete, high-impact v5 compatibility regression affecting a wide class of models, making it the most central exemplar in this set.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43450", - "right": "issue:44617", - "accept": false, - "reason": "Different failures: batched video processor shape mismatch vs video-model OOM. Same broad area, not the same bug." - }, - { - "left": "issue:43525", - "right": "issue:43643", - "accept": false, - "reason": "One is a missing `pad_token_id` on `Llama4Config`; the other is `trust_remote_code` losing fields in `AutoConfig`. Related only at a high level." - }, { "left": "issue:44188", "right": "issue:44336", "accept": false, - "reason": "Attention-kernel divergence under `torch.compile` is unrelated to ANSI codes in `loading_report`." - }, - { - "left": "issue:43784", - "right": "issue:45542", - "accept": false, - "reason": "Both mention import-time breakage, but one is `nn` missing for sentence-transformers and the other is TensorFlow backend handling with only tensorboard installed." - }, - { - "left": "issue:44016", - "right": "issue:44291", - "accept": false, - "reason": "A notebook syntax error and an `init_empty_weights` TypeError are different problems with no shared fix path." + "reason": "Attention-kernel divergence on torch.compile is unrelated to ANSI escape codes in loading_report." }, { "left": "issue:43262", "right": "issue:44521", "accept": false, - "reason": "Both touch `apply_chat_template`, but one is wrong audio sampling-rate default and the other is all-zero assistant masks; different code paths and symptoms." - }, - { - "left": "issue:43673", - "right": "issue:43761", - "accept": false, - "reason": "Chunked-prefill cache handling and CLIP vision hidden-states propagation are unrelated." + "reason": "Audio apply_chat_template sampling-rate default bug is a different code path from multimodal assistant_mask generation." }, { "left": "issue:41628", "right": "issue:43976", "accept": false, - "reason": "Missing `AutoImageProcessor` export and a Python-version compatibility issue are not the same bug." + "reason": "Missing AutoImageProcessor import and Python-version incompatibility are different failures with different causes." }, { "left": "issue:44589", "right": "issue:45362", "accept": false, - "reason": "Float8 storage deserialization failure is unrelated to Qwen3.5 chat crashes." + "reason": "Float8 storage lookup error is unrelated to Qwen3.5 chat crashes." }, { - "left": "issue:44336", - "right": "issue:44373", + "left": "issue:43673", + "right": "issue:43761", "accept": false, - "reason": "Documentation/ANSI output issue versus a wrong docstring for `position_ids`; not the same underlying change." + "reason": "Chunked-prefill cache handling and CLIPVisionModel hidden_states regression are distinct bugs in different subsystems." }, { "left": "issue:16998", "right": "issue:22355", "accept": false, - "reason": "`model_max_length` on DeBERTa-V3 and missing `transformers.onnx` are unrelated." + "reason": "A model_max_length question for DeBERTa-V3 is unrelated to a missing transformers.onnx module import." + }, + { + "left": "issue:44336", + "right": "issue:44373", + "accept": false, + "reason": "Loading-report ANSI codes and a wrong docstring for position_ids are not the same bug." }, { "left": "issue:42491", "right": "issue:43299", "accept": false, - "reason": "Both involve Qwen3 MoE families, but one is LoRA compatibility across hf4.x/hf5.x and the other is a Qwen3VL MoE loading regression; too different to merge." + "reason": "Both concern Qwen MoE loading, but they target different model generations and different failure modes." }, { "left": "issue:44246", "right": "issue:44496", "accept": false, - "reason": "Import slowness and an unrecognized model/config error are unrelated." + "reason": "Import latency sometimes and unrecognized model config are unrelated." }, { "left": "issue:44295", "right": "issue:45397", "accept": false, - "reason": "Buffer registration/`position_ids` access and gemma-4 ZeRO-3 loading are different issues." + "reason": "Buffer access on position_ids is unrelated to Gemma-4 zero3 from_pretrained behavior." }, { "left": "issue:44242", "right": "issue:45593", "accept": false, - "reason": "Both concern losses omitted under flags, but for different models and different toggles (`output_router_logits` vs denoising)." + "reason": "Missing load-balancing loss and missing auxiliary losses are different model-specific training bugs." }, { "left": "issue:44336", @@ -11276,61 +11300,55 @@ "left": "issue:44246", "right": "issue:44871", "accept": false, - "reason": "Import-time performance and Gemma EOS token config mismatch are not the same bug." + "reason": "Import-time slowness is unrelated to Gemma eos_token_id configuration mismatch." }, { "left": "issue:43824", "right": "issue:43976", "accept": false, - "reason": "A missing Qwen2.5 VL export and a Python-version compatibility break are unrelated." + "reason": "Qwen2_5_VL symbol export/import failure is not the same as the Python 3.9+/3.10+ compatibility issue." }, { "left": "issue:41950", "right": "issue:44220", "accept": false, - "reason": "Video-classification pipeline image-processor lookup and `_torch_extract_fbank_features()` are separate subsystems." + "reason": "Video-classification pipeline image-processor lookup and _torch_extract_fbank_features are different pipeline/audio bugs." }, { "left": "issue:43976", "right": "issue:44964", "accept": false, - "reason": "Python 3.9+ compatibility and Phi-4 multimodal model loading are not the same failure." + "reason": "Interpreter-version incompatibility and Phi-4 multimodal loading failure are different problems." }, { - "left": "issue:36246", - "right": "issue:42757", + "left": "issue:42886", + "right": "issue:46132", "accept": false, - "reason": "Missing Qwen2.5 VL symbol and missing `is_offline_mode` from huggingface_hub are different import errors." + "reason": "Offline tokenizer cache loading and AttentionInterface.register behavior change are unrelated." }, { - "left": "issue:42886", - "right": "issue:46132", + "left": "issue:36246", + "right": "issue:42757", "accept": false, - "reason": "Offline tokenizer cache loading and `AttentionInterface.register` behavior changes are unrelated." + "reason": "Missing Qwen2_5_VL image processor export is unrelated to missing is_offline_mode from huggingface_hub." }, { "left": "issue:36246", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but they concern different missing exports and different code paths." + "reason": "These are separate import errors for different symbols and different packages." }, { "left": "issue:44779", "right": "issue:45292", "accept": false, - "reason": "Deepseek tokenizer regression and `resize_token_embeddings` not updating output embeddings are different model behavior bugs." + "reason": "Tokenizer correctness regression and resize_token_embeddings/output_embeddings mismatch are different bugs." }, { "left": "issue:44242", "right": "issue:44857", "accept": false, - "reason": "Different model-specific loss/AMP failures; no common concrete fix." - }, - { - "left": "issue:43673", - "right": "issue:44561", - "accept": false, - "reason": "Cache behavior in chunked prefill and removal of `is_torch_fx_available` affecting trust-remote-code models are unrelated." + "reason": "Training loss omission in D-FINE and LwDetrImageLoss AMP crash are distinct issues." }, { "left": "issue:45081", @@ -11338,17 +11356,53 @@ "accept": false, "reason": "Mistral tokenizer regex patch crash and Qwen3.5 chat crash are different model/tokenizer paths." }, + { + "left": "issue:43673", + "right": "issue:44561", + "accept": false, + "reason": "Both are v5 regressions, but one is GenerationMixin chunked_prefill cache handling and the other is trust_remote_code import breakage." + }, { "left": "issue:44295", "right": "issue:44749", "accept": false, - "reason": "`position_ids` buffer access and Chinese-reported performance regression after upgrade are unrelated." + "reason": "Position_ids buffer access and slowed filtering after upgrade are not the same bug." }, { "left": "issue:43352", "right": "issue:45250", "accept": false, - "reason": "Specific Nemotron FA2 support error versus a generic Flash Attention 2.0 topic; not the same concrete issue." + "reason": "Unsupported Flash Attention 2.0 on Nemotron and a generic Flash Attention 2.0 topic are not the same concrete issue." + }, + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "ESM2 breakage and Gemma 4 mm_token_type_ids defaults are unrelated." + }, + { + "left": "issue:43901", + "right": "issue:45003", + "accept": false, + "reason": "A doc/behavior mismatch for return_all_scores is unrelated to unsafe sys.modules access." + }, + { + "left": "issue:43526", + "right": "issue:45003", + "accept": false, + "reason": "BeitImageProcessorFast reduce_labels and modeling_utils sys.modules access are different code paths." + }, + { + "left": "issue:44496", + "right": "issue:45003", + "accept": false, + "reason": "Unrecognized model config and unsafe sys.modules access are separate failure modes." + }, + { + "left": "issue:45412", + "right": "issue:46123", + "accept": false, + "reason": "RT-DETR memory lifecycle and MaskGenerationPipeline final-batch dropping are unrelated." } ] }, @@ -11364,27 +11418,29 @@ "component_index": 1, "component_count": 18, "review_unit_index": 36, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:36246", + "issue:42886", "issue:43065", "issue:43232", "issue:43257", "issue:43262", "issue:43408", "issue:43475", - "issue:43526", "issue:43540", "issue:43596", "issue:43653", - "issue:43901", + "issue:43673", "issue:43931", + "issue:44060", + "issue:44077", "issue:44162", "issue:44246", "issue:44336", @@ -11392,7 +11448,7 @@ "issue:44466", "issue:44479", "issue:44483", - "issue:44496", + "issue:44734", "issue:44877", "issue:44977", "issue:44987", @@ -11400,11 +11456,11 @@ "issue:45003", "issue:45137", "issue:45200", + "issue:45229", "issue:45292", "issue:45335", "issue:45375", "issue:45397", - "issue:45412", "issue:45459", "issue:45561", "issue:45563", @@ -11416,22 +11472,16 @@ "issue:45759", "issue:45987", "issue:46082", - "issue:46123", "issue:46170", "issue:46202" ], "soft_pairs": [ - "issue:44162|issue:45200", - "issue:43901|issue:45003", - "issue:43526|issue:45003", - "issue:44496|issue:45003", - "issue:45412|issue:46123", + "issue:45375|issue:45759", + "issue:43257|issue:45137", "issue:43262|issue:44246", "issue:45292|issue:45584", - "issue:43257|issue:45137", - "issue:45375|issue:45759", - "issue:45292|issue:45701", "issue:36246|issue:43596", + "issue:45292|issue:45701", "issue:44877|issue:45375", "issue:45561|issue:45563", "issue:43232|issue:44336", @@ -11439,227 +11489,232 @@ "issue:43065|issue:43408", "issue:45987|issue:46082", "issue:44995|issue:45200", - "issue:44466|issue:45335", - "issue:43475|issue:43540", "issue:44479|issue:46170", + "issue:43475|issue:43540", + "issue:44466|issue:45335", "issue:45397|issue:45656", "issue:44483|issue:46202", - "issue:45137|issue:45600", "issue:43653|issue:44162", - "issue:44995|issue:45588", + "issue:45137|issue:45600", "issue:44368|issue:44977", + "issue:44995|issue:45588", "issue:44368|issue:45563", "issue:44877|issue:46170", - "issue:43931|issue:44368" + "issue:43931|issue:44368", + "issue:42886|issue:45229", + "issue:44060|issue:44077", + "issue:45003|issue:45561", + "issue:43673|issue:44734", + "issue:45584|issue:46170" ], - "prepared_review_unit_hash": "af200d3d10a89cf949aeb43f59bc30b82585f5266f323c95cf638beacdda8f1f", + "prepared_review_unit_hash": "209e1ead19b24158784ac127703d21448be6f15e07acee5410ccbd7c92df327e", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36323, - "estimated_input_tokens": 9081, - "estimated_eval_tokens": 18418 + "serialized_chars": 36796, + "estimated_input_tokens": 9199, + "estimated_eval_tokens": 18654 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly a set of unrelated bug reports that only share broad subsystem terms (loading, generation, config, deepspeed). I would not deduplicate any of the soft pairs.", - "confidence": 0.91, - "canonical_issue_reason": "issue:45003 is the strongest issue candidate to represent the cluster only because it is the broadest infrastructure-level bug here and appears in multiple soft-similarity links; however, it still does not match the other issues as a duplicate.", + "summary": "This cluster is heterogeneous: mostly unrelated issue reports spanning multimodal loading, generation, tokenization, deepspeed, config validation, and CI/test utilities. The soft pairs are mostly superficial topical overlaps rather than the same underlying bug.", + "confidence": 0.98, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:45003 is the best single issue to keep as the cluster anchor if one must be chosen, since it is a concrete modeling_utils bug with broader applicability than the more model-specific reports.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44162", - "right": "issue:45200", - "accept": false, - "reason": "ESM2 corruption/broken workflow is unrelated to Gemma 4 mm_token_type_ids defaults." - }, - { - "left": "issue:43901", - "right": "issue:45003", - "accept": false, - "reason": "Docs mentioning return_all_scores do not match a runtime sys.modules access bug." - }, - { - "left": "issue:43526", - "right": "issue:45003", - "accept": false, - "reason": "BeitImageProcessorFast label reduction bug is unrelated to modeling_utils module lookup." - }, - { - "left": "issue:44496", - "right": "issue:45003", + "left": "issue:45375", + "right": "issue:45759", "accept": false, - "reason": "Unrecognized-model config loading and unsafe sys.modules access are different failure modes." + "reason": "Different bugs in composite Qwen multimodal config handling: one drops a config field, the other fails to unwrap text_config in from_config." }, { - "left": "issue:45412", - "right": "issue:46123", + "left": "issue:43257", + "right": "issue:45137", "accept": false, - "reason": "RT-DETR memory cleanup and MaskGenerationPipeline final-batch handling are unrelated." + "reason": "Unrelated failures: Qwen3 MoE loading under accelerate+deepspeed vs DeepSpeed ZeRO3 deque underflow with BertModel." }, { "left": "issue:43262", "right": "issue:44246", "accept": false, - "reason": "Audio chat_template sampling-rate default bug is unrelated to slow import timing." + "reason": "Audio chat-template sampling-rate default bug vs slow transformers import; no shared code path." }, { "left": "issue:45292", "right": "issue:45584", "accept": false, - "reason": "resize_token_embeddings/output_embeddings mismatch is not the same as Whisper empty transcription after token alignment." - }, - { - "left": "issue:43257", - "right": "issue:45137", - "accept": false, - "reason": "Both involve DeepSpeed, but one is Qwen3 MoE weight conversion and the other is a zero3 deque/index error." + "reason": "resize_token_embeddings/output-embeddings bug is unrelated to Whisper empty transcription after align_special_tokens." }, { - "left": "issue:45375", - "right": "issue:45759", + "left": "issue:36246", + "right": "issue:43596", "accept": false, - "reason": "Both are Qwen multimodal config issues, but one drops a missing field and the other fails to unwrap text_config; not the same bug." + "reason": "Qwen2.5 VL import error and a ZeRO3 BertModel index error are different subsystems and failure modes." }, { "left": "issue:45292", "right": "issue:45701", "accept": false, - "reason": "Embedding resize behavior is unrelated to tokenization changes across versions." - }, - { - "left": "issue:36246", - "right": "issue:43596", - "accept": false, - "reason": "Qwen2_5_VL import error and DeepSpeed BertModel index error are unrelated." + "reason": "Embedding-resize synchronization bug vs version-dependent tokenization behavior; not the same issue." }, { "left": "issue:44877", "right": "issue:45375", "accept": false, - "reason": "Granite speech strict-config loading and Qwen config field dropping are different models and code paths." + "reason": "Both are config-related, but one is strict granite_speech loading and the other is a missing Qwen vision config field." }, { "left": "issue:45561", "right": "issue:45563", "accept": false, - "reason": "A pytest-xdist file race is unrelated to a stale generation warning." + "reason": "pytest-xdist file race in testing utils vs a stale generation warning; unrelated." }, { "left": "issue:43232", "right": "issue:44336", "accept": false, - "reason": "Generation kwargs update after sync_gpus has nothing to do with ANSI output in loading_report." + "reason": "Generation kwargs update after sync_gpus is unrelated to ANSI codes emitted by loading_report." }, { "left": "issue:44987", "right": "issue:45459", "accept": false, - "reason": "Model-loading failure and tokenizer error masking are separate issues in different code paths." + "reason": "Loading a model on transformers>=5.1.0 and tokenizer error masking from protobuf exception handling are distinct bugs." }, { "left": "issue:43065", "right": "issue:43408", "accept": false, - "reason": "SAM3 pixel decoder dummy conv and SAM3 tracker model-type warning are not the same underlying defect." + "reason": "Both mention SAM 3, but one is a dummy Conv2d implementation issue and the other is a model-type warning for tracker/video configs." }, { "left": "issue:45987", "right": "issue:46082", "accept": false, - "reason": "StaticCache seq-length return type and LlamaConfig head_dim validation are unrelated." + "reason": "StaticCache seq-length return type contract and LlamaConfig head_dim validation are unrelated." }, { "left": "issue:44995", "right": "issue:45200", "accept": false, - "reason": "GlmMoeDsa stale cache crash is unrelated to Gemma 4 token-type defaults." + "reason": "GlmMoeDsa stale indexer cache on second forward is unrelated to Gemma 4 mm_token_type_ids defaults." }, { - "left": "issue:44466", - "right": "issue:45335", + "left": "issue:44479", + "right": "issue:46170", "accept": false, - "reason": "Tied-weight serialization inconsistency is unrelated to t5gemma resize_token_embeddings propagation." + "reason": "Qwen video-input regression vs Windows ROCm torch.distributed import dependency issue; different areas entirely." }, { "left": "issue:43475", "right": "issue:43540", "accept": false, - "reason": "SAM3 vision encoder output attribute loss and Qwen3OmniMoe video-input validation are different bugs." + "reason": "SAM 3 video encoder attribute error and Qwen3OmniMoe video-processing ValueError are separate model bugs." }, { - "left": "issue:44479", - "right": "issue:46170", + "left": "issue:44466", + "right": "issue:45335", "accept": false, - "reason": "Qwen video-input regression and Windows ROCm torch.distributed import failure are unrelated." + "reason": "Tied-weight serialization inconsistency is not the same as t5gemma resize_token_embeddings failing to update decoder.embed_tokens." }, { "left": "issue:45397", "right": "issue:45656", "accept": false, - "reason": "Gemma-4 zero3 loading and duplicate optimizer.step calls under deepspeed are different problems." + "reason": "Gemma-4 ZeRO3 from_pretrained failure and double optimizer.step under deepspeed are unrelated training/runtime bugs." }, { "left": "issue:44483", "right": "issue:46202", "accept": false, - "reason": "chat/completions request rejection is unrelated to Gemma pipeline roundtrip token normalization." + "reason": "Chat/completions request rejection is unrelated to Gemma tokenization roundtrip failure on the underscore marker." }, { - "left": "issue:45137", - "right": "issue:45600", + "left": "issue:43653", + "right": "issue:44162", "accept": false, - "reason": "DeepSpeed deque failure and stale auto_mappings references to removed SAM3 Lite configs do not share a code path." + "reason": "BigBirdTokenizer special-token registration bug vs a broad ESM2 regression; no concrete shared bug." }, { - "left": "issue:43653", - "right": "issue:44162", + "left": "issue:45137", + "right": "issue:45600", "accept": false, - "reason": "BigBirdTokenizer special-token registration bug is unrelated to the ESM2 breakage report." + "reason": "ZeRO3 empty deque crash and removed Sam3LiteText references in auto_mappings are different issues." }, { - "left": "issue:44995", - "right": "issue:45588", + "left": "issue:44368", + "right": "issue:44977", "accept": false, - "reason": "GlmMoeDsa stale cache crash and flash_attention s_aux=None AttributeError are different failures." + "reason": "Tie-word-embeddings warning during fine-tuning is not the same as Qwen3.5 flash-attention generation failure." }, { - "left": "issue:44368", - "right": "issue:44977", + "left": "issue:44995", + "right": "issue:45588", "accept": false, - "reason": "A tied-word-embeddings warning is not the same as a flash-attention generation regression." + "reason": "GlmMoeDsa cache corruption and flash_attention s_aux=None crash are unrelated code paths." }, { "left": "issue:44368", "right": "issue:45563", "accept": false, - "reason": "Stale num_return_sequences warning and tie_word_embeddings warning are unrelated." + "reason": "A model config warning and a stale num_return_sequences warning are not the same underlying bug." }, { "left": "issue:44877", "right": "issue:46170", "accept": false, - "reason": "Strict granite_speech config loading and Windows ROCm dependency issues are unrelated." + "reason": "Strict config loading for granite_speech is unrelated to a Windows ROCm import/dependency failure." }, { "left": "issue:43931", "right": "issue:44368", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch is unrelated to a tie_word_embeddings warning." + "reason": "Qwen3-VL weight shape mismatch and Qwen3.5 tie_word_embeddings warning are different model-loading problems." + }, + { + "left": "issue:42886", + "right": "issue:45229", + "accept": false, + "reason": "Offline tokenizer cache loading and Gemma 4 multi-GPU OOM are unrelated." + }, + { + "left": "issue:44060", + "right": "issue:44077", + "accept": false, + "reason": "Qwen3-Next tied-weights warning bug and patchtsmixer post_init allowance are separate config/validation issues." + }, + { + "left": "issue:45003", + "right": "issue:45561", + "accept": false, + "reason": "Unsafe sys.modules access in modeling_utils is unrelated to the xdist captured_info.txt race." + }, + { + "left": "issue:43673", + "right": "issue:44734", + "accept": false, + "reason": "Both involve generation/cache, but one is missing GenerationMixin cache in chunked prefill and the other is wrong tensor indexing in serve /v1/responses." + }, + { + "left": "issue:45584", + "right": "issue:46170", + "accept": false, + "reason": "Whisper empty transcription after align_special_tokens has nothing to do with the Windows ROCm torch.distributed import problem." } ] }, @@ -11675,19 +11730,19 @@ "component_index": 1, "component_count": 18, "review_unit_index": 37, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:28282", "issue:30990", "issue:41628", "issue:42371", - "issue:42886", + "issue:42673", "issue:42994", "issue:43232", "issue:43475", @@ -11697,62 +11752,58 @@ "issue:43646", "issue:43673", "issue:43701", + "issue:43704", + "issue:43824", "issue:43827", "issue:44038", - "issue:44060", - "issue:44077", "issue:44162", "issue:44188", + "issue:44222", "issue:44297", + "issue:44351", "issue:44361", "issue:44623", "issue:44661", - "issue:44734", "issue:44749", "issue:44869", "issue:44871", "issue:44933", - "issue:45003", + "issue:44964", + "issue:45081", "issue:45092", - "issue:45229", + "issue:45103", + "issue:45137", "issue:45230", "issue:45310", "issue:45356", "issue:45375", "issue:45381", "issue:45397", + "issue:45406", "issue:45478", - "issue:45561", - "issue:45584", "issue:45656", "issue:46077", "issue:46123", "issue:46132", - "issue:46170", "issue:46202" ], "soft_pairs": [ - "issue:42886|issue:45229", - "issue:44060|issue:44077", - "issue:43673|issue:44734", - "issue:45003|issue:45561", - "issue:45584|issue:46170", "issue:44297|issue:44623", "issue:44869|issue:45356", "issue:44361|issue:44749", "issue:43646|issue:44162", "issue:44749|issue:45656", "issue:44623|issue:45656", - "issue:43673|issue:43827", "issue:44871|issue:46132", "issue:43701|issue:44749", + "issue:43673|issue:43827", "issue:44188|issue:46202", "issue:45230|issue:45397", "issue:42994|issue:43482", - "issue:43475|issue:45381", "issue:44038|issue:44623", - "issue:43475|issue:43643", + "issue:43475|issue:45381", "issue:28282|issue:30990", + "issue:43475|issue:43643", "issue:42371|issue:45092", "issue:43232|issue:43673", "issue:41628|issue:44933", @@ -11760,217 +11811,246 @@ "issue:43475|issue:44661", "issue:42994|issue:43525", "issue:46077|issue:46123", - "issue:45310|issue:45375" + "issue:45310|issue:45375", + "issue:45081|issue:45103", + "issue:45406|issue:46202", + "issue:42673|issue:43704", + "issue:41628|issue:44351", + "issue:44964|issue:45137", + "issue:44222|issue:45310", + "issue:42994|issue:43824", + "issue:44038|issue:45310" ], - "prepared_review_unit_hash": "878631b08c10809bbb978baa77d5b5e73d93b6cf23844298d16d549ca18ac947", + "prepared_review_unit_hash": "dcbb1239ee0eb245a16de6f93ab2187306ce086c1e75f5982070dda919ef9a7c", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 29, - "serialized_chars": 35546, - "estimated_input_tokens": 8887, - "estimated_eval_tokens": 18030 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37108, + "estimated_input_tokens": 9277, + "estimated_eval_tokens": 18810 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Loose cluster around Transformers v5 regressions and adjacent model/tokenizer bugs, but the soft pairs mostly point to different failure modes and subsystems rather than true duplicates.", - "confidence": 0.79, - "canonical_issue_reason": "issue:43646 is the broadest and most representative v5 breakage report in the set, centered on custom/remote model initialization, which overlaps thematically with several other regression reports.", + "summary": "The cluster is mostly heterogeneous, but the Qwen3.5 MoE loading reports form one clear duplicate set: the generic `from_pretrained` failure and the missing `deepstack_visual_indexes` config-field report appear to be the same regression. The rest of the pairs are different bugs or different code paths.", + "confidence": 0.58, + "canonical_issue_reason": "issue:45375 is the best canonical issue for the Qwen3.5 MoE loading regression because it names the likely root cause (`deepstack_visual_indexes` being dropped), while 45310/45478 are symptom-level load errors.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43646 is the best single issue to anchor the cluster because it is concrete, high-signal, and spans a common v5 compatibility path without being overly narrow.", + "best_issue_reason": "issue:45375 is the strongest representative overall among the accepted duplicates: it is specific, root-cause oriented, and explains the broader `from_pretrained` failures on Qwen3.5 MoE.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:42886", - "right": "issue:45229", - "accept": false, - "reason": "Offline tokenizer cache loading and multi-GPU CUDA OOM are unrelated failure modes; same library, different code paths." - }, - { - "left": "issue:44060", - "right": "issue:44077", - "accept": false, - "reason": "One is a tied-weights warning bug in Qwen3-Next, the other is a model API constraint for patchtsmixer; not the same underlying change." - }, - { - "left": "issue:43673", - "right": "issue:44734", - "accept": false, - "reason": "Both involve generation/KV cache behavior, but one is a missing cache object in v5.0.0 during chunked prefill and the other is a serve-time indexing crash in /v1/responses." - }, - { - "left": "issue:45003", - "right": "issue:45561", - "accept": false, - "reason": "Unsafe sys.modules access in modeling_utils is unrelated to a pytest-xdist captured_info.txt race in testing utilities." - }, - { - "left": "issue:45584", - "right": "issue:46170", - "accept": false, - "reason": "Empty transcription generation after token alignment is unrelated to Windows ROCm import/type errors around torch.distributed." - }, { "left": "issue:44297", "right": "issue:44623", "accept": false, - "reason": "Tokenizer save_pretrained metadata mismatch and processor.save_pretrained missing files are both serialization-related, but they affect different save paths and artifacts." + "reason": "Both are save_pretrained issues, but one is tokenizer-class metadata and the other is missing processor files; different failures." }, { "left": "issue:44869", "right": "issue:45356", "accept": false, - "reason": "Whisper timestamp decode crash and Kimi-K2.5 tokenizer codec regression are distinct tokenizer/decoder bugs with different triggers." + "reason": "Both concern tokenizer/codec behavior, but they target different models and distinct decode/regression paths." }, { "left": "issue:44361", "right": "issue:44749", "accept": false, - "reason": "MLukeTokenizer task AttributeError and Chinese performance regression after upgrade are not the same bug." + "reason": "One is a tokenizer AttributeError, the other is a performance regression after upgrade; unrelated bugs." }, { "left": "issue:43646", "right": "issue:44162", "accept": false, - "reason": "Custom model initialization failure and ESM2 breakage may both be v5 regressions, but they are separate model-specific compatibility issues." + "reason": "Custom config instantiation vs ESM2 breakage are different code paths and different symptoms." }, { "left": "issue:44749", "right": "issue:45656", "accept": false, - "reason": "Data filtering slowdown after upgrade and deepspeed optimizer-step duplication are unrelated performance/ \u0627\u0644\u062a\u062f\u0631\u064a\u0628 issues." + "reason": "Data filtering slowdown and double optimizer stepping are unrelated training issues." }, { "left": "issue:44623", "right": "issue:45656", "accept": false, - "reason": "Missing processor save artifacts versus double optimizer stepping are different subsystems and symptoms." - }, - { - "left": "issue:43673", - "right": "issue:43827", - "accept": false, - "reason": "Generation cache bug and docs still referencing pipeline() are not the same underlying issue." + "reason": "Processor save_pretrained file omission is unrelated to DeepSpeed optimizer stepping." }, { "left": "issue:44871", "right": "issue:46132", "accept": false, - "reason": "Gemma eos_token_id mismatch and AttentionInterface.register behavior change are separate configuration/API problems." + "reason": "Eos-token config inconsistency and AttentionInterface behavior changes are different problems." }, { "left": "issue:43701", "right": "issue:44749", "accept": false, - "reason": "resume_from_checkpoint key mismatch is unrelated to the reported data-filtering slowdown." + "reason": "Checkpoint key mismatch is unrelated to the data filtering slowdown report." + }, + { + "left": "issue:43673", + "right": "issue:43827", + "accept": false, + "reason": "Generation cache missing and docs still referencing removed pipeline APIs are not the same bug." }, { "left": "issue:44188", "right": "issue:46202", "accept": false, - "reason": "torch.compile divergence in attention kernels and Gemma pipeline roundtrip failure on a special character are unrelated." + "reason": "Attention-kernel divergence under torch.compile is unrelated to Gemma pipeline roundtrip failure." }, { "left": "issue:45230", "right": "issue:45397", "accept": false, - "reason": "Generic bug report and gemma-4 zero3 from_pretrained error do not indicate the same concrete problem." + "reason": "Placeholder bug report vs gemma-4 zero3 loading issue; no concrete overlap." }, { "left": "issue:42994", "right": "issue:43482", "accept": false, - "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are different directions of the save/load lifecycle." - }, - { - "left": "issue:43475", - "right": "issue:45381", - "accept": false, - "reason": "SAM 3 Video missing attribute and Qwen2.5-VL video position-id error are separate multimodal model bugs." + "reason": "Quantized model save failure and Qwen2.5-GGUF load failure are opposite directions and different code paths." }, { "left": "issue:44038", "right": "issue:44623", "accept": false, - "reason": "Qwen3-VL-Moe breakage and missing processor save files are unrelated." + "reason": "Qwen3-VL-Moe v5 compatibility and processor save_pretrained missing files are not the same bug." }, { "left": "issue:43475", - "right": "issue:43643", + "right": "issue:45381", "accept": false, - "reason": "SAM 3 Video encoder output attribute error and trust_remote_code missing fields are both initialization-related but on different model families and causes." + "reason": "SAM 3 video attribute error and qwen2.5-vl vision position ids mismatch are different model-specific bugs." }, { "left": "issue:28282", "right": "issue:30990", "accept": false, - "reason": "Missing PyTorch import and SentenceTransformers loading hang are unrelated user-environment/runtime issues." + "reason": "Missing PyTorch import error and sentence-transformers loading hang are unrelated failure modes." + }, + { + "left": "issue:43475", + "right": "issue:43643", + "accept": false, + "reason": "SAM 3 video attribute error is unrelated to trust_remote_code missing-fields behavior in AutoConfig." }, { "left": "issue:42371", "right": "issue:45092", "accept": false, - "reason": "TF32 settings docs/API issue and remote-code checkpoint meta-initialization incompatibility are not the same bug." + "reason": "TF32 settings guidance and InternVL2 v5 meta-init incompatibility are unrelated." }, { "left": "issue:43232", "right": "issue:43673", "accept": false, - "reason": "GenerationMixin key update after sync_gpus and missing cache during chunked prefill are both generation-related but different concrete failures." + "reason": "Both are generation-related, but one concerns sync_gpus update ordering and the other chunked_prefill cache absence." }, { "left": "issue:41628", "right": "issue:44933", "accept": false, - "reason": "AutoImageProcessor import failure and nonexistent import from image_utils are distinct API/export regressions." + "reason": "Both are import errors, but they reference different missing symbols and different modules." }, { "left": "issue:45375", "right": "issue:45478", - "accept": false, - "reason": "Missing config field silently dropped by strict handling and Qwen3.5 MoE from_pretrained error are related to the same model family but not the same concrete issue." + "accept": true, + "reason": "Same Qwen3.5 MoE load regression: one report names the missing config field, the other reports the resulting from_pretrained failure." }, { "left": "issue:43475", "right": "issue:44661", "accept": false, - "reason": "SAM 3 Video attribute error and add-new-model-like failure in TOKENIZER_MAPPING_NAMES are different code paths and model infra concerns." + "reason": "SAM 3 video attribute error and tokenizer-mapping add-new-model-like failure are unrelated." }, { "left": "issue:42994", "right": "issue:43525", "accept": false, - "reason": "Quantized model saving and Llama4Config missing pad_token_id are separate bugs." + "reason": "Quantized save failure and Llama4Config missing pad_token_id are different bugs." }, { "left": "issue:46077", "right": "issue:46123", "accept": false, - "reason": "Incorrect attention mask construction and final-partial-batch dropping in MaskGenerationPipeline are unrelated logic errors." + "reason": "Gemma4 attention-mask construction and mask-generation partial-batch handling are separate issues." }, { "left": "issue:45310", "right": "issue:45375", + "accept": true, + "reason": "The generic Qwen3.5 MoE from_pretrained error and the missing config-field report describe the same loading regression." + }, + { + "left": "issue:45081", + "right": "issue:45103", + "accept": false, + "reason": "Tokenizer regex patch crash and auto-docstring annotation crash are unrelated." + }, + { + "left": "issue:45406", + "right": "issue:46202", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer and Gemma pipeline roundtrip on \u2581 are different bugs." + }, + { + "left": "issue:42673", + "right": "issue:43704", + "accept": false, + "reason": "These have the same title, but there is no evidence they are the same underlying VRAM leak instance; only title similarity is too weak here." + }, + { + "left": "issue:41628", + "right": "issue:44351", + "accept": false, + "reason": "AutoImageProcessor import failure and HybridCache import failure are different missing-symbol regressions." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Phi-4 multimodal load failure and DeepSpeed ZeRO3 deque underflow are unrelated." + }, + { + "left": "issue:44222", + "right": "issue:45310", + "accept": false, + "reason": "FP8 save_pretrained for MoE and Qwen3.5 MoE from_pretrained loading error are different directions and different failures." + }, + { + "left": "issue:42994", + "right": "issue:43824", "accept": false, - "reason": "Qwen3.5 MoE from_pretrained error and missing deepstack_visual_indexes in Qwen3_5MoeVisionConfig are different model-specific regressions." + "reason": "Quantized save failure and Qwen2.5-VL import failure are unrelated." + }, + { + "left": "issue:44038", + "right": "issue:45310", + "accept": false, + "reason": "Both mention Qwen MoE and v5-era regressions, but they are different model families and different concrete failures." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The accepted Qwen3.5 MoE pairs are plausibly the same loading regression, with 45375 acting as the root-cause report and 45310/45478 as symptom reports. The remaining pairs are reasonably rejected as different bugs or code paths, and none of the verdicts overstate the evidence." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -11979,13 +12059,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 38, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:30333", "issue:30990", @@ -11995,304 +12075,287 @@ "issue:41628", "issue:42175", "issue:42222", - "issue:42673", "issue:42831", "issue:42898", - "issue:42994", "issue:43065", "issue:43232", "issue:43278", "issue:43377", - "issue:43704", "issue:43723", "issue:43784", "issue:43824", "issue:43874", "issue:43901", - "issue:44038", + "issue:44016", + "issue:44060", + "issue:44075", "issue:44188", "issue:44222", "issue:44242", - "issue:44351", + "issue:44295", "issue:44373", + "issue:44448", "issue:44496", "issue:44568", "issue:44623", + "issue:44734", "issue:44779", "issue:44792", - "issue:44964", - "issue:45081", "issue:45092", "issue:45103", + "issue:45125", "issue:45137", "issue:45200", "issue:45310", "issue:45356", - "issue:45406", "issue:45561", "issue:45600", + "issue:45656", "issue:45712", "issue:45812", "issue:46123", - "issue:46202" + "issue:46231" ], "soft_pairs": [ - "issue:45081|issue:45103", - "issue:45406|issue:46202", - "issue:42673|issue:43704", - "issue:41628|issue:44351", - "issue:44964|issue:45137", - "issue:44222|issue:45310", - "issue:42994|issue:43824", - "issue:44038|issue:45310", "issue:36010|issue:41628", "issue:42831|issue:44222", "issue:42175|issue:45712", "issue:44496|issue:45561", "issue:43377|issue:44242", - "issue:42898|issue:44568", "issue:45561|issue:45600", + "issue:42898|issue:44568", "issue:43901|issue:45561", "issue:43232|issue:44792", "issue:30990|issue:35141", "issue:43278|issue:45137", "issue:42222|issue:45200", - "issue:44623|issue:45310", "issue:43065|issue:46123", - "issue:44188|issue:44373", + "issue:44623|issue:45310", "issue:30333|issue:33290", + "issue:44188|issue:44373", "issue:43232|issue:44188", "issue:42898|issue:45812", "issue:43784|issue:43824", "issue:43723|issue:44568", "issue:43874|issue:45092", "issue:42898|issue:45356", - "issue:42898|issue:44779" + "issue:42898|issue:44779", + "issue:44734|issue:45103", + "issue:44016|issue:46231", + "issue:44075|issue:46231", + "issue:44060|issue:45125", + "issue:44295|issue:45656", + "issue:42898|issue:44448" ], - "prepared_review_unit_hash": "0e2b75db758eff492a45213710f122583209a1fe03b9fefcf2996c53566b2f26", + "prepared_review_unit_hash": "fd78c5b67b259d24208e8f573353dd54210b02efd378a657034cd19b0047e815", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 31, - "serialized_chars": 37043, - "estimated_input_tokens": 9261, - "estimated_eval_tokens": 18778 + "soft_pair_count": 29, + "serialized_chars": 36485, + "estimated_input_tokens": 9122, + "estimated_eval_tokens": 18500 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous. Only one soft pair looks like a true duplicate (the two identical Qwen3ForCausalLM VRAM-leak issues); the rest are at best loosely similar by model family or transformer-v5 regression themes and should stay separate.", - "confidence": 0.9, - "canonical_issue_reason": "42673 is the clearest anchor because 42673/43704 are the only exact duplicate pair in the set (same title and same underlying Qwen3ForCausalLM VRAM leak).", + "summary": "This cluster is highly heterogeneous: most items are unrelated bug reports across different subsystems (tokenizers, generation, training, multimodal, docs, deepspeed, import errors). None of the soft pairs look like the same underlying issue/change, so I would not merge any of them as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "42673 is the best representative issue here: it has the strongest duplicate evidence and the most concrete, self-contained bug report among the cluster.", + "best_issue_reason": "If a representative issue is needed, issue 42898 is the broadest and most concrete regression report (tokenization behavior change in v5), but the set is too mixed to justify a true canonical duplicate.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:45081", - "right": "issue:45103", + "left": "issue:36010", + "right": "issue:41628", "accept": false, - "reason": "Both are crashes, but they are different code paths: Mistral tokenizer regex patching vs auto-docstring parsing with future annotations." + "reason": "Both are import errors, but they concern different missing symbols (`GenerationMixin` vs `AutoImageProcessor`) and different API paths." }, { - "left": "issue:45406", - "right": "issue:46202", + "left": "issue:42831", + "right": "issue:44222", "accept": false, - "reason": "Both touch Gemma, but one is a processor attribute crash and the other is a tokenizer roundtrip failure; not the same bug." + "reason": "Both mention FP8, but one is about accuracy and the other about `save_pretrained` for MoE; different failures." }, { - "left": "issue:42673", - "right": "issue:43704", - "accept": true, - "reason": "Exact same title and same reported Qwen3ForCausalLM VRAM leak in multiple dataloader threads; this is a duplicate." + "left": "issue:42175", + "right": "issue:45712", + "accept": false, + "reason": "One is about TensorFlow missing from a torch-only install, the other about leftover dummy classes; unrelated problems." }, { - "left": "issue:41628", - "right": "issue:44351", + "left": "issue:44496", + "right": "issue:45561", "accept": false, - "reason": "Both are import errors, but for different missing symbols and unrelated code paths." + "reason": "Unrecognized model/config loading is unrelated to the xdist race on `captured_info.txt`." }, { - "left": "issue:44964", - "right": "issue:45137", + "left": "issue:43377", + "right": "issue:44242", "accept": false, - "reason": "Phi-4 multimodal loading failure and DeepSpeed ZeRO3 deque error are unrelated." + "reason": "Different bugs: padding mask support in MIMI encoder vs missing load-balancing loss when router logits are off." }, { - "left": "issue:44222", - "right": "issue:45310", + "left": "issue:45561", + "right": "issue:45600", "accept": false, - "reason": "Both mention MoE/FP8, but one is FP8 save_pretrained and the other is a Qwen3.5 Moe from_pretrained error; different failures." + "reason": "A test file race condition is unrelated to stale `auto_mappings.py` references breaking CI." }, { - "left": "issue:42994", - "right": "issue:43824", + "left": "issue:42898", + "right": "issue:44568", "accept": false, - "reason": "Quantized model saving and a Qwen2.5-VL import error are not the same underlying problem." - }, - { - "left": "issue:44038", - "right": "issue:45310", - "accept": false, - "reason": "Both concern Qwen3-family models, but the concrete failures differ enough that they would not be one merged fix." - }, - { - "left": "issue:36010", - "right": "issue:41628", - "accept": false, - "reason": "Both are import errors in transformers v5, but they involve different APIs and separate breakages." - }, - { - "left": "issue:42831", - "right": "issue:44222", - "accept": false, - "reason": "FP8 accuracy regression and FP8 save_pretrained bug are different issues." - }, - { - "left": "issue:42175", - "right": "issue:45712", - "accept": false, - "reason": "Package-dependency/backend inclusion problem vs dummy-class cleanup/check-repo failure are unrelated." - }, - { - "left": "issue:44496", - "right": "issue:45561", - "accept": false, - "reason": "Model config loading failure and pytest-xdist file race are unrelated." - }, - { - "left": "issue:43377", - "right": "issue:44242", - "accept": false, - "reason": "Missing padding-mask support in MIMI encoder and missing load-balancing loss in MoE are different code paths." - }, - { - "left": "issue:42898", - "right": "issue:44568", - "accept": false, - "reason": "Both are tokenizer regressions in v5, but one is clean_up_tokenization_spaces behavior and the other is add_special_tokens/BOS-EOS handling." - }, - { - "left": "issue:45561", - "right": "issue:45600", - "accept": false, - "reason": "Test-worker file race and broken auto_mappings references are unrelated." + "reason": "Both are tokenizer regressions in v5, but they affect different behaviors and model families (`clean_up_tokenization_spaces` vs BOS/EOS insertion)." }, { "left": "issue:43901", "right": "issue:45561", "accept": false, - "reason": "Docstring mismatch for TextClassificationPipeline and pytest-xdist file race are unrelated." + "reason": "A doc mismatch for `TextClassificationPipeline` does not match a pytest-xdist file race." }, { "left": "issue:43232", "right": "issue:44792", "accept": false, - "reason": "Generation kwargs/update bug and a Janus image-generation test failure do not share the same failure mode." + "reason": "Generation keyword update after `sync_gpus` is not the same as a failed janus image-generation test." }, { "left": "issue:30990", "right": "issue:35141", "accept": false, - "reason": "Sentence-transformers loading stall and embedding reinitialization after resize are unrelated." + "reason": "Loading hangs in sentence-transformers vs embedding reinitialization after resize are different symptoms and code paths." }, { "left": "issue:43278", "right": "issue:45137", "accept": false, - "reason": "BF16-to-FP32 dtype drift during evaluation and ZeRO3 deque underflow are unrelated." + "reason": "BF16\u2192FP32 eval dtype drift is unrelated to the DeepSpeed ZeRO3 deque underflow." }, { "left": "issue:42222", "right": "issue:45200", "accept": false, - "reason": "VitPose breakage and Gemma 4 mm_token_type_ids defaulting are different multimodal issues." - }, - { - "left": "issue:44623", - "right": "issue:45310", - "accept": false, - "reason": "Missing files in processor.save_pretrained and Qwen3.5 Moe from_pretrained error are different problems." + "reason": "Broken VitPose model files and missing `mm_token_type_ids` defaults are different multimodal issues." }, { "left": "issue:43065", "right": "issue:46123", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder and MaskGenerationPipeline partial-batch dropping are unrelated." + "reason": "A dummy `Conv2d` in Sam3PixelDecoder is unrelated to `MaskGenerationPipeline` dropping partial-batch results." }, { - "left": "issue:44188", - "right": "issue:44373", + "left": "issue:44623", + "right": "issue:45310", "accept": false, - "reason": "Torch.compile attention branching bug and a wrong docstring are not duplicates." + "reason": "Missing files from `processor.save_pretrained` is not the same as Qwen3.5 MoE `from_pretrained` failure in newer transformers." }, { "left": "issue:30333", "right": "issue:33290", "accept": false, - "reason": "MLFlow reporting on failed jobs and DeepSpeed OOM with Adafactor are unrelated." + "reason": "MLflow reporting on failed training jobs and DeepSpeed OOM with Adafactor are separate training/runtime issues." + }, + { + "left": "issue:44188", + "right": "issue:44373", + "accept": false, + "reason": "Attention-kernel divergence under `torch.compile` is unrelated to a wrong `position_ids` docstring." }, { "left": "issue:43232", "right": "issue:44188", "accept": false, - "reason": "Both involve generation/runtime behavior, but they are different concrete bugs and code paths." + "reason": "Generation kwargs handling and attention-kernel branching are different code paths and bugs." }, { "left": "issue:42898", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions across v4/v5, but they affect different models and different tokenization behavior." + "reason": "Both are tokenization regressions, but one is about cleanup-space behavior and the other about wrong token IDs for Granite; not the same bug." }, { "left": "issue:43784", "right": "issue:43824", "accept": false, - "reason": "Both are import errors, but for different symbols and different downstream packages." + "reason": "Both are import failures, but they involve different missing names from different modules and different packages." }, { "left": "issue:43723", "right": "issue:44568", "accept": false, - "reason": "Both concern tokenizer behavior in v5, but one is load-time tokenizer issues and the other is special-token insertion semantics." + "reason": "Tokenizer loading issues and missing BOS/EOS on `add_special_tokens=True` are distinct regression reports." }, { "left": "issue:43874", "right": "issue:45092", "accept": false, - "reason": "Both are multimodal model-loading issues, but the missing image-patch method and meta-initialization incompatibility are distinct bugs." + "reason": "A missing image-patch helper in GLM46V is unrelated to InternVL2 meta-initialization incompatibility." }, { "left": "issue:42898", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer regressions around v5, but the codec/fix_mistral_regex problem is separate from clean_up_tokenization_spaces." + "reason": "Both are tokenizer regressions, but they concern different models and different behavioral changes; too broad to be duplicates." }, { "left": "issue:42898", "right": "issue:44779", "accept": false, - "reason": "Granite token ID regression and DeepSeek tokenizer regression are separate model-specific bugs." + "reason": "Different tokenizer regressions for different model families; no evidence they stem from the same underlying change." + }, + { + "left": "issue:44734", + "right": "issue:45103", + "accept": false, + "reason": "KV-cache continuation crash in `transformers serve` is unrelated to auto-docstring parsing with future annotations." + }, + { + "left": "issue:44016", + "right": "issue:46231", + "accept": false, + "reason": "Notebook syntax error and a docs parameter-name mismatch are unrelated documentation problems." + }, + { + "left": "issue:44075", + "right": "issue:46231", + "accept": false, + "reason": "Unused SGD optimizer args and a docs typo in `set_training()` are not the same issue." + }, + { + "left": "issue:44060", + "right": "issue:45125", + "accept": false, + "reason": "Both involve Qwen3 variants, but one is a tied-weights warning bug and the other is a missing tensor-parallel plan." + }, + { + "left": "issue:44295", + "right": "issue:45656", + "accept": false, + "reason": "Registering `position_ids` as a buffer causing errors is unrelated to DeepSpeed double optimizer-step behavior." + }, + { + "left": "issue:42898", + "right": "issue:44448", + "accept": false, + "reason": "A tokenizer cleanup regression is not the same as differing outputs from `google/pegasus-cnn_dailymail` across v4/v5." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The clustering summary is grounded in the packet and stays conservative. The only accepted soft pair, issue:42673 vs issue:43704, is clearly supported by identical titles. The other soft-edge rejections do not overclaim duplicate status based on the provided titles and issue metadata, and the overall summary correctly treats the cluster as mostly heterogeneous." - }, - "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 }, @@ -12301,13 +12364,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 39, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:28282", "issue:30333", @@ -12318,16 +12381,14 @@ "issue:42175", "issue:42757", "issue:42831", - "issue:42898", "issue:43122", + "issue:43232", "issue:43475", "issue:43479", + "issue:43644", "issue:43723", "issue:43873", "issue:43976", - "issue:44016", - "issue:44060", - "issue:44075", "issue:44079", "issue:44190", "issue:44206", @@ -12335,9 +12396,11 @@ "issue:44263", "issue:44295", "issue:44297", + "issue:44373", "issue:44448", + "issue:44484", + "issue:44485", "issue:44623", - "issue:44734", "issue:44869", "issue:44933", "issue:44945", @@ -12345,33 +12408,26 @@ "issue:45071", "issue:45092", "issue:45103", - "issue:45125", "issue:45245", "issue:45310", "issue:45381", "issue:45405", "issue:45656", "issue:45712", + "issue:45834", "issue:46082", "issue:46132", "issue:46209", - "issue:46231", "issue:46305" ], "soft_pairs": [ - "issue:44734|issue:45103", - "issue:44016|issue:46231", - "issue:44075|issue:46231", - "issue:44060|issue:45125", - "issue:42898|issue:44448", - "issue:44295|issue:45656", "issue:45405|issue:45712", "issue:44448|issue:46132", "issue:42175|issue:44933", "issue:44222|issue:44297", "issue:44933|issue:45712", - "issue:43479|issue:44079", "issue:43976|issue:44933", + "issue:43479|issue:44079", "issue:44297|issue:45656", "issue:44222|issue:44623", "issue:30333|issue:34689", @@ -12380,9 +12436,9 @@ "issue:44190|issue:46082", "issue:30333|issue:36331", "issue:43479|issue:44869", - "issue:30333|issue:35141", - "issue:28282|issue:30333", "issue:45071|issue:46082", + "issue:28282|issue:30333", + "issue:30333|issue:35141", "issue:43475|issue:46305", "issue:44297|issue:45310", "issue:41628|issue:42757", @@ -12390,235 +12446,234 @@ "issue:43723|issue:44977", "issue:45381|issue:46305", "issue:44263|issue:45092", + "issue:43873|issue:44945", "issue:43122|issue:46209", - "issue:43873|issue:44945" + "issue:43232|issue:45834", + "issue:44222|issue:44484", + "issue:44373|issue:44485", + "issue:44295|issue:44623", + "issue:43479|issue:43644" ], - "prepared_review_unit_hash": "c8a32b155c873443bcb00115120c4c7334904e954ea325f7b85cfa4b8990eeea", + "prepared_review_unit_hash": "30636f536e68d464285c9913fb0f497a4efd42ec7023f5a9258dd31f849dbb1a", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 33, - "serialized_chars": 37489, - "estimated_input_tokens": 9373, - "estimated_eval_tokens": 19002 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36455, + "estimated_input_tokens": 9114, + "estimated_eval_tokens": 18484 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is a loose, heterogeneous set of issues with only superficial text similarity; none of the soft-pair candidates look like the same underlying bug or change, so there is no true duplicate cluster here.", - "confidence": 0.97, - "canonical_issue_reason": "No single canonical issue stands out: the items span unrelated bugs across serving, docs, tokenization, model loading, config, and save/pretrained code paths, and the soft links are not true duplicates.", + "summary": "The cluster is mostly a set of unrelated issue reports with superficial lexical overlap around serialization, imports, and version regressions. None of the soft pairs look like true duplicates or mergeable PR-equivalents.", + "confidence": 0.86, + "canonical_issue_reason": "Issue 44297 is the clearest self-contained bug report among the candidates: it describes a specific save_pretrained serialization mismatch with a concrete failure mode. It is also the best representative of the small save/pretrained-adjacent subset in this cluster.", "canonical_pr_reason": null, - "best_issue_reason": "No global best issue: none of these issues is broad enough to represent the others or plausibly subsume the rest as the same fix.", + "best_issue_reason": "Issue 44297 is the strongest standalone issue to keep as the representative bug: concrete, reproducible, and centered on a single code path rather than a broad regression description.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44734", - "right": "issue:45103", - "accept": false, - "reason": "Wrong code-path: KV-cache continuation crash in serving vs auto_docstring annotation handling." - }, - { - "left": "issue:44016", - "right": "issue:46231", - "accept": false, - "reason": "Notebook syntax error and a documentation parameter-name mismatch are unrelated." - }, - { - "left": "issue:44075", - "right": "issue:46231", - "accept": false, - "reason": "SGD optimizer args not being used is unrelated to a doc typo about set_training()." - }, - { - "left": "issue:44060", - "right": "issue:45125", - "accept": false, - "reason": "Both mention Qwen3.5-MoE, but one is a tied-weights warning and the other is missing tensor-parallel plan metadata." - }, - { - "left": "issue:42898", - "right": "issue:44448", - "accept": false, - "reason": "Both involve output differences, but the trigger and code path are different models/features." - }, - { - "left": "issue:44295", - "right": "issue:45656", - "accept": false, - "reason": "Buffer read error for position_ids is unrelated to Deepspeed calling optimizer.step twice." - }, { "left": "issue:45405", "right": "issue:45712", "accept": false, - "reason": "Dependency version bump and dummy-class cleanup are different maintenance issues." + "reason": "Different problems: a missing released dependency version vs leftover dummy PT classes and repo-check leakage." }, { "left": "issue:44448", "right": "issue:46132", "accept": false, - "reason": "Pegasus generation regression is unrelated to AttentionInterface.register changing callback behavior." + "reason": "One is a model output regression for Pegasus; the other is a behavior change in AttentionInterface.register. Same subsystem, different bug." }, { "left": "issue:42175", "right": "issue:44933", "accept": false, - "reason": "Pip extra/backends packaging problem is unrelated to a missing image_utils import." + "reason": "Backend packaging/install issue versus a nonexistent import from image_utils; no shared code-path bug." }, { "left": "issue:44222", "right": "issue:44297", "accept": false, - "reason": "Both touch save_pretrained, but one is FP8/MoE model state and the other is tokenizer config metadata." + "reason": "Both mention save_pretrained, but one is FP8 MoE saving and the other is tokenizer_config metadata mismatch. Different failure modes and artifacts." }, { "left": "issue:44933", "right": "issue:45712", "accept": false, - "reason": "Missing import and leftover dummy-class leakage are different symbol/export bugs." + "reason": "Both are import-related, but one is an invalid image_utils import and the other is dummy_pt_objects leakage. Not the same bug." }, { - "left": "issue:43479", - "right": "issue:44079", + "left": "issue:43976", + "right": "issue:44933", "accept": false, - "reason": "Config default initialization with None is unrelated to ModelOutput key assignment semantics." + "reason": "Python version support regression versus a missing import symbol; unrelated." }, { - "left": "issue:43976", - "right": "issue:44933", + "left": "issue:43479", + "right": "issue:44079", "accept": false, - "reason": "Python version compatibility failure is unrelated to a missing image_utils symbol." + "reason": "Config initialization bug versus ModelOutput key assignment bug; different objects and behaviors." }, { "left": "issue:44297", "right": "issue:45656", "accept": false, - "reason": "Tokenizer save metadata mismatch is unrelated to an optimizer step-counting bug." + "reason": "Tokenizer save metadata issue versus optimizer stepping twice under deepspeed; unrelated." }, { "left": "issue:44222", "right": "issue:44623", "accept": false, - "reason": "Both mention save_pretrained, but they affect different object types and failure modes." + "reason": "Both involve save_pretrained, but one is FP8 MoE model saving and the other is processor file emission; not the same code-path problem." }, { "left": "issue:30333", "right": "issue:34689", "accept": false, - "reason": "MLflow job-reporting behavior and model loading regression are unrelated." + "reason": "MLFlow job reporting on failure versus model loading breakage in a specific Transformers version." }, { "left": "issue:42831", "right": "issue:44623", "accept": false, - "reason": "FP8 accuracy issues are unrelated to processor save_pretrained missing files." + "reason": "FP8 accuracy issue versus missing processor save files; different concerns." }, { "left": "issue:45245", "right": "issue:46132", "accept": false, - "reason": "Category-count runtime error and attention registration behavior are unrelated." + "reason": "Category-count limit error versus attention registration behavior change; unrelated." }, { "left": "issue:44190", "right": "issue:46082", "accept": false, - "reason": "Example-script dataset loading and LlamaConfig validation are different bugs." + "reason": "Dataset loading problem versus LlamaConfig validation semantics; not the same bug." }, { "left": "issue:30333", "right": "issue:36331", "accept": false, - "reason": "MLflow failure reporting is unrelated to the compute_loss signature change." + "reason": "MLFlow reporting failure versus Trainer compute_loss signature breakage; different training-path issues." }, { "left": "issue:43479", "right": "issue:44869", "accept": false, - "reason": "Multimodal config defaults and Whisper timestamp decode crash are different code paths." + "reason": "Multimodal config defaulting bug versus Whisper word timestamp decoding crash; unrelated." }, { - "left": "issue:30333", - "right": "issue:35141", + "left": "issue:45071", + "right": "issue:46082", "accept": false, - "reason": "MLflow reporting and token-embedding reinitialization are unrelated." + "reason": "Type-checking regression versus config validation edge case; no shared underlying defect." }, { "left": "issue:28282", "right": "issue:30333", "accept": false, - "reason": "Missing PyTorch import and MLflow job-status reporting are unrelated." + "reason": "PyTorch import availability issue versus MLFlow run-state reporting; unrelated." }, { - "left": "issue:45071", - "right": "issue:46082", + "left": "issue:30333", + "right": "issue:35141", "accept": false, - "reason": "Type-checking regression in PretrainedConfig and LlamaConfig head_dim validation are not the same bug." + "reason": "Training job reporting versus token embedding resize initialization; different code paths." }, { "left": "issue:43475", "right": "issue:46305", "accept": false, - "reason": "SAM3 video encoder attribute error and MoonshineStreaming mask off-by-one are unrelated." + "reason": "SAM3 video encoder attribute error versus Moonshine streaming mask off-by-one; unrelated." }, { "left": "issue:44297", "right": "issue:45310", "accept": false, - "reason": "Tokenizer save metadata mismatch is unrelated to Qwen3.5-MoE from_pretrained regression." + "reason": "Tokenizer save_pretrained metadata mismatch versus Qwen3.5 MoE from_pretrained error; both involve loading/saving but not the same bug." }, { "left": "issue:41628", "right": "issue:42757", "accept": false, - "reason": "Two import errors, but from different symbols/modules and different underlying causes." + "reason": "Both are import errors, but for different missing symbols and different modules; no common underlying fix." }, { "left": "issue:44206", "right": "issue:45103", "accept": false, - "reason": "Feature-extractor center-arg crash is unrelated to auto_docstring annotation handling." + "reason": "Feature extractor parameter regression versus auto_docstring annotations crash; unrelated." }, { "left": "issue:43723", "right": "issue:44977", "accept": false, - "reason": "Tokenizer loading regression and flash-attention generation issue are different failures." + "reason": "Tokenizer loading regression versus flash-attention generation failure; different subsystems and symptoms." }, { "left": "issue:45381", "right": "issue:46305", "accept": false, - "reason": "VL vision_position_ids bug and streaming mask off-by-one are unrelated off-by-one-style reports." + "reason": "Both mention positional/window concepts, but one is VL vision_position_ids and the other is streaming frame masking. Not the same defect." }, { "left": "issue:44263", "right": "issue:45092", "accept": false, - "reason": "torch.split return semantics in GlmMoeDsaIndexer are unrelated to remote-code meta-init incompatibility." + "reason": "torch.split return handling in one model component versus remote-code meta-init incompatibility in another; unrelated." + }, + { + "left": "issue:43873", + "right": "issue:44945", + "accept": false, + "reason": "Quantization/offloading behavior versus incorrect LLM output under pipeline parallelism; same broad area, different bug." }, { "left": "issue:43122", "right": "issue:46209", "accept": false, - "reason": "Tokenization drift across versions is unrelated to FP8 inference failure on B200." + "reason": "Tokenization differences across versions versus FP8 inference failure on B200; unrelated." }, { - "left": "issue:43873", - "right": "issue:44945", + "left": "issue:43232", + "right": "issue:45834", + "accept": false, + "reason": "Generation kwargs update bug versus OCR long-input index error; no shared underlying code-path." + }, + { + "left": "issue:44222", + "right": "issue:44484", "accept": false, - "reason": "Quantization/offloading behavior and pipeline-parallel output correctness are different problems." + "reason": "FP8 MoE save_pretrained bug versus a question about shard size defaults; different issue types and causes." + }, + { + "left": "issue:44373", + "right": "issue:44485", + "accept": false, + "reason": "Docstring mismatch for position_ids versus a RoPE implementation discussion; not a duplicate bug report." + }, + { + "left": "issue:44295", + "right": "issue:44623", + "accept": false, + "reason": "Registered-buffer position_ids read error versus missing processor save files; different serialization problems." + }, + { + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Config default initialization versus non-persistent buffer serialization junk; superficially initialization-related but not the same bug." } ] }, @@ -12634,13 +12689,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 40, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:22355", "issue:28282", @@ -12648,274 +12703,275 @@ "issue:38175", "issue:39401", "issue:41628", + "issue:42371", + "issue:42757", "issue:42831", "issue:42886", "issue:42907", "issue:42994", "issue:43066", - "issue:43232", "issue:43479", + "issue:43576", "issue:43644", + "issue:43646", "issue:43723", "issue:43761", + "issue:43784", "issue:43824", - "issue:43950", "issue:44038", - "issue:44222", + "issue:44117", "issue:44261", "issue:44263", - "issue:44295", "issue:44360", "issue:44368", "issue:44373", + "issue:44464", "issue:44483", - "issue:44484", - "issue:44485", "issue:44514", "issue:44568", - "issue:44623", + "issue:44704", "issue:44829", "issue:44871", "issue:44913", "issue:44945", "issue:45092", "issue:45103", + "issue:45278", "issue:45310", "issue:45357", "issue:45406", "issue:45753", - "issue:45834", "issue:46002", "issue:46050", "issue:46082", "issue:46305" ], "soft_pairs": [ - "issue:43232|issue:45834", - "issue:44222|issue:44484", - "issue:44373|issue:44485", - "issue:44295|issue:44623", - "issue:43479|issue:43644", "issue:44483|issue:45406", "issue:43761|issue:46082", "issue:43723|issue:44871", "issue:45092|issue:45357", "issue:44263|issue:44360", + "issue:42907|issue:46050", "issue:30990|issue:38175", "issue:44261|issue:45092", - "issue:42907|issue:46050", - "issue:43644|issue:46305", "issue:44514|issue:45103", + "issue:43644|issue:46305", "issue:44829|issue:46002", "issue:43479|issue:44913", "issue:42831|issue:45310", - "issue:44263|issue:44373", "issue:22355|issue:28282", + "issue:44263|issue:44373", "issue:41628|issue:43824", "issue:42831|issue:44038", "issue:39401|issue:44568", + "issue:42886|issue:42994", "issue:44368|issue:46305", "issue:42886|issue:43066", - "issue:42886|issue:42994", - "issue:43950|issue:46305", - "issue:44945|issue:45753" + "issue:44945|issue:45753", + "issue:42757|issue:45278", + "issue:44464|issue:46082", + "issue:44117|issue:44704", + "issue:43576|issue:43784", + "issue:43646|issue:46002", + "issue:42371|issue:44263" ], - "prepared_review_unit_hash": "f50d56fb05de64c9bb5d2633f0fec7d2aaa2f0c77dfd6f74df0c244273cdf310", + "prepared_review_unit_hash": "044dabe110ae4d3583be79fb8cf869850c47935d2693f323924eeb7e4dfcd5da", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 35345, - "estimated_input_tokens": 8837, - "estimated_eval_tokens": 17930 + "serialized_chars": 35914, + "estimated_input_tokens": 8979, + "estimated_eval_tokens": 18214 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly a set of unrelated Transformers issues; none of the soft-edge pairs look like true duplicates of the same concrete bug/change.", - "confidence": 0.79, - "canonical_issue_reason": "issue:43950 is the strongest representative of a broad save/load regression theme (non-persistent buffer corruption) and is the closest thing to a shared underlying problem in this set.", + "summary": "This cluster is highly heterogeneous; most pairs share only broad subsystem overlap (imports, tokenizers, configs, saving/loading) and not the same concrete bug. No soft edge should be merged.", + "confidence": 0.92, + "canonical_issue_reason": "issue:45278 is the broadest umbrella-style report in the set, centered on widespread import regressions after a version upgrade, so it is the best single representative issue.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43950 is the best standalone issue here because it is technically specific, reproducible, and closest to the few save/load-related reports in the cluster.", + "best_issue_reason": "issue:45278 is the most general and representative issue here; it can stand in for the import-regression side of the cluster better than the many narrow, model-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43232", - "right": "issue:45834", + "left": "issue:44483", + "right": "issue:45406", "accept": false, - "reason": "Generation kwargs handling and a Kosmos2.5 OCR index error are unrelated code paths." - }, - { - "left": "issue:44222", - "right": "issue:44484", - "accept": false, - "reason": "FP8 MoE save_pretrained failure is unrelated to a question about the default 50GB shard size." - }, - { - "left": "issue:44373", - "right": "issue:44485", - "accept": false, - "reason": "Docstring wording for position_ids is not the same bug as GLM-5 RoPE implementation." - }, - { - "left": "issue:44295", - "right": "issue:44623", - "accept": false, - "reason": "Reading a buffer after registration and processor.save_pretrained missing files are different problems." - }, - { - "left": "issue:43479", - "right": "issue:43644", - "accept": false, - "reason": "Default config initialization for Phi4Multimodal is unrelated to non-persistent buffer junk on load." - }, - { - "left": "issue:44483", - "right": "issue:45406", - "accept": false, - "reason": "Chat/completions request rejection and Gemma4Processor missing _tokenizer are separate failures." + "reason": "Different failures: chat-completions request handling versus a Gemma4Processor attribute/import crash." }, { "left": "issue:43761", "right": "issue:46082", "accept": false, - "reason": "CLIPVisionModel hidden_states regression and LlamaConfig head_dim validation are unrelated." + "reason": "Unrelated code paths: CLIP vision forward regression versus a LlamaConfig validation error." }, { "left": "issue:43723", "right": "issue:44871", "accept": false, - "reason": "Tokenizer loading failure and Gemma-3 eos_token_id mismatch are different tokenizer/config issues." + "reason": "Both involve tokenizers, but one is a generic AutoTokenizer loading issue and the other is a Gemma eos_token_id config mismatch." }, { "left": "issue:45092", "right": "issue:45357", "accept": false, - "reason": "Meta-initialization incompatibility and incorrect saved visual encoder keys are different loading vs saving bugs." + "reason": "Both touch multimodal model persistence, but one is meta-init incompatibility and the other is wrong visual encoder keys on save." }, { "left": "issue:44263", "right": "issue:44360", "accept": false, - "reason": "torch.split return handling in one indexer is not the same as a missing ReLU in another." + "reason": "Same subsystem name, but the bugs differ: torch.split handling versus a missing ReLU in the DSA indexer." + }, + { + "left": "issue:42907", + "right": "issue:46050", + "accept": false, + "reason": "Both concern quantized model save/load, but they appear to be different quantization paths and different failure modes." }, { "left": "issue:30990", "right": "issue:38175", "accept": false, - "reason": "Sentence Transformers loading hang and SigLIP2 zero probabilities are unrelated." + "reason": "No same root cause: a loading hang in Sentence Transformers versus zero probabilities in a specific SigLIP2 model." }, { "left": "issue:44261", "right": "issue:45092", "accept": false, - "reason": "MLA rms_norm_eps precision issue is unrelated to InternVL2 meta-init compatibility." + "reason": "Different model families and bugs: MLA norm epsilon precision error versus remote-code checkpoint/meta-init incompatibility." }, { - "left": "issue:42907", - "right": "issue:46050", + "left": "issue:44514", + "right": "issue:45103", "accept": false, - "reason": "Dequantized model save failure and sinq save/load breakage are both quantization-related but not the same concrete bug." + "reason": "Processor batched chat-template crash and auto-docstring annotation handling are unrelated failures." }, { "left": "issue:43644", "right": "issue:46305", "accept": false, - "reason": "Non-persistent buffer corruption and a sliding-window off-by-one are different subsystems." - }, - { - "left": "issue:44514", - "right": "issue:45103", - "accept": false, - "reason": "apply_chat_template batching crash and auto_docstring annotation handling are unrelated." + "reason": "Different implementation areas and symptoms: non-persistent buffer corruption versus a streaming mask off-by-one." }, { "left": "issue:44829", "right": "issue:46002", "accept": false, - "reason": "FlashAttention-3 training degeneration and NaNs after checkout/init do not share a code path." + "reason": "Both affect training numerics, but one is flash-attention-3 specific and the other is NaNs after init/checkpointing." }, { "left": "issue:43479", "right": "issue:44913", "accept": false, - "reason": "Config defaults for multimodal models and GPTNeoX rotary_pct reload behavior are unrelated." + "reason": "Both are config serialization/init issues, but they concern different config classes and different fields." }, { "left": "issue:42831", "right": "issue:45310", "accept": false, - "reason": "FineGrainedFP8 accuracy and Qwen3.5 MoE from_pretrained are distinct issues." + "reason": "Different problems: FineGrainedFP8 accuracy drift versus a Qwen3.5 MoE from_pretrained error." }, { - "left": "issue:44263", - "right": "issue:44373", + "left": "issue:22355", + "right": "issue:28282", "accept": false, - "reason": "A torch.split return-value bug is unrelated to a docstring typo." + "reason": "Both are import-related, but one is a missing transformers.onnx module and the other is a missing PyTorch dependency." }, { - "left": "issue:22355", - "right": "issue:28282", + "left": "issue:44263", + "right": "issue:44373", "accept": false, - "reason": "Missing transformers.onnx module and missing PyTorch dependency are different import errors." + "reason": "A DSA indexer runtime issue and a docstring bug are not the same underlying change." }, { "left": "issue:41628", "right": "issue:43824", "accept": false, - "reason": "Different missing exports for different classes; same broad area, not the same bug." + "reason": "Both are missing import symbols, but they are different exported objects and likely separate packaging/export regressions." }, { "left": "issue:42831", "right": "issue:44038", "accept": false, - "reason": "FineGrainedFP8 accuracy issue is unrelated to the Qwen3-VL-Moe bug report." + "reason": "Too vague and model-specific to treat as the same bug; one is an accuracy issue, the other a Qwen3-VL-Moe bug report." }, { "left": "issue:39401", "right": "issue:44568", "accept": false, - "reason": "Qwen3 offset_mapping behavior and mDeBERTa special-token insertion are separate tokenizer bugs." + "reason": "Different tokenizer bugs affecting different models and behaviors: offset mapping versus special token insertion." + }, + { + "left": "issue:42886", + "right": "issue:42994", + "accept": false, + "reason": "Offline tokenizer cache loading and quantized model save failures are unrelated." }, { "left": "issue:44368", "right": "issue:46305", "accept": false, - "reason": "tie_word_embeddings warning and streaming mask off-by-one are unrelated." + "reason": "A config warning about tied embeddings is unrelated to the MoonshineStreaming mask off-by-one." }, { "left": "issue:42886", "right": "issue:43066", "accept": false, - "reason": "Offline cache loading failure is not the same as an incorrect tokenizer decoder type." + "reason": "Both involve tokenizers, but one is offline cache loading and the other is a decoder type mismatch." }, { - "left": "issue:42886", - "right": "issue:42994", + "left": "issue:44945", + "right": "issue:45753", "accept": false, - "reason": "Tokenizer offline cache loading and quantized model saving are unrelated." + "reason": "Pipeline-parallel output corruption and image-triggered infinite looping are different runtime failures." }, { - "left": "issue:43950", - "right": "issue:46305", + "left": "issue:42757", + "right": "issue:45278", "accept": false, - "reason": "from_pretrained buffer corruption and Moonshine streaming mask math are different bugs." + "reason": "45278 is a broad umbrella, but this specific huggingface_hub import error is not clearly the same underlying bug." }, { - "left": "issue:44945", - "right": "issue:45753", + "left": "issue:44464", + "right": "issue:46082", + "accept": false, + "reason": "Compiled-forward generation inconsistency and a LlamaConfig head_dim validation error are unrelated." + }, + { + "left": "issue:44117", + "right": "issue:44704", + "accept": false, + "reason": "Both are from_pretrained internals, but one is a tokenizer mapping assumption and the other is kwargs propagation to cached_file." + }, + { + "left": "issue:43576", + "right": "issue:43784", + "accept": false, + "reason": "A broken CLI command and an import-time NameError in sentence-transformers are different issues." + }, + { + "left": "issue:43646", + "right": "issue:46002", + "accept": false, + "reason": "Custom model initialization breakage is not the same as NaNs in classification heads after init." + }, + { + "left": "issue:42371", + "right": "issue:44263", "accept": false, - "reason": "Pipeline parallelism output corruption and a Qwen3_5 image-induced infinite loop are not the same issue." + "reason": "TF32 API guidance and a DSA indexer torch.split issue are unrelated." } ] }, @@ -12931,45 +12987,45 @@ "component_index": 1, "component_count": 18, "review_unit_index": 41, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:28218", "issue:30990", "issue:33357", + "issue:36296", + "issue:39401", "issue:41628", - "issue:42371", - "issue:42757", "issue:42907", "issue:43116", "issue:43122", "issue:43232", "issue:43408", + "issue:43452", "issue:43475", "issue:43550", - "issue:43576", - "issue:43646", "issue:43723", "issue:43756", "issue:43784", + "issue:43824", "issue:43937", "issue:44038", "issue:44060", - "issue:44117", + "issue:44222", "issue:44261", - "issue:44263", "issue:44295", "issue:44297", + "issue:44351", "issue:44360", "issue:44373", "issue:44442", - "issue:44464", "issue:44485", + "issue:44488", "issue:44514", "issue:44521", "issue:44625", @@ -12979,244 +13035,269 @@ "issue:44945", "issue:44964", "issue:45200", + "issue:45216", "issue:45245", "issue:45278", "issue:45290", + "issue:45478", "issue:45593", "issue:45600", - "issue:46002", - "issue:46082", "issue:46132" ], "soft_pairs": [ - "issue:42757|issue:45278", - "issue:44464|issue:46082", - "issue:44117|issue:44704", - "issue:43576|issue:43784", - "issue:43646|issue:46002", - "issue:42371|issue:44263", "issue:43550|issue:46132", "issue:30990|issue:33357", - "issue:43232|issue:45593", "issue:44295|issue:44297", + "issue:43232|issue:45593", "issue:44704|issue:45290", - "issue:44261|issue:44373", - "issue:43723|issue:44442", "issue:42907|issue:44945", + "issue:43723|issue:44442", + "issue:44261|issue:44373", "issue:43723|issue:44779", "issue:42907|issue:43937", "issue:41628|issue:45278", "issue:43756|issue:44060", - "issue:43784|issue:45600", "issue:43408|issue:43475", + "issue:43784|issue:45600", "issue:43122|issue:44514", "issue:42907|issue:44964", "issue:43116|issue:45200", - "issue:44360|issue:44485", "issue:44625|issue:44829", + "issue:44360|issue:44485", "issue:43122|issue:44521", "issue:44038|issue:44297", "issue:43232|issue:45245", "issue:28218|issue:30990", "issue:44038|issue:44295", - "issue:43122|issue:44779" + "issue:42907|issue:43452", + "issue:43122|issue:44779", + "issue:36296|issue:43116", + "issue:44038|issue:44222", + "issue:45216|issue:45478", + "issue:39401|issue:44488", + "issue:39401|issue:44521", + "issue:43824|issue:44351" ], - "prepared_review_unit_hash": "dd506ebcb330fbbd1f004fc26d5eed99361e656266eb9179761680e066c11990", + "prepared_review_unit_hash": "68e039eca2d9368d85eb0bdef569efe99e4b0ae0d05e58c48f73a42b341151da", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 31, - "serialized_chars": 36606, - "estimated_input_tokens": 9152, - "estimated_eval_tokens": 18560 + "soft_pair_count": 32, + "serialized_chars": 36901, + "estimated_input_tokens": 9226, + "estimated_eval_tokens": 18708 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: the soft pairs share broad subsystems (tokenizers, imports, generation, multimodal chat templating) but not the same concrete bug or fix. I would not merge any of the candidate pairs as duplicates.", - "confidence": 0.71, - "canonical_issue_reason": "If forced to pick one reference issue, issue:43122 is the broadest and most representative tokenizer-regression report in the set; however, the cluster as a whole does not have a true canonical duplicate issue.", + "summary": "This cluster is mostly a loose collection of distinct transformer regressions; the strongest common thread is tokenizer/processor loading or behavior changes in v5, but most pairs are not the same underlying bug.", + "confidence": 0.41, + "canonical_issue_reason": "issue:43723 is the broadest representative of the recurring v5 tokenizer-loading regression theme, and is more central than the highly specific reports around it.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43122 is the best single issue to anchor nearby tokenizer-regression discussions because it is relatively broad, clearly reproducible, and has inbound references; still, it does not represent the rest of this mixed cluster.", + "best_issue_reason": "issue:43723 best fits the cluster as a representative issue because it captures the generic AutoTokenizer v5 failure pattern seen across several neighboring reports.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:42757", - "right": "issue:45278", - "accept": false, - "reason": "Both are import-related, but 42757 is one missing symbol from huggingface_hub while 45278 is a broad upgrade breakage with many import errors; not the same underlying bug." - }, - { - "left": "issue:44464", - "right": "issue:46082", - "accept": false, - "reason": "Chunked generation with compiled forward is unrelated to LlamaConfig head_dim validation." - }, - { - "left": "issue:44117", - "right": "issue:44704", - "accept": false, - "reason": "One is TOKENIZER_MAPPING_NAMES returning None; the other is AutoProcessor not forwarding kwargs to cached_file. Different code paths and failures." - }, - { - "left": "issue:43576", - "right": "issue:43784", - "accept": false, - "reason": "A broken CLI env command and a NameError during sentence-transformers import are unrelated issues." - }, - { - "left": "issue:43646", - "right": "issue:46002", - "accept": false, - "reason": "Custom model initialization breakage and NaNs in classification heads are different problems with different symptoms and likely different fixes." - }, - { - "left": "issue:42371", - "right": "issue:44263", - "accept": false, - "reason": "TF32 settings guidance and a torch.split return-value issue are unrelated." - }, { "left": "issue:43550", "right": "issue:46132", "accept": false, - "reason": "torch.compile with SDPA on Bamba and AttentionInterface.register changing function behavior are distinct bugs." + "reason": "Different problems: one is a torch.compile/SDPA failure on a specific model, the other is an AttentionInterface.register behavior change." }, { "left": "issue:30990", "right": "issue:33357", "accept": false, - "reason": "Sentence Transformers loading hangs and a MacOS bus error on a CLIP model are not the same defect." + "reason": "Unrelated symptoms and models: loading hang for Sentence Transformers vs MacOS bus error for a community CLIP model." }, { - "left": "issue:43232", - "right": "issue:45593", + "left": "issue:44295", + "right": "issue:44297", "accept": false, - "reason": "Generation kwargs handling after sync_gpus is unrelated to D-FINE auxiliary losses when denoising is off." + "reason": "One is about reading position_ids after buffer registration; the other is tokenizer_class metadata written by save_pretrained." }, { - "left": "issue:44295", - "right": "issue:44297", + "left": "issue:43232", + "right": "issue:45593", "accept": false, - "reason": "Reading position_ids after buffer registration and tokenizer_class mismatch in save_pretrained are different tokenizer/model serialization issues." + "reason": "Different code paths and failures: generation kwargs after sync_gpus vs missing auxiliary losses in D-FINE." }, { "left": "issue:44704", "right": "issue:45290", "accept": false, - "reason": "Missing kwargs passthrough in AutoProcessor.from_pretrained and chat-template crashes on tool-call assistant messages are separate code paths." + "reason": "Both involve processing APIs, but the bugs differ: cached_file kwargs forwarding vs chat-template crashes with tool-call assistant messages." }, { - "left": "issue:44261", - "right": "issue:44373", + "left": "issue:42907", + "right": "issue:44945", "accept": false, - "reason": "A config epsilon precision issue and a wrong docstring are not duplicates." + "reason": "Different subsystems: dequantized model saving vs incorrect output under pipeline parallelism." }, { "left": "issue:43723", "right": "issue:44442", "accept": false, - "reason": "Both concern tokenizer loading, but one is a general v5 loading issue and the other is a specific FastSpeech2ConformerTokenizer failure; not enough evidence of the same bug." + "reason": "Both are AutoTokenizer load regressions, but the concrete failures and tokenizer targets are different; not the same bug." }, { - "left": "issue:42907", - "right": "issue:44945", + "left": "issue:44261", + "right": "issue:44373", "accept": false, - "reason": "Saving dequantized Ministrals/Devstrals and incorrect output under pipeline parallelism are unrelated." + "reason": "Different issues: config.rms_norm_eps precision mismatch vs a wrong docstring for position_ids." }, { "left": "issue:43723", "right": "issue:44779", "accept": false, - "reason": "Both mention tokenizer regressions in v5, but they affect different tokenizer paths and symptoms; too broad to call the same bug." + "reason": "Both mention tokenizer regressions in v5, but they affect different tokenizers and appear to be separate concrete failures." }, { "left": "issue:42907", "right": "issue:43937", "accept": false, - "reason": "Model-saving of dequantized weights is unrelated to GLM-5 GenerationConfig validation." + "reason": "Different bugs: save_pretrained/dequantization vs GenerationConfig validation for GLM-5." }, { "left": "issue:41628", "right": "issue:45278", "accept": false, - "reason": "AutoImageProcessor import failure may be one example of import breakage, but 45278 is a broad collection of import errors after upgrade, not the same specific issue." + "reason": "One missing import is not the same as a broader collection of import errors after an upgrade." }, { "left": "issue:43756", "right": "issue:44060", "accept": false, - "reason": "Smollm3 RoPE layer loss and Qwen3-Next tied-weights warnings are unrelated model-specific bugs." + "reason": "Different model-specific architecture issues: RoPE layer dropping vs tied-weights warning mapping." }, { - "left": "issue:43784", - "right": "issue:45600", + "left": "issue:43408", + "right": "issue:43475", "accept": false, - "reason": "Missing nn during import and stale auto_mappings references to removed Sam3LiteText configs are different import/config cleanup problems." + "reason": "Both are SAM 3 related, but one is a model-type warning and the other is a missing attribute in a different execution path." }, { - "left": "issue:43408", - "right": "issue:43475", + "left": "issue:43784", + "right": "issue:45600", "accept": false, - "reason": "A model-type mismatch warning and a missing attribute on Sam3VisionEncoderOutput are separate SAM 3 issues." + "reason": "Different failures: NameError importing sentence-transformers vs auto_mappings.py referencing removed Sam3LiteText configs." }, { "left": "issue:43122", "right": "issue:44514", "accept": false, - "reason": "Tokenization drift across versions and a batched apply_chat_template crash with padding=False are different failures." + "reason": "Tokenizer-version regression vs multimodal chat-template crash; related area but not the same underlying bug." + }, + { + "left": "issue:42907", + "right": "issue:44964", + "accept": false, + "reason": "Distinct issues involving model saving/loading and a specific multimodal model load failure." + }, + { + "left": "issue:43116", + "right": "issue:45200", + "accept": false, + "reason": "Different behavior: example script returning empty classification results vs Gemma 4 fine-tuning requiring mm_token_type_ids." }, { "left": "issue:44625", "right": "issue:44829", "accept": false, - "reason": "Qwen3.5 num_labels propagation and flash_attention_3 training degeneration are unrelated." + "reason": "Different model/training bugs: Qwen3.5 num_labels propagation vs flash_attention_3 degenerating sequence classification training." }, { - "left": "issue:44704", + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "Unrelated technical areas: DSA indexer ReLU discussion vs GLM-5 RoPE implementation." + }, + { + "left": "issue:43122", "right": "issue:44521", "accept": false, - "reason": "kwargs forwarding to cached_file and zero assistant_masks for multimodal inputs are distinct chat/template behaviors." + "reason": "Both touch chat/tokenization APIs, but one is a general tokenizer regression and the other is multimodal assistant_masks being all zero." }, { "left": "issue:44038", "right": "issue:44297", "accept": false, - "reason": "A Qwen3-VL-Moe bug and tokenizer save_pretrained metadata mismatch are not the same underlying problem." + "reason": "Different issues: Qwen3-VL-Moe behavior in v5 vs tokenizer_class metadata mismatch on save_pretrained." }, { "left": "issue:43232", "right": "issue:45245", "accept": false, - "reason": "Generation config updates after sync_gpus and a categories-cardinality runtime error are unrelated." + "reason": "No shared code path: generation kwargs handling vs a category-count runtime limit." }, { "left": "issue:28218", "right": "issue:30990", "accept": false, - "reason": "Added-token spacing behavior and Sentence Transformers loading hangs are unrelated." + "reason": "Tokenizer spacing bug vs a loading hang; not the same underlying issue." }, { "left": "issue:44038", - "right": "issue:44295", + "right": "issue:44222", "accept": false, - "reason": "Qwen3-VL-Moe behavior and position_ids buffer access errors are different bugs." + "reason": "Both are MoE-related only at a high level, but one is a model behavior bug and the other is FP8 save_pretrained support." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Different load/save failures: dequantized model saving vs gguf_file breaking AutoTokenizer/AutoModelForCausalLM loading." }, { "left": "issue:43122", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer regressions, but they concern different models and different observable failures; too weak to merge as the same bug." + "reason": "Similar theme of tokenization regressions in v5, but the reports describe different tokenizers and different failure modes." + }, + { + "left": "issue:36296", + "right": "issue:43116", + "accept": false, + "reason": "Tensor parallel training bug is unrelated to the multi-label classification example returning empty results." + }, + { + "left": "issue:45216", + "right": "issue:45478", + "accept": false, + "reason": "Same model family, but one is save_pretrained checkpoint corruption and the other is from_pretrained loading failure." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "Different problems: Qwen3 offset_mapping regression vs a specific model load failure for cjvt/sleng-bert." + }, + { + "left": "issue:39401", + "right": "issue:44521", + "accept": false, + "reason": "Tokenizer offset_mapping bug and multimodal assistant_masks bug are separate issues." + }, + { + "left": "issue:43824", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for unrelated symbols and packages." + }, + { + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "Different subjects entirely; no shared bug or code path is evident." } ] }, @@ -13232,35 +13313,33 @@ "component_index": 1, "component_count": 18, "review_unit_index": 42, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:22355", "issue:28218", "issue:29127", "issue:29159", "issue:33453", - "issue:36296", + "issue:38617", "issue:39401", + "issue:42175", "issue:42222", "issue:42371", "issue:42757", "issue:42831", - "issue:42907", "issue:43066", "issue:43116", "issue:43122", "issue:43441", - "issue:43452", "issue:43493", "issue:43611", "issue:43784", - "issue:43824", "issue:43825", "issue:43874", "issue:43906", @@ -13269,34 +13348,28 @@ "issue:43992", "issue:44038", "issue:44075", - "issue:44222", "issue:44297", "issue:44351", "issue:44355", + "issue:44380", "issue:44485", - "issue:44488", - "issue:44521", + "issue:44625", "issue:44704", "issue:44779", "issue:44936", + "issue:44938", "issue:45083", - "issue:45216", + "issue:45278", "issue:45357", "issue:45446", "issue:45478", "issue:45563", "issue:45753", "issue:46002", + "issue:46077", "issue:46231" ], "soft_pairs": [ - "issue:42907|issue:43452", - "issue:36296|issue:43116", - "issue:44038|issue:44222", - "issue:39401|issue:44488", - "issue:45216|issue:45478", - "issue:39401|issue:44521", - "issue:43824|issue:44351", "issue:22355|issue:28218", "issue:43784|issue:45446", "issue:22355|issue:29159", @@ -13308,8 +13381,8 @@ "issue:43937|issue:44485", "issue:43611|issue:46002", "issue:42371|issue:43874", - "issue:42222|issue:43441", "issue:44075|issue:44936", + "issue:42222|issue:43441", "issue:42831|issue:44297", "issue:43976|issue:45446", "issue:45357|issue:45478", @@ -13320,120 +13393,82 @@ "issue:43116|issue:43493", "issue:43937|issue:45563", "issue:22355|issue:29127", - "issue:33453|issue:39401" + "issue:33453|issue:39401", + "issue:44625|issue:46002", + "issue:44380|issue:46077", + "issue:38617|issue:45278", + "issue:42175|issue:44938" ], - "prepared_review_unit_hash": "733e2a3a8ac1d1a6d9d3ba728380828e2ca30bfae54a3f6dd4837789ebca2f06", + "prepared_review_unit_hash": "becfb9322c52a52a2885f336cedf9f4174e1be0a65b109bec2139aeb901b8fbb", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 31, - "serialized_chars": 36429, - "estimated_input_tokens": 9108, - "estimated_eval_tokens": 18472 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 34992, + "estimated_input_tokens": 8748, + "estimated_eval_tokens": 17752 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly heterogeneous: tokenizer regressions, model loading/saving bugs, import errors, and unrelated training/runtime issues. The only clear duplicate-like matches are the Qwen3.5 save/load serialization regressions.", - "confidence": 0.88, - "canonical_issue_reason": "issue:45216 is the best anchor for the Qwen3.5 checkpoint/save_pretrained regression family: it is broad, regression-tagged, and directly explains the related from_pretrained failures.", + "summary": "This is a loose thematically similar set, not a true duplicate cluster: it mixes tokenizer regressions, import/API breakages, model-specific bugs, and a docs mismatch. None of the soft-edge pairs look like the same underlying bug/change, so I rejected them all. Among the issues, the broadest representative is the tokenizer regression report in 43122.", + "confidence": 0.77, + "canonical_issue_reason": "issue:43122 is the broadest concrete regression report in the dominant tokenizer theme: it describes behavior changing across versions for the same tokenizer, which makes it a better anchor than narrower loading/save/offset-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45216 is the strongest global issue candidate because it captures the underlying serialization bug more generally than the later, narrower symptom reports.", + "best_issue_reason": "issue:43122 is the best representative issue for triage because it is version-anchored, user-facing, and broad enough to sit near several other tokenizer-regression reports in this set.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:42907", - "right": "issue:43452", - "accept": false, - "reason": "Different code paths: dequantized model saving vs gguf loading failure." - }, - { - "left": "issue:36296", - "right": "issue:43116", - "accept": false, - "reason": "Tensor-parallel training bug and multi-label example-script behavior are unrelated." - }, - { - "left": "issue:44038", - "right": "issue:44222", - "accept": false, - "reason": "Both mention MoE, but one is a Qwen3-VL-Moe runtime issue and the other is FP8 save_pretrained; not the same bug." - }, - { - "left": "issue:39401", - "right": "issue:44488", - "accept": false, - "reason": "Wrong offset_mapping in Qwen3 tokenizer vs a separate model-loading failure for cjvt/sleng-bert." - }, - { - "left": "issue:45216", - "right": "issue:45478", - "accept": true, - "reason": "Same Qwen3.5 save/load serialization regression: incorrect saved checkpoint content leads to from_pretrained failure." - }, - { - "left": "issue:39401", - "right": "issue:44521", - "accept": false, - "reason": "Offset mapping bug and all-zero assistant masks are different tokenizer/template problems." - }, - { - "left": "issue:43824", - "right": "issue:44351", - "accept": false, - "reason": "Different missing symbols: Qwen2_5_VLForConditionalGeneration vs HybridCache." - }, { "left": "issue:22355", "right": "issue:28218", "accept": false, - "reason": "ONNX import error and extra-space-after-added-token tokenizer bug are unrelated." + "reason": "Different failure modes: missing module import vs tokenizer spacing behavior; no shared code path or concrete fix." }, { "left": "issue:43784", "right": "issue:45446", "accept": false, - "reason": "Sentence-transformers import NameError is not the same as the flex_attention PyTorch version check bug." + "reason": "Unrelated import problems in different components (sentence-transformers import vs flex_attention version check)." }, { "left": "issue:22355", "right": "issue:29159", "accept": false, - "reason": "ONNX module import failure and slow/fast tokenizer inconsistency are unrelated." + "reason": "One is an ONNX import error, the other is slow/fast tokenizer inconsistency; too broad to be the same bug." }, { "left": "issue:42757", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but they concern different missing APIs and different failure causes." + "reason": "Both are import errors, but for different symbols and different packages/APIs; not the same underlying issue." }, { "left": "issue:43122", "right": "issue:43906", "accept": false, - "reason": "An isolated reproduction can be related, but there is not enough evidence that it is the same concrete tokenizer bug." + "reason": "The second item is only a reproduction reference with no clear matching symptom; insufficient evidence of the same bug." }, { "left": "issue:43992", "right": "issue:44704", "accept": false, - "reason": "Missing embed_tokens.weight in UMT5Encoder and AutoProcessor kwargs forwarding are different loading issues." + "reason": "Different API paths and failures: missing model weight loading vs AutoProcessor kwargs forwarding." }, { "left": "issue:43066", "right": "issue:44779", "accept": false, - "reason": "Wrong tokenizer decoder type and Deepseek tokenization regression are distinct symptoms/code paths." + "reason": "Both are tokenizer regressions, but one is decoder-type metadata and the other is wrong tokenization results for Deepseek; not the same concrete bug." }, { "left": "issue:44355", @@ -13445,107 +13480,128 @@ "left": "issue:43937", "right": "issue:44485", "accept": false, - "reason": "GLM-5 generation config validation and GLM-5 RoPE discussion are not the same bug." + "reason": "Same model family, but one is GenerationConfig validation and the other is RoPE implementation discussion; different code paths." }, { "left": "issue:43611", "right": "issue:46002", "accept": false, - "reason": "base_model_prefix loading breakage and NaNs in classification heads are unrelated." + "reason": "Loading with base_model_prefix is unrelated to NaNs in classification heads." }, { "left": "issue:42371", "right": "issue:43874", "accept": false, - "reason": "TF32 API guidance issue and missing get_number_of_image_patches AttributeError are unrelated." + "reason": "TF32 settings guidance and a missing image-processor method are unrelated issues." }, { - "left": "issue:42222", - "right": "issue:43441", + "left": "issue:44075", + "right": "issue:44936", "accept": false, - "reason": "Vitpose import breakage and Ministral-3 FlashAttention failure are unrelated." + "reason": "Optimizer argument handling and trainer evaluate-after-train behavior are different subsystems and symptoms." }, { - "left": "issue:44075", - "right": "issue:44936", + "left": "issue:42222", + "right": "issue:43441", "accept": false, - "reason": "SGD args not used and trainer.evaluate() failing after training are different trainer/optimizer bugs." + "reason": "Different model breakages (ViTpose model loading vs Ministral FlashAttention) with no shared underlying defect." }, { "left": "issue:42831", "right": "issue:44297", "accept": false, - "reason": "FineGrainedFP8 accuracy issue and tokenizer_class mismatch on save_pretrained are unrelated." + "reason": "FP8 accuracy and tokenizer save_pretrained metadata mismatch are unrelated." }, { "left": "issue:43976", "right": "issue:45446", "accept": false, - "reason": "Python version compatibility regression and flex_attention version-check bug are unrelated." + "reason": "Python-version compatibility regression vs a specific import/version gate bug in flex_attention; not the same issue." }, { "left": "issue:45357", "right": "issue:45478", - "accept": true, - "reason": "Same Qwen3.5 save/load serialization family: incorrect visual-encoder keys in save_pretrained plausibly cause the from_pretrained error." + "accept": false, + "reason": "Both concern Qwen3.5, but one is save_pretrained key persistence and the other is from_pretrained loading; different bugs." }, { "left": "issue:44038", "right": "issue:45478", "accept": false, - "reason": "Both involve Qwen3.5/Qwen3.5-VL-Moe, but one is a transformer/model bug and the other is a save/load serialization failure." + "reason": "Different Qwen model variants and different failure modes; too little to merge as one bug." }, { "left": "issue:43116", "right": "issue:43825", "accept": false, - "reason": "Multi-label example script output and pipeline translation-task messaging are unrelated." + "reason": "Example-script classification output bug and a pipeline error-message issue are unrelated." }, { "left": "issue:43493", "right": "issue:45753", "accept": false, - "reason": "SigLIP2 implementation discrepancy and Qwen3_5 infinite-loop-on-image are unrelated multimodal bugs." + "reason": "Different model families and symptoms: SigLIP2 implementation discrepancy vs Qwen3_5 infinite loop on a specific image." }, { "left": "issue:45083", "right": "issue:45446", "accept": false, - "reason": "Qwen3_omni_moe feature-length helper behavior and flex_attention PyTorch version check are unrelated." + "reason": "Qwen3_omni_moe feature-length helper behavior and a flex_attention import/version check are unrelated." }, { "left": "issue:43116", "right": "issue:43493", "accept": false, - "reason": "Multi-label classification example behavior and SigLIP2 implementation discrepancy are unrelated." + "reason": "Classification example behavior and SigLIP2 implementation parity are different problems." }, { "left": "issue:43937", "right": "issue:45563", "accept": false, - "reason": "GLM-5 generation-config validation and paged generate stale warning are unrelated generation issues." + "reason": "GenerationConfig validation and a stale warning in paged generate() are separate generation-path issues." }, { "left": "issue:22355", "right": "issue:29127", "accept": false, - "reason": "ONNX import failure and layoutlmv3 error-message clarity are unrelated." + "reason": "Importing transformers.onnx and a LayoutLMv3 error-message clarity issue are unrelated." }, { "left": "issue:33453", "right": "issue:39401", "accept": false, - "reason": "Tokenizer loading regression and Qwen3 tokenizer offset_mapping bug are different tokenizer failures." + "reason": "Both are tokenizer regressions, but one is loading regression and the other is wrong offset mapping for Qwen3; not clearly the same bug." + }, + { + "left": "issue:44625", + "right": "issue:46002", + "accept": false, + "reason": "Config propagation for Qwen3.5 and NaNs in classification heads are different defects." + }, + { + "left": "issue:44380", + "right": "issue:46077", + "accept": false, + "reason": "GPT-2 attention scaling under SDPA/FlashAttention and Gemma4 bidirectional attention masking are separate attention-path bugs." + }, + { + "left": "issue:38617", + "right": "issue:45278", + "accept": false, + "reason": "Both are import regressions, but they involve different missing symbols and likely different dependency/version breakages." + }, + { + "left": "issue:42175", + "right": "issue:44938", + "accept": false, + "reason": "Backend extras installation and Python 3.14 loading failure are unrelated environment/package issues." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The analyst output is grounded in the packet and the negative pairings stay appropriately conservative. The only accepted matches are the Qwen3.5 save/load regression family, which is well-supported by the issue titles, and the overall heterogeneity summary fits the set." - }, - "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 }, @@ -13554,19 +13610,19 @@ "component_index": 1, "component_count": 18, "review_unit_index": 43, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:38617", "issue:39401", "issue:41628", - "issue:42175", "issue:42222", + "issue:42371", "issue:42673", "issue:43064", "issue:43066", @@ -13580,20 +13636,20 @@ "issue:43827", "issue:43937", "issue:43986", + "issue:44075", "issue:44117", + "issue:44261", "issue:44291", - "issue:44380", "issue:44410", "issue:44466", "issue:44483", "issue:44484", "issue:44496", - "issue:44625", "issue:44779", "issue:44829", "issue:44877", - "issue:44938", "issue:44945", + "issue:44964", "issue:44977", "issue:45071", "issue:45072", @@ -13605,19 +13661,15 @@ "issue:45446", "issue:45464", "issue:45588", + "issue:45656", "issue:46002", - "issue:46077", "issue:46178", "issue:46231", "issue:46298" ], "soft_pairs": [ - "issue:44625|issue:46002", - "issue:38617|issue:45278", - "issue:44380|issue:46077", - "issue:39401|issue:44779", - "issue:42175|issue:44938", "issue:44117|issue:45588", + "issue:39401|issue:44779", "issue:43064|issue:43673", "issue:43381|issue:44945", "issue:43827|issue:45446", @@ -13625,218 +13677,227 @@ "issue:43937|issue:44977", "issue:44466|issue:46002", "issue:45405|issue:45446", + "issue:42673|issue:45072", "issue:44117|issue:45081", "issue:43122|issue:46298", - "issue:42673|issue:45072", "issue:44410|issue:45310", "issue:44291|issue:45071", "issue:44829|issue:45216", - "issue:39401|issue:43421", - "issue:38617|issue:41628", "issue:42222|issue:46178", + "issue:38617|issue:41628", + "issue:39401|issue:43421", "issue:43408|issue:43742", "issue:44877|issue:46231", "issue:44484|issue:45310", "issue:43937|issue:44496", "issue:43066|issue:43122", - "issue:44483|issue:45464" + "issue:44483|issue:45464", + "issue:44075|issue:45656", + "issue:44964|issue:45278", + "issue:42371|issue:44261" ], - "prepared_review_unit_hash": "ec5e0fe8dac8e08679233342a9d2422a5ff3aab6f935a0c51b72d8b4f615da61", + "prepared_review_unit_hash": "a33378f76a051575f4908ea140c53a7a5128e13cffb71344fec3b158a39e5fe1", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 28, - "serialized_chars": 36088, - "estimated_input_tokens": 9022, - "estimated_eval_tokens": 18300 + "soft_pair_count": 27, + "serialized_chars": 35784, + "estimated_input_tokens": 8946, + "estimated_eval_tokens": 18148 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous, but the strongest shared theme is v5 upgrade regressions around imports/tokenization. No soft pair looks like a true duplicate or the same concrete fix, so all are rejected.", - "confidence": 0.31, - "canonical_issue_reason": "issue:45278 is the broadest representative of the v5 upgrade breakage theme, covering multiple import failures after moving from 4.57.x to 5.5.0.", + "summary": "This cluster is a grab-bag of unrelated Transformers issues. The soft pairs share broad keywords like tokenization, model loading, or optimization, but they do not describe the same concrete bug or change path, so none should be merged as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the most suitable anchor because it is the widest umbrella problem in the set and best matches the likely cross-cutting regression cluster.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44625", - "right": "issue:46002", - "accept": false, - "reason": "Qwen3.5 config propagation vs classification-head NaNs are different symptoms and different code paths." - }, - { - "left": "issue:38617", - "right": "issue:45278", - "accept": false, - "reason": "Both involve imports, but one is a specific missing symbol and the other is a broad post-upgrade import-breakage report." - }, - { - "left": "issue:44380", - "right": "issue:46077", + "left": "issue:44117", + "right": "issue:45588", "accept": false, - "reason": "Different attention bugs: GPT-2 scaling under SDPA/FlashAttention vs Gemma4 bidirectional mask construction." + "reason": "Different subsystems: tokenizer mapping fallback vs flash-attention crash handling for sink-less models." }, { "left": "issue:39401", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different models and different failure modes (offset mapping vs incorrect tokenization output)." - }, - { - "left": "issue:42175", - "right": "issue:44938", - "accept": false, - "reason": "Backend installation/package extras issue is unrelated to Python 3.14 load failure." - }, - { - "left": "issue:44117", - "right": "issue:45588", - "accept": false, - "reason": "Tokenizer mapping None and flash-attention s_aux handling are unrelated subsystems." + "reason": "Both are tokenizer bugs, but one is Qwen3 offset mappings and the other is DeepSeek tokenization output; different concrete failures." }, { "left": "issue:43064", "right": "issue:43673", "accept": false, - "reason": "FSDP/PEFT optimizer-state corruption is unrelated to chunked-prefill cache handling." + "reason": "Optimizer state sharding on FSDP2/PEFT is unrelated to GenerationMixin cache missing during chunked prefill." }, { "left": "issue:43381", "right": "issue:44945", "accept": false, - "reason": "Gradient checkpointing in eval mode is unrelated to pipeline-parallel output correctness." + "reason": "Eval-mode gradient checkpointing behavior is a different code path from pipeline-parallel incorrect outputs." }, { "left": "issue:43827", "right": "issue:45446", "accept": false, - "reason": "Documentation leftovers after pipeline removal are unrelated to a flex_attention version check bug." + "reason": "Docs still mentioning pipeline() is unrelated to a runtime PyTorch version check bug." }, { "left": "issue:43450", "right": "issue:43986", "accept": false, - "reason": "Batched video-shape bug and missing-torchvision crash are different issues." + "reason": "Batched video processor shape handling is not the same as AutoProcessor crashing without torchvision." }, { "left": "issue:43937", "right": "issue:44977", "accept": false, - "reason": "GenerationConfig validation failure and flash-attention generation failure are not the same bug." + "reason": "GenerationConfig validation failure is different from Qwen3.5 flash-attention generation behavior." }, { "left": "issue:44466", "right": "issue:46002", "accept": false, - "reason": "Serialization of tied weights is unrelated to NaNs in classification heads after checkout/init." + "reason": "lm_head serialization/tied-weights issue is unrelated to NaNs in classification heads during checkout/init." }, { "left": "issue:45405", "right": "issue:45446", "accept": false, - "reason": "Released-version dependency bump and PyTorch version gating are unrelated." + "reason": "A PEFT minimum-version bump is not the same bug as an AuxRequest import version check." + }, + { + "left": "issue:42673", + "right": "issue:45072", + "accept": false, + "reason": "VRAM leak in multithreaded Qwen3 inference is unrelated to dtype mismatches in CI inference tests." }, { "left": "issue:44117", "right": "issue:45081", "accept": false, - "reason": "Generic tokenizer mapping failure is not the same as Mistral regex patch crashing on backend_tokenizer access." + "reason": "Tokenizer mapping None and Mistral regex patch crashes affect different failure points." }, { "left": "issue:43122", "right": "issue:46298", "accept": false, - "reason": "Tokenization regression and noisy logging message do not describe the same underlying problem." - }, - { - "left": "issue:42673", - "right": "issue:45072", - "accept": false, - "reason": "VRAM leak in multithreaded loading and dtype mismatch in bfloat16 inference are unrelated." + "reason": "Tokenization output differences and a noisy logging message are unrelated." }, { "left": "issue:44410", "right": "issue:45310", "accept": false, - "reason": "Missing Qwen3Next projections is a model-architecture bug, not the Qwen3.5 MoE from_pretrained loading error." + "reason": "Missing Qwen3Next projections is a model-architecture bug, not the same as Qwen3.5 MoE from_pretrained loading failure." }, { "left": "issue:44291", "right": "issue:45071", "accept": false, - "reason": "Unexpected init argument handling and PreTrainedConfig type checking are different failures." + "reason": "Unexpected _is_hf_initialized argument and PretrainedConfig type-checking are different compatibility issues." }, { "left": "issue:44829", "right": "issue:45216", "accept": false, - "reason": "FlashAttention-3 training degeneration and Qwen3.5 save_pretrained checkpoint corruption are different bugs." + "reason": "Flash-attention training degeneration is not the same as save_pretrained checkpoint serialization regression." }, { - "left": "issue:39401", - "right": "issue:43421", + "left": "issue:42222", + "right": "issue:46178", "accept": false, - "reason": "Wrong offset mapping and runtime post-processor updates are both tokenizer-related but not the same defect." + "reason": "VitPose model breakage and backend-mapping errors for tensorflow_text are unrelated load-path problems." }, { "left": "issue:38617", "right": "issue:41628", "accept": false, - "reason": "Different missing-import symbols with no evidence of the same fix path." + "reason": "Both are import errors, but they involve different missing symbols and different removal/compatibility paths." }, { - "left": "issue:42222", - "right": "issue:46178", + "left": "issue:39401", + "right": "issue:43421", "accept": false, - "reason": "Vitpose model breakage and tensorflow_text backend mapping error are unrelated." + "reason": "Wrong offset mapping and runtime post-processor refresh behavior are different tokenizer issues." }, { "left": "issue:43408", "right": "issue:43742", "accept": false, - "reason": "Model-type warning for sam3_tracker is unrelated to loading MobileLLM-125M with a key error." + "reason": "Model-type mismatch warning for SAM3 trackers is unrelated to the MobileLLM config key error." }, { "left": "issue:44877", "right": "issue:46231", "accept": false, - "reason": "Strict config loading and documentation parameter mismatch are unrelated." + "reason": "Strict config loading for granite_speech is unrelated to a docs parameter-name mismatch." }, { "left": "issue:44484", - "right": "issue:45464", + "right": "issue:45310", "accept": false, - "reason": "Both mention chat/completions, but one is a general API regression while the other is a specific Qwen3.5 streaming failure." + "reason": "A question about max_shard_size has no overlap with a Qwen3.5 MoE loading error." }, { "left": "issue:43937", "right": "issue:44496", "accept": false, - "reason": "GenerationConfig validation and unrecognized model/config loading are different problems." + "reason": "Invalid GenerationConfig and unrecognized model/config.json issues are distinct loading failures." }, { "left": "issue:43066", "right": "issue:43122", "accept": false, - "reason": "Wrong decoder type and changed tokenization output are related only at a broad version-regression level, not the same concrete bug." + "reason": "Wrong tokenizer decoder type and changed tokenization output across versions are not the same concrete defect." }, { "left": "issue:44483", "right": "issue:45464", "accept": false, - "reason": "Chat/completions request rejection and Qwen3.5 streaming failure are not the same underlying bug." + "reason": "Chat/completions request rejection and streaming inference failure on Qwen3.5-0.8B are different API-path bugs." + }, + { + "left": "issue:44075", + "right": "issue:45656", + "accept": false, + "reason": "SGD args being ignored is unrelated to deepspeed calling optimizer.step() twice." + }, + { + "left": "issue:44964", + "right": "issue:45278", + "accept": false, + "reason": "A specific Phi-4 multimodal loading failure is not the same as broad import errors after upgrading transformers." + }, + { + "left": "issue:42371", + "right": "issue:44261", + "accept": false, + "reason": "TF32 configuration warnings are unrelated to MLA q_a_layernorm precision/config handling." + }, + { + "left": "issue:45278", + "right": "issue:41628", + "accept": false, + "reason": "General import errors after upgrade are too broad to equate with one specific missing import symbol." + }, + { + "left": "issue:44977", + "right": "issue:45081", + "accept": false, + "reason": "Qwen3.5 flash-attention generation problems and Mistral regex patch crashes are different tokenizer/model paths." } ] }, @@ -13852,13 +13913,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 44, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:38617", "issue:41628", @@ -13866,8 +13927,10 @@ "issue:42222", "issue:42371", "issue:42673", + "issue:42757", "issue:42831", "issue:43012", + "issue:43064", "issue:43097", "issue:43335", "issue:43408", @@ -13880,11 +13943,12 @@ "issue:43976", "issue:43986", "issue:44038", - "issue:44075", + "issue:44062", "issue:44117", "issue:44220", "issue:44261", "issue:44263", + "issue:44315", "issue:44360", "issue:44368", "issue:44410", @@ -13894,29 +13958,24 @@ "issue:44829", "issue:44933", "issue:44945", - "issue:44964", "issue:44977", "issue:44987", "issue:45071", - "issue:45278", + "issue:45200", "issue:45440", "issue:45478", "issue:45542", "issue:45588", - "issue:45656", "issue:46050", "issue:46172", "issue:46231" ], "soft_pairs": [ - "issue:44075|issue:45656", - "issue:44964|issue:45278", - "issue:42371|issue:44261", "issue:42222|issue:45542", "issue:43012|issue:43577", "issue:44410|issue:45478", - "issue:38617|issue:44220", "issue:42175|issue:42222", + "issue:38617|issue:44220", "issue:43874|issue:44261", "issue:43097|issue:44493", "issue:41628|issue:45071", @@ -13924,8 +13983,8 @@ "issue:45071|issue:45542", "issue:42673|issue:44977", "issue:42673|issue:43906", - "issue:44263|issue:46172", "issue:43012|issue:43408", + "issue:44263|issue:46172", "issue:44360|issue:46172", "issue:44484|issue:45478", "issue:43441|issue:44829", @@ -13940,236 +13999,246 @@ "issue:43976|issue:44945", "issue:43986|issue:44484", "issue:45440|issue:46172", + "issue:43097|issue:45588", + "issue:43064|issue:44315", "issue:43097|issue:46050", - "issue:43097|issue:45588" + "issue:42673|issue:45200", + "issue:38617|issue:42757", + "issue:44062|issue:44117" ], - "prepared_review_unit_hash": "5d05464b4f90a55d125641f12af2268869f06cd700d584333ffcf10df06d7723", + "prepared_review_unit_hash": "5bbef32ea2fb55d41feb3d8ba479ae302a0f7ff09ddebeb063740bbf238145e6", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 33, - "serialized_chars": 36829, - "estimated_input_tokens": 9208, - "estimated_eval_tokens": 18672 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37842, + "estimated_input_tokens": 9461, + "estimated_eval_tokens": 19178 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is heterogeneous, with mostly unrelated transformer regressions and API/behavior issues. None of the soft pairs look like the same underlying bug or fix, so I would not merge any of them as duplicates.", + "summary": "No soft edge looks like a true duplicate; most pairs are only loosely related by subsystem, model family, or symptom type. I\u2019d anchor the cluster on issue:43097 as the clearest high-signal regression report, but there is no mergeable PR here.", "confidence": 0.61, - "canonical_issue_reason": "issue:45278 is the broadest and most umbrella-like issue in the set, centered on widespread import breakage after upgrading Transformers, which makes it the best representative target for this mixed cluster.", + "canonical_issue_reason": "issue:43097 is the clearest, most broadly impactful regression report in this set, with explicit references and a concrete API-break symptom.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 best captures the general user-facing regression theme in the cluster and is broad enough to subsume related import/load failures better than the narrower, model-specific reports.", + "best_issue_reason": "issue:43097 is the strongest representative issue: it is specific, reproducible, and likely to attract the most relevant fix context.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44075", - "right": "issue:45656", - "accept": false, - "reason": "Both concern optimizers, but one says SGD args are ignored while the other says deepspeed calls optimizer.step twice; different failure modes and fixes." - }, - { - "left": "issue:44964", - "right": "issue:45278", - "accept": false, - "reason": "One is a specific Phi-4 multimodal loading failure; the other is a broad post-upgrade import-error report. Same area, not the same bug." - }, - { - "left": "issue:42371", - "right": "issue:44261", - "accept": false, - "reason": "TF32 API guidance and an MLA rms_norm_eps precision issue are unrelated problems." - }, { "left": "issue:42222", "right": "issue:45542", "accept": false, - "reason": "VitPose breakage and missing TensorFlow backend from a tensorboard-only install are different issues." + "reason": "Different failures: vitpose model breakage vs missing TF backend wiring." }, { "left": "issue:43012", "right": "issue:43577", "accept": false, - "reason": "Both mention dtype/precision, but one is a compile warning on bfloat16 models and the other is a Blip2 loading bug where dtype fields stay float32." + "reason": "Both involve dtype/precision, but the model paths and bugs differ." }, { "left": "issue:44410", "right": "issue:45478", "accept": false, - "reason": "Both involve Qwen3-family models, but one is a missing projection in qwen3next and the other is a Qwen3.5 MoE from_pretrained error; not the same code path." + "reason": "Different model families and different loader failures; not the same bug." }, { - "left": "issue:38617", - "right": "issue:44220", + "left": "issue:42175", + "right": "issue:42222", "accept": false, - "reason": "An import error from configuration_utils and a feature-extraction fbank issue are unrelated." + "reason": "Packaging/backend dependency issue vs a specific model regression." }, { - "left": "issue:42175", - "right": "issue:42222", + "left": "issue:38617", + "right": "issue:44220", "accept": false, - "reason": "TensorFlow backend packaging and broken VitPose models are different bugs." + "reason": "Unrelated import error versus audio feature extraction bug." }, { "left": "issue:43874", "right": "issue:44261", "accept": false, - "reason": "A missing image-patch helper in a fast image processor and an MLA layernorm epsilon precision issue do not share the same underlying defect." + "reason": "Different GLM subcomponents and different failure modes." }, { "left": "issue:43097", "right": "issue:44493", "accept": false, - "reason": "One is a config removal/regression around tied embeddings, the other is unexpected position-id key warnings; different symptoms and fixes." + "reason": "Config-removal warning is not the same as unexpected position-id key errors." }, { "left": "issue:41628", "right": "issue:45071", "accept": false, - "reason": "Missing AutoImageProcessor import and PretrainedConfig type-checking regressions are unrelated API breakages." + "reason": "Import failure for AutoImageProcessor vs a type-checking regression." }, { "left": "issue:44263", "right": "issue:44485", "accept": false, - "reason": "Both mention GLM/DSA, but one is about torch.split return values and the other is a RoPE implementation bug." + "reason": "Tokenizer/indexer split bug and RoPE implementation bug are distinct." }, { "left": "issue:45071", "right": "issue:45542", "accept": false, - "reason": "PretrainedConfig type checking and an undefined tf backend error from tensorboard-only installs are separate problems." + "reason": "Type checking regression is unrelated to TF backend installation problems." }, { "left": "issue:42673", "right": "issue:44977", "accept": false, - "reason": "VRAM leakage in multi-threaded dataloaders and Qwen3.5 flash-attention generation failure are different bugs." + "reason": "VRAM leak in threaded loading is not the flash-attention generation bug." }, { "left": "issue:42673", "right": "issue:43906", "accept": false, - "reason": "The VRAM leak report is about Qwen3ForCausalLM threading; the linked reproduction issue is not the same concrete failure." + "reason": "No evidence they describe the same underlying bug." }, { - "left": "issue:44263", - "right": "issue:46172", + "left": "issue:43012", + "right": "issue:43408", "accept": false, - "reason": "Different model families and different code paths: a DSA indexer split issue versus a DeepSeek V4 dtype/transpose bug." + "reason": "Different model-specific warnings with different causes." }, { - "left": "issue:43012", - "right": "issue:43408", + "left": "issue:44263", + "right": "issue:46172", "accept": false, - "reason": "A bfloat16 precision warning during compile and a SAM3 model-type mismatch warning are unrelated." + "reason": "Different tensor-shape/indexing issues in different model code paths." }, { "left": "issue:44360", "right": "issue:46172", "accept": false, - "reason": "A missing ReLU in the DSA indexer and a DeepSeek V4 comb dtype/transpose issue are not the same bug." + "reason": "Activation omission and dtype/transpose bug are unrelated." }, { "left": "issue:44484", "right": "issue:45478", "accept": false, - "reason": "A save_pretrained shard-size question and a Qwen3.5 MoE loading error have no shared underlying fix." + "reason": "Shard-size question vs Qwen3.5 model-loading failure." }, { "left": "issue:43441", "right": "issue:44829", "accept": false, - "reason": "Both involve FlashAttention, but one is a Ministral-3 failure and the other is degenerate training with flash_attention_3; not mergeable as one PR." + "reason": "Both mention flash attention, but the models and symptoms differ." }, { "left": "issue:42371", "right": "issue:42831", "accept": false, - "reason": "TF32 settings guidance and FineGrainedFP8 accuracy loss are separate numeric-behavior issues." + "reason": "TF32 warning/config discussion is not the same as FineGrainedFP8 accuracy loss." }, { "left": "issue:43931", "right": "issue:44038", "accept": false, - "reason": "A Qwen3-VL checkpoint shape mismatch and a Qwen3-VL-Moe issue on Transformers 5.0 are related only by model family, not by the same concrete bug." + "reason": "Related model family, but different load-time errors and code paths." }, { "left": "issue:43335", "right": "issue:45071", "accept": false, - "reason": "SwitchTransformers sparse-layer config bug and PretrainedConfig type-checking regression are unrelated." + "reason": "Sparse-layer config bug is unrelated to PretrainedConfig type checking." }, { "left": "issue:43097", "right": "issue:43452", "accept": false, - "reason": "Tied-embedding config removal and gguf_file loading failures are different loading regressions." + "reason": "API removal warning and gguf loading breakage are different issues." }, { "left": "issue:43874", "right": "issue:44263", "accept": false, - "reason": "Missing get_number_of_image_patches and torch.split return handling are different implementation bugs." + "reason": "Different GLM bugs: missing method vs incorrect split outputs." }, { "left": "issue:44933", "right": "issue:46231", "accept": false, - "reason": "A missing image_utils import and a documentation typo in set_training are unrelated." + "reason": "Missing import and documentation typo are not duplicates." }, { "left": "issue:44117", "right": "issue:44987", "accept": false, - "reason": "Tokenizer mapping returning None and a physical-intelligence/fast load failure are different issues." + "reason": "Tokenizer mapping assumption bug versus a specific model loading failure." }, { "left": "issue:43408", "right": "issue:44368", "accept": false, - "reason": "Both are warnings, but one is a model-type mismatch and the other is a tied-embeddings config warning; not the same bug." + "reason": "Different warning text and different model/config contexts." }, { "left": "issue:43976", "right": "issue:44945", "accept": false, - "reason": "Python compatibility breakage and incorrect output under pipeline parallelism are unrelated." + "reason": "Python version compatibility issue vs pipeline-parallelism output corruption." }, { "left": "issue:43986", "right": "issue:44484", "accept": false, - "reason": "A crash without torchvision and a save_pretrained shard-size question do not share the same underlying defect." + "reason": "AutoProcessor torchvision dependency issue is unrelated to max shard size." }, { "left": "issue:45440", "right": "issue:46172", "accept": false, - "reason": "Both mention DeepSeek, but one is a native-vs-remote implementation divergence and the other is a dtype/transpose bug in DeepSeek V4." + "reason": "Different DeepSeek model generations and different implementation bugs." }, { "left": "issue:43097", - "right": "issue:46050", + "right": "issue:45588", "accept": false, - "reason": "Tie-embedding config removal and sinq quantization save/load breakage are unrelated regressions." + "reason": "Config-removal warning is unrelated to flash-attention AttributeError." + }, + { + "left": "issue:43064", + "right": "issue:44315", + "accept": false, + "reason": "Distributed optimizer-state bug is not the same as Liger kernel application." }, { "left": "issue:43097", - "right": "issue:45588", + "right": "issue:46050", "accept": false, - "reason": "Tied-embedding config removal and a flash_attention crash on s_aux=None are different code-path failures." - } + "reason": "API removal warning and quantized save/load breakage are distinct." + }, + { + "left": "issue:42673", + "right": "issue:45200", + "accept": false, + "reason": "Threaded VRAM leak has nothing to do with Gemma 4 multimodal token IDs." + }, + { + "left": "issue:38617", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but for different symbols and modules." + }, + { + "left": "issue:44062", + "right": "issue:44117", + "accept": false, + "reason": "Tokenizer AddedToken construction bug is unrelated to tokenizer-mapping None handling." + } ] }, "evaluator_result": null, @@ -14184,22 +14253,19 @@ "component_index": 1, "component_count": 18, "review_unit_index": 45, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:38617", "issue:42617", - "issue:42673", - "issue:42757", "issue:42890", "issue:42907", "issue:43012", - "issue:43064", "issue:43066", "issue:43097", "issue:43208", @@ -14213,25 +14279,27 @@ "issue:43986", "issue:43992", "issue:44038", - "issue:44062", "issue:44117", "issue:44162", + "issue:44222", "issue:44261", - "issue:44315", "issue:44351", "issue:44368", "issue:44410", + "issue:44483", "issue:44484", "issue:44485", + "issue:44514", "issue:44655", "issue:44704", + "issue:44938", "issue:44964", "issue:45103", - "issue:45200", "issue:45290", "issue:45341", "issue:45372", "issue:45375", + "issue:45478", "issue:45561", "issue:45588", "issue:45712", @@ -14239,31 +14307,28 @@ "issue:45859", "issue:46123", "issue:46172", + "issue:46178", "issue:46298" ], "soft_pairs": [ - "issue:43064|issue:44315", - "issue:42673|issue:45200", - "issue:38617|issue:42757", - "issue:44062|issue:44117", "issue:44410|issue:45375", "issue:42890|issue:46298", - "issue:44038|issue:44410", "issue:43931|issue:44410", + "issue:44038|issue:44410", "issue:43502|issue:43992", "issue:45341|issue:45859", - "issue:43097|issue:43873", "issue:38617|issue:44351", + "issue:43097|issue:43873", "issue:44261|issue:44485", "issue:44964|issue:45341", "issue:44704|issue:45103", "issue:43986|issue:44117", "issue:42617|issue:45341", - "issue:43066|issue:43906", "issue:43986|issue:45372", + "issue:43066|issue:43906", "issue:43986|issue:45290", - "issue:45341|issue:45588", "issue:45341|issue:45561", + "issue:45341|issue:45588", "issue:43208|issue:43550", "issue:43986|issue:45341", "issue:43992|issue:44484", @@ -14275,240 +14340,237 @@ "issue:42907|issue:45753", "issue:43097|issue:46298", "issue:44162|issue:46172", - "issue:43792|issue:46123" + "issue:43792|issue:46123", + "issue:44938|issue:46178", + "issue:44483|issue:44514", + "issue:44222|issue:45478" ], - "prepared_review_unit_hash": "1b05eb4b1c664d447527d9646f02fec84ccc8692f1cbb5eb33c7cb929450235c", + "prepared_review_unit_hash": "6eb95c7b03f7ba1c3987155558fff7edb69417820347e94a9b840b90506edf9b", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 34, - "serialized_chars": 37751, - "estimated_input_tokens": 9438, - "estimated_eval_tokens": 19132 + "soft_pair_count": 33, + "serialized_chars": 37333, + "estimated_input_tokens": 9334, + "estimated_eval_tokens": 18924 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items span many unrelated Transformers bugs, warnings, and model-specific regressions. The soft edges are only superficial text overlap (imports, warnings, save/load, model names), so none look like the same underlying issue.", - "confidence": 0.93, - "canonical_issue_reason": "No true duplicate cluster is evident; if one issue must anchor the set, issue:43873 is the most broadly discussed active bug report and the least arbitrary representative.", + "summary": "This cluster is highly heterogeneous: the soft pairs mostly share only broad subsystem keywords (imports, loading, save_pretrained, testing, Qwen/processor paths) rather than the same concrete bug. I rejected all soft edges and found no clear canonical issue/PR within this set.", + "confidence": 0.97, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 is the best representative issue by activity and breadth, but it should be treated as a standalone bug report rather than a duplicate hub.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43064", - "right": "issue:44315", - "accept": false, - "reason": "Different bugs: FSDP2/PEFT optimizer-state corruption vs model_init not applying Liger Kernel." - }, - { - "left": "issue:42673", - "right": "issue:45200", - "accept": false, - "reason": "VRAM leak in multithreaded dataloaders is unrelated to Gemma 4 mm_token_type_ids defaults." - }, - { - "left": "issue:38617", - "right": "issue:42757", - "accept": false, - "reason": "Both are import errors, but for different missing symbols and different dependency paths." - }, - { - "left": "issue:44062", - "right": "issue:44117", - "accept": false, - "reason": "Tokenizer AddedToken constructor error and TOKENIZER_MAPPING_NAMES None are distinct failure modes." - }, { "left": "issue:44410", "right": "issue:45375", "accept": false, - "reason": "Different model issues: missing Qwen3Next projections vs a Qwen3_5MoeVisionConfig field being dropped." + "reason": "Both are Qwen3-family config/model issues, but they describe different missing fields/projection problems and likely different code paths." }, { "left": "issue:42890", "right": "issue:46298", "accept": false, - "reason": "A flaky test seed issue is unrelated to a noisy logging message." + "reason": "One is a seeding-related test flake; the other is a noisy logging message. Same testing area only at a very broad level." }, { - "left": "issue:44038", + "left": "issue:43931", "right": "issue:44410", "accept": false, - "reason": "Qwen3-VL-Moe bug and qwen3next missing projections affect different models/code paths." + "reason": "Model weight-shape mismatch for Qwen3-VL is not the same bug as missing attention projections in qwen3next." }, { - "left": "issue:43931", + "left": "issue:44038", "right": "issue:44410", "accept": false, - "reason": "Weight-shape mismatch on model load is not the same as missing attention projections." + "reason": "Qwen3-VL-Moe loading breakage and qwen3next missing projections are distinct model-specific failures." }, { "left": "issue:43502", "right": "issue:43992", "accept": false, - "reason": "local_files_only network access bug is unrelated to missing embed_tokens.weight during load." + "reason": "local_files_only causing network requests is a loader/network bug; missing embed_tokens.weight is a model checkpoint loading mismatch." }, { "left": "issue:45341", "right": "issue:45859", "accept": false, - "reason": "Testing utility bug vs Qwen3_5MoeTextRotaryEmbedding CPU-offload incompatibility are unrelated." + "reason": "A testing_utils bug and a Qwen3.5 rotary embedding CPU-offload bug are unrelated." }, { - "left": "issue:43097", - "right": "issue:43873", + "left": "issue:38617", + "right": "issue:44351", "accept": false, - "reason": "Removed config attribute warning and quantization/offloading behavior are different regressions." + "reason": "Both are ImportErrors, but for different symbols and likely different release/compatibility regressions." }, { - "left": "issue:38617", - "right": "issue:44351", + "left": "issue:43097", + "right": "issue:43873", "accept": false, - "reason": "Different missing imports: layer_type_validation vs HybridCache." + "reason": "Removed config flag documentation/behavior and quantization offloading issues are different problems." }, { "left": "issue:44261", "right": "issue:44485", "accept": false, - "reason": "RMS norm epsilon precision issue and GLM-5 RoPE implementation are separate model internals bugs." + "reason": "Different model internals: MLA layernorm epsilon handling vs GLM-5 RoPE implementation." }, { "left": "issue:44964", "right": "issue:45341", "accept": false, - "reason": "Phi-4 multimodal loading failure has nothing in common with a testing_utils bug." + "reason": "Phi-4 multimodal loading failure is unrelated to a testing_utils bug." }, { "left": "issue:44704", "right": "issue:45103", "accept": false, - "reason": "AutoProcessor kwargs forwarding and auto_docstring annotation handling are different utilities." + "reason": "AutoProcessor kwargs forwarding and an auto_docstring annotation crash are separate code paths." }, { "left": "issue:43986", "right": "issue:44117", "accept": false, - "reason": "AutoProcessor/torchvision crash and tokenizer mapping None are unrelated." + "reason": "Video-model AutoProcessor crash without torchvision and TOKENIZER_MAPPING_NAMES returning None are different loader failures." }, { "left": "issue:42617", "right": "issue:45341", "accept": false, - "reason": "3d_parallel.py execution failure is not the same as a testing_utils issue." + "reason": "Running 3d_parallel.py and a testing_utils bug do not describe the same underlying defect." }, { - "left": "issue:43066", - "right": "issue:43906", + "left": "issue:43986", + "right": "issue:45372", "accept": false, - "reason": "Wrong tokenizer decoder type and an isolated reproduction of another issue are not the same concrete bug." + "reason": "Both affect processor loading, but one is a missing torchvision path and the other is a missing mistral_common symbol." }, { - "left": "issue:43986", - "right": "issue:45372", + "left": "issue:43066", + "right": "issue:43906", "accept": false, - "reason": "Missing torchvision crash is unrelated to the mistral_common ReasoningEffort import error." + "reason": "Wrong tokenizer decoder type in v5 is not clearly the same as an isolated reproduction of another issue." }, { "left": "issue:43986", "right": "issue:45290", "accept": false, - "reason": "AutoProcessor video-model crash and chat-template tool-call crash use different code paths." + "reason": "Processor loading without torchvision and apply_chat_template tool-call handling are distinct failures." }, { "left": "issue:45341", - "right": "issue:45588", + "right": "issue:45561", "accept": false, - "reason": "Testing utilities bug and flash_attention s_aux=None crash are unrelated." + "reason": "A helper bug in testing_utils and an xdist race on captured_info.txt are not the same bug." }, { "left": "issue:45341", - "right": "issue:45561", + "right": "issue:45588", "accept": false, - "reason": "Testing utilities bug and pytest-xdist file race are different concurrency problems." + "reason": "Testing utils vs flash_attention sinkless-model crash: different modules and failure modes." }, { "left": "issue:43208", "right": "issue:43550", "accept": false, - "reason": "xLSTM training blockers and Bamba torch.compile SDPA failure are model-specific and distinct." + "reason": "xLSTM training bugs and a Bamba SDPA torch.compile failure are unrelated model/runtime issues." }, { "left": "issue:43986", "right": "issue:45341", "accept": false, - "reason": "AutoProcessor/torchvision crash is unrelated to the testing_utils issue." + "reason": "Processor/torchvision loading failure is unrelated to the testing_utils issue despite both being utility-level bugs." }, { "left": "issue:43992", "right": "issue:44484", "accept": false, - "reason": "Missing embed_tokens.weight during load is unrelated to max_shard_size save behavior." + "reason": "Missing checkpoint weights and a question about save_pretrained sharding size are not the same bug." }, { "left": "issue:43012", "right": "issue:43550", "accept": false, - "reason": "A PyTorch precision warning during compile is not the same as a SDPA runtime failure." + "reason": "A bfloat16 compile warning and a Bamba torch.compile SDPA failure are different PyTorch behaviors." }, { "left": "issue:43825", "right": "issue:44655", "accept": false, - "reason": "pipeline() translation messaging and saving Pipeline objects are different pipeline bugs." + "reason": "A misleading pipeline error message and inability to save Pipeline objects are separate pipeline issues." }, { "left": "issue:43012", "right": "issue:44368", "accept": false, - "reason": "Both mention warnings, but they come from different code paths and underlying issues." + "reason": "The bfloat16 precision warning is unrelated to a tie_word_embeddings warning from fine-tuning." }, { "left": "issue:45712", "right": "issue:46298", "accept": false, - "reason": "Leftover dummy classes and noisy logging are unrelated cleanup vs logging problems." + "reason": "Leftover dummy classes in dummy_pt_objects.py and a noisy logging message are not the same defect." }, { "left": "issue:42907", "right": "issue:44655", "accept": false, - "reason": "Dequantized model save failure is not the same as Pipeline.save_pretrained serialization." + "reason": "Saving dequantized models and saving Pipeline objects both involve serialization, but they are different code paths and failures." }, { "left": "issue:42907", "right": "issue:45753", "accept": false, - "reason": "Model save failure and an infinite-loop image input bug are unrelated." + "reason": "Model save failure for dequantized Ministrals/Devstrals is unrelated to an infinite-loop inference bug." }, { "left": "issue:43097", "right": "issue:46298", "accept": false, - "reason": "Removed tie_embeddings_and_encoder_decoder and a noisy logging message are not the same issue." + "reason": "A removed config behavior and noisy logging do not point to the same underlying issue." }, { "left": "issue:44162", "right": "issue:46172", "accept": false, - "reason": "ESM2 breakage and deepseek-v4 comb transpose bug are separate model-specific failures." + "reason": "ESM2 breakage and a DeepSeek-v4 dtype/transpose bug are unrelated model-specific regressions." }, { "left": "issue:43792", "right": "issue:46123", "accept": false, - "reason": "Whisper loading failure and MaskGenerationPipeline partial-batch drop are unrelated." + "reason": "Whisper-large-v2 runtime failure and MaskGenerationPipeline partial-batch loss are different pipeline/model bugs." + }, + { + "left": "issue:44938", + "right": "issue:46178", + "accept": false, + "reason": "Python 3.14 import/load failure and tensorflow_text backend mapping error are different environment/backend issues." + }, + { + "left": "issue:44483", + "right": "issue:44514", + "accept": false, + "reason": "Chat-completions request rejection and apply_chat_template batching crash are separate API-path bugs." + }, + { + "left": "issue:44222", + "right": "issue:45478", + "accept": false, + "reason": "FP8 save_pretrained for moe models and a Qwen3.5 MoE from_pretrained error are not the same concrete failure." } ] }, @@ -14524,13 +14586,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 46, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:36010", "issue:36683", @@ -14554,37 +14616,34 @@ "issue:43761", "issue:43824", "issue:43867", + "issue:43881", "issue:43901", + "issue:43906", "issue:43937", "issue:44038", - "issue:44222", "issue:44230", + "issue:44242", "issue:44380", "issue:44483", "issue:44485", - "issue:44514", "issue:44625", "issue:44655", "issue:44683", - "issue:44938", "issue:45216", "issue:45276", "issue:45290", - "issue:45478", "issue:45600", "issue:45698", + "issue:45753", "issue:46077", "issue:46082", "issue:46123", + "issue:46132", "issue:46172", - "issue:46178", "issue:46209", "issue:46298" ], "soft_pairs": [ - "issue:44938|issue:46178", - "issue:44483|issue:44514", - "issue:44222|issue:45478", "issue:43208|issue:43526", "issue:43012|issue:44683", "issue:42490|issue:43012", @@ -14600,221 +14659,238 @@ "issue:43299|issue:43867", "issue:43066|issue:46209", "issue:36683|issue:43475", - "issue:45276|issue:46077", "issue:42994|issue:44655", + "issue:45276|issue:46077", "issue:43901|issue:44655", "issue:44625|issue:46077", - "issue:36010|issue:42757", "issue:43352|issue:43937", "issue:43611|issue:44655", + "issue:36010|issue:42757", "issue:43867|issue:45216", "issue:46077|issue:46082", "issue:43502|issue:44485", "issue:36683|issue:43525", - "issue:43618|issue:45216" + "issue:43618|issue:45216", + "issue:36683|issue:44242", + "issue:44380|issue:46132", + "issue:36683|issue:43881", + "issue:43906|issue:45753", + "issue:43299|issue:45753" ], - "prepared_review_unit_hash": "54c995427b89620d8f53e33d6aae32a9d796a773760871b304d026c85e399138", + "prepared_review_unit_hash": "f12a4213e247b18bdd010f3fd71cec91345de9e9cf12b9cf7f29151db0308b30", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36257, - "estimated_input_tokens": 9065, - "estimated_eval_tokens": 18386 + "soft_pair_count": 32, + "serialized_chars": 36833, + "estimated_input_tokens": 9209, + "estimated_eval_tokens": 18674 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The items are a loose collection of distinct Transformers bugs and regressions rather than a single duplicate cluster; the soft pairs are mostly superficial matches on broad terms like import, save, or model loading.", - "confidence": 0.97, - "canonical_issue_reason": "Issue 45698 is the strongest representative only by signal strength: it is open, has the most discussion/inbound attention, and describes a concrete, user-visible regression around from_pretrained/save_pretrained behavior.", + "summary": "This cluster is not a true duplicate set: the items span unrelated Transformers regressions, import failures, config/schema issues, saving/loading bugs, and model-specific behavior changes. None of the soft edges look like the same concrete bug or change.", + "confidence": 0.95, + "canonical_issue_reason": "No single canonical issue cleanly represents the cluster; the topics are too heterogeneous. If forced to pick one representative, issue 45698 is the broadest active save/load regression, but it does not cover the rest.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45698 is the best single issue to anchor the cluster if one must be chosen, because it is active and centrally phrased as a reproducible regression with broader impact than the mostly narrow model-specific reports.", + "best_issue_reason": "Issue 45698 is the best representative only in a loose sense because it is an active, broadly described save/load regression with multiple references, but it is not a duplicate umbrella for the other items.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44938", - "right": "issue:46178", - "accept": false, - "reason": "Different failures: Python 3.14 import/load issue vs an invalid backend name in backend mapping." - }, - { - "left": "issue:44483", - "right": "issue:44514", - "accept": false, - "reason": "Different surfaces and bugs: chat/completions request validation vs chat template batching crash." - }, - { - "left": "issue:44222", - "right": "issue:45478", - "accept": false, - "reason": "Both are Qwen3.5 MoE-related, but one is FP8 save_pretrained and the other is from_pretrained loading; not the same concrete bug." - }, { "left": "issue:43208", "right": "issue:43526", "accept": false, - "reason": "Completely unrelated: xLSTM training bugs vs BeitImageProcessorFast label reduction behavior." + "reason": "Different bugs: xLSTM training blockers vs a BeitImageProcessorFast label-reduction bug." }, { "left": "issue:43012", "right": "issue:44683", "accept": false, - "reason": "Different code paths: a PyTorch precision warning during compile vs a flex_attention failure on newer torch." + "reason": "Different failures: a PyTorch warning during bfloat16 compilation vs a hard compile error in flex_attention on torch>=2.9." }, { "left": "issue:42490", "right": "issue:43012", "accept": false, - "reason": "Local-path import warning is unrelated to bf16 compile warnings." + "reason": "Unrelated symptoms: local-path slash import warning vs bfloat16 precision warning." }, { "left": "issue:42757", "right": "issue:43824", "accept": false, - "reason": "Both are import errors, but for different symbols and different dependency surfaces." + "reason": "Both are import errors, but for different symbols and modules; not the same underlying breakage." }, { "left": "issue:45600", "right": "issue:46298", "accept": false, - "reason": "CI break from removed config references is unrelated to noisy logging." + "reason": "CI break from removed Sam3LiteText configs vs a noisy logging message; not the same bug." }, { "left": "issue:38617", "right": "issue:43502", "accept": false, - "reason": "Missing import symbol vs local_files_only network requests are different bugs." + "reason": "ImportError for layer_type_validation vs API requests despite local_files_only; unrelated code paths." }, { "left": "issue:45698", "right": "issue:46172", "accept": false, - "reason": "save_pretrained/from_pretrained regression is unrelated to the DeepSeek tensor transpose error." + "reason": "Save/load custom module regression vs a deepseek-v4 tensor/dtype transpose issue; different code paths." }, { "left": "issue:44038", "right": "issue:44230", "accept": false, - "reason": "Same model family, but one is a loading bug and the other is FP8 inference support; not one fixable change." + "reason": "Both mention Qwen3-VL-Moe, but one is a loading regression and the other is fp8 inference support; not the same bug." }, { "left": "issue:43673", "right": "issue:46123", "accept": false, - "reason": "GenerationMixin cache handling during chunked prefill is unrelated to pipeline final-batch bookkeeping." + "reason": "Generation cache missing in chunked prefill vs batch-finalization dropping results in MaskGenerationPipeline; unrelated." }, { "left": "issue:44483", "right": "issue:45290", "accept": false, - "reason": "Both involve chat APIs, but one is request acceptance and the other is assistant tool-call template tokenization." + "reason": "Both touch chat APIs, but one is a v5.3 /v1/chat/completions request rejection and the other is a chat-template crash with tool calls." }, { "left": "issue:43064", "right": "issue:44380", "accept": false, - "reason": "Distributed optimizer-state corruption is unrelated to GPT-2 attention scaling under SDPA/FlashAttention." + "reason": "Distributed optimizer-state corruption vs GPT2 attention-scaling being ignored in SDPA/FlashAttention; different problems." }, { "left": "issue:43618", "right": "issue:43761", "accept": false, - "reason": "Both are CLIP v5 regressions, but they concern different missing output fields and are not clearly the same concrete defect." + "reason": "Both are CLIP v5 regressions, but one is missing attentions and the other missing hidden_states; too different to merge as one bug." }, { "left": "issue:43299", "right": "issue:43867", "accept": false, - "reason": "Qwen3VL MoE loading breakage and sorted-state_dict load failure are not the same underlying problem." + "reason": "Qwen3VL-Moe loading regression vs a state_dict-sorting load error; not the same underlying failure." }, { "left": "issue:43066", "right": "issue:46209", "accept": false, - "reason": "Tokenizer decoder type regression is unrelated to FP8 inference on B200." + "reason": "Wrong tokenizer decoder type vs FP8 inference on B200; unrelated issues." }, { "left": "issue:36683", "right": "issue:43475", "accept": false, - "reason": "Gemma3Config vocab_size missing is unrelated to SAM3 vision encoder output attributes." + "reason": "Different missing attributes on different model objects: Gemma3Config.vocab_size vs Sam3VisionEncoderOutput.fpn_position_embeddings." + }, + { + "left": "issue:42994", + "right": "issue:44655", + "accept": false, + "reason": "Quantized model saving failure vs Pipeline save_pretrained serialization; related area but different concrete bugs." }, { "left": "issue:45276", "right": "issue:46077", "accept": false, - "reason": "Both are Gemma4 issues, but one is embedding resize propagation and the other is bidirectional attention masking." + "reason": "Gemma4 embedding resize propagation bug vs Gemma4 attention-mask construction bug; different code paths." }, { - "left": "issue:42994", + "left": "issue:43901", "right": "issue:44655", "accept": false, - "reason": "Quantized model serialization failure and pipeline save_pretrained failure are different save paths." + "reason": "Docs/behavior mismatch for return_all_scores vs inability to save Pipeline objects; unrelated." }, { - "left": "issue:43901", - "right": "issue:44655", + "left": "issue:44625", + "right": "issue:46077", "accept": false, - "reason": "Documentation mismatch for return_all_scores is unrelated to pipeline serialization." + "reason": "Qwen3.5 num_labels propagation issue vs Gemma4 attention masking; different models and bugs." }, { "left": "issue:43352", "right": "issue:43937", "accept": false, - "reason": "FlashAttention support limitation for Nemotron is unrelated to GLM-5 GenerationConfig validation." + "reason": "Flash Attention 2 support gap in Nemotron vs invalid GenerationConfig in GLM-5; unrelated." }, { "left": "issue:43611", "right": "issue:44655", "accept": false, - "reason": "base_model_prefix loading regression is unrelated to saving Pipeline objects." + "reason": "Loading with base_model_prefix regression vs Pipeline save_pretrained failure; different lifecycle stages and failures." + }, + { + "left": "issue:36010", + "right": "issue:42757", + "accept": false, + "reason": "Different import errors from different packages and symbols; not the same duplicate." }, { "left": "issue:43867", "right": "issue:45216", "accept": false, - "reason": "Both touch loading/saving, but one is a state_dict ordering load error and the other is incorrect Qwen3.5 checkpoint saving." + "reason": "State_dict-sorting load error vs incorrect checkpoint saved by save_pretrained; related to serialization but not the same bug." }, { "left": "issue:46077", "right": "issue:46082", "accept": false, - "reason": "Gemma4 attention mask bug and LlamaConfig head_dim validation are unrelated." + "reason": "Gemma4 causal-mask regression vs LlamaConfig head_dim validation; unrelated." }, { "left": "issue:43502", "right": "issue:44485", "accept": false, - "reason": "Offline/local-files behavior is unrelated to GLM-5 RoPE implementation discussion." + "reason": "Offline/local-files-only network leak vs GLM-5 RoPE implementation discussion; no shared concrete bug." }, { - "left": "issue:36010", - "right": "issue:42757", + "left": "issue:36683", + "right": "issue:43525", "accept": false, - "reason": "Both are import errors, but for different missing APIs from different packages." + "reason": "Two different missing config attributes (vocab_size vs pad_token_id); same pattern, but not the same underlying bug." }, { "left": "issue:36683", - "right": "issue:43525", + "right": "issue:44242", "accept": false, - "reason": "Different config attributes on different models, no shared bug." + "reason": "Gemma3Config attribute error vs missing load-balancing loss when output_router_logits=False; unrelated." }, { - "left": "issue:43618", - "right": "issue:45216", + "left": "issue:44380", + "right": "issue:46132", + "accept": false, + "reason": "GPT2 attention scaling ignored in backend kernels vs AttentionInterface.register changing function behavior; different features and bugs." + }, + { + "left": "issue:36683", + "right": "issue:43881", "accept": false, - "reason": "CLIP output regression and Qwen3.5 checkpoint regression are unrelated." + "reason": "Gemma3Config missing vocab_size vs glm-4v-9b loading failed; both are load/config problems but not the same one." + }, + { + "left": "issue:43906", + "right": "issue:45753", + "accept": false, + "reason": "An isolated reproduction of another issue vs an infinite-loop bug on a specific image; insufficient evidence of the same underlying defect." + }, + { + "left": "issue:43299", + "right": "issue:45753", + "accept": false, + "reason": "Qwen3VL-Moe loading regression vs Qwen3_5 infinite loop on one image; different symptoms and code paths." } ] }, @@ -14830,44 +14906,44 @@ "component_index": 1, "component_count": 18, "review_unit_index": 47, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:28282", "issue:33453", "issue:36296", "issue:36683", "issue:41669", + "issue:42548", "issue:42890", "issue:43064", "issue:43097", - "issue:43299", "issue:43352", "issue:43441", "issue:43502", "issue:43575", + "issue:43576", "issue:43618", "issue:43867", - "issue:43881", - "issue:43906", "issue:43976", "issue:43992", "issue:44062", "issue:44112", "issue:44117", "issue:44222", - "issue:44242", "issue:44246", + "issue:44351", "issue:44355", "issue:44360", "issue:44380", "issue:44492", "issue:44655", + "issue:44749", "issue:44855", "issue:44857", "issue:44938", @@ -14882,25 +14958,19 @@ "issue:45479", "issue:45561", "issue:45600", - "issue:45753", "issue:46132", "issue:46202", "issue:46285" ], "soft_pairs": [ - "issue:36683|issue:44242", - "issue:44380|issue:46132", - "issue:36683|issue:43881", - "issue:43906|issue:45753", - "issue:43299|issue:45753", "issue:43352|issue:43976", "issue:43992|issue:45310", "issue:43441|issue:44380", "issue:43352|issue:45478", "issue:44355|issue:44855", "issue:43867|issue:46202", - "issue:43867|issue:45375", "issue:44380|issue:45479", + "issue:43867|issue:45375", "issue:43352|issue:44380", "issue:36683|issue:44117", "issue:43618|issue:44380", @@ -14923,240 +14993,250 @@ "issue:42890|issue:44964", "issue:42890|issue:44112", "issue:42890|issue:45600", - "issue:44222|issue:45125" + "issue:44222|issue:45125", + "issue:43576|issue:46132", + "issue:44355|issue:44749", + "issue:45125|issue:45478", + "issue:42548|issue:44351" ], - "prepared_review_unit_hash": "8aa57892d8fae62adcd8ec080c7f31aa6975323616e01658b48a46ae5d9c6dda", + "prepared_review_unit_hash": "c211c314d48ef32cd94d158030c34668bac6708acf565e33931c77cc453b445e", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 36, - "serialized_chars": 38087, - "estimated_input_tokens": 9522, - "estimated_eval_tokens": 19300 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 35, + "serialized_chars": 37289, + "estimated_input_tokens": 9323, + "estimated_eval_tokens": 18902 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a heterogeneous mix of Transformers issues (model loading, attention backends, tensor parallelism, CI flakiness, and import-time regressions). None of the soft-linked pairs look like the same underlying bug/change, so I would not merge any of them as duplicates.", - "confidence": 0.46, - "canonical_issue_reason": "Issue 36683 is the most active open item with a concrete runtime failure and broad model-loading relevance; among this mixed set it is the best anchor issue, even though the cluster is not a true duplicate group.", + "summary": "This set is a loose collection of Transformers issue reports spanning unrelated regressions (imports, tokenizer/model loading, flash-attention, tensor parallelism, save/load, CI, and Python-version compatibility). None of the soft pairs look like the same underlying defect, so there is no clear canonical duplicate anchor.", + "confidence": 0.96, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue 36683 is the strongest representative issue here: active discussion, clear failure mode, and enough specificity to be actionable without being just a test or minor cleanup report.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:36683", - "right": "issue:44242", - "accept": false, - "reason": "Different bugs: Gemma3 config missing vocab_size vs MoE load-balancing loss handling." - }, - { - "left": "issue:44380", - "right": "issue:46132", - "accept": false, - "reason": "Both touch attention code, but one is GPT-2 scaling under SDPA/FlashAttention and the other changes AttentionInterface.register behavior." - }, - { - "left": "issue:36683", - "right": "issue:43881", - "accept": false, - "reason": "Both are load-time failures, but they affect different models and stem from different missing attributes/code paths." - }, - { - "left": "issue:43906", - "right": "issue:45753", - "accept": false, - "reason": "An isolated reproduction of an earlier issue is not the same as a specific Qwen3.5 image-loop regression." - }, - { - "left": "issue:43299", - "right": "issue:45753", - "accept": false, - "reason": "Both mention Qwen3.5-era models, but one is MoE loading and the other is multimodal infinite-loop generation." - }, { "left": "issue:43352", "right": "issue:43976", "accept": false, - "reason": "Nemotron FlashAttention 2 support gating is unrelated to Python version compatibility." + "reason": "Both mention model/runtime breakage, but one is a FlashAttention support issue for NemotronH and the other is a Python-version compatibility problem for Transformers 5.1.0." }, { "left": "issue:43992", "right": "issue:45310", "accept": false, - "reason": "UMT5 missing embed_tokens weights and Qwen3.5 MoE from_pretrained errors are different model-specific load failures." + "reason": "Different failures: UMT5Encoder missing embed_tokens.weight vs Qwen3.5 MoE from_pretrained error." }, { "left": "issue:43441", "right": "issue:44380", "accept": false, - "reason": "Both involve attention, but one is unsupported FlashAttention on Ministral-3 and the other is ignored attention scaling on GPT-2 backends." + "reason": "Both involve attention backends, but one is Ministral-3 failing with FlashAttention while the other is GPT2 ignoring attention scaling config under SDPA/FlashAttention; different code paths." }, { "left": "issue:43352", "right": "issue:45478", "accept": false, - "reason": "Same broad topic family, but 45478 is a different Qwen3.5 MoE loading regression, not the Nemotron FlashAttention issue." + "reason": "Same broad FlashAttention theme, but distinct model families and reported symptoms; not the same concrete bug." }, { "left": "issue:44355", "right": "issue:44855", "accept": false, - "reason": "Both are import errors, but one is about compiled Python files and the other is a torch.jit parsing problem in DebertaV2Model." + "reason": "Both are import-time failures, but one is about compiled Python files and the other is a Python 3.13 torch.jit.script parsing/IndentationError issue." }, { "left": "issue:43867", "right": "issue:46202", "accept": false, - "reason": "State-dict ordering load failures and Gemma pipeline roundtrip failures are unrelated." + "reason": "Both are model loading/roundtrip-related, but the specific failures differ: sorted state_dict loading vs Gemma pipeline roundtrip on the \u2581 token." }, { - "left": "issue:43867", - "right": "issue:45375", + "left": "issue:44380", + "right": "issue:45479", "accept": false, - "reason": "Generic state_dict loading errors are not the same as a missing Qwen3_5MoeVisionConfig field." + "reason": "Attention/ML terms overlap, but one is backend-specific config being ignored and the other is a sequence-classification loss bug." }, { - "left": "issue:44380", - "right": "issue:45479", + "left": "issue:43867", + "right": "issue:45375", "accept": false, - "reason": "Attention backend config bug vs degenerate zero-loss classification behavior." + "reason": "A load-model issue vs a missing vision config field; unrelated root causes." }, { "left": "issue:43352", "right": "issue:44380", "accept": false, - "reason": "Both are attention-related, but they affect different models and different backend behavior." + "reason": "Both reference FlashAttention, but one is lack of support and the other is ignored scaling config; not mergeable as one fix." }, { "left": "issue:36683", "right": "issue:44117", "accept": false, - "reason": "Different missing-attribute regressions in different parts of the loading stack." + "reason": "Different objects and failures: Gemma3Config missing vocab_size vs tokenizer mapping sometimes returning None." }, { "left": "issue:43618", "right": "issue:44380", "accept": false, - "reason": "CLIPOutput attentions regression is not the same bug as GPT-2 attention scaling being ignored." + "reason": "Different model API regressions: CLIPOutput attentions missing vs GPT2 attention scaling ignored under certain backends." }, { "left": "issue:41669", "right": "issue:44246", "accept": false, - "reason": "Import-time slowdown from wildcard imports and intermittent import slowness are related only at a very high level, not the same defect." + "reason": "Import-* cleanup/performance discussion vs intermittent slow import of transformers; related only at a very high level." }, { "left": "issue:43097", "right": "issue:43352", "accept": false, - "reason": "Tie-embeddings API removal and FlashAttention support are unrelated regressions." + "reason": "One is an API removal/change notice, the other is a model FlashAttention support bug; not the same defect." }, { "left": "issue:42890", "right": "issue:45561", "accept": false, - "reason": "Both are test flakiness, but one is missing set_seed and the other is an xdist file-race issue." + "reason": "A flaky integration test missing set_seed() vs a pytest-xdist race in captured_info.txt; both test-related but different issues." }, { "left": "issue:43064", "right": "issue:44492", "accept": false, - "reason": "Trainer optimizer-state corruption and a cache-strategy typo are unrelated." + "reason": "Training/FSDP optimizer-state corruption vs a typo in cache strategies; unrelated." }, { "left": "issue:43976", "right": "issue:44938", "accept": false, - "reason": "Both are Python compatibility reports, but for different versions and likely different root causes." + "reason": "Both are Python-version compatibility complaints, but they target different versions and likely different breakpoints." }, { "left": "issue:43867", "right": "issue:45357", "accept": false, - "reason": "Different model save/load regressions: state_dict ordering versus incorrect visual encoder keys." + "reason": "Load-time state_dict ordering problem vs save_pretrained writing incorrect visual encoder keys." + }, + { + "left": "issue:43992", + "right": "issue:45478", + "accept": false, + "reason": "Different models and failures; no shared code path beyond both being from_pretrained reports." }, { "left": "issue:36296", "right": "issue:43575", "accept": false, - "reason": "Tensor-parallel training bug and tp-related OOM when loading a large model are not the same underlying issue." + "reason": "Tensor-parallel training bug vs OOM when loading a large Qwen model with tp; related area, different underlying problems." }, { "left": "issue:43064", "right": "issue:44655", "accept": false, - "reason": "Wrong optimizer states during training is unrelated to saving Pipeline objects." + "reason": "Wrong optimizer state under v5/FSDP2/PEFT vs inability to save Pipeline objects; both serialization/training-adjacent but distinct." }, { "left": "issue:28282", "right": "issue:33453", "accept": false, - "reason": "Missing PyTorch ImportError and tokenizer-loading regression are unrelated." + "reason": "ImportError about missing PyTorch vs tokenizer loading regression; different subsystems and symptoms." }, { "left": "issue:43618", "right": "issue:46285", "accept": false, - "reason": "Both are CLIP-related, but one is missing attentions output and the other is structural divergence between two CLIP text models." + "reason": "Both concern CLIP model structure, but one is output attentions and the other is inconsistent text model structure; not the same bug." }, { "left": "issue:43618", "right": "issue:44857", "accept": false, - "reason": "Attention-output regression and AMP/CUDA loss crash are different code paths and symptoms." + "reason": "CLIPOutput attentions regression vs LwDetrImageLoss crashing under float16 AMP; unrelated." }, { - "left": "issue:43618", - "right": "issue:44977", + "left": "issue:44655", + "right": "issue:44938", + "accept": false, + "reason": "Saving Pipeline objects vs Python 3.14 import/load failure; different failure modes." + }, + { + "left": "issue:43618", + "right": "issue:44977", "accept": false, - "reason": "CLIP attentions regression is unrelated to Qwen3.5 flash-attention generation failure." + "reason": "Different attention-related reports on different models; one is missing attentions in CLIPOutput, the other is Qwen3.5 generation failing with flash-attention." }, { "left": "issue:44360", "right": "issue:45440", "accept": false, - "reason": "DSA indexer implementation details are unrelated to DeepseekV3MoE diverging from the remote implementation." + "reason": "A DSA/indexer discussion vs DeepseekV3MoE implementation divergence; no shared bug." }, { "left": "issue:43502", "right": "issue:44062", "accept": false, - "reason": "local_files_only network leakage and AddedToken keyword conflicts are different bugs." + "reason": "Local-files-only still making API requests vs AddedToken constructor argument conflict; unrelated." }, { "left": "issue:42890", "right": "issue:44964", "accept": false, - "reason": "Seed-related CI flakiness is unrelated to a specific multimodal model load failure." + "reason": "Flaky test seeding issue vs inability to load a specific multimodal model on latest Transformers." }, { "left": "issue:42890", "right": "issue:44112", "accept": false, - "reason": "Missing set_seed in tests and stale device override behavior are different CI issues." + "reason": "Both are test failures, but one is missing set_seed coverage and the other is a stale device-override CI failure." }, { "left": "issue:42890", "right": "issue:45600", "accept": false, - "reason": "A flaky test seed issue is not the same as broken auto_mappings references causing CI failures." + "reason": "Test flakiness vs auto_mappings referencing removed configs; different root causes." }, { "left": "issue:44222", "right": "issue:45125", "accept": false, - "reason": "FP8 save_pretrained behavior and a missing tensor-parallel plan are separate model-export/parallelism bugs." + "reason": "FP8 save_pretrained bug for MoE vs missing _tp_plan for tensor parallelism; same model family, different functionality." + }, + { + "left": "issue:43576", + "right": "issue:46132", + "accept": false, + "reason": "Broken transformers env command vs AttentionInterface.register behavior change; unrelated command/API issues." + }, + { + "left": "issue:44355", + "right": "issue:44749", + "accept": false, + "reason": "Compiled-file import errors vs a performance regression after upgrading Transformers; different problems." + }, + { + "left": "issue:45125", + "right": "issue:45478", + "accept": false, + "reason": "Missing tensor-parallel plan vs Qwen3.5 MoE from_pretrained error; likely adjacent but not the same defect." + }, + { + "left": "issue:42548", + "right": "issue:44351", + "accept": false, + "reason": "Top-level import errors, but different missing symbols: PreTrainedModel vs HybridCache." } ] }, @@ -15172,13 +15252,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 48, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:30333", "issue:30990", @@ -15202,10 +15282,11 @@ "issue:43874", "issue:43901", "issue:43976", + "issue:44038", "issue:44222", "issue:44230", + "issue:44273", "issue:44315", - "issue:44351", "issue:44355", "issue:44360", "issue:44368", @@ -15213,27 +15294,22 @@ "issue:44556", "issue:44623", "issue:44655", - "issue:44749", "issue:44908", "issue:44938", "issue:44961", + "issue:45003", "issue:45071", "issue:45125", "issue:45230", "issue:45278", "issue:45375", "issue:45468", - "issue:45478", + "issue:45542", "issue:45656", "issue:45850", - "issue:46132", "issue:46209" ], "soft_pairs": [ - "issue:43576|issue:46132", - "issue:44355|issue:44749", - "issue:45125|issue:45478", - "issue:42548|issue:44351", "issue:44387|issue:44556", "issue:30333|issue:33453", "issue:45125|issue:45375", @@ -15243,18 +15319,18 @@ "issue:43502|issue:44222", "issue:42831|issue:43874", "issue:43010|issue:45850", - "issue:43502|issue:44961", "issue:43010|issue:44368", + "issue:43502|issue:44961", "issue:30333|issue:36296", - "issue:43827|issue:44961", "issue:44355|issue:45071", + "issue:43827|issue:44961", "issue:44355|issue:44655", "issue:44315|issue:44360", "issue:33453|issue:34689", "issue:43299|issue:44556", "issue:42757|issue:43576", - "issue:43576|issue:45071", "issue:33453|issue:36296", + "issue:43576|issue:45071", "issue:43010|issue:45656", "issue:43576|issue:43824", "issue:31795|issue:38617", @@ -15264,267 +15340,282 @@ "issue:43576|issue:44938", "issue:44908|issue:44961", "issue:44315|issue:46209", - "issue:41628|issue:42548", "issue:36296|issue:43901", + "issue:41628|issue:42548", "issue:36296|issue:43526", - "issue:44623|issue:45230" + "issue:44623|issue:45230", + "issue:44273|issue:45542", + "issue:44038|issue:45125", + "issue:44273|issue:45003" ], - "prepared_review_unit_hash": "e2b7280b697fa473078d4a3a5b5dd904c396eb613d827832bb7c0eac9f13d811", + "prepared_review_unit_hash": "4d11498e82557383630a78f9e85bf2f6e27316f3038cfbfdc497b33deff10384", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 38, - "serialized_chars": 38487, - "estimated_input_tokens": 9622, - "estimated_eval_tokens": 19500 + "soft_pair_count": 37, + "serialized_chars": 38083, + "estimated_input_tokens": 9521, + "estimated_eval_tokens": 19298 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These issues are mostly a grab bag of unrelated Transformers regressions, import errors, docs bugs, and model-specific failures. The soft pairs are only superficially similar by wording; none look like the same concrete defect. The broadest representative issue is the v5 regression umbrella.", - "confidence": 0.63, - "canonical_issue_reason": "issue:45850 is the broadest umbrella here: it captures the general v5 breakage theme that many of the other regression-style reports superficially resemble.", + "summary": "This cluster is mostly a grab-bag of distinct transformers issues: import/load regressions, FP8/MoE bugs, tensor-parallel problems, docs mismatches, and unrelated runtime errors. None of the soft pairs look like the same concrete underlying bug or change.", + "confidence": 0.66, + "canonical_issue_reason": "Issue 45278 is the broadest umbrella-style report here (v4.57 -> v5.5 import failures), so it is the closest canonical issue among otherwise heterogeneous reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45850 best represents the cluster because it is the most general, current, and central user-facing breakage report among otherwise unrelated issues.", + "best_issue_reason": "Issue 45278 is the best representative issue because it is the most general v5 regression report and overlaps conceptually with several other loading/import complaints, even though the cluster is not a true duplicate set.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43576", - "right": "issue:46132", - "accept": false, - "reason": "Different bugs: broken `transformers env` CLI vs `AttentionInterface.register` behavior change." - }, - { - "left": "issue:44355", - "right": "issue:44749", - "accept": false, - "reason": "Unrelated symptoms: compiled Python file errors vs a data-filtering slowdown after upgrade." - }, - { - "left": "issue:45125", - "right": "issue:45478", - "accept": false, - "reason": "Both are Qwen3.5 Moe-related, but one is missing tensor-parallel plan data and the other is a `from_pretrained` loading failure; not the same code path." - }, - { - "left": "issue:42548", - "right": "issue:44351", - "accept": false, - "reason": "Different missing exports: `PreTrainedModel` vs `HybridCache`." - }, { "left": "issue:44387", "right": "issue:44556", "accept": false, - "reason": "Memory/OOM regression under int4 quantization is not the same as failing to reload a checkpoint after upgrading." + "reason": "Both are v5 regressions, but one is an int4/OOM memory issue and the other is a checkpoint reload/loading failure; different bugs and code paths." }, { "left": "issue:30333", "right": "issue:33453", "accept": false, - "reason": "MLFlow reporting on failed training jobs is unrelated to tokenizer loading regression." + "reason": "MLFlow failed-job reporting and tokenizer loading regression are unrelated." }, { "left": "issue:45125", "right": "issue:45375", "accept": false, - "reason": "Tensor-parallel plan omission is separate from `deepstack_visual_indexes` being dropped by strict config handling." + "reason": "Both involve Qwen3_5Moe, but one is tensor-parallel plan support and the other is a config field being dropped; not the same bug." }, { "left": "issue:43010", "right": "issue:45278", "accept": false, - "reason": "`@torch.no_grad` cache update issue is unrelated to broad import failures after upgrade." + "reason": "No-grad decoration for cache/layer updates is unrelated to broad import errors after upgrading to v5.5.0." }, { "left": "issue:30990", "right": "issue:31795", "accept": false, - "reason": "A loading hang in Sentence Transformers is not the same as docs confusion around `input_ids` and `past_key_values`." + "reason": "Sentence Transformers hanging on load is not the same as a documentation issue about model.forward arguments." }, { "left": "issue:43723", "right": "issue:44556", "accept": false, - "reason": "Both mention loading, but one is tokenizer loading in v5 and the other is checkpoint reloading after upgrade; different failures." + "reason": "Both are loading problems, but one is AutoTokenizer loading in v5 and the other is a checkpoint reload regression; too different to merge." }, { "left": "issue:43502", "right": "issue:44222", "accept": false, - "reason": "Unexpected hub API access with `local_files_only=True` is unrelated to FP8 `save_pretrained` behavior." + "reason": "Local-files-only network leakage and FP8 save_pretrained issues are unrelated." }, { "left": "issue:42831", "right": "issue:43874", "accept": false, - "reason": "FP8 accuracy regression and a missing image-patch helper causing an AttributeError are different bugs." + "reason": "An FP8 accuracy issue is not the same as a missing image-processor method causing an AttributeError." }, { "left": "issue:43010", "right": "issue:45850", "accept": false, - "reason": "A specific cache-update decorator bug is not the same as a broad claim that v5 breaks many models." + "reason": "A specific no-grad bug is not the same as a broad complaint that v5 breaks many models." }, { - "left": "issue:43502", - "right": "issue:44961", + "left": "issue:43010", + "right": "issue:44368", "accept": false, - "reason": "`local_files_only` network access bug is unrelated to the placeholder issue titled `racoon`." + "reason": "Cache update gradients and a tie_word_embeddings warning are unrelated." }, { - "left": "issue:43010", - "right": "issue:44368", + "left": "issue:43502", + "right": "issue:44961", "accept": false, - "reason": "Cache update no-grad behavior has nothing to do with a warning about `tie_word_embeddings`." + "reason": "local_files_only API requests and an unrelated 'racoon' issue do not match." }, { "left": "issue:30333", "right": "issue:36296", "accept": false, - "reason": "MLFlow status reporting and tensor-parallel training behavior are distinct problems." + "reason": "MLFlow reporting and tensor-parallel training are distinct issues." }, { - "left": "issue:43827", - "right": "issue:44961", + "left": "issue:44355", + "right": "issue:45071", "accept": false, - "reason": "Docs still referencing `pipeline()` after removals is unrelated to the placeholder `racoon` issue." + "reason": "Compiled Python file errors and PretrainedConfig type checking are different regressions." }, { - "left": "issue:44355", - "right": "issue:45071", + "left": "issue:43827", + "right": "issue:44961", "accept": false, - "reason": "Compiled-file errors and `PretrainedConfig` type-checking breakage are different failure modes." + "reason": "Docs still referencing pipeline() and an unrelated 'racoon' issue are not duplicates." }, { "left": "issue:44355", "right": "issue:44655", "accept": false, - "reason": "Compiled Python file handling is unrelated to saving Pipeline objects." + "reason": "Compiled-file errors and saving Pipeline objects are separate save/load problems." }, { "left": "issue:44315", "right": "issue:44360", "accept": false, - "reason": "Liger Kernel not applied with `model_init` is unrelated to a missing ReLU in the DSA indexer." + "reason": "Liger Kernel application and a DSA indexer ReLU complaint are unrelated." }, { "left": "issue:33453", "right": "issue:34689", "accept": false, - "reason": "Tokenizer-loading regression and Llama 3.2 Vision model-loading regression are different model/code-path issues." + "reason": "Tokenizer loading regression and Llama 3.2 model-loading failure are both load-related but not the same underlying bug." }, { "left": "issue:43299", "right": "issue:44556", "accept": false, - "reason": "A Qwen3VL MoE loading breakage is not the same as a general checkpoint reload regression after upgrade." + "reason": "Qwen3VL MoE loading breakage and generic checkpoint reload breakage are different failures." }, { "left": "issue:42757", "right": "issue:43576", "accept": false, - "reason": "Missing `is_offline_mode` export and broken `transformers env` CLI are unrelated." + "reason": "A missing huggingface_hub import and the transformers env command bug are unrelated." }, { - "left": "issue:43576", - "right": "issue:45071", + "left": "issue:33453", + "right": "issue:36296", "accept": false, - "reason": "Broken environment command and `PretrainedConfig` type-checking regression do not describe the same defect." + "reason": "Tokenizer loading regression and tensor-parallel training bug do not share the same code path." }, { - "left": "issue:33453", - "right": "issue:36296", + "left": "issue:43576", + "right": "issue:45071", "accept": false, - "reason": "Tokenizer loading regression and tensor-parallel training bug are different issues." + "reason": "Broken env CLI and PretrainedConfig type checking are unrelated." }, { "left": "issue:43010", "right": "issue:45656", "accept": false, - "reason": "Cache update no-grad behavior is unrelated to DeepSpeed calling optimizer.step twice." + "reason": "Cache/layer update no-grad handling is not the same as double optimizer stepping under deepspeed." }, { "left": "issue:43576", "right": "issue:43824", "accept": false, - "reason": "CLI `transformers env` breakage is unrelated to missing Qwen2.5-VL import symbols." + "reason": "Broken env CLI and a missing Qwen2_5_VLForConditionalGeneration import are different issues." }, { "left": "issue:31795", "right": "issue:38617", "accept": false, - "reason": "Documentation confusion and an `ImportError` for `layer_type_validation` are not the same bug." + "reason": "Documentation confusion in model.forward and an ImportError for layer_type_validation are unrelated." }, { "left": "issue:44961", "right": "issue:45468", "accept": false, - "reason": "Placeholder issue `racoon` has no clear relation to Gemma-4 audio positional encoding." + "reason": "An empty/placeholder issue and a Gemma-4 audio positional-encoding bug are not duplicates." }, { "left": "issue:44222", "right": "issue:44230", "accept": false, - "reason": "Both touch FP8/MoE, but one is a save/load bug and the other is an inference-support request; not mergeable as one fix." + "reason": "Both mention FP8/MoE, but one is save_pretrained behavior and the other is inference support; different concrete bugs." }, { "left": "issue:43010", "right": "issue:43976", "accept": false, - "reason": "Cache update decorator behavior is unrelated to the Python version compatibility report." + "reason": "Cache update no-grad behavior is unrelated to Python 3.9/3.10 compatibility." }, { - "left": "issue:43576", - "right": "issue:44938", + "left": "issue:44273", + "right": "issue:45542", "accept": false, - "reason": "Broken CLI command and Python 3.14 load failure are separate problems." + "reason": "Lazy loading problems and a TensorBoard-without-TensorFlow backend error are not the same bug." }, { - "left": "issue:44908", - "right": "issue:44961", + "left": "issue:44038", + "right": "issue:45125", "accept": false, - "reason": "Scheduler kwargs bug and placeholder `racoon` issue do not match." + "reason": "Both are Qwen3-Moe-related, but one is a broad v5/Qwen3-VL-Moe bug and the other is a missing tensor-parallel plan; not enough to call them duplicates." }, { - "left": "issue:44315", - "right": "issue:46209", + "left": "issue:44273", + "right": "issue:45003", "accept": false, - "reason": "Liger Kernel application with `model_init` is unrelated to FP8 inference failing on B200 GPUs." + "reason": "Lazy loading issues and sys.modules access in modeling_utils may be related conceptually, but they are not clearly the same concrete bug." }, { - "left": "issue:41628", - "right": "issue:42548", + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "processor.save_pretrained missing files and a generic bug report are not duplicates." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Same as above: possible relation to import/lazy-loading, but not a clear same-bug match." + }, + { + "left": "issue:44315", + "right": "issue:46209", "accept": false, - "reason": "Different import errors for different symbols from `transformers`." + "reason": "Liger Kernel application and FP8 inference on B200 are separate issues." }, { "left": "issue:36296", "right": "issue:43901", "accept": false, - "reason": "Tensor-parallel training bug and pipeline docs mismatch are unrelated." + "reason": "Tensor-parallel training and docs about return_all_scores are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:42548", + "accept": false, + "reason": "Import errors for AutoImageProcessor and PreTrainedModel are both namespace issues, but not the same symbol or failure mode." }, { "left": "issue:36296", "right": "issue:43526", "accept": false, - "reason": "Tensor-parallel training bug is unrelated to `reduce_labels` returning the wrong shape/value in BeitImageProcessorFast." + "reason": "Tensor-parallel training and BeitImageProcessorFast reduce_labels behavior are unrelated." }, { "left": "issue:44623", "right": "issue:45230", "accept": false, - "reason": "Missing files from `processor.save_pretrained(...)` is specific; the other item is an empty generic bug report." + "reason": "Processor save_pretrained file omissions and a generic bug report are not the same issue." + }, + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Lazy loading and TensorBoard/TensorFlow backend configuration are different problems." + }, + { + "left": "issue:44038", + "right": "issue:45125", + "accept": false, + "reason": "Qwen3-VL-Moe breakage and missing _tp_plan are related only at a high level; not the same underlying defect." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Lazy loading and unsafe sys.modules access are too loosely related to merge as duplicates." } ] }, @@ -15540,13 +15631,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 49, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:31795", "issue:33453", @@ -15562,7 +15653,6 @@ "issue:43827", "issue:43874", "issue:43976", - "issue:44038", "issue:44077", "issue:44230", "issue:44246", @@ -15571,25 +15661,26 @@ "issue:44297", "issue:44315", "issue:44351", + "issue:44360", "issue:44373", "issue:44393", "issue:44492", "issue:44556", + "issue:44655", "issue:44704", "issue:44829", "issue:44861", "issue:44908", "issue:44961", - "issue:45003", "issue:45071", "issue:45092", - "issue:45125", "issue:45230", "issue:45278", + "issue:45310", "issue:45341", "issue:45362", + "issue:45468", "issue:45478", - "issue:45542", "issue:45753", "issue:45850", "issue:46095", @@ -15598,22 +15689,19 @@ "issue:46218" ], "soft_pairs": [ - "issue:44273|issue:45542", - "issue:44038|issue:45125", - "issue:44273|issue:45003", "issue:44393|issue:45071", "issue:44230|issue:46095", "issue:44373|issue:44961", - "issue:43576|issue:44556", "issue:44393|issue:45753", + "issue:43576|issue:44556", "issue:41669|issue:43673", "issue:43525|issue:45850", "issue:44556|issue:45092", "issue:43572|issue:45850", "issue:45362|issue:45850", "issue:41669|issue:46218", - "issue:44246|issue:44273", "issue:42890|issue:44908", + "issue:44246|issue:44273", "issue:43408|issue:44315", "issue:44704|issue:44908", "issue:43352|issue:44393", @@ -15636,277 +15724,268 @@ "issue:41669|issue:44492", "issue:44315|issue:44829", "issue:42371|issue:44492", - "issue:43576|issue:44273" + "issue:43576|issue:44273", + "issue:44655|issue:44861", + "issue:44360|issue:45468", + "issue:45230|issue:45310", + "issue:44273|issue:44908", + "issue:43576|issue:46218", + "issue:43827|issue:44492" ], - "prepared_review_unit_hash": "44c8d08d29d64678ff928112b2f1aab3bdb582492d4a4b4add479835883b1be8", + "prepared_review_unit_hash": "ed8235708529f42e87927a2891745a401a0bdea062e7d56ad15d4d95a17105b0", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 39, - "serialized_chars": 38785, - "estimated_input_tokens": 9697, - "estimated_eval_tokens": 19650 + "soft_pair_count": 42, + "serialized_chars": 39625, + "estimated_input_tokens": 9907, + "estimated_eval_tokens": 20070 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a noisy cluster of unrelated issues. The only coherent theme is import/lazy-loading/v5 import regressions, but none of the soft pairs are close enough to treat as duplicates or the same fix.", - "confidence": 0.86, - "canonical_issue_reason": "issue:44273 is the broadest umbrella for the import/lazy-loading regression theme and best matches the shared symptom family across the relevant issues.", + "summary": "Mostly closed/open issue reports about Transformers v5 regressions: import/lazy-loading slowdowns, model-loading/config compatibility breakages, and API changes. No PRs are present in the cluster.", + "confidence": 0.78, + "canonical_issue_reason": "issue:45850 is the broadest umbrella report in the set: it explicitly frames a wide v5 regression affecting many models, making it the best single issue to anchor the cluster even though it is vague.", "canonical_pr_reason": null, - "best_issue_reason": "issue:41669 is the most concrete root-cause report in the cluster: it identifies import * usage as the source of the import-time slowdown and is the most actionable issue to anchor the duplicate set.", + "best_issue_reason": "issue:45850 is the best global representative because it is broad, current, and covers the same general family of v5 breakage complaints seen across the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44273", - "right": "issue:45542", - "accept": false, - "reason": "Both are runtime/import-related at a high level, but one is about lazy loading and the other is a TensorFlow backend warning path; different bugs and fixes." - }, - { - "left": "issue:44038", - "right": "issue:45125", - "accept": false, - "reason": "Different model families and different failures: Qwen3-VL-Moe inference bug versus missing tensor-parallel plan in Qwen3.5 MoE." - }, - { - "left": "issue:44273", - "right": "issue:45003", - "accept": false, - "reason": "One is a broad lazy-loading complaint; the other is a specific sys.modules access bug in modeling_utils. Related area, not the same underlying defect." - }, { "left": "issue:44393", "right": "issue:45071", "accept": false, - "reason": "Completely different surfaces: Qwen3-VL 2D bounding-box output bug versus PretrainedConfig type-checking breakage." + "reason": "Different problems: Qwen3-VL multimodal output failure vs PretrainedConfig type checking regression." }, { "left": "issue:44230", "right": "issue:46095", "accept": false, - "reason": "Different features and code paths: FP8 inference support for Qwen models versus a DeepSeek weight-conversion question." + "reason": "Unrelated model/features: fp8 inference support for Qwen3-VL/Qwen3.5 MoE vs DeepSeekV4 weight conversion question." }, { "left": "issue:44373", "right": "issue:44961", "accept": false, - "reason": "A docstring typo about position_ids is unrelated to the placeholder 'racoon' issue." + "reason": "Docstring bug vs a completely unrelated issue titled 'racoon'." }, { - "left": "issue:43576", - "right": "issue:44556", + "left": "issue:44393", + "right": "issue:45753", "accept": false, - "reason": "Both mention v5, but one is the transformers env command and the other is checkpoint reload compatibility; not the same bug." + "reason": "Both are Qwen-family vision issues, but the failure modes differ materially: bbox hallucination/error vs infinite loop on a specific image." }, { - "left": "issue:44393", - "right": "issue:45753", + "left": "issue:43576", + "right": "issue:44556", "accept": false, - "reason": "Both involve Qwen3-VL/Qwen3.5, but the symptoms differ: 2D box output failure versus an infinite-loop case on a specific image." + "reason": "Both are v5 regressions, but one is the `transformers env` command and the other is checkpoint reload compatibility; not the same code path." }, { "left": "issue:41669", "right": "issue:43673", "accept": false, - "reason": "Import-performance regression versus missing GenerationMixin cache during chunked prefill; different root causes and fixes." + "reason": "Import-star cleanup/performance issue vs missing generation cache during chunked prefill; different subsystems and fixes." }, { "left": "issue:43525", "right": "issue:45850", "accept": false, - "reason": "Specific Llama4Config pad_token_id error versus a broad complaint that v5 breaks many models." + "reason": "Specific `pad_token_id`/config attribute error vs a broad complaint about many models breaking in v5." }, { "left": "issue:44556", "right": "issue:45092", "accept": false, - "reason": "Both are upgrade-compatibility problems, but they affect different model-loading paths and are not clearly the same defect." + "reason": "Both concern v5 checkpoint loading, but one is generic reload failure and the other is remote-code/meta-init incompatibility; too different to treat as duplicates." }, { "left": "issue:43572", "right": "issue:45850", "accept": false, - "reason": "StableLmConfig pad_token_idx regression is a specific config issue, not the same as the broad v5 breakage complaint." + "reason": "Specific StableLmConfig `pad_token_idx` regression vs a broad v5 breakage umbrella issue." }, { "left": "issue:45362", "right": "issue:45850", "accept": false, - "reason": "A Qwen3.5 chat crash is a model-specific failure, not the same as the generic 'v5 breaks too many models' report." - }, - { - "left": "issue:41669", - "right": "issue:46218", - "accept": false, - "reason": "Import * slowdown and a missing Qwen3_5ForCausalLM import error may both involve imports, but the underlying problems are different." - }, - { - "left": "issue:44246", - "right": "issue:44273", - "accept": false, - "reason": "Both concern import behavior, but one reports intermittent slow imports while the other claims lazy loading is broken; not enough evidence of the same bug." + "reason": "Qwen3.5 chat crash is a concrete model bug, not the same as a general v5 breakage report." }, { - "left": "issue:42890", - "right": "issue:44908", - "accept": false, - "reason": "Test seed omissions in SAM HQ tests have nothing to do with the inverse_sqrt scheduler ignoring kwargs." - }, - { - "left": "issue:43408", - "right": "issue:44315", + "left": "issue:45278", + "right": "issue:45850", "accept": false, - "reason": "Model type mismatch warning for SAM3 versus Liger Kernel not applied in model_init are unrelated." + "reason": "Broad overlap in release complaints, but one is about import errors after upgrade and the other is a generic 'too many models' regression; not a single concrete bug." }, { - "left": "issue:44704", + "left": "issue:44273", "right": "issue:44908", "accept": false, - "reason": "AutoProcessor cached_file kwargs forwarding and scheduler kwargs handling are separate code paths." + "reason": "Lazy loading/import behavior vs inverse_sqrt scheduler kwargs; unrelated code paths." }, { - "left": "issue:43352", - "right": "issue:44393", + "left": "issue:44297", + "right": "issue:45230", "accept": false, - "reason": "Flash Attention 2 support error in Nemotron is unrelated to Qwen3-VL 2D bounding-box output behavior." + "reason": "Tokenizer save_pretrained metadata mismatch vs an empty generic bug report; no evidence they are the same defect." }, { - "left": "issue:44297", - "right": "issue:45230", + "left": "issue:46143", + "right": "issue:46144", "accept": false, - "reason": "Tokenizer save_pretrained metadata mismatch is unrelated to a generic bug report placeholder." + "reason": "Both involve RoFormer, but one is kwargs propagation and the other is attention-interface usage; related area, not the same bug." }, { - "left": "issue:43976", - "right": "issue:44393", + "left": "issue:43827", + "right": "issue:44492", "accept": false, - "reason": "Python version compatibility regression is unrelated to the Qwen3-VL bounding-box bug." + "reason": "Documentation reference to removed pipelines vs a typo in cache strategies; unrelated." }, { "left": "issue:44295", "right": "issue:45230", "accept": false, - "reason": "Position_ids buffer access error is unrelated to a generic bug report placeholder." + "reason": "position_ids buffer access error vs generic bug report; no shared concrete failure." }, { - "left": "issue:43576", + "left": "issue:44655", "right": "issue:44908", "accept": false, - "reason": "Broken env command and inverse_sqrt scheduler kwargs are separate issues." - }, - { - "left": "issue:44908", - "right": "issue:45071", - "accept": false, - "reason": "Scheduler keyword handling and PretrainedConfig type-checking are unrelated." + "reason": "Pipeline save/load issue vs scheduler kwargs handling; unrelated." }, { "left": "issue:41669", - "right": "issue:44351", - "accept": false, - "reason": "Import * cleanup and missing HybridCache export may both touch imports, but they are different missing-symbol/import surfaces." - }, - { - "left": "issue:41669", - "right": "issue:45341", + "right": "issue:46218", "accept": false, - "reason": "Import-time slowdown is unrelated to a small testing_utils bug." + "reason": "Import performance cleanup vs an import error for a specific Qwen3_5 symbol; same broad import area, but not the same bug." }, { - "left": "issue:45278", - "right": "issue:45850", + "left": "issue:44360", + "right": "issue:45468", "accept": false, - "reason": "General import errors after upgrade and a broad claim that v5 breaks many models are too vague to be the same underlying issue." + "reason": "DSA indexer/ReLU discussion vs Gemma4Audio positional encoding bug; different models and failures." }, { - "left": "issue:43673", - "right": "issue:44492", + "left": "issue:45230", + "right": "issue:45310", "accept": false, - "reason": "Chunked-prefill cache regression and a typo in cache strategies are unrelated." + "reason": "A generic bug report vs a specific Qwen3.5 MoE from_pretrained error; no concrete match." }, { - "left": "issue:46143", - "right": "issue:46144", - "accept": false, - "reason": "Missing kwargs passthrough and attention-interface noncompliance in RoFormer are related to the same model family but not the same concrete defect." + "left": "issue:44246", + "right": "issue:44273", + "accept": true, + "reason": "Both point at the same underlying import/lazy-loading regression: import transformers becoming slow and lazy loading not functioning properly." }, { "left": "issue:43874", "right": "issue:45850", "accept": false, - "reason": "A specific GLM46V image-processor method bug is not the same as the broad v5 breakage complaint." - }, - { - "left": "issue:44393", - "right": "issue:45478", - "accept": false, - "reason": "Different Qwen-related failures with different symptoms and code paths." + "reason": "Specific GLM46V multimodal token counting failure vs a broad v5 breakage umbrella issue." }, { "left": "issue:44077", "right": "issue:44861", "accept": false, - "reason": "Optional post_init validation and tied-weight-key AttributeError are distinct issues." + "reason": "Different defects in different utilities: patchtsmixer post_init handling vs tied-weight key attribute crash." }, { "left": "issue:43976", "right": "issue:46218", "accept": false, - "reason": "Python version incompatibility and a missing Qwen3_5 import are separate failures." + "reason": "Python-version compatibility complaint vs a specific missing import symbol; not the same issue." }, { "left": "issue:43827", "right": "issue:44908", "accept": false, - "reason": "Documentation still referencing pipeline() is unrelated to scheduler kwargs handling." + "reason": "Docs referencing removed pipeline APIs vs scheduler kwargs handling; unrelated." }, { "left": "issue:31795", "right": "issue:33453", "accept": false, - "reason": "Model.forward documentation confusion and tokenizer loading regression are unrelated." + "reason": "Documentation confusion vs tokenizer loading regression; not the same underlying bug." }, { "left": "issue:41669", "right": "issue:44492", "accept": false, - "reason": "Import * performance regression and a cache-strategy typo are not the same bug." + "reason": "Import-star performance regression vs a typo in cache-strategy docs; unrelated." }, { "left": "issue:44315", "right": "issue:44829", "accept": false, - "reason": "Liger Kernel not applied in model_init and flash_attention_3 causing degenerate training are different changes/issues." + "reason": "Liger Kernel application bug vs degenerate training with flash_attention_3; different behavior and fixes." }, { "left": "issue:42371", "right": "issue:44492", "accept": false, - "reason": "TF32 API guidance and a cache-strategy typo are unrelated." + "reason": "TF32 settings migration issue vs cache-strategy typo; unrelated." }, { "left": "issue:43576", "right": "issue:44273", "accept": false, - "reason": "Broken transformers env command and lazy loading issues overlap on version context only; they do not look like the same concrete bug." + "reason": "Broken `transformers env` command vs lazy loading not working properly; both v5 issues, but different code paths." + }, + { + "left": "issue:44655", + "right": "issue:44861", + "accept": false, + "reason": "Saving Pipeline objects vs `_tied_weight_keys` attribute crash; unrelated." + }, + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "Different problem domains and no shared code path." + }, + { + "left": "issue:45230", + "right": "issue:45310", + "accept": false, + "reason": "Generic report vs specific Qwen3.5 MoE loading failure; not duplicates." + }, + { + "left": "issue:44273", + "right": "issue:44908", + "accept": false, + "reason": "Lazy loading regression vs scheduler kwargs bug; unrelated." + }, + { + "left": "issue:43576", + "right": "issue:46218", + "accept": false, + "reason": "Broken env command vs missing Qwen3_5 import; not the same issue." + }, + { + "left": "issue:43827", + "right": "issue:44492", + "accept": false, + "reason": "Docs about removed pipelines vs cache-strategy typo; unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: the cluster is dominated by closed/open Transformers v5 regression issues, and issue:45850 is a reasonable broad umbrella representative. The soft-edge verdicts are conservative overall, with one plausible accept on the import/lazy-loading pair and the rest appropriately rejected based on distinct titles/code paths. No PRs are present." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -15915,13 +15994,13 @@ "component_index": 1, "component_count": 18, "review_unit_index": 50, - "review_unit_count": 51, + "review_unit_count": 50, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", + "cluster_id": "cluster-10000-441", "nodes": [ "issue:30333", "issue:31515", @@ -15929,39 +16008,35 @@ "issue:33453", "issue:34689", "issue:36296", + "issue:37934", "issue:38617", "issue:41669", "issue:42371", "issue:42913", "issue:42936", "issue:43066", + "issue:43240", "issue:43298", "issue:43352", "issue:43502", "issue:43519", "issue:43525", "issue:43572", - "issue:43576", "issue:43606", "issue:43668", "issue:43704", "issue:43717", - "issue:43827", + "issue:43813", "issue:44031", "issue:44230", "issue:44261", "issue:44263", - "issue:44273", "issue:44351", - "issue:44360", "issue:44485", - "issue:44492", "issue:44623", - "issue:44655", "issue:44861", "issue:44908", "issue:44998", - "issue:45230", "issue:45290", "issue:45310", "issue:45468", @@ -15973,12 +16048,6 @@ "issue:46218" ], "soft_pairs": [ - "issue:44360|issue:45468", - "issue:44655|issue:44861", - "issue:45230|issue:45310", - "issue:44273|issue:44908", - "issue:43576|issue:46218", - "issue:43827|issue:44492", "issue:46018|issue:46172", "issue:43519|issue:44485", "issue:43519|issue:44908", @@ -16016,418 +16085,327 @@ "issue:43298|issue:44230", "issue:43717|issue:46144", "issue:42936|issue:46018", - "issue:43668|issue:44031" + "issue:43668|issue:44031", + "issue:42936|issue:43240", + "issue:43240|issue:46144", + "issue:42936|issue:44031", + "issue:42936|issue:46144", + "issue:42936|issue:43813", + "issue:43240|issue:43668", + "issue:43668|issue:43813", + "issue:37934|issue:43668", + "issue:42936|issue:43668" ], - "prepared_review_unit_hash": "614ce861ade4edddad97d1c3049cda05ce228da564bbd3fc7eafdf10e05eb78c", + "prepared_review_unit_hash": "04ff93f98507054db59ea683e1f2a01d441f814f00ef1ec4da933e5355fd3870", "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 44, - "serialized_chars": 39956, - "estimated_input_tokens": 9989, - "estimated_eval_tokens": 20234 + "node_count": 44, + "item_count": 44, + "soft_pair_count": 47, + "serialized_chars": 38701, + "estimated_input_tokens": 9676, + "estimated_eval_tokens": 19608 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: most items are separate regressions, API changes, docs issues, or model-specific bugs that only share a broad Transformers/v5 similarity. I found no soft-edge pair that clearly describes the same underlying defect/change, so all candidates should be rejected.", - "confidence": 0.91, - "canonical_issue_reason": "issue:45850 is the broadest representative of the cluster\u2019s recurring v5 breakage theme, even though it is more of an umbrella complaint than a precise duplicate target.", + "summary": "This is a loose Transformers v5 regression/loading/tokenizer cluster with many superficially similar but distinct issues. I found no true duplicate soft edges; the candidates mostly share only broad subsystem or version-regression wording. No PRs are present.", + "confidence": 0.89, + "canonical_issue_reason": "issue:45850 is the broadest umbrella report for the v5 breakage/regression theme and can serve as the cluster anchor better than the more specific failures.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45850 best matches the overall cluster topic because it captures the general \u201cv5 breaks many models\u201d regression pattern seen across several items.", + "best_issue_reason": "issue:45850 has the widest scope and strongest discussion signal, so it is the best representative issue even though the cluster is too heterogeneous for true deduping.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44360", - "right": "issue:45468", - "accept": false, - "reason": "Different model families and failures: DSA indexer/ReLU vs Gemma audio positional encoding." - }, - { - "left": "issue:44655", - "right": "issue:44861", - "accept": false, - "reason": "One is save_pretrained for Pipeline objects; the other is _get_tied_weight_keys crashing on a list." - }, - { - "left": "issue:45230", - "right": "issue:45310", - "accept": false, - "reason": "Generic bug report vs a specific Qwen3.5 MoE from_pretrained error; no concrete match." - }, - { - "left": "issue:44273", - "right": "issue:44908", - "accept": false, - "reason": "Lazy loading behavior and inverse_sqrt scheduler kwargs are unrelated code paths." - }, - { - "left": "issue:43576", - "right": "issue:46218", - "accept": false, - "reason": "Broken env command vs import failure for Qwen3_5ForCausalLM are different issues." - }, - { - "left": "issue:43827", - "right": "issue:44492", - "accept": false, - "reason": "Docs referencing pipeline() and a cache-strategy typo are not the same defect." - }, { "left": "issue:46018", "right": "issue:46172", "accept": false, - "reason": "DeepSeek-V4 shared expert gating vs a deepseek-v4 tensor transpose issue are distinct." + "reason": "Different model families and bugs: DeepSeek-V4 gating vs a deepseek-v4 dtype/transpose problem." }, { "left": "issue:43519", "right": "issue:44485", "accept": false, - "reason": "Qwen3VL timestamp calculation and GLM-5 RoPE implementation are separate model bugs." + "reason": "Qwen3VL timestamp math is unrelated to GLM-5 RoPE implementation." }, { "left": "issue:43519", "right": "issue:44908", "accept": false, - "reason": "Processor timestamp math and scheduler kwargs handling are unrelated." + "reason": "Processor timestamp calculation and scheduler kwargs handling are unrelated code paths." }, { "left": "issue:41669", "right": "issue:43704", "accept": false, - "reason": "Import-star cleanup in models does not match a VRAM leak in Qwen3ForCausalLM." + "reason": "Import-star performance/footprint cleanup is unrelated to a VRAM leak in dataloader threads." }, { "left": "issue:31795", "right": "issue:34689", "accept": false, - "reason": "Documentation confusion and a Llama 3.2 model-loading regression are different problems." + "reason": "Documentation confusion in forward args is not the same as a model-loading regression." }, { "left": "issue:31515", "right": "issue:31795", "accept": false, - "reason": "Slow checkpoint loading is not the same as confusing forward docs." + "reason": "Slow checkpoint loading and forward-doc confusion are different issues." }, { "left": "issue:44485", "right": "issue:45468", "accept": false, - "reason": "RoPE implementation and Gemma4AudioRelPositionalEncoding are distinct model-specific changes." + "reason": "Both touch positional/attention-style internals, but they are different models and different failures." }, { "left": "issue:43704", "right": "issue:44485", "accept": false, - "reason": "A VRAM leak in multi-threaded loading and a RoPE bug do not share the same failure mode." + "reason": "VRAM leak in threaded loading is unrelated to GLM-5 RoPE math." }, { "left": "issue:43298", "right": "issue:46095", "accept": false, - "reason": "Continuous batching support for audio models is unrelated to a DeepSeek-V4 weight conversion script request." + "reason": "Audio continuous batching support and DeepSeek-V4 weight conversion are different feature requests." }, { "left": "issue:45850", "right": "issue:46218", "accept": false, - "reason": "Broad v5 breakage complaint vs a specific Qwen3_5 import failure; too coarse to merge." + "reason": "Broad v5 breakage versus a specific import failure for Qwen3_5ForCausalLM." }, { "left": "issue:43502", "right": "issue:43519", "accept": false, - "reason": "local_files_only network requests and Qwen3VL timestamp calculation are unrelated." + "reason": "Offline file-only network calls are unrelated to Qwen3VL timestamp calculation." }, { "left": "issue:42371", "right": "issue:43704", "accept": false, - "reason": "TF32 settings API guidance vs VRAM leakage are different issues." + "reason": "TF32 settings guidance and a dataloader-thread VRAM leak are unrelated." }, { "left": "issue:43519", "right": "issue:44623", "accept": false, - "reason": "Processor timestamp logic and missing save_pretrained files are separate bugs." + "reason": "Processor timestamp math and processor.save_pretrained missing files are different bugs." }, { "left": "issue:43519", "right": "issue:43704", "accept": false, - "reason": "Different components: Qwen3VL processor timestamps vs Qwen3ForCausalLM memory leak." + "reason": "Qwen3VL timestamp math and multi-thread VRAM leakage do not share the same failure." }, { "left": "issue:43606", "right": "issue:45468", "accept": false, - "reason": "CPU offload device mismatch in bark-small is unrelated to Gemma audio positional encoding." + "reason": "CPU offload device mismatch is unrelated to Gemma4 audio positional encoding." }, { "left": "issue:44263", "right": "issue:45468", "accept": false, - "reason": "torch.split return values in GlmMoeDsaIndexer do not match Gemma audio positional encoding." + "reason": "torch.split return-value handling in GLM-MoE is unrelated to Gemma4 audio positional encoding." }, { "left": "issue:44261", "right": "issue:45468", "accept": false, - "reason": "MLA q_a_layernorm epsilon precision issue is a different model-specific bug." + "reason": "Missing rms_norm_eps in MLA q_a_layernorm is a different model/config bug." }, { "left": "issue:44351", "right": "issue:46218", "accept": false, - "reason": "HybridCache import failure and Qwen3_5ForCausalLM import failure are different symbols and code paths." + "reason": "Different missing imports for different symbols/modules." }, { "left": "issue:43352", "right": "issue:46218", "accept": false, - "reason": "Flash Attention support gap in NemotronHForCausalLM is not the same as Qwen3_5 import failure." + "reason": "Flash Attention support gating and a Qwen3_5 import failure are unrelated." }, { "left": "issue:42913", "right": "issue:43066", "accept": false, - "reason": "Both are tokenizer-related but describe different v5 symptoms and likely different code paths." + "reason": "Both are tokenizer/v5 related, but one is behavior drift and the other is a decoder type bug." }, { "left": "issue:44908", "right": "issue:44998", "accept": false, - "reason": "Scheduler kwargs bug and an unrelated non-technical issue are not duplicates." + "reason": "Scheduler kwargs bug and an off-topic issue are unrelated." }, { "left": "issue:44861", "right": "issue:45290", "accept": false, - "reason": "_tied_weights_keys list crash and apply_chat_template tool-call crash are unrelated." + "reason": "Tied-weight key handling and chat-template tool-call crashing are different code paths." }, { "left": "issue:43525", "right": "issue:44861", "accept": false, - "reason": "Missing pad_token_id in Llama4Config is not the same as tied-weights key handling." + "reason": "Missing pad_token_id in config is unrelated to tied-weight key extraction." }, { "left": "issue:43704", "right": "issue:44998", "accept": false, - "reason": "VRAM leak vs unrelated non-technical issue." + "reason": "VRAM leak and off-topic issue are unrelated." + }, + { + "left": "issue:43572", + "right": "issue:44861", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to tied-weight key crashes." }, { "left": "issue:41669", "right": "issue:44998", "accept": false, - "reason": "Import-star cleanup issue vs unrelated non-technical issue." + "reason": "Import-star cleanup and off-topic issue are unrelated." }, { "left": "issue:43066", "right": "issue:44998", "accept": false, - "reason": "Tokenizer decoder type bug vs unrelated non-technical issue." + "reason": "Tokenizer decoder-type regression and off-topic issue are unrelated." }, { "left": "issue:31515", "right": "issue:38617", "accept": false, - "reason": "Slow from_pretrained checkpoint loading and missing layer_type_validation import are different." - }, - { - "left": "issue:44230", - "right": "issue:44623", - "accept": false, - "reason": "fp8 support request for qwen3-vl/qwen3.5 MoE and missing processor save files are unrelated." - }, - { - "left": "issue:44998", - "right": "issue:46218", - "accept": false, - "reason": "Unrelated non-technical issue vs Qwen3_5 import failure." - }, - { - "left": "issue:30333", - "right": "issue:31515", - "accept": false, - "reason": "MLFlow job-status reporting and checkpoint load slowness are unrelated." - }, - { - "left": "issue:31515", - "right": "issue:36296", - "accept": false, - "reason": "Checkpoint loading slowness and tensor-parallel training bug are different concerns." + "reason": "Checkpoint load slowness is unrelated to an ImportError for layer_type_validation." }, { - "left": "issue:44230", - "right": "issue:45310", - "accept": false, - "reason": "fp8 inference support request and Qwen3.5 MoE from_pretrained failure are not the same bug." - }, - { - "left": "issue:31515", - "right": "issue:33453", - "accept": false, - "reason": "Checkpoint loading performance regression and tokenizer loading regression are distinct." - }, - { - "left": "issue:43298", - "right": "issue:44230", - "accept": false, - "reason": "Audio-model continuous batching support and qwen3-vl/qwen3.5 fp8 inference support are different feature gaps." - }, - { - "left": "issue:43717", - "right": "issue:46144", - "accept": false, - "reason": "Mamba init_weights usage and RoFormer attention interface are unrelated model implementations." - }, - { - "left": "issue:42936", - "right": "issue:46018", - "accept": false, - "reason": "Mask2former ignore_value handling is not the same as DeepSeek-V4 shared-expert gating." - }, - { - "left": "issue:43668", - "right": "issue:44031", - "accept": false, - "reason": "Incorrect type hint in ModernBERTConfig and tokenizer regex warning are unrelated." - } - ] - }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, - "accepted_nontrivial_soft_edge": false, - "error_kind": null, - "error_message": null - }, - { - "label": "issue", - "component_index": 1, - "component_count": 18, - "review_unit_index": 51, - "review_unit_count": 51, - "status": "reviewed", - "reason": null, - "source": "cache", - "cache_hit": true, - "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-439", - "nodes": [ - "issue:37934", - "issue:42936", - "issue:43240", - "issue:43668", - "issue:43813", - "issue:44031", - "issue:46144" - ], - "soft_pairs": [ - "issue:42936|issue:43240", - "issue:43240|issue:46144", - "issue:42936|issue:44031", - "issue:42936|issue:46144", - "issue:42936|issue:43813", - "issue:43240|issue:43668", - "issue:43668|issue:43813", - "issue:37934|issue:43668", - "issue:42936|issue:43668" - ], - "prepared_review_unit_hash": "2b10f80c507cdcef3e1eb7c83b3f0fac260952bb51775e571bc950351fc6d7b7", - "component_budget": { - "node_count": 439, - "item_count": 439, - "soft_pair_count": 1616, - "serialized_chars": 837663, - "estimated_input_tokens": 209416, - "estimated_eval_tokens": 419088 - }, - "budget": { - "node_count": 7, - "item_count": 7, - "soft_pair_count": 9, - "serialized_chars": 6639, - "estimated_input_tokens": 1660, - "estimated_eval_tokens": 3576 - }, - "overflow_policy": "truncate_then_skip", - "trimmed": false, - "aggressively_trimmed": false, - "split": true, - "analyst_result": { - "summary": "These issues are not duplicates; they span unrelated areas including a norm-layer question, Mask2Former ignore_value handling, loss kwargs propagation, a config type hint, a typo, tokenizer regex warnings, and RoFormer attention interface usage.", - "confidence": 0.96, - "canonical_issue_reason": "No true canonical duplicate exists in this set. If one issue must anchor the cluster, issue 43240 is the broadest/general framework bug, but it is still unrelated to the other reports.", - "canonical_pr_reason": null, - "best_issue_reason": "Issue 43240 is the closest thing to a cluster anchor because it concerns generic loss-call plumbing rather than a single model-specific typo or configuration detail.", - "best_pr_reason": null, - "soft_edge_verdicts": [ + "left": "issue:44230", + "right": "issue:44623", + "accept": false, + "reason": "FP8 support request is unrelated to missing processor save files." + }, + { + "left": "issue:44998", + "right": "issue:46218", + "accept": false, + "reason": "Off-topic issue and a Qwen3_5 import failure are unrelated." + }, + { + "left": "issue:30333", + "right": "issue:31515", + "accept": false, + "reason": "MLflow reporting after failed training and slow from_pretrained checkpoint loading are unrelated." + }, + { + "left": "issue:31515", + "right": "issue:36296", + "accept": false, + "reason": "Checkpoint load slowness and tensor-parallel training bug are different failures." + }, + { + "left": "issue:44230", + "right": "issue:45310", + "accept": false, + "reason": "FP8 inference support and Qwen3.5 MoE from_pretrained error are not the same bug." + }, + { + "left": "issue:31515", + "right": "issue:33453", + "accept": false, + "reason": "Slow checkpoint loading and tokenizer-loading regression are distinct issues." + }, + { + "left": "issue:43298", + "right": "issue:44230", + "accept": false, + "reason": "Audio continuous batching support and FP8 inference support are unrelated." + }, + { + "left": "issue:43717", + "right": "issue:46144", + "accept": false, + "reason": "Mamba init_weights behavior is unrelated to RoFormer attention interface usage." + }, + { + "left": "issue:42936", + "right": "issue:46018", + "accept": false, + "reason": "Mask2former ignore_value handling is unrelated to DeepSeek-V4 shared-expert gating." + }, + { + "left": "issue:43668", + "right": "issue:44031", + "accept": false, + "reason": "A type-hint issue in ModernBERTConfig is unrelated to tokenizer regex warnings." + }, { "left": "issue:42936", "right": "issue:43240", "accept": false, - "reason": "Mask2Former ignore_value handling and loss-kwargs propagation are different bugs in different code paths." + "reason": "Ignore_value usage in Mask2former and loss-kwargs forwarding are unrelated." }, { "left": "issue:43240", "right": "issue:46144", "accept": false, - "reason": "Loss kwargs forwarding and RoFormer attention interface usage are unrelated implementation issues." + "reason": "Loss kwargs forwarding and RoFormer attention interface are unrelated." }, { "left": "issue:42936", "right": "issue:44031", "accept": false, - "reason": "A Mask2Former loss-parameter bug is not the same as tokenizer regex warning behavior." + "reason": "Mask2former ignore_value and tokenizer regex warning are unrelated." }, { "left": "issue:42936", "right": "issue:46144", "accept": false, - "reason": "These concern different subsystems: segmentation loss handling vs attention interface usage in RoFormer." + "reason": "Mask2former ignore_value and RoFormer attention interface are unrelated." }, { "left": "issue:42936", "right": "issue:43813", "accept": false, - "reason": "Mask2Former ignore_value logic and a typo in a quantization attribute are unrelated." + "reason": "Mask2former ignore_value and a quantization-operation typo are unrelated." }, { "left": "issue:43240", "right": "issue:43668", "accept": false, - "reason": "Passing kwargs to a loss function and an incorrect type hint in ModernBERTConfig are not the same underlying bug." + "reason": "Loss kwargs forwarding and a config type-hint typo are unrelated." }, { "left": "issue:43668", "right": "issue:43813", "accept": false, - "reason": "A config type-hint error is unrelated to a typo in a quantization field name." + "reason": "Config type-hint correctness and a quantization-operation typo are different issues." }, { "left": "issue:37934", "right": "issue:43668", "accept": false, - "reason": "A question about Llama4TextL2Norm vs RMSNorm is unrelated to a ModernBERTConfig annotation problem." + "reason": "Llama4TextL2Norm semantics and ModernBERTConfig type hints are unrelated." }, { "left": "issue:42936", "right": "issue:43668", "accept": false, - "reason": "Mask2Former loss-parameter handling and a ModernBERT config typing issue do not share the same code path or fix." + "reason": "Mask2former ignore_value and a ModernBERTConfig type hint are unrelated." } ] }, @@ -16446,8 +16424,8 @@ "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-10519-3", "nodes": [ @@ -16460,52 +16438,52 @@ "issue:10519|issue:11005", "issue:11005|issue:11307" ], - "prepared_review_unit_hash": "5ff925bcca17d8cae1050ee7f139f04be2298d337c4cdda928b13aeb67b7837d", + "prepared_review_unit_hash": "c96d02fdde0fc3e7748f502effef9f622ce56178adb91d01270bf86cad702203", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 2876, - "estimated_input_tokens": 719, - "estimated_eval_tokens": 1694 + "serialized_chars": 2877, + "estimated_input_tokens": 720, + "estimated_eval_tokens": 1696 }, "budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 2876, - "estimated_input_tokens": 719, - "estimated_eval_tokens": 1694 + "serialized_chars": 2877, + "estimated_input_tokens": 720, + "estimated_eval_tokens": 1696 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These three issues are unrelated: one is about truncation behavior in tokenization, one about ReduceLROnPlateau-like scheduler functionality, and one about Wav2Vec2 word time offsets. No duplicate cluster emerges.", + "summary": "These three issues are about unrelated areas: token truncation behavior, learning-rate scheduler functionality, and Wav2Vec2 word time offsets. They do not describe the same underlying bug or feature request, so the cluster should not be merged as duplicates.", "confidence": 0.98, - "canonical_issue_reason": "Issue #10519 is the clearest concrete product request and best represents the only coherent topic in the set; the other two issues are unrelated feature requests in different areas.", + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "#10519 is the most specific and actionable issue; it is the best standalone representative, but not a duplicate of the others.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:10519", "right": "issue:11307", "accept": false, - "reason": "Token truncation options and Wav2Vec2 word boundary timestamps are different features with no shared code-path or concrete bug overlap." + "reason": "Different feature areas and requests: truncation strategy in tokenization vs word start/end timestamps in Wav2Vec2. No shared code-path or concrete duplicate symptom." }, { "left": "issue:10519", "right": "issue:11005", "accept": false, - "reason": "Truncation-from-beginning is unrelated to ReduceLROnPlateau-like scheduler behavior; they concern different subsystems and problems." + "reason": "Unrelated topics: token truncation direction vs ReduceLROnPlateau-like scheduler behavior. They do not appear to be the same bug or change." }, { "left": "issue:11005", "right": "issue:11307", "accept": false, - "reason": "A learning-rate scheduler feature request and Wav2Vec2 time-offset extraction are unrelated issues with no plausible duplication." + "reason": "Different ML features: learning-rate scheduling vs extracting token time offsets from speech models. No plausible single fix or merged PR target." } ] }, @@ -16524,8 +16502,8 @@ "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": [ @@ -16537,7 +16515,7 @@ "issue:13244|issue:16618", "issue:15980|issue:16618" ], - "prepared_review_unit_hash": "abe46e12df50bd2b8975547c25fe78a4b98496b57276d0e9899e0c931253d464", + "prepared_review_unit_hash": "2e45e3e60be5be66820c62f40ee2c17efa7ef14551655c89acd275263076d6b7", "component_budget": { "node_count": 3, "item_count": 3, @@ -16559,24 +16537,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster looks mixed: one Tapas tokenization issue and two model-loading issues. The two model-loading tickets are superficially similar, but the available evidence is not strong enough to treat them as the same concrete bug, so I would not merge any soft edges.", - "confidence": 0.84, - "canonical_issue_reason": "Issue #16618 is the broadest and most central model-loading report in the set, with the most discussion and the clearest user-facing failure around loading a pretrained model.", + "summary": "This cluster does not look like a single bug family: one issue is about TAPAS tokenization differing from TensorFlow code, one about a private-model download/login error, and one about failing to load bert-base-uncased. The soft links are likely superficial Hugging Face model-related similarity rather than the same underlying problem.", + "confidence": 0.89, + "canonical_issue_reason": "issue:16618 is the most concrete and broadly representative user-facing failure in the set (a model loading bug), whereas the others concern unrelated tokenization behavior and private-model authentication messaging.", "canonical_pr_reason": null, - "best_issue_reason": "#16618 is the best representative of the cluster because it captures the main model-loading problem theme, whereas #13244 is unrelated and #15980 is narrower and more specific to private-model login/error messaging.", + "best_issue_reason": "issue:16618 is the best standalone issue because it describes a specific, actionable load failure with a common model name and is easier to triage than the more specialized or ancillary reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:13244", "right": "issue:16618", "accept": false, - "reason": "Different bug class and code path: Tapas tokenization behavior versus inability to load a pretrained model." + "reason": "Tokenization mismatch in TAPAS and inability to load bert-base-uncased are different code paths and different failure modes; they do not appear to be the same underlying bug." }, { "left": "issue:15980", "right": "issue:16618", "accept": false, - "reason": "Both concern model access/loading, but one is specifically about a private-model download/auth error and the other is a generic failure to load 'bert-base-uncased'; not enough evidence they are the same bug." + "reason": "Private-model download/login error is an authentication/message issue, while bert-base-uncased load failure is a separate model-loading problem; no clear shared root cause." } ] }, @@ -16595,8 +16573,8 @@ "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": [ @@ -16606,12 +16584,12 @@ "soft_pairs": [ "issue:26350|issue:33260" ], - "prepared_review_unit_hash": "89dea94304e87b13a1de33358b66b54f86d807632925567de5212ed71dd7aa1e", + "prepared_review_unit_hash": "2c9b694aa1bef3b70a6daeefa69a15e923d7d3fa679b71e776bb8c94d07b709f", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1590, + "serialized_chars": 1589, "estimated_input_tokens": 398, "estimated_eval_tokens": 1052 }, @@ -16619,7 +16597,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1590, + "serialized_chars": 1589, "estimated_input_tokens": 398, "estimated_eval_tokens": 1052 }, @@ -16628,18 +16606,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two separate community contribution feature requests: one for Flash Attention 2 support on more architectures, the other for GGUF support on more architectures. They are only loosely similar at a high level and should not be treated as duplicates.", - "confidence": 0.97, - "canonical_issue_reason": "Issue 26350 is the better anchor only because it is older and has substantially more discussion activity, but it is not a duplicate of 33260.", + "summary": "These are two separate community contribution requests: one for Flash Attention 2 support across more architectures, and one for GGUF support across more architectures. They share a broad theme of architecture support, but not the same underlying bug or change.", + "confidence": 0.98, + "canonical_issue_reason": "No canonical issue: the issues track different feature requests for different code paths and are not duplicates.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 26350 is the strongest representative issue in this pair due to age and higher engagement, though the topics are distinct.", + "best_issue_reason": "No single best issue for deduplication; each issue represents a distinct enhancement area rather than a duplicate of the other.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:26350", - "right": "issue:33260", + "left": "issue:26350 (Community contribution: Adding Flash Attention 2 support for more architectures)", + "right": "issue:33260 (Community contribution: Adding GGUF support for more architectures)", "accept": false, - "reason": "Different requested features in different inference/runtime paths: Flash Attention 2 vs GGUF support. The similarity is only broad subsystem-level language about architectures, not the same underlying bug or change." + "reason": "Different capabilities and likely different implementation paths; they share only a broad \"support more architectures\" theme, so they are not the same underlying request." } ] }, @@ -16658,8 +16636,8 @@ "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-27088-8", "nodes": [ @@ -16698,184 +16676,184 @@ "issue:44370|issue:46226", "issue:46222|issue:46226" ], - "prepared_review_unit_hash": "6c8001445ec5c67915bfad843678756c3797f1a5c1093594fb73e2567b96a6a9", + "prepared_review_unit_hash": "48debc4ac5c4516ca7ae7ae2544714a2d73ea8679f6fef2a0396b60e5b960593", "component_budget": { "node_count": 8, "item_count": 8, "soft_pair_count": 24, - "serialized_chars": 13529, - "estimated_input_tokens": 3383, - "estimated_eval_tokens": 7022 + "serialized_chars": 13528, + "estimated_input_tokens": 3382, + "estimated_eval_tokens": 7020 }, "budget": { "node_count": 8, "item_count": 8, "soft_pair_count": 24, - "serialized_chars": 13529, - "estimated_input_tokens": 3383, - "estimated_eval_tokens": 7022 + "serialized_chars": 13528, + "estimated_input_tokens": 3382, + "estimated_eval_tokens": 7020 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly a set of i18n docs-translation issues, but the evidence points to different language requests or template-like variants rather than one true duplicate set. Only the two exact Thai-title issues look like a likely duplicate pair.", - "confidence": 0.71, - "canonical_issue_reason": "issue:27088 is the earliest concrete, language-specific request (Turkish) and is clearer than the later template-like variants.", + "summary": "Mixed i18n cluster: Turkish/Spanish/Romanian translation requests are distinct, while the three generic placeholder issues form one duplicate subgroup and the two Thai-text issues form another exact duplicate pair. No single issue cleanly canonicalizes the whole set.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue fits the full cluster because it contains separate language-specific requests plus two independent duplicate subgroups.", "canonical_pr_reason": null, - "best_issue_reason": "issue:27088 is the most representative single issue in the group because it has a specific language target and avoids the placeholder/generic titles.", + "best_issue_reason": "issue:43834 is the best representative of the placeholder-template duplicate subgroup, but it is not a good global representative for the language-specific requests or the Thai duplicate pair.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:27088", "right": "issue:44370", "accept": false, - "reason": "Same translation-issue template, but different issue instances and no evidence they report the same specific language request." + "reason": "Different target language/request; Turkish vs generic placeholder text." }, { "left": "issue:27088", "right": "issue:44365", "accept": false, - "reason": "Template similarity is high, but the issues are not shown to target the same underlying translation request." + "reason": "Different target language/request; Turkish vs generic placeholder text." }, { "left": "issue:27088", "right": "issue:43834", "accept": false, - "reason": "The placeholder title makes this look generic, but there is no concrete evidence it is the same request as the Turkish issue." + "reason": "Turkish translation request is separate from the generic placeholder issue." }, { "left": "issue:27088", "right": "issue:38435", "accept": false, - "reason": "Different language target (Turkish vs Romanian), so these are separate requests, not duplicates." + "reason": "Different language requests: Turkish vs Romanian." }, { "left": "issue:27088", "right": "issue:28936", "accept": false, - "reason": "Different language target (Turkish vs Spanish), so these are separate requests, not duplicates." + "reason": "Different language requests: Turkish vs Spanish." }, { "left": "issue:28936", "right": "issue:43834", "accept": false, - "reason": "High text similarity appears to come from shared template structure; no evidence of the same underlying request." + "reason": "Spanish translation request is separate from the generic placeholder issue." }, { "left": "issue:28936", "right": "issue:44365", "accept": false, - "reason": "Similar template text only; the language-specific request is not shown to be the same." + "reason": "Spanish translation request is separate from the generic placeholder issue." }, { "left": "issue:28936", "right": "issue:38435", "accept": false, - "reason": "Spanish and Romanian translation requests are different issues." + "reason": "Different language requests: Spanish vs Romanian." }, { "left": "issue:28936", "right": "issue:44370", "accept": false, - "reason": "Shared i18n template, but no concrete evidence of the same underlying translation task." + "reason": "Spanish translation request is separate from the generic placeholder issue." }, { "left": "issue:38435", "right": "issue:44370", "accept": false, - "reason": "Romanian issue vs a generic placeholder-title issue; similarity is not enough to treat them as the same report." + "reason": "Romanian translation request is separate from the generic placeholder issue." }, { "left": "issue:38435", "right": "issue:44365", "accept": false, - "reason": "Template overlap only; no clear sign these are duplicate reports of the same request." + "reason": "Romanian translation request is separate from the generic placeholder issue." }, { "left": "issue:38435", "right": "issue:43834", "accept": false, - "reason": "Both are translation-related, but the placeholder title makes this too ambiguous to merge as duplicates." + "reason": "Romanian translation request is separate from the generic placeholder issue." }, { "left": "issue:38435", "right": "issue:46222", "accept": false, - "reason": "The Thai-titled issue is unrelated in language/content to the Romanian request." + "reason": "Different content: Romanian translation request vs Thai-text issue." }, { "left": "issue:38435", "right": "issue:46226", "accept": false, - "reason": "The Thai-titled issue is unrelated in language/content to the Romanian request." + "reason": "Different content: Romanian translation request vs Thai-text issue." }, { "left": "issue:43834", "right": "issue:44370", - "accept": false, - "reason": "Both use the same placeholder template, but the available evidence does not show the same concrete issue." + "accept": true, + "reason": "Same generic placeholder translation issue; looks like a duplicate copy of the same template artifact." }, { "left": "issue:43834", "right": "issue:44365", - "accept": false, - "reason": "Placeholder-title similarity alone is not enough to call these duplicates." + "accept": true, + "reason": "Same generic placeholder translation issue; looks like a duplicate copy of the same template artifact." }, { "left": "issue:43834", "right": "issue:46222", "accept": false, - "reason": "Different title content; no evidence they are the same request." + "reason": "Generic placeholder issue vs distinct Thai-text issue; not the same underlying request." }, { "left": "issue:43834", "right": "issue:46226", "accept": false, - "reason": "Different title content; no evidence they are the same request." + "reason": "Generic placeholder issue vs distinct Thai-text issue; not the same underlying request." }, { "left": "issue:44365", "right": "issue:44370", - "accept": false, - "reason": "Near-identical template text, but no proof they describe the same underlying issue." + "accept": true, + "reason": "Same generic placeholder translation issue; looks like a duplicate copy of the same template artifact." }, { "left": "issue:44365", "right": "issue:46222", "accept": false, - "reason": "Different title content and no shared issue target or code-path evidence." + "reason": "Generic placeholder issue vs distinct Thai-text issue; not the same underlying request." }, { "left": "issue:44365", "right": "issue:46226", "accept": false, - "reason": "Different title content and no shared issue target or code-path evidence." + "reason": "Generic placeholder issue vs distinct Thai-text issue; not the same underlying request." }, { "left": "issue:44370", "right": "issue:46222", "accept": false, - "reason": "Different title content and no shared issue target or code-path evidence." + "reason": "Generic placeholder issue vs distinct Thai-text issue; not the same underlying request." }, { "left": "issue:44370", "right": "issue:46226", "accept": false, - "reason": "Different title content and no shared issue target or code-path evidence." + "reason": "Generic placeholder issue vs distinct Thai-text issue; not the same underlying request." }, { "left": "issue:46222", "right": "issue:46226", "accept": true, - "reason": "These are exact title matches with the same body length and very close timestamps, which strongly suggests a duplicate issue." + "reason": "Exact duplicate Thai-text issue; same title/content and same underlying artifact." } ] }, "evaluator_result": { "accept": true, - "feedback": "The analysis stays grounded in the packet and is appropriately conservative on the soft edges. The only accepted duplicate pair, issue:46222 vs issue:46226, is well supported by exact title match, identical body length, and close timestamps. The remaining false verdicts are cautious given the evidence, and the summary does not overstate duplicate status across different language-targeted requests." + "feedback": "The summary is grounded in the packet: the three language-specific translation requests are reasonably treated as separate, the three placeholder-template issues form a coherent duplicate subgroup, and the two Thai-title issues are an exact duplicate pair. The soft-edge verdicts are conservative overall, with no clear overstatement of evidence." }, "evaluator_used": true, "retried": false, @@ -16891,8 +16869,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-36817-16", "nodes": [ @@ -16951,250 +16929,250 @@ "issue:43837|issue:45177", "issue:45322|issue:46133" ], - "prepared_review_unit_hash": "482d6d60d95736084fea705a6cc5c0f844421b25c1cf112d8a5c24e6a8aeef2d", + "prepared_review_unit_hash": "907f945b8a32ec29c695556c788f027f25f7cb421aae0585240a85f07e97c890", "component_budget": { "node_count": 16, "item_count": 16, "soft_pair_count": 36, - "serialized_chars": 21658, - "estimated_input_tokens": 5415, - "estimated_eval_tokens": 11086 + "serialized_chars": 21661, + "estimated_input_tokens": 5416, + "estimated_eval_tokens": 11088 }, "budget": { "node_count": 16, "item_count": 16, "soft_pair_count": 36, - "serialized_chars": 21658, - "estimated_input_tokens": 5415, - "estimated_eval_tokens": 11086 + "serialized_chars": 21661, + "estimated_input_tokens": 5416, + "estimated_eval_tokens": 11088 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are mostly distinct model-request issues spanning unrelated model families and modalities. The titles are superficially similar, but there is no shared underlying bug/change to collapse into duplicates. No soft edges accepted.", - "confidence": 0.96, - "canonical_issue_reason": "No single canonical issue: each item targets a different model or capability (e.g. EuroBert, RF-DETR, NeoBERT, MiniCPM3, Qwen3 TTS/ASR, vision-language encoders), so they should remain separate tickets.", + "summary": "These are all distinct model-request issues that share a generic 'add model/support' pattern, but they target different models and code paths. No soft edge looks like a true duplicate or same underlying change.", + "confidence": 0.93, + "canonical_issue_reason": "issue:43671 is the most central and broadly representative request in this cluster, with the widest set of soft links and a generic model-support framing.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43671 is one of the clearest self-contained requests, but only as an exemplar; it is not representative enough to serve as a duplicate anchor for the whole cluster.", + "best_issue_reason": "issue:43671 is the best umbrella representative for the cluster because it is broad, central, and least tied to a single narrow model variant.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36817", "right": "issue:41211", "accept": false, - "reason": "EuroBert config vs DEIMv2: different model additions with no shared code-path change." + "reason": "Different requested models and architectures; only the 'add model' wording overlaps." }, { "left": "issue:36817", "right": "issue:43671", "accept": false, - "reason": "EuroBert model config and Qwen3-TTS support are unrelated model requests." + "reason": "EuroBert config vs Qwen3-TTS support are unrelated model additions." }, { "left": "issue:36817", "right": "issue:36879", "accept": false, - "reason": "EuroBert config and RF-DETR support are different model families and tasks." + "reason": "EuroBert config and RF-DETR target different model families and code paths." }, { "left": "issue:36879", "right": "issue:41211", "accept": false, - "reason": "RF-DETR and DEIMv2 are distinct detection models; not the same request." + "reason": "RF-DETR and DEIMv2 are distinct detection models, not the same change." }, { "left": "issue:36879", "right": "issue:43909", "accept": false, - "reason": "Object detection model support vs audio model support are different changes." + "reason": "Object detection support vs an audio model request are unrelated." }, { "left": "issue:36879", "right": "issue:44121", "accept": false, - "reason": "RF-DETR support and sparse-transformer support are unrelated model additions." + "reason": "RF-DETR and weight-sparse transformer support are different model changes." }, { "left": "issue:36879", "right": "issue:43837", "accept": false, - "reason": "RF-DETR detection support and Qwen3-ASR support are different model requests." + "reason": "RF-DETR and Qwen3-ASR are different tasks and model families." }, { "left": "issue:36879", "right": "issue:43671", "accept": false, - "reason": "RF-DETR detection support and Qwen3-TTS support are unrelated." + "reason": "RF-DETR and Qwen3-TTS are unrelated requests beyond generic model support." }, { "left": "issue:37015", "right": "issue:43837", "accept": false, - "reason": "NeoBERT and Qwen3-ASR target different model architectures and tasks." + "reason": "NeoBERT and Qwen3-ASR are different model families and capabilities." }, { "left": "issue:37015", "right": "issue:44121", "accept": false, - "reason": "NeoBERT and the sparse transformer request do not describe the same code change." + "reason": "NeoBERT addition is not the same as sparse-transformer support." }, { "left": "issue:37015", "right": "issue:43909", "accept": false, - "reason": "NeoBERT and LFM2.5 Audio are unrelated model requests." + "reason": "A BERT-style text model and an audio model are unrelated." }, { "left": "issue:36879", "right": "issue:37015", "accept": false, - "reason": "RF-DETR and NeoBERT are unrelated model additions." + "reason": "RF-DETR and NeoBERT are distinct requests with no shared concrete bug/change." }, { "left": "issue:37015", "right": "issue:42503", "accept": false, - "reason": "NeoBERT and ModernVBERT are different model families; not the same underlying change." + "reason": "NeoBERT and ModernVBERT are different model-specific requests, not a duplicate fix." }, { "left": "issue:41115", "right": "issue:43837", "accept": false, - "reason": "MiniCPM3 architecture support and Qwen3-ASR support are different requests." + "reason": "MiniCPM3 architecture support and Qwen3-ASR support are separate model integrations." }, { "left": "issue:41115", "right": "issue:44121", "accept": false, - "reason": "MiniCPM3 and the sparse transformer proposal are not the same model change." + "reason": "MiniCPM3 and a weight-sparse transformer request do not describe the same change." }, { "left": "issue:41115", "right": "issue:42503", "accept": false, - "reason": "MiniCPM3 and ModernVBERT are distinct architecture/model requests." + "reason": "MiniCPM3 and ModernVBERT are different model additions." }, { "left": "issue:41115", "right": "issue:45027", "accept": false, - "reason": "MiniCPM3 support and Voxtral TTS support are different models and code paths." + "reason": "MiniCPM3 and Voxtral TTS target different model families and tasks." }, { "left": "issue:41211", "right": "issue:43671", "accept": false, - "reason": "DEIMv2 and Qwen3-TTS are unrelated model support tickets." + "reason": "DEIMv2 detection support and Qwen3-TTS support are unrelated." }, { "left": "issue:42503", "right": "issue:43837", "accept": false, - "reason": "ModernVBERT and Qwen3-ASR are different model families." + "reason": "ModernVBERT models and Qwen3-ASR are different requests." }, { "left": "issue:42503", "right": "issue:43671", "accept": false, - "reason": "ModernVBERT and Qwen3-TTS are unrelated model requests." + "reason": "ModernVBERT and Qwen3-TTS are not the same underlying change." }, { "left": "issue:42503", "right": "issue:42738", "accept": false, - "reason": "ModernVBERT models and BERT-like models with RoPE are not the same concrete addition." + "reason": "Both are BERT-themed, but one is a specific model request and the other is a general RoPE discussion; not a duplicate." }, { "left": "issue:42738", "right": "issue:45027", "accept": false, - "reason": "BERT-like RoPE support and Voxtral TTS support are unrelated changes." + "reason": "A general BERT-like RoPE request and Voxtral TTS support are unrelated." }, { "left": "issue:43671", "right": "issue:43837", "accept": false, - "reason": "Qwen3-TTS and Qwen3-ASR are related by family name but are different capabilities and not the same underlying request." + "reason": "TTS and ASR are different capabilities; same family, but not the same concrete change." }, { "left": "issue:43671", "right": "issue:45027", "accept": false, - "reason": "Both mention TTS, but they are different models and do not map to one mergeable change." + "reason": "Qwen3-TTS and Voxtral TTS are different model integrations." }, { "left": "issue:43671", "right": "issue:45177", "accept": false, - "reason": "Qwen3-TTS support and DeepSeek-OCR2 are different model requests." + "reason": "Audio TTS support and DeepSeek-OCR2 are unrelated model requests." }, { "left": "issue:43671", "right": "issue:44121", "accept": false, - "reason": "Qwen3-TTS support and the sparse-transformer proposal are unrelated." + "reason": "Qwen3-TTS and sparse-transformer support are different code paths and model families." }, { "left": "issue:43837", "right": "issue:43909", "accept": false, - "reason": "Qwen3-ASR and LFM2.5 Audio are different audio-model requests, not the same change." + "reason": "Qwen3-ASR and LFM2.5 Audio are both audio-related, but they are distinct model requests." }, { "left": "issue:43837", "right": "issue:45322", "accept": false, - "reason": "ASR support and EUPE vision-language encoder support are unrelated model additions." + "reason": "ASR support and EUPE vision-language encoder support are unrelated." }, { "left": "issue:43909", "right": "issue:44205", "accept": false, - "reason": "LFM2.5 Audio and SAM3-LiteText are different model families and tasks." + "reason": "Audio model support and SAM3-LiteText are different model domains." }, { "left": "issue:43671", "right": "issue:43909", "accept": false, - "reason": "Qwen3-TTS and LFM2.5 Audio are different TTS/audio-model additions." + "reason": "Qwen3-TTS and LFM2.5 Audio 1.5B are different audio model integrations." }, { "left": "issue:44121", "right": "issue:45322", "accept": false, - "reason": "Sparse transformer support and EUPE support are unrelated model requests." + "reason": "Sparse transformer support and EUPE are different model requests." }, { "left": "issue:43837", "right": "issue:44121", "accept": false, - "reason": "Qwen3-ASR support and sparse-transformer support are different code paths." + "reason": "ASR support and weight-sparse transformer support are unrelated." }, { "left": "issue:43837", "right": "issue:45027", "accept": false, - "reason": "Qwen3-ASR and Voxtral TTS are different audio-model requests." + "reason": "Both are audio-related, but ASR and TTS requests are distinct model changes." }, { "left": "issue:37015", "right": "issue:45027", "accept": false, - "reason": "NeoBERT and Voxtral TTS are unrelated model additions." + "reason": "NeoBERT and Voxtral-4B-TTS are unrelated model additions." }, { "left": "issue:43837", "right": "issue:45177", "accept": false, - "reason": "Qwen3-ASR support and DeepSeek-OCR2 are unrelated." + "reason": "ASR support and OCR model support are different features." }, { "left": "issue:45322", "right": "issue:46133", "accept": false, - "reason": "EUPE and TIPSv2 are distinct vision-language model requests; similar domain, different changes." + "reason": "EUPE and TIPSv2 are separate vision-language model requests, not the same change." } ] }, @@ -17213,8 +17191,8 @@ "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": [ @@ -17224,7 +17202,7 @@ "soft_pairs": [ "issue:41084|issue:43037" ], - "prepared_review_unit_hash": "06ab3184c07df60a20cebabd2fa10ca4cb48c3dfa69200c2d2426d734e32e7d9", + "prepared_review_unit_hash": "e940da8bb581782bf6d4b3ef3f8b36c70f45fb8a583a2bfe387241a705d44e1c", "component_budget": { "node_count": 2, "item_count": 2, @@ -17246,18 +17224,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are unrelated: one is about block decoding, the other is a DeepSeek v3.2 support request. No duplicate/canonical relationship is evident.", + "summary": "The two issues are not duplicates: one is about block decoding, the other is about adding DeepSeek v3.2 support. They describe different features/problems with no shared target or code path.", "confidence": 0.98, - "canonical_issue_reason": "No canonical issue: the titles and likely problem domains do not match.", + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a true duplicate, but 43037 is the more specific user-facing feature request.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41084", "right": "issue:43037", "accept": false, - "reason": "Different topics and no shared target or code-path evidence; this does not look like the same underlying bug or change." + "reason": "Different topics and likely different code paths; block decoding is unrelated to DeepSeek model support." } ] }, @@ -17276,8 +17254,8 @@ "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": [ @@ -17287,12 +17265,12 @@ "soft_pairs": [ "issue:43598|issue:43600" ], - "prepared_review_unit_hash": "7f149f6076aacee460fc076144ce5b59c3eae407aee954396870ea2ac2a99267", + "prepared_review_unit_hash": "3b73fc8e8f885e0913655ff4a86d78ff1ced2108635afee613a6dc41ea6aa593", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1583, + "serialized_chars": 1584, "estimated_input_tokens": 396, "estimated_eval_tokens": 1048 }, @@ -17300,7 +17278,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1583, + "serialized_chars": 1584, "estimated_input_tokens": 396, "estimated_eval_tokens": 1048 }, @@ -17309,18 +17287,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related to Trainer customizability, but they describe different problems: one is about `remove_unused_column`, the other about FLOPS estimation for custom trainers. They should not be merged as duplicates.", - "confidence": 0.92, - "canonical_issue_reason": "Issue #43598 is the better canonical issue only if the cluster is centered on Trainer customizability, because it is a concrete request to revisit a specific Trainer behavior (`remove_unused_column`).", + "summary": "These two closed issues both mention Trainer customization, but they describe different problems: one is about `remove_unused_column` behavior, the other about FLOPS estimation for custom trainers. They should not be treated as duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "No strong canonical issue emerges because the topics are distinct rather than overlapping reports of the same bug.", "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a strong global canonical for the pair; if forced to pick one, #43598 is slightly more concrete and narrowly scoped than the FLOPS-estimation request.", + "best_issue_reason": "Issue #43598 is slightly more specific and actionable around a concrete Trainer setting, so it is the better representative if one must be chosen.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43598", "right": "issue:43600", "accept": false, - "reason": "They are adjacent Trainer customization topics, but not the same underlying bug or change. One concerns column removal behavior, the other concerns FLOPS counting for custom trainers." + "reason": "Different underlying concerns: one is about column-removal customizability in Trainer, the other about FLOPS accounting for custom trainers. Shared Trainer context is too broad to treat as the same issue." } ] }, @@ -17339,8 +17317,8 @@ "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": [ @@ -17350,7 +17328,7 @@ "soft_pairs": [ "issue:43684|issue:44405" ], - "prepared_review_unit_hash": "a5037a0e4d9d2af8461aa0dfb211a502ae01d0fc29824daccf816c7691b989b3", + "prepared_review_unit_hash": "3c603234892ba3ead0b7c30c3952d891c4d5084dd7c129512050878b5cc25296", "component_budget": { "node_count": 2, "item_count": 2, @@ -17372,18 +17350,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates: one requests general AutoModel/AutoModelForConditionalGeneration registration for Qwen3-Omni, while the other requests AutoModelForSequenceClassification support for Qwen3.5. They involve different model families and different auto classes/capabilities.", - "confidence": 0.94, - "canonical_issue_reason": "No true canonical duplicate here; if one must be chosen as the representative issue, #43684 is broader and more central because it concerns model registration in core auto classes.", + "summary": "The two issues are related only at the very broad level of adding AutoModel support, but they target different model families and different auto classes, so they are not duplicates.", + "confidence": 0.93, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "#43684 is the better representative issue because it describes the broader auto-model registration gap, whereas #44405 is a separate, more specific sequence-classification support request for a different config/model line.", + "best_issue_reason": "Issue 43684 is the broader request: it covers Qwen3-Omni registration for both AutoModel and AutoModelForConditionalGeneration, whereas 44405 is a narrower, separate request for sequence classification on Qwen3.5.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43684", "right": "issue:44405", "accept": false, - "reason": "Different underlying requests: Qwen3-Omni auto-registration vs Qwen3.5 sequence-classification support. Different model/config targets and different auto classes, so they should not be merged as duplicates." + "reason": "Different underlying features/models: Qwen3-Omni auto-registration vs Qwen3.5 sequence-classification support. They do not describe the same bug or change path and would not be merged into one fix." } ] }, @@ -17402,8 +17380,8 @@ "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": [ @@ -17420,12 +17398,12 @@ "issue:43845|issue:43859", "issue:43846|issue:43859" ], - "prepared_review_unit_hash": "549bef0f323a46daa752b8ecd0190a9ad53433bbb35a421f48d9f50bcdafcbc5", + "prepared_review_unit_hash": "3461e2b8c6ce78d0b69667ade54a7017554095fb126ad3695b9c3001cc0f9e11", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4264, + "serialized_chars": 4263, "estimated_input_tokens": 1066, "estimated_eval_tokens": 2388 }, @@ -17433,7 +17411,7 @@ "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4264, + "serialized_chars": 4263, "estimated_input_tokens": 1066, "estimated_eval_tokens": 2388 }, @@ -17442,54 +17420,54 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Three issues share the same title/body pattern (\u201chuggingface \u201d) and look like duplicate placeholders; the unrelated issue \u201cFett\u201d does not belong in that cluster.", - "confidence": 0.79, - "canonical_issue_reason": "issue:43845 is the earliest of the three near-identical \u201chuggingface \u201d issues and is the best representative of that duplicate set.", + "summary": "The cluster appears to contain one unrelated issue ('Fett') and a duplicate trio of near-identical 'huggingface ' issues. With no substantive bodies, the safest canonical choice is the earliest huggingface issue.", + "confidence": 0.92, + "canonical_issue_reason": "issue:43845 is the earliest of the three identical 'huggingface ' issues and is the best representative for that duplicate set; issue:43835 ('Fett') looks unrelated.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43845 best fits the cluster because it matches the repeated title and is the earliest of the matching issues; issue:43835 appears unrelated.", + "best_issue_reason": "issue:43845 is the strongest global issue representative because it matches the other two huggingface issues exactly in title and is earliest, while the only other title in the cluster is unrelated.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43835", "right": "issue:43845", "accept": false, - "reason": "Different titles with no shared content beyond generic similarity; no evidence they describe the same issue." + "reason": "Different titles with no shared substantive details; nothing indicates the same underlying report." }, { "left": "issue:43835", "right": "issue:43846", "accept": false, - "reason": "Different titles with no concrete overlap in description or targets." + "reason": "Different titles with no shared substantive details; nothing indicates the same underlying report." }, { "left": "issue:43835", "right": "issue:43859", "accept": false, - "reason": "Different titles and no shared issue details; not the same underlying bug." + "reason": "Different titles with no shared substantive details; nothing indicates the same underlying report." }, { "left": "issue:43845", "right": "issue:43846", "accept": true, - "reason": "Same title (\u201chuggingface \u201d) and essentially identical sparse content, so they look like duplicates." + "reason": "Same normalized title ('huggingface ') and close creation times, with no evidence of distinct topics." }, { "left": "issue:43845", "right": "issue:43859", "accept": true, - "reason": "Same title and no differentiating details; likely the same duplicate issue." + "reason": "Same normalized title ('huggingface ') and close creation times; likely duplicate submissions of the same issue." }, { "left": "issue:43846", "right": "issue:43859", "accept": true, - "reason": "Same title and matching minimal content, indicating the same underlying issue." + "reason": "Same normalized title ('huggingface ') and close creation times, consistent with duplicates of one report." } ] }, "evaluator_result": { "accept": true, - "feedback": "The clustering is grounded in the packet: issues 43845/43846/43859 share the same title and have strong soft similarity, while 43835 is distinct and only weakly similar. The chosen canonical issue is the earliest among the matching set, and the soft-edge verdicts are conservative." + "feedback": "The summary is well grounded in the packet: the three \"huggingface \" issues are title-identical and closely timed, while \"Fett\" is distinct. The canonical choice of the earliest huggingface issue is conservative, and the soft-edge verdicts do not overstate evidence." }, "evaluator_used": true, "retried": false, @@ -17505,8 +17483,8 @@ "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": [ @@ -17527,7 +17505,7 @@ "issue:44572|issue:44937", "issue:44573|issue:44937" ], - "prepared_review_unit_hash": "d6d401e5dfe42ffe0846e5094c05d1bf062e3e8e89c2b9fbd597572392c2a7f9", + "prepared_review_unit_hash": "d53f1092e43db99dd1fa31b3ae01c9423542bf97eae17fc59f5a890c967f15ba", "component_budget": { "node_count": 5, "item_count": 5, @@ -17549,66 +17527,66 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster looks like unrelated low-signal/spam issues rather than true duplicates. None of the soft pairs should be merged.", - "confidence": 0.93, - "canonical_issue_reason": "issue:44418 is the only item with a non-spam, vaguely substantive title, so it is the best representative if one must be chosen.", + "summary": "This cluster is dominated by near-empty/spam-like issues with no concrete bug or change details. Issue 44418 is the only somewhat substantive topic, but there is still not enough evidence that any pair is a true duplicate.", + "confidence": 0.34, + "canonical_issue_reason": "issue:44418 has the most meaningful, non-spam title and is the only item that looks like a real triage target; the others are generic spam placeholders.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44418 is the most suitable issue to anchor the cluster because it is at least a recognizable documentation request, unlike the placeholder spam titles.", + "best_issue_reason": "issue:44418 is the best representative of the cluster because it is the only issue with an identifiable subject matter and some potential triage value.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44371", "right": "issue:44572", "accept": false, - "reason": "Both are titled '', but that is not enough to show the same underlying issue; they read as separate spam submissions with no concrete shared bug or change." + "reason": "Both are just '' with no body detail; identical placeholder titles are not enough to prove the same underlying issue." }, { "left": "issue:44371", "right": "issue:44573", "accept": false, - "reason": "Same placeholder spam title, but no evidence of one underlying report; treat as distinct spam issues." + "reason": "Both are just '' with no actionable description, so there is no concrete evidence they report the same problem." }, { "left": "issue:44371", "right": "issue:44418", "accept": false, - "reason": "'' vs a documentation suggestion are clearly different topics." + "reason": "One is spam-like and the other is a generic documentation request; the titles do not indicate the same underlying bug or change." }, { "left": "issue:44371", "right": "issue:44937", "accept": false, - "reason": "'' and 'Check out \"Google LLC\"' are unrelated and look like separate spam-like submissions." + "reason": "The placeholder spam title and the specific 'Google LLC' title are unrelated on their face, with no body evidence tying them together." }, { "left": "issue:44418", "right": "issue:44572", "accept": false, - "reason": "A documentation enhancement request is not the same underlying issue as a spam placeholder." + "reason": "A documentation enhancement request and a spam placeholder do not describe the same issue; there is no shared concrete problem statement." }, { "left": "issue:44418", "right": "issue:44573", "accept": false, - "reason": "These titles point to different kinds of reports; no duplicate evidence beyond generic similarity." + "reason": "The documentation suggestion is too generic, and the other item is empty spam; no basis to merge as duplicates." }, { "left": "issue:44572", "right": "issue:44573", "accept": false, - "reason": "Identical spam titles alone do not establish the same bug or change; they could be independent junk submissions." + "reason": "Matching spam placeholders alone do not establish a duplicate; there is no description or target to confirm the same underlying report." }, { "left": "issue:44572", "right": "issue:44937", "accept": false, - "reason": "Placeholder spam and a branded promotional title are not the same underlying issue." + "reason": "A spam placeholder and a specific entity-name title are not the same reported problem, and there is no supporting detail." }, { "left": "issue:44573", "right": "issue:44937", "accept": false, - "reason": "These look like different spam artifacts, not one fixable problem." + "reason": "These titles are not describing the same concrete bug or request, and the bodies provide no evidence of a shared root cause." } ] }, @@ -17627,8 +17605,8 @@ "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": [ @@ -17638,7 +17616,7 @@ "soft_pairs": [ "issue:45145|issue:45151" ], - "prepared_review_unit_hash": "1e5819a01de7f68854674bdae2d138309ce027dc7110d6bd478612817547dc0a", + "prepared_review_unit_hash": "a64c90beeda203331664a7c1862ba6065c02a171e9900ce7eeb28bc31b169da5", "component_budget": { "node_count": 2, "item_count": 2, @@ -17660,28 +17638,28 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two closed issues share an identical title and soft similarity signals, so they likely describe the same report; however, the available packet does not expose the issue bodies, so the duplicate conclusion should be kept conservative.", - "confidence": 0.86, - "canonical_issue_reason": "#45145 is the better canonical issue because it was created first and already has an inbound reference, which makes it the more likely original thread.", + "summary": "The two closed issues share an identical title and some textual overlap, but the evidence is not strong enough to call them near-identical duplicates or to consolidate them as the same report.", + "confidence": 0.9, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "#45145 is the best issue candidate because it is the earliest of the two and has the only inbound reference, while the evidence available here does not distinguish a different scope.", + "best_issue_reason": "issue #45145 is the slightly better cluster representative because it is earlier and already has an inbound reference, but this is only a mild preference, not a duplicate determination.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45145", "right": "issue:45151", - "accept": true, - "reason": "The issues have the same title and a soft similarity signal; that is enough to treat them as likely duplicates, but the packet does not show the bodies, so the match is based on title/similarity rather than any confirmed identical proposal." + "accept": false, + "reason": "Identical titles and moderate similarity suggest related reports, but the packet does not show sufficiently similar bodies or evidence of the same underlying issue to justify duplicate consolidation." } ] }, "evaluator_result": { "accept": false, - "feedback": "The packet supports that these two closed issues are highly similar and share the same title, but the analyst overstates the evidence in a few places. In particular, the packet does not show the bodies themselves, only similar body lengths, so claims like \u201cnearly identical body size,\u201d \u201csame broad proposal,\u201d and \u201cno evidence of different scope\u201d go beyond what is grounded. The duplicate-issue conclusion may be plausible, but the soft-edge verdict should be stated more conservatively based on title match and soft similarity only." + "feedback": "The pairing is only supported as a soft similarity match with identical titles and moderate Jaccard similarity; the packet does not show the bodies are \"highly similar\" or that the issues are \"near-identical duplicates.\" The consolidation/duplicate language is a bit too strong for the evidence, so this should be rejected as overstated." }, "evaluator_used": true, "retried": true, - "accepted_nontrivial_soft_edge": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -17747,8 +17725,8 @@ "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": [ @@ -17758,7 +17736,7 @@ "soft_pairs": [ "issue:45672|issue:45674" ], - "prepared_review_unit_hash": "842e6ab531aaabd33816066e9a3abce215014ae236d6614d858d3fa166350a5c", + "prepared_review_unit_hash": "0e6650b004bf55a238b3d2999a96a9ab90dcb286490a30050c01223489a249c6", "component_budget": { "node_count": 2, "item_count": 2, @@ -17780,18 +17758,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are related only at a high level (4-bit inference in the same stack) but describe different failures and code paths: one is a Gemma4 audio-module `torch.finfo()` TypeError on `uint8` weights, the other is a BitsAndBytesConfig `llm_int8_skip_modules` regression that removes the default `lm_head` exclusion and triggers an AssertionError.", - "confidence": 0.95, - "canonical_issue_reason": null, + "summary": "These are related only at a high level (both involve 4-bit inference), but they describe different bugs: one is a torch.finfo() failure on uint8 audio-module weights, the other is a BitsAndBytesConfig skip-module regression that drops the default lm_head exclusion and triggers an AssertionError.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 45674 is the better canonical representative because it is the more general, currently open regression with clearer reproduction around BitsAndBytesConfig behavior and 4-bit inference.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45674 is the more general and actionable report: it names a concrete BitsAndBytesConfig regression, states the observed assertion failure, and is framed in a way that is easier to map to a fix than the narrower Gemma4-specific audio-module TypeError.", + "best_issue_reason": "Issue 45674 is the best overall issue in the cluster: it is broader, more actionable, and likely to gather the most useful triage attention for the 4-bit inference path.", "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 a BitsAndBytesConfig skip-modules regression that clears `lm_head` exclusion and causes an AssertionError. Same general area, not the same fix." + "reason": "Reject: these appear to be distinct bugs with different symptoms and code paths. 45672 is a torch.finfo() TypeError on uint8 audio-module weights, while 45674 is a BitsAndBytesConfig skip-module handling regression that removes the default lm_head exclusion and causes an AssertionError." } ] }, @@ -17810,8 +17788,8 @@ "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": [ @@ -17821,7 +17799,7 @@ "soft_pairs": [ "issue:45762|issue:45763" ], - "prepared_review_unit_hash": "cde04dc22fae4edbfc9b7902a316cb3c1d7aa61e8411e532d6e578c81d402acf", + "prepared_review_unit_hash": "aeb748a30b21ab9a3edbacf594a998ee1d4142181cd086af6f8393d45addcc32", "component_budget": { "node_count": 2, "item_count": 2, @@ -17843,18 +17821,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are unrelated placeholder/spam-like reports with different personal names and dates; there is no evidence they describe the same bug or task.", - "confidence": 0.98, + "summary": "The two issues are not duplicates of the same underlying bug; they appear to be separate name/date entries with no technical context. The soft similarity is superficial only.", + "confidence": 0.93, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a good canonical representative because neither contains a real problem description or actionable context.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45762", "right": "issue:45763", "accept": false, - "reason": "Different named entities and dates with no shared technical context, code path, or problem statement; they do not appear to be the same underlying issue." + "reason": "Titles share a similar personal-name/date pattern, but there is no evidence they describe the same problem or request; they look like distinct entries." } ] }, @@ -17873,8 +17851,8 @@ "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": [ @@ -17884,7 +17862,7 @@ "soft_pairs": [ "issue:45764|issue:45765" ], - "prepared_review_unit_hash": "c396abe2d662c192ee161454d9b074c3fd11fe3167013d7c72a97b3ce7ecb29e", + "prepared_review_unit_hash": "f6a653532046d023650df77efcfc20b583ac910ec692a97d99e6cc5806e017a2", "component_budget": { "node_count": 2, "item_count": 2, @@ -17906,18 +17884,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are just title-only personal-name/date entries and do not describe the same underlying problem; the similarity is superficial only.", - "confidence": 0.98, - "canonical_issue_reason": "No clear canonical issue: both are unrelated, low-context entries with no substantive problem description.", + "summary": "These two closed issues only share a loose textual similarity in personal-name/date formatting and do not show any evidence of the same underlying bug or request.", + "confidence": 0.97, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "45764 is only marginally preferable as the earliest created issue, but it is not a meaningful representative of the pair.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45764", "right": "issue:45765", "accept": false, - "reason": "The titles share only a name/date pattern; there is no evidence they refer to the same bug, request, or code path." + "reason": "The titles refer to different people and different dates; there is no issue body, linkage, or code context indicating a shared defect or duplicate report." } ] }, @@ -17936,8 +17914,8 @@ "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-45810-2", "nodes": [ @@ -17947,7 +17925,7 @@ "soft_pairs": [ "issue:45810|issue:46295" ], - "prepared_review_unit_hash": "67356b72308b09767f276b3499a0f60c2a086f4d8ec0be629a5ac0e0cf0a579e", + "prepared_review_unit_hash": "3e8d6bf0b1330236a40e752aede04a703a87cef620dbf1666a0e1b32ebe3b08e", "component_budget": { "node_count": 2, "item_count": 2, @@ -17969,18 +17947,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are not duplicates: one requests a new token-classification head for Qwen3_5, while the other asks for FP8 training support. They touch different features and code paths.", + "summary": "The two issues are not duplicates: one asks for a new token-classification class for Qwen3.5, while the other requests FP8 training support. They touch different features and code paths.", "confidence": 0.97, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue 45810 is the more concrete and narrowly scoped request, so it is the better cluster representative, but it is not the same underlying problem as issue 46295.", + "best_issue_reason": "issue:46295 is the broader, more general feature request, but it is still unrelated to issue:45810 and should not be treated as a duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45810", "right": "issue:46295", "accept": false, - "reason": "Different feature requests with no shared concrete bug or implementation target: model head addition for token classification versus FP8 training support." + "reason": "Different underlying requests: model-specific token classification support versus FP8 training support. They do not describe the same bug or change." } ] }, @@ -17999,8 +17977,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45869-5", "nodes": [ @@ -18020,7 +17998,7 @@ "issue:45871|issue:45873", "issue:45872|issue:45873" ], - "prepared_review_unit_hash": "ca896772fffd2448fe667557ae093990a4a0554627487050933b3e2b4d627449", + "prepared_review_unit_hash": "ac47b6e4fb64cd07a9076b6f05ff817a1fd77993cdbdc64e724c4f5f14b55528", "component_budget": { "node_count": 5, "item_count": 5, @@ -18042,66 +18020,66 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The set splits into two exact-duplicate issue pairs: 45870/45871 and 45872/45873. Issue 45869 is related only at a broad financial-data theme level and should not be merged with the others.", - "confidence": 0.95, - "canonical_issue_reason": "45870 is the best canonical representative to keep for the first duplicate pair because 45870 and 45871 are identical, and 45870 is the earlier of the two.", + "summary": "This cluster is not a single duplicate set: it contains two exact-duplicate issue pairs about different DataForge examples, plus one unrelated dataset promotion issue. Accept the two exact-title duplicates and reject the cross-topic soft similarities.", + "confidence": 0.96, + "canonical_issue_reason": "If forced to pick one representative, issue #45870 is the cleanest canonical for the exact duplicate pair #45870/#45871; however, the cluster actually splits into two separate issue topics, so a single true canonical does not exist.", "canonical_pr_reason": null, - "best_issue_reason": "45870 is the strongest single issue to retain overall: it is an exact duplicate pair member, clearly representative of one concrete topic, and there is no PR evidence in the cluster.", + "best_issue_reason": "#45870 is the best issue candidate because it is the clearest, most general example issue and has an exact duplicate partner. Still, it should not absorb #45872/#45873 since those describe a different example/change.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45869", "right": "issue:45870", "accept": false, - "reason": "Related financial-domain wording, but one is a dataset bundle listing and the other is an example request; not the same underlying issue." + "reason": "Dataset bundle promo and a DataForge example are related only by theme, not the same underlying issue." }, { "left": "issue:45869", "right": "issue:45871", "accept": false, - "reason": "Same as 45869/45870: broad theme overlap only, with different concrete topic and no evidence of an exact duplicate." + "reason": "Same as above; no evidence they describe the same bug or change." }, { "left": "issue:45870", "right": "issue:45871", "accept": true, - "reason": "Exact title/body duplicate pair for the same example request." + "reason": "Exact same title and clearly the same example issue; this is a straightforward duplicate." }, { "left": "issue:45870", "right": "issue:45872", "accept": false, - "reason": "Both mention financial time series, but 45872 adds a scraping service component and is a different concrete request." + "reason": "Both are DataForge-related, but one is an example issue and the other is a separate time-series + scraping service topic." }, { "left": "issue:45870", "right": "issue:45873", "accept": false, - "reason": "Same as 45870/45872: overlapping domain, but different underlying change/request." + "reason": "Different concrete topic despite similar keywords; not the same underlying change." }, { "left": "issue:45871", "right": "issue:45872", "accept": false, - "reason": "Broad thematic similarity only; not the same issue or change." + "reason": "These are two distinct issue topics, not one duplicate bug/report." }, { "left": "issue:45871", "right": "issue:45873", "accept": false, - "reason": "Broad thematic similarity only; not the same issue or change." + "reason": "Same as above; similarity is only at the broad subsystem/topic level." }, { "left": "issue:45872", "right": "issue:45873", "accept": true, - "reason": "Exact duplicate pair with matching title and body length, describing the same request." + "reason": "Exact same title and evidently the same issue; strong duplicate match." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet: 45870/45871 and 45872/45873 are the only clear duplicate pairs, while 45869 is only soft-related by theme. The canonical choice of the earlier duplicate is conservative, and the soft-edge rejects do not overstate the evidence." + "feedback": "The summary is grounded in the packet: 45870/45871 are identical, 45872/45873 are identical, and 45869 is a distinct dataset-promo item. The canonicalization note is conservative, and the soft-edge verdicts do not overclaim beyond the available evidence." }, "evaluator_used": true, "retried": false, @@ -18140,8 +18118,8 @@ ], "soft_pairs": [ "pull_request:39895|pull_request:43424", - "pull_request:41855|pull_request:43424", "pull_request:41855|pull_request:43823", + "pull_request:41855|pull_request:43424", "pull_request:42668|pull_request:43823", "pull_request:43251|pull_request:43823", "pull_request:43251|pull_request:44827", @@ -18160,160 +18138,160 @@ "pull_request:46048|pull_request:46268", "pull_request:43251|pull_request:43254" ], - "prepared_review_unit_hash": "ce29ba6f70d1119876e0c5a140047f2bf57bd0ed7f83daee0b6f4c28cc9abe66", + "prepared_review_unit_hash": "4f3877877ddca4d741ef42f29ca2e42b06a6962e958c70b022fd4daf238ebb5c", "component_budget": { "node_count": 15, "item_count": 15, "soft_pair_count": 20, - "serialized_chars": 40311, - "estimated_input_tokens": 10078, - "estimated_eval_tokens": 20412 + "serialized_chars": 40304, + "estimated_input_tokens": 10076, + "estimated_eval_tokens": 20408 }, "budget": { "node_count": 15, "item_count": 15, "soft_pair_count": 20, - "serialized_chars": 40311, - "estimated_input_tokens": 10078, - "estimated_eval_tokens": 20412 + "serialized_chars": 40304, + "estimated_input_tokens": 10076, + "estimated_eval_tokens": 20408 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Most items are unrelated PRs spanning distinct models, docs, kernel routing, and loss utilities. The only true duplicate-like pair is the two fixed_cross_entropy PRs (43251/43254), which address the same loss helper and same issue target.", + "summary": "One clear duplicate cluster exists around `fixed_cross_entropy` kwargs forwarding (`43251` and `43254`). The rest are distinct model additions, tests, documentation, or unrelated infra changes and should remain separate.", "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43251 is the stronger representative for the loss-helper fix: it directly targets issue 43240, changes the same code path in loss_utils.py, and has substantially more review/discussion activity than 43254.", + "canonical_pr_reason": "`pull_request:43251` is the cleaner, more complete fix for the loss kwargs bug: it explicitly forwards `weight` and `label_smoothing` to `nn.functional.cross_entropy` and is linked to the tracked issue.", "best_issue_reason": null, - "best_pr_reason": "PR 43251 is the best single representative of the cluster because it is the clearest concrete bug fix with an explicit issue target and the only pair with strong duplicate evidence.", + "best_pr_reason": "`pull_request:43251` is the best representative of the cluster because it directly addresses the concrete cross-entropy kwargs omission and is the more polished implementation compared with `43254`.", "soft_edge_verdicts": [ { "left": "pull_request:39895", "right": "pull_request:43424", "accept": false, - "reason": "Different features in different areas: VideoPrism model addition vs. an Executorch dynamic-shape test. No shared code path or concrete duplicate bug." + "reason": "VideoPrism model addition and an executorch dynamic-shapes test are unrelated changes; same broad theme of testing/exportability is not enough." }, { "left": "pull_request:41855", - "right": "pull_request:43424", + "right": "pull_request:43823", "accept": false, - "reason": "Unrelated tokenizer method addition vs. Executorch export testing. Same broad repo, but not the same underlying change." + "reason": "Mistral tokenizer method additions and a new MobileLLM model implementation are unrelated code paths." }, { "left": "pull_request:41855", - "right": "pull_request:43823", + "right": "pull_request:43424", "accept": false, - "reason": "Mistral tokenizer methods and a new MobileLLM model are unrelated changes with no shared failing path." + "reason": "Tokenizer API additions and executorch export tests fix different problems in different subsystems." }, { "left": "pull_request:42668", "right": "pull_request:43823", "accept": false, - "reason": "Processor robustness fixes vs. adding MobileLLM; different models and different problems." + "reason": "Processor auto-loading robustness and adding MobileLLM are different changes; no shared concrete bug." }, { "left": "pull_request:43251", "right": "pull_request:43823", "accept": false, - "reason": "Loss-helper kwargs fix and MobileLLM model addition are unrelated." + "reason": "Cross-entropy kwargs forwarding and a new model implementation are unrelated." }, { "left": "pull_request:43251", "right": "pull_request:44827", "accept": false, - "reason": "The loss utility fix is unrelated to Mistral4 test/model changes." + "reason": "The loss helper fix and Mistral4 test/kernel changes do not target the same bug or code path." }, { "left": "pull_request:43424", "right": "pull_request:44827", "accept": false, - "reason": "Executorch dynamic-shape export test and Mistral4 fixes do not address the same code path." + "reason": "Executorch dynamic-shape testing and Mistral4 tests are unrelated test concerns, not the same underlying issue." }, { "left": "pull_request:43424", "right": "pull_request:43823", "accept": false, - "reason": "Executorch export testing and MobileLLM model implementation are different features." + "reason": "Dynamic-shape export tests and MobileLLM implementation are unrelated." }, { "left": "pull_request:43823", "right": "pull_request:45655", "accept": false, - "reason": "New MobileLLM model and config-resolution ordering fix are unrelated." + "reason": "A new model implementation and fixing `cls.config` resolution are unrelated changes." }, { "left": "pull_request:43823", "right": "pull_request:43995", "accept": false, - "reason": "MobileLLM model addition vs. Falcon output-collection refactor; no common bug or mergeable overlap." + "reason": "MobileLLM implementation and Falcon output-interface refactoring are different model-specific changes." }, { "left": "pull_request:43995", "right": "pull_request:45213", "accept": false, - "reason": "Falcon refactor and a model-creation skill are unrelated; one is code, the other is tooling/docs." + "reason": "Falcon refactoring and a model-creation skill are not the same bug/change." }, { "left": "pull_request:43995", "right": "pull_request:44827", "accept": false, - "reason": "Falcon output-interface refactor does not overlap with Mistral4 test fixes." + "reason": "Falcon output-collection refactor and Mistral4 test fixes do not fix the same concrete issue." }, { "left": "pull_request:44827", "right": "pull_request:45170", "accept": false, - "reason": "Mistral4 test fixes and layrnorm->layernorm renaming are different changes in different subsystems." + "reason": "Mistral4 test fixes and `layrnorm` spelling cleanup across CLIP-like models are unrelated." }, { "left": "pull_request:44827", "right": "pull_request:46267", "accept": false, - "reason": "Mistral4 fixes and chat-response parsing update are unrelated." + "reason": "Mistral4 test fixes and chat response parsing updates are completely different areas." }, { "left": "pull_request:44827", "right": "pull_request:46048", "accept": false, - "reason": "Mistral4 tests/fixes and TDT loss kernel integration are different concrete changes." + "reason": "Mistral4 test/integration changes and TDT loss kernel integration are unrelated." }, { "left": "pull_request:44827", "right": "pull_request:45655", "accept": false, - "reason": "Mistral4 test fixes and cls.config resolution ordering fix are unrelated." + "reason": "Mistral4 test changes and config-resolution order fixes are unrelated." }, { "left": "pull_request:44827", "right": "pull_request:45213", "accept": false, - "reason": "Mistral4 fixes and the model-creation skill are not the same underlying bug or feature." + "reason": "Model-specific test fixes and the model-creation skill draft are not the same underlying change." }, { "left": "pull_request:43823", "right": "pull_request:45213", "accept": false, - "reason": "MobileLLM model addition and the model-creation skill are unrelated." + "reason": "Adding MobileLLM and creating a model-generation skill are unrelated." }, { "left": "pull_request:46048", "right": "pull_request:46268", "accept": false, - "reason": "Different hub-kernel additions: TDT loss kernel vs. ROCm AITER rope routing. Same integration area, but not the same change." + "reason": "Both touch hub-kernel plumbing, but they target different kernels and different behaviors; they are not the same concrete change." }, { "left": "pull_request:43251", "right": "pull_request:43254", "accept": true, - "reason": "Both PRs modify fixed_cross_entropy in loss_utils.py for the same issue 43240 and the same concrete cross-entropy kwargs support, making them true duplicates/overlapping fixes." + "reason": "Both PRs fix the same `fixed_cross_entropy` kwargs-forwarding bug in the same file, and one could plausibly merge the other\u2019s intent into a single PR." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded and conservative overall. The summary matches the packet\u2019s evidence: most pairs are clearly unrelated by title/files, and the only mergeable same-bugfix overlap is 43251/43254, which share the loss_utils.py path and issue target 43240. No soft-edge verdicts seem overstated." + "feedback": "The cluster summary is well grounded in the packet: `43251` and `43254` are the only clear same-bug pair, both touching `src/transformers/loss/loss_utils.py` and sharing issue `43240`. The remaining soft-edge rejections are conservative and consistent with the diff previews. The only slightly subjective phrasing is the preference for `43251` as the cleaner representative, but it is still supported by the explicit kwargs-forwarding change." }, "evaluator_used": true, "retried": false, @@ -18340,7 +18318,7 @@ "soft_pairs": [ "pull_request:40546|pull_request:42781" ], - "prepared_review_unit_hash": "2a57226dae04814bf16dc85f32d2e22d993ca88ec95b447c1c2484e5968dd116", + "prepared_review_unit_hash": "c195102c5d8f4c12bd6a08e2fd3b0c0ecd5130c20ee5dc408aebab8cfdf44754", "component_budget": { "node_count": 2, "item_count": 2, @@ -18362,18 +18340,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs are about VibeVoice, but they are not duplicates: #40546 is the base VibeVoice integration, while #42781 adds VibeVoice Realtime plus acoustic-tokenizer/feature-extraction changes. They share model scaffolding, but the concrete changes and code paths differ.", - "confidence": 0.93, + "summary": "Both PRs target VibeVoice support, but they are adjacent, not duplicate, changes: #40546 implements core VibeVoice integration, while #42781 adds a separate realtime variant plus an acoustic tokenizer and related updates.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "#40546 is the broader, foundational VibeVoice implementation and the best anchor for the cluster.", + "canonical_pr_reason": "No true duplicate PR is evident. If one must serve as the representative, #42781 is the more complete and newer VibeVoice integration, but it extends the earlier work rather than fixing the same concrete code path.", "best_issue_reason": null, - "best_pr_reason": "#40546 is the most representative PR for the overall VibeVoice integration; #42781 is a later, more specialized realtime extension.", + "best_pr_reason": "#42781 is the strongest cluster representative because it covers the broader, newer VibeVoice integration surface and includes the extra realtime/tokenizer components, whereas #40546 is the initial core implementation.", "soft_edge_verdicts": [ { "left": "pull_request:40546", "right": "pull_request:42781", "accept": false, - "reason": "Related but not the same underlying change: #40546 implements initial VibeVoice support, whereas #42781 adds realtime-specific functionality and acoustic-tokenizer/feature-extraction work. They touch overlapping files, but the code-path changes are distinct and would not plausibly be merged as one PR." + "reason": "They overlap on VibeVoice support and several shared files, but #42781 adds realtime and acoustic-tokenizer work, so it is not the same concrete change and would not be merged as one PR." } ] }, @@ -18403,12 +18381,12 @@ "soft_pairs": [ "pull_request:40755|pull_request:41224" ], - "prepared_review_unit_hash": "bfd5f8bbc27ff260f016c80722903b8ef8eb1857881a4c9ebe239b82c8ae8324", + "prepared_review_unit_hash": "5fe4d439564b8f4da6ab04fdc8d78efe1df762eb02fb46c6ee560e811fd4c8e9", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4675, + "serialized_chars": 4674, "estimated_input_tokens": 1169, "estimated_eval_tokens": 2594 }, @@ -18416,7 +18394,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4675, + "serialized_chars": 4674, "estimated_input_tokens": 1169, "estimated_eval_tokens": 2594 }, @@ -18425,18 +18403,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two unrelated pull requests: one adds TimesFM covariate forecasting, the other adds DINOv3 image classification support. The similarity is only superficial (both are model feature additions), so they should not be clustered as duplicates.", + "summary": "The two pull requests are unrelated: one adds TimesFM covariate forecasting support, the other adds DINOv3 image-classification support. They touch different model families, code paths, and filenames, so they should not be clustered as duplicates.", "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the changes target different model families and different code paths, so there is no duplicate relationship to collapse into one representative PR.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "If a single representative is needed, PR 40755 is the broader standalone feature addition (docs, model code, utils, tests) and is easier to treat as the cluster\u2019s exemplar; PR 41224 is a separate model-specific classification support change.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:40755", "right": "pull_request:41224", "accept": false, - "reason": "Different features for different models (TimesFM covariates vs DINOv3 image classification). They do not fix the same code-path problem and would not plausibly be merged into one PR." + "reason": "Different underlying changes: TimesFM forecasting with covariates versus DINOv3 classification support. No shared bug or mergeable code-path overlap; they belong to separate PRs." } ] }, @@ -18531,12 +18509,12 @@ "pull_request:41797|pull_request:45186", "pull_request:45186|pull_request:45630" ], - "prepared_review_unit_hash": "68838b396e01cdc739de733c52570f8f5406fae3f1894940c1a468455e5cd16b", + "prepared_review_unit_hash": "b098863a567bbd3fbbc235ed1496d7fefa0a89f3a5280b0d8815f4746edd17b7", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 10435, + "serialized_chars": 10436, "estimated_input_tokens": 2609, "estimated_eval_tokens": 5474 }, @@ -18544,7 +18522,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 10435, + "serialized_chars": 10436, "estimated_input_tokens": 2609, "estimated_eval_tokens": 5474 }, @@ -18553,24 +18531,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three separate model-addition pull requests for different model families (DeepSeek OCR, Isaac, Kimi2-6). They overlap in Transformers boilerplate and docs registration, but they do not address the same underlying change or bug.", + "summary": "All three PRs are similar in structure because they add new model integrations, docs, auto-mappings, and tests, but they are for different models (DeepSeek OCR, Isaac, Kimi2-6) and do not fix the same underlying change.", "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: each PR introduces a different model implementation and related registration/docs, so they are not duplicates of one another.", + "canonical_pr_reason": "No canonical duplicate PR here: each PR introduces a distinct model integration with its own files and model-specific code paths.", "best_issue_reason": null, - "best_pr_reason": "If one PR must be used as the representative of the cluster, #45630 is the strongest candidate by maturity and discussion activity, but it is still not a duplicate-equivalent of the others.", + "best_pr_reason": "PR #45630 is the best representative only because it is the most comprehensive and recent of the three model-addition PRs, but it should not be treated as a duplicate of the others.", "soft_edge_verdicts": [ { "left": "pull_request:41797", "right": "pull_request:45186", "accept": false, - "reason": "Different model additions (DeepSeek OCR vs Isaac). Shared auto-registration/docs boilerplate is not enough to make them the same underlying change." + "reason": "Different target models and codebases: DeepSeek OCR vs Isaac. The overlap is just the standard new-model integration template, not the same concrete change." }, { "left": "pull_request:45186", "right": "pull_request:45630", "accept": false, - "reason": "Different model additions (Isaac vs Kimi2-6) with different model directories and code paths. They share framework plumbing, not a single fix or feature." + "reason": "Different models and different implementation targets: Isaac vs Kimi2-6. Shared auto/docs/test scaffolding is too broad to treat as the same PR." } ] }, @@ -18600,7 +18578,7 @@ "soft_pairs": [ "pull_request:42975|pull_request:42976" ], - "prepared_review_unit_hash": "ea34e46b433dab20cb9c2586351e1409e352cbf5c06cd82372fbe14dab2a6354", + "prepared_review_unit_hash": "9eb3835338f5f389a9d8c4c3841dd9b6adcb2de9b4bd6446cba5d3f8ee98df48", "component_budget": { "node_count": 2, "item_count": 2, @@ -18622,18 +18600,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two unrelated workflow-maintenance PRs: one broad GitHub Actions runner/version compatibility upgrade, the other a narrower docker/build-push-action bump. They overlap in a few workflows but do not fix the same underlying problem.", - "confidence": 0.94, + "summary": "Two separate workflow-maintenance PRs: one upgrades common GitHub Actions for Node 24 compatibility, the other bumps docker/build-push-action across Docker image workflows. They overlap in broad CI upkeep but target different actions and different failure modes.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 42975 is the broader and more central maintenance change, explicitly targeting Node 24 compatibility across many workflows; it better represents the cluster theme than the narrower docker-action version bump.", + "canonical_pr_reason": "No clear duplicate pair here; if one PR must be treated as the representative, PR #42975 is the broader GitHub Actions compatibility update across many workflows and the more central maintenance change.", "best_issue_reason": null, - "best_pr_reason": "PR 42975 is the best representative because it covers the larger, explicit compatibility upgrade across the workflow set.", + "best_pr_reason": "PR #42975 is the better representative because it touches more workflow files and addresses a concrete platform compatibility goal, while PR #42976 is a narrower docker-action version bump.", "soft_edge_verdicts": [ { "left": "pull_request:42975", "right": "pull_request:42976", "accept": false, - "reason": "Reject: these are different upgrades to different GitHub Actions dependencies. PR 42975 focuses on Node 24 compatibility for checkout/cache/upload-artifact and related workflow actions, while PR 42976 updates docker/build-push-action to v6. Shared workflow files are not enough to make them the same change." + "reason": "Both are GitHub Actions version updates, but they modify different actions (checkout/cache/upload-artifact vs docker/build-push-action) and fix different concerns. They could be merged independently and do not look like the same underlying change." } ] }, @@ -18670,12 +18648,12 @@ "pull_request:43448|pull_request:45490", "pull_request:43451|pull_request:45490" ], - "prepared_review_unit_hash": "e314897c3566c8ad0ddda52aef5968200273df440df2b723d3721b47128e97f6", + "prepared_review_unit_hash": "6da1e23d73484be7f7f11082c1355628621c9111a22617bc65278764412d43d6", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 14182, + "serialized_chars": 14184, "estimated_input_tokens": 3546, "estimated_eval_tokens": 7348 }, @@ -18683,7 +18661,7 @@ "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 14182, + "serialized_chars": 14184, "estimated_input_tokens": 3546, "estimated_eval_tokens": 7348 }, @@ -18692,48 +18670,48 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are four distinct model-addition pull requests with overlapping boilerplate (docs, auto mappings, tests) but different underlying models and code paths. They are not duplicates of one another.", - "confidence": 0.98, + "summary": "All four PRs are independent model-addition efforts. They share generic Transformers boilerplate (docs, auto mappings, toctree) but target different architectures and code paths, so none should be clustered as duplicates.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "No single canonical PR: each request adds a different model family (ViT NEPA, Molmo, Molmo2, CTSM) with separate configs, model files, processors, and tests.", + "canonical_pr_reason": "PR #43451 (Molmo2) is the strongest representative of the cluster by scope and discussion volume, with the broadest implementation surface across modeling, processing, video support, auto-mappings, and tests.", "best_issue_reason": null, - "best_pr_reason": "PR 43451 (Molmo2) is the broadest and most substantial of the set, but it is still not a good duplicate canonical because it targets a different model and implementation than the others.", + "best_pr_reason": "PR #43451 is the best single representative because it is the most substantial and complete model-addition change in the set.", "soft_edge_verdicts": [ { "left": "pull_request:42978", "right": "pull_request:43448", "accept": false, - "reason": "Different models and code paths: ViT NEPA vs Molmo. Shared docs/auto-registry changes are generic model-addition boilerplate, not the same change." + "reason": "ViT NEPA and Molmo are different model families with separate configs, model code, and tests; shared docs/auto-registration boilerplate is not enough to indicate the same underlying change." }, { "left": "pull_request:42978", "right": "pull_request:43451", "accept": false, - "reason": "Different model additions: ViT NEPA vs Molmo2. Similar framework files are touched, but the underlying implementation, configs, and tests are unrelated." + "reason": "ViT NEPA and Molmo2 are unrelated model additions affecting different architectures and code paths despite both touching Transformer registry/docs files." }, { "left": "pull_request:42978", "right": "pull_request:45490", "accept": false, - "reason": "CTSM time-series model addition is unrelated to ViT NEPA. Shared auto/docs updates do not indicate the same bug or feature." + "reason": "ViT NEPA and CTSM are distinct model implementations for different domains, and the overlap is only in generic integration files." }, { "left": "pull_request:43448", "right": "pull_request:43451", "accept": false, - "reason": "Molmo and Molmo2 are separate model versions with different architecture/processor files and tests; they are not one mergeable PR for the same concrete change." + "reason": "Molmo and Molmo2 are separate model variants with different processing and modeling stacks; they are not the same concrete bug or change and would not plausibly merge into one PR." }, { "left": "pull_request:43448", "right": "pull_request:45490", "accept": false, - "reason": "Molmo vs CTSM are different model families. The overlap is only in registry/docs plumbing, which is too broad to treat as a duplicate." + "reason": "Molmo and CTSM are unrelated model additions; the shared filenames are just registry/docs updates common to many PRs." }, { "left": "pull_request:43451", "right": "pull_request:45490", "accept": false, - "reason": "Molmo2 and CTSM address different models and code paths; shared auto-mapping and docs updates are incidental, not evidence of the same underlying change." + "reason": "Molmo2 and CTSM target different model architectures and implementation paths, so this is only broad subsystem overlap, not a duplicate fix/change." } ] }, @@ -18766,7 +18744,7 @@ "pull_request:43291|pull_request:44815", "pull_request:43665|pull_request:44815" ], - "prepared_review_unit_hash": "e89c62311ebb1685ed4a24d207c3609aa9de9f0c2e8835acbf527e13525d0e68", + "prepared_review_unit_hash": "76545cff41088254b3faa6d885cec3ebf3f6bb097f38c2143471c382e5b5c332", "component_budget": { "node_count": 3, "item_count": 3, @@ -18788,30 +18766,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are not true duplicates. PR #44815 is a substantive FP8 dequant/loading fix, PR #43291 is a Whisper tokenizer/test fix, and PR #43665 appears to be a test-only change with intentional failing asserts and no shared code path. The only apparent similarity is superficial.", - "confidence": 0.91, + "summary": "These three PRs are not duplicates of one another: one targets Whisper decoding/tests, one is just unrelated CLIP/ViT test edits, and one is a real FP8 dequant/model-loading fix. The shared similarity is only superficial.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #44815 is the most substantive and concrete fix in the set, with actual implementation changes in model loading and FP8 integration rather than test-only edits.", + "canonical_pr_reason": "PR #44815 is the only substantive code fix here; it addresses a concrete FP8 dequant/loading path and is much more representative than the test-only or test-noise PRs.", "best_issue_reason": null, - "best_pr_reason": "PR #44815 is the best representative of a real underlying change because it fixes a concrete code path and has the clearest production impact; the other PRs are either unrelated or mostly test scaffolding.", + "best_pr_reason": "PR #44815 is the best standalone PR in the set because it contains an actual implementation change with a clear bug fix, whereas #43291 and #43665 are test-focused and appear unrelated to the same underlying problem.", "soft_edge_verdicts": [ { "left": "pull_request:43291", "right": "pull_request:43665", "accept": false, - "reason": "Whisper tokenization/test cleanup and CLIP/ViT test edits are unrelated; no shared bug or code path, only generic test-focused similarity." + "reason": "Both touch tests, but they target different models and different problems: Whisper tokenizer/decode tests vs CLIP/ViT test assertions. No shared concrete bug or change path." }, { "left": "pull_request:43291", "right": "pull_request:44815", "accept": false, - "reason": "One is Whisper decode behavior, the other is FP8/model-loading dequant logic; different subsystems and different failure modes." + "reason": "Whisper tokenization/test changes are unrelated to FP8 dequantization and model-loading logic. Different subsystem, different bug, no plausible single merged fix." }, { "left": "pull_request:43665", "right": "pull_request:44815", "accept": false, - "reason": "The similarity is only superficial. #43665 is test-only and does not fix the same concrete problem as the FP8 dequant/loading changes in #44815." + "reason": "CLIP/ViT test-only edits do not match the FP8 dequant/loading implementation fix. They are not the same underlying change and should not be merged as duplicates." } ] }, @@ -18848,64 +18826,64 @@ "pull_request:45932|pull_request:46100", "pull_request:46100|pull_request:46114" ], - "prepared_review_unit_hash": "9295b82d381f86618d1c69d3516a7bdd86765fc7e7996f7b85e146d6851b848f", + "prepared_review_unit_hash": "5c8d59a11221d973edb8b9fff8c86e97513d815c6ed79f307b54267859bd05b5", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 5, - "serialized_chars": 12260, - "estimated_input_tokens": 3065, - "estimated_eval_tokens": 6386 + "serialized_chars": 12263, + "estimated_input_tokens": 3066, + "estimated_eval_tokens": 6388 }, "budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 5, - "serialized_chars": 12260, - "estimated_input_tokens": 3065, - "estimated_eval_tokens": 6386 + "serialized_chars": 12263, + "estimated_input_tokens": 3066, + "estimated_eval_tokens": 6388 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are mostly unrelated \"don't merge\" test/fake-change PRs; they share superficial CI/test-file overlap but not the same concrete bug or code change. No soft edge should be accepted.", - "confidence": 0.94, + "summary": "These are mostly unrelated synthetic \"do not merge\" PRs used to test bot/CI behavior. The only mildly similar pair (45932, 46100) still targets different test-fixture goals, so I do not treat any pair as true duplicates.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "No exact duplicate PR exists here; if one PR must anchor the cluster, #46114 is the broadest CI/test-migration style change and overlaps most with the other noisy test-focused PRs.", + "canonical_pr_reason": "PR 46114 is the broadest representative of the cluster\u2019s synthetic test/migration theme, with CI, model, and test-file changes that best summarize the fixture-like nature of the set.", "best_issue_reason": null, - "best_pr_reason": "#46114 is the best representative only in a loose sense because it touches several CI/test files and sits near the center of the superficial similarity graph, but it is not a true duplicate of the others.", + "best_pr_reason": "PR 46114 is the best global representative because it spans the widest mix of CI and test scaffolding among the dummy PRs, making it the clearest canonical example of the cluster.", "soft_edge_verdicts": [ { "left": "pull_request:43488", "right": "pull_request:46114", "accept": false, - "reason": "Both are test/bot-style PRs, but #43488 is a setup/version/dependency tweak while #46114 is a wider CI/test-migration change; different underlying changes." + "reason": "Both are synthetic test PRs, but they touch different scaffolding: 43488 is a repo-bot/version/setup tweak around ViT tests, while 46114 is a CI/OTEL migration fixture. Not the same underlying change." }, { "left": "pull_request:43488", "right": "pull_request:43532", "accept": false, - "reason": "#43488 changes packaging/version and pytest deps, while #43532 rewires CI workflows and test utilities; they do not fix the same code-path problem." + "reason": "Shared Vit test file is incidental; 43488 is setup/version plumbing, while 43532 rewires comment-driven workflow and notification logic. Different dummy changes." }, { "left": "pull_request:43532", "right": "pull_request:46114", "accept": false, - "reason": "Shared test/CI themes and one overlapping test file are too broad; the actual edits are different workflow/utilities changes versus unrelated test migration and model/CI tweaks." + "reason": "Both are bot/CI-related, but one is a self-comment workflow rewrite and the other is a forked migration/OTEL test PR. They do not fix the same code-path problem." }, { "left": "pull_request:45932", "right": "pull_request:46100", "accept": false, - "reason": "#45932 is a broad model/CI instrumentation change, while #46100 only injects failing assertions into ViT tests; not the same concrete bug or fix." + "reason": "Both are test-oriented, but 45932 changes multiple CI/model files for a Grafana/OTEL-style fixture, whereas 46100 only injects failing assertions into Vit tests. Not mergeable as one PR." }, { "left": "pull_request:46100", "right": "pull_request:46114", "accept": false, - "reason": "Both touch test code, but #46100 is just a small ViT test mutation and #46114 is a larger CI/test-migration PR with unrelated model/config changes." + "reason": "Shared test-labelling is too broad; 46100 is a narrow Vit test-migration stub, while 46114 is a wider CI/model/test migration fixture. Different underlying changes." } ] }, @@ -18935,12 +18913,12 @@ "soft_pairs": [ "pull_request:43612|pull_request:43613" ], - "prepared_review_unit_hash": "355ef9e5e48f8886effdbb47f5846b976686cd38ed3345515ea3ef77ad6ea54c", + "prepared_review_unit_hash": "a1d6914a2f6a41f8edc1e78e6a77d9a421ce987b54491c07fe3b03917d2f44bf", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6083, + "serialized_chars": 6082, "estimated_input_tokens": 1521, "estimated_eval_tokens": 3298 }, @@ -18948,7 +18926,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6083, + "serialized_chars": 6082, "estimated_input_tokens": 1521, "estimated_eval_tokens": 3298 }, @@ -18957,18 +18935,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two separate pipeline-addition PRs in the same segmentation area, but they implement different tasks and integrate different model families. They are similar in structure, not the same underlying change.", - "confidence": 0.96, + "summary": "These are related but distinct pull requests for different segmentation pipelines: one adds Promptable Concept Segmentation, the other adds Promptable Visual Segmentation. Their overlap is mostly shared documentation and registration plumbing, not the same underlying bug or change.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "No true canonical duplicate here; if forced to pick one standalone reference, #43613 is the broader/more complete PR because it wires in more model integrations and tests, but it is not a substitute for #43612.", + "canonical_pr_reason": "No canonical PR as a duplicate target: the two PRs implement different pipelines with different model/support files and different concrete behavior.", "best_issue_reason": null, - "best_pr_reason": "#43613 is the stronger standalone PR example: it adds promptable visual segmentation across multiple model files and has a wider test surface. #43612 is a separate feature for promptable concept segmentation.", + "best_pr_reason": "PR 43613 is the broader representative of the pair because it adds the more expansive Promptable Visual Segmentation pipeline and touches more related model/tests/docs, but it is still a separate change from 43612.", "soft_edge_verdicts": [ { "left": "pull_request:43612", "right": "pull_request:43613", "accept": false, - "reason": "Different features: promptable concept segmentation vs promptable visual segmentation. They share scaffolding/docs, but the changed model files and pipeline code target different tasks and code paths, so they should not be merged as duplicates." + "reason": "Reject: they target different pipeline features and different code paths/models. Shared docs, pipeline registration, and metadata updates are incidental and not enough to treat them as one mergeable PR." } ] }, @@ -18998,12 +18976,12 @@ "soft_pairs": [ "pull_request:43649|pull_request:46301" ], - "prepared_review_unit_hash": "c82b7cc6ba93c9e67f03cebd882bb7a33d782205ee67ddd3e1d89a3890164e2a", + "prepared_review_unit_hash": "fa56513447ac7cdc943a0f517b73ee28337e9a3979c00f844c23feb3d0720a26", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4549, + "serialized_chars": 4550, "estimated_input_tokens": 1138, "estimated_eval_tokens": 2532 }, @@ -19011,7 +18989,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4549, + "serialized_chars": 4550, "estimated_input_tokens": 1138, "estimated_eval_tokens": 2532 }, @@ -19020,18 +18998,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two PRs are unrelated: one revises self-comment CI / failure-reporting workflows, while the other changes CircleCI job markers and tensor-parallel test handling. There is no sign they address the same underlying bug or could be merged into one PR.", - "confidence": 0.97, + "summary": "These two pull requests are not duplicates: one is about CI/self-comment failure reporting workflows, while the other changes CircleCI test selection and tensor-parallel test handling. They touch different code paths and would not plausibly be merged into one PR.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the PRs target different subsystems and different changesets, so neither is a duplicate of the other.", + "canonical_pr_reason": "No canonical PR: the changes address unrelated areas (workflow/reporting automation vs. CircleCI marker/test behavior) and do not fix the same concrete bug or implement the same change.", "best_issue_reason": null, - "best_pr_reason": "No best PR for deduplication purposes; each PR is specific to a different code-path problem and neither serves as a representative duplicate target.", + "best_pr_reason": "PR 43649 is the better representative only because it has a clearer, coherent title and a more self-contained CI/reporting theme; PR 46301 is very generic ('check') and mixes CircleCI config with tensor-parallel test adjustments.", "soft_edge_verdicts": [ { "left": "pull_request:43649", "right": "pull_request:46301", "accept": false, - "reason": "Different changes: 43649 touches self-comment CI, notification, and model-test reporting; 46301 changes CircleCI selection and tensor-parallel test infrastructure. They do not fix the same bug/change." + "reason": "Different underlying changes: 43649 modifies workflow/comment-based failure reporting, while 46301 adjusts CircleCI job markers and tensor-parallel test execution. They are not the same bug/change." } ] }, @@ -19061,40 +19039,40 @@ "soft_pairs": [ "pull_request:43651|pull_request:43663" ], - "prepared_review_unit_hash": "7d57f1ea86c7bede16a922717759a2882f4e638bc8b973e019c8b25d2e30c9a1", + "prepared_review_unit_hash": "5e68d8c7ae6bf657bb26c8831a86307bebedba11bc8cd0b5e75b0a7076a5b27e", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4293, - "estimated_input_tokens": 1074, - "estimated_eval_tokens": 2404 + "serialized_chars": 4292, + "estimated_input_tokens": 1073, + "estimated_eval_tokens": 2402 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4293, - "estimated_input_tokens": 1074, - "estimated_eval_tokens": 2404 + "serialized_chars": 4292, + "estimated_input_tokens": 1073, + "estimated_eval_tokens": 2402 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs modify `trainer.py` to add extensibility hooks, but they address different code paths: #43651 changes gradient-accumulation loss scaling, while #43663 refactors signature-column filtering. They are not duplicates and should stay separate.", - "confidence": 0.97, + "summary": "Two independent Trainer enhancements in the same file: one changes gradient-accumulation loss scaling, the other factors out signature-column filtering for custom trainers/PEFT. They do not look like the same bug or mergeable change.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "No issue is present. As a cluster representative, #43663 is the cleaner API refactor because it extracts a reusable override point for column filtering.", + "canonical_pr_reason": "No issue is present. Among the PRs, 43651 is the more canonical representative because it changes a core training-path behavior, has more discussion activity, and is the more substantive trainer semantics change.", "best_issue_reason": null, - "best_pr_reason": "#43663 is the better representative PR for this cluster because it is a self-contained Trainer extensibility change with a clear override hook and broader utility.", + "best_pr_reason": "43651 is the best representative PR for the cluster because it touches the main training_step loss-scaling path and has broader behavioral impact than the column-filtering refactor in 43663.", "soft_edge_verdicts": [ { "left": "pull_request:43651", "right": "pull_request:43663", "accept": false, - "reason": "They change different behavior in `Trainer`: one alters loss normalization for gradient accumulation, the other changes how signature columns are computed for data filtering. Same file and similar extensibility intent, but not the same underlying bug or mergeable PR." + "reason": "Different changes in the same subsystem but not the same underlying problem: 43651 is about gradient accumulation loss scaling, while 43663 is about extracting signature-column filtering for custom trainers/PEFT. They would not plausibly be merged into one PR." } ] }, @@ -19124,12 +19102,12 @@ "soft_pairs": [ "pull_request:43656|pull_request:43842" ], - "prepared_review_unit_hash": "9552c440d910cc5926882d779891b9520e4298a97300971354ffce2b9deda804", + "prepared_review_unit_hash": "7e972a73e3f88f477ca7ecb3ef80ec3135f5e95b7f3f7811792a32a5478b9d38", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 3664, + "serialized_chars": 3663, "estimated_input_tokens": 916, "estimated_eval_tokens": 2088 }, @@ -19137,7 +19115,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 3664, + "serialized_chars": 3663, "estimated_input_tokens": 916, "estimated_eval_tokens": 2088 }, @@ -19146,18 +19124,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs touch `src/transformers/cli/serve.py` and mention `TypeAdapter`, but they target different failure modes: one is a broader type-annotation/runtime typing cleanup, the other specifically fixes `TypeAdapter` being unavailable when pydantic is not installed. They should not be merged as duplicates.", - "confidence": 0.88, + "summary": "Two PRs touch the same file and both mention a TypeAdapter NameError in the CLI, but they appear to address different underlying fixes rather than the same exact bug/change.", + "confidence": 0.82, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #43842 is the better canonical PR: it directly addresses the concrete `TypeAdapter`/pydantic availability NameError with a minimal, targeted fix and an explicit bug description.", + "canonical_pr_reason": "PR 43656 is the broader, more central fix in src/transformers/cli/serve.py and appears to address the annotation-evaluation NameError directly, with wider typing cleanup. PR 43842 is a narrower follow-up focused on the optional pydantic import path.", "best_issue_reason": null, - "best_pr_reason": "PR #43842 is the strongest representative because its change is narrowly scoped to the runtime `TypeAdapter` import/annotation problem and is clearly tied to a specific user-facing failure.", + "best_pr_reason": "PR 43656 is the best representative of the cluster because it is the more complete change to the CLI typing/runtime-annotation problem, while PR 43842 is a targeted variant for a specific dependency scenario.", "soft_edge_verdicts": [ { "left": "pull_request:43656", "right": "pull_request:43842", "accept": false, - "reason": "They overlap in file and type-annotation concerns, but the fixes are not the same underlying bug. #43842 guards `TypeAdapter` when pydantic is absent; #43656 appears to be a broader annotation/runtime typing adjustment and could be merged independently." + "reason": "Same file and similar symptom, but the fixes are not the same concrete change: 43656 adds future-annotations/typing cleanup, while 43842 explicitly moves TypeAdapter under TYPE_CHECKING and quotes the annotation to handle the pydantic-absent case. These are related but distinct solutions and not clearly mergeable into one PR." } ] }, @@ -19349,20 +19327,20 @@ "pull_request:44044|pull_request:44129", "pull_request:44044|pull_request:44154" ], - "prepared_review_unit_hash": "36f8038c6e6fd28681e2b324c7f3dd51bfcfe187c2895f42024a819f740edd9b", + "prepared_review_unit_hash": "2de69f7eaaed316090d1bf8248f1c5198acfd6a4c2ad347fe7bc15d1d760ad83", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 78, - "serialized_chars": 84900, - "estimated_input_tokens": 21225, - "estimated_eval_tokens": 42706 + "serialized_chars": 84896, + "estimated_input_tokens": 21224, + "estimated_eval_tokens": 42704 }, "budget": { "node_count": 27, "item_count": 27, "soft_pair_count": 72, - "serialized_chars": 75096, + "serialized_chars": 75095, "estimated_input_tokens": 18774, "estimated_eval_tokens": 37804 }, @@ -19371,288 +19349,240 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a heterogeneous cluster of output-tracing refactors across unrelated model families. The only clear duplicate subclusters are GPT-Neo (44018/44068) and GPT-J/CodeGen (44066/44085/44722), with 44722 looking like the broadest later GPT-J/CodeGen representative.", - "confidence": 0.76, - "canonical_issue_reason": "No standalone issue artifact is present in the cluster; #43979 is only a shared tracker for 10/27 PRs, so it is the closest cluster-level issue reference but not a majority signal.", - "canonical_pr_reason": "PR 44722 is the best canonical PR because it is the broadest and cleanest GPT-J/CodeGen standardized-output refactor, and it subsumes the narrower GPT-J-only changes seen in 44066 and 44085.", + "summary": "This is mostly a thematically similar wave of output-tracing refactors across many different models, not a single duplicate PR. The clearest true duplicates are the GPT-J/CodeGen revisions and the GPT-Neo revision pair; the rest are model-specific changes that should stay separate.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44722 is the cleanest representative of the main true-duplicate subcluster: it targets the same GPT-J/CodeGen output-tracing refactor, has an explicit tracking issue, and is the broadest/latest version of that concrete change.", "best_issue_reason": null, - "best_pr_reason": "44722 is the strongest representative PR for the GPT-J/CodeGen duplicate subcluster: same output-tracing refactor, same core code path, and later/broader than the narrower GPT-J-only PRs.", + "best_pr_reason": "PR 44722 is the best anchor for deduping the GPT-J/CodeGen subcluster because it addresses the same concrete code-path change with the most complete scope among the duplicate candidates.", "soft_edge_verdicts": [ { "left": "pull_request:43996", "right": "pull_request:44085", "accept": false, - "reason": "Different model families and code paths; same tracking issue is not enough." + "reason": "Different models and code paths (CVT/FNet vs GPT-J); same issue target alone is not enough." }, { "left": "pull_request:43996", "right": "pull_request:44044", "accept": false, - "reason": "Unrelated models; both are output-tracing refactors, but not the same bug or change." + "reason": "Different model families and refactor scopes (CVT/FNet vs DeBERTa)." }, { "left": "pull_request:44066", "right": "pull_request:44085", "accept": true, - "reason": "Both are GPT-J output-tracing refactors in modeling_gptj.py, so they look like the same change split across PRs." + "reason": "Same GPT-J output-tracing refactor in the same model file; these are overlapping revisions of the same concrete change." }, { "left": "pull_request:44007", "right": "pull_request:44072", "accept": false, - "reason": "ResNet and EfficientNet are different model implementations; same refactor pattern only." + "reason": "Both are output-tracing refactors, but for different CNN architectures and separate code paths." }, { "left": "pull_request:44072", "right": "pull_request:44722", "accept": false, - "reason": "EfficientNet and GPT-J/CodeGen are unrelated code paths." + "reason": "EfficientNet and GPT-J are unrelated model code paths; broad theme overlap is not a duplicate." }, { "left": "pull_request:44066", "right": "pull_request:44072", "accept": false, - "reason": "GPT-J/CodeGen vs EfficientNet; only the broad refactor theme overlaps." + "reason": "GPT-J/CodeGen refactor vs EfficientNet refactor; different underlying change." }, { "left": "pull_request:44013", "right": "pull_request:44044", "accept": false, - "reason": "MobileNetV2 and DeBERTa are unrelated model paths." + "reason": "MobileNetV2 and DeBERTa are unrelated implementations." }, { "left": "pull_request:44066", "right": "pull_request:44086", "accept": false, - "reason": "GPT-J/CodeGen and MGP-STR are different model code paths." + "reason": "GPT-J/CodeGen output tracing vs MGP-STR output tracing; different model-specific fixes." }, { "left": "pull_request:44066", "right": "pull_request:44071", "accept": false, - "reason": "GPT-J/CodeGen and MPT are separate model implementations." + "reason": "GPT-J/CodeGen and MPT are distinct model code paths." }, { "left": "pull_request:44018", "right": "pull_request:44068", "accept": true, - "reason": "Both are GPT-Neo output-tracing refactors on the same model file/test surface and appear to be the same underlying change." + "reason": "Same GPT-Neo output-tracing refactor in the same model file; looks like duplicate revisions of one change." }, { "left": "pull_request:44018", "right": "pull_request:44066", "accept": false, - "reason": "GPT-Neo is unrelated to GPT-J/CodeGen." + "reason": "GPT-Neo vs GPT-J/CodeGen; same refactor pattern but different models." }, { "left": "pull_request:44066", "right": "pull_request:44068", "accept": false, - "reason": "GPT-J/CodeGen and GPT-Neo are different model families." + "reason": "GPT-J/CodeGen vs GPT-Neo; related theme, not the same code-path change." }, { "left": "pull_request:44019", "right": "pull_request:44722", "accept": false, - "reason": "ResNet and GPT-J/CodeGen are unrelated code paths." + "reason": "ResNet refactor vs GPT-J/CodeGen refactor; different model code paths." }, { "left": "pull_request:44086", "right": "pull_request:44722", "accept": false, - "reason": "MGP-STR and GPT-J/CodeGen are different models and different fixes." + "reason": "MGP-STR and GPT-J/CodeGen are unrelated model implementations." }, { "left": "pull_request:44019", "right": "pull_request:44085", "accept": false, - "reason": "ResNet and GPT-J are unrelated; only the general refactor style is shared." + "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 model implementations." + "reason": "GPT-Neo and ResNet are unrelated code paths." }, { "left": "pull_request:44019", "right": "pull_request:44071", "accept": false, - "reason": "ResNet and MPT are unrelated model paths." + "reason": "ResNet and MPT are different models; only the general refactor theme overlaps." }, { "left": "pull_request:44068", "right": "pull_request:44116", "accept": false, - "reason": "GPT-Neo and Flaubert/XLM are unrelated." + "reason": "GPT-Neo vs Flaubert; different model-specific output tracing changes." }, { "left": "pull_request:44018", "right": "pull_request:44116", "accept": false, - "reason": "GPT-Neo and Flaubert/XLM are unrelated." + "reason": "GPT-Neo and Flaubert are unrelated implementations." }, { "left": "pull_request:44068", "right": "pull_request:44722", "accept": false, - "reason": "GPT-Neo and GPT-J/CodeGen are different model families." + "reason": "GPT-Neo and GPT-J/CodeGen are separate model code paths." }, { "left": "pull_request:44071", "right": "pull_request:44722", "accept": false, - "reason": "MPT and GPT-J/CodeGen are unrelated code paths." + "reason": "MPT and GPT-J/CodeGen are different models." }, { "left": "pull_request:44018", "right": "pull_request:44085", "accept": false, - "reason": "GPT-Neo and GPT-J are different model code paths." + "reason": "GPT-Neo and GPT-J are different model-specific refactors." }, { "left": "pull_request:44018", "right": "pull_request:44086", "accept": false, - "reason": "GPT-Neo and MGP-STR are unrelated." + "reason": "GPT-Neo vs MGP-STR; not the same underlying bug or change." }, { "left": "pull_request:44068", "right": "pull_request:44086", "accept": false, - "reason": "GPT-Neo and MGP-STR are unrelated." + "reason": "GPT-Neo and MGP-STR are unrelated code paths." }, { "left": "pull_request:44071", - "right": "pull_request:44085", + "right": "pull_request:44086", "accept": false, - "reason": "MPT and GPT-J are unrelated model families." + "reason": "MPT and MGP-STR are distinct model implementations." }, { "left": "pull_request:44059", "right": "pull_request:44086", "accept": false, - "reason": "GPT2 and MGP-STR are unrelated." + "reason": "GPT-2 and MGP-STR are unrelated model code paths." }, { "left": "pull_request:44085", "right": "pull_request:44086", "accept": false, - "reason": "Different model families; same high-level refactor theme only." + "reason": "GPT-J and MGP-STR are different models; same output-tracing theme is too broad." }, { "left": "pull_request:44098", "right": "pull_request:44101", "accept": false, - "reason": "ViLT and Flaubert/XLM are unrelated model paths." + "reason": "ViLT and Flaubert are different model families." }, { "left": "pull_request:44059", "right": "pull_request:44722", "accept": false, - "reason": "GPT2 and GPT-J/CodeGen are separate implementations." + "reason": "GPT-2 vs GPT-J/CodeGen are separate code paths." }, { "left": "pull_request:44017", "right": "pull_request:44722", "accept": false, - "reason": "Segformer and GPT-J/CodeGen are unrelated." + "reason": "SegFormer and GPT-J/CodeGen are unrelated model changes." }, { "left": "pull_request:44017", "right": "pull_request:44161", "accept": false, - "reason": "Segformer and LongT5 are different model families." + "reason": "SegFormer and LongT5 are different model-specific refactors." }, { "left": "pull_request:44017", "right": "pull_request:44073", "accept": false, - "reason": "Segformer and VisualBert are unrelated." - }, - { - "left": "pull_request:43997", - "right": "pull_request:44073", - "accept": false, - "reason": "RegNet and VisualBert are unrelated." + "reason": "SegFormer and VisualBERT are unrelated code paths." }, { "left": "pull_request:43997", "right": "pull_request:44056", "accept": false, - "reason": "RegNet and MPNet are unrelated." - }, - { - "left": "pull_request:44056", - "right": "pull_request:44161", - "accept": false, - "reason": "MPNet and LongT5 are different code paths." - }, - { - "left": "pull_request:43997", - "right": "pull_request:44018", - "accept": false, - "reason": "RegNet and GPT-Neo are unrelated." - }, - { - "left": "pull_request:44010", - "right": "pull_request:44071", - "accept": false, - "reason": "SqueezeBERT and MPT are unrelated." - }, - { - "left": "pull_request:44074", - "right": "pull_request:44101", - "accept": false, - "reason": "TextNet and Flaubert/XLM are unrelated." + "reason": "RegNet and MPNet are different model implementations." }, { "left": "pull_request:43997", "right": "pull_request:44161", "accept": false, - "reason": "RegNet and LongT5 are unrelated." + "reason": "RegNet and LongT5 are unrelated model-specific changes." }, { - "left": "pull_request:44074", - "right": "pull_request:44161", - "accept": false, - "reason": "TextNet and LongT5 are unrelated." - }, - { - "left": "pull_request:44018", + "left": "pull_request:44056", "right": "pull_request:44161", "accept": false, - "reason": "GPT-Neo and LongT5 are unrelated." + "reason": "MPNet and LongT5 are different code paths." }, { "left": "pull_request:43997", - "right": "pull_request:44071", - "accept": false, - "reason": "RegNet and MPT are unrelated." - }, - { - "left": "pull_request:44010", - "right": "pull_request:44044", - "accept": false, - "reason": "SqueezeBERT and DeBERTa are unrelated." - }, - { - "left": "pull_request:44044", - "right": "pull_request:44074", + "right": "pull_request:44017", "accept": false, - "reason": "DeBERTa and TextNet are unrelated." + "reason": "RegNet and SegFormer are distinct models." }, { "left": "pull_request:44066", "right": "pull_request:44722", "accept": true, - "reason": "Both target the same GPT-J/CodeGen standardized output-tracing refactor; 44722 looks like the later broader version." + "reason": "Same GPT-J/CodeGen output-tracing refactor with the same concrete files; 44722 is a later, broader revision of the same change." }, { "left": "pull_request:44085", "right": "pull_request:44722", "accept": true, - "reason": "Both touch GPT-J output tracing in modeling_gptj.py, and 44722 is a plausible superset that also folds in CodeGen." + "reason": "Same GPT-J output-tracing change in the same code path; 44722 subsumes this narrower revision." }, { "left": "pull_request:44085", @@ -19664,34 +19594,34 @@ "left": "pull_request:43996", "right": "pull_request:44007", "accept": false, - "reason": "FNet/CVT and ResNet are unrelated model implementations." + "reason": "CVT/FNet vs RegNet/ResNet; same refactor theme but not the same underlying change." }, { "left": "pull_request:43996", "right": "pull_request:44129", "accept": false, - "reason": "FNet/CVT and SpeechT5 are unrelated." + "reason": "CVT/FNet and SpeechT5 are different model implementations." }, { "left": "pull_request:44044", "right": "pull_request:44129", "accept": false, - "reason": "DeBERTa and SpeechT5 are unrelated." + "reason": "DeBERTa and SpeechT5 are unrelated model code paths." }, { "left": "pull_request:44044", "right": "pull_request:44154", "accept": false, - "reason": "DeBERTa and VITS are different model families." + "reason": "DeBERTa and VITS are different models; not mergeable as one duplicate PR." } ] }, "evaluator_result": { - "accept": false, - "feedback": "Rejecting because the cluster summary overstates the evidence for the issue link: #43979 is not referenced by 'most' PRs in the packet, only a minority of them (10/27). The rest of the reasoning is mostly conservative and the 44066\u219444722 duplicate call is grounded, but the canonical_issue_reason is not precise enough as written." + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative about the non-duplicate pairs. The accepted duplicate pairs are plausibly mergeable as revisions of the same GPT-J/CodeGen and GPT-Neo fixes, and the rest are correctly treated as model-specific changes rather than one shared bugfix. The only minor caveat is that phrases like \"most complete scope\" are somewhat interpretive, but not overstated enough to reject." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -19726,70 +19656,70 @@ "pull_request:43997|pull_request:44010", "pull_request:44010|pull_request:44129" ], - "prepared_review_unit_hash": "8a8f6ea5a677e6a15cbb4d04c7a1bfe435d2ebf5c3411cd7583bf9a3c45a1025", + "prepared_review_unit_hash": "0a504e473fc0ff7034db79e7708bc1ed08a1ca619939031f4b5f0f75436fea71", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 78, - "serialized_chars": 84900, - "estimated_input_tokens": 21225, - "estimated_eval_tokens": 42706 + "serialized_chars": 84896, + "estimated_input_tokens": 21224, + "estimated_eval_tokens": 42704 }, "budget": { "node_count": 8, "item_count": 8, "soft_pair_count": 6, - "serialized_chars": 17125, - "estimated_input_tokens": 4282, - "estimated_eval_tokens": 8820 + "serialized_chars": 17122, + "estimated_input_tokens": 4281, + "estimated_eval_tokens": 8818 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These are not duplicate PRs; they are a cluster of similar standardized-output-tracing refactors applied to different model implementations. They share a common pattern and tooling, but each PR touches a separate code path and model-specific output logic.", + "summary": "This cluster is a set of model-specific refactors to standardized output tracing/capture_outputs across different architectures, not duplicate fixes of the same underlying bug or change. They share a theme but target different code paths and would not be merged as one PR.", "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44129 is the strongest canonical anchor because it is the only one with an explicit issue target and a concrete, model-specific refactor scope.", + "canonical_pr_reason": "No single PR is a true duplicate canonical for the cluster; the closest representative is the most complete standardized-output refactor among the model-specific changes.", "best_issue_reason": null, - "best_pr_reason": "PR 44129 is the most clearly scoped and traceable change in the set, with an explicit issue link and a substantial output-tracing refactor in one model.", + "best_pr_reason": "PR 44073 is the clearest representative of the shared pattern because it explicitly wires output-capture metadata and migrates a model to the standardized tracing decorators in a self-contained way.", "soft_edge_verdicts": [ { "left": "pull_request:44024", "right": "pull_request:44076", "accept": false, - "reason": "Both are output-tracing refactors, but they target different models (FocalNet vs ImageGPT) and different forward paths; not the same underlying bug/change." + "reason": "Different models (FocalNet vs ImageGPT) and different implementation details. Same refactor theme, but not the same concrete code-path change." }, { "left": "pull_request:44073", "right": "pull_request:44074", "accept": false, - "reason": "VisualBert and TextNet are separate model implementations; this is the same refactor pattern, not the same concrete code-path issue." + "reason": "Both are output-tracing refactors, but they affect different architectures and separate model logic; not mergeable as one PR." }, { "left": "pull_request:44056", "right": "pull_request:44073", "accept": false, - "reason": "MPNet and VisualBert have distinct encoder/attention code paths; shared standardized-output refactoring alone is not enough to merge them as duplicates." + "reason": "MPNet and VisualBert are distinct model implementations. Shared capture_outputs plumbing is a pattern, not a duplicate change." }, { "left": "pull_request:44056", "right": "pull_request:44074", "accept": false, - "reason": "These affect different models and different output-capture wiring; they are related maintenance changes, not the same fix." + "reason": "Different models and different forward/output handling. Similar modernization, but not the same bug or change." }, { "left": "pull_request:43997", "right": "pull_request:44010", "accept": false, - "reason": "RegNet and SqueezeBert are unrelated model paths; the PRs share a decorator migration theme but not a single underlying defect." + "reason": "RegNet and SqueezeBert are unrelated model code paths. These are separate migrations to standardized output handling, not duplicates." }, { "left": "pull_request:44010", "right": "pull_request:44129", "accept": false, - "reason": "SqueezeBert and SpeechT5 have different attention/output implementations; these are separate model-specific refactors, not one mergeable PR." + "reason": "SqueezeBert vs SpeechT5: distinct implementations and refactor scopes. They share output-capture modernization but are not the same underlying change." } ] }, @@ -19868,268 +19798,268 @@ "pull_request:44028|pull_request:44030", "pull_request:44029|pull_request:44030" ], - "prepared_review_unit_hash": "80335fcee18827dbc5eacc6eb767c70883534bfb00de57dc3195c295df9928ce", + "prepared_review_unit_hash": "df9f3158a8bec9efd4d9df3e885aa0ae47d0f018e66d7daaf1ce8596550417f7", "component_budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 39, - "serialized_chars": 39767, - "estimated_input_tokens": 9942, - "estimated_eval_tokens": 20140 + "serialized_chars": 39793, + "estimated_input_tokens": 9949, + "estimated_eval_tokens": 20154 }, "budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 39, - "serialized_chars": 39767, - "estimated_input_tokens": 9942, - "estimated_eval_tokens": 20140 + "serialized_chars": 39793, + "estimated_input_tokens": 9949, + "estimated_eval_tokens": 20154 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs share a broad \"refactor output tracing\" theme, but they are model-specific changes across different code paths, so none of the soft edges are true duplicates. I\u2019d treat them as a family of related refactors, not one mergeable change.", - "confidence": 0.86, + "summary": "The cluster is a set of model-specific 'refactor output tracing' pull requests. They are thematically similar but target different architectures and code paths, so they should not be treated as duplicate fixes. The best representative PR is 44025 because it is a clear, multi-file example of the shared refactor pattern.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "No single PR is a true duplicate target; 44025 is the closest representative because it makes a coherent can_return_tuple refactor across two closely related depth-anything files.", + "canonical_pr_reason": "No single PR is a true canonical duplicate for the cluster; 44025 is the closest representative because it is a concrete output-tracing refactor spanning two closely related files in one model family.", "best_issue_reason": null, - "best_pr_reason": "44025 is the strongest standalone PR in the set: it has a focused, internally consistent model-family refactor spanning related files rather than a wrapper-only or unrelated architecture change.", + "best_pr_reason": "44025 is the best global representative because it shows the shared pattern cleanly and covers two related depth-anything implementations, while still being a specific, mergeable code change.", "soft_edge_verdicts": [ { "left": "pull_request:43998", "right": "pull_request:44002", "accept": false, - "reason": "Different model families and forward paths; same refactor style only." + "reason": "Different model families (timm_backbone vs upernet); same refactor theme, but not the same concrete code-path change." }, { "left": "pull_request:43998", "right": "pull_request:44000", "accept": false, - "reason": "Timm backbone vs vision-text dual encoder are unrelated code paths." + "reason": "Different architectures (timm_backbone vs vision_text_dual_encoder); similar tracing refactor, not a duplicate fix." }, { "left": "pull_request:43998", "right": "pull_request:44001", "accept": false, - "reason": "Different architectures; not the same concrete output-handling bug/change." + "reason": "Unrelated model implementations; both refactor output handling, but they are separate code paths." }, { "left": "pull_request:43998", "right": "pull_request:44003", "accept": false, - "reason": "Timm backbone and Falcon/Mamba are separate implementations with no shared fix." + "reason": "timm_backbone and mamba/falcon_mamba are unrelated model families; not the same underlying bug or change." }, { "left": "pull_request:43998", "right": "pull_request:44029", "accept": false, - "reason": "Backbone refactor vs RWKV refactor; only the theme matches." + "reason": "Different model families and output mechanics; only the high-level refactor motif matches." }, { "left": "pull_request:43999", "right": "pull_request:44001", "accept": false, - "reason": "MobileNetV1 and UnivNet are unrelated model code paths." + "reason": "MobileNetV1 and UnivNet are unrelated models; the overlap is only the generic output-tracing refactor pattern." }, { "left": "pull_request:43999", "right": "pull_request:44002", "accept": false, - "reason": "Different model families; not the same underlying change." + "reason": "Different subsystems and model code paths; not a single mergeable fix." }, { "left": "pull_request:43999", "right": "pull_request:44025", "accept": false, - "reason": "MobileNetV1 vs depth-anything; similar decorator refactor, different code path." + "reason": "MobileNetV1 vs depth_anything are separate model families; same style of refactor, not duplicate behavior." }, { "left": "pull_request:43999", "right": "pull_request:44026", "accept": false, - "reason": "Different wrapper/model families; no shared concrete bug." + "reason": "MobileNetV1 and vision_encoder_decoder do not share the same concrete code path." }, { "left": "pull_request:43999", "right": "pull_request:44027", "accept": false, - "reason": "MobileNetV1 and speech encoder-decoder are separate refactors." + "reason": "MobileNetV1 and speech_encoder_decoder are unrelated; only the generic tracing refactor overlaps." }, { "left": "pull_request:44000", "right": "pull_request:44001", "accept": false, - "reason": "Vision-text dual encoder and UnivNet are unrelated implementations." + "reason": "vision_text_dual_encoder and univnet are different models with different forward paths; not the same change." }, { "left": "pull_request:44000", "right": "pull_request:44002", "accept": false, - "reason": "Different model families; same general tracing cleanup only." + "reason": "Different model families; both adjust output tracing, but not the same underlying fix." }, { "left": "pull_request:44000", "right": "pull_request:44003", "accept": false, - "reason": "Vision-text dual encoder and Falcon/Mamba fix different code paths." + "reason": "vision_text_dual_encoder and mamba/codegen are unrelated code paths." }, { "left": "pull_request:44000", "right": "pull_request:44025", "accept": false, - "reason": "Different models; no single shared bug or mergeable fix." + "reason": "No shared concrete bug: one is vision-text dual encoder, the other is depth-anything refactoring." }, { "left": "pull_request:44001", "right": "pull_request:44002", "accept": false, - "reason": "UnivNet and UperNet are distinct model families." + "reason": "Unrelated model families; the similarity is only in output tracing cleanup." }, { "left": "pull_request:44001", "right": "pull_request:44003", "accept": false, - "reason": "UnivNet vs Falcon/Mamba; same meta-pattern, not same change." + "reason": "UnivNet and mamba/codegen do not appear to fix the same code-path problem." }, { "left": "pull_request:44001", "right": "pull_request:44025", "accept": false, - "reason": "Different architectures and output types; not duplicates." + "reason": "Different model families and different forward implementations; not mergeable as one PR." }, { "left": "pull_request:44002", "right": "pull_request:44029", "accept": false, - "reason": "UperNet and RWKV touch unrelated forward paths." + "reason": "UperNet and RWKV are unrelated architectures; same refactor motif, not duplicate behavior." }, { "left": "pull_request:44002", "right": "pull_request:44003", "accept": false, - "reason": "UperNet and Falcon/Mamba are separate model changes." + "reason": "Different subsystems and fixes; not the same concrete change." }, { "left": "pull_request:44003", "right": "pull_request:44004", "accept": false, - "reason": "Falcon/Mamba and CodeGen are different language-model implementations." + "reason": "Mamba/falcon_mamba versus CodeGen are unrelated code paths; the overlap is only the output-tracing refactor theme." }, { "left": "pull_request:43998", "right": "pull_request:44004", "accept": false, - "reason": "Timm backbone and CodeGen do not share the same code path." + "reason": "timm_backbone and codegen are different model families with different forward logic." }, { "left": "pull_request:44001", "right": "pull_request:44004", "accept": false, - "reason": "UnivNet and CodeGen are unrelated changes." + "reason": "UnivNet and CodeGen are unrelated; similar refactor pattern is not enough to duplicate them." }, { "left": "pull_request:44002", "right": "pull_request:44004", "accept": false, - "reason": "UperNet and CodeGen are not the same underlying fix." + "reason": "UperNet and CodeGen are separate implementations; not the same underlying bug." }, { "left": "pull_request:44000", "right": "pull_request:44004", "accept": false, - "reason": "Vision-text dual encoder and CodeGen are separate implementations." + "reason": "vision_text_dual_encoder and codegen are different code paths and different model domains." }, { "left": "pull_request:44025", "right": "pull_request:44026", "accept": false, - "reason": "DepthAnything and vision-encoder-decoder are different model families." + "reason": "DepthAnything and vision_encoder_decoder are different model families; same refactor style only." }, { "left": "pull_request:44025", "right": "pull_request:44027", "accept": false, - "reason": "DepthAnything and speech-encoder-decoder are analogous refactors, not duplicates." + "reason": "DepthAnything and speech_encoder_decoder are unrelated code paths." }, { "left": "pull_request:44025", "right": "pull_request:44028", "accept": false, - "reason": "DepthAnything vs SuperPoint: different model internals and outputs." + "reason": "DepthAnything and SuperPoint are different models; not a single concrete fix." }, { "left": "pull_request:44025", "right": "pull_request:44029", "accept": false, - "reason": "DepthAnything and RWKV are unrelated code paths." + "reason": "DepthAnything and RWKV do not share the same implementation issue." }, { "left": "pull_request:44025", "right": "pull_request:44030", "accept": false, - "reason": "DepthAnything and DPR fix different model families." + "reason": "DepthAnything and DPR are unrelated architectures; same refactor motif is too broad." }, { "left": "pull_request:44026", "right": "pull_request:44029", "accept": false, - "reason": "Vision-encoder-decoder wrapper vs RWKV core model; not the same change." + "reason": "VisionEncoderDecoder and RWKV are different model families with different output paths." }, { "left": "pull_request:44026", "right": "pull_request:44027", "accept": false, - "reason": "Vision and speech encoder-decoder wrappers are similar, but separate code paths." + "reason": "VisionEncoderDecoder and SpeechEncoderDecoder are analogous wrapper models, but still separate code paths and not one mergeable fix." }, { "left": "pull_request:44026", "right": "pull_request:44028", "accept": false, - "reason": "Different architectures and tracing implementations." + "reason": "Different model implementations; no shared concrete bug beyond tracing cleanup." }, { "left": "pull_request:44026", "right": "pull_request:44030", "accept": false, - "reason": "Vision-encoder-decoder and DPR are distinct models." + "reason": "VisionEncoderDecoder and DPR are unrelated; not the same change." }, { "left": "pull_request:44027", "right": "pull_request:44028", "accept": false, - "reason": "Speech encoder-decoder and SuperPoint are unrelated subsystems." + "reason": "SpeechEncoderDecoder and SuperPoint are unrelated models; only the generic output-tracing pattern matches." }, { "left": "pull_request:44027", "right": "pull_request:44029", "accept": false, - "reason": "Speech encoder-decoder and RWKV do not share a concrete fix." + "reason": "SpeechEncoderDecoder and RWKV are different code paths and architectures." }, { "left": "pull_request:44027", "right": "pull_request:44030", "accept": false, - "reason": "Different model families and output contracts." + "reason": "SpeechEncoderDecoder and DPR do not appear to fix the same concrete issue." }, { "left": "pull_request:44028", "right": "pull_request:44029", "accept": false, - "reason": "SuperPoint and RWKV are unrelated code paths." + "reason": "SuperPoint and RWKV are unrelated model families." }, { "left": "pull_request:44028", "right": "pull_request:44030", "accept": false, - "reason": "SuperPoint and DPR are different model implementations." + "reason": "SuperPoint and DPR are different code paths; not duplicate changes." }, { "left": "pull_request:44029", "right": "pull_request:44030", "accept": false, - "reason": "RWKV and DPR are distinct changes with no shared underlying bug." + "reason": "RWKV and DPR are unrelated architectures, so this is not the same underlying fix." } ] }, @@ -20162,12 +20092,12 @@ "pull_request:44662|pull_request:45075", "pull_request:45075|pull_request:45401" ], - "prepared_review_unit_hash": "c78d0470c027ec722e8a0fe3acd3dd948dcda0ef5e2d2c64899640fb4c218f87", + "prepared_review_unit_hash": "728e7b17fb2371a5cd141f8b2b6d7c46730ba48586e2465d16504f187cb9a760", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 10913, + "serialized_chars": 10914, "estimated_input_tokens": 2729, "estimated_eval_tokens": 5714 }, @@ -20175,7 +20105,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 10913, + "serialized_chars": 10914, "estimated_input_tokens": 2729, "estimated_eval_tokens": 5714 }, @@ -20184,30 +20114,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All three pull requests are separate model-integration additions for different models (PenguinVL, DeepSeek-OCR-2, and Voxtral TTS). They share common Transformers scaffolding files because each registers a new model, but they do not target the same underlying bug or change.", + "summary": "All three pull requests are separate model-addition implementations, not duplicates. They share generic Transformers scaffolding (auto registration, docs, package wiring), but each targets a different model family and code path: PenguinVL, DeepSeek-OCR-2, and Voxtral TTS.", "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "No single PR is a duplicate canonical representative; these are distinct feature additions for different models, so none should be collapsed into one cluster.", + "canonical_pr_reason": "No canonical PR for a duplicate cluster: the three PRs introduce different models and do not appear to fix or implement the same concrete change.", "best_issue_reason": null, - "best_pr_reason": "No best PR for duplication purposes: each PR is a separate model implementation and should remain independent.", + "best_pr_reason": "PR 45075 is the strongest representative only by maturity/activity, but it is still a distinct DeepSeek-OCR-2 model-addition PR rather than a duplicate of the others.", "soft_edge_verdicts": [ { "left": "pull_request:44662", "right": "pull_request:45401", "accept": false, - "reason": "Both are new model additions and share auto-registration/docs plumbing, but they implement different models with different code paths, tests, and model-specific files. Not the same underlying change." + "reason": "Different model additions and different problem domains: PenguinVL (vision-language) vs Voxtral TTS (text-to-speech). Shared auto/docs wiring is generic and not evidence of the same underlying change." }, { "left": "pull_request:44662", "right": "pull_request:45075", "accept": false, - "reason": "PenguinVL and DeepSeek-OCR-2 are unrelated model integrations. Overlap is limited to common framework registration files, which is expected for separate PRs." + "reason": "PenguinVL and DeepSeek-OCR-2 are separate model implementations with different architectures and assets; overlap is only in framework registration/docs scaffolding, so they are not the same code-path fix." }, { "left": "pull_request:45075", "right": "pull_request:45401", "accept": false, - "reason": "DeepSeek-OCR-2 and Voxtral TTS add different model families with distinct modeling/processing implementations. They are not plausibly mergeable into one PR as the same fix/change." + "reason": "DeepSeek-OCR-2 and Voxtral TTS address unrelated model families and tasks. They may touch common auto-mapping files, but they are not plausibly mergeable as one PR for the same bug/change." } ] }, @@ -20237,7 +20167,7 @@ "soft_pairs": [ "pull_request:44979|pull_request:45363" ], - "prepared_review_unit_hash": "80f7636eafb95ab658a30929b350983f3e426b20cbf666405d241f87aeb6f98e", + "prepared_review_unit_hash": "9f481e346682d77dd3ef7bdba072e53e771a4bc0b3046a6d67d4d0b5ece73b8d", "component_budget": { "node_count": 2, "item_count": 2, @@ -20259,18 +20189,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two PRs are related to module/kernel fusion, but they are not duplicates: one adds a standalone Module Fusion API, while the other wires n-to-1 fusion through KernelConfig and model loading.", - "confidence": 0.92, + "summary": "These PRs are related to module/kernel fusion, but they are not duplicates: one defines a general Module Fusion API, while the other implements n-to-1 kernel fusion through KernelConfig and model/runtime integration.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44979 is the broader foundational change: it introduces a generic Module Fusion API in a new module and tests it, whereas PR 45363 is a separate KernelConfig-based integration on top of a different code path.", + "canonical_pr_reason": "PR #44979 is the broader, more foundational change: it introduces the new Module Fusion API and core module_fusion implementation, which is the best representative of this cluster.", "best_issue_reason": null, - "best_pr_reason": "PR 44979 is the best cluster representative because it defines the underlying fusion abstraction; PR 45363 is a narrower, implementation-specific feature/example and not as central to the shared theme.", + "best_pr_reason": "PR #44979 is the best single PR to represent the cluster because it centers the underlying fusion abstraction rather than one specific KernelConfig application path.", "soft_edge_verdicts": [ { "left": "pull_request:44979", "right": "pull_request:45363", "accept": false, - "reason": "Related topic, but not the same underlying change. 44979 adds a general Module Fusion API; 45363 implements n-to-1 kernel fusion via KernelConfig and touches integration/model-loading paths. They are not the same fix and would not plausibly merge into one PR." + "reason": "They share the general fusion theme, but they address different concrete changes: #44979 adds the module fusion API, while #45363 wires n-to-1 kernel fusion through KernelConfig and related integrations. They would not plausibly be merged as one PR." } ] }, @@ -20300,7 +20230,7 @@ "soft_pairs": [ "pull_request:45086|pull_request:45317" ], - "prepared_review_unit_hash": "f5965cbf9cf73456e52ae7a8e260c38d145533df9986e9e53622b4d68af56767", + "prepared_review_unit_hash": "8b47e36ac54c6c8679ab15a45d64bd382586bfadde51d65496ef1ce64d151cab", "component_budget": { "node_count": 2, "item_count": 2, @@ -20322,18 +20252,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open PRs address the same Mistral regex AttributeError fix in the same code path; PR 45317 is the more complete version with test coverage, while 45086 is an earlier narrower variant.", + "summary": "Both pull requests fix the same Mistral regex patch bug in `tokenization_utils_tokenizers.py` by switching from `tokenizer.backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer`; 45317 is the more complete version because it adds tests.", "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45317 is the better canonical PR because it implements the same fix as 45086 but adds tests and appears to be the more complete, discussion-backed variant.", + "canonical_pr_reason": "45317 is the better canonical PR because it implements the same fix as 45086 and also includes regression test coverage.", "best_issue_reason": null, - "best_pr_reason": "PR 45317 is the best representative of the cluster: same underlying bug, same target issue, same file, and it includes the broader fix plus tests.", + "best_pr_reason": "45317 is the best PR candidate: same concrete fix, plus tests, making it the most complete mergeable representation of the change.", "soft_edge_verdicts": [ { "left": "pull_request:45086", "right": "pull_request:45317", "accept": true, - "reason": "Same concrete AttributeError fix in _patch_mistral_regex, same target issue, and identical code-path change in tokenization_utils_tokenizers.py; 45317 is just the expanded version with tests." + "reason": "Same underlying bug fix, same file, same code-path change, and one PR is essentially a test-augmented version of the other." } ] }, @@ -20363,7 +20293,7 @@ "soft_pairs": [ "pull_request:45115|pull_request:45569" ], - "prepared_review_unit_hash": "565cea04c5d9161e5103e4ae3d09cf8485919457f380e8af0181a5c3eb05b82b", + "prepared_review_unit_hash": "b3d9772d1cf179d5e9b1a7f299fc68fa7042e74f91d6d7cfbeee89861baabd21", "component_budget": { "node_count": 2, "item_count": 2, @@ -20385,18 +20315,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related NemotronH PRs, but they address different changes: one refactors conversion/mapping/inheritance, while the other introduces the dense/sparse model split plus docs and tests. They are not duplicates and should not be merged as one PR.", - "confidence": 0.93, + "summary": "The two PRs are related to NemotronH, but they tackle different changes: one is a conversion/refactor toward GraniteMoeHybrid-compatible naming, while the other splits NemotronH into dense/sparse variants with docs and tests. Shared files reflect the same subsystem, not the same underlying fix.", + "confidence": 0.88, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45569 is the broader, user-facing model split and likely the best representative of the NemotronH family work in this cluster.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR #45569 is the clearest standalone change set: it introduces the new dense/sparse model variants, updates auto-mapping/docs, and adds tests.", + "best_pr_reason": "PR 45569 is the more complete and user-facing change: it adds the dense/sparse model split, auto-mapping, documentation, and tests, so it is the better representative of this cluster.", "soft_edge_verdicts": [ { "left": "pull_request:45115", "right": "pull_request:45569", "accept": false, - "reason": "They share NemotronH files, but the concrete changes differ: #45115 focuses on conversion key renaming and inheritance refactoring, while #45569 adds new dense/sparse model classes, docs, auto-mappings, and tests. Same subsystem, different underlying change." + "reason": "Reject: both touch NemotronH internals, but 45115 is a narrow structural/conversion refactor, while 45569 is a broader model-family split with docs/tests. They are not the same concrete change and would not plausibly merge as one PR." } ] }, @@ -20426,7 +20356,7 @@ "soft_pairs": [ "pull_request:45332|pull_request:45333" ], - "prepared_review_unit_hash": "fa4f802391298b55760d4121403b6ff0066cd8f610193b05cb0f2dad25412d24", + "prepared_review_unit_hash": "4f4eaba59426177f515be4cc2c2cc2577223bc91721ab50fed4fbd43db7a7580", "component_budget": { "node_count": 2, "item_count": 2, @@ -20448,18 +20378,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two PRs are related pieces of a heterogeneous-model feature, but they are not the same change: one introduces per-layer config support, while the other adds modeling/runtime support that consumes that config. The overlap in shared files is infrastructure, not a duplicate code-path fix.", - "confidence": 0.92, + "summary": "These are related but not duplicates: PR #45333 adds the heterogeneous per-layer config plumbing, while PR #45332 builds on that to add heterogeneous model/cache/masking support. They target different layers of the same feature and would not be merged as one change set.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45332 is the broader, more central change set: it covers heterogeneous model support in modeling/runtime code and also touches the config plumbing it needs, so it best represents the cluster.", + "canonical_pr_reason": "PR #45332 is the broader umbrella change: it includes the modeling/cache integration work and depends on the config machinery added elsewhere, so it is the better representative of the overall feature cluster.", "best_issue_reason": null, - "best_pr_reason": "PR #45332 is the best representative because it spans the main feature area rather than only the config layer, making it the most informative cluster anchor.", + "best_pr_reason": "PR #45333 is the cleanest standalone representative for the foundational config API because it is narrowly scoped to per-layer configuration plumbing and tests.", "soft_edge_verdicts": [ { "left": "pull_request:45332", "right": "pull_request:45333", "accept": false, - "reason": "These are complementary but distinct PRs: #45333 adds per-layer configuration plumbing, while #45332 adds heterogeneous model/cache/masking behavior that uses it. They do not fix the same concrete bug or change the same code path, so they should not be merged as duplicates." + "reason": "Not the same concrete change: #45333 introduces heterogeneous config support, while #45332 adds model/cache behavior that consumes that config. They are complementary feature pieces rather than duplicate fixes and would not be plausibly merged into one PR." } ] }, @@ -20489,12 +20419,12 @@ "soft_pairs": [ "pull_request:45415|pull_request:45425" ], - "prepared_review_unit_hash": "9edda8aa431c5c4d648931ca9b21796008bcc8833da7a0df15c8f26ed5bcf1f5", + "prepared_review_unit_hash": "c8a66d8ccc7babff6aa2ba8fb35a4d3eb7aed5f6e55f58463b5eb68c0e890050", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5514, + "serialized_chars": 5513, "estimated_input_tokens": 1379, "estimated_eval_tokens": 3014 }, @@ -20502,7 +20432,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5514, + "serialized_chars": 5513, "estimated_input_tokens": 1379, "estimated_eval_tokens": 3014 }, @@ -20511,18 +20441,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open PRs both relate to typing, but they are not the same change: #45415 is a broad repo-wide type-checking sweep, while #45425 is a focused typing-infra update around `modeling_utils` and related protocols/utilities.", - "confidence": 0.91, + "summary": "Two open PRs in the typing/type-checking area, but they address different changes: one is a broad pass to add type checking across many `src/transformers/*.py` files, while the other introduces typing protocols/helpers and modeling-related typing fixes. They overlap in a few files, but not the same concrete bug/change.", + "confidence": 0.87, "canonical_issue_reason": null, - "canonical_pr_reason": "#45415 is the broader, more representative typing sweep across many `src/transformers/*.py` files, so it better serves as the cluster\u2019s canonical PR.", + "canonical_pr_reason": "PR 45425 is the more concrete, self-contained typing refactor with specific helper protocol additions and targeted modeling-related fixes; it better represents a distinct change than the broad type-checking sweep in 45415.", "best_issue_reason": null, - "best_pr_reason": "#45425 is the better focused candidate if you want a concrete typing-infra change centered on `modeling_utils` and shared protocols, rather than a large, diffuse type-checking cleanup.", + "best_pr_reason": "PR 45425 is the best representative of the cluster because it has a focused technical goal and clearer code-level scope, whereas 45415 is a broad, cross-file type-checking cleanup.", "soft_edge_verdicts": [ { "left": "pull_request:45415", "right": "pull_request:45425", "accept": false, - "reason": "Reject: both are typing-related, but #45415 is a broad type-checking pass while #45425 adds specific protocols and typing support for `modeling_utils`-adjacent code. They do not look like the same underlying fix and would not plausibly merge into one PR." + "reason": "Reject: both are typing-related and share a few files, but 45415 is a broad type-checking sweep while 45425 adds specific typing protocols and modeling fixes; they are not the same underlying change and would not plausibly merge as one PR." } ] }, @@ -20552,7 +20482,7 @@ "soft_pairs": [ "pull_request:45513|pull_request:46084" ], - "prepared_review_unit_hash": "8ce6a51e36d3b2e40cfc8fcddb0196ea43b4bc43e7db1c707ec786ec091077d6", + "prepared_review_unit_hash": "8c3bff436b63fa6c2eba44d9eb09e8d0e39604eb6d09b246a1ad8a7d4b3aef78", "component_budget": { "node_count": 2, "item_count": 2, @@ -20574,18 +20504,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two PRs are not duplicates: #45513 fixes multi-token cached forward behavior in Qwen3.5/Qwen3-next/Olmo-hybrid linear attention, while #46084 fixes use_cache behavior for Mamba2 when seq_len > 1. They target different model families and different code paths.", - "confidence": 0.98, + "summary": "Both pull requests address cached-forward bugs when `seq_len > 1`, but in different model families and codepaths: OLMo/Qwen hybrid linear attention vs. Mamba2 CUDA kernels. They are thematically similar, not the same underlying change, so they should not be clustered as duplicates.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "PR 46084 is the better representative because it has a concrete linked issue, a narrowly scoped fix, and a clear bug statement about `use_cache` with `seq_len > 1` in one specific implementation.", "best_issue_reason": null, - "best_pr_reason": "PR #46084 is the cleaner standalone representative: it is narrowly scoped, explicitly tied to issue #46032, and fixes one concrete cache/sequence-length bug in Mamba2.", + "best_pr_reason": "PR 46084 is the stronger canonical PR for the cluster since it is explicit, targeted, and less broad than the other PR; PR 45513 spans multiple model variants and a broader refactor-like fix across several files.", "soft_edge_verdicts": [ { "left": "pull_request:45513", "right": "pull_request:46084", "accept": false, - "reason": "Different bugs in different architectures: Qwen3.5/Qwen3-next/Olmo-hybrid cached linear attention vs. Mamba2 cache handling for seq_len > 1. They would not plausibly be merged into one PR." + "reason": "Both involve cached multi-token forward behavior, but they fix different concrete implementations and likely different bugs: OLMo/Qwen hybrid linear attention versus Mamba2 state-space CUDA kernels. The overlap is only high-level theme, not a shared underlying change." } ] }, @@ -20615,7 +20545,7 @@ "soft_pairs": [ "pull_request:45548|pull_request:45662" ], - "prepared_review_unit_hash": "5d4fbdab8acbe90a5bd4e3621f32e4069e1774b890fd2d11c80a79af20dd8d5b", + "prepared_review_unit_hash": "77464b2b886a37677f5e97fe4ae356f789e6eaf7613a25e921dc9ced185d2b14", "component_budget": { "node_count": 2, "item_count": 2, @@ -20637,18 +20567,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs are about expert-parallel loading, but they target different failure modes and backends: #45548 fixes EP + DeepSpeed ZeRO-3/accelerate config serialization, while #45662 fixes EP + FSDP2 rank-0 broadcast overwriting experts. They are related, but not the same underlying bug/change.", - "confidence": 0.92, + "summary": "These two PRs are related by expert-parallelism, but they target different bugs and should not be merged as duplicates: one fixes EP/ZeRO-3 loading and config serialization, the other fixes expert parameters being overwritten by rank-0 broadcast in EP+FSDP2.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "No issue artifacts are present; if one PR must represent the cluster, #45662 is the better canonical PR because it addresses a concrete EP parameter-loading corruption path with explicit code changes in the sharding/broadcast flow.", + "canonical_pr_reason": "PR 45662 is the better representative of the cluster because it addresses a more concrete runtime correctness bug in the EP loading path, with changes across the actual code paths that overwrite expert parameters.", "best_issue_reason": null, - "best_pr_reason": "#45662 is the strongest representative of the cluster because it fixes a specific expert-parallel loading bug in the actual parameter assignment path, with a clearer symptom and broader implementation impact than #45548.", + "best_pr_reason": "PR 45662 is the strongest standalone candidate: it fixes a specific expert-overwrite failure mode and has broader, coherent code-path coverage than 45548.", "soft_edge_verdicts": [ { "left": "pull_request:45548", "right": "pull_request:45662", "accept": false, - "reason": "Related EP loading area, but different concrete bugs: ZeRO-3/accelerate config handling vs FSDP2 experts being overwritten by rank-0 broadcast. They would not plausibly merge into one PR." + "reason": "Both mention EP, but they fix different concrete problems: 45548 is about EP + DeepSpeed ZeRO-3 loading via accelerate and config handling, while 45662 is about experts being silently overwritten by rank-0 broadcast in EP + FSDP2. They are not the same underlying change." } ] }, @@ -20678,7 +20608,7 @@ "soft_pairs": [ "pull_request:45639|pull_request:45645" ], - "prepared_review_unit_hash": "904d176f204fe3ba4c2769524c282005497eec710007ad04b506793106b28534", + "prepared_review_unit_hash": "e8aa6ef632066fe7e69cba204e2b10af3e56f847474b4b8948e811b3cca35836", "component_budget": { "node_count": 2, "item_count": 2, @@ -20700,18 +20630,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs target xdist-safe handling of captured_info artifacts, but one is a narrow testing_utils fix while the other is a broader end-to-end CI/artifact handling change.", - "confidence": 0.74, + "summary": "Both PRs address the same xdist-related captured_info collision in testing_utils; 45645 is the more complete, cluster-representative fix because it also updates CI log collection for the new per-worker files.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45645 is the more complete candidate: it includes the underlying xdist-safe output-file fix plus the CI/workflow changes needed to consume multiple captured_info files.", + "canonical_pr_reason": "PR 45645 is the broader, more complete implementation of the same underlying bug: it makes captured_info output xdist-safe and updates downstream CI handling to aggregate multiple files.", "best_issue_reason": null, - "best_pr_reason": "PR 45645 best represents the cluster because it addresses the full artifact-collision problem and the downstream CI/debug-log handling, not just the local code-path fix.", + "best_pr_reason": "PR 45645 is the best representative of the cluster because it covers the core code-path fix plus the necessary CI-side adjustment, whereas 45639 is the narrower code-only slice.", "soft_edge_verdicts": [ { "left": "pull_request:45639", "right": "pull_request:45645", - "accept": false, - "reason": "They overlap on the same testing_utils bug, but 45645 is a broader superset that also changes workflow and notification-service behavior; that scope difference makes them not a clean duplicate PR pair." + "accept": true, + "reason": "Same underlying xdist captured_info collision fix; 45645 extends the same code-path change with compatible CI log aggregation, so they could plausibly be merged into one PR." } ] }, @@ -20743,12 +20673,12 @@ "pull_request:45720|pull_request:46259", "pull_request:46023|pull_request:46259" ], - "prepared_review_unit_hash": "33cb8c246ab97db81cbc5787608b996d47ae8d6519123225bac855a42a9de491", + "prepared_review_unit_hash": "ac459517ab879e8296a6f755af7cb937a2aa374f3602a511ed324c4a67764004", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 10911, + "serialized_chars": 10912, "estimated_input_tokens": 2728, "estimated_eval_tokens": 5712 }, @@ -20756,7 +20686,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 10911, + "serialized_chars": 10912, "estimated_input_tokens": 2728, "estimated_eval_tokens": 5712 }, @@ -20765,24 +20695,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is centered on linter/modeling-rule work, with one PR for a different rule (TRF003) and two PRs that appear to target the same TRF016-style dead-flag checks in image/video processors.", - "confidence": 0.91, + "summary": "These are three separate linter-rule PRs, not duplicates: one implements TRF003 for generation outputs, one fixes TRF016 for processor boolean flags, and one adds a different TRF018 modeling rule. The shared docs/setup files and overlapping model areas are too broad to treat any pair as the same change.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46259 is the broadest and most complete of the matching linter-rule fixes, with overlapping modeling-rule docs and many of the same processor/model files as 46023, making it the best representative of the duplicate change set.", + "canonical_pr_reason": "PR 46023 is the clearest standalone change: it targets a specific linter rule (TRF016), has a focused description, and the touched files align with that rule\u2019s processor-override checks.", "best_issue_reason": null, - "best_pr_reason": "PR 46259 is the best global PR candidate because it covers the same concrete linter rule area as 46023 and appears to subsume it with additional affected models/files.", + "best_pr_reason": "PR 46023 is the best representative of the cluster because it is the most self-contained and concretely described implementation of a single rule, whereas the others are separate rule implementations.", "soft_edge_verdicts": [ { "left": "pull_request:45720", "right": "pull_request:46259", "accept": false, - "reason": "These are about different concrete rules and code paths: 45720 implements part of TRF003 in Bloom/sequence-model forward signatures, while 46259 is about TRF016/TRF018-style processor flag-consumption checks. Shared subsystem is not enough to treat them as the same change." + "reason": "Different underlying rules and code paths: TRF003 changes model forward behavior for tuple returns, while TRF018 is a separate modeling rule. Shared linter/docs infrastructure and overlapping filenames are not enough to merge them." }, { "left": "pull_request:46023", "right": "pull_request:46259", - "accept": true, - "reason": "Both PRs target the same underlying dead-flag modeling-rule fix for image/video processors, share the modeling-rules docs change, and touch overlapping model processor files. They look mergeable into one broader linter-rule PR." + "accept": false, + "reason": "Both are linter-rule PRs, but they implement different rules (TRF016 vs TRF018) with different target classes and fixes. They are related infrastructure work, not the same bug/change." } ] }, @@ -20881,12 +20811,12 @@ "pull_request:46073|pull_request:46196", "pull_request:46154|pull_request:46196" ], - "prepared_review_unit_hash": "ecb3bbc299776ba01fe6624bc7cab7c52bf7c1bc6b695128a6927d78076f9c37", + "prepared_review_unit_hash": "b55001d1a33e05900e9b52d10dbb02dcf58fc8408fa8e49c5f353f1ed902272a", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 8083, + "serialized_chars": 8081, "estimated_input_tokens": 2021, "estimated_eval_tokens": 4298 }, @@ -20894,7 +20824,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 8083, + "serialized_chars": 8081, "estimated_input_tokens": 2021, "estimated_eval_tokens": 4298 }, @@ -20903,30 +20833,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All three PRs are about adding `attn_implementation=\"sdpa\"` support, but they target different model families and code paths. They are thematically similar rather than true duplicates, so none of the soft edges should be accepted.", - "confidence": 0.91, + "summary": "Three open PRs all add SDPA dispatch support, but they target different model families and code paths. They are related by theme, not duplicates of the same fix.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #46196 is the broadest and most representative of the shared SDPA-dispatch theme, spanning multiple model families and tests.", + "canonical_pr_reason": "PR #46196 is the broadest representative of the SDPA-dispatch theme, covering multiple model implementations and test updates, so it best represents the cluster even though it is not a duplicate of the others.", "best_issue_reason": null, - "best_pr_reason": "PR #46196 has the widest scope and strongest evidence of the common change pattern, covering several related model implementations in one place.", + "best_pr_reason": "PR #46196 is the strongest standalone PR in the set because it spans the most models and includes the most complete implementation/test coverage for the same feature area.", "soft_edge_verdicts": [ { "left": "pull_request:46073", "right": "pull_request:46154", "accept": false, - "reason": "Both add SDPA support, but for different models (CTRL vs T5) with distinct attention code paths; they are not the same concrete bug/change and would not be merged as one PR." + "reason": "Both add SDPA dispatch support, but one is for CTRL and the other for T5; they touch different model code paths and are not the same concrete bug or mergeable change." }, { "left": "pull_request:46073", "right": "pull_request:46196", "accept": false, - "reason": "CTRL-specific SDPA dispatch is separate from the wav2vec2/seamless M4T family changes in #46196; shared theme only, not the same underlying fix." + "reason": "These are model-specific SDPA enablement PRs for different architectures (CTRL vs wav2vec2/seamless variants); same broad feature goal, but not the same underlying fix." }, { "left": "pull_request:46154", "right": "pull_request:46196", "accept": false, - "reason": "T5 SDPA dispatch and the multi-model audio/seq2seq SDPA changes in #46196 touch different implementations and test sets, so they are not duplicates." + "reason": "T5 SDPA dispatch and the speech/multimodal SDPA work in #46196 are separate implementation efforts in different code paths, so they should not be merged as duplicates." } ] }, @@ -20956,7 +20886,7 @@ "soft_pairs": [ "pull_request:46126|pull_request:46269" ], - "prepared_review_unit_hash": "3bdeb5f1a054a56cecc887b38d2fd16f453b71198a1672539d7bf7c09dbbbb85", + "prepared_review_unit_hash": "91a6873f4f72ffde6168aef1f96d4074f6deb790013b30abc2255177000a4da1", "component_budget": { "node_count": 2, "item_count": 2, @@ -20978,18 +20908,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two pull requests overlap on distributed tensor-parallel files, but they are not the same underlying change: one is a focused EP/trainer integration feature, while the other is a broad distributed branch base spanning many unrelated distributed updates.", - "confidence": 0.94, + "summary": "The two pull requests are not duplicates: one is a focused EP/Trainer integration change, while the other is a broad distributed branch base with many unrelated edits. They overlap in a few distributed files, but not the same concrete code-path fix.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46126 is the more canonical change because it is narrowly scoped to EP/Trainer integration and a specific tensor-parallel routing implementation, making it a coherent fix/feature rather than a broad branch aggregation.", + "canonical_pr_reason": "PR 46126 is the more specific, self-contained change centered on EP routing and Trainer integration; PR 46269 is a broad branch-base PR touching many distributed areas and is not a clean canonical duplicate target.", "best_issue_reason": null, - "best_pr_reason": "PR 46126 is the best representative because it targets a concrete code path and has a clear, self-contained purpose. PR 46269 is too broad and mixed to serve as a duplicate-triage anchor.", + "best_pr_reason": "PR 46126 best represents the cluster because it has a clear, narrow purpose and concrete implementation details, whereas PR 46269 is too broad and mixed-purpose to serve as a good representative.", "soft_edge_verdicts": [ { "left": "pull_request:46126", "right": "pull_request:46269", "accept": false, - "reason": "Shared filenames are not enough here. 46126 implements EP routing/trainer integration, while 46269 is a large distributed branch base touching many subsystems; they do not look like the same mergeable fix or the same underlying bug/change." + "reason": "Reject: shared distributed filenames are not enough here. 46126 is about EP router masking and trainer integration; 46269 is a large distributed branch-base PR with many additional changes, so they do not appear to be the same underlying bug/change or plausibly merge into one PR." } ] }, @@ -21019,7 +20949,7 @@ "soft_pairs": [ "pull_request:46152|pull_request:46207" ], - "prepared_review_unit_hash": "f98eac615fd89dfa9be0e7e6ac7e0dd5c08b70f05bdf1ae60202dcd4205ab69e", + "prepared_review_unit_hash": "c8db8088063215f3b854ec41f80a0e28754df8c6713d624e3c246501634ae36b", "component_budget": { "node_count": 2, "item_count": 2, @@ -21041,18 +20971,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open PRs that both remove direct file-open source inspection, but they target different code paths and likely separate bugs. One changes `import_utils.py` to avoid opening Transformers' own source file; the other changes `modeling_utils.py` to use `inspect.getsource` in attention-implementation detection.", - "confidence": 0.88, + "summary": "These two PRs share only a broad theme of avoiding direct file-open source inspection, but they target different code paths and fixes. They should not be treated as duplicates or merged as one change.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #46207 is the more specific and directly motivated fix: it replaces `open()` with `inspect.getsource()` in the attention-implementation heuristic and includes a test. PR #46152 is broader and addresses a different source-resolution path in import utilities.", + "canonical_pr_reason": "No canonical PR here: PR 46152 updates import_utils source/traversable resolution, while PR 46207 changes _can_set_attn_implementation in modeling_utils. They address different behaviors and are not the same fix.", "best_issue_reason": null, - "best_pr_reason": "PR #46207 is the best representative change because it fixes a concrete, self-contained code path with a clear behavior change and test coverage.", + "best_pr_reason": "Neither PR is a good global representative of the cluster as a duplicate target; the overlap is only thematic, not a shared concrete bug or change.", "soft_edge_verdicts": [ { "left": "pull_request:46152", "right": "pull_request:46207", "accept": false, - "reason": "Both avoid opening source files, but they modify different modules and different behaviors: one resolves package resources in `import_utils`, the other changes `inspect.getsource` usage in `modeling_utils`. They do not look like the same underlying bug or mergeable change." + "reason": "Reject: both avoid file-open-based source reading, but they fix different code paths in different modules with different mechanics. They do not look mergeable into one PR for the same underlying bug." } ] }, diff --git a/analysis/current/manifest.json b/analysis/current/manifest.json index b6cca7f546a1ba13f2a0633e394b4ba5aa519bff..4319f110e329ba887c7048565488c267266ca955 100644 --- a/analysis/current/manifest.json +++ b/analysis/current/manifest.json @@ -1,8 +1,8 @@ { - "analysis_id": "hybrid-model-20260601t000028z", + "analysis_id": "hybrid-model-20260601t060039z", "archived_artifacts": { - "hybrid": "snapshots/20260601T000028Z/analysis-runs/hybrid-model-20260601t000028z/analysis-report-hybrid.json", - "hybrid_reviews": "snapshots/20260601T000028Z/analysis-runs/hybrid-model-20260601t000028z/analysis-report-hybrid.llm-reviews.json" + "hybrid": "snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/analysis-report-hybrid.llm-reviews.json" }, "artifacts": { "hybrid": "analysis/current/analysis-report-hybrid.json", @@ -10,9 +10,9 @@ }, "channel": "canonical", "model": null, - "published_at": "2026-06-01T00:04:45Z", + "published_at": "2026-06-01T06:09:00Z", "repo": "huggingface/transformers", "schema_version": 1, - "snapshot_id": "20260601T000028Z", + "snapshot_id": "20260601T060039Z", "variant": "hybrid" } diff --git a/snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/analysis-report-hybrid.json b/snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/analysis-report-hybrid.json new file mode 100644 index 0000000000000000000000000000000000000000..7e62e3fee4869e908142685a9ee730027ac35f10 --- /dev/null +++ b/snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/analysis-report-hybrid.json @@ -0,0 +1,1997 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260601T060039Z", + "generated_at": "2026-06-01T06:08:58Z", + "evidence_quality": "full", + "llm_enrichment": true, + "meta_bugs": [ + { + "cluster_id": "cluster-43979-11", + "summary": "Cluster of 1 issues and 10 PRs centered on issue #43979.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 43979, + "canonical_pr_number": 44007, + "issue_numbers": [ + 43979 + ], + "pr_numbers": [ + 43996, + 44007, + 44013, + 44044, + 44066, + 44072, + 44085, + 44129, + 44154, + 44722 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 43996, + "right_pr_number": 44007, + "code_similarity": 0.179, + "size_similarity": 0.576, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.429, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44013, + "code_similarity": 0.122, + "size_similarity": 0.318, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.392, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44044, + "code_similarity": 0.245, + "size_similarity": 0.864, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.479, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44066, + "code_similarity": 0.225, + "size_similarity": 0.818, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.408, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44072, + "code_similarity": 0.14, + "size_similarity": 0.303, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.528, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44085, + "code_similarity": 0.216, + "size_similarity": 0.783, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.398, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44129, + "code_similarity": 0.163, + "size_similarity": 0.643, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.229, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44154, + "code_similarity": 0.153, + "size_similarity": 0.535, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.31, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44722, + "code_similarity": 0.225, + "size_similarity": 0.848, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.368, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44013, + "code_similarity": 0.19, + "size_similarity": 0.553, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.531, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44044, + "code_similarity": 0.186, + "size_similarity": 0.667, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.354, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44066, + "code_similarity": 0.188, + "size_similarity": 0.704, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.315, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44072, + "code_similarity": 0.212, + "size_similarity": 0.526, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.708, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44085, + "code_similarity": 0.195, + "size_similarity": 0.735, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.318, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44129, + "code_similarity": 0.103, + "size_similarity": 0.37, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.191, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44154, + "code_similarity": 0.238, + "size_similarity": 0.93, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.344, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44722, + "code_similarity": 0.178, + "size_similarity": 0.679, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.28, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44044, + "code_similarity": 0.126, + "size_similarity": 0.368, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.351, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44066, + "code_similarity": 0.127, + "size_similarity": 0.389, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.325, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44072, + "code_similarity": 0.29, + "size_similarity": 0.952, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.667, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44085, + "code_similarity": 0.131, + "size_similarity": 0.406, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.329, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44129, + "code_similarity": 0.07, + "size_similarity": 0.205, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.192, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44154, + "code_similarity": 0.177, + "size_similarity": 0.594, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.389, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44722, + "code_similarity": 0.118, + "size_similarity": 0.375, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.287, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44066, + "code_similarity": 0.25, + "size_similarity": 0.947, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.404, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44072, + "code_similarity": 0.136, + "size_similarity": 0.351, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.442, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44085, + "code_similarity": 0.24, + "size_similarity": 0.906, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.394, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44129, + "code_similarity": 0.147, + "size_similarity": 0.555, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.243, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44154, + "code_similarity": 0.17, + "size_similarity": 0.62, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.306, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44722, + "code_similarity": 0.257, + "size_similarity": 0.982, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.402, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44072, + "code_similarity": 0.133, + "size_similarity": 0.37, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.393, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44085, + "code_similarity": 0.763, + "size_similarity": 0.957, + "file_overlap": 0.5, + "area_overlap": 0.825, + "patch_similarity": 0.887, + "shared_filenames": [ + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 41 + ], + [ + 174, + 181 + ], + [ + 250, + 257 + ], + [ + 398, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 466 + ], + [ + 471, + 488 + ], + [ + 518, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 606 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ], + "right_ranges": [ + [ + 33, + 40 + ], + [ + 173, + 180 + ], + [ + 249, + 256 + ], + [ + 397, + 405 + ], + [ + 408, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 465 + ], + [ + 470, + 489 + ], + [ + 517, + 539 + ], + [ + 553, + 559 + ], + [ + 565, + 574 + ], + [ + 579, + 597 + ], + [ + 600, + 611 + ], + [ + 633, + 639 + ], + [ + 645, + 652 + ], + [ + 657, + 673 + ], + [ + 716, + 728 + ], + [ + 737, + 743 + ], + [ + 748, + 771 + ], + [ + 789, + 794 + ] + ] + } + ] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44129, + "code_similarity": 0.145, + "size_similarity": 0.526, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.263, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44154, + "code_similarity": 0.174, + "size_similarity": 0.654, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.286, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44722, + "code_similarity": 0.907, + "size_similarity": 0.964, + "file_overlap": 1.0, + "area_overlap": 0.808, + "patch_similarity": 0.874, + "shared_filenames": [ + "src/transformers/models/codegen/modeling_codegen.py", + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/codegen/modeling_codegen.py", + "left_ranges": [ + [ + 245, + 266 + ] + ], + "right_ranges": [ + [ + 228, + 234 + ] + ] + }, + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 41 + ], + [ + 174, + 181 + ], + [ + 250, + 257 + ], + [ + 398, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 466 + ], + [ + 471, + 488 + ], + [ + 518, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 606 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ], + "right_ranges": [ + [ + 33, + 42 + ], + [ + 175, + 182 + ], + [ + 251, + 258 + ], + [ + 399, + 420 + ], + [ + 426, + 435 + ], + [ + 460, + 467 + ], + [ + 472, + 489 + ], + [ + 519, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 612 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 663 + ], + [ + 666, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ] + } + ] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44085, + "code_similarity": 0.137, + "size_similarity": 0.387, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.398, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44129, + "code_similarity": 0.074, + "size_similarity": 0.195, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.231, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44154, + "code_similarity": 0.175, + "size_similarity": 0.566, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.414, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44722, + "code_similarity": 0.124, + "size_similarity": 0.357, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.347, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44129, + "code_similarity": 0.141, + "size_similarity": 0.503, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.272, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44154, + "code_similarity": 0.18, + "size_similarity": 0.684, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.289, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44722, + "code_similarity": 0.728, + "size_similarity": 0.923, + "file_overlap": 0.5, + "area_overlap": 0.791, + "patch_similarity": 0.78, + "shared_filenames": [ + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 40 + ], + [ + 173, + 180 + ], + [ + 249, + 256 + ], + [ + 397, + 405 + ], + [ + 408, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 465 + ], + [ + 470, + 489 + ], + [ + 517, + 539 + ], + [ + 553, + 559 + ], + [ + 565, + 574 + ], + [ + 579, + 597 + ], + [ + 600, + 611 + ], + [ + 633, + 639 + ], + [ + 645, + 652 + ], + [ + 657, + 673 + ], + [ + 716, + 728 + ], + [ + 737, + 743 + ], + [ + 748, + 771 + ], + [ + 789, + 794 + ] + ], + "right_ranges": [ + [ + 33, + 42 + ], + [ + 175, + 182 + ], + [ + 251, + 258 + ], + [ + 399, + 420 + ], + [ + 426, + 435 + ], + [ + 460, + 467 + ], + [ + 472, + 489 + ], + [ + 519, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 612 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 663 + ], + [ + 666, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ] + } + ] + }, + { + "left_pr_number": 44129, + "right_pr_number": 44154, + "code_similarity": 0.099, + "size_similarity": 0.344, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.199, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44129, + "right_pr_number": 44722, + "code_similarity": 0.146, + "size_similarity": 0.545, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.247, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44154, + "right_pr_number": 44722, + "code_similarity": 0.164, + "size_similarity": 0.631, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.255, + "shared_filenames": [], + "shared_file_areas": [] + } + ] + }, + { + "cluster_id": "cluster-41211-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #41211.", + "status": "open", + "confidence": 0.55, + "canonical_issue_number": 41211, + "canonical_pr_number": 44339, + "issue_numbers": [ + 41211 + ], + "pr_numbers": [ + 41356, + 44339 + ], + "evidence_types": [ + "closing_reference" + ], + "pr_comparisons": [ + { + "left_pr_number": 41356, + "right_pr_number": 44339, + "code_similarity": 0.155, + "size_similarity": 0.094, + "file_overlap": 0.273, + "area_overlap": 0.078, + "patch_similarity": 0.183, + "shared_filenames": [ + "docs/source/en/model_doc/deimv2.md", + "src/transformers/models/deimv2/__init__.py", + "src/transformers/models/deimv2/configuration_deimv2.py", + "src/transformers/models/deimv2/modeling_deimv2.py", + "tests/models/deimv2/__init__.py", + "tests/models/deimv2/test_modeling_deimv2.py" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/model_doc/deimv2.md", + "left_ranges": [ + [ + 1, + 132 + ] + ], + "right_ranges": [ + [ + 1, + 65 + ] + ] + }, + { + "filename": "src/transformers/models/deimv2/__init__.py", + "left_ranges": [ + [ + 1, + 15 + ] + ], + "right_ranges": [ + [ + 1, + 29 + ] + ] + }, + { + "filename": "src/transformers/models/deimv2/configuration_deimv2.py", + "left_ranges": [ + [ + 1, + 74 + ] + ], + "right_ranges": [ + [ + 1, + 266 + ] + ] + }, + { + "filename": "tests/models/deimv2/test_modeling_deimv2.py", + "left_ranges": [ + [ + 1, + 15 + ] + ], + "right_ranges": [ + [ + 1, + 1734 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-43240-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #43240.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 43240, + "canonical_pr_number": 43251, + "issue_numbers": [ + 43240 + ], + "pr_numbers": [ + 43251, + 43254 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 43251, + "right_pr_number": 43254, + "code_similarity": 0.794, + "size_similarity": 0.64, + "file_overlap": 1.0, + "area_overlap": 0.667, + "patch_similarity": 0.882, + "shared_filenames": [ + "src/transformers/loss/loss_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/loss/loss_utils.py", + "left_ranges": [ + [ + 30, + 50 + ] + ], + "right_ranges": [ + [ + 30, + 43 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-41115-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #41115.", + "status": "open", + "confidence": 0.55, + "canonical_issue_number": 41115, + "canonical_pr_number": 45613, + "issue_numbers": [ + 41115 + ], + "pr_numbers": [ + 41116, + 45613 + ], + "evidence_types": [ + "closing_reference" + ], + "pr_comparisons": [ + { + "left_pr_number": 41116, + "right_pr_number": 45613, + "code_similarity": 0.797, + "size_similarity": 0.885, + "file_overlap": 0.75, + "area_overlap": 0.831, + "patch_similarity": 0.693, + "shared_filenames": [ + "docs/source/en/_toctree.yml", + "docs/source/en/model_doc/minicpm3.md", + "src/transformers/models/auto/modeling_auto.py", + "src/transformers/models/minicpm3/__init__.py", + "src/transformers/models/minicpm3/configuration_minicpm3.py", + "src/transformers/models/minicpm3/modeling_minicpm3.py", + "src/transformers/models/minicpm3/modular_minicpm3.py", + "tests/models/minicpm3/__init__.py", + "tests/models/minicpm3/test_modeling_minicpm3.py" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/_toctree.yml", + "left_ranges": [ + [ + 713, + 720 + ] + ], + "right_ranges": [ + [ + 730, + 737 + ] + ] + }, + { + "filename": "docs/source/en/model_doc/minicpm3.md", + "left_ranges": [ + [ + 1, + 67 + ] + ], + "right_ranges": [ + [ + 1, + 45 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 285, + 291 + ], + [ + 700, + 706 + ], + [ + 1298, + 1304 + ] + ], + "right_ranges": [ + [ + 720, + 726 + ], + [ + 1331, + 1337 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/__init__.py", + "left_ranges": [ + [ + 1, + 29 + ] + ], + "right_ranges": [ + [ + 1, + 29 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/configuration_minicpm3.py", + "left_ranges": [ + [ + 1, + 141 + ] + ], + "right_ranges": [ + [ + 1, + 126 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/modeling_minicpm3.py", + "left_ranges": [ + [ + 1, + 544 + ] + ], + "right_ranges": [ + [ + 1, + 522 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/modular_minicpm3.py", + "left_ranges": [ + [ + 1, + 444 + ] + ], + "right_ranges": [ + [ + 1, + 342 + ] + ] + }, + { + "filename": "tests/models/minicpm3/test_modeling_minicpm3.py", + "left_ranges": [ + [ + 1, + 109 + ] + ], + "right_ranges": [ + [ + 1, + 136 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-45647-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45647.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 45647, + "canonical_pr_number": 45738, + "issue_numbers": [ + 45647 + ], + "pr_numbers": [ + 45737, + 45738 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 45737, + "right_pr_number": 45738, + "code_similarity": 0.526, + "size_similarity": 0.11, + "file_overlap": 0.5, + "area_overlap": 1.0, + "patch_similarity": 0.024, + "shared_filenames": [ + "src/transformers/models/musicgen_melody/modeling_musicgen_melody.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/musicgen_melody/modeling_musicgen_melody.py", + "left_ranges": [ + [ + 491, + 497 + ] + ], + "right_ranges": [ + [ + 491, + 497 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-45561-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45561.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 45561, + "canonical_pr_number": 45645, + "issue_numbers": [ + 45561 + ], + "pr_numbers": [ + 45639, + 45645 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 45639, + "right_pr_number": 45645, + "code_similarity": 0.494, + "size_similarity": 0.619, + "file_overlap": 0.5, + "area_overlap": 0.434, + "patch_similarity": 0.451, + "shared_filenames": [ + "src/transformers/testing_utils.py", + "tests/utils/test_testing_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/testing_utils.py", + "left_ranges": [ + [ + 3595, + 3602 + ], + [ + 3819, + 3853 + ] + ], + "right_ranges": [ + [ + 3525, + 3558 + ], + [ + 3782, + 3788 + ] + ] + }, + { + "filename": "tests/utils/test_testing_utils.py", + "left_ranges": [ + [ + 1, + 86 + ] + ], + "right_ranges": [ + [ + 1, + 114 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-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-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-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 13 discussion comments plus 8 inbound references.", + "score": 164.0 + }, + "best_pr": { + "cluster_id": "cluster-43979-11", + "pr_number": 44007, + "reason": "PR #44007 is the strongest global PR candidate because it is open, belongs to a cluster with 11 artifacts, links to 1 issues, and carries 2 review/discussion events.", + "score": 238.886 + } +} diff --git a/snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/analysis-report-hybrid.llm-reviews.json b/snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/analysis-report-hybrid.llm-reviews.json new file mode 100644 index 0000000000000000000000000000000000000000..e5bc9f0e6392900462bfeca791f1a9453b6ebcf9 --- /dev/null +++ b/snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/analysis-report-hybrid.llm-reviews.json @@ -0,0 +1,20997 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260601T060039Z", + "generated_at": "2026-06-01T06:08:58Z", + "model": "gpt-5.4-mini?service_tier=flex", + "reviews": [ + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 1, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:24643", + "issue:30277", + "issue:34634", + "issue:34928", + "issue:35545", + "issue:36010", + "issue:36331", + "issue:39290", + "issue:41720", + "issue:41762", + "issue:42915", + "issue:43039", + "issue:43296", + "issue:43425", + "issue:43541", + "issue:43572", + "issue:43596", + "issue:43638", + "issue:43716", + "issue:43827", + "issue:43828", + "issue:43901", + "issue:43975", + "issue:44292", + "issue:44322", + "issue:44457", + "issue:44509", + "issue:44512", + "issue:44560", + "issue:44589", + "issue:44805", + "issue:44918", + "issue:45081", + "issue:45161", + "issue:45237", + "issue:45464", + "issue:45507", + "issue:45685", + "issue:45800", + "issue:45823", + "issue:45859", + "issue:45874", + "issue:45901", + "issue:45902", + "issue:46015", + "issue:46165", + "issue:46294", + "issue:46304" + ], + "soft_pairs": [ + "issue:44322|issue:45464", + "issue:45237|issue:45685", + "issue:43716|issue:45237", + "issue:43828|issue:45237", + "issue:45237|issue:45507", + "issue:44292|issue:45237", + "issue:30277|issue:35545", + "issue:35545|issue:42915", + "issue:35545|issue:41720", + "issue:45901|issue:46015", + "issue:44509|issue:44512", + "issue:35545|issue:36010", + "issue:44805|issue:45823", + "issue:44918|issue:46165", + "issue:43638|issue:44805", + "issue:43638|issue:46165", + "issue:44805|issue:46294", + "issue:43596|issue:44805", + "issue:44805|issue:45161", + "issue:44560|issue:46165", + "issue:43596|issue:46165", + "issue:43572|issue:46165", + "issue:34634|issue:35545", + "issue:30277|issue:43638", + "issue:43596|issue:43638", + "issue:43975|issue:44457", + "issue:43039|issue:45161", + "issue:44292|issue:44457", + "issue:24643|issue:30277", + "issue:34928|issue:44805", + "issue:45859|issue:45902", + "issue:43638|issue:45161", + "issue:34928|issue:36331", + "issue:43596|issue:45161", + "issue:43716|issue:45507", + "issue:45874|issue:46294", + "issue:39290|issue:45081", + "issue:45823|issue:45874", + "issue:43425|issue:44292", + "issue:34928|issue:41720", + "issue:43828|issue:45161", + "issue:43541|issue:43716", + "issue:43828|issue:44292", + "issue:43716|issue:45685", + "issue:44918|issue:45161", + "issue:43716|issue:44292", + "issue:45507|issue:45685", + "issue:44292|issue:44918", + "issue:43039|issue:44918", + "issue:41762|issue:43638", + "issue:43827|issue:44512", + "issue:43716|issue:44560", + "issue:43828|issue:45685", + "issue:44918|issue:45685", + "issue:43039|issue:43828", + "issue:30277|issue:36331", + "issue:34928|issue:36010", + "issue:45237|issue:46015", + "issue:43039|issue:44805", + "issue:45800|issue:46304", + "issue:43296|issue:46304", + "issue:41762|issue:43596", + "issue:39290|issue:41720", + "issue:43901|issue:44512", + "issue:45823|issue:46294", + "issue:44589|issue:46304", + "issue:45237|issue:45901" + ], + "prepared_review_unit_hash": "df2da42a21c2ca132200e8d735a737a9d4e84ffd93a0e77528a5294eddbf6dfe", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 67, + "serialized_chars": 47824, + "estimated_input_tokens": 11956, + "estimated_eval_tokens": 24168 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly noisy and heterogeneous. The only clear duplicate family is the ZeRO-3 \"index 0 is out of bounds for dimension 0 with size 0\" loading failure (43596/43638, with 41762 as a close same-bug report), plus one plausible Qwen3.5 MoE rotary-embedding/offload duplicate (45859/45902). Most other soft links are false positives across different models, APIs, or docs topics.", + "confidence": 0.76, + "canonical_issue_reason": "No single canonical fits the whole mixed set; if one issue must represent the real duplicate family, 43638 is the best pick because it is the clearest ZeRO-3 empty-index loading failure and aligns with the exact duplicate reports around the same bug.", + "canonical_pr_reason": null, + "best_issue_reason": "43638 is the strongest issue to keep: it states the concrete ZeRO-3 index-out-of-bounds failure clearly, is reproducible, and best captures the main duplicate family in this set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44322", + "right": "issue:45464", + "accept": false, + "reason": "Different failure surface: config attribute access vs streaming chat/completions inference." + }, + { + "left": "issue:45237", + "right": "issue:45685", + "accept": false, + "reason": "Unrelated model/runtime bugs: AMD GPU failure vs MPS histogram op limitation." + }, + { + "left": "issue:43716", + "right": "issue:45237", + "accept": false, + "reason": "Different models and code paths; image preprocessor dtype mismatch is not the same as GPT-OSS AMD GPU failure." + }, + { + "left": "issue:43828", + "right": "issue:45237", + "accept": false, + "reason": "Both are runtime dtype-related only at a high level; the concrete model/code path is different." + }, + { + "left": "issue:45237", + "right": "issue:45507", + "accept": false, + "reason": "Different MoE/model failures; one is AMD GPU support, the other is an invalid method call." + }, + { + "left": "issue:44292", + "right": "issue:45237", + "accept": false, + "reason": "Different model families and different errors; NVFP4 runtime issue is not the same bug as GPT-OSS AMD GPU failure." + }, + { + "left": "issue:30277", + "right": "issue:35545", + "accept": false, + "reason": "DeepSpeed collectives mismatch and ONNX export failure are unrelated." + }, + { + "left": "issue:35545", + "right": "issue:42915", + "accept": false, + "reason": "ONNX export error vs FP8 config failure are different issues." + }, + { + "left": "issue:35545", + "right": "issue:41720", + "accept": false, + "reason": "Different problems: ONNX export vs auto device mapping CUDA assert." + }, + { + "left": "issue:45901", + "right": "issue:46015", + "accept": false, + "reason": "Table QA crash and RAG docs updates are unrelated." + }, + { + "left": "issue:44509", + "right": "issue:44512", + "accept": false, + "reason": "Both are v5 docs cleanups, but they target different obsolete references and not one concrete fix." + }, + { + "left": "issue:35545", + "right": "issue:36010", + "accept": false, + "reason": "ONNX export error and GenerationMixin import error are unrelated." + }, + { + "left": "issue:44805", + "right": "issue:45823", + "accept": false, + "reason": "Mask/tensor shape mismatch is a different failure than Gemma4 device-mapping mismatch." + }, + { + "left": "issue:44918", + "right": "issue:46165", + "accept": false, + "reason": "Different training failures; TRL SFT embedding unpacking is not the same as resume/trust-remote-code issues." + }, + { + "left": "issue:43638", + "right": "issue:44805", + "accept": false, + "reason": "Both are IndexError/shape-related, but the concrete bug and code path differ." + }, + { + "left": "issue:43638", + "right": "issue:46165", + "accept": false, + "reason": "ZeRO-3 loading failure and resume/trust-remote-code failure are unrelated." + }, + { + "left": "issue:44805", + "right": "issue:46294", + "accept": false, + "reason": "Mask shape mismatch and Qwen3.5 video chat template mismatch are different bugs." + }, + { + "left": "issue:43596", + "right": "issue:44805", + "accept": false, + "reason": "Same broad category of tensor errors, but not the same underlying failure." + }, + { + "left": "issue:44805", + "right": "issue:45161", + "accept": false, + "reason": "Mask shape mismatch is unrelated to GPT-OSS tensor parallelism." + }, + { + "left": "issue:44560", + "right": "issue:46165", + "accept": false, + "reason": "Video StopIteration and resume/trust-remote-code are different issues." + }, + { + "left": "issue:43596", + "right": "issue:46165", + "accept": false, + "reason": "ZeRO-3 index-out-of-bounds is unrelated to model resume/trust-remote-code." + }, + { + "left": "issue:43572", + "right": "issue:46165", + "accept": false, + "reason": "StableLm pad_token_idx regression is unrelated to the training resume issue." + }, + { + "left": "issue:34634", + "right": "issue:35545", + "accept": false, + "reason": "Processor voice_preset bug and ONNX export failure are unrelated." + }, + { + "left": "issue:30277", + "right": "issue:43638", + "accept": false, + "reason": "Different DeepSpeed failures: collectives mismatch vs empty-index loading error." + }, + { + "left": "issue:43596", + "right": "issue:43638", + "accept": true, + "reason": "Same ZeRO-3 index-0-out-of-bounds loading bug; the titles differ only in model/context wording." + }, + { + "left": "issue:43975", + "right": "issue:44457", + "accept": false, + "reason": "Different tokenization/LoRA issues with no shared concrete failure mode." + }, + { + "left": "issue:43039", + "right": "issue:45161", + "accept": false, + "reason": "Liger Kernel cross-entropy invocation is unrelated to GPT-OSS tensor parallel failures." + }, + { + "left": "issue:44292", + "right": "issue:44457", + "accept": false, + "reason": "Different model/runtime bugs; NVFP4 execution error is not the same as LoRA merge/save mismatch." + }, + { + "left": "issue:24643", + "right": "issue:30277", + "accept": false, + "reason": "Both mention DeepSpeed, but one is a 2-D weight error and the other is a collective mismatch." + }, + { + "left": "issue:34928", + "right": "issue:44805", + "accept": false, + "reason": "Activation-checkpointing/FSDP recompute mismatch is not the same as the mask-shape IndexError." + }, + { + "left": "issue:45859", + "right": "issue:45902", + "accept": true, + "reason": "Same Qwen3.5 MoE rotary-embedding/offload bug; one report describes the symptom, the other pinpoints the uninitialized inv_freq materialization path." + }, + { + "left": "issue:43638", + "right": "issue:45161", + "accept": false, + "reason": "ZeRO-3 loading failure and GPT-OSS tensor-parallel issue are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36331", + "accept": false, + "reason": "FSDP checkpointing mismatch and compute_loss signature change are different issues." + }, + { + "left": "issue:43596", + "right": "issue:45161", + "accept": false, + "reason": "ZeRO-3 loading error and tensor-parallel failure are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:45507", + "accept": false, + "reason": "Different models and different bugs; dtype mismatch is not the same as invalid method call." + }, + { + "left": "issue:45874", + "right": "issue:46294", + "accept": false, + "reason": "Both are Gemma/Qwen multimodal issues, but the code paths are different (reverse embedding lookup vs video template token mismatch)." + }, + { + "left": "issue:39290", + "right": "issue:45081", + "accept": false, + "reason": "Different model-specific regressions: Gemma3 config attribute vs Mistral tokenizer regex crash." + }, + { + "left": "issue:45823", + "right": "issue:45874", + "accept": false, + "reason": "Both are Gemma4 issues, but they affect different mechanisms and are not the same concrete bug." + }, + { + "left": "issue:43425", + "right": "issue:44292", + "accept": false, + "reason": "Torch version incompatibility and Qwen-3 NVFP4 runtime error are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:41720", + "accept": false, + "reason": "Different distributed-training failures with different triggers and error modes." + }, + { + "left": "issue:43828", + "right": "issue:45161", + "accept": false, + "reason": "Autocast dtype mismatch and GPT-OSS tensor-parallel failure are different bugs." + }, + { + "left": "issue:43541", + "right": "issue:43716", + "accept": false, + "reason": "Grouped_mm tracing failure and image-preprocessor dtype mismatch are unrelated." + }, + { + "left": "issue:43828", + "right": "issue:44292", + "accept": false, + "reason": "Both are runtime model errors, but the concrete model and failure mode differ." + }, + { + "left": "issue:43716", + "right": "issue:45685", + "accept": false, + "reason": "Different model/runtime backends and different error types." + }, + { + "left": "issue:44918", + "right": "issue:45161", + "accept": false, + "reason": "TRL SFT embedding unpacking and GPT-OSS tensor-parallel failures are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:44292", + "accept": false, + "reason": "Different model families and failure mechanisms." + }, + { + "left": "issue:45507", + "right": "issue:45685", + "accept": false, + "reason": "Invalid method call is not the same bug as the MPS histogram limitation." + }, + { + "left": "issue:44292", + "right": "issue:44918", + "accept": false, + "reason": "Different Qwen/GPT-OSS training/runtime bugs with no shared concrete code path." + }, + { + "left": "issue:43039", + "right": "issue:44918", + "accept": false, + "reason": "Liger Kernel cross-entropy call and TRL embedding unpacking are unrelated." + }, + { + "left": "issue:41762", + "right": "issue:43638", + "accept": true, + "reason": "Same ZeRO-3 index-out-of-bounds loading failure; different model names, same concrete failure mode." + }, + { + "left": "issue:43827", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs regressions from v5 removals, but they point to different obsolete references." + }, + { + "left": "issue:43716", + "right": "issue:44560", + "accept": false, + "reason": "Different multimodal bugs: dtype mismatch vs video StopIteration." + }, + { + "left": "issue:43828", + "right": "issue:45685", + "accept": false, + "reason": "Different runtime issues with no shared underlying bug." + }, + { + "left": "issue:44918", + "right": "issue:45685", + "accept": false, + "reason": "TRL trainer unpacking and MPS histogram limitations are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:43828", + "accept": false, + "reason": "Different failures: Liger Kernel cross-entropy vs autocast dtype mismatch." + }, + { + "left": "issue:30277", + "right": "issue:36331", + "accept": false, + "reason": "DeepSpeed collectives mismatch and compute_loss signature mismatch are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36010", + "accept": false, + "reason": "Activation-checkpointing/FSDP issue and GenerationMixin import failure are unrelated." + }, + { + "left": "issue:45237", + "right": "issue:46015", + "accept": false, + "reason": "GPT-OSS AMD GPU runtime issue and outdated RAG examples are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:44805", + "accept": false, + "reason": "Liger Kernel cross-entropy issue and mask-shape IndexError are unrelated." + }, + { + "left": "issue:45800", + "right": "issue:46304", + "accept": false, + "reason": "Torch/transformers compatibility problems, but different versions and different breakage." + }, + { + "left": "issue:43296", + "right": "issue:46304", + "accept": false, + "reason": "PaddleOCR-VL/vLLM loading failure and USE_DISTRIBUTED=0 regression are unrelated." + }, + { + "left": "issue:45823", + "right": "issue:46294", + "accept": false, + "reason": "Gemma4 device-mapping and Qwen3.5 video chat-template issues are different code paths." + }, + { + "left": "issue:44589", + "right": "issue:46304", + "accept": false, + "reason": "Float8 storage lookup error and distributed-disabled compatibility break are unrelated." + }, + { + "left": "issue:45237", + "right": "issue:45901", + "accept": false, + "reason": "GPT-OSS AMD GPU runtime bug and table-question-answering crash are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet and stays conservative. The ZeRO-3 index-out-of-bounds family (43596/43638, with 41762 as a near-match) is clearly supported by the titles, and the Qwen3.5 MoE rotary/offload pair (45859/45902) is a plausible same-bug cluster rather than an overclaim. The remaining false links are appropriately treated as heterogeneous across models, APIs, runtime backends, or docs." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 2, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:36010", + "issue:39290", + "issue:41093", + "issue:41762", + "issue:43039", + "issue:43296", + "issue:43366", + "issue:43404", + "issue:43425", + "issue:43531", + "issue:43541", + "issue:43572", + "issue:43596", + "issue:43638", + "issue:43716", + "issue:43792", + "issue:43828", + "issue:43854", + "issue:43866", + "issue:43872", + "issue:43975", + "issue:44291", + "issue:44292", + "issue:44387", + "issue:44479", + "issue:44560", + "issue:44589", + "issue:44805", + "issue:44841", + "issue:44863", + "issue:44918", + "issue:45005", + "issue:45020", + "issue:45070", + "issue:45084", + "issue:45161", + "issue:45362", + "issue:45464", + "issue:45685", + "issue:45800", + "issue:45812", + "issue:45823", + "issue:45901", + "issue:45920", + "issue:46015", + "issue:46294", + "issue:46304" + ], + "soft_pairs": [ + "issue:44841|issue:45823", + "issue:43039|issue:44560", + "issue:43975|issue:46015", + "issue:45812|issue:45920", + "issue:44863|issue:46015", + "issue:45020|issue:46304", + "issue:43866|issue:44863", + "issue:44292|issue:46015", + "issue:43872|issue:46304", + "issue:39290|issue:43531", + "issue:43854|issue:43866", + "issue:41093|issue:43638", + "issue:43792|issue:45901", + "issue:43425|issue:45070", + "issue:43596|issue:45800", + "issue:43975|issue:45901", + "issue:43792|issue:45070", + "issue:45362|issue:45464", + "issue:44292|issue:45901", + "issue:39290|issue:45823", + "issue:43404|issue:43866", + "issue:45464|issue:45685", + "issue:43541|issue:45685", + "issue:43541|issue:43828", + "issue:44841|issue:45084", + "issue:36010|issue:39290", + "issue:45800|issue:45823", + "issue:41093|issue:41762", + "issue:43296|issue:43366", + "issue:43572|issue:45070", + "issue:43541|issue:45161", + "issue:44387|issue:45005", + "issue:44805|issue:44918", + "issue:43828|issue:45464", + "issue:44291|issue:45070", + "issue:43366|issue:45464", + "issue:43716|issue:46294", + "issue:43531|issue:45070", + "issue:45020|issue:45800", + "issue:44479|issue:46294", + "issue:43541|issue:44589", + "issue:41093|issue:43596" + ], + "prepared_review_unit_hash": "1e21385230bc1cfe73c9ee0668f983d7b06af469c67bb25e4d4a4cd93c233e96", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 42, + "serialized_chars": 39564, + "estimated_input_tokens": 9891, + "estimated_eval_tokens": 20038 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most soft pairs are false positives across unrelated model/runtime bugs. Two pairs look like genuine duplicates: the AutoTokenizer wrong-token-ID regression and the Qwen video-input regression.", + "confidence": 0.82, + "canonical_issue_reason": "issue:45920 is the strongest canonical issue candidate because it describes the broadest and clearest AutoTokenizer v5 regression, spanning multiple model families and subsuming the more specific Granite-only report in issue:45812.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45920 is the best overall issue representative: broad impact, explicit regression framing, and the most general statement of the shared bug in this set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44841", + "right": "issue:45823", + "accept": false, + "reason": "Different failures on different models: Voxtral processor load issue vs Gemma4 device-mismatch during forward." + }, + { + "left": "issue:43039", + "right": "issue:44560", + "accept": false, + "reason": "Unrelated bugs: Liger Kernel cross-entropy path vs Qwen3-VL video StopIteration." + }, + { + "left": "issue:43975", + "right": "issue:46015", + "accept": false, + "reason": "Detokenization regression vs unrelated RAG documentation issue." + }, + { + "left": "issue:45812", + "right": "issue:45920", + "accept": true, + "reason": "Same underlying AutoTokenizer v5 regression producing wrong token IDs; 45920 is the broader report across more model families." + }, + { + "left": "issue:44863", + "right": "issue:46015", + "accept": false, + "reason": "NemotronH checkpoint-loading failure is unrelated to outdated RAG examples." + }, + { + "left": "issue:45020", + "right": "issue:46304", + "accept": false, + "reason": "Both mention version breakage, but one is about remote_code models and the other about USE_DISTRIBUTED=0 support; not the same bug." + }, + { + "left": "issue:43866", + "right": "issue:44863", + "accept": false, + "reason": "Both are load failures, but for different model families and different symptoms; not the same underlying issue." + }, + { + "left": "issue:44292", + "right": "issue:46015", + "accept": false, + "reason": "init_empty_weights/_is_hf_initialized TypeError is unrelated to RAG examples." + }, + { + "left": "issue:43872", + "right": "issue:46304", + "accept": false, + "reason": "bitsandbytes Int8Params constructor incompatibility is different from USE_DISTRIBUTED=0 breakage." + }, + { + "left": "issue:39290", + "right": "issue:43531", + "accept": false, + "reason": "Both involve sliding-window concepts, but the affected configs and failure modes differ (Gemma3TextConfig vs Qwen3-MoE)." + }, + { + "left": "issue:43854", + "right": "issue:43866", + "accept": false, + "reason": "Different models and different loading problems; no evidence of one shared bug." + }, + { + "left": "issue:41093", + "right": "issue:43638", + "accept": false, + "reason": "Both are zero3/index errors, but the concrete failures differ: mask shape mismatch vs size-0 index out of bounds." + }, + { + "left": "issue:43792", + "right": "issue:45901", + "accept": false, + "reason": "Whisper model load failure is unrelated to table-question-answering crashes." + }, + { + "left": "issue:43975", + "right": "issue:45901", + "accept": false, + "reason": "Detokenization bug and table QA crash are unrelated." + }, + { + "left": "issue:43792", + "right": "issue:45070", + "accept": false, + "reason": "Whisper runtime issue is unrelated to a PretrainedConfig pydantic regression." + }, + { + "left": "issue:45362", + "right": "issue:45464", + "accept": false, + "reason": "Both are Qwen3.5-related, but one is a transformers chat crash and the other is a streaming chat/completions API failure." + }, + { + "left": "issue:44292", + "right": "issue:45901", + "accept": false, + "reason": "Different subsystems and different errors; no shared bug." + }, + { + "left": "issue:39290", + "right": "issue:45823", + "accept": false, + "reason": "Gemma3 sliding_window_pattern issue is unrelated to Gemma4 device mismatch." + }, + { + "left": "issue:43404", + "right": "issue:43866", + "accept": false, + "reason": "Tied lm_head weights bug vs corrupted checkpoint/loading failure; not the same issue." + }, + { + "left": "issue:45464", + "right": "issue:45685", + "accept": false, + "reason": "Qwen3.5 streaming inference failure is unrelated to the MPS histogram Int error." + }, + { + "left": "issue:43541", + "right": "issue:45685", + "accept": false, + "reason": "Grouped_mm tracing error and MPS histogram error are different code paths and symptoms." + }, + { + "left": "issue:43541", + "right": "issue:43828", + "accept": false, + "reason": "Both mention dtype-ish behavior, but one is torch dynamo grouped_mm tracing and the other is an autocast mismatch." + }, + { + "left": "issue:44841", + "right": "issue:45084", + "accept": false, + "reason": "Processor failure for Voxtral is unrelated to template-node compilation errors." + }, + { + "left": "issue:36010", + "right": "issue:39290", + "accept": false, + "reason": "ImportError for GenerationMixin and Gemma3 sliding-window config are unrelated." + }, + { + "left": "issue:45800", + "right": "issue:45823", + "accept": false, + "reason": "General torch/transformers version incompatibility is unrelated to Gemma4 device placement." + }, + { + "left": "issue:41093", + "right": "issue:41762", + "accept": false, + "reason": "Both are index errors, but the concrete contexts and failure signatures differ." + }, + { + "left": "issue:43296", + "right": "issue:43366", + "accept": false, + "reason": "Different model-support requests: PaddleOCR-VL in vLLM vs GGUF gpt-oss architecture support." + }, + { + "left": "issue:43572", + "right": "issue:45070", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is not the same as a broken PretrainedConfig pydantic field." + }, + { + "left": "issue:43541", + "right": "issue:45161", + "accept": false, + "reason": "Different MoE/parallelism issues with no shared concrete code path." + }, + { + "left": "issue:44387", + "right": "issue:45005", + "accept": false, + "reason": "Int4 quantization OOM and tied-weights translation bugs are unrelated." + }, + { + "left": "issue:44805", + "right": "issue:44918", + "accept": false, + "reason": "Both are tensor-shape style errors, but the model/trainer contexts differ and the bug signatures do not match closely enough." + }, + { + "left": "issue:43828", + "right": "issue:45464", + "accept": false, + "reason": "Autocast dtype mismatch for Phi-tiny-MoE is unrelated to Qwen3.5 streaming API failure." + }, + { + "left": "issue:44291", + "right": "issue:45070", + "accept": false, + "reason": "Unexpected _is_hf_initialized argument during init_empty_weights is not the same as the PretrainedConfig pydantic-field regression." + }, + { + "left": "issue:43366", + "right": "issue:45464", + "accept": false, + "reason": "GGUF gpt-oss support request is unrelated to Qwen3.5 streaming inference failure." + }, + { + "left": "issue:43716", + "right": "issue:46294", + "accept": false, + "reason": "Both are multimodal, but Mistral image-preprocessor dtype mismatch and Qwen3.5 video token/feature mismatch are different bugs." + }, + { + "left": "issue:43531", + "right": "issue:45070", + "accept": false, + "reason": "Qwen3-MoE sliding_window bug is unrelated to the PretrainedConfig pydantic regression." + }, + { + "left": "issue:45020", + "right": "issue:45800", + "accept": false, + "reason": "Remote_code breakage and torch 2.4.1/transformers 5.8.0 incompatibility are only loosely related by versioning." + }, + { + "left": "issue:44479", + "right": "issue:46294", + "accept": true, + "reason": "Same Qwen video-input regression: both report video-feature/video-token mismatches in the v5.3.x line, with 44479 being the broader umbrella report." + }, + { + "left": "issue:43541", + "right": "issue:44589", + "accept": false, + "reason": "Grouped_mm tracing failure is unrelated to missing Float8 storage support." + }, + { + "left": "issue:41093", + "right": "issue:43596", + "accept": false, + "reason": "Both are zero3-related index errors, but the concrete failure modes and contexts differ too much to treat as duplicates." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The canonical issue choice (45920) is well supported by the titles, and the two accepted soft pairs (45812/45920 and 44479/46294) look like plausible duplicates based on shared bug descriptions. The remaining verdicts are phrased cautiously and do not overclaim beyond the available evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 3, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:36331", + "issue:38175", + "issue:39186", + "issue:39692", + "issue:40990", + "issue:41553", + "issue:41762", + "issue:42915", + "issue:42947", + "issue:43054", + "issue:43404", + "issue:43644", + "issue:43645", + "issue:43716", + "issue:43828", + "issue:43866", + "issue:43872", + "issue:43873", + "issue:43883", + "issue:43950", + "issue:43975", + "issue:43994", + "issue:44292", + "issue:44360", + "issue:44387", + "issue:44403", + "issue:44492", + "issue:44512", + "issue:44530", + "issue:44534", + "issue:44589", + "issue:44863", + "issue:44910", + "issue:44918", + "issue:44928", + "issue:44960", + "issue:45005", + "issue:45399", + "issue:45507", + "issue:45542", + "issue:45632", + "issue:45663", + "issue:45698", + "issue:45701", + "issue:45759", + "issue:45800", + "issue:45859", + "issue:45902" + ], + "soft_pairs": [ + "issue:43054|issue:43994", + "issue:45759|issue:45902", + "issue:45542|issue:45800", + "issue:43716|issue:44918", + "issue:39186|issue:44918", + "issue:45632|issue:45698", + "issue:43645|issue:45399", + "issue:43404|issue:45005", + "issue:40990|issue:44960", + "issue:44530|issue:45005", + "issue:42915|issue:45399", + "issue:43873|issue:45005", + "issue:44360|issue:44512", + "issue:44863|issue:45005", + "issue:44918|issue:45507", + "issue:44492|issue:44512", + "issue:43975|issue:44292", + "issue:44292|issue:45507", + "issue:43950|issue:44534", + "issue:39692|issue:43054", + "issue:43644|issue:43950", + "issue:43828|issue:44928", + "issue:36331|issue:41762", + "issue:43950|issue:45902", + "issue:43872|issue:44589", + "issue:43866|issue:44403", + "issue:38175|issue:43054", + "issue:41553|issue:43883", + "issue:43054|issue:45701", + "issue:45663|issue:45902", + "issue:44910|issue:45859", + "issue:42947|issue:44387" + ], + "prepared_review_unit_hash": "18ce39af92b151dc6c9b23b49b3468a555a76e3656885efe581f1be7b438ad3c", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37142, + "estimated_input_tokens": 9286, + "estimated_eval_tokens": 18828 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous. The only clear duplicate subgroup is the Transformers v5 non-persistent-buffer corruption regression; all other soft pairs look like separate bugs or unrelated model-specific issues.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 43950 is the most explicit and descriptive statement of the non-persistent-buffer corruption regression, making it the best canonical anchor for that duplicate subgroup.", + "canonical_pr_reason": null, + "best_issue_reason": "43950 best represents the only clearly duplicated bug family in this cluster: it names the regression, the affected API, and the bad buffer behavior directly.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43054", + "right": "issue:43994", + "accept": false, + "reason": "Both involve SigLIP2 output quality, but the reported symptoms and contexts differ enough that they do not clearly describe the same concrete bug." + }, + { + "left": "issue:45759", + "right": "issue:45902", + "accept": false, + "reason": "One is about AutoModelForCausalLM config unwrapping; the other is about uninitialized rotary embedding memory after materialization. Different code paths and failures." + }, + { + "left": "issue:45542", + "right": "issue:45800", + "accept": false, + "reason": "TensorBoard-without-TensorFlow backend detection is unrelated to the torch/transformers compatibility report." + }, + { + "left": "issue:43716", + "right": "issue:44918", + "accept": false, + "reason": "Different models and failures: image preprocessor dtype mismatch vs Qwen3.5 embedding unpacking with TRL SFTTrainer." + }, + { + "left": "issue:39186", + "right": "issue:44918", + "accept": false, + "reason": "FSDP weight-dimension failure is not the same bug as Qwen3.5 input-embedding unpacking." + }, + { + "left": "issue:45632", + "right": "issue:45698", + "accept": false, + "reason": "Both touch custom-module loading, but one is a leaf-directory cache collision and the other is wrong module selection after save_pretrained; not the same defect." + }, + { + "left": "issue:43645", + "right": "issue:45399", + "accept": false, + "reason": "Notebook custom-model initialization and flash-attn2 fallback gating are unrelated issues." + }, + { + "left": "issue:43404", + "right": "issue:45005", + "accept": false, + "reason": "Both mention tied weights, but these are different models and different failure modes." + }, + { + "left": "issue:40990", + "right": "issue:44960", + "accept": false, + "reason": "Different models and symptoms; no concrete shared bug is evident." + }, + { + "left": "issue:44530", + "right": "issue:45005", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash is unrelated to tied-weight handling." + }, + { + "left": "issue:42915", + "right": "issue:45399", + "accept": false, + "reason": "FineGrainedFP8 failure on Qwen3Moe is unrelated to the flash-attn2 fallback check ordering bug." + }, + { + "left": "issue:43873", + "right": "issue:45005", + "accept": false, + "reason": "Quantization/offloading behavior is not the same as tied-weight handling in translation models." + }, + { + "left": "issue:44360", + "right": "issue:44512", + "accept": false, + "reason": "DSA indexer behavior is unrelated to the removed run-command documentation issue." + }, + { + "left": "issue:44863", + "right": "issue:45005", + "accept": false, + "reason": "NemotronH checkpoint loading failure is unrelated to translation-model tied weights." + }, + { + "left": "issue:44918", + "right": "issue:45507", + "accept": false, + "reason": "Qwen3.5 embedding unpacking and GraniteMoEHybrid invalid-method calls are different problems." + }, + { + "left": "issue:44492", + "right": "issue:44512", + "accept": false, + "reason": "A cache-strategy typo is not the same as stale documentation about a removed command." + }, + { + "left": "issue:43975", + "right": "issue:44292", + "accept": false, + "reason": "Detokenization regression and NVFP4 runtime error are different failures." + }, + { + "left": "issue:44292", + "right": "issue:45507", + "accept": false, + "reason": "Different models and code paths; no duplicate relationship is apparent." + }, + { + "left": "issue:43950", + "right": "issue:44534", + "accept": true, + "reason": "These are the same non-persistent-buffer corruption regression described with different wording." + }, + { + "left": "issue:39692", + "right": "issue:43054", + "accept": false, + "reason": "Documentation example errors are not the same as the reported SigLIP2 embedding-quality issue." + }, + { + "left": "issue:43644", + "right": "issue:43950", + "accept": true, + "reason": "Both report the same v5 regression where non-persistent buffers are filled/corrupted with junk." + }, + { + "left": "issue:43828", + "right": "issue:44928", + "accept": false, + "reason": "Both mention dtype/BF16 issues, but the model, trigger, and failure mode are different." + }, + { + "left": "issue:36331", + "right": "issue:41762", + "accept": false, + "reason": "Trainer signature mismatch and ZeRO-3 load-time index error are unrelated." + }, + { + "left": "issue:43950", + "right": "issue:45902", + "accept": false, + "reason": "Both involve buffer/materialization concerns, but 45902 is a model-specific uninitialized-memory bug, not the same regression as 43950." + }, + { + "left": "issue:43872", + "right": "issue:44589", + "accept": false, + "reason": "Different low-level quantization/storage incompatibilities." + }, + { + "left": "issue:43866", + "right": "issue:44403", + "accept": false, + "reason": "Corrupted checkpoint and loading noise are unrelated symptoms." + }, + { + "left": "issue:38175", + "right": "issue:43054", + "accept": false, + "reason": "Same model family, but the observed problems differ too much to treat as the same underlying bug." + }, + { + "left": "issue:41553", + "right": "issue:43883", + "accept": false, + "reason": "Tokenizer error messaging and missing tied-weight metadata are unrelated." + }, + { + "left": "issue:43054", + "right": "issue:45701", + "accept": false, + "reason": "SigLIP2 embedding quality and version-dependent tokenization changes are different issues." + }, + { + "left": "issue:45663", + "right": "issue:45902", + "accept": false, + "reason": "Gemma4 FSDP2 shared_kv_states handling is unrelated to Qwen3_5Moe rotary embedding initialization." + }, + { + "left": "issue:44910", + "right": "issue:45859", + "accept": false, + "reason": "FlashAttention position_ids misinterpretation and CPU-offload incompatibility are different bugs." + }, + { + "left": "issue:42947", + "right": "issue:44387", + "accept": false, + "reason": "Gradient checkpointing ineffectiveness and int4 quantization memory growth are different issues." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative on the soft edges. The accepted duplicate subgroup around issue 43950 is well supported by the titles, and the other pair rejections do not overclaim sameness. No clear evidence of an overstated link." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 4, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:36331", + "issue:41093", + "issue:41553", + "issue:41762", + "issue:42491", + "issue:42915", + "issue:42947", + "issue:43054", + "issue:43257", + "issue:43404", + "issue:43493", + "issue:43643", + "issue:43742", + "issue:43782", + "issue:43792", + "issue:43824", + "issue:43827", + "issue:43828", + "issue:43856", + "issue:43866", + "issue:43872", + "issue:43881", + "issue:43883", + "issue:43940", + "issue:43950", + "issue:43975", + "issue:44220", + "issue:44291", + "issue:44368", + "issue:44403", + "issue:44451", + "issue:44488", + "issue:44509", + "issue:44661", + "issue:44960", + "issue:44991", + "issue:45020", + "issue:45081", + "issue:45305", + "issue:45313", + "issue:45356", + "issue:45399", + "issue:45520", + "issue:45685", + "issue:45715", + "issue:45823", + "issue:45923", + "issue:46315" + ], + "soft_pairs": [ + "issue:44403|issue:44451", + "issue:41553|issue:43881", + "issue:43975|issue:45520", + "issue:43054|issue:43493", + "issue:42947|issue:43856", + "issue:44403|issue:44488", + "issue:43643|issue:45020", + "issue:41762|issue:42915", + "issue:45685|issue:45823", + "issue:42491|issue:44368", + "issue:44991|issue:45399", + "issue:43824|issue:43872", + "issue:43881|issue:43883", + "issue:45020|issue:45823", + "issue:44661|issue:45020", + "issue:45715|issue:46315", + "issue:43828|issue:45823", + "issue:45081|issue:45399", + "issue:43782|issue:44960", + "issue:45020|issue:45356", + "issue:43742|issue:45399", + "issue:43404|issue:43950", + "issue:43827|issue:44509", + "issue:43792|issue:44220", + "issue:43872|issue:44291", + "issue:43257|issue:43866", + "issue:42915|issue:43872", + "issue:42947|issue:45923", + "issue:43940|issue:45313", + "issue:36331|issue:41093", + "issue:43257|issue:45305" + ], + "prepared_review_unit_hash": "32d19a0456792aad16c941d631f377220fcbd59f17d45fad59245dee87ba53ad", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37241, + "estimated_input_tokens": 9311, + "estimated_eval_tokens": 18878 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly a loose set of transformer/model-loading regressions and tokenizer/model-compatibility issues, but the soft edges are generally only thematic, not true duplicates. No PRs are present.", + "confidence": 0.86, + "canonical_issue_reason": "Issue 45020 is the broadest and most central issue here: it covers recent transformers regressions breaking `remote_code`-based model loading, which overlaps the general loading/compatibility theme of the cluster better than the narrower model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45020 is the best global issue candidate because it is the most umbrella-like problem statement and could reasonably subsume several of the model-loading/tokenizer regressions, whereas the others are mostly narrow model-specific failures or unrelated training/runtime bugs.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44403", + "right": "issue:44451", + "accept": false, + "reason": "Both concern loading behavior, but one is about warning/noise during loading while the other is a specific model load failure; not the same underlying bug." + }, + { + "left": "issue:41553", + "right": "issue:43881", + "accept": false, + "reason": "Different models and different failure modes: bad AutoTokenizer error for Voxtral vs a model loading failure for glm-4v-9b." + }, + { + "left": "issue:43975", + "right": "issue:45520", + "accept": false, + "reason": "One is a tokenizer detokenization regression, the other is a missing `flash_attn` import/KeyError on Python 3.13; unrelated code paths." + }, + { + "left": "issue:43054", + "right": "issue:43493", + "accept": false, + "reason": "Both mention SigLIP2, but one is a performance/quality complaint and the other is a correctness discrepancy versus JAX; not a duplicate bug." + }, + { + "left": "issue:42947", + "right": "issue:43856", + "accept": false, + "reason": "Both are training-related, but one is gradient checkpointing with PEFT LoRA and the other is memory usage in Qwen3 MoE training; different problems." + }, + { + "left": "issue:44403", + "right": "issue:44488", + "accept": false, + "reason": "Both are load-related, but they report different models and different symptoms; this looks like a family of regressions, not one bug." + }, + { + "left": "issue:43643", + "right": "issue:45020", + "accept": false, + "reason": "`trust_remote_code` returning missing fields is a specific config-object bug; 45020 is a broader loading regression across remote-code models." + }, + { + "left": "issue:41762", + "right": "issue:42915", + "accept": false, + "reason": "Different failures in different contexts: ZeRO-3 load-time IndexError for Gemma3 vs FineGrainedFP8Config failure for Qwen3 MoE." + }, + { + "left": "issue:45685", + "right": "issue:45823", + "accept": false, + "reason": "Both involve device/backend issues, but one is an MPS histogram implementation error in MoE and the other is a device mismatch in Gemma4 forward." + }, + { + "left": "issue:42491", + "right": "issue:44368", + "accept": false, + "reason": "One is a Qwen3 MoE LoRA compatibility break between versions; the other is a warning about `tie_word_embeddings` during fine-tuning. Different root causes." + }, + { + "left": "issue:44991", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer loading for a specific model is not the same as FA2 fallback being blocked by other checks; only loosely related through model loading." + }, + { + "left": "issue:43824", + "right": "issue:43872", + "accept": false, + "reason": "Different incompatibilities: missing Qwen2_5_VL import vs bitsandbytes `Int8Params` keyword mismatch." + }, + { + "left": "issue:43881", + "right": "issue:43883", + "accept": false, + "reason": "Both are loading failures, but one is model loading and the other is missing `all_tied_weights_keys` on a different model class; no shared concrete bug." + }, + { + "left": "issue:45020", + "right": "issue:45823", + "accept": false, + "reason": "`remote_code` loading regressions are not the same as a Gemma4 device_map runtime mismatch." + }, + { + "left": "issue:44661", + "right": "issue:45020", + "accept": false, + "reason": "`add-new-model-like` failing in `TOKENIZER_MAPPING_NAMES` is a tooling/extensibility bug, while 45020 is a runtime remote-code loading regression." + }, + { + "left": "issue:45715", + "right": "issue:46315", + "accept": false, + "reason": "Both involve tokenizer speed, but one is `convert_ids_to_tokens` rebuilding special-id sets; the other is slower `convert_tokens_to_ids` with many added tokens. Different hot paths." + }, + { + "left": "issue:43828", + "right": "issue:45823", + "accept": false, + "reason": "Autocast dtype mismatch in Phi MoE and Gemma4 device mismatch are different forward-pass bugs." + }, + { + "left": "issue:45081", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer regex patch crash and FA2 fallback gating are unrelated code paths." + }, + { + "left": "issue:43782", + "right": "issue:44960", + "accept": false, + "reason": "Qwen3VL `weight_only=True` loading error and generic GLM5 issue do not point to the same underlying bug." + }, + { + "left": "issue:45020", + "right": "issue:45356", + "accept": false, + "reason": "Remote-code loading regressions and Kimi-K2.5 tokenizer codec handling are different issues; the latter only mentions a misleading warning as a secondary symptom." + }, + { + "left": "issue:43742", + "right": "issue:45399", + "accept": false, + "reason": "Loading facebook/MobileLLM-125M failing is unrelated to the FA2 fallback gating issue." + }, + { + "left": "issue:43404", + "right": "issue:43950", + "accept": false, + "reason": "Untied `lm_head` weights in Mistral3 and silent corruption of non-persistent buffers are different model state bugs." + }, + { + "left": "issue:43827", + "right": "issue:44509", + "accept": false, + "reason": "Both are docs about removed pipeline tasks, but they concern different documentation locations/symptoms and are not the same concrete issue." + }, + { + "left": "issue:43792", + "right": "issue:44220", + "accept": false, + "reason": "Whisper load failure and `_torch_extract_fbank_features()` issue are not the same bug; only both live in audio/model loading." + }, + { + "left": "issue:43872", + "right": "issue:44291", + "accept": false, + "reason": "Both mention `_is_hf_initialized`, but one is a bitsandbytes incompatibility and the other is `init_empty_weights`/unexpected kwarg during loading; related but not the same concrete failure." + }, + { + "left": "issue:43257", + "right": "issue:43866", + "accept": false, + "reason": "Qwen3 MoE weight conversion with accelerate/deepspeed and corrupted Ovis2 checkpoint are distinct issues." + }, + { + "left": "issue:42915", + "right": "issue:43872", + "accept": false, + "reason": "FineGrainedFP8Config failure and bitsandbytes `_is_hf_initialized` incompatibility are different model-loading/runtime incompatibilities." + }, + { + "left": "issue:42947", + "right": "issue:45923", + "accept": false, + "reason": "Gradient checkpointing ineffective with LoRA and missing `supports_gradient_checkpointing` flag are related by topic, but not the same bug." + }, + { + "left": "issue:43940", + "right": "issue:45313", + "accept": false, + "reason": "Both involve DeepSpeed ZeRO-3 loading failures for Qwen variants, but they affect different model families and likely different missing-weight paths." + }, + { + "left": "issue:36331", + "right": "issue:41093", + "accept": false, + "reason": "`compute_loss()` signature mismatch and mask shape IndexError are completely unrelated training errors." + }, + { + "left": "issue:43257", + "right": "issue:45305", + "accept": false, + "reason": "One is Qwen3 MoE weight conversion with accelerate/deepspeed; the other is gradient averaging with GAS under DeepSpeed and `model_accepts_loss_kwargs=True`. Different code paths." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 5, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:24643", + "issue:30064", + "issue:36010", + "issue:38175", + "issue:39692", + "issue:41093", + "issue:41553", + "issue:42617", + "issue:42915", + "issue:43317", + "issue:43366", + "issue:43531", + "issue:43550", + "issue:43596", + "issue:43632", + "issue:43638", + "issue:43645", + "issue:43646", + "issue:43824", + "issue:43828", + "issue:43854", + "issue:43950", + "issue:43957", + "issue:43994", + "issue:44155", + "issue:44479", + "issue:44488", + "issue:44530", + "issue:44560", + "issue:44589", + "issue:44661", + "issue:44756", + "issue:44805", + "issue:44863", + "issue:44877", + "issue:44910", + "issue:44918", + "issue:44928", + "issue:44991", + "issue:45030", + "issue:45325", + "issue:45584", + "issue:45588", + "issue:45663", + "issue:45685", + "issue:45823", + "issue:45834", + "issue:45902" + ], + "soft_pairs": [ + "issue:43854|issue:44863", + "issue:44928|issue:45588", + "issue:41553|issue:44991", + "issue:41093|issue:44805", + "issue:44479|issue:44560", + "issue:44530|issue:44863", + "issue:41553|issue:44488", + "issue:24643|issue:30064", + "issue:45325|issue:45902", + "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:44560|issue:45685", + "issue:43366|issue:44589", + "issue:42915|issue:43957", + "issue:44560|issue:44918", + "issue:43824|issue:43957", + "issue:42617|issue:43366", + "issue:36010|issue:42915", + "issue:44661|issue:45823", + "issue:43632|issue:45834", + "issue:38175|issue:43994", + "issue:44155|issue:45584", + "issue:44560|issue:44805", + "issue:44877|issue:45030", + "issue:43638|issue:44661", + "issue:43596|issue:44661", + "issue:44661|issue:44805", + "issue:43317|issue:44756", + "issue:38175|issue:39692" + ], + "prepared_review_unit_hash": "466adf75844cdee42297c10caed381906fb4b9cb2d07445312b0f9b60abe3eff", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37723, + "estimated_input_tokens": 9431, + "estimated_eval_tokens": 19118 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous and does not look like one true duplicate set. The only convincing soft duplicate is the Transformers 5.0 custom model initialization regression pair (43645/43646).", + "confidence": 0.83, + "canonical_issue_reason": "issue:43646 is the broader report of the Transformers 5.0 custom model initialization regression; issue:43645 appears to be the same bug reproduced in a Jupyter notebook context.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 is the clearest representative of the only real duplicate-like problem in the set, and it states the underlying regression more generally than the notebook-specific variant.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43854", + "right": "issue:44863", + "accept": false, + "reason": "Different models and failure modes: GLM-4.7 loading in tests vs NemotronH checkpoint loading." + }, + { + "left": "issue:44928", + "right": "issue:45588", + "accept": false, + "reason": "Both are Qwen-related runtime bugs, but they hit different code paths: RLHF/3D position_ids vs flash_attention s_aux handling." + }, + { + "left": "issue:41553", + "right": "issue:44991", + "accept": false, + "reason": "Both involve tokenizer loading, but the reported models and symptoms differ; error-message quality vs outright tokenizer load failure." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": false, + "reason": "The errors look similarly worded, but there is no evidence they come from the same underlying off-by-one bug or code path." + }, + { + "left": "issue:44479", + "right": "issue:44560", + "accept": false, + "reason": "Both concern video inputs, but one is a v5.3.0 regression across Qwen VL models and the other is a Qwen3-vl-embedding StopIteration issue." + }, + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash and NemotronH checkpoint loading are unrelated." + }, + { + "left": "issue:41553", + "right": "issue:44488", + "accept": false, + "reason": "Different models and different failures: Voxtral tokenizer error messaging vs sleng-bert loading failure." + }, + { + "left": "issue:24643", + "right": "issue:30064", + "accept": false, + "reason": "Completely unrelated: DeepSpeed training weight shape error vs image processor void-segmentation-map handling." + }, + { + "left": "issue:45325", + "right": "issue:45902", + "accept": false, + "reason": "Both involve positional embeddings/indices, but they affect different models and distinct bugs." + }, + { + "left": "issue:43366", + "right": "issue:43828", + "accept": false, + "reason": "Unrelated: GGUF gpt-oss support vs Phi MoE autocast dtype mismatch." + }, + { + "left": "issue:45663", + "right": "issue:45823", + "accept": false, + "reason": "Same model family, but different concrete bugs: FSDP2 shared_kv_states rebuild vs device_map auto device mismatch." + }, + { + "left": "issue:43366", + "right": "issue:43531", + "accept": false, + "reason": "Different issues entirely: GGUF architecture support vs Qwen3-MoE sliding_window behavior." + }, + { + "left": "issue:43950", + "right": "issue:45834", + "accept": false, + "reason": "Non-persistent buffer regression and Kosmos2.5 long-OCR index error are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:44910", + "accept": false, + "reason": "Different models and failure modes: torch.compile/SDPA on Bamba vs Flash Attention illegal memory access on Qwen3.5." + }, + { + "left": "issue:43645", + "right": "issue:43646", + "accept": true, + "reason": "Same underlying Transformers 5.0 custom model initialization regression; 43645 is a Jupyter notebook-specific reproduction of the broader 43646 report." + }, + { + "left": "issue:44560", + "right": "issue:45685", + "accept": false, + "reason": "Video StopIteration in Qwen3-vl-embedding is unrelated to an MPS histogram kernel limitation in moe." + }, + { + "left": "issue:43366", + "right": "issue:44589", + "accept": false, + "reason": "GGUF gpt-oss support and Float8 storage lookup failure are unrelated bugs." + }, + { + "left": "issue:42915", + "right": "issue:43957", + "accept": false, + "reason": "Different regressions: Qwen3Moe FineGrainedFP8Config handling vs meta-device model loading on Transformers 5.x." + }, + { + "left": "issue:44560", + "right": "issue:44918", + "accept": false, + "reason": "Both mention Qwen3.5/video-ish inputs, but one is a video StopIteration and the other is an SFT trainer embedding unpacking failure." + }, + { + "left": "issue:43824", + "right": "issue:43957", + "accept": false, + "reason": "Importing a missing class and meta-device loading failures are different problems." + }, + { + "left": "issue:42617", + "right": "issue:43366", + "accept": false, + "reason": "3d_parallel.py execution failure and GGUF gpt-oss support do not share a concrete bug." + }, + { + "left": "issue:36010", + "right": "issue:42915", + "accept": false, + "reason": "GenerationMixin import error and Qwen3Moe FP8 failure are unrelated." + }, + { + "left": "issue:44661", + "right": "issue:45823", + "accept": false, + "reason": "Tokenizer mapping add-new-model-like logic and Gemma4 PLE device mismatch are different issues." + }, + { + "left": "issue:43632", + "right": "issue:45834", + "accept": false, + "reason": "Transformers v5 _is_hf_initialized regression is unrelated to Kosmos2.5 OCR indexing." + }, + { + "left": "issue:38175", + "right": "issue:43994", + "accept": false, + "reason": "Same model family, but one is unexpected zero probabilities and the other is a docs/pipeline misuse report with nonsensical outputs." + }, + { + "left": "issue:44155", + "right": "issue:45584", + "accept": false, + "reason": "AudioFlamingo3 track leakage and Whisper empty-transcription handling are unrelated." + }, + { + "left": "issue:44560", + "right": "issue:44805", + "accept": false, + "reason": "Both are index-ish failures, but the models and code paths differ; no evidence of a shared root cause." + }, + { + "left": "issue:44877", + "right": "issue:45030", + "accept": false, + "reason": "Strict config loading for granite_speech and tiny-random glm4v config validation are different config regressions." + }, + { + "left": "issue:43638", + "right": "issue:44661", + "accept": false, + "reason": "DeepSpeed zero3 Bert initialization and TOKENIZER_MAPPING_NAMES add-new-model-like behavior are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:44661", + "accept": false, + "reason": "Different bugs: zero3 init with BertModel vs tokenizer-mapping-related add-new-model-like failure." + }, + { + "left": "issue:44661", + "right": "issue:44805", + "accept": false, + "reason": "Tokenizer-mapping logic and mask-shape IndexError are unrelated." + }, + { + "left": "issue:43317", + "right": "issue:44756", + "accept": false, + "reason": "device_map auto dequantized load failure and Strix Halo mmap OOM are different issues." + }, + { + "left": "issue:38175", + "right": "issue:39692", + "accept": false, + "reason": "SigLIP2 zero-probabilities bug and SigLIP2 docs example errors are related only by model family, not the same concrete failure." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet and stays conservative. The only accepted soft duplicate, issue:43645/issue:43646, is supported by the titles and the similarity signal. The other pairwise rejections are phrased cautiously 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": 1, + "component_count": 18, + "review_unit_index": 6, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:36010", + "issue:39404", + "issue:39692", + "issue:41720", + "issue:42491", + "issue:42915", + "issue:42947", + "issue:43257", + "issue:43404", + "issue:43425", + "issue:43454", + "issue:43644", + "issue:43645", + "issue:43708", + "issue:43720", + "issue:43854", + "issue:43856", + "issue:43873", + "issue:43931", + "issue:43957", + "issue:43994", + "issue:44075", + "issue:44164", + "issue:44291", + "issue:44387", + "issue:44402", + "issue:44451", + "issue:44488", + "issue:44534", + "issue:44756", + "issue:44910", + "issue:44960", + "issue:44977", + "issue:44991", + "issue:45276", + "issue:45292", + "issue:45313", + "issue:45335", + "issue:45357", + "issue:45362", + "issue:45406", + "issue:45488", + "issue:45588", + "issue:45663", + "issue:45704", + "issue:45759", + "issue:45812", + "issue:45859" + ], + "soft_pairs": [ + "issue:42491|issue:43720", + "issue:44960|issue:44977", + "issue:41720|issue:42915", + "issue:45313|issue:45357", + "issue:44451|issue:44488", + "issue:42491|issue:45276", + "issue:43708|issue:44075", + "issue:43644|issue:44534", + "issue:42915|issue:44756", + "issue:39692|issue:43994", + "issue:45488|issue:45812", + "issue:43854|issue:43931", + "issue:43257|issue:43854", + "issue:43856|issue:43873", + "issue:42491|issue:43931", + "issue:43856|issue:44387", + "issue:43644|issue:43645", + "issue:42491|issue:45292", + "issue:44910|issue:45588", + "issue:44488|issue:44991", + "issue:44164|issue:44291", + "issue:45362|issue:45406", + "issue:45276|issue:45335", + "issue:43404|issue:43454", + "issue:36010|issue:39404", + "issue:39692|issue:44960", + "issue:42947|issue:45704", + "issue:43425|issue:43957", + "issue:44402|issue:44488", + "issue:45663|issue:45759", + "issue:45663|issue:45859", + "issue:43873|issue:44387" + ], + "prepared_review_unit_hash": "aa4b0e9d112cc083e61c6254ed7e5762155bd95ad979e4ae50c92124658384d8", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37349, + "estimated_input_tokens": 9338, + "estimated_eval_tokens": 18932 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly a heterogeneous collection of unrelated Transformers issues, with one clear duplicate pair around the v5 non-persistent buffer regression. Most other soft edges are only loosely similar by subsystem or model family, not the same concrete bug.", + "confidence": 0.82, + "canonical_issue_reason": "issue:43644 is the strongest canonical representative because it has the clearest near-verbatim duplicate in issue:44534 and directly names the underlying Transformers v5 buffer initialization regression.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43644 is the best issue candidate overall because it is the most clearly supported by a duplicate-style twin and describes the bug in the broadest, most reusable way.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "Different bugs: qwen3_moe LoRA incompatibility across HF versions vs BitNet packed-weight handling during accelerate loading." + }, + { + "left": "issue:44960", + "right": "issue:44977", + "accept": false, + "reason": "Different models and failure modes: GLM5 vs Qwen3.5 flash-attention generation behavior." + }, + { + "left": "issue:41720", + "right": "issue:42915", + "accept": false, + "reason": "Different code paths: device-mapping cuda assert on Qwen3 vs FineGrainedFP8Config failure on Qwen3Moe." + }, + { + "left": "issue:45313", + "right": "issue:45357", + "accept": false, + "reason": "Different regressions: DeepSpeed ZeRO-3 loading of language_model weights vs save_pretrained writing wrong visual encoder keys." + }, + { + "left": "issue:44451", + "right": "issue:44488", + "accept": false, + "reason": "Both are load failures for different BERT models, but the titles do not indicate the same underlying cause." + }, + { + "left": "issue:42491", + "right": "issue:45276", + "accept": false, + "reason": "Unrelated issues: qwen3_moe LoRA on hf4.x vs Gemma4 resize_token_embeddings propagation." + }, + { + "left": "issue:43708", + "right": "issue:44075", + "accept": false, + "reason": "Trainer checkpoint math and optimizer argument usage are separate problems." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "Near-identical regression about non-persistent buffers being filled with junk in Transformers v5." + }, + { + "left": "issue:42915", + "right": "issue:44756", + "accept": false, + "reason": "Different failures: Qwen3Moe FP8 config vs mmap-related OOM on Strix Halo." + }, + { + "left": "issue:39692", + "right": "issue:43994", + "accept": false, + "reason": "Both involve SigLIP2, but one is a docs/example mismatch and quantization failure while the other is bad inference output." + }, + { + "left": "issue:45488", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and different tokenization behaviors." + }, + { + "left": "issue:43854", + "right": "issue:43931", + "accept": false, + "reason": "Different models and distinct loading errors: GLM-4.7-Flash unit-test loading vs Qwen3-VL shape mismatch." + }, + { + "left": "issue:43257", + "right": "issue:43854", + "accept": false, + "reason": "Different bugs: Qwen3 MoE weight conversion with accelerate/deepspeed vs GLM-4.7-Flash loading in tests." + }, + { + "left": "issue:43856", + "right": "issue:43873", + "accept": false, + "reason": "Both are memory-related, but one is Qwen3 MoE training inefficiency and the other is quantization offloading behavior." + }, + { + "left": "issue:42491", + "right": "issue:43931", + "accept": false, + "reason": "Different model families and distinct failures: qwen3_moe LoRA portability vs Qwen3-VL weight shape mismatch." + }, + { + "left": "issue:43856", + "right": "issue:44387", + "accept": false, + "reason": "Both mention memory, but the causes differ: MoE training inefficiency vs int4 quantization reserved-memory/OOM regression." + }, + { + "left": "issue:43644", + "right": "issue:43645", + "accept": false, + "reason": "Related Transformer 5 regressions, but one is buffer initialization and the other is custom-model initialization in Jupyter." + }, + { + "left": "issue:42491", + "right": "issue:45292", + "accept": false, + "reason": "Different issues: qwen3_moe LoRA compatibility vs resize_token_embeddings not updating output embeddings." + }, + { + "left": "issue:44910", + "right": "issue:45588", + "accept": false, + "reason": "Both are flash-attention crashes, but the concrete failure modes differ: illegal memory access from packed-sequence handling vs AttributeError on s_aux=None." + }, + { + "left": "issue:44488", + "right": "issue:44991", + "accept": false, + "reason": "Different tokenizer/loading bugs affecting different models." + }, + { + "left": "issue:44164", + "right": "issue:44291", + "accept": false, + "reason": "Different mechanics: extra_state save/load handling vs init_empty_weights argument compatibility." + }, + { + "left": "issue:45362", + "right": "issue:45406", + "accept": false, + "reason": "Separate crashes in different serving/chat paths and different model classes." + }, + { + "left": "issue:45276", + "right": "issue:45335", + "accept": false, + "reason": "Both concern resize_token_embeddings, but the affected model structures differ and the titles do not show the same concrete bug." + }, + { + "left": "issue:43404", + "right": "issue:43454", + "accept": false, + "reason": "Both are lm_head weight-tying issues, but they affect different architectures and are not clearly the same code-path bug." + }, + { + "left": "issue:36010", + "right": "issue:39404", + "accept": false, + "reason": "Completely different problems: import failure for GenerationMixin vs Whisper pipeline return_language regression." + }, + { + "left": "issue:39692", + "right": "issue:44960", + "accept": false, + "reason": "Different issues and model families; only superficial similarity in being model-related." + }, + { + "left": "issue:42947", + "right": "issue:45704", + "accept": false, + "reason": "Unrelated regressions: LoRA gradient checkpointing behavior vs apex RMSNorm memory leak." + }, + { + "left": "issue:43425", + "right": "issue:43957", + "accept": false, + "reason": "Torch version compatibility issue vs meta-device loading regression; different root causes." + }, + { + "left": "issue:44402", + "right": "issue:44488", + "accept": false, + "reason": "Token-vocab mismatch in one model vs inability to load a different model." + }, + { + "left": "issue:45663", + "right": "issue:45759", + "accept": false, + "reason": "Different bugs in different model families: FSDP2 shared_kv_states rebuild vs composite-config unwrapping." + }, + { + "left": "issue:45663", + "right": "issue:45859", + "accept": false, + "reason": "Different issues: Gemma-4 FSDP2 KeyError vs Qwen3_5MoeTextRotaryEmbedding CPU-offload incompatibility." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The only accepted soft edge is well supported by the near-identical 43644/44534 titles, and the rejected pairs are generally justified by distinct model families or failure modes." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 7, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:39404", + "issue:41720", + "issue:42947", + "issue:43257", + "issue:43335", + "issue:43381", + "issue:43425", + "issue:43611", + "issue:43645", + "issue:43646", + "issue:43708", + "issue:43716", + "issue:43828", + "issue:43854", + "issue:43906", + "issue:43957", + "issue:44155", + "issue:44164", + "issue:44291", + "issue:44292", + "issue:44387", + "issue:44402", + "issue:44451", + "issue:44479", + "issue:44530", + "issue:44589", + "issue:44683", + "issue:44756", + "issue:44863", + "issue:44877", + "issue:44912", + "issue:44928", + "issue:44960", + "issue:45084", + "issue:45216", + "issue:45276", + "issue:45292", + "issue:45335", + "issue:45588", + "issue:45663", + "issue:45684", + "issue:45698", + "issue:45701", + "issue:45812", + "issue:45823", + "issue:46050", + "issue:46315" + ], + "soft_pairs": [ + "issue:44155|issue:45823", + "issue:44291|issue:44756", + "issue:44292|issue:44912", + "issue:44960|issue:46050", + "issue:43335|issue:45663", + "issue:44387|issue:44756", + "issue:42947|issue:43381", + "issue:45276|issue:45292", + "issue:43854|issue:45216", + "issue:43381|issue:44928", + "issue:44164|issue:44756", + "issue:45684|issue:45698", + "issue:41720|issue:44155", + "issue:44479|issue:44877", + "issue:45588|issue:45663", + "issue:43957|issue:44756", + "issue:43646|issue:43708", + "issue:45812|issue:46315", + "issue:45292|issue:45335", + "issue:43645|issue:44164", + "issue:44402|issue:44451", + "issue:43828|issue:43957", + "issue:43425|issue:43828", + "issue:44589|issue:44756", + "issue:43257|issue:44863", + "issue:44683|issue:45823", + "issue:39404|issue:45701", + "issue:43611|issue:43645", + "issue:43425|issue:45823", + "issue:44530|issue:45084", + "issue:43425|issue:43716", + "issue:43906|issue:44530" + ], + "prepared_review_unit_hash": "27f3d8c709d93eafe18e6775a5c2630326634fe35af3910febe78e03ca8e08c8", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36690, + "estimated_input_tokens": 9173, + "estimated_eval_tokens": 18602 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is loosely related: it mixes several distinct Transformers v5 regressions (loading/saving, tokenization, quantization, checkpointing, and a small resize_token_embeddings subfamily). Most soft links are false positives; only the resize-token-embeddings reports look like the same underlying bug family.", + "confidence": 0.34, + "canonical_issue_reason": "Issue 45701 is the broadest and most central open regression report here, with multiple inbound references and the widest umbrella over v5 behavior changes, even though the cluster is not a single exact duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45701 is the best overall representative because it is the most general, active, and referenced issue among these mixed regression reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44155", + "right": "issue:45823", + "accept": false, + "reason": "Different bugs: batched AudioFlamingo3 leakage between tracks vs Gemma4 PLE device mismatch in forward." + }, + { + "left": "issue:44291", + "right": "issue:44756", + "accept": false, + "reason": "Unexpected _is_hf_initialized arg during init_empty_weights is unrelated to Strix Halo mmap-related OOM." + }, + { + "left": "issue:44292", + "right": "issue:44912", + "accept": false, + "reason": "Both are quantization/loading regressions, but the concrete failures and model paths differ (NVFP4 runtime error vs MXFP4 fallback to bf16)." + }, + { + "left": "issue:44960", + "right": "issue:46050", + "accept": false, + "reason": "GLM5 loading failure and sinq save/load corruption are different underlying bugs." + }, + { + "left": "issue:43335", + "right": "issue:45663", + "accept": false, + "reason": "Sparse-layer creation in SwitchTransformers is unrelated to Gemma4 FSDP2 shared_kv_states rebuilding." + }, + { + "left": "issue:44387", + "right": "issue:44756", + "accept": false, + "reason": "Int4 CUDA memory regression and Strix Halo mmap OOM are both memory issues, but not the same code-path bug." + }, + { + "left": "issue:42947", + "right": "issue:43381", + "accept": false, + "reason": "Gradient checkpointing ineffective with PEFT LoRA is distinct from the eval-mode restriction on checkpointing." + }, + { + "left": "issue:45276", + "right": "issue:45292", + "accept": true, + "reason": "Same resize_token_embeddings bug family: both say resized embeddings are not propagated, with 45276 adding a Gemma4-specific per-layer/output-embedding manifestation." + }, + { + "left": "issue:43854", + "right": "issue:45216", + "accept": false, + "reason": "Model-loading failure in unit tests and Qwen3.5 save_pretrained regression are different issues." + }, + { + "left": "issue:43381", + "right": "issue:44928", + "accept": false, + "reason": "Eval-mode checkpointing prohibition is unrelated to the RLHF NaN/SDPA fallback gradient explosion." + }, + { + "left": "issue:44164", + "right": "issue:44756", + "accept": false, + "reason": "save/from_pretrained extra_state handling is a serialization bug, not an mmap/OOM memory issue." + }, + { + "left": "issue:45684", + "right": "issue:45698", + "accept": false, + "reason": "Read-only permission propagation during save_pretrained and wrong custom module loading after save_pretrained are different failure modes." + }, + { + "left": "issue:41720", + "right": "issue:44155", + "accept": false, + "reason": "CUDA assert on Qwen3 auto device mapping is unrelated to AudioFlamingo3 token/embedding leakage between tracks." + }, + { + "left": "issue:44479", + "right": "issue:44877", + "accept": false, + "reason": "Video-input regression across Qwen models and strict granite_speech config loading are not the same bug." + }, + { + "left": "issue:45588", + "right": "issue:45663", + "accept": false, + "reason": "flash_attention sink-less crash and Gemma4 FSDP2 KeyError are different code paths." + }, + { + "left": "issue:43957", + "right": "issue:44756", + "accept": false, + "reason": "Meta-device loading breakage across some models is unrelated to Strix Halo mmap OOM." + }, + { + "left": "issue:43646", + "right": "issue:43708", + "accept": false, + "reason": "Custom model initialization in v5 and resume_from_checkpoint max_steps calculation are separate regressions." + }, + { + "left": "issue:45812", + "right": "issue:46315", + "accept": false, + "reason": "Both involve tokenization changes, but one is wrong token IDs for Granite models and the other is a slow convert_tokens_to_ids regression; not the same bug." + }, + { + "left": "issue:45292", + "right": "issue:45335", + "accept": true, + "reason": "Same underlying resize_token_embeddings propagation bug: 45335 is the t5gemma-specific case of output_embeddings not being updated." + }, + { + "left": "issue:43645", + "right": "issue:44164", + "accept": false, + "reason": "Notebook custom-model initialization breakage is unrelated to extra_state save/from_pretrained handling." + }, + { + "left": "issue:44402", + "right": "issue:44451", + "accept": false, + "reason": "Tokenizer vocab-size mismatch on one model and inability to load ScandiBERT are different load-time issues." + }, + { + "left": "issue:43828", + "right": "issue:43957", + "accept": false, + "reason": "Autocast dtype mismatch on Phi-tiny-MoE and meta-device loading failures are distinct model/runtime bugs." + }, + { + "left": "issue:43425", + "right": "issue:43828", + "accept": false, + "reason": "Torch 2.10 incompatibility is a version-compatibility issue, not the same as the Phi autocast dtype mismatch." + }, + { + "left": "issue:44589", + "right": "issue:44756", + "accept": false, + "reason": "Float8 storage lookup failure and Strix Halo OOM are different runtime problems." + }, + { + "left": "issue:43257", + "right": "issue:44863", + "accept": false, + "reason": "Qwen3 MOE weights not converted under accelerate+deepspeed is unrelated to NemotronH checkpoint loading." + }, + { + "left": "issue:44683", + "right": "issue:45823", + "accept": false, + "reason": "Compiled flex_attention failing on torch>=2.9 is unrelated to Gemma4 device mismatch during forward." + }, + { + "left": "issue:39404", + "right": "issue:45701", + "accept": false, + "reason": "Whisper pipeline return_language regression and generic tokenizer-version regression are not the same bug." + }, + { + "left": "issue:43611", + "right": "issue:43645", + "accept": false, + "reason": "base_model_prefix loading regression and notebook custom model initialization regression are distinct issues." + }, + { + "left": "issue:43425", + "right": "issue:45823", + "accept": false, + "reason": "Torch 2.10 incompatibility is a broad version issue, not the Gemma4 PLE device-mismatch bug." + }, + { + "left": "issue:44530", + "right": "issue:45084", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash and non-template-node compile error are different failures." + }, + { + "left": "issue:43425", + "right": "issue:43716", + "accept": false, + "reason": "Torch 2.10 incompatibility and Mistral-3 image-preprocessor/model dtype mismatch are unrelated." + }, + { + "left": "issue:43906", + "right": "issue:44530", + "accept": false, + "reason": "An isolated reproduction of another issue is not the same bug as the Qwen3.5 PagedAttentionCache crash." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary stays grounded in the packet: 45701 is a reasonable broad representative given its open state, multiple inbound references, and generic tokenization-regression framing. The soft-edge calls are conservative overall, and the only accepted cluster is the resize_token_embeddings subfamily, which is supported by the issue titles. No PR-mergeability concerns apply here." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 8, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:32090", + "issue:34567", + "issue:36032", + "issue:39404", + "issue:40444", + "issue:40990", + "issue:42915", + "issue:43299", + "issue:43421", + "issue:43425", + "issue:43452", + "issue:43541", + "issue:43582", + "issue:43596", + "issue:43611", + "issue:43632", + "issue:43643", + "issue:43644", + "issue:43716", + "issue:43856", + "issue:43901", + "issue:43931", + "issue:43994", + "issue:44079", + "issue:44164", + "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:44811", + "issue:44898", + "issue:44912", + "issue:44918", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45584", + "issue:45759", + "issue:45859" + ], + "soft_pairs": [ + "issue:44265|issue:44898", + "issue:39404|issue:44912", + "issue:44479|issue:44805", + "issue:43901|issue:44509", + "issue:44589|issue:45084", + "issue:43425|issue:44918", + "issue:44164|issue:45759", + "issue:36032|issue:43452", + "issue:43299|issue:43931", + "issue:43596|issue:44683", + "issue:43994|issue:44079", + "issue:43856|issue:44756", + "issue:44671|issue:44991", + "issue:44991|issue:45081", + "issue:39404|issue:44671", + "issue:44737|issue:45859", + "issue:44164|issue:44479", + "issue:44488|issue:45081", + "issue:43582|issue:43632", + "issue:43611|issue:43644", + "issue:44661|issue:44683", + "issue:40990|issue:43421", + "issue:32090|issue:39404", + "issue:43716|issue:45084", + "issue:44488|issue:44987", + "issue:43643|issue:44530", + "issue:34567|issue:40444", + "issue:36032|issue:42915", + "issue:44811|issue:45584", + "issue:42915|issue:44530", + "issue:43541|issue:45084" + ], + "prepared_review_unit_hash": "f5318ceea4b8a1c74b81cfefd260fed8508972aaf2364a1dbdb9260f5da68cc2", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37062, + "estimated_input_tokens": 9266, + "estimated_eval_tokens": 18788 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is a loose collection of unrelated Transformers regressions, mostly around v5 loading/tokenizer/model behavior. None of the soft pairs look like the same concrete bug or change, so all soft edges are rejected.", + "confidence": 0.37, + "canonical_issue_reason": "Issue 43611 is the broadest and most representative v5 regression in this set: a generic model-loading breakage. The other issues are narrower, model-specific, tokenizer-specific, or unrelated performance/runtime failures.", + "canonical_pr_reason": null, + "best_issue_reason": "43611 is the strongest global issue candidate because it is phrased broadly and matches the dominant theme of v5 compatibility breakages without being overly narrow to one model family.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44265", + "right": "issue:44898", + "accept": false, + "reason": "Both are model/runtime failures, but they involve different APIs and likely different code paths; not the same underlying bug." + }, + { + "left": "issue:39404", + "right": "issue:44912", + "accept": false, + "reason": "Whisper pipeline language return vs MXFP4 quantization fallback are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:44805", + "accept": false, + "reason": "Video-input regression and mask-shape IndexError do not describe the same failure mode." + }, + { + "left": "issue:43901", + "right": "issue:44509", + "accept": false, + "reason": "Both are docs-related, but they concern different removed/incorrectly documented pipeline behavior; not one bug." + }, + { + "left": "issue:44589", + "right": "issue:45084", + "accept": false, + "reason": "Different TypeErrors in different subsystems; no shared concrete code path." + }, + { + "left": "issue:43425", + "right": "issue:44918", + "accept": false, + "reason": "Torch version incompatibility is a broad environment issue, while the other is a specific SFT trainer embedding unpacking failure." + }, + { + "left": "issue:44164", + "right": "issue:45759", + "accept": false, + "reason": "Serialization extra_state handling and from_config text_config unwrapping are distinct loading/saving bugs." + }, + { + "left": "issue:36032", + "right": "issue:43452", + "accept": false, + "reason": "Tokenizer special-token method conflict and gguf_file from_pretrained breakage are different tokenizer/model-loading problems." + }, + { + "left": "issue:43299", + "right": "issue:43931", + "accept": false, + "reason": "Both involve Qwen3-VL/Qwen3VL loading, but one is a version-specific loading breakage and the other is a weight-shape mismatch on a specific model; not clearly the same bug." + }, + { + "left": "issue:43596", + "right": "issue:44683", + "accept": false, + "reason": "DeepSpeed ZeRO-3 init index error and compiled flex_attention torch version failure are unrelated." + }, + { + "left": "issue:43994", + "right": "issue:44079", + "accept": false, + "reason": "A nonsensical model/pipeline output report is not the same as a ModelOutput key-assignment bug." + }, + { + "left": "issue:43856", + "right": "issue:44756", + "accept": false, + "reason": "MoE training memory usage and mmap OOM on Strix Halo are different performance/memory issues." + }, + { + "left": "issue:44671", + "right": "issue:44991", + "accept": false, + "reason": "Different model families and different symptoms: incorrect masked LM predictions vs tokenizer loading failure." + }, + { + "left": "issue:44991", + "right": "issue:45081", + "accept": false, + "reason": "Both are tokenizer-load regressions, but the affected models and failure mechanisms differ." + }, + { + "left": "issue:39404", + "right": "issue:44671", + "accept": false, + "reason": "Whisper pipeline behavior and CamemBERT masked-LM predictions are unrelated." + }, + { + "left": "issue:44737", + "right": "issue:45859", + "accept": false, + "reason": "CPU device placement bug in XLNet is not the same as Qwen3.5 MoE rotary embedding CPU offload incompatibility." + }, + { + "left": "issue:44164", + "right": "issue:44479", + "accept": false, + "reason": "save/from_pretrained extra_state handling is unrelated to the Qwen video-input regression." + }, + { + "left": "issue:44488", + "right": "issue:45081", + "accept": false, + "reason": "Two different tokenizer loading issues for different models; no shared concrete fix." + }, + { + "left": "issue:43582", + "right": "issue:43632", + "accept": false, + "reason": "Apple Silicon warmup TypeError and `_is_hf_initialized` flag breakage are different regression classes." + }, + { + "left": "issue:43611", + "right": "issue:43644", + "accept": false, + "reason": "Both are v5 regressions, but one is model loading via base_model_prefix and the other is non-persistent buffer initialization; not the same bug." + }, + { + "left": "issue:44661", + "right": "issue:44683", + "accept": false, + "reason": "Tokenizer mapping-name handling and flex_attention compilation are unrelated." + }, + { + "left": "issue:40990", + "right": "issue:43421", + "accept": false, + "reason": "Perplexity on a specific model and runtime special-token post-processing are not the same issue." + }, + { + "left": "issue:32090", + "right": "issue:39404", + "accept": false, + "reason": "Trainer GPU broadcast NoneType error is unrelated to Whisper return_language behavior." + }, + { + "left": "issue:43716", + "right": "issue:45084", + "accept": false, + "reason": "Image preprocessor/model dtype mismatch and Jinja/template compilation error are unrelated." + }, + { + "left": "issue:44488", + "right": "issue:44987", + "accept": false, + "reason": "Both are model loading failures, but the affected models and symptoms differ." + }, + { + "left": "issue:43643", + "right": "issue:44530", + "accept": false, + "reason": "AutoConfig trust_remote_code missing fields and PagedAttentionCache linear_attention crash are unrelated." + }, + { + "left": "issue:34567", + "right": "issue:40444", + "accept": false, + "reason": "TrainerState token counting and Qwen2.5-VL iterable dataset multi-image finetuning are different bugs." + }, + { + "left": "issue:36032", + "right": "issue:42915", + "accept": false, + "reason": "T5 tokenizer class-method conflict and Qwen3 MoE FineGrainedFP8Config failure are unrelated." + }, + { + "left": "issue:44811", + "right": "issue:45584", + "accept": false, + "reason": "Whisper skip_special_tokens handling and empty-transcription generation after align_special_tokens are related by theme but not clearly the same concrete bug." + }, + { + "left": "issue:42915", + "right": "issue:44530", + "accept": false, + "reason": "Different Qwen-related failures with different triggers: FP8 config vs cache group type error." + }, + { + "left": "issue:43541", + "right": "issue:45084", + "accept": false, + "reason": "Torch dynamo tracing grouped_mm error and template-node compilation failure are unrelated code paths." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 9, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:36032", + "issue:42175", + "issue:42915", + "issue:43089", + "issue:43262", + "issue:43317", + "issue:43322", + "issue:43377", + "issue:43404", + "issue:43452", + "issue:43482", + "issue:43540", + "issue:43632", + "issue:43644", + "issue:43645", + "issue:43646", + "issue:43742", + "issue:43819", + "issue:43883", + "issue:43994", + "issue:44062", + "issue:44206", + "issue:44220", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44451", + "issue:44464", + "issue:44466", + "issue:44488", + "issue:44534", + "issue:44589", + "issue:44610", + "issue:44683", + "issue:44756", + "issue:44849", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45127", + "issue:45216", + "issue:45325", + "issue:45357", + "issue:45507", + "issue:45987", + "issue:46178", + "issue:46315" + ], + "soft_pairs": [ + "issue:44451|issue:44987", + "issue:43644|issue:43646", + "issue:44206|issue:44220", + "issue:44464|issue:45987", + "issue:43452|issue:43482", + "issue:43317|issue:43482", + "issue:43883|issue:44488", + "issue:44849|issue:45325", + "issue:36032|issue:43645", + "issue:45084|issue:45507", + "issue:42175|issue:46178", + "issue:43262|issue:43377", + "issue:44987|issue:44991", + "issue:43646|issue:44534", + "issue:43089|issue:45987", + "issue:43404|issue:44466", + "issue:43322|issue:43540", + "issue:43994|issue:44610", + "issue:44403|issue:45081", + "issue:43883|issue:44991", + "issue:42915|issue:45084", + "issue:43377|issue:43819", + "issue:43645|issue:44534", + "issue:44265|issue:44610", + "issue:44991|issue:46315", + "issue:44451|issue:45081", + "issue:44987|issue:45081", + "issue:45216|issue:45357", + "issue:44220|issue:44683", + "issue:36032|issue:43742", + "issue:42915|issue:44291", + "issue:43632|issue:44756", + "issue:42915|issue:44589", + "issue:43404|issue:45127", + "issue:44062|issue:44987" + ], + "prepared_review_unit_hash": "9beae67f2838759d00ffa982ca734055f34c3982f694799ef94c9d05f40661cf", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38263, + "estimated_input_tokens": 9566, + "estimated_eval_tokens": 19388 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous v5/model-loading regressions that are not true duplicates. The only clear duplicate-looking pair is the Qwen3.5 save_pretrained regression (45216/45357).", + "confidence": 0.86, + "canonical_issue_reason": "Pick issue 45216 as the canonical issue: it is the clearest, broadest description of the underlying Qwen3.5 save_pretrained checkpoint regression, and 45357 reads like a later/finer-grained follow-up on the same bug.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45216 is the best representative issue in the cluster because it states the regression window and the concrete failure mode most generally, making it the most useful triage target for the duplicate pocket.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44451", + "right": "issue:44987", + "accept": false, + "reason": "Both are load failures on v5, but they name different models and there is no clear shared concrete bug path." + }, + { + "left": "issue:43644", + "right": "issue:43646", + "accept": false, + "reason": "Related to custom-model initialization, but one is about non-persistent buffers and the other about notebook/custom init behavior; too different to merge." + }, + { + "left": "issue:44206", + "right": "issue:44220", + "accept": false, + "reason": "Both are audio-feature bugs, but the titles point to different failing entry points and symptoms." + }, + { + "left": "issue:44464", + "right": "issue:45987", + "accept": false, + "reason": "Both touch generation/cache behavior, but one is about chunked generation with compiled forward and the other is a StaticCache length contract issue." + }, + { + "left": "issue:43452", + "right": "issue:43482", + "accept": false, + "reason": "Both involve GGUF/loading regressions, but one is a broad from_pretrained API break and the other is a model-specific Qwen2.5-GGUF failure." + }, + { + "left": "issue:43317", + "right": "issue:43482", + "accept": false, + "reason": "Different failure modes: dequantized model device_map offload versus GGUF model loading." + }, + { + "left": "issue:43883", + "right": "issue:44488", + "accept": false, + "reason": "Different models and different missing attributes/loading symptoms." + }, + { + "left": "issue:44849", + "right": "issue:45325", + "accept": false, + "reason": "Both are Qwen-family regressions, but they affect different model variants and different code paths." + }, + { + "left": "issue:36032", + "right": "issue:43645", + "accept": false, + "reason": "Tokenizer init conflict versus custom model initialization breakage; not the same bug." + }, + { + "left": "issue:45084", + "right": "issue:45507", + "accept": false, + "reason": "Different subsystems and different error types; no shared concrete fix path." + }, + { + "left": "issue:42175", + "right": "issue:46178", + "accept": false, + "reason": "Packaging/backend mapping issue versus backend lookup for tensorflow_text; not the same defect." + }, + { + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Audio chat-template sampling-rate bug versus MIMI batching/padding-mask discrepancy." + }, + { + "left": "issue:44987", + "right": "issue:44991", + "accept": false, + "reason": "Both are v5 load regressions, but they target different models and likely different underlying tokenizer/model problems." + }, + { + "left": "issue:43646", + "right": "issue:44534", + "accept": false, + "reason": "Both mention custom-model behavior in v5, but one is initialization and the other is buffer serialization/junk; not the same bug." + }, + { + "left": "issue:43089", + "right": "issue:45987", + "accept": false, + "reason": "Generation performance overhead is unrelated to the StaticCache return-type issue." + }, + { + "left": "issue:43404", + "right": "issue:44466", + "accept": false, + "reason": "Weight tying in a multimodal model versus inconsistent lm_head serialization; related theme, different concrete bugs." + }, + { + "left": "issue:43322", + "right": "issue:43540", + "accept": false, + "reason": "Llava Next loading crash and Qwen3OmniMoe video-processing ValueError are unrelated." + }, + { + "left": "issue:43994", + "right": "issue:44610", + "accept": false, + "reason": "Different models and different incorrect-output mechanisms." + }, + { + "left": "issue:44403", + "right": "issue:45081", + "accept": false, + "reason": "Noise/logging on load versus a tokenizer regex patch crash; not the same code-path problem." + }, + { + "left": "issue:43883", + "right": "issue:44991", + "accept": false, + "reason": "Different model families and different loading failures." + }, + { + "left": "issue:42915", + "right": "issue:45084", + "accept": false, + "reason": "FineGrainedFP8Config loading failure versus template-node compilation failure; unrelated." + }, + { + "left": "issue:43377", + "right": "issue:43819", + "accept": false, + "reason": "Both are model path discrepancies, but they concern different models and different missing-feature causes." + }, + { + "left": "issue:43645", + "right": "issue:44534", + "accept": false, + "reason": "Custom-model notebook init breakage and non-persistent buffer junk are different regression types." + }, + { + "left": "issue:44265", + "right": "issue:44610", + "accept": false, + "reason": "torch.export/exportability issue versus processor input-size mismatch; not the same bug." + }, + { + "left": "issue:44991", + "right": "issue:46315", + "accept": false, + "reason": "Tokenizer loading regression versus tokenizer speed regression; different defects." + }, + { + "left": "issue:44451", + "right": "issue:45081", + "accept": false, + "reason": "Model load failure versus Mistral regex patch crash; distinct code paths." + }, + { + "left": "issue:44987", + "right": "issue:45081", + "accept": false, + "reason": "Different affected models and different failure sites." + }, + { + "left": "issue:45216", + "right": "issue:45357", + "accept": true, + "reason": "Same underlying Qwen3.5 save_pretrained checkpoint serialization regression; 45357 is a later manifestation about incorrect visual encoder keys." + }, + { + "left": "issue:44220", + "right": "issue:44683", + "accept": false, + "reason": "Audio fbank extraction issue versus compiled flex_attention failure on newer torch; unrelated." + }, + { + "left": "issue:36032", + "right": "issue:43742", + "accept": false, + "reason": "Tokenizer special-token conflict versus model loading key error; different bugs." + }, + { + "left": "issue:42915", + "right": "issue:44291", + "accept": false, + "reason": "Both involve loading/init regressions, but they are different failure modes and not a single concrete defect." + }, + { + "left": "issue:43632", + "right": "issue:44756", + "accept": false, + "reason": "Different issues: _is_hf_initialized flag breakage versus mmap/OOM on Strix Halo." + }, + { + "left": "issue:42915", + "right": "issue:44589", + "accept": false, + "reason": "FineGrainedFP8Config failure versus float8 storage deserialization error; unrelated." + }, + { + "left": "issue:43404", + "right": "issue:45127", + "accept": false, + "reason": "Different tied-weights/model-collapse regressions with different triggers and surfaces." + }, + { + "left": "issue:44062", + "right": "issue:44987", + "accept": false, + "reason": "AddedToken special-argument bug versus a model loading regression; not the same issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is grounded in the packet and stays conservative overall. The Qwen3.5 save_pretrained pair (45216/45357) is the only clearly supported duplicate-looking match from the listed soft pairs, and the other rejections are reasonable given only title-level evidence. One mild caveat: the summary is broad and does not attempt exhaustive recall, but it 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": 1, + "component_count": 18, + "review_unit_index": 10, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:33666", + "issue:36032", + "issue:40444", + "issue:40990", + "issue:41553", + "issue:41720", + "issue:42175", + "issue:43296", + "issue:43317", + "issue:43377", + "issue:43381", + "issue:43421", + "issue:43425", + "issue:43482", + "issue:43525", + "issue:43531", + "issue:43582", + "issue:43611", + "issue:43632", + "issue:43792", + "issue:43818", + "issue:43856", + "issue:44079", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44462", + "issue:44479", + "issue:44534", + "issue:44568", + "issue:44589", + "issue:44683", + "issue:44811", + "issue:44829", + "issue:44877", + "issue:44991", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45325", + "issue:45357", + "issue:45479", + "issue:45491", + "issue:45538", + "issue:45542", + "issue:45663", + "issue:45812", + "issue:45923" + ], + "soft_pairs": [ + "issue:43611|issue:44534", + "issue:43531|issue:44877", + "issue:43296|issue:43531", + "issue:36032|issue:41720", + "issue:44479|issue:45325", + "issue:43425|issue:44403", + "issue:40990|issue:44811", + "issue:33666|issue:40444", + "issue:43792|issue:44589", + "issue:43421|issue:44568", + "issue:44403|issue:44991", + "issue:43632|issue:44534", + "issue:44079|issue:45663", + "issue:43381|issue:45923", + "issue:44683|issue:45020", + "issue:44462|issue:45812", + "issue:44829|issue:45479", + "issue:40444|issue:45325", + "issue:43482|issue:45538", + "issue:43317|issue:43856", + "issue:44291|issue:44589", + "issue:43482|issue:43525", + "issue:43818|issue:45357", + "issue:42175|issue:45542", + "issue:41553|issue:43582", + "issue:43377|issue:45491", + "issue:41720|issue:45081", + "issue:43531|issue:44291", + "issue:44265|issue:45072" + ], + "prepared_review_unit_hash": "02a6615e43e4513f11069898ee06a0b741f13c8d480abc4ebb87b90615d8f2aa", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36598, + "estimated_input_tokens": 9150, + "estimated_eval_tokens": 18556 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly heterogeneous rather than one tight duplicate cluster. Only two soft edges look plausibly duplicate-level: one around Qwen-VL temporal/rope indexing, and one around AutoTokenizer/tokenizer.json handling.", + "confidence": 0.61, + "canonical_issue_reason": "No single canonical issue: the items split into several unrelated regressions. If one representative must be chosen, issue:45812 is the strongest because it is broad, concrete, and has inbound references.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45812 is the best overall representative among the issues because it describes a clear, user-visible regression with broad impact and linked discussion.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43611", + "right": "issue:44534", + "accept": false, + "reason": "Both mention Transformers v5, but the failures are different: model-loading/base_model_prefix vs non-persistent buffer initialization." + }, + { + "left": "issue:43531", + "right": "issue:44877", + "accept": false, + "reason": "Different bugs in different models: Qwen3-MoE sliding-window behavior vs Granite speech config loading." + }, + { + "left": "issue:43296", + "right": "issue:43531", + "accept": false, + "reason": "Both are Qwen-family issues, but one is PaddleOCR-VL loading in vLLM and the other is Qwen3-MoE sliding-window behavior; not the same bug." + }, + { + "left": "issue:36032", + "right": "issue:41720", + "accept": false, + "reason": "Tokenizer method conflict vs device-mapping cuda assert are unrelated failure modes." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": true, + "reason": "Both describe the same Qwen-VL temporal/rope-indexing regression in v5.3.0, affecting visual/video position IDs in closely related code paths." + }, + { + "left": "issue:43425", + "right": "issue:44403", + "accept": false, + "reason": "Torch 2.10 compatibility and noisy loading output are unrelated issues." + }, + { + "left": "issue:40990", + "right": "issue:44811", + "accept": false, + "reason": "Perplexity quality regression is unrelated to Whisper batch_decode skip_special_tokens handling." + }, + { + "left": "issue:33666", + "right": "issue:40444", + "accept": false, + "reason": "Both are Qwen2-VL training issues, but the concrete failures differ: multi-GPU training vs IterableDataset with multiple images per prompt." + }, + { + "left": "issue:43792", + "right": "issue:44589", + "accept": false, + "reason": "Whisper runtime failure and missing Float8 storage object are different code-path failures." + }, + { + "left": "issue:43421", + "right": "issue:44568", + "accept": false, + "reason": "Runtime post-processor updates for modified special tokens is not the same as add_special_tokens failing to add BOS/EOS." + }, + { + "left": "issue:44403", + "right": "issue:44991", + "accept": false, + "reason": "Generic loading noise and a tokenizer-loading regression for est-roberta are not the same underlying bug." + }, + { + "left": "issue:43632", + "right": "issue:44534", + "accept": false, + "reason": "_is_hf_initialized flag breakage and non-persistent buffer junk are separate Transformers v5 regressions." + }, + { + "left": "issue:44079", + "right": "issue:45663", + "accept": false, + "reason": "ModelOutput key assignment and Gemma-4 FSDP2 shared_kv_states rebuilding are unrelated." + }, + { + "left": "issue:43381", + "right": "issue:45923", + "accept": false, + "reason": "Eval-mode gradient checkpointing behavior is distinct from a missing supports_gradient_checkpointing flag on a remote-code model." + }, + { + "left": "issue:44683", + "right": "issue:45020", + "accept": false, + "reason": "Compiled flex_attention on torch>=2.9 and remote_code model breakage are different problems." + }, + { + "left": "issue:44462", + "right": "issue:45812", + "accept": true, + "reason": "Both point to the same AutoTokenizer/tokenizer.json selection regression, with 45812 being a model-specific manifestation of the broader 44462 bug." + }, + { + "left": "issue:44829", + "right": "issue:45479", + "accept": false, + "reason": "Degenerate training with flash_attention_3 and zero-loss single-label classification are different classification-loss bugs." + }, + { + "left": "issue:40444", + "right": "issue:45325", + "accept": false, + "reason": "Qwen2.5-VL iterable-dataset multi-image failure is not the same as the still-image temporal position-id scaling bug." + }, + { + "left": "issue:43482", + "right": "issue:45538", + "accept": false, + "reason": "GGUF loading failure and CLIPTokenizer max_length behavior are unrelated." + }, + { + "left": "issue:43317", + "right": "issue:43856", + "accept": false, + "reason": "GPU+CPU offload loading failure and Qwen3 MoE training memory usage are different issues." + }, + { + "left": "issue:44291", + "right": "issue:44589", + "accept": false, + "reason": "Unexpected _is_hf_initialized argument and missing Float8 storage object are unrelated Transformers v5 breakages." + }, + { + "left": "issue:43482", + "right": "issue:43525", + "accept": false, + "reason": "Qwen2.5-GGUF loading and Llama4Config missing pad_token_id are different model-specific regressions." + }, + { + "left": "issue:43818", + "right": "issue:45357", + "accept": false, + "reason": "Video-LLaVA video_tower/temporal-attention bug and Qwen3.5 save_pretrained key regression do not share the same code path." + }, + { + "left": "issue:42175", + "right": "issue:45542", + "accept": false, + "reason": "Both concern TF packaging, but one is missing TensorFlow in the torch extra while the other is a tensorboard-only install causing an undefined backend error." + }, + { + "left": "issue:41553", + "right": "issue:43582", + "accept": false, + "reason": "AutoTokenizer error messaging for Voxtral and Apple Silicon caching_allocator_warmup TypeError are unrelated." + }, + { + "left": "issue:43377", + "right": "issue:45491", + "accept": false, + "reason": "Missing padding-mask support in MIMI encoder and Gemma3 NaN embeddings from sliding-window all-padding windows are distinct bugs." + }, + { + "left": "issue:41720", + "right": "issue:45081", + "accept": false, + "reason": "Qwen3 auto device mapping cudaErrorAssert and Mistral regex patch crashing are unrelated." + }, + { + "left": "issue:43531", + "right": "issue:44291", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and init_empty_weights/_is_hf_initialized are different regressions." + }, + { + "left": "issue:44265", + "right": "issue:45072", + "accept": false, + "reason": "torch.export with torch_compilable_check and dtype mismatches in bfloat16 inference are separate issues." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall: the cluster is mostly heterogeneous, and the two accepted soft edges are plausibly related duplicate-level regressions based on the titles. The conservative false verdicts are also reasonable. The representative issue choice and summary stay within the packet evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 11, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:35707", + "issue:39692", + "issue:40444", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43262", + "issue:43296", + "issue:43377", + "issue:43454", + "issue:43482", + "issue:43550", + "issue:43606", + "issue:43611", + "issue:43632", + "issue:43644", + "issue:43646", + "issue:43701", + "issue:43746", + "issue:43761", + "issue:43792", + "issue:43819", + "issue:43881", + "issue:43957", + "issue:43994", + "issue:44079", + "issue:44186", + "issue:44265", + "issue:44291", + "issue:44387", + "issue:44466", + "issue:44479", + "issue:44488", + "issue:44534", + "issue:44568", + "issue:44610", + "issue:44617", + "issue:44877", + "issue:44898", + "issue:45072", + "issue:45325", + "issue:45381", + "issue:45800", + "issue:45823", + "issue:45834", + "issue:45920", + "issue:46305", + "issue:46315" + ], + "soft_pairs": [ + "issue:43611|issue:43646", + "issue:43632|issue:45072", + "issue:43994|issue:44898", + "issue:42175|issue:43296", + "issue:45325|issue:45834", + "issue:43792|issue:43957", + "issue:43454|issue:44466", + "issue:44079|issue:44534", + "issue:42915|issue:43482", + "issue:40444|issue:45381", + "issue:45325|issue:45381", + "issue:43644|issue:44079", + "issue:35707|issue:40444", + "issue:43761|issue:44079", + "issue:44479|issue:45823", + "issue:44387|issue:44617", + "issue:43606|issue:44265", + "issue:44291|issue:44877", + "issue:43296|issue:45800", + "issue:43701|issue:45325", + "issue:44568|issue:45920", + "issue:43606|issue:43819", + "issue:41720|issue:45823", + "issue:44186|issue:44265", + "issue:43454|issue:43550", + "issue:43262|issue:43746", + "issue:46305|issue:46315", + "issue:39692|issue:40444", + "issue:43881|issue:44488", + "issue:43377|issue:44610", + "issue:43454|issue:45072" + ], + "prepared_review_unit_hash": "9f90680cb9dcc4154e73cf6b8cedeaf1b7ed709d9d86e07f0757f60e3e390e0b", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37162, + "estimated_input_tokens": 9291, + "estimated_eval_tokens": 18838 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a grab-bag of mostly unrelated Transformers issues, with many superficial similarities around v5 regressions, loading, multimodal inputs, or device/memory errors. None of the soft pairs look like true duplicates of the same concrete bug.", + "confidence": 0.88, + "canonical_issue_reason": "Issue 43646 is the broadest representative of the cluster\u2019s central theme: a Transformers v5 model-initialization regression affecting loading behavior. It is not a duplicate of the others, but it best anchors the shared v5 regression pattern.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43646 is the most representative issue for the overall cluster because it describes a core v5 initialization breakage, which is the closest common thread among many other regression reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43611", + "right": "issue:43646", + "accept": false, + "reason": "Both involve v5 loading/init breakage, but one is about `base_model_prefix` and the other about custom model initialization; different failure modes and code paths." + }, + { + "left": "issue:43632", + "right": "issue:45072", + "accept": false, + "reason": "`_is_hf_initialized` flag handling and bfloat16 dtype mismatch are unrelated bugs." + }, + { + "left": "issue:43994", + "right": "issue:44898", + "accept": false, + "reason": "Different models and different symptoms: SigLIP2 nonsensical outputs vs Perceiver resolution/interpolation failure." + }, + { + "left": "issue:42175", + "right": "issue:43296", + "accept": false, + "reason": "A backend packaging/install issue is unrelated to a PaddleOCR-VL model load failure." + }, + { + "left": "issue:45325", + "right": "issue:45834", + "accept": false, + "reason": "Both mention indexing/positioning, but one is Qwen2.5-VL rope position logic and the other is Kosmos2.5 OCR length handling; not the same bug." + }, + { + "left": "issue:43792", + "right": "issue:43957", + "accept": false, + "reason": "Whisper load/runtime failure and meta-device loading regression are different issues." + }, + { + "left": "issue:43454", + "right": "issue:44466", + "accept": false, + "reason": "Both mention tied `lm_head` weights, but one is a model config not tying weights and the other is inconsistent serialization behavior." + }, + { + "left": "issue:44079", + "right": "issue:44534", + "accept": false, + "reason": "ModelOutput key assignment and non-persistent buffer junk are separate internals bugs." + }, + { + "left": "issue:42915", + "right": "issue:43482", + "accept": false, + "reason": "FineGrainedFP8 failure and GGUF loading failure are unrelated model-loading problems." + }, + { + "left": "issue:40444", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen2.5-VL related, but one is finetuning with multiple images and the other is video position-id regression; not enough to treat as the same bug." + }, + { + "left": "issue:45325", + "right": "issue:45381", + "accept": false, + "reason": "Both concern Qwen2.5-VL positional indexing, but one targets still-image temporal positions and the other video inputs; related, not the same concrete defect." + }, + { + "left": "issue:43644", + "right": "issue:44079", + "accept": false, + "reason": "A junk-buffer serialization regression and a ModelOutput keying bug are different code paths." + }, + { + "left": "issue:35707", + "right": "issue:40444", + "accept": false, + "reason": "Progressive generation with `inputs_embeds`/`past_key_values` is unrelated to multimodal finetuning with multiple images." + }, + { + "left": "issue:43761", + "right": "issue:44079", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and ModelOutput key assignment are separate issues." + }, + { + "left": "issue:44479", + "right": "issue:45823", + "accept": false, + "reason": "Video input regression in Qwen family and Gemma4 device mismatch are different bugs." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Both are OOM reports, but one is int4 reserved-memory behavior and the other is a model-specific video OOM." + }, + { + "left": "issue:43606", + "right": "issue:44265", + "accept": false, + "reason": "CPU offload device mismatch and torch.export export failure with `torch_compilable_check` are unrelated." + }, + { + "left": "issue:44291", + "right": "issue:44877", + "accept": false, + "reason": "Unexpected `_is_hf_initialized` argument and strict config rejection are different loading regressions." + }, + { + "left": "issue:43296", + "right": "issue:45800", + "accept": false, + "reason": "A specific model load failure and a broad torch/Transformers compatibility issue are not the same bug." + }, + { + "left": "issue:43701", + "right": "issue:45325", + "accept": false, + "reason": "Checkpoint key mismatch and Qwen2.5-VL rope-index regression are unrelated." + }, + { + "left": "issue:44568", + "right": "issue:45920", + "accept": false, + "reason": "Tokenizer special-token omission and wrong token IDs across models are tokenizer regressions, but not the same concrete defect." + }, + { + "left": "issue:43606", + "right": "issue:43819", + "accept": false, + "reason": "CPU offload device mismatch is unrelated to DAC latents/STE forward mismatch." + }, + { + "left": "issue:41720", + "right": "issue:45823", + "accept": false, + "reason": "Both are device-related, but one is an auto device-mapping CUDA assert and the other is a PLE forward device mismatch." + }, + { + "left": "issue:44186", + "right": "issue:44265", + "accept": false, + "reason": "Tokenizer crash on padding/truncation and export failures with `torch_compilable_check` are different subsystems." + }, + { + "left": "issue:43454", + "right": "issue:43550", + "accept": false, + "reason": "Weight tying in AyaVisionConfig is unrelated to Bamba-9B-v2 torch.compile/SDPA failure." + }, + { + "left": "issue:43262", + "right": "issue:43746", + "accept": false, + "reason": "Audio chat-template sampling-rate default and PEFT adapter checkpoint loading are unrelated." + }, + { + "left": "issue:46305", + "right": "issue:46315", + "accept": false, + "reason": "Sliding-window mask off-by-one and slow `convert_tokens_to_ids` are separate regressions." + }, + { + "left": "issue:39692", + "right": "issue:40444", + "accept": false, + "reason": "Documentation/example errors for SigLIP2 are unrelated to Qwen2.5-VL finetuning failures." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "Different models and different load failures; no shared concrete bug." + }, + { + "left": "issue:43377", + "right": "issue:44610", + "accept": false, + "reason": "Padding-mask batch discrepancy in MIMI and OmDet-Turbo processor/model size mismatch are unrelated." + }, + { + "left": "issue:43454", + "right": "issue:45072", + "accept": false, + "reason": "lm_head weight tying/serialization and bfloat16 dtype mismatch are different problems." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 12, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:34634", + "issue:35707", + "issue:36331", + "issue:40990", + "issue:41720", + "issue:42175", + "issue:42617", + "issue:42915", + "issue:42947", + "issue:43257", + "issue:43317", + "issue:43377", + "issue:43404", + "issue:43450", + "issue:43454", + "issue:43526", + "issue:43550", + "issue:43572", + "issue:43577", + "issue:43708", + "issue:43720", + "issue:43873", + "issue:43881", + "issue:43883", + "issue:44008", + "issue:44186", + "issue:44387", + "issue:44423", + "issue:44466", + "issue:44610", + "issue:44756", + "issue:44877", + "issue:44898", + "issue:44977", + "issue:44991", + "issue:45072", + "issue:45325", + "issue:45357", + "issue:45406", + "issue:45584", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45823", + "issue:45923", + "issue:46291", + "issue:46315" + ], + "soft_pairs": [ + "issue:40990|issue:45584", + "issue:43257|issue:45923", + "issue:43317|issue:43873", + "issue:43572|issue:45759", + "issue:40990|issue:42947", + "issue:43377|issue:44898", + "issue:43873|issue:45701", + "issue:43454|issue:43720", + "issue:45325|issue:46315", + "issue:41720|issue:44977", + "issue:44466|issue:45357", + "issue:44991|issue:46291", + "issue:43526|issue:43883", + "issue:42617|issue:42915", + "issue:35707|issue:41720", + "issue:34634|issue:35707", + "issue:43881|issue:44991", + "issue:43577|issue:43708", + "issue:44008|issue:45823", + "issue:43404|issue:45357", + "issue:43450|issue:45701", + "issue:44186|issue:45072", + "issue:44877|issue:44991", + "issue:43550|issue:43720", + "issue:36331|issue:42915", + "issue:44423|issue:45406", + "issue:36331|issue:42617", + "issue:42175|issue:46291", + "issue:43873|issue:44756", + "issue:44387|issue:45704", + "issue:44610|issue:44898" + ], + "prepared_review_unit_hash": "bbd1be4a3cecee02f4664352c8a67aa5e5df8f450c9707ee354c0810e48a35be", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36692, + "estimated_input_tokens": 9173, + "estimated_eval_tokens": 18602 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "All soft edges are false positives: the titles often overlap on broad themes like quantization, offloading, multimodal serving, or save/load behavior, but each report points to a different concrete bug and code path.", + "confidence": 0.74, + "canonical_issue_reason": "No single issue is a clear canonical for this set; if forced, issue 43873 is the broadest open report with the most discussion/inbound references, but it is still distinct from most others.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43873 is the best representative only in a loose sense because it is open, broadly scoped, and has the most cross-references; however, the cluster is too heterogeneous for a true umbrella issue.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:40990", + "right": "issue:45584", + "accept": false, + "reason": "Different symptoms and models: perplexity on gpt-oss-20b vs Whisper empty-transcription generation after align_special_tokens." + }, + { + "left": "issue:43257", + "right": "issue:45923", + "accept": false, + "reason": "One is MoE weight conversion during accelerate+deepspeed loading; the other is a missing gradient-checkpointing flag in a trust_remote_code variant." + }, + { + "left": "issue:43317", + "right": "issue:43873", + "accept": false, + "reason": "Both mention offloading/quantization, but one is dequantized model loading with gpu+cpu offload while the other is general offloading behavior under quantization." + }, + { + "left": "issue:43572", + "right": "issue:45759", + "accept": false, + "reason": "Unrelated config issues: missing pad_token_idx in StableLmConfig vs failing to unwrap text_config for composite Qwen multimodal configs." + }, + { + "left": "issue:40990", + "right": "issue:42947", + "accept": false, + "reason": "Perplexity regression on a specific model/data pair is not the same bug as PEFT LoRA gradient checkpointing being ineffective." + }, + { + "left": "issue:43377", + "right": "issue:44898", + "accept": false, + "reason": "Both involve batched inputs and shape differences, but one is MIMI padding-mask behavior and the other is Perceiver image-resolution handling." + }, + { + "left": "issue:43873", + "right": "issue:45701", + "accept": false, + "reason": "Offloading/quantization is unrelated to tokenization behavior changes across versions." + }, + { + "left": "issue:43454", + "right": "issue:43720", + "accept": false, + "reason": "One is missing lm_head tying in a multimodal model; the other is packed-weight unpacking during accelerate loading for BitNet." + }, + { + "left": "issue:45325", + "right": "issue:46315", + "accept": false, + "reason": "Qwen2.5-VL rope-index temporal scaling and slow tokenizer conversion performance are unrelated code paths." + }, + { + "left": "issue:41720", + "right": "issue:44977", + "accept": false, + "reason": "Both are Qwen-family failures, but one is auto device mapping cuda assert and the other is flash-attention generation correctness." + }, + { + "left": "issue:44466", + "right": "issue:45357", + "accept": false, + "reason": "Both touch save/load serialization, but one is tied lm_head.weight behavior by device and the other is incorrect visual encoder keys in Qwen3.5." + }, + { + "left": "issue:44991", + "right": "issue:46291", + "accept": false, + "reason": "Tokenizer loading for est-roberta and kernels v0.15 version/revision requirements are separate dependency/config issues." + }, + { + "left": "issue:43526", + "right": "issue:43883", + "accept": false, + "reason": "A label-reduction bug in BeitImageProcessorFast is unrelated to a missing all_tied_weights_keys attribute in MolmoForCausalLM." + }, + { + "left": "issue:42617", + "right": "issue:42915", + "accept": false, + "reason": "3d_parallel.py startup failure and Qwen3Moe FineGrainedFP8Config failure are different model/runtime problems." + }, + { + "left": "issue:35707", + "right": "issue:41720", + "accept": false, + "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to Qwen3 auto device mapping cuda asserts." + }, + { + "left": "issue:34634", + "right": "issue:35707", + "accept": false, + "reason": "BarkProcessor voice_preset is a processor issue, while the other is generation with inputs_embeds and past_key_values." + }, + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "glm-4v-9b loading failure and tokenizer loading failure for est-roberta are different models and failure modes." + }, + { + "left": "issue:43577", + "right": "issue:43708", + "accept": false, + "reason": "Model dtype propagation on Blip2 is unrelated to Trainer resume_from_checkpoint max_steps calculation." + }, + { + "left": "issue:44008", + "right": "issue:45823", + "accept": false, + "reason": "Both are Gemma-family forward issues, but one is a variable-name collision in Gemma3nModel.forward and the other is a device mismatch in Gemma4 PLE with device_map=auto." + }, + { + "left": "issue:43404", + "right": "issue:45357", + "accept": false, + "reason": "Missing lm_head tying in Mistral3 differs from incorrect visual encoder keys during Qwen3.5 save_pretrained." + }, + { + "left": "issue:43450", + "right": "issue:45701", + "accept": false, + "reason": "Video processor output shape mismatch is unrelated to tokenizer behavior changes." + }, + { + "left": "issue:44186", + "right": "issue:45072", + "accept": false, + "reason": "LayoutLMv2 tokenization crashes on NER/batching are unrelated to dtype mismatches in SwitchTransformers and TimmWrapperModel." + }, + { + "left": "issue:44877", + "right": "issue:44991", + "accept": false, + "reason": "Strict config blocking granite_speech loading is a config-class issue, not the same as tokenizer loading for est-roberta." + }, + { + "left": "issue:43550", + "right": "issue:43720", + "accept": false, + "reason": "Torch.compile with SDPA on Bamba-9B-v2 and packed-weight unpacking during accelerate loading are different failure paths." + }, + { + "left": "issue:36331", + "right": "issue:42915", + "accept": false, + "reason": "Unexpected num_items_in_batch in CustomTrainer.compute_loss is unrelated to FineGrainedFP8Config failures in Qwen3Moe." + }, + { + "left": "issue:44423", + "right": "issue:45406", + "accept": false, + "reason": "Both hit transformers serve, but one crashes on multimodal Qwen3.5 inputs and the other on Gemma4Processor missing _tokenizer." + }, + { + "left": "issue:36331", + "right": "issue:42617", + "accept": false, + "reason": "CustomTrainer compute_loss signature mismatch and inability to run 3d_parallel.py are unrelated problems." + }, + { + "left": "issue:42175", + "right": "issue:46291", + "accept": false, + "reason": "TensorFlow backend packaging and kernels version/revision requirements are unrelated dependency issues." + }, + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "General quantization offloading bugs are not the same as disabling mmap on Strix Halo to avoid OOM." + }, + { + "left": "issue:44387", + "right": "issue:45704", + "accept": false, + "reason": "CUDA reserved-memory OOM under int4 quantization is unrelated to T5 using apex.FusedRMSNorm and its memory leak." + }, + { + "left": "issue:44610", + "right": "issue:44898", + "accept": false, + "reason": "Processor/model resolution mismatch in OmDet-Turbo and Perceiver interpolation behavior are different image-preprocessing bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 13, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:29159", + "issue:30064", + "issue:34634", + "issue:36010", + "issue:36032", + "issue:40990", + "issue:41720", + "issue:42617", + "issue:42915", + "issue:43295", + "issue:43316", + "issue:43317", + "issue:43322", + "issue:43421", + "issue:43425", + "issue:43525", + "issue:43526", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43572", + "issue:43606", + "issue:43653", + "issue:43720", + "issue:43749", + "issue:43782", + "issue:43881", + "issue:44186", + "issue:44361", + "issue:44492", + "issue:44509", + "issue:44530", + "issue:44534", + "issue:44610", + "issue:44671", + "issue:44743", + "issue:44913", + "issue:45081", + "issue:45161", + "issue:45406", + "issue:45464", + "issue:45632", + "issue:45684", + "issue:45701", + "issue:45759", + "issue:45812", + "issue:46050", + "issue:46291" + ], + "soft_pairs": [ + "issue:43749|issue:44534", + "issue:43550|issue:43606", + "issue:44186|issue:44610", + "issue:43322|issue:46050", + "issue:45632|issue:45684", + "issue:42915|issue:43317", + "issue:43295|issue:45701", + "issue:41720|issue:43322", + "issue:40990|issue:45701", + "issue:41720|issue:42617", + "issue:41720|issue:45161", + "issue:36010|issue:42617", + "issue:43425|issue:46291", + "issue:45081|issue:45812", + "issue:43525|issue:46291", + "issue:43316|issue:44913", + "issue:43531|issue:43572", + "issue:44671|issue:44743", + "issue:43540|issue:43782", + "issue:43421|issue:45812", + "issue:44186|issue:44361", + "issue:43526|issue:43881", + "issue:29159|issue:36032", + "issue:45406|issue:45464", + "issue:30064|issue:34634", + "issue:43550|issue:43653", + "issue:43606|issue:43720", + "issue:43421|issue:45701", + "issue:44530|issue:46291", + "issue:42915|issue:43322", + "issue:43606|issue:43653", + "issue:43531|issue:45759", + "issue:44492|issue:44509" + ], + "prepared_review_unit_hash": "a5f121fa25e4ec2b23e304644b0c1a2fb911c19f0cfc3777834470f06c5f95e7", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "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": "The set is mostly a collection of superficially similar but distinct bugs. I found no soft edge that looks like the same concrete issue/change, so all pairwise candidates are rejected. The closest anchor is the tokenizer-regression subgroup, with issue 45701 as the broadest representative and issue 45812 as the most actionable regression report.", + "confidence": 0.58, + "canonical_issue_reason": "issue:45701 is the broadest tokenizer-regression report and is the most plausible umbrella anchor for the tokenizer-related subset; the rest are mostly model- or component-specific failures.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45812 is the clearest high-impact regression report, with a specific silent tokenization break across Granite models and strong triage value.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43749", + "right": "issue:44534", + "accept": false, + "reason": "Different failure modes: FSDP CPU RAM efficient loading vs non-persistent buffer initialization; same training/runtime area, not the same bug." + }, + { + "left": "issue:43550", + "right": "issue:43606", + "accept": false, + "reason": "Both are runtime/model-loading failures, but one is torch.compile+SDPA on Bamba and the other is CPU-offload device mismatch on Bark." + }, + { + "left": "issue:44186", + "right": "issue:44610", + "accept": false, + "reason": "Tokenizer NER/padding crash in LayoutLMv2 is unrelated to OmDet-Turbo processor size mismatch." + }, + { + "left": "issue:43322", + "right": "issue:46050", + "accept": false, + "reason": "Segfault while loading Llava Next is a different bug from quantized-save/load corruption in sinq." + }, + { + "left": "issue:45632", + "right": "issue:45684", + "accept": false, + "reason": "Both involve save/load workflows, but one is a cache-path collision and the other is file permission propagation." + }, + { + "left": "issue:42915", + "right": "issue:43317", + "accept": false, + "reason": "Both touch model loading/quantization, but FineGrainedFP8 failure and dequantized gpu+cpu offload failure are different code paths." + }, + { + "left": "issue:43295", + "right": "issue:45701", + "accept": false, + "reason": "Both mention tokenization regressions, but one is about processor.tokenizer/images in v4.57.5 and the other is a broader tokenization change report." + }, + { + "left": "issue:41720", + "right": "issue:43322", + "accept": false, + "reason": "Different model families and symptoms: cudaErrorAssert in Qwen3 auto device mapping vs segfault on Llava Next load." + }, + { + "left": "issue:40990", + "right": "issue:45701", + "accept": false, + "reason": "High perplexity on one model is not the same as a generic tokenization-version complaint, even if tokenization could be a factor." + }, + { + "left": "issue:41720", + "right": "issue:42617", + "accept": false, + "reason": "Qwen3 auto device-map assert is unrelated to failing to run 3d_parallel.py." + }, + { + "left": "issue:41720", + "right": "issue:45161", + "accept": false, + "reason": "Both are Qwen3-related, but auto device mapping failure and TP-only MoE failure are distinct issues." + }, + { + "left": "issue:36010", + "right": "issue:42617", + "accept": false, + "reason": "ImportError for GenerationMixin and a 3d_parallel.py runtime failure are unrelated." + }, + { + "left": "issue:43425", + "right": "issue:46291", + "accept": false, + "reason": "Torch 2.10 incompatibility is not the same as kernels requiring version/revision metadata." + }, + { + "left": "issue:45081", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but one is a Mistral regex/backend crash and the other is wrong token IDs for Granite models." + }, + { + "left": "issue:43525", + "right": "issue:46291", + "accept": false, + "reason": "Missing pad_token_id in Llama4Config is unrelated to the kernels import/version requirement." + }, + { + "left": "issue:43316", + "right": "issue:44913", + "accept": false, + "reason": "Both are config/API discrepancies, but they affect different fields and model configs." + }, + { + "left": "issue:43531", + "right": "issue:43572", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE and missing pad_token_idx in StableLmConfig are different regressions." + }, + { + "left": "issue:44671", + "right": "issue:44743", + "accept": false, + "reason": "CamemBERT masked-LM prediction errors and Qwen3 modular recurrent-state reset are unrelated model behaviors." + }, + { + "left": "issue:43540", + "right": "issue:43782", + "accept": false, + "reason": "Video-processing ValueError in Qwen3OmniMoe is not the same as Qwen3VL weight_only loading failure." + }, + { + "left": "issue:43421", + "right": "issue:45812", + "accept": false, + "reason": "Runtime special-token post-processor update bug is not the same as Granite-wide wrong token IDs on AutoTokenizer." + }, + { + "left": "issue:44186", + "right": "issue:44361", + "accept": false, + "reason": "Two tokenizer bugs, but different models and failure surfaces: batched padding/truncation vs task-time AttributeError." + }, + { + "left": "issue:43526", + "right": "issue:43881", + "accept": false, + "reason": "A BEiTImageProcessorFast label reduction bug is unrelated to a glm-4v-9b load failure." + }, + { + "left": "issue:29159", + "right": "issue:36032", + "accept": false, + "reason": "Slow/fast tokenizer inconsistency is a different issue from T5Tokenizer's add_special_tokens method conflict." + }, + { + "left": "issue:45406", + "right": "issue:45464", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer in serve and Qwen3.5 streaming inference failures are separate bugs." + }, + { + "left": "issue:30064", + "right": "issue:34634", + "accept": false, + "reason": "Image processor void-segmentation handling and BarkProcessor voice_preset are unrelated processor bugs." + }, + { + "left": "issue:43550", + "right": "issue:43653", + "accept": false, + "reason": "torch.compile+SDPA on Bamba and BigBird mask-token registration/decode issues do not share the same underlying code path." + }, + { + "left": "issue:43606", + "right": "issue:43720", + "accept": false, + "reason": "CPU-offload device mismatch for Bark-small and BitNet packed-weight unpacking during accelerate loading are distinct loading bugs." + }, + { + "left": "issue:43421", + "right": "issue:45701", + "accept": false, + "reason": "Special-token runtime post-processing and broad tokenization-version changes overlap only loosely; not the same concrete bug." + }, + { + "left": "issue:44530", + "right": "issue:46291", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash on Qwen3.5 is unrelated to kernels import metadata requirements." + }, + { + "left": "issue:42915", + "right": "issue:43322", + "accept": false, + "reason": "FineGrainedFP8Config failure and Llava Next segfault are different model-loading problems." + }, + { + "left": "issue:43606", + "right": "issue:43653", + "accept": false, + "reason": "CPU offload device mismatch is not the same as missing BigBird special-token registration." + }, + { + "left": "issue:43531", + "right": "issue:45759", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior and AutoModelForCausalLM from_config text_config unwrapping are unrelated." + }, + { + "left": "issue:44492", + "right": "issue:44509", + "accept": false, + "reason": "A cache-strategy typo and outdated pipeline docs are both doc-ish, but not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 14, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:33666", + "issue:34567", + "issue:34634", + "issue:35707", + "issue:39692", + "issue:41720", + "issue:42490", + "issue:42886", + "issue:43329", + "issue:43381", + "issue:43388", + "issue:43421", + "issue:43525", + "issue:43531", + "issue:43540", + "issue:43653", + "issue:43746", + "issue:43844", + "issue:43873", + "issue:43927", + "issue:44062", + "issue:44186", + "issue:44368", + "issue:44451", + "issue:44514", + "issue:44545", + "issue:44589", + "issue:44610", + "issue:44843", + "issue:44898", + "issue:44918", + "issue:44936", + "issue:44987", + "issue:44991", + "issue:45042", + "issue:45059", + "issue:45072", + "issue:45081", + "issue:45356", + "issue:45357", + "issue:45588", + "issue:45663", + "issue:45704", + "issue:45759", + "issue:45859", + "issue:45923", + "issue:46294" + ], + "soft_pairs": [ + "issue:34567|issue:35707", + "issue:43388|issue:45059", + "issue:44514|issue:44545", + "issue:43873|issue:45704", + "issue:42490|issue:43329", + "issue:41720|issue:43746", + "issue:43421|issue:43927", + "issue:33666|issue:35707", + "issue:44898|issue:45072", + "issue:45759|issue:45859", + "issue:44451|issue:44991", + "issue:43531|issue:44589", + "issue:45588|issue:45759", + "issue:44368|issue:45357", + "issue:44062|issue:44451", + "issue:44987|issue:45042", + "issue:39692|issue:44610", + "issue:45704|issue:45923", + "issue:43421|issue:43653", + "issue:43381|issue:43844", + "issue:45663|issue:45923", + "issue:42886|issue:44843", + "issue:44918|issue:44936", + "issue:45704|issue:45859", + "issue:34634|issue:41720", + "issue:43421|issue:44186", + "issue:43381|issue:45704", + "issue:43525|issue:43531", + "issue:45081|issue:45356", + "issue:43540|issue:46294", + "issue:39692|issue:43873" + ], + "prepared_review_unit_hash": "13a649a197437fc631bb71fb9dd08134a58ef351b34c859888cd6ae7d8f48d10", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36896, + "estimated_input_tokens": 9224, + "estimated_eval_tokens": 18704 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab-bag of unrelated issue reports; the only clear duplicate pair is the Qwen2_5_VLProcessor apply_chat_template crash (#44514/#44545).", + "confidence": 0.92, + "canonical_issue_reason": "#44545 is the cleaner canonical record: same bug as #44514, later report, and slightly more referenced.", + "canonical_pr_reason": null, + "best_issue_reason": "#44545 is the best representative issue in this set because it is the exact duplicate target for the only strong match and has the clearest, most specific problem statement.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:34567", + "right": "issue:35707", + "accept": false, + "reason": "Different subsystems and failures: TrainerState token counting vs progressive generation with inputs_embeds/past_key_values." + }, + { + "left": "issue:43388", + "right": "issue:45059", + "accept": false, + "reason": "Different code paths: metric gathering/label truncation vs SAM3 text+box prompt behavior." + }, + { + "left": "issue:44514", + "right": "issue:44545", + "accept": true, + "reason": "Same issue title and same crash in Qwen2_5_VLProcessor.apply_chat_template with batched input and padding=False; clear duplicate." + }, + { + "left": "issue:43873", + "right": "issue:45704", + "accept": false, + "reason": "Unrelated bugs: quantization offloading behavior vs T5/apex FusedRMSNorm memory leak." + }, + { + "left": "issue:42490", + "right": "issue:43329", + "accept": false, + "reason": "Local-path import warning and multimodal token video-branch undefined variables are unrelated." + }, + { + "left": "issue:41720", + "right": "issue:43746", + "accept": false, + "reason": "Different failures: Qwen3 device mapping cuda assert vs PEFT local checkpoint loading." + }, + { + "left": "issue:43421", + "right": "issue:43927", + "accept": false, + "reason": "Both token-related, but one is runtime post-processor updating and the other is save/load preserving custom token IDs; not the same bug." + }, + { + "left": "issue:33666", + "right": "issue:35707", + "accept": false, + "reason": "Multi-GPU training question for Qwen2-VL is not the same bug as progressive generation with inputs_embeds/past_key_values." + }, + { + "left": "issue:44898", + "right": "issue:45072", + "accept": false, + "reason": "Different models and failure modes: Perceiver interpolation vs dtype mismatch in inference." + }, + { + "left": "issue:45759", + "right": "issue:45859", + "accept": false, + "reason": "Composite config unwrapping bug vs CPU-offload incompatibility in rotary embedding; not the same code-path problem." + }, + { + "left": "issue:44451", + "right": "issue:44991", + "accept": false, + "reason": "Both tokenizer-loading issues, but they affect different models and likely different root causes." + }, + { + "left": "issue:43531", + "right": "issue:44589", + "accept": false, + "reason": "Qwen3-MoE sliding_window issue is unrelated to Float8 storage lookup failure." + }, + { + "left": "issue:45588", + "right": "issue:45759", + "accept": false, + "reason": "Flash-attention sink handling crash and Qwen multimodal config unwrapping are unrelated." + }, + { + "left": "issue:44368", + "right": "issue:45357", + "accept": false, + "reason": "A warning about tie_word_embeddings during LoRA fine-tuning is unrelated to saving incorrect visual encoder keys." + }, + { + "left": "issue:44062", + "right": "issue:44451", + "accept": false, + "reason": "AddedToken keyword conflict and ScandiBERT load failure are different bugs." + }, + { + "left": "issue:44987", + "right": "issue:45042", + "accept": false, + "reason": "Model-loading failure and PIL image processor torchvision dependency are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:44610", + "accept": false, + "reason": "Doc/example mismatch and processor/model resolution mismatch are different image-processing issues." + }, + { + "left": "issue:45704", + "right": "issue:45923", + "accept": false, + "reason": "Apex RMSNorm memory leak and missing supports_gradient_checkpointing flag are unrelated." + }, + { + "left": "issue:43421", + "right": "issue:43653", + "accept": false, + "reason": "Runtime post-processor refresh bug and BigBirdTokenizer special-token registration bug are different tokenizer issues." + }, + { + "left": "issue:43381", + "right": "issue:43844", + "accept": false, + "reason": "Eval-mode gradient checkpointing restriction and ZeRO-3 gradient explosion are not the same defect." + }, + { + "left": "issue:45663", + "right": "issue:45923", + "accept": false, + "reason": "Gemma-4 FSDP2 KeyError and missing gradient-checkpointing flag on a trust_remote_code variant are unrelated." + }, + { + "left": "issue:42886", + "right": "issue:44843", + "accept": false, + "reason": "Both concern offline/tokenizer loading, but one is cache lookup failure and the other is an unconditional model_info call in a Mistral regex patch; distinct root causes." + }, + { + "left": "issue:44918", + "right": "issue:44936", + "accept": false, + "reason": "Different trainer failures: SFT unpacking of Qwen3.5 embeddings vs evaluate() after train()." + }, + { + "left": "issue:45704", + "right": "issue:45859", + "accept": false, + "reason": "Same as above: unrelated T5 RMSNorm leak vs Qwen3_5Moe CPU offload incompatibility." + }, + { + "left": "issue:34634", + "right": "issue:41720", + "accept": false, + "reason": "BarkProcessor voice preset and Qwen3 auto device mapping are unrelated." + }, + { + "left": "issue:43421", + "right": "issue:44186", + "accept": false, + "reason": "Different tokenizer families and different failure modes: post-processor update vs LayoutLMv2 batching/padding crash." + }, + { + "left": "issue:43381", + "right": "issue:45704", + "accept": false, + "reason": "Gradient checkpointing eval restriction is unrelated to T5 apex memory leak." + }, + { + "left": "issue:43525", + "right": "issue:43531", + "accept": false, + "reason": "Missing pad_token_id on Llama4Config is unrelated to Qwen3-MoE sliding_window behavior." + }, + { + "left": "issue:45081", + "right": "issue:45356", + "accept": false, + "reason": "Both involve Mistral/Kimi tokenizer handling, but one crashes in _patch_mistral_regex while the other is a codec-handling regression with warning text; not clearly the same bug." + }, + { + "left": "issue:43540", + "right": "issue:46294", + "accept": false, + "reason": "Both are Qwen video-input issues, but one is a ValueError in Qwen3OmniMoe and the other is a feature/token mismatch in Qwen3.5-9B; likely different root causes." + }, + { + "left": "issue:39692", + "right": "issue:43873", + "accept": false, + "reason": "SigLIP2 docs/example errors are unrelated to quantization offloading behavior." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the only accepted pair (#44514/#44545) is strongly supported by the identical titles plus matching crash description. The other soft-edge rejections are conservative and do not overclaim duplicate status. No PR-pair mergeability concerns apply here." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 15, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:30064", + "issue:36010", + "issue:42915", + "issue:43295", + "issue:43335", + "issue:43388", + "issue:43421", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43596", + "issue:43618", + "issue:43643", + "issue:43653", + "issue:43697", + "issue:43716", + "issue:43720", + "issue:43746", + "issue:43756", + "issue:43782", + "issue:43819", + "issue:43873", + "issue:43950", + "issue:43994", + "issue:44008", + "issue:44062", + "issue:44077", + "issue:44186", + "issue:44220", + "issue:44488", + "issue:44625", + "issue:44661", + "issue:44671", + "issue:44811", + "issue:44843", + "issue:44877", + "issue:44898", + "issue:45072", + "issue:45081", + "issue:45183", + "issue:45356", + "issue:45381", + "issue:45479", + "issue:45834", + "issue:45907", + "issue:45987", + "issue:46032", + "issue:46050" + ], + "soft_pairs": [ + "issue:44625|issue:45479", + "issue:43782|issue:45381", + "issue:45356|issue:45907", + "issue:43994|issue:44671", + "issue:43746|issue:46050", + "issue:43720|issue:43819", + "issue:30064|issue:36010", + "issue:44077|issue:44625", + "issue:43596|issue:44220", + "issue:43653|issue:43720", + "issue:42915|issue:43746", + "issue:43540|issue:44008", + "issue:44220|issue:44661", + "issue:43618|issue:43697", + "issue:43550|issue:45072", + "issue:43746|issue:43873", + "issue:43295|issue:45834", + "issue:43550|issue:43819", + "issue:45183|issue:45479", + "issue:43388|issue:46032", + "issue:44186|issue:44898", + "issue:43335|issue:43756", + "issue:43716|issue:44008", + "issue:44062|issue:44488", + "issue:43653|issue:43819", + "issue:43531|issue:43643", + "issue:44843|issue:45081", + "issue:43950|issue:45987", + "issue:43643|issue:44877", + "issue:43421|issue:44811", + "issue:43540|issue:45381" + ], + "prepared_review_unit_hash": "60e1449be85798848ae9b983ef463237b7af30b778dbd919d2b741d7984466a2", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37453, + "estimated_input_tokens": 9364, + "estimated_eval_tokens": 18984 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is mostly heterogeneous: the soft-similarity pairs span different models, APIs, and failure modes, so there are no strong duplicate merges here. I would keep the cluster anchored on one concrete, well-documented bug report rather than merging across these pairs.", + "confidence": 0.91, + "canonical_issue_reason": "issue:46032 is the strongest representative in this set: it has a clear, concrete code-path failure, active discussion, and inbound references, making it the best anchor for the cluster. The other issues are largely orthogonal or model-specific.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:46032 is the most actionable and clearly reproduced bug here, with the strongest signal that it represents a real underlying defect worth deduplicating against if a matching report appears.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44625", + "right": "issue:45479", + "accept": false, + "reason": "Both involve labels/classification, but one is a config propagation bug and the other is a loss-function regression across sequence-classification models; different underlying defects." + }, + { + "left": "issue:43782", + "right": "issue:45381", + "accept": false, + "reason": "Different models and different failure modes: a Qwen3VL load/error path versus a video-input vision_position_ids bug in Qwen2.5-VL." + }, + { + "left": "issue:45356", + "right": "issue:45907", + "accept": false, + "reason": "Completely unrelated topics: tokenizer regression for Kimi-K2.5 versus hub API forwarding of revision in list_repo_files." + }, + { + "left": "issue:43994", + "right": "issue:44671", + "accept": false, + "reason": "Both are wrong-output reports, but they concern different models and likely different root causes; too broad to treat as the same bug." + }, + { + "left": "issue:43746", + "right": "issue:46050", + "accept": false, + "reason": "Both are load/save related, but one is PEFT adapter loading and the other is quantized model persistence; different code paths and symptoms." + }, + { + "left": "issue:43720", + "right": "issue:43819", + "accept": false, + "reason": "Different model families and failures: packed-weight loading versus DAC latent conversion / missing STE behavior." + }, + { + "left": "issue:30064", + "right": "issue:36010", + "accept": false, + "reason": "No meaningful overlap: image processor segmentation-map handling versus a GenerationMixin import error." + }, + { + "left": "issue:44077", + "right": "issue:44625", + "accept": false, + "reason": "One is about patchtsmixer post_init policy, the other about num_labels propagation; unrelated implementation bugs." + }, + { + "left": "issue:43596", + "right": "issue:44220", + "accept": false, + "reason": "Different subsystems entirely: DeepSpeed/BertModel initialization versus audio feature extraction." + }, + { + "left": "issue:43653", + "right": "issue:43720", + "accept": false, + "reason": "Tokenizer special-token registration bug versus BitNet packed-weight unpacking; not the same underlying issue." + }, + { + "left": "issue:42915", + "right": "issue:43746", + "accept": false, + "reason": "Different models and components: Qwen3Moe FP8 configuration versus GraniteSpeech PEFT checkpoint loading." + }, + { + "left": "issue:43540", + "right": "issue:44008", + "accept": false, + "reason": "Both are multimodal, but one is a Qwen3OmniMoe video-processing ValueError and the other is a Gemma3n variable-name collision; not duplicates." + }, + { + "left": "issue:44220", + "right": "issue:44661", + "accept": false, + "reason": "Audio fbank extraction and add-new-model-like / TOKENIZER_MAPPING_NAMES logic are unrelated." + }, + { + "left": "issue:43618", + "right": "issue:43697", + "accept": false, + "reason": "One is a CLIPOutput field omission, the other is RTDetrV2 behavioral drift in v5; different bug classes." + }, + { + "left": "issue:43550", + "right": "issue:45072", + "accept": false, + "reason": "Different dtype/compile problems in different models and execution paths; too broad to merge." + }, + { + "left": "issue:43746", + "right": "issue:43873", + "accept": false, + "reason": "Checkpoint loading with PEFT adapters is distinct from offloading/quantization behavior." + }, + { + "left": "issue:43295", + "right": "issue:45834", + "accept": false, + "reason": "Processor/tokenizer regression and long OCR index error are different model-specific failures." + }, + { + "left": "issue:43550", + "right": "issue:43819", + "accept": false, + "reason": "torch.compile/SDPA failure versus DAC.from_latents mismatch; unrelated code paths." + }, + { + "left": "issue:45183", + "right": "issue:45479", + "accept": false, + "reason": "One is input validation/error-message logic, the other is a zero-loss regression; not the same defect." + }, + { + "left": "issue:43388", + "right": "issue:46032", + "accept": false, + "reason": "Metric-gathering label truncation and Mamba2 cache correctness are unrelated training/runtime bugs." + }, + { + "left": "issue:44186", + "right": "issue:44898", + "accept": false, + "reason": "Tokenizer crash on NER padding/truncation versus Perceiver image-classification resolution handling; different subsystems." + }, + { + "left": "issue:43335", + "right": "issue:43756", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation and Smollm3 RoPE-layer retention are separate architecture/config issues." + }, + { + "left": "issue:43716", + "right": "issue:44008", + "accept": false, + "reason": "Different failures: dtype mismatch between image preprocessor/model versus an audio_mel_mask attribute collision." + }, + { + "left": "issue:44062", + "right": "issue:44488", + "accept": false, + "reason": "AddedToken keyword conflict is unrelated to loading a specific tokenizer/model checkpoint." + }, + { + "left": "issue:43653", + "right": "issue:43819", + "accept": false, + "reason": "Tokenizer special-token registration and DAC latent/forward mismatch are different domains and code paths." + }, + { + "left": "issue:43531", + "right": "issue:43643", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and trust_remote_code AutoConfig field loss are unrelated." + }, + { + "left": "issue:44843", + "right": "issue:45081", + "accept": false, + "reason": "Both mention _patch_mistral_regex, but the failures are different: unconditional model_info offline breakage versus backend_tokenizer AttributeError when fix_mistral_regex=True." + }, + { + "left": "issue:43950", + "right": "issue:45987", + "accept": false, + "reason": "Non-persistent buffer corruption on from_pretrained is unrelated to StaticCache.get_seq_length returning the wrong type." + }, + { + "left": "issue:43643", + "right": "issue:44877", + "accept": false, + "reason": "One is missing fields with trust_remote_code; the other is strict config blocking granite_speech loading. Similar area, different bug." + }, + { + "left": "issue:43421", + "right": "issue:44811", + "accept": false, + "reason": "Tokenizers post-processor refresh and Whisper batch_decode skip_special_tokens handling are different tokenizer/processor issues." + }, + { + "left": "issue:43540", + "right": "issue:45381", + "accept": false, + "reason": "Different Qwen video/multimodal bugs with distinct symptoms and likely different fixes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 16, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:28218", + "issue:29159", + "issue:30064", + "issue:34634", + "issue:36010", + "issue:36331", + "issue:39692", + "issue:42175", + "issue:43322", + "issue:43404", + "issue:43452", + "issue:43577", + "issue:43582", + "issue:43606", + "issue:43720", + "issue:43782", + "issue:43818", + "issue:43825", + "issue:43827", + "issue:43906", + "issue:43994", + "issue:44186", + "issue:44442", + "issue:44448", + "issue:44479", + "issue:44485", + "issue:44509", + "issue:44561", + "issue:44610", + "issue:44671", + "issue:44805", + "issue:44821", + "issue:44857", + "issue:44936", + "issue:44977", + "issue:44991", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45406", + "issue:45584", + "issue:45685", + "issue:45701", + "issue:45834", + "issue:45987", + "issue:46192", + "issue:46209" + ], + "soft_pairs": [ + "issue:44448|issue:44671", + "issue:30064|issue:36331", + "issue:43782|issue:44977", + "issue:43577|issue:43818", + "issue:39692|issue:45834", + "issue:45584|issue:45834", + "issue:43322|issue:45685", + "issue:44821|issue:45084", + "issue:43606|issue:45072", + "issue:44479|issue:45834", + "issue:43825|issue:43994", + "issue:44186|issue:44442", + "issue:34634|issue:36010", + "issue:44485|issue:44509", + "issue:45987|issue:46192", + "issue:44561|issue:45020", + "issue:28218|issue:29159", + "issue:43577|issue:43582", + "issue:44805|issue:44936", + "issue:44610|issue:45072", + "issue:43720|issue:45072", + "issue:43452|issue:44991", + "issue:43825|issue:43827", + "issue:43582|issue:44857", + "issue:43404|issue:43782", + "issue:45081|issue:45406", + "issue:43906|issue:46209", + "issue:42175|issue:45081", + "issue:44671|issue:45701" + ], + "prepared_review_unit_hash": "cc678a99f1829732f5cfe0268613c9f128518e202721d48486ebd7e55fa85e8c", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36406, + "estimated_input_tokens": 9102, + "estimated_eval_tokens": 18460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items do not form a single duplicate family; they span many unrelated areas (tokenizers, vision, model loading, trainer, docs, hardware-specific bugs). The closest overlap is the v5 pipeline-removal/docs subgroup, but even there the artifacts are not the same concrete bug or change.", + "confidence": 0.97, + "canonical_issue_reason": "Issue #43825 is the closest thing to a representative anchor because it has the clearest v5 pipeline-removal angle and some inbound traffic, but there is no true cluster-wide canonical issue here.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #43825 is the best available issue representative for the small v5 pipeline/doc overlap, though it is not a strong canonical choice for the overall set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44448", + "right": "issue:44671", + "accept": false, + "reason": "Both are model-output complaints, but they involve different models and different failure modes; not the same underlying bug." + }, + { + "left": "issue:30064", + "right": "issue:36331", + "accept": false, + "reason": "Image segmentation-map processing and trainer custom loss kwargs are unrelated code paths." + }, + { + "left": "issue:43782", + "right": "issue:44977", + "accept": false, + "reason": "One is a weight_only loading error for Qwen3VL, the other is a flash-attention generation issue for Qwen3.5; different bugs." + }, + { + "left": "issue:43577", + "right": "issue:43818", + "accept": false, + "reason": "BLIP2 dtype propagation and Video-LLaVA temporal-attention/weight-sharing are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:45834", + "accept": false, + "reason": "SigLIP2 docs errors and Kosmos2.5 OCR index error are different problems." + }, + { + "left": "issue:45584", + "right": "issue:45834", + "accept": false, + "reason": "Whisper empty transcription failure after special-token alignment is unrelated to Kosmos2.5 long-OCR indexing." + }, + { + "left": "issue:43322", + "right": "issue:45685", + "accept": false, + "reason": "Llava Next segfault and MPS histogram int support error are different models and backends." + }, + { + "left": "issue:44821", + "right": "issue:45084", + "accept": false, + "reason": "AutoImageProcessor URL loading and template-node compilation are unrelated failures." + }, + { + "left": "issue:43606", + "right": "issue:45072", + "accept": false, + "reason": "CPU offload device mismatch in bark-small and bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel are not the same bug." + }, + { + "left": "issue:44479", + "right": "issue:45834", + "accept": false, + "reason": "Video-input regression for Qwen VL models and Kosmos2.5 OCR indexing are unrelated." + }, + { + "left": "issue:43825", + "right": "issue:43994", + "accept": false, + "reason": "v5 pipeline error messaging and SigLIP2 nonsensical outputs are different issues." + }, + { + "left": "issue:44186", + "right": "issue:44442", + "accept": false, + "reason": "LayoutLMv2 tokenizer crashes on NER/batched padding and AutoTokenizer failing to load FastSpeech2ConformerTokenizer are distinct tokenizer bugs." + }, + { + "left": "issue:34634", + "right": "issue:36010", + "accept": false, + "reason": "BarkProcessor voice_preset handling and GenerationMixin import failure are unrelated." + }, + { + "left": "issue:44485", + "right": "issue:44509", + "accept": false, + "reason": "GLM-5 RoPE implementation discussion and docs mentioning removed pipeline tasks are not the same change." + }, + { + "left": "issue:45987", + "right": "issue:46192", + "accept": false, + "reason": "StaticCache.get_seq_length contract mismatch and return_tensors being ignored are separate API issues." + }, + { + "left": "issue:44561", + "right": "issue:45020", + "accept": false, + "reason": "Removal of is_torch_fx_available breaking trust_remote_code and generic remote_code breakages are related only broadly, not the same concrete bug." + }, + { + "left": "issue:28218", + "right": "issue:29159", + "accept": false, + "reason": "Tokenizer spacing after added tokens and slow/fast tokenizer inconsistency overlap conceptually, but the concrete behaviors and code paths are not clearly the same." + }, + { + "left": "issue:43577", + "right": "issue:43582", + "accept": false, + "reason": "BLIP2 dtype propagation and AppleSilicon warmup TypeError are unrelated." + }, + { + "left": "issue:44805", + "right": "issue:44936", + "accept": false, + "reason": "Mask/index shape mismatch and trainer.evaluate() after train() are different runtime failures." + }, + { + "left": "issue:44610", + "right": "issue:45072", + "accept": false, + "reason": "OmDet-Turbo input-size mismatch and dtype mismatch in other models are unrelated." + }, + { + "left": "issue:43720", + "right": "issue:45072", + "accept": false, + "reason": "BitNet packed-weight unpacking during accelerate loading and bfloat16 dtype mismatch are different code-path problems." + }, + { + "left": "issue:43452", + "right": "issue:44991", + "accept": false, + "reason": "gguf_file loading breaks and tokenizer loading for EMBEDDIA/est-roberta are different tokenizer/model-loading issues." + }, + { + "left": "issue:43825", + "right": "issue:43827", + "accept": false, + "reason": "Both mention v5 pipeline removals, but one is an error-message bug and the other is a docs reference cleanup; not the same underlying bug or change." + }, + { + "left": "issue:43582", + "right": "issue:44857", + "accept": false, + "reason": "AppleSilicon warmup TypeError and LwDetrImageLoss float16 AMP crash are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:43782", + "accept": false, + "reason": "Mistral3 lm_head weight tying and Qwen3VL weight_only loading are different model-loading defects." + }, + { + "left": "issue:45081", + "right": "issue:45406", + "accept": false, + "reason": "Mistral regex patch crash and Gemma4Processor missing _tokenizer are unrelated tokenizer/processor bugs." + }, + { + "left": "issue:43906", + "right": "issue:46209", + "accept": false, + "reason": "An isolated reproduction of a prior issue and an FP8 B200 inference failure are not the same bug." + }, + { + "left": "issue:42175", + "right": "issue:45081", + "accept": false, + "reason": "Missing TensorFlow backend in torch install and Mistral tokenizer regex crash are unrelated." + }, + { + "left": "issue:44671", + "right": "issue:45701", + "accept": false, + "reason": "CamemBERT masked LM predictions in v5 and version-dependent tokenization changes are too broad and not the same concrete defect." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 17, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:34634", + "issue:36010", + "issue:36331", + "issue:38175", + "issue:39692", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43454", + "issue:43526", + "issue:43531", + "issue:43550", + "issue:43577", + "issue:43638", + "issue:43643", + "issue:43644", + "issue:43645", + "issue:43650", + "issue:43653", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43746", + "issue:43749", + "issue:43819", + "issue:43854", + "issue:43906", + "issue:43927", + "issue:44060", + "issue:44291", + "issue:44368", + "issue:44554", + "issue:44589", + "issue:44610", + "issue:44811", + "issue:44912", + "issue:44936", + "issue:44977", + "issue:45003", + "issue:45020", + "issue:45042", + "issue:45084", + "issue:45459", + "issue:45538", + "issue:45600", + "issue:45759", + "issue:45803", + "issue:46285" + ], + "soft_pairs": [ + "issue:43262|issue:43819", + "issue:45003|issue:46285", + "issue:43526|issue:43697", + "issue:43653|issue:44912", + "issue:44060|issue:44368", + "issue:43650|issue:43906", + "issue:43742|issue:45020", + "issue:45084|issue:45538", + "issue:43329|issue:45600", + "issue:43746|issue:44811", + "issue:43644|issue:43749", + "issue:44589|issue:45042", + "issue:43638|issue:44936", + "issue:43550|issue:44912", + "issue:43819|issue:43927", + "issue:34634|issue:36331", + "issue:43854|issue:44977", + "issue:39692|issue:44912", + "issue:38175|issue:44610", + "issue:36010|issue:43645", + "issue:43720|issue:44610", + "issue:45459|issue:45803", + "issue:43454|issue:43577", + "issue:43526|issue:44554", + "issue:36010|issue:36331", + "issue:43262|issue:44811", + "issue:43749|issue:44291", + "issue:43531|issue:45042", + "issue:43257|issue:43329", + "issue:43643|issue:45759" + ], + "prepared_review_unit_hash": "515588f8868a93dd25def96dd2acdb501da8f8e76a40d1add9fe9fae24d78bcb", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 37019, + "estimated_input_tokens": 9255, + "estimated_eval_tokens": 18766 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a heterogeneous mix of unrelated bugs; none of the proposed soft pairs looks like the same concrete underlying issue, so all soft edges should be rejected.", + "confidence": 0.94, + "canonical_issue_reason": "Issue 45020 is the broadest and most representative loading-regression report in the set, but the cluster is too mixed to treat any pair as a true duplicate group.", + "canonical_pr_reason": null, + "best_issue_reason": "45020 is the best cluster anchor because it is the most general current regression report; the others are narrow, model-specific edge cases rather than duplicates of it.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43262", + "right": "issue:43819", + "accept": false, + "reason": "Both are audio-related, but one is chat-template sampling-rate handling and the other is a DAC latent/STE correctness bug; different code paths." + }, + { + "left": "issue:45003", + "right": "issue:46285", + "accept": false, + "reason": "Both involve modeling internals, but one is unsafe sys.modules access and the other is CLIP text-model structure changes; not the same bug." + }, + { + "left": "issue:43526", + "right": "issue:43697", + "accept": false, + "reason": "BEiT label reduction and RTDetrV2 output drift are separate vision-model issues with no shared concrete failure mode." + }, + { + "left": "issue:43653", + "right": "issue:44912", + "accept": false, + "reason": "Tokenizer special-token registration is unrelated to MXFP4 quantization loading behavior." + }, + { + "left": "issue:44060", + "right": "issue:44368", + "accept": false, + "reason": "Both mention tied-weights warnings, but they concern different models and different underlying warnings, so they are not duplicates." + }, + { + "left": "issue:43650", + "right": "issue:43906", + "accept": false, + "reason": "A vague placeholder issue is unrelated to an isolated reproduction of another bug." + }, + { + "left": "issue:43742", + "right": "issue:45020", + "accept": false, + "reason": "A specific MobileLLM loading KeyError is not the same concrete issue as the broader remote_code regression." + }, + { + "left": "issue:45084", + "right": "issue:45538", + "accept": false, + "reason": "These are unrelated: template-node compilation versus tokenizer max length defaults." + }, + { + "left": "issue:43329", + "right": "issue:45600", + "accept": false, + "reason": "One is a missing video-token-count implementation path; the other is stale auto-mapping config references." + }, + { + "left": "issue:43746", + "right": "issue:44811", + "accept": false, + "reason": "PEFT local-checkpoint loading and Whisper batch decoding are unrelated code paths." + }, + { + "left": "issue:43644", + "right": "issue:43749", + "accept": false, + "reason": "Non-persistent buffer junk and FSDP CPU RAM-efficient loading are both loading-related, but not the same bug or fix." + }, + { + "left": "issue:44589", + "right": "issue:45042", + "accept": false, + "reason": "Float8 storage lookup failure is unrelated to PIL image processor torchvision requirements." + }, + { + "left": "issue:43638", + "right": "issue:44936", + "accept": false, + "reason": "Zero3/non-pretrained-Bert index errors and train-then-evaluate failure are different trainer/runtime problems." + }, + { + "left": "issue:43550", + "right": "issue:44912", + "accept": false, + "reason": "torch.compile/SDPA failures are unrelated to MXFP4 quantization fallback behavior." + }, + { + "left": "issue:43819", + "right": "issue:43927", + "accept": false, + "reason": "DAC latent reconstruction mismatch and DiaConfig token-ID persistence are distinct model bugs." + }, + { + "left": "issue:34634", + "right": "issue:36331", + "accept": false, + "reason": "BarkProcessor voice_preset and CustomTrainer.compute_loss signature mismatch are unrelated." + }, + { + "left": "issue:43854", + "right": "issue:44977", + "accept": false, + "reason": "These are different model-specific generation/loading regressions, not one shared code-path bug." + }, + { + "left": "issue:39692", + "right": "issue:44912", + "accept": false, + "reason": "SigLIP2 doc/example errors are unrelated to git-oss MXFP4 quantization loading." + }, + { + "left": "issue:38175", + "right": "issue:44610", + "accept": false, + "reason": "Unexpected zero probabilities and OmDet-Turbo resolution mismatch are separate vision-model issues." + }, + { + "left": "issue:36010", + "right": "issue:43645", + "accept": false, + "reason": "GenerationMixin import breakage is not the same as the Jupyter custom-model initialization regression." + }, + { + "left": "issue:43720", + "right": "issue:44610", + "accept": false, + "reason": "BitNet packed-weight loading and OmDet-Turbo image-size mismatch are unrelated." + }, + { + "left": "issue:45459", + "right": "issue:45803", + "accept": false, + "reason": "Both involve broad exception handling, but they target different components and different failure modes." + }, + { + "left": "issue:43454", + "right": "issue:43577", + "accept": false, + "reason": "Weight tying garbage-generation and Blip2 dtype handling are different model-loading bugs." + }, + { + "left": "issue:43526", + "right": "issue:44554", + "accept": false, + "reason": "BEiT label reduction and MPS attention correctness are unrelated issues." + }, + { + "left": "issue:36010", + "right": "issue:36331", + "accept": false, + "reason": "Importing GenerationMixin and compute_loss keyword-argument incompatibility are separate API breakages." + }, + { + "left": "issue:43262", + "right": "issue:44811", + "accept": false, + "reason": "Audio chat-template sampling rate and Whisper batch_decode skip_special_tokens are different processor behaviors." + }, + { + "left": "issue:43749", + "right": "issue:44291", + "accept": false, + "reason": "FSDP CPU RAM-efficient loading and init_empty_weights _is_hf_initialized are related to low-memory loading, but not the same concrete bug." + }, + { + "left": "issue:43531", + "right": "issue:45042", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior is unrelated to PIL backend torchvision dependency requirements." + }, + { + "left": "issue:43257", + "right": "issue:43329", + "accept": false, + "reason": "MoE weight conversion with accelerate/deepspeed and multimodal token counting video-path bugs are unrelated." + }, + { + "left": "issue:43643", + "right": "issue:45759", + "accept": false, + "reason": "Both touch config/model loading, but one is missing fields from trust_remote_code and the other is text_config unwrapping for composite Qwen configs; not the same fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 18, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:24643", + "issue:29159", + "issue:33357", + "issue:34567", + "issue:38175", + "issue:40990", + "issue:41720", + "issue:42490", + "issue:42915", + "issue:43257", + "issue:43317", + "issue:43329", + "issue:43450", + "issue:43452", + "issue:43540", + "issue:43577", + "issue:43582", + "issue:43606", + "issue:43696", + "issue:43697", + "issue:43742", + "issue:43927", + "issue:43931", + "issue:44062", + "issue:44291", + "issue:44479", + "issue:44560", + "issue:44568", + "issue:44737", + "issue:44821", + "issue:44849", + "issue:44898", + "issue:44912", + "issue:44987", + "issue:44991", + "issue:45042", + "issue:45072", + "issue:45137", + "issue:45356", + "issue:45357", + "issue:45381", + "issue:45405", + "issue:45488", + "issue:45538", + "issue:45588", + "issue:45701", + "issue:45859", + "issue:45907" + ], + "soft_pairs": [ + "issue:42490|issue:43257", + "issue:43329|issue:43450", + "issue:43606|issue:45859", + "issue:34567|issue:40990", + "issue:44568|issue:45356", + "issue:29159|issue:34567", + "issue:43329|issue:44479", + "issue:42915|issue:43742", + "issue:43257|issue:43931", + "issue:44849|issue:45701", + "issue:44291|issue:45042", + "issue:43452|issue:43582", + "issue:43927|issue:44912", + "issue:43697|issue:45357", + "issue:44568|issue:45488", + "issue:41720|issue:43540", + "issue:42915|issue:44821", + "issue:33357|issue:34567", + "issue:42915|issue:45538", + "issue:44987|issue:45538", + "issue:44479|issue:45381", + "issue:43540|issue:44479", + "issue:43317|issue:45538", + "issue:43540|issue:44560", + "issue:45137|issue:45405", + "issue:44987|issue:45907", + "issue:24643|issue:29159", + "issue:43742|issue:44291", + "issue:43577|issue:45072", + "issue:38175|issue:44898", + "issue:44062|issue:44991", + "issue:43257|issue:43696", + "issue:44737|issue:45588" + ], + "prepared_review_unit_hash": "4e58f44ef3ce4f591246d3add0b5af86ebb988e17545e1beb8bb3cfd80cfcbd3", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37720, + "estimated_input_tokens": 9430, + "estimated_eval_tokens": 19116 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These issues are mostly unrelated; the soft-similarity links do not rise to duplicate-level matches. I\u2019d keep one concrete tokenizer regression as the best representative issue, but no PR is present.", + "confidence": 0.89, + "canonical_issue_reason": "issue:45488 is the clearest standalone report: a specific tokenizer regression with concrete impact on DeepSeek V3/R1-style models. It is more actionable than the vaguer tokenization reports in the set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45488 is the best issue to anchor this cluster because it is specific, reproducible, and describes a real user-facing regression rather than a broad symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42490", + "right": "issue:43257", + "accept": false, + "reason": "Both involve loading, but one is a path/import warning and the other is Qwen3 MoE weight conversion with accelerate+deepspeed; different failures and code paths." + }, + { + "left": "issue:43329", + "right": "issue:43450", + "accept": false, + "reason": "Both are video/multimodal related, but one is an undefined-helper bug in video token counting and the other is a batched-shape bug in video processors; not the same defect." + }, + { + "left": "issue:43606", + "right": "issue:45859", + "accept": false, + "reason": "Both mention CPU offload, but they affect different models and different modules; no evidence they are the same concrete bug." + }, + { + "left": "issue:34567", + "right": "issue:40990", + "accept": false, + "reason": "TrainerState token counting not updating is unrelated to GPT-OSS perplexity behavior." + }, + { + "left": "issue:44568", + "right": "issue:45356", + "accept": false, + "reason": "Both are tokenizer regressions, but they describe different model-specific issues and different tokenizer behaviors." + }, + { + "left": "issue:29159", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer consistency versus TrainerState token counting are unrelated areas." + }, + { + "left": "issue:43329", + "right": "issue:44479", + "accept": false, + "reason": "The latter is a broad video-input regression across Qwen VL families; the former is a specific missing-undefined-helper bug. They may be in the same area, but not clearly the same underlying defect." + }, + { + "left": "issue:42915", + "right": "issue:43742", + "accept": false, + "reason": "Different model-loading errors: FineGrainedFP8Config for Qwen3 MoE versus a key error on MobileLLM-125M." + }, + { + "left": "issue:43257", + "right": "issue:43931", + "accept": false, + "reason": "Both are model loading problems, but one is MOE weight conversion with deepspeed while the other is a Qwen3-VL shape mismatch; different root causes." + }, + { + "left": "issue:44849", + "right": "issue:45701", + "accept": false, + "reason": "Qwen3.5 hidden-state output bug is unrelated to the generic tokenization-change report." + }, + { + "left": "issue:44291", + "right": "issue:45042", + "accept": false, + "reason": "Unexpected init_empty_weights argument and PIL image-processor torchvision dependency are unrelated." + }, + { + "left": "issue:43452", + "right": "issue:43582", + "accept": false, + "reason": "GGUF loading failures and Apple Silicon caching_allocator_warmup TypeError are unrelated." + }, + { + "left": "issue:43927", + "right": "issue:44912", + "accept": false, + "reason": "DiaConfig save/load token IDs and MXFP4 quantization fallback are different bugs." + }, + { + "left": "issue:43697", + "right": "issue:45357", + "accept": false, + "reason": "RTDetrV2 output changes and Qwen3.5 save_pretrained visual-key corruption are different code paths." + }, + { + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "Both are tokenizer regressions, but one is about add_special_tokens/BOS-EOS behavior and the other is LlamaTokenizer overriding tokenizer.json pre-tokenizer settings; different concrete bugs." + }, + { + "left": "issue:41720", + "right": "issue:43540", + "accept": false, + "reason": "Auto device-mapping CUDA assert and Qwen3OmniMoe video ValueError are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:44821", + "accept": false, + "reason": "Qwen3 MoE FP8 loading and AutoImageProcessor-from-URL are different failures." + }, + { + "left": "issue:33357", + "right": "issue:34567", + "accept": false, + "reason": "MacOS bus error with CLIP and TrainerState token counting are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:45538", + "accept": false, + "reason": "Qwen3 MoE FP8 load failure is unrelated to CLIPTokenizer's max-length bug." + }, + { + "left": "issue:44987", + "right": "issue:45538", + "accept": false, + "reason": "Loading physical-intelligence/fast and CLIPTokenizer model_max_length are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:45381", + "accept": false, + "reason": "Both concern video input in Qwen VL models, but they report different regressions and symptoms; not enough to treat as the same bug." + }, + { + "left": "issue:43540", + "right": "issue:44479", + "accept": false, + "reason": "Both are video-related, but one is a Qwen3OmniMoe processing error and the other is a broader v5.3.0 regression across multiple VL families; different concrete problems." + }, + { + "left": "issue:43317", + "right": "issue:45538", + "accept": false, + "reason": "Dequantized model offload failure and CLIPTokenizer max-length bug are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44560", + "accept": false, + "reason": "Both involve video inputs, but they affect different models and describe different failures (ValueError vs StopIteration)." + }, + { + "left": "issue:45137", + "right": "issue:45405", + "accept": false, + "reason": "DeepSpeed ZeRO3 deque error and a PEFT version bump are unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45907", + "accept": false, + "reason": "Model loading regression and list_repo_files not forwarding revision are unrelated." + }, + { + "left": "issue:24643", + "right": "issue:29159", + "accept": false, + "reason": "DeepSpeed training weight shape error and tokenizer slow/fast inconsistency are unrelated." + }, + { + "left": "issue:43742", + "right": "issue:44291", + "accept": false, + "reason": "MobileLLM key error and init_empty_weights argument handling are different loading bugs." + }, + { + "left": "issue:43577", + "right": "issue:45072", + "accept": false, + "reason": "Both mention dtype issues, but they affect different architectures and code paths; not the same bug." + }, + { + "left": "issue:38175", + "right": "issue:44898", + "accept": false, + "reason": "SIGLIP zero probabilities and Perceiver non-default-resolution failure are different vision-model issues." + }, + { + "left": "issue:44062", + "right": "issue:44991", + "accept": false, + "reason": "AddedToken keyword collision and est-roberta tokenizer loading failure are not the same defect." + }, + { + "left": "issue:43257", + "right": "issue:43696", + "accept": false, + "reason": "Qwen3 MoE weight conversion and GPT-OSS out-of-memory are unrelated." + }, + { + "left": "issue:44737", + "right": "issue:45588", + "accept": false, + "reason": "XLNet CPU placement bug and flash_attention s_aux=None crash are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 19, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:28282", + "issue:29127", + "issue:34567", + "issue:34689", + "issue:36010", + "issue:36331", + "issue:42175", + "issue:43257", + "issue:43404", + "issue:43452", + "issue:43493", + "issue:43526", + "issue:43531", + "issue:43582", + "issue:43643", + "issue:43646", + "issue:43697", + "issue:43742", + "issue:43749", + "issue:43756", + "issue:43824", + "issue:43906", + "issue:43994", + "issue:44062", + "issue:44291", + "issue:44488", + "issue:44589", + "issue:44625", + "issue:44756", + "issue:44792", + "issue:44821", + "issue:44877", + "issue:44913", + "issue:44964", + "issue:44977", + "issue:45042", + "issue:45081", + "issue:45216", + "issue:45325", + "issue:45357", + "issue:45375", + "issue:45381", + "issue:45405", + "issue:45479", + "issue:45600", + "issue:45920", + "issue:45987", + "issue:46121" + ], + "soft_pairs": [ + "issue:43404|issue:45987", + "issue:34567|issue:36331", + "issue:43646|issue:43749", + "issue:44062|issue:45081", + "issue:44913|issue:46121", + "issue:43906|issue:45381", + "issue:43493|issue:43994", + "issue:43749|issue:44756", + "issue:43257|issue:43756", + "issue:44792|issue:44964", + "issue:44291|issue:44821", + "issue:28282|issue:34689", + "issue:36010|issue:42175", + "issue:42175|issue:43824", + "issue:45042|issue:45081", + "issue:45357|issue:45920", + "issue:43452|issue:44488", + "issue:43742|issue:44589", + "issue:45405|issue:45600", + "issue:28282|issue:29127", + "issue:43643|issue:45375", + "issue:43526|issue:45479", + "issue:43756|issue:45325", + "issue:43531|issue:44821", + "issue:44625|issue:44877", + "issue:43697|issue:45216", + "issue:43582|issue:44977" + ], + "prepared_review_unit_hash": "686b7a32efa7d16ad8e92f75474cc1478af555c4641c4f9e5540a0659eb49b95", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 36070, + "estimated_input_tokens": 9018, + "estimated_eval_tokens": 18292 + }, + "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 loading, configs, trainers, tokenizers, and multimodal models. The SigLIP2 pair is only adjacent/possibly related, not a confirmed duplicate; the rest should remain separate.", + "confidence": 0.74, + "canonical_issue_reason": "No single canonical issue cleanly represents the cluster because the items are largely unrelated. If one must be chosen, issue 43493 is the broadest technical report and the most plausible anchor, but it does not unify the whole set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43493 is the best representative issue for the cluster only in the limited sense that it is the most detailed substantive report among the adjacent SigLIP2 items; it should not be read as a true cluster-wide duplicate anchor.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43404", + "right": "issue:45987", + "accept": false, + "reason": "Different bugs: Mistral lm_head tying vs StaticCache sequence-length return type. Same area, not the same defect." + }, + { + "left": "issue:34567", + "right": "issue:36331", + "accept": false, + "reason": "Both touch trainer behavior, but one is token counting and the other is a new compute_loss kwarg. Different code paths." + }, + { + "left": "issue:43646", + "right": "issue:43749", + "accept": false, + "reason": "One is a Transformers 5 custom model init regression; the other is FSDP CPU RAM efficient loading. Too different to merge." + }, + { + "left": "issue:44062", + "right": "issue:45081", + "accept": false, + "reason": "Different tokenizer-related failures: AddedToken argument handling vs mistral regex patching/backend access." + }, + { + "left": "issue:44913", + "right": "issue:46121", + "accept": false, + "reason": "Both involve RoPE config/validation, but one is silent reload persistence and the other is a TypeError in rope dict conversion. Not the same bug." + }, + { + "left": "issue:43906", + "right": "issue:45381", + "accept": false, + "reason": "Isolated reproduction of an old issue versus a Qwen2.5-VL video position-id regression. Different models and symptoms." + }, + { + "left": "issue:43493", + "right": "issue:43994", + "accept": false, + "reason": "These are SigLIP2-adjacent but not clearly the same bug: one reports a JAX/HF implementation discrepancy, the other nonsensical AutoModel/pipeline outputs. Related, but unconfirmed duplicate." + }, + { + "left": "issue:43749", + "right": "issue:44756", + "accept": false, + "reason": "Both concern memory/loading, but one is an FSDP loading regression and the other is a Strix Halo mmap OOM workaround. Different fixes." + }, + { + "left": "issue:43257", + "right": "issue:43756", + "accept": false, + "reason": "Qwen3 MoE weight conversion with accelerate/deepspeed is unrelated to Smollm3 RoPE layer dropping." + }, + { + "left": "issue:44792", + "right": "issue:44964", + "accept": false, + "reason": "Different multimodal model issues: Janus image-generation test failure vs Phi-4 multimodal model loading failure." + }, + { + "left": "issue:44291", + "right": "issue:44821", + "accept": false, + "reason": "init_empty_weights/_is_hf_initialized argument handling is unrelated to AutoImageProcessor URL loading." + }, + { + "left": "issue:28282", + "right": "issue:34689", + "accept": false, + "reason": "PyTorch-not-found ImportError versus a specific Llama 3.2 vision loading regression. Not the same defect." + }, + { + "left": "issue:36010", + "right": "issue:42175", + "accept": false, + "reason": "GenerationMixin import failure and missing TensorFlow backend in a torch install are unrelated packaging/import problems." + }, + { + "left": "issue:42175", + "right": "issue:43824", + "accept": false, + "reason": "Missing TensorFlow backend versus missing Qwen2_5_VL symbol. Different import/export issues." + }, + { + "left": "issue:45042", + "right": "issue:45081", + "accept": false, + "reason": "PIL image processors wrongly requiring torchvision is unrelated to Mistral regex patch crashes." + }, + { + "left": "issue:45357", + "right": "issue:45920", + "accept": false, + "reason": "Qwen3.5 save_pretrained checkpoint key regression is unrelated to AutoTokenizer producing wrong token IDs across multiple models." + }, + { + "left": "issue:43452", + "right": "issue:44488", + "accept": false, + "reason": "gguf_file loading breakage is a different root cause from the cjvt/sleng-bert load failure." + }, + { + "left": "issue:43742", + "right": "issue:44589", + "accept": false, + "reason": "MobileLLM-125M key error and missing Float8 storage are distinct loading/runtime failures." + }, + { + "left": "issue:45405", + "right": "issue:45600", + "accept": false, + "reason": "PEFT version bump and auto_mappings referencing removed configs are unrelated maintenance issues." + }, + { + "left": "issue:28282", + "right": "issue:29127", + "accept": false, + "reason": "PyTorch missing ImportError and LayoutLMv3 box validation messaging are unrelated." + }, + { + "left": "issue:43643", + "right": "issue:45375", + "accept": false, + "reason": "trust_remote_code missing fields is a general config-return issue; Qwen3_5MoeVisionConfig missing deepstack_visual_indexes is a specific schema omission." + }, + { + "left": "issue:43526", + "right": "issue:45479", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug and single-label classification zero-loss regression are unrelated." + }, + { + "left": "issue:43756", + "right": "issue:45325", + "accept": false, + "reason": "Both involve RoPE, but they affect different models and different calculations (layer dropping vs position-id scaling)." + }, + { + "left": "issue:43531", + "right": "issue:44821", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior is unrelated to AutoImageProcessor URL loading." + }, + { + "left": "issue:44625", + "right": "issue:44877", + "accept": false, + "reason": "Num-label propagation bug and strict config blocking granite_speech loading are different configuration issues." + }, + { + "left": "issue:43697", + "right": "issue:45216", + "accept": false, + "reason": "RTDetrV2 output differences in v5 and Qwen3.5 save_pretrained checkpoint corruption are separate regressions." + }, + { + "left": "issue:43582", + "right": "issue:44977", + "accept": false, + "reason": "Apple Silicon caching_allocator_warmup TypeError and Qwen3.5 flash-attention generation failure are unrelated." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The summary is mostly grounded, but it overstates one soft-edge pair: issue 43493 vs 43994 is marked deterministic_accept=false in the packet, so calling it the same underlying bug is too strong without more evidence. Please keep that verdict conservative (e.g. 'possibly related' or 'adjacent but unconfirmed') and avoid elevating the pair to accepted." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 20, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:24643", + "issue:29127", + "issue:29159", + "issue:30064", + "issue:33357", + "issue:33666", + "issue:38175", + "issue:41720", + "issue:41762", + "issue:43295", + "issue:43452", + "issue:43531", + "issue:43550", + "issue:43577", + "issue:43653", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43819", + "issue:43906", + "issue:43927", + "issue:44361", + "issue:44442", + "issue:44464", + "issue:44466", + "issue:44484", + "issue:44493", + "issue:44514", + "issue:44568", + "issue:44704", + "issue:44743", + "issue:44811", + "issue:44843", + "issue:44898", + "issue:44936", + "issue:44977", + "issue:44991", + "issue:45005", + "issue:45290", + "issue:45356", + "issue:45372", + "issue:45381", + "issue:45542", + "issue:45823", + "issue:46050", + "issue:46170", + "issue:46178", + "issue:46202" + ], + "soft_pairs": [ + "issue:43720|issue:44898", + "issue:44514|issue:45381", + "issue:24643|issue:29127", + "issue:41762|issue:44936", + "issue:43452|issue:43577", + "issue:44466|issue:45005", + "issue:43906|issue:46202", + "issue:45372|issue:46170", + "issue:43452|issue:44843", + "issue:43819|issue:44811", + "issue:43295|issue:45356", + "issue:30064|issue:33666", + "issue:44936|issue:45823", + "issue:44484|issue:44704", + "issue:43653|issue:44361", + "issue:44361|issue:44442", + "issue:38175|issue:43295", + "issue:44514|issue:45290", + "issue:45542|issue:46178", + "issue:43653|issue:44568", + "issue:43577|issue:43697", + "issue:44493|issue:46050", + "issue:29159|issue:33357", + "issue:44843|issue:44991", + "issue:44464|issue:44743", + "issue:43653|issue:43927", + "issue:41720|issue:43906", + "issue:43531|issue:43742", + "issue:43550|issue:43927", + "issue:44464|issue:44977", + "issue:43550|issue:44361" + ], + "prepared_review_unit_hash": "f7c42ba2f36d82ed2d43516914f8b848bbb4656ab991e142654151986afa6179", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37237, + "estimated_input_tokens": 9310, + "estimated_eval_tokens": 18876 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: the issues share broad Transformers/tokenizer/model-loading themes, but the soft pairs describe different concrete bugs and code paths. I would not merge any of the proposed soft edges; no PRs are present.", + "confidence": 0.74, + "canonical_issue_reason": "If a single issue must stand in for the cluster, issue:44843 is the best representative because it is a broad framework-level loading regression (AutoTokenizer/_patch_mistral_regex/offline mode) rather than a model-specific edge case.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue:44843 is the most suitable global representative because it concerns a central loading path with wider impact across models and environments, while most other issues are narrower model-specific failures.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43720", + "right": "issue:44898", + "accept": false, + "reason": "Different failures in different code paths: BitNet accelerate loading/unpacking vs Perceiver image classification with interpolation." + }, + { + "left": "issue:44514", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen2.5-VL related, but one is chat-template batching/padding and the other is video vision_position_ids; not the same bug." + }, + { + "left": "issue:24643", + "right": "issue:29127", + "accept": false, + "reason": "Unrelated errors: DeepSpeed training weight shape vs LayoutLMv3 box-content validation/message clarity." + }, + { + "left": "issue:41762", + "right": "issue:44936", + "accept": false, + "reason": "Different paths and symptoms: Gemma3 ZeRO-3 loading failure vs trainer.evaluate() breaking after train()." + }, + { + "left": "issue:43452", + "right": "issue:43577", + "accept": false, + "reason": "Tokenizer/AutoModel gguf_file loading breakage is unrelated to Blip2 dtype remaining float32 on load." + }, + { + "left": "issue:44466", + "right": "issue:45005", + "accept": false, + "reason": "Both mention tied weights, but one is device-dependent serialization and the other is translation-model behavior; too broad to treat as the same bug." + }, + { + "left": "issue:43906", + "right": "issue:46202", + "accept": false, + "reason": "Different reproductions: issue 38071 replay vs Gemma pipeline roundtrip failure on a specific token." + }, + { + "left": "issue:45372", + "right": "issue:46170", + "accept": false, + "reason": "Unrelated import/dependency failures: mistral_common ReasoningEffort for Gemma 4 vs Windows/ROCm torch.distributed issues." + }, + { + "left": "issue:43452", + "right": "issue:44843", + "accept": false, + "reason": "Both affect model loading, but gguf_file handling and offline model_info() calls are distinct failure modes." + }, + { + "left": "issue:43819", + "right": "issue:44811", + "accept": false, + "reason": "DAC latents/STE mismatch is unrelated to Whisper batch_decode skipping special tokens." + }, + { + "left": "issue:43295", + "right": "issue:45356", + "accept": false, + "reason": "Different regressions in processor/tokenizer behavior: processor.tokenizer access vs Kimi-K2.5 codec/regex warning handling." + }, + { + "left": "issue:30064", + "right": "issue:33666", + "accept": false, + "reason": "Void segmentation maps in image processing vs Qwen2-VL multi-GPU training are unrelated." + }, + { + "left": "issue:44936", + "right": "issue:45823", + "accept": false, + "reason": "Trainer state after training vs Gemma4 PLE device mismatch during forward are different bugs." + }, + { + "left": "issue:44484", + "right": "issue:44704", + "accept": false, + "reason": "Shard-size configuration question is unrelated to AutoProcessor kwargs not reaching cached_file." + }, + { + "left": "issue:43653", + "right": "issue:44361", + "accept": false, + "reason": "BigBirdTokenizer special-token registration bug is unrelated to MLukeTokenizer attribute errors." + }, + { + "left": "issue:44361", + "right": "issue:44442", + "accept": false, + "reason": "Different tokenizer failures: MLukeTokenizer task handling vs AutoTokenizer loading FastSpeech2ConformerTokenizer." + }, + { + "left": "issue:38175", + "right": "issue:43295", + "accept": false, + "reason": "Siglip2 zero probabilities are unrelated to the processor.tokenizer regression in v4.57.5." + }, + { + "left": "issue:44514", + "right": "issue:45290", + "accept": false, + "reason": "Same API surface, but different crash triggers: batched padding=False vs assistant tool-call messages without content." + }, + { + "left": "issue:45542", + "right": "issue:46178", + "accept": false, + "reason": "Different backend registration problems: missing TensorFlow backend vs offending tensorflow_text backend mapping." + }, + { + "left": "issue:43653", + "right": "issue:44568", + "accept": false, + "reason": "Different tokenizer bugs affecting different models and token types." + }, + { + "left": "issue:43577", + "right": "issue:43697", + "accept": false, + "reason": "Blip2 dtype persistence and RTDetrV2 output drift are unrelated." + }, + { + "left": "issue:44493", + "right": "issue:46050", + "accept": false, + "reason": "Position-id key warnings and sinq quantization save/load breakage are different serialization/compatibility issues." + }, + { + "left": "issue:29159", + "right": "issue:33357", + "accept": false, + "reason": "Slow-vs-fast tokenizer inconsistency is unrelated to the MacOS CLIP bus error." + }, + { + "left": "issue:44843", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer-loading regressions, but one is offline model_info() patching and the other is a specific model tokenizer load failure." + }, + { + "left": "issue:44464", + "right": "issue:44743", + "accept": false, + "reason": "Chunked generation with compiled forward is a different problem from recurrent states being reset with cache and seq_len>1." + }, + { + "left": "issue:43653", + "right": "issue:43927", + "accept": false, + "reason": "BigBirdTokenizer special token registration and DiaConfig custom token ID persistence are unrelated." + }, + { + "left": "issue:41720", + "right": "issue:43906", + "accept": false, + "reason": "Qwen3 auto device mapping cudaErrorAssert is unrelated to an isolated reproduction of issue 38071." + }, + { + "left": "issue:43531", + "right": "issue:43742", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and MobileLLM key errors are different code paths." + }, + { + "left": "issue:43550", + "right": "issue:43927", + "accept": false, + "reason": "torch.compile SDPA failure in Bamba-9B-v2 is unrelated to DiaConfig save/load token ID loss." + }, + { + "left": "issue:44464", + "right": "issue:44977", + "accept": false, + "reason": "Different generation regressions: compiled-forward chunking vs flash-attention behavior in Qwen3.5." + }, + { + "left": "issue:43550", + "right": "issue:44361", + "accept": false, + "reason": "Compilation/SDPA failure and tokenizer AttributeError are not the same underlying bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 21, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:28282", + "issue:32090", + "issue:33357", + "issue:33666", + "issue:34567", + "issue:35141", + "issue:36010", + "issue:36331", + "issue:39692", + "issue:41720", + "issue:43262", + "issue:43295", + "issue:43452", + "issue:43493", + "issue:43504", + "issue:43525", + "issue:43550", + "issue:43572", + "issue:43577", + "issue:43606", + "issue:43643", + "issue:43688", + "issue:43720", + "issue:43742", + "issue:43756", + "issue:43825", + "issue:43927", + "issue:43994", + "issue:44361", + "issue:44451", + "issue:44464", + "issue:44466", + "issue:44561", + "issue:44610", + "issue:44821", + "issue:44849", + "issue:44913", + "issue:44977", + "issue:45357", + "issue:45440", + "issue:45491", + "issue:45588", + "issue:45823", + "issue:46002", + "issue:46050", + "issue:46192", + "issue:46209" + ], + "soft_pairs": [ + "issue:44977|issue:46209", + "issue:43643|issue:44561", + "issue:32090|issue:34567", + "issue:43577|issue:45588", + "issue:28282|issue:35141", + "issue:45491|issue:46002", + "issue:32090|issue:33666", + "issue:43504|issue:43550", + "issue:43504|issue:43720", + "issue:44466|issue:46050", + "issue:43504|issue:44610", + "issue:41720|issue:44464", + "issue:33666|issue:36010", + "issue:43688|issue:45440", + "issue:43742|issue:44821", + "issue:45823|issue:46050", + "issue:28282|issue:32090", + "issue:28282|issue:33357", + "issue:43262|issue:43295", + "issue:43720|issue:43927", + "issue:44849|issue:46209", + "issue:43606|issue:43927", + "issue:43994|issue:46192", + "issue:43452|issue:44451", + "issue:33666|issue:36331", + "issue:43756|issue:44913", + "issue:43720|issue:44361", + "issue:43525|issue:43572", + "issue:43493|issue:43825", + "issue:44849|issue:45357", + "issue:39692|issue:43493" + ], + "prepared_review_unit_hash": "847c0d6dc4362ab8e392a7078022c42ee3bec7f7679e3105330f40260afe4b68", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36554, + "estimated_input_tokens": 9139, + "estimated_eval_tokens": 18534 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab-bag of unrelated issue reports that only share superficial regression/loading/model keywords. I don\u2019t see any soft pair that looks like the same underlying bug or change, and there are no PRs in the set.", + "confidence": 0.96, + "canonical_issue_reason": "Issue 44561 is the broadest, clearest representative of the cluster\u2019s overall v5/trust_remote_code regression theme, but it is still not a duplicate hub for the other reports.", + "canonical_pr_reason": null, + "best_issue_reason": "44561 is the most general and actionable issue in the set, with a concrete compatibility regression and a clearer scope than the other mostly model-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44977", + "right": "issue:46209", + "accept": false, + "reason": "Different bugs: Qwen3.5 flash-attention generation failure vs FP8 inference failure on B200." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "Both touch trust_remote_code, but one is missing config fields and the other is removal of a helper breaking models; not the same code-path bug." + }, + { + "left": "issue:32090", + "right": "issue:34567", + "accept": false, + "reason": "Unrelated Trainer distributed broadcast TypeError vs TrainerState token-count tracking bug." + }, + { + "left": "issue:43577", + "right": "issue:45588", + "accept": false, + "reason": "Different subsystems: Blip2 dtype propagation vs flash_attention handling of s_aux=None." + }, + { + "left": "issue:28282", + "right": "issue:35141", + "accept": false, + "reason": "Missing PyTorch import and embedding reinitialization are unrelated failures." + }, + { + "left": "issue:45491", + "right": "issue:46002", + "accept": false, + "reason": "Both mention NaNs, but one is Gemma3 sliding-window attention and the other is classification-head init/checkpointing." + }, + { + "left": "issue:32090", + "right": "issue:33666", + "accept": false, + "reason": "Distributed broadcast type error is not the same as multi-GPU training behavior for Qwen2-VL." + }, + { + "left": "issue:43504", + "right": "issue:43550", + "accept": false, + "reason": "One is a legacy-field load bug in Beit; the other is a torch.compile/SDPA failure in Bamba." + }, + { + "left": "issue:43504", + "right": "issue:43720", + "accept": false, + "reason": "Both are loading issues, but the concrete causes differ: legacy preset field vs unpacking packed BitNet weights." + }, + { + "left": "issue:44466", + "right": "issue:46050", + "accept": false, + "reason": "Both concern save/load behavior, but one is tied-weight serialization and the other is sinq quantization persistence." + }, + { + "left": "issue:43504", + "right": "issue:44610", + "accept": false, + "reason": "Model preset loading bug vs processor/model input-size mismatch; different layers of the stack." + }, + { + "left": "issue:41720", + "right": "issue:44464", + "accept": false, + "reason": "CUDA assert under auto device mapping is unrelated to inconsistent chunked generation with compiled forward." + }, + { + "left": "issue:33666", + "right": "issue:36010", + "accept": false, + "reason": "Multi-GPU training issue and GenerationMixin import failure are different problems." + }, + { + "left": "issue:43688", + "right": "issue:45440", + "accept": false, + "reason": "Both involve MoE models, but one is auxiliary-loss normalization and the other is a native-vs-remote implementation divergence." + }, + { + "left": "issue:43742", + "right": "issue:44821", + "accept": false, + "reason": "Key error loading MobileLLM-125M is unrelated to loading AutoImageProcessor from a URL." + }, + { + "left": "issue:45823", + "right": "issue:46050", + "accept": false, + "reason": "Device mismatch with device_map=\"auto\" is not the same as quantized save/load breakage." + }, + { + "left": "issue:28282", + "right": "issue:32090", + "accept": false, + "reason": "ImportError for PyTorch and GPU broadcast TypeError are unrelated." + }, + { + "left": "issue:28282", + "right": "issue:33357", + "accept": false, + "reason": "PyTorch import missing vs Mac bus error in CLIP are different failure modes and code paths." + }, + { + "left": "issue:43262", + "right": "issue:43295", + "accept": false, + "reason": "Audio chat-template sampling-rate default bug and processor.tokenizer/image-passing regression are separate issues." + }, + { + "left": "issue:43720", + "right": "issue:43927", + "accept": false, + "reason": "BitNet loading with packed weights and DiaConfig custom-token-ID loss are unrelated save/load bugs." + }, + { + "left": "issue:44849", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3.5 hidden-states bug is unrelated to FP8 inference failure on B200." + }, + { + "left": "issue:43606", + "right": "issue:43927", + "accept": false, + "reason": "CPU-offload device mismatch and config token-ID loss do not share the same underlying bug." + }, + { + "left": "issue:43994", + "right": "issue:46192", + "accept": false, + "reason": "SigLIP2 inference correctness issues are unrelated to return_tensors being ignored when text_kwargs is passed." + }, + { + "left": "issue:43452", + "right": "issue:44451", + "accept": false, + "reason": "GGUF-related tokenizer/model loading breakage is not the same as ScandiBERT loading failure." + }, + { + "left": "issue:33666", + "right": "issue:36331", + "accept": false, + "reason": "Qwen2-VL multi-GPU training and CustomTrainer.compute_loss signature mismatch are different Trainer problems." + }, + { + "left": "issue:43756", + "right": "issue:44913", + "accept": false, + "reason": "Smollm3 RoPE-layer behavior and GPTNeoXConfig reload of rotary_pct are related only thematically, not as the same bug." + }, + { + "left": "issue:43720", + "right": "issue:44361", + "accept": false, + "reason": "BitNet model loading and MLukeTokenizer task AttributeError are unrelated." + }, + { + "left": "issue:43525", + "right": "issue:43572", + "accept": false, + "reason": "Similar symptom class, but different configs and missing fields: Llama4 pad_token_id vs StableLm pad_token_idx." + }, + { + "left": "issue:43493", + "right": "issue:43825", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and pipeline translation-task messaging are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:43493", + "accept": false, + "reason": "Documentation-example errors are not the same underlying bug as the SigLIP2 implementation discrepancy." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 22, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:33357", + "issue:33666", + "issue:39692", + "issue:43257", + "issue:43278", + "issue:43295", + "issue:43299", + "issue:43334", + "issue:43493", + "issue:43504", + "issue:43540", + "issue:43572", + "issue:43575", + "issue:43577", + "issue:43606", + "issue:43646", + "issue:43653", + "issue:43688", + "issue:43749", + "issue:43756", + "issue:43824", + "issue:43844", + "issue:43927", + "issue:44112", + "issue:44291", + "issue:44361", + "issue:44466", + "issue:44479", + "issue:44488", + "issue:44493", + "issue:44554", + "issue:44568", + "issue:44589", + "issue:44625", + "issue:44843", + "issue:44964", + "issue:44977", + "issue:45083", + "issue:45200", + "issue:45229", + "issue:45405", + "issue:45520", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:46285" + ], + "soft_pairs": [ + "issue:44964|issue:45405", + "issue:44479|issue:44625", + "issue:43577|issue:44554", + "issue:43688|issue:43749", + "issue:44568|issue:45812", + "issue:43540|issue:44977", + "issue:45200|issue:45229", + "issue:44488|issue:44843", + "issue:39692|issue:43295", + "issue:43688|issue:45704", + "issue:44493|issue:46285", + "issue:43756|issue:45083", + "issue:43493|issue:43844", + "issue:43334|issue:43572", + "issue:43646|issue:44466", + "issue:45229|issue:45823", + "issue:33357|issue:33666", + "issue:43334|issue:45759", + "issue:43824|issue:44589", + "issue:43927|issue:44361", + "issue:43278|issue:43844", + "issue:43504|issue:43606", + "issue:44589|issue:45520", + "issue:44625|issue:45759", + "issue:43257|issue:43299", + "issue:43572|issue:44291", + "issue:43575|issue:45405", + "issue:43688|issue:45701", + "issue:43606|issue:44112", + "issue:43504|issue:43653" + ], + "prepared_review_unit_hash": "23323ac74f7c20ee6c7c4ef873357eb1972a8ee9d2e77effe6d23bad4ffe9ca4", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36905, + "estimated_input_tokens": 9227, + "estimated_eval_tokens": 18710 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is not a duplicate set; it mixes many unrelated model-loading, tokenizer, config, dtype, and training/runtime regressions across different model families. None of the soft edges look like the same underlying bug/change, so all should be rejected.", + "confidence": 0.94, + "canonical_issue_reason": "If a single issue must represent the cluster, issue 43646 is the broadest umbrella ('Transformers 5.0.0 breaks custom model initialization'), but it still does not actually duplicate most of the others.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43646 is the most generic version-wide regression in the set and best serves as an approximate umbrella, though the cluster is too heterogeneous for a true canonical duplicate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44964", + "right": "issue:45405", + "accept": false, + "reason": "Different problems: Phi-4 multimodal loading regression vs an unreleased PEFT version bump." + }, + { + "left": "issue:44479", + "right": "issue:44625", + "accept": false, + "reason": "One is a video-input regression in Qwen VL models; the other is num_labels propagation in Qwen3.5 config plumbing." + }, + { + "left": "issue:43577", + "right": "issue:44554", + "accept": false, + "reason": "Blip2 dtype-casting bug is unrelated to the MPS attention correctness issue." + }, + { + "left": "issue:43688", + "right": "issue:43749", + "accept": false, + "reason": "Aux-loss normalization and FSDP CPU RAM efficient loading are different code paths and failure modes." + }, + { + "left": "issue:44568", + "right": "issue:45812", + "accept": false, + "reason": "Both involve tokenization, but one is BOS/EOS insertion for mdeberta while the other is Granite token ID regression; not the same bug." + }, + { + "left": "issue:43540", + "right": "issue:44977", + "accept": false, + "reason": "Qwen3Omni video input processing is unrelated to Qwen3.5 flash-attention generation behavior." + }, + { + "left": "issue:45200", + "right": "issue:45229", + "accept": false, + "reason": "Gemma4 token-type defaults for fine-tuning vs multi-GPU inference OOM are distinct issues." + }, + { + "left": "issue:44488", + "right": "issue:44843", + "accept": false, + "reason": "Model load failure for a specific repo is not the same as the offline model_info call regression." + }, + { + "left": "issue:39692", + "right": "issue:43295", + "accept": false, + "reason": "Documentation/example mistakes are not the same as a runtime processor.tokenizer API regression." + }, + { + "left": "issue:43688", + "right": "issue:45704", + "accept": false, + "reason": "Different model-specific bugs: auxiliary loss normalization vs T5 apex RMSNorm memory leak." + }, + { + "left": "issue:44493", + "right": "issue:46285", + "accept": false, + "reason": "Unexpected position-id keys during loading is unrelated to CLIP text-model attribute structure changes." + }, + { + "left": "issue:43756", + "right": "issue:45083", + "accept": false, + "reason": "Smollm3 RoPE-layer dropping and qwen3_omni_moe feature-length helper behavior are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:43844", + "accept": false, + "reason": "HF-vs-JAX SigLIP2 discrepancy is not the same as the ZeRO-3 gradient blow-up bug." + }, + { + "left": "issue:43334", + "right": "issue:43572", + "accept": false, + "reason": "Both mention missing token IDs, but they affect different configs/models and different fields." + }, + { + "left": "issue:43646", + "right": "issue:44466", + "accept": false, + "reason": "Custom model initialization breakage is separate from device-dependent lm_head serialization." + }, + { + "left": "issue:45229", + "right": "issue:45823", + "accept": false, + "reason": "OOM in Gemma4 inference and PLE device mismatch during forward are different failures." + }, + { + "left": "issue:33357", + "right": "issue:33666", + "accept": false, + "reason": "MacOS CLIP bus error and Qwen2-VL multi-GPU training are unrelated." + }, + { + "left": "issue:43334", + "right": "issue:45759", + "accept": false, + "reason": "Missing pad_token_id in Qwen3-VL config is not the same as failing to unwrap text_config for composite multimodal configs." + }, + { + "left": "issue:43824", + "right": "issue:44589", + "accept": false, + "reason": "ImportError for a missing Qwen2.5-VL class and a float8 storage lookup error are unrelated." + }, + { + "left": "issue:43927", + "right": "issue:44361", + "accept": false, + "reason": "Config save/load losing token IDs is a different bug from MLukeTokenizer AttributeError on tasks." + }, + { + "left": "issue:43278", + "right": "issue:43844", + "accept": false, + "reason": "Embedding dtype mismatch at eval time is unrelated to the ZeRO-3 gradient increase issue." + }, + { + "left": "issue:43504", + "right": "issue:43606", + "accept": false, + "reason": "Legacy config field loading and CPU-offload device mismatch are different classes of bugs." + }, + { + "left": "issue:44589", + "right": "issue:45520", + "accept": false, + "reason": "Float8 storage resolution and flash_attn import key errors are different backend/import issues." + }, + { + "left": "issue:44625", + "right": "issue:45759", + "accept": false, + "reason": "Num_labels propagation and text_config unwrapping are related to config plumbing but not the same concrete failure." + }, + { + "left": "issue:43257", + "right": "issue:43299", + "accept": false, + "reason": "Accelerate+DeepSpeed weight conversion bug is distinct from the Qwen3VL MOE loading regression." + }, + { + "left": "issue:43572", + "right": "issue:44291", + "accept": false, + "reason": "StableLm pad_token_idx loss and unexpected _is_hf_initialized argument are unrelated v5 regressions." + }, + { + "left": "issue:43575", + "right": "issue:45405", + "accept": false, + "reason": "Tensor-parallel OOM for Qwen2-57B is unrelated to the PEFT version pin issue." + }, + { + "left": "issue:43688", + "right": "issue:45701", + "accept": false, + "reason": "Auxiliary loss normalization and tokenization changes are entirely different issues." + }, + { + "left": "issue:43606", + "right": "issue:44112", + "accept": false, + "reason": "Bark CPU-offload device mismatch and GraniteSpeech stale-device CI test are not the same bug." + }, + { + "left": "issue:43504", + "right": "issue:43653", + "accept": false, + "reason": "Beit legacy-field loading failure and BigBirdTokenizer special-token decoding are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 23, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:15354", + "issue:28218", + "issue:29159", + "issue:29942", + "issue:30333", + "issue:32090", + "issue:33357", + "issue:34567", + "issue:38175", + "issue:39692", + "issue:43232", + "issue:43278", + "issue:43295", + "issue:43450", + "issue:43493", + "issue:43540", + "issue:43688", + "issue:43701", + "issue:43844", + "issue:43873", + "issue:43994", + "issue:44186", + "issue:44423", + "issue:44464", + "issue:44493", + "issue:44521", + "issue:44554", + "issue:44610", + "issue:44734", + "issue:44792", + "issue:44849", + "issue:44964", + "issue:45072", + "issue:45084", + "issue:45198", + "issue:45200", + "issue:45290", + "issue:45356", + "issue:45405", + "issue:45406", + "issue:45446", + "issue:45520", + "issue:45584", + "issue:45588", + "issue:45823", + "issue:46050", + "issue:46121", + "issue:46192" + ], + "soft_pairs": [ + "issue:44792|issue:45405", + "issue:44423|issue:44734", + "issue:44464|issue:45584", + "issue:45588|issue:46050", + "issue:43994|issue:44493", + "issue:29942|issue:30333", + "issue:43844|issue:44964", + "issue:43688|issue:43873", + "issue:28218|issue:34567", + "issue:29159|issue:32090", + "issue:44521|issue:45290", + "issue:44554|issue:45446", + "issue:15354|issue:38175", + "issue:45084|issue:45520", + "issue:43540|issue:45200", + "issue:43873|issue:46050", + "issue:45200|issue:45823", + "issue:43278|issue:43701", + "issue:45072|issue:45198", + "issue:43232|issue:44849", + "issue:44186|issue:46192", + "issue:45406|issue:46121", + "issue:45823|issue:46192", + "issue:28218|issue:33357", + "issue:43493|issue:43701", + "issue:43295|issue:43450", + "issue:43295|issue:44610", + "issue:39692|issue:46192", + "issue:44554|issue:45356" + ], + "prepared_review_unit_hash": "2e5d353488406f25c9f83a07ce3d79bcb3b51cc0525213b47bf802c08eb2ced2", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36481, + "estimated_input_tokens": 9121, + "estimated_eval_tokens": 18498 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a set of unrelated Transformers bugs that only share broad keywords like tokenizers, multimodal processors, generation, or serving. I found no soft pair that clearly refers to the same underlying defect, so no merge-worthy duplicates here.", + "confidence": 0.91, + "canonical_issue_reason": "issue:44521 is the best anchor if one must pick a canonical issue: it is active, broadly scoped to multimodal `apply_chat_template`, and has inbound references; however, it does not actually unify the rest of the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44521 is the strongest standalone issue in this set because it is clearly described, currently open, and broad enough to represent a real multimodal template regression.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44792", + "right": "issue:45405", + "accept": false, + "reason": "Completely unrelated: a Janus test failure vs a PEFT version-pin release issue." + }, + { + "left": "issue:44423", + "right": "issue:44734", + "accept": false, + "reason": "Both are serve crashes, but the endpoints and failure modes differ (`str.to` on multimodal batching vs KV-cache tensor indexing)." + }, + { + "left": "issue:44464", + "right": "issue:45584", + "accept": false, + "reason": "Different subsystems and bugs: compiled-forward chunked generation vs Whisper empty-transcription generation." + }, + { + "left": "issue:45588", + "right": "issue:46050", + "accept": false, + "reason": "Flash-attention `s_aux=None` crash is unrelated to sinq save/load quantization breakage." + }, + { + "left": "issue:43994", + "right": "issue:44493", + "accept": false, + "reason": "One is a SigLIP2 model-output bug; the other is a generic key/position-id regression across models." + }, + { + "left": "issue:29942", + "right": "issue:30333", + "accept": false, + "reason": "FlashAttention test failures and MLflow job-finalization issues are unrelated." + }, + { + "left": "issue:43844", + "right": "issue:44964", + "accept": false, + "reason": "ZeRO-3 gradient explosion is unrelated to a Phi-4 multimodal load failure." + }, + { + "left": "issue:43688", + "right": "issue:43873", + "accept": false, + "reason": "Aux-loss normalization and quantization/offloading are different bug classes." + }, + { + "left": "issue:28218", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer spacing behavior and TrainerState token counting are unrelated." + }, + { + "left": "issue:29159", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer parity issue vs Trainer NoneType broadcast error; no shared code path." + }, + { + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Both involve chat templates, but one is zero assistant masks for multimodal inputs while the other is a crash on tool-call assistant messages." + }, + { + "left": "issue:44554", + "right": "issue:45446", + "accept": false, + "reason": "MPS attention correctness and a flex_attention PyTorch version check are unrelated." + }, + { + "left": "issue:15354", + "right": "issue:38175", + "accept": false, + "reason": "TorchScript export failure on ViT is unrelated to SigLIP2 zero-probability outputs." + }, + { + "left": "issue:45084", + "right": "issue:45520", + "accept": false, + "reason": "Template-node compilation error and a Python 3.13 flash_attn import KeyError are distinct." + }, + { + "left": "issue:43540", + "right": "issue:45200", + "accept": false, + "reason": "Video-input processing error in Qwen3OmniMoe is unrelated to Gemma4 text-only mm_token_type_ids handling." + }, + { + "left": "issue:43873", + "right": "issue:46050", + "accept": false, + "reason": "Both touch quantization/offload, but one is offloading not working and the other is save/load corruption after sinq quantization." + }, + { + "left": "issue:45200", + "right": "issue:45823", + "accept": false, + "reason": "Gemma4 input-defaulting bug and Gemma4 device-map forward mismatch are different failures." + }, + { + "left": "issue:43278", + "right": "issue:43701", + "accept": false, + "reason": "Embedding dtype drift and resume checkpoint key mismatch do not share a concrete bug." + }, + { + "left": "issue:45072", + "right": "issue:45198", + "accept": false, + "reason": "Bfloat16 inference dtype mismatches and Wav2Vec2 save/tokenization failure are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:44849", + "accept": false, + "reason": "Generation kwarg update after sync_gpus and Qwen3.5 output_hidden_states bug are separate generation issues." + }, + { + "left": "issue:44186", + "right": "issue:46192", + "accept": false, + "reason": "LayoutLMv2 NER/padding crash and `return_tensors` being ignored are different tokenizer bugs." + }, + { + "left": "issue:45406", + "right": "issue:46121", + "accept": false, + "reason": "Gemma4 serve processor missing `_tokenizer` is unrelated to rope-parameter JSON list handling." + }, + { + "left": "issue:45823", + "right": "issue:46192", + "accept": false, + "reason": "Device-mapping mismatch in Gemma4 and `return_tensors` precedence are not the same defect." + }, + { + "left": "issue:28218", + "right": "issue:33357", + "accept": false, + "reason": "Added-token spacing and CLIP bus error on MacOS are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:43701", + "accept": false, + "reason": "SigLIP2 implementation discrepancy is unrelated to checkpoint key mismatches." + }, + { + "left": "issue:43295", + "right": "issue:43450", + "accept": false, + "reason": "Processor/tokenizer regression and batched video shape bug are different multimodal issues." + }, + { + "left": "issue:43295", + "right": "issue:44610", + "accept": false, + "reason": "Custom processor/tokenizer regression and OmDet-Turbo input-size mismatch are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:46192", + "accept": false, + "reason": "SigLIP2 docs/model mismatch and ignored `return_tensors` are separate problems." + }, + { + "left": "issue:44554", + "right": "issue:45356", + "accept": false, + "reason": "Attention correctness on MPS and Kimi-K2.5 tokenizer regression are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 24, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:37428", + "issue:38175", + "issue:41720", + "issue:42490", + "issue:42915", + "issue:43262", + "issue:43278", + "issue:43295", + "issue:43334", + "issue:43450", + "issue:43493", + "issue:43525", + "issue:43526", + "issue:43550", + "issue:43572", + "issue:43575", + "issue:43653", + "issue:43696", + "issue:43824", + "issue:43844", + "issue:43856", + "issue:43873", + "issue:43950", + "issue:43957", + "issue:43975", + "issue:44112", + "issue:44442", + "issue:44466", + "issue:44514", + "issue:44521", + "issue:44561", + "issue:44898", + "issue:44987", + "issue:45127", + "issue:45198", + "issue:45200", + "issue:45325", + "issue:45381", + "issue:45412", + "issue:45520", + "issue:45759", + "issue:46170", + "issue:46178", + "issue:46192", + "issue:46209", + "issue:46234", + "issue:46285" + ], + "soft_pairs": [ + "issue:45381|issue:46209", + "issue:43295|issue:43873", + "issue:43572|issue:44987", + "issue:44898|issue:45412", + "issue:43950|issue:46285", + "issue:43572|issue:43957", + "issue:43550|issue:43844", + "issue:44466|issue:45127", + "issue:45200|issue:45325", + "issue:43450|issue:43575", + "issue:45520|issue:46178", + "issue:43334|issue:43525", + "issue:37428|issue:43957", + "issue:42490|issue:43696", + "issue:43824|issue:43975", + "issue:43278|issue:43856", + "issue:43526|issue:46192", + "issue:38175|issue:43493", + "issue:43653|issue:44112", + "issue:42915|issue:43824", + "issue:44442|issue:45198", + "issue:43295|issue:43493", + "issue:38175|issue:45412", + "issue:43262|issue:43450", + "issue:44514|issue:44521", + "issue:37428|issue:43824", + "issue:45520|issue:46170", + "issue:45759|issue:46285", + "issue:41720|issue:46209", + "issue:44561|issue:46234" + ], + "prepared_review_unit_hash": "f64c2b3d7b39f048de2e04fe73e63c9cfe16a7cc95ad95b1c0f5b94652cfd4aa", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36449, + "estimated_input_tokens": 9113, + "estimated_eval_tokens": 18482 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly noisy and spans unrelated Transformers regressions. The only soft edge that looks like a true duplicate is the pair of pad_token_id AttributeErrors in different config classes; the rest are distinct bugs affecting different models, modalities, or code paths.", + "confidence": 0.37, + "canonical_issue_reason": "issue:43334 is the clearest representative of the shared regression: a load-time AttributeError caused by missing pad_token_id on model/config objects. It is more general than the Llama4-specific report and best captures the underlying config API breakage.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43334 is the best global issue because it describes a concrete, reproducible regression in a widely shared config access path and is broad enough to cover the closely related Llama4 report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45381", + "right": "issue:46209", + "accept": false, + "reason": "Different bugs: one is wrong vision_position_ids for Qwen2.5-VL video/still image handling, the other is FP8 inference failure on B200 GPUs." + }, + { + "left": "issue:43295", + "right": "issue:43873", + "accept": false, + "reason": "Unrelated symptoms and code paths: processor.tokenizer regression vs quantization/offloading behavior." + }, + { + "left": "issue:43572", + "right": "issue:44987", + "accept": false, + "reason": "Both are load failures, but they point to different root causes and different models; not the same bug." + }, + { + "left": "issue:44898", + "right": "issue:45412", + "accept": false, + "reason": "Perceiver positional-encoding failure and RT-DETR memory release issue are unrelated." + }, + { + "left": "issue:43950", + "right": "issue:46285", + "accept": false, + "reason": "Both involve model structure/loading changes in v5, but the specific defects differ: non-persistent buffer corruption vs CLIP text_model structure mismatch." + }, + { + "left": "issue:43572", + "right": "issue:43957", + "accept": false, + "reason": "Different classes of regressions: missing pad_token_idx in a config vs meta-device loading breakage." + }, + { + "left": "issue:43550", + "right": "issue:43844", + "accept": false, + "reason": "Bamba torch.compile/SDPA failure is unrelated to DeepSpeed ZeRO-3 gradient growth." + }, + { + "left": "issue:44466", + "right": "issue:45127", + "accept": false, + "reason": "Both mention tied weights/serialization-adjacent behavior, but one is inconsistent lm_head.weight saving and the other is post-LoRA model collapse with extended vocab." + }, + { + "left": "issue:45200", + "right": "issue:45325", + "accept": false, + "reason": "Both are Qwen2.5-VL multimodal issues, but one is missing mm_token_type_ids defaults while the other is incorrect rope/temporal position scaling." + }, + { + "left": "issue:43450", + "right": "issue:43575", + "accept": false, + "reason": "Video processor batch-shape bug and Qwen2-57B tensor-parallel OOM are unrelated." + }, + { + "left": "issue:45520", + "right": "issue:46178", + "accept": false, + "reason": "Different import/backend issues: flash_attn lookup on Python 3.13 vs tensorflow_text backend mapping." + }, + { + "left": "issue:43334", + "right": "issue:43525", + "accept": true, + "reason": "Same underlying load-time regression: both configs lack pad_token_id and crash when Transformers code expects that attribute." + }, + { + "left": "issue:37428", + "right": "issue:43957", + "accept": false, + "reason": "Both are loading/import issues, but one is a missing helper import and the other is meta-device model loading breakage." + }, + { + "left": "issue:42490", + "right": "issue:43696", + "accept": false, + "reason": "Trailing-slash local path warning and GPT-oss CUDA OOM are unrelated." + }, + { + "left": "issue:43824", + "right": "issue:43975", + "accept": false, + "reason": "ImportError for Qwen2_5_VL class and DeepSeek detokenization bug are different defects." + }, + { + "left": "issue:43278", + "right": "issue:43856", + "accept": false, + "reason": "Training/eval dtype drift is unrelated to Qwen3 MoE memory inefficiency." + }, + { + "left": "issue:43526", + "right": "issue:46192", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug and return_tensors being ignored are distinct processor issues." + }, + { + "left": "issue:38175", + "right": "issue:43493", + "accept": false, + "reason": "SigLIP2 zero-probabilities and HF-vs-JAX discrepancy are related to the same model family but not the same concrete bug." + }, + { + "left": "issue:43653", + "right": "issue:44112", + "accept": false, + "reason": "BigBirdTokenizer special-token registration and GraniteSpeech CI test staleness are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:43824", + "accept": false, + "reason": "Qwen3Moe FP8 failure and Qwen2_5_VL import error are different model-specific issues." + }, + { + "left": "issue:44442", + "right": "issue:45198", + "accept": false, + "reason": "Tokenizer load failure and Wav2Vec2 save_pretrained/tokenization failure are separate bugs." + }, + { + "left": "issue:43295", + "right": "issue:43493", + "accept": false, + "reason": "Processor.tokenizer regression and SigLIP2 implementation discrepancy do not share the same failing code path." + }, + { + "left": "issue:38175", + "right": "issue:45412", + "accept": false, + "reason": "SigLIP2 probability issue and RT-DETR memory cleanup bug are unrelated." + }, + { + "left": "issue:43262", + "right": "issue:43450", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting and video batch-shape handling are different processor bugs." + }, + { + "left": "issue:44514", + "right": "issue:44521", + "accept": false, + "reason": "Both involve Qwen2.5-VL apply_chat_template, but one is a batched-input crash and the other is incorrect assistant_masks output; not clearly the same bug." + }, + { + "left": "issue:37428", + "right": "issue:43824", + "accept": false, + "reason": "Different import errors for different symbols and root causes." + }, + { + "left": "issue:45520", + "right": "issue:46170", + "accept": false, + "reason": "flash_attn import lookup on Python 3.13 and Windows ROCm torch.distributed dependency issues are unrelated." + }, + { + "left": "issue:45759", + "right": "issue:46285", + "accept": false, + "reason": "Composite Qwen multimodal config unwrapping and CLIP text_model structure mismatch are different regressions." + }, + { + "left": "issue:41720", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3 auto-device-map cudaErrorAssert and FP8 inference failure on B200 are separate hardware/runtime problems." + }, + { + "left": "issue:44561", + "right": "issue:46234", + "accept": false, + "reason": "Removal of is_torch_fx_available and missing is_falcon_mamba_ssm_available are different API compatibility issues." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded and conservative overall. The only accepted soft edge is well supported by the titles: both issues are load-time `pad_token_id` AttributeErrors on config objects, so treating them as the same regression is reasonable. The remaining pairings are phrased cautiously as distinct bugs and do not overstate evidence from the packet." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 25, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:28218", + "issue:29942", + "issue:32090", + "issue:35141", + "issue:37428", + "issue:39692", + "issue:43278", + "issue:43335", + "issue:43388", + "issue:43425", + "issue:43504", + "issue:43526", + "issue:43550", + "issue:43606", + "issue:43611", + "issue:43653", + "issue:43720", + "issue:43819", + "issue:43825", + "issue:43827", + "issue:43901", + "issue:43931", + "issue:44112", + "issue:44206", + "issue:44351", + "issue:44448", + "issue:44493", + "issue:44610", + "issue:44743", + "issue:44792", + "issue:44855", + "issue:44857", + "issue:45083", + "issue:45127", + "issue:45200", + "issue:45245", + "issue:45290", + "issue:45310", + "issue:45381", + "issue:45459", + "issue:45478", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45593", + "issue:45823", + "issue:45859", + "issue:46234" + ], + "soft_pairs": [ + "issue:44610|issue:44857", + "issue:43819|issue:44112", + "issue:43827|issue:43901", + "issue:43819|issue:45083", + "issue:44206|issue:45584", + "issue:43335|issue:43550", + "issue:44857|issue:45593", + "issue:43606|issue:45083", + "issue:44792|issue:45563", + "issue:44448|issue:44743", + "issue:37428|issue:43425", + "issue:45200|issue:45588", + "issue:45127|issue:45245", + "issue:43611|issue:43931", + "issue:45290|issue:45381", + "issue:43335|issue:43653", + "issue:44855|issue:45083", + "issue:43550|issue:44112", + "issue:43819|issue:44493", + "issue:44206|issue:45823", + "issue:29942|issue:32090", + "issue:43335|issue:43504", + "issue:45588|issue:45859", + "issue:37428|issue:44351", + "issue:43278|issue:43825", + "issue:45459|issue:46234", + "issue:39692|issue:43526", + "issue:43388|issue:43526", + "issue:43335|issue:45083", + "issue:45310|issue:45478", + "issue:39692|issue:44493", + "issue:28218|issue:35141", + "issue:43720|issue:44112", + "issue:43335|issue:44112", + "issue:43335|issue:45593" + ], + "prepared_review_unit_hash": "84da5fe9a3f434dff1060f313eb4dc4deffa6ad83056961cf96a7da8a383d3c0", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38417, + "estimated_input_tokens": 9605, + "estimated_eval_tokens": 19466 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most soft pairs are only loosely related by subsystem or release context. The only clear duplicate is the pair of Qwen3.5 MoE `from_pretrained` issues (45310 and 45478).", + "confidence": 0.86, + "canonical_issue_reason": "Issue 45310 is the clearest canonical issue because it shares the exact same title as 45478, appears earlier, and already has an inbound reference; 45478 reads like a later duplicate report of the same Qwen3.5 MoE `from_pretrained` failure.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45310 is the best overall representative for the duplicated Qwen3.5 MoE `from_pretrained` regression because it is the earlier report and already has an inbound reference.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44610", + "right": "issue:44857", + "accept": false, + "reason": "Different failures: processor/model input-size mismatch vs AMP float16 crash in a loss module." + }, + { + "left": "issue:43819", + "right": "issue:44112", + "accept": false, + "reason": "Both are runtime bugs, but they affect different code paths and symptoms." + }, + { + "left": "issue:43827", + "right": "issue:43901", + "accept": false, + "reason": "Both involve docs around v5 changes, but they are different pipeline/documentation issues." + }, + { + "left": "issue:43819", + "right": "issue:45083", + "accept": false, + "reason": "Unrelated components: DAC latent decoding vs a qwen3_omni_moe helper function." + }, + { + "left": "issue:44206", + "right": "issue:45584", + "accept": false, + "reason": "Different bugs: unsupported `center` argument in feature extraction vs empty-transcription generation failure." + }, + { + "left": "issue:43335", + "right": "issue:43550", + "accept": false, + "reason": "Different models and failures: SwitchTransformers sparse-layer construction vs Bamba SDPA/torch.compile." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "Both are training-loss related, but one is an AMP crash and the other is missing auxiliary losses." + }, + { + "left": "issue:43606", + "right": "issue:45083", + "accept": false, + "reason": "Different subsystems and root causes: CPU offload device mismatch vs qwen3_omni_moe output-length helper behavior." + }, + { + "left": "issue:44792", + "right": "issue:45563", + "accept": false, + "reason": "A failed image-generation test and a stale warning in generation are unrelated." + }, + { + "left": "issue:44448", + "right": "issue:44743", + "accept": false, + "reason": "Different models and code paths: Pegasus output regression vs Qwen recurrent-state reset in cache handling." + }, + { + "left": "issue:37428", + "right": "issue:43425", + "accept": false, + "reason": "Import error for a missing helper is not the same as general Torch 2.10 incompatibility." + }, + { + "left": "issue:45200", + "right": "issue:45588", + "accept": false, + "reason": "Gemma4 token-type-id issue and a flash-attention `s_aux=None` crash are separate problems." + }, + { + "left": "issue:45127", + "right": "issue:45245", + "accept": false, + "reason": "LoRA/vocab collapse and category-count overflow are unrelated bugs." + }, + { + "left": "issue:43611", + "right": "issue:43931", + "accept": false, + "reason": "Both are loading issues, but they involve different models and mismatch mechanisms." + }, + { + "left": "issue:45290", + "right": "issue:45381", + "accept": false, + "reason": "Chat-template tool-call handling and Qwen2.5-VL video position IDs are distinct failures." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "Different components: SwitchTransformers config vs BigBird tokenizer special-token registration." + }, + { + "left": "issue:44855", + "right": "issue:45083", + "accept": false, + "reason": "Python 3.13 import parsing error is unrelated to qwen3_omni_moe helper behavior." + }, + { + "left": "issue:43550", + "right": "issue:44112", + "accept": false, + "reason": "A compile/SDPA model bug and a CI stale-device-test failure are not the same issue." + }, + { + "left": "issue:43819", + "right": "issue:44493", + "accept": false, + "reason": "Different symptoms and code paths: DAC latent mismatch vs position-id key warnings." + }, + { + "left": "issue:44206", + "right": "issue:45823", + "accept": false, + "reason": "Different models and failures: feature-extractor crash vs Gemma4 PLE device mismatch." + }, + { + "left": "issue:29942", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention test failures and Trainer `_gpu_broadcast_one` TypeError are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:43504", + "accept": false, + "reason": "Different model families and bugs; only broad ML/model-loading similarity." + }, + { + "left": "issue:45588", + "right": "issue:45859", + "accept": false, + "reason": "Flash-attention sinkless-model crash and Qwen rotary embedding CPU-offload incompatibility are separate issues." + }, + { + "left": "issue:37428", + "right": "issue:44351", + "accept": false, + "reason": "Two import errors, but for different symbols and different causes." + }, + { + "left": "issue:43278", + "right": "issue:43825", + "accept": false, + "reason": "Training/eval dtype drift is unrelated to pipeline v5 error messaging." + }, + { + "left": "issue:45459", + "right": "issue:46234", + "accept": false, + "reason": "Tokenizer protobuf-error masking and missing import_utils availability helper are distinct API issues." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "Docs-example errors are unrelated to Beit fast processor label reduction." + }, + { + "left": "issue:43388", + "right": "issue:43526", + "accept": false, + "reason": "Both affect labels, but one is metric gathering and the other is image processor label reduction." + }, + { + "left": "issue:43335", + "right": "issue:45083", + "accept": false, + "reason": "No concrete overlap beyond both being internal model helpers." + }, + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Same reported bug: identical Qwen3.5 MoE `from_pretrained` failure title, with 45478 looking like a later duplicate report." + }, + { + "left": "issue:39692", + "right": "issue:44493", + "accept": false, + "reason": "Documentation example errors are unrelated to key/position-id loading warnings." + }, + { + "left": "issue:28218", + "right": "issue:35141", + "accept": false, + "reason": "Tokenizer spacing bug and embedding reinitialization after resize are different issues." + }, + { + "left": "issue:43720", + "right": "issue:44112", + "accept": false, + "reason": "Packed-weight loading in BitNet and a GraniteSpeech stale-device CI test are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:44112", + "accept": false, + "reason": "Different model/config bug versus CI test flakiness; not mergeable as one fix." + }, + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation and D-FINE auxiliary-loss omission are separate model bugs." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The duplicate call on issues 45310/45478 is plausible from the identical titles and ordering, and the soft-edge rejections are generally conservative. However, the canonical_issue_reason overstates the packet evidence by claiming 45310 has the \"more informative title/body\"\u2014the packet only provides titles and body lengths, not the body content or relative informativeness. Please keep the justification limited to what is explicitly supported (identical title, later duplicate-like report, inbound reference on 45310)." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 26, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:25251", + "issue:29127", + "issue:29942", + "issue:32090", + "issue:34689", + "issue:35141", + "issue:38175", + "issue:40444", + "issue:42490", + "issue:43278", + "issue:43381", + "issue:43412", + "issue:43526", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43611", + "issue:43653", + "issue:43701", + "issue:43746", + "issue:43854", + "issue:43927", + "issue:44360", + "issue:44448", + "issue:44466", + "issue:44493", + "issue:44561", + "issue:44568", + "issue:44743", + "issue:44779", + "issue:44857", + "issue:44977", + "issue:45072", + "issue:45083", + "issue:45198", + "issue:45229", + "issue:45245", + "issue:45372", + "issue:45491", + "issue:45563", + "issue:45588", + "issue:45698", + "issue:45803", + "issue:45823", + "issue:45920", + "issue:46192", + "issue:46209", + "issue:46285" + ], + "soft_pairs": [ + "issue:44448|issue:44779", + "issue:42490|issue:43854", + "issue:43606|issue:44857", + "issue:45920|issue:46192", + "issue:44360|issue:44561", + "issue:38175|issue:44493", + "issue:44977|issue:45229", + "issue:44857|issue:45072", + "issue:44448|issue:44977", + "issue:34689|issue:35141", + "issue:43701|issue:45083", + "issue:29127|issue:32090", + "issue:43550|issue:45083", + "issue:43927|issue:45698", + "issue:40444|issue:43526", + "issue:29942|issue:35141", + "issue:42490|issue:43575", + "issue:44561|issue:44779", + "issue:44977|issue:45563", + "issue:44857|issue:45588", + "issue:43746|issue:46209", + "issue:44568|issue:45491", + "issue:43412|issue:45588", + "issue:45372|issue:45588", + "issue:43278|issue:43381", + "issue:45198|issue:45803", + "issue:45588|issue:45823", + "issue:43611|issue:46285", + "issue:44448|issue:44466", + "issue:45245|issue:45563", + "issue:25251|issue:32090", + "issue:43653|issue:44743" + ], + "prepared_review_unit_hash": "dcfebee9fe332132c88ca179358d6d98a2439e4ed0bf5806e257ec047ed923b1", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37506, + "estimated_input_tokens": 9377, + "estimated_eval_tokens": 19010 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are largely unrelated, spanning tokenizer regressions, model-loading/saving issues, device/dtype mismatches, export failures, and model-specific runtime bugs. None of the soft pairs look like the same underlying bug or change, so all soft edges should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": "Issue 45920 is the best representative because it is the broadest, most explicit v5 tokenizer regression umbrella and covers multiple affected models, making it the strongest canonical issue for a tokenizer-bug cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45920 is the strongest standalone issue: it is specific enough to be actionable, but broad enough to subsume related tokenizer reports across several models.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44448", + "right": "issue:44779", + "accept": false, + "reason": "Both are v5 output regressions, but one is Pegasus output drift and the other is a DeepSeek tokenizer error; different code paths and symptoms." + }, + { + "left": "issue:42490", + "right": "issue:43854", + "accept": false, + "reason": "One is a local-path import warning with trailing slash, the other is a model-loading failure for GLM-4.7; not the same defect." + }, + { + "left": "issue:43606", + "right": "issue:44857", + "accept": false, + "reason": "CPU offload device mismatch in bark-small is unrelated to an AMP float16 crash in LwDetrImageLoss." + }, + { + "left": "issue:45920", + "right": "issue:46192", + "accept": false, + "reason": "Tokenizer wrong IDs across multiple models is different from return_tensors being ignored when text_kwargs is passed." + }, + { + "left": "issue:44360", + "right": "issue:44561", + "accept": false, + "reason": "DSA indexer ReLU discussion has no overlap with the removal of is_torch_fx_available breaking trust_remote_code models." + }, + { + "left": "issue:38175", + "right": "issue:44493", + "accept": false, + "reason": "Siglip2 zero probabilities and unexpected position-id keys are different model-specific inference bugs." + }, + { + "left": "issue:44977", + "right": "issue:45229", + "accept": false, + "reason": "Qwen3.5 flash-attention generation failure and Gemma4 multi-GPU OOM are unrelated failures." + }, + { + "left": "issue:44857", + "right": "issue:45072", + "accept": false, + "reason": "Both mention dtype-related problems, but one is an AMP crash in a loss module and the other is a bfloat16 mismatch in different models." + }, + { + "left": "issue:44448", + "right": "issue:44977", + "accept": false, + "reason": "Pegasus v4/v5 output differences and Qwen3.5 flash-attention failures affect different models and code paths." + }, + { + "left": "issue:34689", + "right": "issue:35141", + "accept": false, + "reason": "Model loading breakage for Llama 3.2 vision is unrelated to output embeddings being reinitialized after resize when weights are untied." + }, + { + "left": "issue:43701", + "right": "issue:45083", + "accept": false, + "reason": "Checkpoint key mismatch and qwen3_omni_moe feature-length helper behavior are distinct issues." + }, + { + "left": "issue:29127", + "right": "issue:32090", + "accept": false, + "reason": "LayoutLMv3 error-message clarity and Trainer _gpu_broadcast_one NoneType failure are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:45083", + "accept": false, + "reason": "Bamba torch.compile/SDPA failure is not the same underlying bug as a qwen3_omni_moe output-length helper issue." + }, + { + "left": "issue:43927", + "right": "issue:45698", + "accept": false, + "reason": "Both involve save/load behavior, but one loses custom token IDs and the other loads the wrong custom module; different metadata paths." + }, + { + "left": "issue:40444", + "right": "issue:43526", + "accept": false, + "reason": "IterableDataset with multiple images per prompt and BeitImageProcessorFast reduce_labels are different preprocessing bugs." + }, + { + "left": "issue:29942", + "right": "issue:35141", + "accept": false, + "reason": "Flash Attention 2 test failures are unrelated to embedding reinitialization during post_init." + }, + { + "left": "issue:42490", + "right": "issue:43575", + "accept": false, + "reason": "Trailing-slash local-model warning is unrelated to Qwen2-57B TP out-of-memory." + }, + { + "left": "issue:44561", + "right": "issue:44779", + "accept": false, + "reason": "A missing helper in v5 trust_remote_code support is not the same bug as DeepSeek tokenizer regression." + }, + { + "left": "issue:44977", + "right": "issue:45563", + "accept": false, + "reason": "Flash-attention generation failure and a stale warning for num_return_sequences are different generate-path issues." + }, + { + "left": "issue:44857", + "right": "issue:45588", + "accept": false, + "reason": "Float16 AMP crash in LwDetrImageLoss and flash_attention.py crashing on s_aux=None are separate defects." + }, + { + "left": "issue:43746", + "right": "issue:46209", + "accept": false, + "reason": "PEFT checkpoint loading for GraniteSpeech and FP8 inference failure on B200 are unrelated." + }, + { + "left": "issue:44568", + "right": "issue:45491", + "accept": false, + "reason": "Tokenizer add_special_tokens behavior and Gemma3 NaN embeddings with sliding-window attention are not the same bug." + }, + { + "left": "issue:43412", + "right": "issue:45588", + "accept": false, + "reason": "Executorch export guard/support issues are unrelated to flash_attention.py AttributeError on sink-less models." + }, + { + "left": "issue:45372", + "right": "issue:45588", + "accept": false, + "reason": "Gemma4 processor import failure and flash_attention.py s_aux=None crash are different failures." + }, + { + "left": "issue:43278", + "right": "issue:43381", + "accept": false, + "reason": "Embedding dtype drift in eval and gradient checkpointing being unusable in eval mode are distinct." + }, + { + "left": "issue:45198", + "right": "issue:45803", + "accept": false, + "reason": "Wav2Vec2 save/tokenization failure and Fuyu bare-except swallowing errors are unrelated." + }, + { + "left": "issue:45588", + "right": "issue:45823", + "accept": false, + "reason": "Flash-attention sink-less crash and Gemma4 PLE device mismatch are different runtime bugs." + }, + { + "left": "issue:43611", + "right": "issue:46285", + "accept": false, + "reason": "base_model_prefix loading breakage and CLIPTextModel structural inconsistency are both v5 model-structure issues but not the same underlying defect." + }, + { + "left": "issue:44448", + "right": "issue:44466", + "accept": false, + "reason": "Pegasus v4/v5 output mismatch and lm_head.weight serialization varying by device are separate regression classes." + }, + { + "left": "issue:45245", + "right": "issue:45563", + "accept": false, + "reason": "Category-cardinality limit error and stale generate warning are unrelated." + }, + { + "left": "issue:25251", + "right": "issue:32090", + "accept": false, + "reason": "Pipeline top_k nesting change and Trainer _gpu_broadcast_one NoneType error affect different subsystems and code paths." + }, + { + "left": "issue:43653", + "right": "issue:44743", + "accept": false, + "reason": "BigBirdTokenizer special-token decoding issue and qwen3_5 recurrent-state reset bug are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 27, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:10000", + "issue:15354", + "issue:29127", + "issue:29942", + "issue:30064", + "issue:32090", + "issue:33357", + "issue:34567", + "issue:36010", + "issue:37428", + "issue:39692", + "issue:41720", + "issue:42490", + "issue:42915", + "issue:42994", + "issue:43065", + "issue:43066", + "issue:43278", + "issue:43381", + "issue:43421", + "issue:43575", + "issue:43611", + "issue:43653", + "issue:43701", + "issue:43873", + "issue:43976", + "issue:43994", + "issue:44186", + "issue:44206", + "issue:44265", + "issue:44361", + "issue:44442", + "issue:44462", + "issue:44479", + "issue:44610", + "issue:44734", + "issue:44871", + "issue:44977", + "issue:45198", + "issue:45229", + "issue:45372", + "issue:45406", + "issue:45412", + "issue:45440", + "issue:45459", + "issue:45491", + "issue:45712", + "issue:46121" + ], + "soft_pairs": [ + "issue:36010|issue:37428", + "issue:43065|issue:45412", + "issue:45459|issue:45712", + "issue:43611|issue:46121", + "issue:44186|issue:45198", + "issue:44361|issue:45198", + "issue:29127|issue:34567", + "issue:33357|issue:39692", + "issue:44871|issue:44977", + "issue:45372|issue:45459", + "issue:44206|issue:44479", + "issue:42915|issue:42994", + "issue:43994|issue:45412", + "issue:29942|issue:30064", + "issue:43976|issue:45440", + "issue:43278|issue:43575", + "issue:43066|issue:44462", + "issue:44265|issue:45198", + "issue:10000|issue:15354", + "issue:42490|issue:43873", + "issue:43873|issue:45229", + "issue:43065|issue:43994", + "issue:44734|issue:45406", + "issue:43381|issue:43701", + "issue:43421|issue:45491", + "issue:43653|issue:44442", + "issue:44610|issue:45412", + "issue:41720|issue:45229", + "issue:32090|issue:33357" + ], + "prepared_review_unit_hash": "129375939ff4ce0b3e72ee0d2e5ebe18320f88e0a1b45157669aa27c3144aec0", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36454, + "estimated_input_tokens": 9114, + "estimated_eval_tokens": 18484 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is not a real duplicate set: it mixes unrelated issues across tokenizers, generation, training, serving, quantization, and model-specific regressions. All soft edges look like false positives.", + "confidence": 0.99, + "canonical_issue_reason": "No safe canonical issue: the items span many unrelated code paths and would not collapse into one duplicate cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "No single issue is a good global representative; the cluster is too heterogeneous to use one issue as the canonical target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36010", + "right": "issue:37428", + "accept": false, + "reason": "Different import failures in different modules: GenerationMixin vs flash-attention utility." + }, + { + "left": "issue:43065", + "right": "issue:45412", + "accept": false, + "reason": "Sam3PixelDecoder dummy layer vs RT-DETR memory leak; unrelated bugs." + }, + { + "left": "issue:45459", + "right": "issue:45712", + "accept": false, + "reason": "Tokenizer error masking with missing protobuf vs leftover dummy classes; no shared bug." + }, + { + "left": "issue:43611", + "right": "issue:46121", + "accept": false, + "reason": "base_model_prefix loading regression vs rope validation TypeError; different code paths." + }, + { + "left": "issue:44186", + "right": "issue:45198", + "accept": false, + "reason": "LayoutLMv2 NER/padding crash vs Wav2Vec2 save/tokenization failure; different tokenizer behavior." + }, + { + "left": "issue:44361", + "right": "issue:45198", + "accept": false, + "reason": "MLukeTokenizer AttributeError is unrelated to Wav2Vec2 tokenization/save_pretrained issues." + }, + { + "left": "issue:29127", + "right": "issue:34567", + "accept": false, + "reason": "LayoutLMv3 error messaging vs TrainerState token counter update; unrelated training/model issues." + }, + { + "left": "issue:33357", + "right": "issue:39692", + "accept": false, + "reason": "MacOS CLIP bus error vs SigLIP2 documentation/example errors; different failure modes." + }, + { + "left": "issue:44871", + "right": "issue:44977", + "accept": false, + "reason": "Gemma-3 eos_token mismatch vs Qwen3.5 flash-attention generation bug; different model/config paths." + }, + { + "left": "issue:45372", + "right": "issue:45459", + "accept": false, + "reason": "Gemma4 processor import dependency issue vs protobuf decode-error handling; unrelated." + }, + { + "left": "issue:44206", + "right": "issue:44479", + "accept": false, + "reason": "LasrFeatureExtractor crash vs video-input regression for Qwen VL models; not the same bug." + }, + { + "left": "issue:42915", + "right": "issue:42994", + "accept": false, + "reason": "Qwen3Moe FP8 config failure vs quantized model saving failure; separate quantization problems." + }, + { + "left": "issue:43994", + "right": "issue:45412", + "accept": false, + "reason": "SigLIP2 bad outputs vs RT-DETR memory leak; unrelated subsystems." + }, + { + "left": "issue:29942", + "right": "issue:30064", + "accept": false, + "reason": "Flash Attention 2 test failures vs void segmentation-map processing; different regressions." + }, + { + "left": "issue:43976", + "right": "issue:45440", + "accept": false, + "reason": "Python version compatibility issue vs DeepseekV3MoE divergence; unrelated." + }, + { + "left": "issue:43278", + "right": "issue:43575", + "accept": false, + "reason": "Embedding dtype drift vs tensor-parallel OOM; different bugs." + }, + { + "left": "issue:43066", + "right": "issue:44462", + "accept": false, + "reason": "Tokenizer decoder type regression vs AutoTokenizer ignoring tokenizer.json; both tokenizer-related but distinct." + }, + { + "left": "issue:44265", + "right": "issue:45198", + "accept": false, + "reason": "torch.export/torch_compilable_check failure vs Wav2Vec2 tokenization/save failure; unrelated." + }, + { + "left": "issue:10000", + "right": "issue:15354", + "accept": false, + "reason": "DistilBertModel issue vs GeneratorExp unsupported in torch.jit.script; unrelated model/export bugs." + }, + { + "left": "issue:42490", + "right": "issue:43873", + "accept": false, + "reason": "Trailing-slash local model warning vs quantization offloading regression; different behavior." + }, + { + "left": "issue:43873", + "right": "issue:45229", + "accept": false, + "reason": "Offloading-with-quantization problem vs multi-GPU CUDA OOM; both memory-related but not the same bug." + }, + { + "left": "issue:43065", + "right": "issue:43994", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder vs SigLIP2 inference issues; unrelated." + }, + { + "left": "issue:44734", + "right": "issue:45406", + "accept": false, + "reason": "KV-cache continuation indexing crash vs Gemma4Processor missing _tokenizer; different code paths." + }, + { + "left": "issue:43381", + "right": "issue:43701", + "accept": false, + "reason": "Eval-mode checkpointing restriction vs resume_from_checkpoint key mismatch; unrelated training-state bugs." + }, + { + "left": "issue:43421", + "right": "issue:45491", + "accept": false, + "reason": "Runtime special-token update feature vs Gemma3 NaN embeddings; not the same issue." + }, + { + "left": "issue:43653", + "right": "issue:44442", + "accept": false, + "reason": "BigBirdTokenizer special-token registration vs AutoTokenizer loading FastSpeech2ConformerTokenizer; distinct tokenizer problems." + }, + { + "left": "issue:44610", + "right": "issue:45412", + "accept": false, + "reason": "Processor/model size mismatch vs RT-DETR memory leak; unrelated." + }, + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "Qwen3 auto device-mapping assert vs Gemma4 multi-GPU OOM; different hardware/runtime failures." + }, + { + "left": "issue:32090", + "right": "issue:33357", + "accept": false, + "reason": "Trainer NoneType broadcast error vs MacOS bus error with CLIP; unrelated issues." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 28, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:29127", + "issue:29942", + "issue:33357", + "issue:35141", + "issue:36010", + "issue:38175", + "issue:42490", + "issue:42915", + "issue:43262", + "issue:43278", + "issue:43329", + "issue:43334", + "issue:43381", + "issue:43452", + "issue:43526", + "issue:43531", + "issue:43575", + "issue:43646", + "issue:43701", + "issue:43723", + "issue:43756", + "issue:43761", + "issue:43824", + "issue:43931", + "issue:44062", + "issue:44077", + "issue:44186", + "issue:44206", + "issue:44279", + "issue:44387", + "issue:44442", + "issue:44554", + "issue:44743", + "issue:44811", + "issue:44857", + "issue:44977", + "issue:45072", + "issue:45103", + "issue:45245", + "issue:45372", + "issue:45406", + "issue:45440", + "issue:45584", + "issue:45588", + "issue:45593", + "issue:45834", + "issue:45859", + "issue:45907" + ], + "soft_pairs": [ + "issue:43756|issue:45440", + "issue:43278|issue:43329", + "issue:45372|issue:45406", + "issue:43526|issue:45834", + "issue:29942|issue:36010", + "issue:43723|issue:44554", + "issue:43761|issue:45859", + "issue:45440|issue:45593", + "issue:45103|issue:45372", + "issue:43526|issue:44186", + "issue:43575|issue:43701", + "issue:44977|issue:45245", + "issue:43526|issue:44857", + "issue:43575|issue:44387", + "issue:42490|issue:43381", + "issue:43334|issue:43531", + "issue:43824|issue:44279", + "issue:44442|issue:45072", + "issue:43646|issue:44077", + "issue:29127|issue:35141", + "issue:42915|issue:43931", + "issue:44743|issue:44811", + "issue:43262|issue:44206", + "issue:43452|issue:45588", + "issue:44977|issue:45584", + "issue:33357|issue:38175", + "issue:44062|issue:45907" + ], + "prepared_review_unit_hash": "d01e2f2fe6f73195c83f191b188be16c4931549874503038603d48aadd4a0573", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35862, + "estimated_input_tokens": 8966, + "estimated_eval_tokens": 18188 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The items do not form a single duplicate cluster; they span many unrelated bugs and regressions across loading, tokenization, vision, audio, attention, and training paths. All soft-edge matches look like topic-level similarity only, not the same underlying defect.", + "confidence": 0.98, + "canonical_issue_reason": "No clear canonical issue: the set is heterogeneous, and the apparent matches are false positives rather than one underlying bug.", + "canonical_pr_reason": null, + "best_issue_reason": "No single issue is a strong cluster anchor; at best, issue 43646 is the broadest/generalized regression report, but it still does not represent the rest of the items.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43756", + "right": "issue:45440", + "accept": false, + "reason": "Both mention model internals, but one is about Smollm3 RoPE layer handling and the other is DeepSeekV3MoE implementation divergence; different code paths and fixes." + }, + { + "left": "issue:43278", + "right": "issue:43329", + "accept": false, + "reason": "Different bugs: BF16->FP32 embedding dtype shift in eval vs undefined video-token helper/merge_size in multimodal token counting." + }, + { + "left": "issue:45372", + "right": "issue:45406", + "accept": false, + "reason": "Same Gemma 4 area, but one is a mistral_common import/dependency failure and the other is a missing _tokenizer attribute in serve; not the same defect." + }, + { + "left": "issue:43526", + "right": "issue:45834", + "accept": false, + "reason": "BeitImageProcessorFast reduce_labels bug vs Kosmos2.5 long OCR index error; unrelated processor paths." + }, + { + "left": "issue:29942", + "right": "issue:36010", + "accept": false, + "reason": "Flash Attention 2 test failures and GenerationMixin import errors are unrelated regression types." + }, + { + "left": "issue:43723", + "right": "issue:44554", + "accept": false, + "reason": "Tokenizer loading in v5 vs an MPS attention correctness issue; different subsystems and failure modes." + }, + { + "left": "issue:43761", + "right": "issue:45859", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression vs Qwen3_MoeTextRotaryEmbedding CPU-offload incompatibility; not the same code path." + }, + { + "left": "issue:45440", + "right": "issue:45593", + "accept": false, + "reason": "DeepSeekV3MoE divergence and D-FINE auxiliary-loss behavior are unrelated model-specific bugs." + }, + { + "left": "issue:45103", + "right": "issue:45372", + "accept": false, + "reason": "auto_docstring annotation parsing crash vs Gemma 4 ReasoningEffort import breakage; unrelated." + }, + { + "left": "issue:43526", + "right": "issue:44186", + "accept": false, + "reason": "Both are tokenizer/image-processing issues, but BEiT reduce_labels and LayoutLMv2 NER/padding crashes are different implementations." + }, + { + "left": "issue:43575", + "right": "issue:43701", + "accept": false, + "reason": "TP loading OOM and resume_from_checkpoint key mismatch are different training/runtime problems." + }, + { + "left": "issue:44977", + "right": "issue:45245", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug vs category-count limit in another pipeline; no shared underlying bug." + }, + { + "left": "issue:43526", + "right": "issue:44857", + "accept": false, + "reason": "BEiT label reduction and LwDetrImageLoss float16 CUDA crash are unrelated." + }, + { + "left": "issue:43575", + "right": "issue:44387", + "accept": false, + "reason": "Both involve memory pressure, but one is tensor-parallel loading OOM and the other is increased reserved memory under int4 quantization; different root causes." + }, + { + "left": "issue:42490", + "right": "issue:43381", + "accept": false, + "reason": "Local-path import warning vs gradient checkpointing in eval mode; unrelated." + }, + { + "left": "issue:43334", + "right": "issue:43531", + "accept": false, + "reason": "Qwen3-VL pad_token_id loading error and Qwen3-MoE sliding_window behavior are different model/configuration bugs." + }, + { + "left": "issue:43824", + "right": "issue:44279", + "accept": false, + "reason": "Qwen2_5_VL import failure vs a generic dependency issue; too broad to be the same concrete bug." + }, + { + "left": "issue:44442", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer load failure and BF16 dtype mismatch in inference are unrelated." + }, + { + "left": "issue:43646", + "right": "issue:44077", + "accept": false, + "reason": "General Transformers 5 initialization breakage and patchtsmixer optional post_init restriction are related only at a broad API level, not the same fixable defect." + }, + { + "left": "issue:29127", + "right": "issue:35141", + "accept": false, + "reason": "LayoutLMv3 error messaging and token-embedding reinitialization in post_init are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:43931", + "accept": false, + "reason": "Qwen3Moe FineGrainedFP8Config failure and Qwen3-VL weight-shape mismatch are different loading/compatibility issues." + }, + { + "left": "issue:44743", + "right": "issue:44811", + "accept": false, + "reason": "Qwen3_5 recurrent-state cache bug vs Whisper batch_decode skip_special_tokens handling; unrelated model families and code paths." + }, + { + "left": "issue:43262", + "right": "issue:44206", + "accept": false, + "reason": "Audio chat-template sample-rate default bug vs LasrFeatureExtractor unsupported center arg; different processors and fixes." + }, + { + "left": "issue:43452", + "right": "issue:45588", + "accept": false, + "reason": "gguf_file loading breakage and flash_attention sink-less s_aux=None crash are unrelated." + }, + { + "left": "issue:44977", + "right": "issue:45584", + "accept": false, + "reason": "Qwen3.5 flash-attention generation failure vs Whisper empty-transcription/alignment failure; different models and root causes." + }, + { + "left": "issue:33357", + "right": "issue:38175", + "accept": false, + "reason": "Both are model inference issues, but MacOS bus error on community CLIP and zero probabilities on siglip2-base are not the same underlying bug." + }, + { + "left": "issue:44062", + "right": "issue:45907", + "accept": false, + "reason": "AddedToken special-argument duplication and list_repo_files revision forwarding are unrelated utility/API bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 29, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:28218", + "issue:32090", + "issue:33290", + "issue:34567", + "issue:34689", + "issue:36010", + "issue:42175", + "issue:42907", + "issue:43316", + "issue:43381", + "issue:43493", + "issue:43525", + "issue:43550", + "issue:43611", + "issue:43643", + "issue:43650", + "issue:43720", + "issue:43761", + "issue:43827", + "issue:43873", + "issue:43931", + "issue:44077", + "issue:44188", + "issue:44190", + "issue:44442", + "issue:44464", + "issue:44479", + "issue:44496", + "issue:44568", + "issue:44743", + "issue:44855", + "issue:44871", + "issue:44898", + "issue:44913", + "issue:44964", + "issue:45137", + "issue:45216", + "issue:45290", + "issue:45440", + "issue:45479", + "issue:45491", + "issue:45698", + "issue:45759", + "issue:45803", + "issue:46202", + "issue:46209", + "issue:46234" + ], + "soft_pairs": [ + "issue:28218|issue:32090", + "issue:44464|issue:45491", + "issue:33290|issue:34567", + "issue:44442|issue:44898", + "issue:43381|issue:43873", + "issue:44964|issue:45440", + "issue:43316|issue:43525", + "issue:34689|issue:43611", + "issue:32090|issue:33290", + "issue:43720|issue:44442", + "issue:44479|issue:45290", + "issue:43931|issue:45216", + "issue:34689|issue:36010", + "issue:44442|issue:45803", + "issue:43550|issue:44743", + "issue:43493|issue:43827", + "issue:42907|issue:45698", + "issue:42175|issue:46234", + "issue:43873|issue:44871", + "issue:43643|issue:44913", + "issue:43761|issue:44188", + "issue:43316|issue:45759", + "issue:45290|issue:46209", + "issue:33290|issue:34689", + "issue:44190|issue:44496", + "issue:44568|issue:45479", + "issue:44077|issue:44479", + "issue:44855|issue:45137", + "issue:43650|issue:46202" + ], + "prepared_review_unit_hash": "791e08a582ef2b611b2bd7469a92c4b6104f140f69d32d6c4d8491c375290867", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 36053, + "estimated_input_tokens": 9014, + "estimated_eval_tokens": 18284 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are mostly unrelated Transformers issues that were linked only by loose textual similarity. No soft edge looks like the same underlying bug/change, so none should be merged as duplicates.", + "confidence": 0.91, + "canonical_issue_reason": "If a representative issue is needed, issue 45698 is the broadest and most active loading/saving report here, with multiple references and discussion around from_pretrained/save_pretrained behavior.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45698 is the strongest standalone triage candidate because it is open, has higher discussion/inbound refs, and concerns a general model-loading regression rather than a model-specific edge case.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:28218", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer added-token spacing is unrelated to Trainer's _gpu_broadcast_one NoneType error." + }, + { + "left": "issue:44464", + "right": "issue:45491", + "accept": false, + "reason": "Chunked generation with compiled forward is a different failure mode from Gemma3 mixed-length NaN embeddings." + }, + { + "left": "issue:33290", + "right": "issue:34567", + "accept": false, + "reason": "Deepspeed Adafactor OOM and num_input_tokens_seen not updating are distinct trainer/optimizer bugs." + }, + { + "left": "issue:44442", + "right": "issue:44898", + "accept": false, + "reason": "Tokenizer loading failure and Perceiver image-classification resolution failure are unrelated code paths." + }, + { + "left": "issue:43381", + "right": "issue:43873", + "accept": false, + "reason": "Gradient checkpointing eval-mode behavior does not match quantization offloading issues." + }, + { + "left": "issue:44964", + "right": "issue:45440", + "accept": false, + "reason": "Phi-4 multimodal loading regression and DeepseekV3MoE implementation divergence are different model-specific bugs." + }, + { + "left": "issue:43316", + "right": "issue:43525", + "accept": false, + "reason": "Gemma3TextConfig API mismatch is not the same as Llama4Config missing pad_token_id." + }, + { + "left": "issue:34689", + "right": "issue:43611", + "accept": false, + "reason": "Llama 3.2 Vision loading failure and base_model_prefix loading regression are separate model-loading problems." + }, + { + "left": "issue:32090", + "right": "issue:33290", + "accept": false, + "reason": "Trainer GPU broadcast type error and Adafactor/deepspeed OOM are different training failures." + }, + { + "left": "issue:43720", + "right": "issue:44442", + "accept": false, + "reason": "BitNet packed-weight loading bug is unrelated to FastSpeech2ConformerTokenizer auto-loading." + }, + { + "left": "issue:44479", + "right": "issue:45290", + "accept": false, + "reason": "Video-input regression for Qwen variants is not the same as chat-template crashes on tool-call assistant messages." + }, + { + "left": "issue:43931", + "right": "issue:45216", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch and Qwen3.5 save_pretrained regression are different load/save failures." + }, + { + "left": "issue:34689", + "right": "issue:36010", + "accept": false, + "reason": "Llama 3.2 Vision loading and GenerationMixin import failure are unrelated import/config issues." + }, + { + "left": "issue:44442", + "right": "issue:45803", + "accept": false, + "reason": "Tokenizer loading and Fuyu batch tensor conversion error are different subsystems and symptoms." + }, + { + "left": "issue:43550", + "right": "issue:44743", + "accept": false, + "reason": "Bamba torch.compile/SDPA failure is unrelated to Qwen3.5 recurrent-state cache reset." + }, + { + "left": "issue:43493", + "right": "issue:43827", + "accept": false, + "reason": "SigLIP2 implementation divergence is not the same as docs still referencing removed pipeline()." + }, + { + "left": "issue:42907", + "right": "issue:45698", + "accept": false, + "reason": "Dequantized save failure for Ministral/Devstral is unrelated to wrong custom-module loading after save_pretrained." + }, + { + "left": "issue:42175", + "right": "issue:46234", + "accept": false, + "reason": "TensorFlow missing from a torch extra install and a missing import_utils symbol are different packaging/API issues." + }, + { + "left": "issue:43873", + "right": "issue:44871", + "accept": false, + "reason": "Quantization offloading behavior is unrelated to Gemma-3 eos_token_id config inconsistency." + }, + { + "left": "issue:43643", + "right": "issue:44913", + "accept": false, + "reason": "trust_remote_code missing fields and GPTNeoX rotary_pct reload fallback are different config-loading bugs." + }, + { + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "CLIPVision hidden_states regression is unrelated to attention kernel divergence under torch.compile." + }, + { + "left": "issue:43316", + "right": "issue:45759", + "accept": false, + "reason": "Gemma3TextConfig API discrepancy and Qwen multimodal text_config unwrapping are different config-shape issues." + }, + { + "left": "issue:45290", + "right": "issue:46209", + "accept": false, + "reason": "Chat template tool-call crash and FP8 inference failure on B200 are unrelated failure modes." + }, + { + "left": "issue:33290", + "right": "issue:34689", + "accept": false, + "reason": "Deepspeed Adafactor OOM and Llama 3.2 Vision loading error are not the same bug." + }, + { + "left": "issue:44190", + "right": "issue:44496", + "accept": false, + "reason": "Local dataset loading in a training script and unrecognized model/config.json errors are unrelated." + }, + { + "left": "issue:44568", + "right": "issue:45479", + "accept": false, + "reason": "mDeBERTa special-token regression and sequence-classification zero-loss bug affect different model/task paths." + }, + { + "left": "issue:44077", + "right": "issue:44479", + "accept": false, + "reason": "patchtsmixer post_init policy and Qwen video-input regression are unrelated model-specific changes." + }, + { + "left": "issue:44855", + "right": "issue:45137", + "accept": false, + "reason": "TorchScript import IndentationError and DeepSpeed ZeRO3 deque error are completely different runtime failures." + }, + { + "left": "issue:43650", + "right": "issue:46202", + "accept": false, + "reason": "Vague data request and Gemma pipeline roundtrip failure on \u2581 are not the same issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 30, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:25251", + "issue:28282", + "issue:29127", + "issue:33290", + "issue:33357", + "issue:35141", + "issue:41950", + "issue:42617", + "issue:43065", + "issue:43232", + "issue:43295", + "issue:43316", + "issue:43388", + "issue:43404", + "issue:43441", + "issue:43452", + "issue:43475", + "issue:43504", + "issue:43540", + "issue:43550", + "issue:43572", + "issue:43575", + "issue:43606", + "issue:43646", + "issue:43701", + "issue:43716", + "issue:43742", + "issue:43761", + "issue:43825", + "issue:44162", + "issue:44479", + "issue:44496", + "issue:44514", + "issue:44617", + "issue:44792", + "issue:44871", + "issue:44877", + "issue:44913", + "issue:44987", + "issue:45127", + "issue:45200", + "issue:45584", + "issue:45656", + "issue:45698", + "issue:45803", + "issue:46050", + "issue:46170", + "issue:46178" + ], + "soft_pairs": [ + "issue:29127|issue:33357", + "issue:44987|issue:46178", + "issue:43295|issue:43825", + "issue:25251|issue:28282", + "issue:43550|issue:43701", + "issue:42617|issue:43575", + "issue:43606|issue:43701", + "issue:43540|issue:44514", + "issue:33290|issue:35141", + "issue:43575|issue:43742", + "issue:33290|issue:45656", + "issue:43316|issue:43716", + "issue:43404|issue:43441", + "issue:44162|issue:44792", + "issue:44871|issue:45584", + "issue:43316|issue:44877", + "issue:43295|issue:44871", + "issue:45803|issue:46170", + "issue:43065|issue:44617", + "issue:41950|issue:43475", + "issue:43646|issue:44913", + "issue:44871|issue:46050", + "issue:45127|issue:45200", + "issue:43452|issue:44496", + "issue:43761|issue:44479", + "issue:43232|issue:43388", + "issue:43504|issue:45698", + "issue:43572|issue:46178" + ], + "prepared_review_unit_hash": "5920f4656bb8c5b20e6462e925a8c543f3fa35ff8bbd324fd5d59451a2456851", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36078, + "estimated_input_tokens": 9020, + "estimated_eval_tokens": 18296 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No true duplicate cluster emerges here: the soft edges span unrelated bugs across loading, pipelines, configs, multimodal processors, and training/runtime issues. I would anchor the cluster on the broad from_pretrained/save_pretrained regression, but I accept no soft merges.", + "confidence": 0.88, + "canonical_issue_reason": "issue:45698 is the broadest, most representative user-facing loading/serialization regression in the set, with the most discussion and inbound references.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45698 best represents the cluster because it is a general from_pretrained/save_pretrained correctness bug with wide practical impact.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:29127", + "right": "issue:33357", + "accept": false, + "reason": "Different bugs: LayoutLMv3 error clarity vs CLIP bus error on macOS." + }, + { + "left": "issue:44987", + "right": "issue:46178", + "accept": false, + "reason": "Different failure modes: model loading regression vs unsupported backend mapping for tensorflow_text." + }, + { + "left": "issue:43295", + "right": "issue:43825", + "accept": false, + "reason": "Different code paths: processor/tokenizer regression vs pipeline translation-task messaging." + }, + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Unrelated: pipeline top_k nesting bug vs missing PyTorch ImportError." + }, + { + "left": "issue:43550", + "right": "issue:43701", + "accept": false, + "reason": "Different problems: torch.compile/SDPA failure vs checkpoint key mismatch." + }, + { + "left": "issue:42617", + "right": "issue:43575", + "accept": false, + "reason": "Different load/runtime issues: 3d_parallel.py failure vs tensor-parallel OOM." + }, + { + "left": "issue:43606", + "right": "issue:43701", + "accept": false, + "reason": "Different bugs: CPU offload device mismatch vs resume_from_checkpoint key mismatch." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Both multimodal, but different paths: Qwen3OmniMoe video input processing vs batched chat template padding crash." + }, + { + "left": "issue:33290", + "right": "issue:35141", + "accept": false, + "reason": "Different areas: DeepSpeed/Adafactor OOM vs embedding reinitialization after resize." + }, + { + "left": "issue:43575", + "right": "issue:43742", + "accept": false, + "reason": "Different load failures: huge-model OOM vs missing key for MobileLLM." + }, + { + "left": "issue:33290", + "right": "issue:45656", + "accept": false, + "reason": "Both involve DeepSpeed, but one is OOM while the other is duplicate optimizer stepping." + }, + { + "left": "issue:43316", + "right": "issue:43716", + "accept": false, + "reason": "Different issues: config API discrepancy vs image preprocessor/model dtype mismatch." + }, + { + "left": "issue:43404", + "right": "issue:43441", + "accept": false, + "reason": "Different model bugs: lm_head tying vs FlashAttention failure." + }, + { + "left": "issue:44162", + "right": "issue:44792", + "accept": false, + "reason": "Unrelated models and symptoms: ESM2 breakage vs janus image-generation test failure." + }, + { + "left": "issue:44871", + "right": "issue:45584", + "accept": false, + "reason": "Different token/generation bugs: Gemma eos_token config mismatch vs Whisper empty transcription regression." + }, + { + "left": "issue:43316", + "right": "issue:44877", + "accept": false, + "reason": "Different config issues: Gemma3TextConfig API mismatch vs granite_speech strict config load." + }, + { + "left": "issue:43295", + "right": "issue:44871", + "accept": false, + "reason": "Different regressions: custom processor/tokenizer handling vs Gemma eos_token_id inconsistency." + }, + { + "left": "issue:45803", + "right": "issue:46170", + "accept": false, + "reason": "Different concerns: swallowed exceptions in Fuyu tensor conversion vs Windows ROCm torch.distributed dependency." + }, + { + "left": "issue:43065", + "right": "issue:44617", + "accept": false, + "reason": "Different Sam3 issues: dummy Conv2d placeholder vs CUDA OOM." + }, + { + "left": "issue:41950", + "right": "issue:43475", + "accept": false, + "reason": "Different multimodal bugs: pipeline processor selection vs missing fpn_position_embeddings attribute." + }, + { + "left": "issue:43646", + "right": "issue:44913", + "accept": false, + "reason": "Both are init/config-related, but the model-init breakage and rotary_pct reload bug are distinct." + }, + { + "left": "issue:44871", + "right": "issue:46050", + "accept": false, + "reason": "Different serialization bugs: eos_token config mismatch vs quantized model save/load breakage." + }, + { + "left": "issue:45127", + "right": "issue:45200", + "accept": false, + "reason": "Different training/model-behavior issues: LoRA merge collapse vs missing mm_token_type_ids default." + }, + { + "left": "issue:43452", + "right": "issue:44496", + "accept": false, + "reason": "Different loading failures: gguf handling vs missing model_type in config.json." + }, + { + "left": "issue:43761", + "right": "issue:44479", + "accept": false, + "reason": "Different regressions in different models: CLIPVision hidden_states vs Qwen VL video input handling." + }, + { + "left": "issue:43232", + "right": "issue:43388", + "accept": false, + "reason": "Different utilities and symptoms: generation kwargs update vs metrics gathering dropping labels." + }, + { + "left": "issue:43504", + "right": "issue:45698", + "accept": false, + "reason": "Both are loading-related, but legacy preset-field handling and wrong custom-module loading are different root causes." + }, + { + "left": "issue:43572", + "right": "issue:46178", + "accept": false, + "reason": "Unrelated: missing StableLm pad_token_idx vs unsupported tensorflow_text backend mapping." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 31, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:35141", + "issue:41720", + "issue:42490", + "issue:43065", + "issue:43116", + "issue:43335", + "issue:43540", + "issue:43550", + "issue:43644", + "issue:43673", + "issue:43761", + "issue:43827", + "issue:43931", + "issue:43976", + "issue:43994", + "issue:44060", + "issue:44077", + "issue:44112", + "issue:44190", + "issue:44242", + "issue:44462", + "issue:44479", + "issue:44521", + "issue:44534", + "issue:44610", + "issue:44617", + "issue:44734", + "issue:44737", + "issue:44743", + "issue:44792", + "issue:44855", + "issue:44877", + "issue:44913", + "issue:44964", + "issue:44977", + "issue:45127", + "issue:45245", + "issue:45357", + "issue:45381", + "issue:45542", + "issue:45584", + "issue:45588", + "issue:45600", + "issue:45632", + "issue:45803", + "issue:45907", + "issue:45987", + "issue:46234" + ], + "soft_pairs": [ + "issue:44077|issue:44534", + "issue:45542|issue:46234", + "issue:43673|issue:44743", + "issue:43761|issue:44242", + "issue:44792|issue:44977", + "issue:44617|issue:45584", + "issue:44534|issue:44913", + "issue:44190|issue:44964", + "issue:44792|issue:45600", + "issue:44479|issue:44734", + "issue:43827|issue:43994", + "issue:44855|issue:45588", + "issue:41720|issue:45245", + "issue:43065|issue:44610", + "issue:43931|issue:44060", + "issue:45803|issue:45987", + "issue:35141|issue:42490", + "issue:43644|issue:44913", + "issue:43976|issue:44855", + "issue:43335|issue:44855", + "issue:43550|issue:44190", + "issue:45632|issue:46234", + "issue:44077|issue:44877", + "issue:43540|issue:44734", + "issue:45127|issue:45357", + "issue:44462|issue:45907", + "issue:44521|issue:45381", + "issue:43335|issue:44190", + "issue:44737|issue:45803", + "issue:44112|issue:45600", + "issue:43116|issue:44190" + ], + "prepared_review_unit_hash": "7b8e91d8e58f17f7729042704219caa70bb7ca6bf5bbd6f295048922775f2c2d", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37102, + "estimated_input_tokens": 9276, + "estimated_eval_tokens": 18808 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The items are not a duplicate cluster; they span many unrelated bugs/regressions across different models, configs, and utilities. All soft-edge pairs are false positives on broad vocabulary overlap rather than the same underlying change or bug.", + "confidence": 0.98, + "canonical_issue_reason": "No single canonical issue fits this set because the issues are semantically unrelated; there is no clear duplicate-center artifact to anchor the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "No global best issue is appropriate for a duplicate cluster here. If one had to be picked as a representative, it would be a self-contained regression like issue:44534, but it does not represent the rest of the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44077", + "right": "issue:44534", + "accept": false, + "reason": "Both mention v5 behavior changes, but one is about optional post_init in patchtsmixer and the other about non-persistent buffers being junk; different code paths and failures." + }, + { + "left": "issue:45542", + "right": "issue:46234", + "accept": false, + "reason": "Completely different areas: TensorBoard/TensorFlow backend installation vs a missing import utility for Falcon Mamba SSM." + }, + { + "left": "issue:43673", + "right": "issue:44743", + "accept": false, + "reason": "Both involve cache/state handling, but one is a GenerationMixin chunked-prefill cache bug and the other is Qwen3.5 recurrent state reset logic; not the same defect." + }, + { + "left": "issue:43761", + "right": "issue:44242", + "accept": false, + "reason": "Unrelated model forward regression in CLIPVisionModel vs MoE load-balancing loss omission." + }, + { + "left": "issue:44792", + "right": "issue:44977", + "accept": false, + "reason": "Different models and symptoms: a janus image-generation test failure vs Qwen3.5 flash-attention generation issues." + }, + { + "left": "issue:44617", + "right": "issue:45584", + "accept": false, + "reason": "CUDA OOM in Sam3Video is not the same as Whisper empty-transcription failure after align_special_tokens." + }, + { + "left": "issue:44534", + "right": "issue:44913", + "accept": false, + "reason": "One is a v5 non-persistent buffer initialization regression; the other is GPTNeoXConfig rotary_pct not persisting on reload." + }, + { + "left": "issue:44190", + "right": "issue:44964", + "accept": false, + "reason": "Dataset loading in an example script is unrelated to loading a specific multimodal model." + }, + { + "left": "issue:44792", + "right": "issue:45600", + "accept": false, + "reason": "A failed janus generation test and stale auto_mappings references to removed configs are different problems." + }, + { + "left": "issue:44479", + "right": "issue:44734", + "accept": false, + "reason": "Different subsystems: video-input regression for Qwen VL models vs serve/KV-cache tensor indexing." + }, + { + "left": "issue:43827", + "right": "issue:43994", + "accept": false, + "reason": "Docs still referencing pipeline() is not the same as a broken pipeline inference result for SigLIP2." + }, + { + "left": "issue:44855", + "right": "issue:45588", + "accept": false, + "reason": "Python 3.13 import parsing error in DebertaV2 is unrelated to flash_attention.py crashing on s_aux=None." + }, + { + "left": "issue:41720", + "right": "issue:45245", + "accept": false, + "reason": "Qwen3 auto device mapping CUDA assert and a categories-limit runtime error are unrelated." + }, + { + "left": "issue:43065", + "right": "issue:44610", + "accept": false, + "reason": "A dummy Conv2d in Sam3PixelDecoder is not the same as OmDet-Turbo processor/model input-size mismatch." + }, + { + "left": "issue:43931", + "right": "issue:44060", + "accept": false, + "reason": "Weight-shape mismatch while loading Qwen3-VL is unrelated to an incorrect tied-weights warning in Qwen3-Next." + }, + { + "left": "issue:45803", + "right": "issue:45987", + "accept": false, + "reason": "A broad exception swallowing KeyboardInterrupt in FuyuBatchFeature is unrelated to StaticCache.get_seq_length type mismatch." + }, + { + "left": "issue:35141", + "right": "issue:42490", + "accept": false, + "reason": "Token embedding resizing/post_init reinit bug has nothing to do with local model path trailing-slash import warnings." + }, + { + "left": "issue:43644", + "right": "issue:44913", + "accept": false, + "reason": "Non-persistent buffer junk initialization and GPTNeoX rotary_pct reload persistence are different regressions." + }, + { + "left": "issue:43976", + "right": "issue:44855", + "accept": false, + "reason": "Python-version support breakage is not the same as an IndentationError from a TorchScript-decorated function comment." + }, + { + "left": "issue:43335", + "right": "issue:44855", + "accept": false, + "reason": "SwitchTransformers sparse-layer config bug is unrelated to DebertaV2 import parsing on Python 3.13." + }, + { + "left": "issue:43550", + "right": "issue:44190", + "accept": false, + "reason": "Bamba torch.compile/SDPA failure and local dataset loading are entirely different code paths." + }, + { + "left": "issue:45632", + "right": "issue:46234", + "accept": false, + "reason": "A trust_remote_code cache-path collision is not the same as a missing import_utils symbol." + }, + { + "left": "issue:44077", + "right": "issue:44877", + "accept": false, + "reason": "Optional post_init policy for patchtsmixer and strict config rejection of granite_speech are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44734", + "accept": false, + "reason": "Video input validation in Qwen3OmniMoe does not match a KV-cache continuation tensor-indexing crash in serve." + }, + { + "left": "issue:45127", + "right": "issue:45357", + "accept": false, + "reason": "LoRA merge with extended vocabulary and incorrect visual-encoder save_pretrained keys are different model-state bugs." + }, + { + "left": "issue:44462", + "right": "issue:45907", + "accept": false, + "reason": "AutoTokenizer ignoring tokenizer.json is unrelated to list_repo_files failing to forward revision." + }, + { + "left": "issue:44521", + "right": "issue:45381", + "accept": false, + "reason": "All-zero assistant_masks for multimodal inputs and wrong vision_position_ids for qwen2.5-vl video inputs are not the same bug." + }, + { + "left": "issue:43335", + "right": "issue:44190", + "accept": false, + "reason": "A SwitchTransformers config edge case is unrelated to a local dataset loading failure in an example script." + }, + { + "left": "issue:44737", + "right": "issue:45803", + "accept": false, + "reason": "CPU device placement bug in XLNet positional encoding is unrelated to a bare-except error swallowing KeyboardInterrupt." + }, + { + "left": "issue:44112", + "right": "issue:45600", + "accept": false, + "reason": "CI stale test failure and auto_mappings referencing removed configs are different maintenance issues." + }, + { + "left": "issue:43116", + "right": "issue:44190", + "accept": false, + "reason": "Empty multi-label results in an image-classification example are not the same as local dataset loading problems." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 32, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:36246", + "issue:38175", + "issue:38617", + "issue:41628", + "issue:42907", + "issue:43065", + "issue:43295", + "issue:43388", + "issue:43404", + "issue:43441", + "issue:43479", + "issue:43526", + "issue:43531", + "issue:43643", + "issue:43650", + "issue:43688", + "issue:43854", + "issue:43901", + "issue:44112", + "issue:44186", + "issue:44220", + "issue:44246", + "issue:44464", + "issue:44496", + "issue:44561", + "issue:44683", + "issue:44734", + "issue:44933", + "issue:44938", + "issue:44964", + "issue:45003", + "issue:45042", + "issue:45216", + "issue:45245", + "issue:45276", + "issue:45335", + "issue:45362", + "issue:45375", + "issue:45446", + "issue:45542", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45600", + "issue:45632", + "issue:46082", + "issue:46121" + ], + "soft_pairs": [ + "issue:46082|issue:46121", + "issue:43479|issue:45375", + "issue:44938|issue:45542", + "issue:45276|issue:45584", + "issue:43295|issue:45276", + "issue:45245|issue:45584", + "issue:44464|issue:45563", + "issue:43650|issue:44220", + "issue:44186|issue:44734", + "issue:36246|issue:41628", + "issue:43404|issue:45216", + "issue:44246|issue:45003", + "issue:44683|issue:45446", + "issue:38175|issue:43065", + "issue:44933|issue:45362", + "issue:43643|issue:44496", + "issue:43854|issue:44964", + "issue:44964|issue:45600", + "issue:44112|issue:44964", + "issue:45563|issue:45584", + "issue:36246|issue:38617", + "issue:44734|issue:45632", + "issue:44561|issue:45632", + "issue:42907|issue:43441", + "issue:43531|issue:45600", + "issue:43526|issue:43901", + "issue:43295|issue:45335", + "issue:43441|issue:45588", + "issue:44496|issue:45042", + "issue:43388|issue:43688" + ], + "prepared_review_unit_hash": "d4c0fba6fb770d0493b52b9fb911d491930ebd85db585f23a365c6e8d1087afc", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36137, + "estimated_input_tokens": 9035, + "estimated_eval_tokens": 18326 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the pairs share only broad lexical similarity (imports, loading, config, generation, embeddings, CI) but not the same concrete bug. No soft-edge pair is safe to merge as a duplicate or same PR/change.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue is a good fit; the items span many unrelated regressions across different subsystems, so choosing one as the representative would be misleading.", + "canonical_pr_reason": "No PRs are present in this cluster.", + "best_issue_reason": "No issue is a strong global representative for duplicate triage. If forced to pick a nearby theme, issue 44964 is the most user-facing model-loading regression, but it still does not unify the rest of the cluster.", + "best_pr_reason": "No PRs are present in this cluster.", + "soft_edge_verdicts": [ + { + "left": "issue:46082", + "right": "issue:46121", + "accept": false, + "reason": "Different bugs: one is LlamaConfig validation around explicit head_dim, the other is rope-validation handling of a JSON-loaded list." + }, + { + "left": "issue:43479", + "right": "issue:45375", + "accept": false, + "reason": "Both are multimodal config issues, but they affect different config classes and failure modes; not the same underlying bug." + }, + { + "left": "issue:44938", + "right": "issue:45542", + "accept": false, + "reason": "Both mention backend/loading failures, but one is a Python 3.14 import/load issue and the other is a TensorBoard-without-TF backend error." + }, + { + "left": "issue:45276", + "right": "issue:45584", + "accept": false, + "reason": "Completely different code paths: Gemma resize_token_embeddings vs Whisper empty-transcription generation." + }, + { + "left": "issue:43295", + "right": "issue:45276", + "accept": false, + "reason": "Both involve embedding/tokenizer behavior, but one is a regression in processor/tokenizer access and the other is model embedding resizing; not the same bug." + }, + { + "left": "issue:45245", + "right": "issue:45584", + "accept": false, + "reason": "Unrelated issues: category-count limits versus Whisper generation on empty transcription." + }, + { + "left": "issue:44464", + "right": "issue:45563", + "accept": false, + "reason": "Both are generation-related, but one is inconsistent compiled forward outputs and the other is a stale warning in paged generate; different defects." + }, + { + "left": "issue:43650", + "right": "issue:44220", + "accept": false, + "reason": "The titles do not describe the same problem; one is a vague data request, the other is an fbank feature extraction bug." + }, + { + "left": "issue:44186", + "right": "issue:44734", + "accept": false, + "reason": "Different failures in different paths: LayoutLMv2 tokenizer padding/truncation vs transformers serve KV-cache continuation indexing." + }, + { + "left": "issue:36246", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but for different missing symbols in different modules; not the same underlying breakage." + }, + { + "left": "issue:43404", + "right": "issue:45216", + "accept": false, + "reason": "Different bugs: lm_head weight tying in Mistral3 vs Qwen3.5 save_pretrained checkpoint correctness." + }, + { + "left": "issue:44246", + "right": "issue:45003", + "accept": false, + "reason": "Both touch import/runtime behavior, but one is slow import performance and the other is unsafe sys.modules access." + }, + { + "left": "issue:44683", + "right": "issue:45446", + "accept": false, + "reason": "Both involve flex_attention, but one is compiled flex_attention on newer torch and the other is a version-check bug for AuxRequest import; not the same fix." + }, + { + "left": "issue:38175", + "right": "issue:43065", + "accept": false, + "reason": "Completely different model behaviors: zero probabilities in SigLIP2 vs dummy Conv2d in Sam3PixelDecoder." + }, + { + "left": "issue:44933", + "right": "issue:45362", + "accept": false, + "reason": "One is an image_utils import reference issue, the other is a Qwen3.5-35B chat crash; no shared concrete code-path bug." + }, + { + "left": "issue:43643", + "right": "issue:44496", + "accept": false, + "reason": "Both are config/loading related, but one is trust_remote_code field loss and the other is an unrecognized model/config.json model_type error." + }, + { + "left": "issue:43854", + "right": "issue:44964", + "accept": false, + "reason": "Both are model loading failures, but they concern different models and different causes; not mergeable as one bug." + }, + { + "left": "issue:44964", + "right": "issue:45600", + "accept": false, + "reason": "Different failure modes: Phi-4 multimodal loading versus stale auto_mappings references to removed Sam3LiteText configs." + }, + { + "left": "issue:44112", + "right": "issue:44964", + "accept": false, + "reason": "CI test flakiness in GraniteSpeech is unrelated to loading Phi-4 multimodal with latest transformers." + }, + { + "left": "issue:45563", + "right": "issue:45584", + "accept": false, + "reason": "Stale warning in paged generate is not the same as Whisper failing on empty transcription after align_special_tokens." + }, + { + "left": "issue:36246", + "right": "issue:38617", + "accept": false, + "reason": "Both are ImportError reports, but they reference different symbols and modules; not the same missing export." + }, + { + "left": "issue:44734", + "right": "issue:45632", + "accept": false, + "reason": "Different issues: server KV-cache tensor indexing versus trust_remote_code cache-path collisions." + }, + { + "left": "issue:44561", + "right": "issue:45632", + "accept": false, + "reason": "Both involve trust_remote_code, but one is a removed utility breaking remote-code models and the other is a local cache-path collision; distinct bugs." + }, + { + "left": "issue:42907", + "right": "issue:43441", + "accept": false, + "reason": "Ministral/Devstral save-dequantization and FlashAttention failures are unrelated." + }, + { + "left": "issue:43531", + "right": "issue:45600", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior is unrelated to Sam3LiteText config references in auto_mappings." + }, + { + "left": "issue:43526", + "right": "issue:43901", + "accept": false, + "reason": "BeitImageProcessorFast label reduction and TextClassificationPipeline docs mismatch are unrelated." + }, + { + "left": "issue:43295", + "right": "issue:45335", + "accept": false, + "reason": "Both mention embedding/tokenizer effects, but they target different models and different embedding tensors." + }, + { + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "Different flash-attention-related problems in different models/code paths; not the same underlying bug." + }, + { + "left": "issue:44496", + "right": "issue:45042", + "accept": false, + "reason": "Unrelated model/config loading error versus image processor torchvision dependency regression." + }, + { + "left": "issue:43388", + "right": "issue:43688", + "accept": false, + "reason": "Different training/evaluation bugs: gather_for_metrics label truncation vs auxiliary-loss normalization." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 33, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:32090", + "issue:35141", + "issue:38175", + "issue:39692", + "issue:41950", + "issue:42175", + "issue:43295", + "issue:43299", + "issue:43381", + "issue:43479", + "issue:43525", + "issue:43577", + "issue:43650", + "issue:43673", + "issue:43824", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44062", + "issue:44075", + "issue:44220", + "issue:44242", + "issue:44279", + "issue:44336", + "issue:44351", + "issue:44368", + "issue:44496", + "issue:44617", + "issue:44683", + "issue:44779", + "issue:44933", + "issue:45020", + "issue:45200", + "issue:45216", + "issue:45278", + "issue:45335", + "issue:45357", + "issue:45372", + "issue:45405", + "issue:45542", + "issue:45584", + "issue:45593", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:46209", + "issue:46234" + ], + "soft_pairs": [ + "issue:43673|issue:44336", + "issue:41950|issue:44016", + "issue:44075|issue:44368", + "issue:43299|issue:45216", + "issue:42175|issue:44279", + "issue:43381|issue:45593", + "issue:43901|issue:44496", + "issue:44016|issue:44220", + "issue:44496|issue:45759", + "issue:43650|issue:46209", + "issue:45372|issue:46234", + "issue:45335|issue:45584", + "issue:32090|issue:35141", + "issue:43479|issue:43525", + "issue:44242|issue:45200", + "issue:43577|issue:45216", + "issue:43299|issue:45357", + "issue:38175|issue:44617", + "issue:43824|issue:44933", + "issue:44779|issue:45812", + "issue:44279|issue:45405", + "issue:39692|issue:43901", + "issue:44351|issue:45278", + "issue:41950|issue:45020", + "issue:43295|issue:44779", + "issue:44016|issue:44062", + "issue:41950|issue:45823", + "issue:43299|issue:43854", + "issue:43976|issue:45405", + "issue:44683|issue:45542" + ], + "prepared_review_unit_hash": "ace6640e99a7322675b5eae934b072f5f5a0bccb48e55653fe42712d156ad88d", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36648, + "estimated_input_tokens": 9162, + "estimated_eval_tokens": 18580 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose mix of recent Transformers regressions, import/load failures, and unrelated model/docs bugs. The soft pairs do not look like true duplicates of the same concrete code-path issue, so none should be merged.", + "confidence": 0.31, + "canonical_issue_reason": "issue:45278 is the broadest and most representative user-facing regression in the set: an upgrade-triggered import breakage affecting many symbols. It fits the dominant recent-version breakage theme better than the narrower model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the best triage anchor because it states a concrete, version-scoped failure mode and has the clearest broad scope among the issues listed.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43673", + "right": "issue:44336", + "accept": false, + "reason": "One is a GenerationMixin cache regression; the other is ANSI formatting in loading_report. Different subsystems and failure modes." + }, + { + "left": "issue:41950", + "right": "issue:44016", + "accept": false, + "reason": "A pipeline processor-selection bug is unrelated to a notebook syntax error." + }, + { + "left": "issue:44075", + "right": "issue:44368", + "accept": false, + "reason": "SGD optimizer arguments being ignored is unrelated to a tie_word_embeddings warning." + }, + { + "left": "issue:43299", + "right": "issue:45216", + "accept": false, + "reason": "Qwen3VL model-loading regression and Qwen3.5 save_pretrained checkpoint corruption are different concrete bugs." + }, + { + "left": "issue:42175", + "right": "issue:44279", + "accept": false, + "reason": "A missing TensorFlow backend install and a generic dependency issue are not the same underlying problem." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Gradient checkpointing in eval mode and D-FINE auxiliary-loss behavior are unrelated." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "A documentation mismatch in TextClassificationPipeline is unrelated to an unrecognized model/config loading error." + }, + { + "left": "issue:44016", + "right": "issue:44220", + "accept": false, + "reason": "Notebook syntax error and _torch_extract_fbank_features() failure do not share the same bug." + }, + { + "left": "issue:44496", + "right": "issue:45759", + "accept": false, + "reason": "Unrecognized model/config loading and composite multimodal config unwrapping are different code paths." + }, + { + "left": "issue:43650", + "right": "issue:46209", + "accept": false, + "reason": "A placeholder request is unrelated to an FP8 inference failure on B200." + }, + { + "left": "issue:45372", + "right": "issue:46234", + "accept": false, + "reason": "Gemma processor loading import failure and a missing falcon_mamba utility export are separate import bugs." + }, + { + "left": "issue:45335", + "right": "issue:45584", + "accept": false, + "reason": "Tokenizer/embedding resize behavior is unrelated to Whisper empty-transcription generation after token alignment." + }, + { + "left": "issue:32090", + "right": "issue:35141", + "accept": false, + "reason": "Trainer _gpu_broadcast_one NoneType error and embedding reinitialization after resize are different failures." + }, + { + "left": "issue:43479", + "right": "issue:43525", + "accept": false, + "reason": "Default multimodal config initialization and missing pad_token_id are unrelated config issues." + }, + { + "left": "issue:44242", + "right": "issue:45200", + "accept": false, + "reason": "Missing load-balancing loss and mm_token_type_ids defaulting are different modeling/training bugs." + }, + { + "left": "issue:43577", + "right": "issue:45216", + "accept": false, + "reason": "Dtype loading for Blip2 and incorrect Qwen3.5 save_pretrained output are unrelated regressions." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Both are Qwen3.5/Qwen3VL-related, but one is loading failure and the other is save_pretrained key serialization; not the same bug." + }, + { + "left": "issue:38175", + "right": "issue:44617", + "accept": false, + "reason": "Zero probabilities in SigLIP2 and CUDA OOM in Sam3Video are completely different issues." + }, + { + "left": "issue:43824", + "right": "issue:44933", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols and different modules." + }, + { + "left": "issue:44779", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but for different model families and likely different code paths." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "A dependency installation problem is not the same as bumping MIN_PEFT_VERSION to an unreleased version." + }, + { + "left": "issue:39692", + "right": "issue:43901", + "accept": false, + "reason": "Separate documentation problems in different sections with different concrete defects." + }, + { + "left": "issue:44351", + "right": "issue:45278", + "accept": false, + "reason": "A missing HybridCache export may be related to broader import breakage, but the evidence is too weak to treat it as the same concrete bug as the multi-import regression." + }, + { + "left": "issue:41950", + "right": "issue:45020", + "accept": false, + "reason": "A video-classification processor lookup bug is not the same as a broad remote_code compatibility regression." + }, + { + "left": "issue:43295", + "right": "issue:44779", + "accept": false, + "reason": "Processor/tokenizer API regression and Deepseek tokenization errors are different model-specific bugs." + }, + { + "left": "issue:44016", + "right": "issue:44062", + "accept": false, + "reason": "Notebook syntax error and AddedToken constructor argument conflict are unrelated." + }, + { + "left": "issue:41950", + "right": "issue:45823", + "accept": false, + "reason": "Video-classification processor lookup and Gemma4 PLE device mismatch are unrelated." + }, + { + "left": "issue:43299", + "right": "issue:43854", + "accept": false, + "reason": "Qwen3VL MoE loading and GLM-4.7-Flash unit-test loading are different model-specific failures." + }, + { + "left": "issue:43976", + "right": "issue:45405", + "accept": false, + "reason": "Python version compatibility and an unreleased PEFT version requirement are unrelated." + }, + { + "left": "issue:44683", + "right": "issue:45542", + "accept": false, + "reason": "Compiled flex_attention on torch>=2.9 and TensorBoard-without-TF backend errors have no shared underlying bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 34, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:42994", + "issue:43065", + "issue:43317", + "issue:43441", + "issue:43450", + "issue:43525", + "issue:43643", + "issue:43650", + "issue:43653", + "issue:43723", + "issue:43784", + "issue:43881", + "issue:43901", + "issue:43976", + "issue:43994", + "issue:44016", + "issue:44162", + "issue:44246", + "issue:44261", + "issue:44263", + "issue:44279", + "issue:44291", + "issue:44297", + "issue:44361", + "issue:44462", + "issue:44466", + "issue:44521", + "issue:44617", + "issue:44623", + "issue:44671", + "issue:44683", + "issue:44749", + "issue:44779", + "issue:44898", + "issue:44987", + "issue:45081", + "issue:45276", + "issue:45325", + "issue:45397", + "issue:45542", + "issue:45593", + "issue:45600", + "issue:45712", + "issue:45823", + "issue:45834", + "issue:45907", + "issue:46132" + ], + "soft_pairs": [ + "issue:43650|issue:46132", + "issue:44462|issue:45081", + "issue:43976|issue:44987", + "issue:44297|issue:44462", + "issue:43976|issue:44279", + "issue:43450|issue:44279", + "issue:45081|issue:45907", + "issue:44521|issue:45325", + "issue:44466|issue:45276", + "issue:43441|issue:45834", + "issue:42994|issue:43317", + "issue:43653|issue:45593", + "issue:45600|issue:45712", + "issue:44623|issue:44749", + "issue:44623|issue:45397", + "issue:43065|issue:44898", + "issue:44261|issue:44263", + "issue:44361|issue:45593", + "issue:44671|issue:44779", + "issue:44016|issue:45823", + "issue:43441|issue:44683", + "issue:43723|issue:44162", + "issue:43901|issue:43994", + "issue:44297|issue:45397", + "issue:43525|issue:43881", + "issue:43450|issue:44617", + "issue:43901|issue:44246", + "issue:43525|issue:43643", + "issue:43784|issue:45542", + "issue:44016|issue:44291" + ], + "prepared_review_unit_hash": "5620ea6855bfa79c5cab3185d15daaceabdf6310635dc4a615e570d1d39f600a", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 35943, + "estimated_input_tokens": 8986, + "estimated_eval_tokens": 18228 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a heterogeneous cluster of mostly unrelated closed issues across tokenizer loading/saving, model-specific regressions, multimodal preprocessing, config/serialization bugs, and CI/docs noise. None of the soft pairs look like true duplicates of the same concrete bug or change.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue for the full cluster; the only plausible anchor is issue 44462, but the rest of the items span too many unrelated failure modes to treat as duplicates of it.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44462 is the broadest tokenizer-loading report in the set and the closest thing to a representative issue, but it still does not cover the cluster as a whole.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43650", + "right": "issue:46132", + "accept": false, + "reason": "Placeholder/low-signal issue vs an AttentionInterface.register behavior change; unrelated artifacts." + }, + { + "left": "issue:44462", + "right": "issue:45081", + "accept": false, + "reason": "Both involve tokenizers, but one is AutoTokenizer ignoring tokenizer.json while the other is a Mistral regex patch crash; different code paths and bugs." + }, + { + "left": "issue:43976", + "right": "issue:44987", + "accept": false, + "reason": "Python version compatibility problem vs a model loading failure; no concrete shared bug." + }, + { + "left": "issue:44297", + "right": "issue:44462", + "accept": false, + "reason": "Tokenizer save_pretrained class metadata mismatch vs AutoTokenizer ignoring tokenizer.json; adjacent area, but distinct failures." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Version-support issue vs vague dependency issue; not the same underlying bug." + }, + { + "left": "issue:43450", + "right": "issue:44279", + "accept": false, + "reason": "Video processor batch-shape bug vs dependency issue; unrelated." + }, + { + "left": "issue:45081", + "right": "issue:45907", + "accept": false, + "reason": "Mistral tokenizer regex crash vs list_repo_files revision forwarding; completely different functionality." + }, + { + "left": "issue:44521", + "right": "issue:45325", + "accept": false, + "reason": "Both multimodal, but assistant mask generation and temporal rope index scaling are different code paths." + }, + { + "left": "issue:44466", + "right": "issue:45276", + "accept": false, + "reason": "Both touch embeddings/weights, but serialization of lm_head.weight and resize_token_embeddings updating modules are different bugs." + }, + { + "left": "issue:43441", + "right": "issue:45834", + "accept": false, + "reason": "FlashAttention loading failure vs OCR long-input index error; unrelated model-specific issues." + }, + { + "left": "issue:42994", + "right": "issue:43317", + "accept": false, + "reason": "Quantized saving vs dequantized/offload loading; same broad area, but not the same concrete defect." + }, + { + "left": "issue:43653", + "right": "issue:45593", + "accept": false, + "reason": "Tokenizer special-token registration bug vs D-FINE auxiliary-loss behavior; unrelated." + }, + { + "left": "issue:45600", + "right": "issue:45712", + "accept": false, + "reason": "Both are repo hygiene/auto-generated-file issues, but they affect different files and failures." + }, + { + "left": "issue:44623", + "right": "issue:44749", + "accept": false, + "reason": "Missing files in processor.save_pretrained vs performance regression during data filtering; unrelated." + }, + { + "left": "issue:44623", + "right": "issue:45397", + "accept": false, + "reason": "Processor save bug vs Gemma4 zero3 loading bug; different code paths." + }, + { + "left": "issue:43065", + "right": "issue:44898", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder vs Perceiver interpolation failure; different models and errors." + }, + { + "left": "issue:44261", + "right": "issue:44263", + "accept": false, + "reason": "Missing rms_norm_eps causing precision drift vs torch.split return-value handling; unrelated math bugs." + }, + { + "left": "issue:44361", + "right": "issue:45593", + "accept": false, + "reason": "MLukeTokenizer AttributeError vs D-FINE auxiliary-loss issue; no shared underlying defect." + }, + { + "left": "issue:44671", + "right": "issue:44779", + "accept": false, + "reason": "Both are v5 regressions, but one is CamemBERT masked-LM output corruption and the other is Deepseek tokenizer output corruption; not the same bug." + }, + { + "left": "issue:44016", + "right": "issue:45823", + "accept": false, + "reason": "Notebook syntax error vs Gemma4 device mismatch; unrelated." + }, + { + "left": "issue:43441", + "right": "issue:44683", + "accept": false, + "reason": "FlashAttention model failure vs compiled flex_attention failure on torch>=2.9; different implementations and causes." + }, + { + "left": "issue:43723", + "right": "issue:44162", + "accept": false, + "reason": "Tokenizer loading regression vs ESM2 broken; both loading-related, but not the same failure mode." + }, + { + "left": "issue:43901", + "right": "issue:43994", + "accept": false, + "reason": "Docs mismatch for return_all_scores vs bad model outputs in pipeline; different problem class." + }, + { + "left": "issue:44297", + "right": "issue:45397", + "accept": false, + "reason": "Tokenizer serialization metadata bug vs Gemma4 zero3 loading bug; unrelated." + }, + { + "left": "issue:43525", + "right": "issue:43881", + "accept": false, + "reason": "Missing pad_token_id in Llama4Config vs glm-4v-9b loading failure; different model/config issues." + }, + { + "left": "issue:43450", + "right": "issue:44617", + "accept": false, + "reason": "Incorrect batched video processor shape vs CUDA OOM in Sam3Video; not the same bug." + }, + { + "left": "issue:43901", + "right": "issue:44246", + "accept": false, + "reason": "Documentation mismatch vs intermittent import slowness; unrelated." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": false, + "reason": "Specific missing config field vs trust_remote_code returning incomplete config fields; related theme, but not the same concrete issue." + }, + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "Sentence-transformers import NameError vs missing TensorFlow backend handling; different optional-dependency paths." + }, + { + "left": "issue:44016", + "right": "issue:44291", + "accept": false, + "reason": "Notebook syntax error vs init_empty_weights TypeError from unexpected argument; unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 35, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:16998", + "issue:22355", + "issue:36246", + "issue:41628", + "issue:41950", + "issue:42491", + "issue:42757", + "issue:42886", + "issue:43262", + "issue:43299", + "issue:43352", + "issue:43526", + "issue:43673", + "issue:43761", + "issue:43824", + "issue:43901", + "issue:43976", + "issue:44162", + "issue:44188", + "issue:44220", + "issue:44242", + "issue:44246", + "issue:44295", + "issue:44336", + "issue:44351", + "issue:44373", + "issue:44485", + "issue:44496", + "issue:44521", + "issue:44561", + "issue:44589", + "issue:44749", + "issue:44779", + "issue:44857", + "issue:44871", + "issue:44964", + "issue:45003", + "issue:45081", + "issue:45200", + "issue:45250", + "issue:45292", + "issue:45362", + "issue:45397", + "issue:45412", + "issue:45593", + "issue:46123", + "issue:46132" + ], + "soft_pairs": [ + "issue:44188|issue:44336", + "issue:43262|issue:44521", + "issue:41628|issue:43976", + "issue:44589|issue:45362", + "issue:43673|issue:43761", + "issue:16998|issue:22355", + "issue:44336|issue:44373", + "issue:42491|issue:43299", + "issue:44246|issue:44496", + "issue:44295|issue:45397", + "issue:44242|issue:45593", + "issue:44336|issue:44485", + "issue:44246|issue:44871", + "issue:43824|issue:43976", + "issue:41950|issue:44220", + "issue:43976|issue:44964", + "issue:42886|issue:46132", + "issue:36246|issue:42757", + "issue:36246|issue:44351", + "issue:44779|issue:45292", + "issue:44242|issue:44857", + "issue:45081|issue:45362", + "issue:43673|issue:44561", + "issue:44295|issue:44749", + "issue:43352|issue:45250", + "issue:44162|issue:45200", + "issue:43901|issue:45003", + "issue:43526|issue:45003", + "issue:44496|issue:45003", + "issue:45412|issue:46123" + ], + "prepared_review_unit_hash": "780239c227bd63f1c777810292e1030f9b2a859a733a5cc6a83f8459de6eecd2", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36013, + "estimated_input_tokens": 9004, + "estimated_eval_tokens": 18264 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous, with mostly unrelated regression/import/model-loading issues that only loosely share Transformers-v5 breakage themes. None of the soft pairs looks like the same underlying bug or change.", + "confidence": 0.42, + "canonical_issue_reason": "Issue 44561 is the broadest core v5 regression here: a low-level API removal that breaks trust_remote_code models and best matches the cluster\u2019s general breakage/import theme.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44561 is the best representative issue because it is a concrete, high-impact v5 compatibility regression affecting a wide class of models, making it the most central exemplar in this set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "Attention-kernel divergence on torch.compile is unrelated to ANSI escape codes in loading_report." + }, + { + "left": "issue:43262", + "right": "issue:44521", + "accept": false, + "reason": "Audio apply_chat_template sampling-rate default bug is a different code path from multimodal assistant_mask generation." + }, + { + "left": "issue:41628", + "right": "issue:43976", + "accept": false, + "reason": "Missing AutoImageProcessor import and Python-version incompatibility are different failures with different causes." + }, + { + "left": "issue:44589", + "right": "issue:45362", + "accept": false, + "reason": "Float8 storage lookup error is unrelated to Qwen3.5 chat crashes." + }, + { + "left": "issue:43673", + "right": "issue:43761", + "accept": false, + "reason": "Chunked-prefill cache handling and CLIPVisionModel hidden_states regression are distinct bugs in different subsystems." + }, + { + "left": "issue:16998", + "right": "issue:22355", + "accept": false, + "reason": "A model_max_length question for DeBERTa-V3 is unrelated to a missing transformers.onnx module import." + }, + { + "left": "issue:44336", + "right": "issue:44373", + "accept": false, + "reason": "Loading-report ANSI codes and a wrong docstring for position_ids are not the same bug." + }, + { + "left": "issue:42491", + "right": "issue:43299", + "accept": false, + "reason": "Both concern Qwen MoE loading, but they target different model generations and different failure modes." + }, + { + "left": "issue:44246", + "right": "issue:44496", + "accept": false, + "reason": "Import latency sometimes and unrecognized model config are unrelated." + }, + { + "left": "issue:44295", + "right": "issue:45397", + "accept": false, + "reason": "Buffer access on position_ids is unrelated to Gemma-4 zero3 from_pretrained behavior." + }, + { + "left": "issue:44242", + "right": "issue:45593", + "accept": false, + "reason": "Missing load-balancing loss and missing auxiliary losses are different model-specific training bugs." + }, + { + "left": "issue:44336", + "right": "issue:44485", + "accept": false, + "reason": "ANSI code output and GLM-5 RoPE implementation are unrelated." + }, + { + "left": "issue:44246", + "right": "issue:44871", + "accept": false, + "reason": "Import-time slowness is unrelated to Gemma eos_token_id configuration mismatch." + }, + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "Qwen2_5_VL symbol export/import failure is not the same as the Python 3.9+/3.10+ compatibility issue." + }, + { + "left": "issue:41950", + "right": "issue:44220", + "accept": false, + "reason": "Video-classification pipeline image-processor lookup and _torch_extract_fbank_features are different pipeline/audio bugs." + }, + { + "left": "issue:43976", + "right": "issue:44964", + "accept": false, + "reason": "Interpreter-version incompatibility and Phi-4 multimodal loading failure are different problems." + }, + { + "left": "issue:42886", + "right": "issue:46132", + "accept": false, + "reason": "Offline tokenizer cache loading and AttentionInterface.register behavior change are unrelated." + }, + { + "left": "issue:36246", + "right": "issue:42757", + "accept": false, + "reason": "Missing Qwen2_5_VL image processor export is unrelated to missing is_offline_mode from huggingface_hub." + }, + { + "left": "issue:36246", + "right": "issue:44351", + "accept": false, + "reason": "These are separate import errors for different symbols and different packages." + }, + { + "left": "issue:44779", + "right": "issue:45292", + "accept": false, + "reason": "Tokenizer correctness regression and resize_token_embeddings/output_embeddings mismatch are different bugs." + }, + { + "left": "issue:44242", + "right": "issue:44857", + "accept": false, + "reason": "Training loss omission in D-FINE and LwDetrImageLoss AMP crash are distinct issues." + }, + { + "left": "issue:45081", + "right": "issue:45362", + "accept": false, + "reason": "Mistral tokenizer regex patch crash and Qwen3.5 chat crash are different model/tokenizer paths." + }, + { + "left": "issue:43673", + "right": "issue:44561", + "accept": false, + "reason": "Both are v5 regressions, but one is GenerationMixin chunked_prefill cache handling and the other is trust_remote_code import breakage." + }, + { + "left": "issue:44295", + "right": "issue:44749", + "accept": false, + "reason": "Position_ids buffer access and slowed filtering after upgrade are not the same bug." + }, + { + "left": "issue:43352", + "right": "issue:45250", + "accept": false, + "reason": "Unsupported Flash Attention 2.0 on Nemotron and a generic Flash Attention 2.0 topic are not the same concrete issue." + }, + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "ESM2 breakage and Gemma 4 mm_token_type_ids defaults are unrelated." + }, + { + "left": "issue:43901", + "right": "issue:45003", + "accept": false, + "reason": "A doc/behavior mismatch for return_all_scores is unrelated to unsafe sys.modules access." + }, + { + "left": "issue:43526", + "right": "issue:45003", + "accept": false, + "reason": "BeitImageProcessorFast reduce_labels and modeling_utils sys.modules access are different code paths." + }, + { + "left": "issue:44496", + "right": "issue:45003", + "accept": false, + "reason": "Unrecognized model config and unsafe sys.modules access are separate failure modes." + }, + { + "left": "issue:45412", + "right": "issue:46123", + "accept": false, + "reason": "RT-DETR memory lifecycle and MaskGenerationPipeline final-batch dropping are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 36, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:36246", + "issue:42886", + "issue:43065", + "issue:43232", + "issue:43257", + "issue:43262", + "issue:43408", + "issue:43475", + "issue:43540", + "issue:43596", + "issue:43653", + "issue:43673", + "issue:43931", + "issue:44060", + "issue:44077", + "issue:44162", + "issue:44246", + "issue:44336", + "issue:44368", + "issue:44466", + "issue:44479", + "issue:44483", + "issue:44734", + "issue:44877", + "issue:44977", + "issue:44987", + "issue:44995", + "issue:45003", + "issue:45137", + "issue:45200", + "issue:45229", + "issue:45292", + "issue:45335", + "issue:45375", + "issue:45397", + "issue:45459", + "issue:45561", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45600", + "issue:45656", + "issue:45701", + "issue:45759", + "issue:45987", + "issue:46082", + "issue:46170", + "issue:46202" + ], + "soft_pairs": [ + "issue:45375|issue:45759", + "issue:43257|issue:45137", + "issue:43262|issue:44246", + "issue:45292|issue:45584", + "issue:36246|issue:43596", + "issue:45292|issue:45701", + "issue:44877|issue:45375", + "issue:45561|issue:45563", + "issue:43232|issue:44336", + "issue:44987|issue:45459", + "issue:43065|issue:43408", + "issue:45987|issue:46082", + "issue:44995|issue:45200", + "issue:44479|issue:46170", + "issue:43475|issue:43540", + "issue:44466|issue:45335", + "issue:45397|issue:45656", + "issue:44483|issue:46202", + "issue:43653|issue:44162", + "issue:45137|issue:45600", + "issue:44368|issue:44977", + "issue:44995|issue:45588", + "issue:44368|issue:45563", + "issue:44877|issue:46170", + "issue:43931|issue:44368", + "issue:42886|issue:45229", + "issue:44060|issue:44077", + "issue:45003|issue:45561", + "issue:43673|issue:44734", + "issue:45584|issue:46170" + ], + "prepared_review_unit_hash": "209e1ead19b24158784ac127703d21448be6f15e07acee5410ccbd7c92df327e", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36796, + "estimated_input_tokens": 9199, + "estimated_eval_tokens": 18654 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: mostly unrelated issue reports spanning multimodal loading, generation, tokenization, deepspeed, config validation, and CI/test utilities. The soft pairs are mostly superficial topical overlaps rather than the same underlying bug.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45375", + "right": "issue:45759", + "accept": false, + "reason": "Different bugs in composite Qwen multimodal config handling: one drops a config field, the other fails to unwrap text_config in from_config." + }, + { + "left": "issue:43257", + "right": "issue:45137", + "accept": false, + "reason": "Unrelated failures: Qwen3 MoE loading under accelerate+deepspeed vs DeepSpeed ZeRO3 deque underflow with BertModel." + }, + { + "left": "issue:43262", + "right": "issue:44246", + "accept": false, + "reason": "Audio chat-template sampling-rate default bug vs slow transformers import; no shared code path." + }, + { + "left": "issue:45292", + "right": "issue:45584", + "accept": false, + "reason": "resize_token_embeddings/output-embeddings bug is unrelated to Whisper empty transcription after align_special_tokens." + }, + { + "left": "issue:36246", + "right": "issue:43596", + "accept": false, + "reason": "Qwen2.5 VL import error and a ZeRO3 BertModel index error are different subsystems and failure modes." + }, + { + "left": "issue:45292", + "right": "issue:45701", + "accept": false, + "reason": "Embedding-resize synchronization bug vs version-dependent tokenization behavior; not the same issue." + }, + { + "left": "issue:44877", + "right": "issue:45375", + "accept": false, + "reason": "Both are config-related, but one is strict granite_speech loading and the other is a missing Qwen vision config field." + }, + { + "left": "issue:45561", + "right": "issue:45563", + "accept": false, + "reason": "pytest-xdist file race in testing utils vs a stale generation warning; unrelated." + }, + { + "left": "issue:43232", + "right": "issue:44336", + "accept": false, + "reason": "Generation kwargs update after sync_gpus is unrelated to ANSI codes emitted by loading_report." + }, + { + "left": "issue:44987", + "right": "issue:45459", + "accept": false, + "reason": "Loading a model on transformers>=5.1.0 and tokenizer error masking from protobuf exception handling are distinct bugs." + }, + { + "left": "issue:43065", + "right": "issue:43408", + "accept": false, + "reason": "Both mention SAM 3, but one is a dummy Conv2d implementation issue and the other is a model-type warning for tracker/video configs." + }, + { + "left": "issue:45987", + "right": "issue:46082", + "accept": false, + "reason": "StaticCache seq-length return type contract and LlamaConfig head_dim validation are unrelated." + }, + { + "left": "issue:44995", + "right": "issue:45200", + "accept": false, + "reason": "GlmMoeDsa stale indexer cache on second forward is unrelated to Gemma 4 mm_token_type_ids defaults." + }, + { + "left": "issue:44479", + "right": "issue:46170", + "accept": false, + "reason": "Qwen video-input regression vs Windows ROCm torch.distributed import dependency issue; different areas entirely." + }, + { + "left": "issue:43475", + "right": "issue:43540", + "accept": false, + "reason": "SAM 3 video encoder attribute error and Qwen3OmniMoe video-processing ValueError are separate model bugs." + }, + { + "left": "issue:44466", + "right": "issue:45335", + "accept": false, + "reason": "Tied-weight serialization inconsistency is not the same as t5gemma resize_token_embeddings failing to update decoder.embed_tokens." + }, + { + "left": "issue:45397", + "right": "issue:45656", + "accept": false, + "reason": "Gemma-4 ZeRO3 from_pretrained failure and double optimizer.step under deepspeed are unrelated training/runtime bugs." + }, + { + "left": "issue:44483", + "right": "issue:46202", + "accept": false, + "reason": "Chat/completions request rejection is unrelated to Gemma tokenization roundtrip failure on the underscore marker." + }, + { + "left": "issue:43653", + "right": "issue:44162", + "accept": false, + "reason": "BigBirdTokenizer special-token registration bug vs a broad ESM2 regression; no concrete shared bug." + }, + { + "left": "issue:45137", + "right": "issue:45600", + "accept": false, + "reason": "ZeRO3 empty deque crash and removed Sam3LiteText references in auto_mappings are different issues." + }, + { + "left": "issue:44368", + "right": "issue:44977", + "accept": false, + "reason": "Tie-word-embeddings warning during fine-tuning is not the same as Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:44995", + "right": "issue:45588", + "accept": false, + "reason": "GlmMoeDsa cache corruption and flash_attention s_aux=None crash are unrelated code paths." + }, + { + "left": "issue:44368", + "right": "issue:45563", + "accept": false, + "reason": "A model config warning and a stale num_return_sequences warning are not the same underlying bug." + }, + { + "left": "issue:44877", + "right": "issue:46170", + "accept": false, + "reason": "Strict config loading for granite_speech is unrelated to a Windows ROCm import/dependency failure." + }, + { + "left": "issue:43931", + "right": "issue:44368", + "accept": false, + "reason": "Qwen3-VL weight shape mismatch and Qwen3.5 tie_word_embeddings warning are different model-loading problems." + }, + { + "left": "issue:42886", + "right": "issue:45229", + "accept": false, + "reason": "Offline tokenizer cache loading and Gemma 4 multi-GPU OOM are unrelated." + }, + { + "left": "issue:44060", + "right": "issue:44077", + "accept": false, + "reason": "Qwen3-Next tied-weights warning bug and patchtsmixer post_init allowance are separate config/validation issues." + }, + { + "left": "issue:45003", + "right": "issue:45561", + "accept": false, + "reason": "Unsafe sys.modules access in modeling_utils is unrelated to the xdist captured_info.txt race." + }, + { + "left": "issue:43673", + "right": "issue:44734", + "accept": false, + "reason": "Both involve generation/cache, but one is missing GenerationMixin cache in chunked prefill and the other is wrong tensor indexing in serve /v1/responses." + }, + { + "left": "issue:45584", + "right": "issue:46170", + "accept": false, + "reason": "Whisper empty transcription after align_special_tokens has nothing to do with the Windows ROCm torch.distributed import problem." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 37, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:28282", + "issue:30990", + "issue:41628", + "issue:42371", + "issue:42673", + "issue:42994", + "issue:43232", + "issue:43475", + "issue:43482", + "issue:43525", + "issue:43643", + "issue:43646", + "issue:43673", + "issue:43701", + "issue:43704", + "issue:43824", + "issue:43827", + "issue:44038", + "issue:44162", + "issue:44188", + "issue:44222", + "issue:44297", + "issue:44351", + "issue:44361", + "issue:44623", + "issue:44661", + "issue:44749", + "issue:44869", + "issue:44871", + "issue:44933", + "issue:44964", + "issue:45081", + "issue:45092", + "issue:45103", + "issue:45137", + "issue:45230", + "issue:45310", + "issue:45356", + "issue:45375", + "issue:45381", + "issue:45397", + "issue:45406", + "issue:45478", + "issue:45656", + "issue:46077", + "issue:46123", + "issue:46132", + "issue:46202" + ], + "soft_pairs": [ + "issue:44297|issue:44623", + "issue:44869|issue:45356", + "issue:44361|issue:44749", + "issue:43646|issue:44162", + "issue:44749|issue:45656", + "issue:44623|issue:45656", + "issue:44871|issue:46132", + "issue:43701|issue:44749", + "issue:43673|issue:43827", + "issue:44188|issue:46202", + "issue:45230|issue:45397", + "issue:42994|issue:43482", + "issue:44038|issue:44623", + "issue:43475|issue:45381", + "issue:28282|issue:30990", + "issue:43475|issue:43643", + "issue:42371|issue:45092", + "issue:43232|issue:43673", + "issue:41628|issue:44933", + "issue:45375|issue:45478", + "issue:43475|issue:44661", + "issue:42994|issue:43525", + "issue:46077|issue:46123", + "issue:45310|issue:45375", + "issue:45081|issue:45103", + "issue:45406|issue:46202", + "issue:42673|issue:43704", + "issue:41628|issue:44351", + "issue:44964|issue:45137", + "issue:44222|issue:45310", + "issue:42994|issue:43824", + "issue:44038|issue:45310" + ], + "prepared_review_unit_hash": "dcbb1239ee0eb245a16de6f93ab2187306ce086c1e75f5982070dda919ef9a7c", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37108, + "estimated_input_tokens": 9277, + "estimated_eval_tokens": 18810 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous, but the Qwen3.5 MoE loading reports form one clear duplicate set: the generic `from_pretrained` failure and the missing `deepstack_visual_indexes` config-field report appear to be the same regression. The rest of the pairs are different bugs or different code paths.", + "confidence": 0.58, + "canonical_issue_reason": "issue:45375 is the best canonical issue for the Qwen3.5 MoE loading regression because it names the likely root cause (`deepstack_visual_indexes` being dropped), while 45310/45478 are symptom-level load errors.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45375 is the strongest representative overall among the accepted duplicates: it is specific, root-cause oriented, and explains the broader `from_pretrained` failures on Qwen3.5 MoE.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44297", + "right": "issue:44623", + "accept": false, + "reason": "Both are save_pretrained issues, but one is tokenizer-class metadata and the other is missing processor files; different failures." + }, + { + "left": "issue:44869", + "right": "issue:45356", + "accept": false, + "reason": "Both concern tokenizer/codec behavior, but they target different models and distinct decode/regression paths." + }, + { + "left": "issue:44361", + "right": "issue:44749", + "accept": false, + "reason": "One is a tokenizer AttributeError, the other is a performance regression after upgrade; unrelated bugs." + }, + { + "left": "issue:43646", + "right": "issue:44162", + "accept": false, + "reason": "Custom config instantiation vs ESM2 breakage are different code paths and different symptoms." + }, + { + "left": "issue:44749", + "right": "issue:45656", + "accept": false, + "reason": "Data filtering slowdown and double optimizer stepping are unrelated training issues." + }, + { + "left": "issue:44623", + "right": "issue:45656", + "accept": false, + "reason": "Processor save_pretrained file omission is unrelated to DeepSpeed optimizer stepping." + }, + { + "left": "issue:44871", + "right": "issue:46132", + "accept": false, + "reason": "Eos-token config inconsistency and AttentionInterface behavior changes are different problems." + }, + { + "left": "issue:43701", + "right": "issue:44749", + "accept": false, + "reason": "Checkpoint key mismatch is unrelated to the data filtering slowdown report." + }, + { + "left": "issue:43673", + "right": "issue:43827", + "accept": false, + "reason": "Generation cache missing and docs still referencing removed pipeline APIs are not the same bug." + }, + { + "left": "issue:44188", + "right": "issue:46202", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile is unrelated to Gemma pipeline roundtrip failure." + }, + { + "left": "issue:45230", + "right": "issue:45397", + "accept": false, + "reason": "Placeholder bug report vs gemma-4 zero3 loading issue; no concrete overlap." + }, + { + "left": "issue:42994", + "right": "issue:43482", + "accept": false, + "reason": "Quantized model save failure and Qwen2.5-GGUF load failure are opposite directions and different code paths." + }, + { + "left": "issue:44038", + "right": "issue:44623", + "accept": false, + "reason": "Qwen3-VL-Moe v5 compatibility and processor save_pretrained missing files are not the same bug." + }, + { + "left": "issue:43475", + "right": "issue:45381", + "accept": false, + "reason": "SAM 3 video attribute error and qwen2.5-vl vision position ids mismatch are different model-specific bugs." + }, + { + "left": "issue:28282", + "right": "issue:30990", + "accept": false, + "reason": "Missing PyTorch import error and sentence-transformers loading hang are unrelated failure modes." + }, + { + "left": "issue:43475", + "right": "issue:43643", + "accept": false, + "reason": "SAM 3 video attribute error is unrelated to trust_remote_code missing-fields behavior in AutoConfig." + }, + { + "left": "issue:42371", + "right": "issue:45092", + "accept": false, + "reason": "TF32 settings guidance and InternVL2 v5 meta-init incompatibility are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:43673", + "accept": false, + "reason": "Both are generation-related, but one concerns sync_gpus update ordering and the other chunked_prefill cache absence." + }, + { + "left": "issue:41628", + "right": "issue:44933", + "accept": false, + "reason": "Both are import errors, but they reference different missing symbols and different modules." + }, + { + "left": "issue:45375", + "right": "issue:45478", + "accept": true, + "reason": "Same Qwen3.5 MoE load regression: one report names the missing config field, the other reports the resulting from_pretrained failure." + }, + { + "left": "issue:43475", + "right": "issue:44661", + "accept": false, + "reason": "SAM 3 video attribute error and tokenizer-mapping add-new-model-like failure are unrelated." + }, + { + "left": "issue:42994", + "right": "issue:43525", + "accept": false, + "reason": "Quantized save failure and Llama4Config missing pad_token_id are different bugs." + }, + { + "left": "issue:46077", + "right": "issue:46123", + "accept": false, + "reason": "Gemma4 attention-mask construction and mask-generation partial-batch handling are separate issues." + }, + { + "left": "issue:45310", + "right": "issue:45375", + "accept": true, + "reason": "The generic Qwen3.5 MoE from_pretrained error and the missing config-field report describe the same loading regression." + }, + { + "left": "issue:45081", + "right": "issue:45103", + "accept": false, + "reason": "Tokenizer regex patch crash and auto-docstring annotation crash are unrelated." + }, + { + "left": "issue:45406", + "right": "issue:46202", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer and Gemma pipeline roundtrip on \u2581 are different bugs." + }, + { + "left": "issue:42673", + "right": "issue:43704", + "accept": false, + "reason": "These have the same title, but there is no evidence they are the same underlying VRAM leak instance; only title similarity is too weak here." + }, + { + "left": "issue:41628", + "right": "issue:44351", + "accept": false, + "reason": "AutoImageProcessor import failure and HybridCache import failure are different missing-symbol regressions." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Phi-4 multimodal load failure and DeepSpeed ZeRO3 deque underflow are unrelated." + }, + { + "left": "issue:44222", + "right": "issue:45310", + "accept": false, + "reason": "FP8 save_pretrained for MoE and Qwen3.5 MoE from_pretrained loading error are different directions and different failures." + }, + { + "left": "issue:42994", + "right": "issue:43824", + "accept": false, + "reason": "Quantized save failure and Qwen2.5-VL import failure are unrelated." + }, + { + "left": "issue:44038", + "right": "issue:45310", + "accept": false, + "reason": "Both mention Qwen MoE and v5-era regressions, but they are different model families and different concrete failures." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The accepted Qwen3.5 MoE pairs are plausibly the same loading regression, with 45375 acting as the root-cause report and 45310/45478 as symptom reports. The remaining pairs are reasonably rejected as different bugs or code paths, and none of the verdicts overstate the evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 38, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:30333", + "issue:30990", + "issue:33290", + "issue:35141", + "issue:36010", + "issue:41628", + "issue:42175", + "issue:42222", + "issue:42831", + "issue:42898", + "issue:43065", + "issue:43232", + "issue:43278", + "issue:43377", + "issue:43723", + "issue:43784", + "issue:43824", + "issue:43874", + "issue:43901", + "issue:44016", + "issue:44060", + "issue:44075", + "issue:44188", + "issue:44222", + "issue:44242", + "issue:44295", + "issue:44373", + "issue:44448", + "issue:44496", + "issue:44568", + "issue:44623", + "issue:44734", + "issue:44779", + "issue:44792", + "issue:45092", + "issue:45103", + "issue:45125", + "issue:45137", + "issue:45200", + "issue:45310", + "issue:45356", + "issue:45561", + "issue:45600", + "issue:45656", + "issue:45712", + "issue:45812", + "issue:46123", + "issue:46231" + ], + "soft_pairs": [ + "issue:36010|issue:41628", + "issue:42831|issue:44222", + "issue:42175|issue:45712", + "issue:44496|issue:45561", + "issue:43377|issue:44242", + "issue:45561|issue:45600", + "issue:42898|issue:44568", + "issue:43901|issue:45561", + "issue:43232|issue:44792", + "issue:30990|issue:35141", + "issue:43278|issue:45137", + "issue:42222|issue:45200", + "issue:43065|issue:46123", + "issue:44623|issue:45310", + "issue:30333|issue:33290", + "issue:44188|issue:44373", + "issue:43232|issue:44188", + "issue:42898|issue:45812", + "issue:43784|issue:43824", + "issue:43723|issue:44568", + "issue:43874|issue:45092", + "issue:42898|issue:45356", + "issue:42898|issue:44779", + "issue:44734|issue:45103", + "issue:44016|issue:46231", + "issue:44075|issue:46231", + "issue:44060|issue:45125", + "issue:44295|issue:45656", + "issue:42898|issue:44448" + ], + "prepared_review_unit_hash": "fd78c5b67b259d24208e8f573353dd54210b02efd378a657034cd19b0047e815", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36485, + "estimated_input_tokens": 9122, + "estimated_eval_tokens": 18500 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: most items are unrelated bug reports across different subsystems (tokenizers, generation, training, multimodal, docs, deepspeed, import errors). None of the soft pairs look like the same underlying issue/change, so I would not merge any of them as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "If a representative issue is needed, issue 42898 is the broadest and most concrete regression report (tokenization behavior change in v5), but the set is too mixed to justify a true canonical duplicate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36010", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols (`GenerationMixin` vs `AutoImageProcessor`) and different API paths." + }, + { + "left": "issue:42831", + "right": "issue:44222", + "accept": false, + "reason": "Both mention FP8, but one is about accuracy and the other about `save_pretrained` for MoE; different failures." + }, + { + "left": "issue:42175", + "right": "issue:45712", + "accept": false, + "reason": "One is about TensorFlow missing from a torch-only install, the other about leftover dummy classes; unrelated problems." + }, + { + "left": "issue:44496", + "right": "issue:45561", + "accept": false, + "reason": "Unrecognized model/config loading is unrelated to the xdist race on `captured_info.txt`." + }, + { + "left": "issue:43377", + "right": "issue:44242", + "accept": false, + "reason": "Different bugs: padding mask support in MIMI encoder vs missing load-balancing loss when router logits are off." + }, + { + "left": "issue:45561", + "right": "issue:45600", + "accept": false, + "reason": "A test file race condition is unrelated to stale `auto_mappings.py` references breaking CI." + }, + { + "left": "issue:42898", + "right": "issue:44568", + "accept": false, + "reason": "Both are tokenizer regressions in v5, but they affect different behaviors and model families (`clean_up_tokenization_spaces` vs BOS/EOS insertion)." + }, + { + "left": "issue:43901", + "right": "issue:45561", + "accept": false, + "reason": "A doc mismatch for `TextClassificationPipeline` does not match a pytest-xdist file race." + }, + { + "left": "issue:43232", + "right": "issue:44792", + "accept": false, + "reason": "Generation keyword update after `sync_gpus` is not the same as a failed janus image-generation test." + }, + { + "left": "issue:30990", + "right": "issue:35141", + "accept": false, + "reason": "Loading hangs in sentence-transformers vs embedding reinitialization after resize are different symptoms and code paths." + }, + { + "left": "issue:43278", + "right": "issue:45137", + "accept": false, + "reason": "BF16\u2192FP32 eval dtype drift is unrelated to the DeepSpeed ZeRO3 deque underflow." + }, + { + "left": "issue:42222", + "right": "issue:45200", + "accept": false, + "reason": "Broken VitPose model files and missing `mm_token_type_ids` defaults are different multimodal issues." + }, + { + "left": "issue:43065", + "right": "issue:46123", + "accept": false, + "reason": "A dummy `Conv2d` in Sam3PixelDecoder is unrelated to `MaskGenerationPipeline` dropping partial-batch results." + }, + { + "left": "issue:44623", + "right": "issue:45310", + "accept": false, + "reason": "Missing files from `processor.save_pretrained` is not the same as Qwen3.5 MoE `from_pretrained` failure in newer transformers." + }, + { + "left": "issue:30333", + "right": "issue:33290", + "accept": false, + "reason": "MLflow reporting on failed training jobs and DeepSpeed OOM with Adafactor are separate training/runtime issues." + }, + { + "left": "issue:44188", + "right": "issue:44373", + "accept": false, + "reason": "Attention-kernel divergence under `torch.compile` is unrelated to a wrong `position_ids` docstring." + }, + { + "left": "issue:43232", + "right": "issue:44188", + "accept": false, + "reason": "Generation kwargs handling and attention-kernel branching are different code paths and bugs." + }, + { + "left": "issue:42898", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenization regressions, but one is about cleanup-space behavior and the other about wrong token IDs for Granite; not the same bug." + }, + { + "left": "issue:43784", + "right": "issue:43824", + "accept": false, + "reason": "Both are import failures, but they involve different missing names from different modules and different packages." + }, + { + "left": "issue:43723", + "right": "issue:44568", + "accept": false, + "reason": "Tokenizer loading issues and missing BOS/EOS on `add_special_tokens=True` are distinct regression reports." + }, + { + "left": "issue:43874", + "right": "issue:45092", + "accept": false, + "reason": "A missing image-patch helper in GLM46V is unrelated to InternVL2 meta-initialization incompatibility." + }, + { + "left": "issue:42898", + "right": "issue:45356", + "accept": false, + "reason": "Both are tokenizer regressions, but they concern different models and different behavioral changes; too broad to be duplicates." + }, + { + "left": "issue:42898", + "right": "issue:44779", + "accept": false, + "reason": "Different tokenizer regressions for different model families; no evidence they stem from the same underlying change." + }, + { + "left": "issue:44734", + "right": "issue:45103", + "accept": false, + "reason": "KV-cache continuation crash in `transformers serve` is unrelated to auto-docstring parsing with future annotations." + }, + { + "left": "issue:44016", + "right": "issue:46231", + "accept": false, + "reason": "Notebook syntax error and a docs parameter-name mismatch are unrelated documentation problems." + }, + { + "left": "issue:44075", + "right": "issue:46231", + "accept": false, + "reason": "Unused SGD optimizer args and a docs typo in `set_training()` are not the same issue." + }, + { + "left": "issue:44060", + "right": "issue:45125", + "accept": false, + "reason": "Both involve Qwen3 variants, but one is a tied-weights warning bug and the other is a missing tensor-parallel plan." + }, + { + "left": "issue:44295", + "right": "issue:45656", + "accept": false, + "reason": "Registering `position_ids` as a buffer causing errors is unrelated to DeepSpeed double optimizer-step behavior." + }, + { + "left": "issue:42898", + "right": "issue:44448", + "accept": false, + "reason": "A tokenizer cleanup regression is not the same as differing outputs from `google/pegasus-cnn_dailymail` across v4/v5." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 39, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:28282", + "issue:30333", + "issue:34689", + "issue:35141", + "issue:36331", + "issue:41628", + "issue:42175", + "issue:42757", + "issue:42831", + "issue:43122", + "issue:43232", + "issue:43475", + "issue:43479", + "issue:43644", + "issue:43723", + "issue:43873", + "issue:43976", + "issue:44079", + "issue:44190", + "issue:44206", + "issue:44222", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44373", + "issue:44448", + "issue:44484", + "issue:44485", + "issue:44623", + "issue:44869", + "issue:44933", + "issue:44945", + "issue:44977", + "issue:45071", + "issue:45092", + "issue:45103", + "issue:45245", + "issue:45310", + "issue:45381", + "issue:45405", + "issue:45656", + "issue:45712", + "issue:45834", + "issue:46082", + "issue:46132", + "issue:46209", + "issue:46305" + ], + "soft_pairs": [ + "issue:45405|issue:45712", + "issue:44448|issue:46132", + "issue:42175|issue:44933", + "issue:44222|issue:44297", + "issue:44933|issue:45712", + "issue:43976|issue:44933", + "issue:43479|issue:44079", + "issue:44297|issue:45656", + "issue:44222|issue:44623", + "issue:30333|issue:34689", + "issue:42831|issue:44623", + "issue:45245|issue:46132", + "issue:44190|issue:46082", + "issue:30333|issue:36331", + "issue:43479|issue:44869", + "issue:45071|issue:46082", + "issue:28282|issue:30333", + "issue:30333|issue:35141", + "issue:43475|issue:46305", + "issue:44297|issue:45310", + "issue:41628|issue:42757", + "issue:44206|issue:45103", + "issue:43723|issue:44977", + "issue:45381|issue:46305", + "issue:44263|issue:45092", + "issue:43873|issue:44945", + "issue:43122|issue:46209", + "issue:43232|issue:45834", + "issue:44222|issue:44484", + "issue:44373|issue:44485", + "issue:44295|issue:44623", + "issue:43479|issue:43644" + ], + "prepared_review_unit_hash": "30636f536e68d464285c9913fb0f497a4efd42ec7023f5a9258dd31f849dbb1a", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36455, + "estimated_input_tokens": 9114, + "estimated_eval_tokens": 18484 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly a set of unrelated issue reports with superficial lexical overlap around serialization, imports, and version regressions. None of the soft pairs look like true duplicates or mergeable PR-equivalents.", + "confidence": 0.86, + "canonical_issue_reason": "Issue 44297 is the clearest self-contained bug report among the candidates: it describes a specific save_pretrained serialization mismatch with a concrete failure mode. It is also the best representative of the small save/pretrained-adjacent subset in this cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44297 is the strongest standalone issue to keep as the representative bug: concrete, reproducible, and centered on a single code path rather than a broad regression description.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45405", + "right": "issue:45712", + "accept": false, + "reason": "Different problems: a missing released dependency version vs leftover dummy PT classes and repo-check leakage." + }, + { + "left": "issue:44448", + "right": "issue:46132", + "accept": false, + "reason": "One is a model output regression for Pegasus; the other is a behavior change in AttentionInterface.register. Same subsystem, different bug." + }, + { + "left": "issue:42175", + "right": "issue:44933", + "accept": false, + "reason": "Backend packaging/install issue versus a nonexistent import from image_utils; no shared code-path bug." + }, + { + "left": "issue:44222", + "right": "issue:44297", + "accept": false, + "reason": "Both mention save_pretrained, but one is FP8 MoE saving and the other is tokenizer_config metadata mismatch. Different failure modes and artifacts." + }, + { + "left": "issue:44933", + "right": "issue:45712", + "accept": false, + "reason": "Both are import-related, but one is an invalid image_utils import and the other is dummy_pt_objects leakage. Not the same bug." + }, + { + "left": "issue:43976", + "right": "issue:44933", + "accept": false, + "reason": "Python version support regression versus a missing import symbol; unrelated." + }, + { + "left": "issue:43479", + "right": "issue:44079", + "accept": false, + "reason": "Config initialization bug versus ModelOutput key assignment bug; different objects and behaviors." + }, + { + "left": "issue:44297", + "right": "issue:45656", + "accept": false, + "reason": "Tokenizer save metadata issue versus optimizer stepping twice under deepspeed; unrelated." + }, + { + "left": "issue:44222", + "right": "issue:44623", + "accept": false, + "reason": "Both involve save_pretrained, but one is FP8 MoE model saving and the other is processor file emission; not the same code-path problem." + }, + { + "left": "issue:30333", + "right": "issue:34689", + "accept": false, + "reason": "MLFlow job reporting on failure versus model loading breakage in a specific Transformers version." + }, + { + "left": "issue:42831", + "right": "issue:44623", + "accept": false, + "reason": "FP8 accuracy issue versus missing processor save files; different concerns." + }, + { + "left": "issue:45245", + "right": "issue:46132", + "accept": false, + "reason": "Category-count limit error versus attention registration behavior change; unrelated." + }, + { + "left": "issue:44190", + "right": "issue:46082", + "accept": false, + "reason": "Dataset loading problem versus LlamaConfig validation semantics; not the same bug." + }, + { + "left": "issue:30333", + "right": "issue:36331", + "accept": false, + "reason": "MLFlow reporting failure versus Trainer compute_loss signature breakage; different training-path issues." + }, + { + "left": "issue:43479", + "right": "issue:44869", + "accept": false, + "reason": "Multimodal config defaulting bug versus Whisper word timestamp decoding crash; unrelated." + }, + { + "left": "issue:45071", + "right": "issue:46082", + "accept": false, + "reason": "Type-checking regression versus config validation edge case; no shared underlying defect." + }, + { + "left": "issue:28282", + "right": "issue:30333", + "accept": false, + "reason": "PyTorch import availability issue versus MLFlow run-state reporting; unrelated." + }, + { + "left": "issue:30333", + "right": "issue:35141", + "accept": false, + "reason": "Training job reporting versus token embedding resize initialization; different code paths." + }, + { + "left": "issue:43475", + "right": "issue:46305", + "accept": false, + "reason": "SAM3 video encoder attribute error versus Moonshine streaming mask off-by-one; unrelated." + }, + { + "left": "issue:44297", + "right": "issue:45310", + "accept": false, + "reason": "Tokenizer save_pretrained metadata mismatch versus Qwen3.5 MoE from_pretrained error; both involve loading/saving but not the same bug." + }, + { + "left": "issue:41628", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but for different missing symbols and different modules; no common underlying fix." + }, + { + "left": "issue:44206", + "right": "issue:45103", + "accept": false, + "reason": "Feature extractor parameter regression versus auto_docstring annotations crash; unrelated." + }, + { + "left": "issue:43723", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer loading regression versus flash-attention generation failure; different subsystems and symptoms." + }, + { + "left": "issue:45381", + "right": "issue:46305", + "accept": false, + "reason": "Both mention positional/window concepts, but one is VL vision_position_ids and the other is streaming frame masking. Not the same defect." + }, + { + "left": "issue:44263", + "right": "issue:45092", + "accept": false, + "reason": "torch.split return handling in one model component versus remote-code meta-init incompatibility in another; unrelated." + }, + { + "left": "issue:43873", + "right": "issue:44945", + "accept": false, + "reason": "Quantization/offloading behavior versus incorrect LLM output under pipeline parallelism; same broad area, different bug." + }, + { + "left": "issue:43122", + "right": "issue:46209", + "accept": false, + "reason": "Tokenization differences across versions versus FP8 inference failure on B200; unrelated." + }, + { + "left": "issue:43232", + "right": "issue:45834", + "accept": false, + "reason": "Generation kwargs update bug versus OCR long-input index error; no shared underlying code-path." + }, + { + "left": "issue:44222", + "right": "issue:44484", + "accept": false, + "reason": "FP8 MoE save_pretrained bug versus a question about shard size defaults; different issue types and causes." + }, + { + "left": "issue:44373", + "right": "issue:44485", + "accept": false, + "reason": "Docstring mismatch for position_ids versus a RoPE implementation discussion; not a duplicate bug report." + }, + { + "left": "issue:44295", + "right": "issue:44623", + "accept": false, + "reason": "Registered-buffer position_ids read error versus missing processor save files; different serialization problems." + }, + { + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Config default initialization versus non-persistent buffer serialization junk; superficially initialization-related but not the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 40, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:22355", + "issue:28282", + "issue:30990", + "issue:38175", + "issue:39401", + "issue:41628", + "issue:42371", + "issue:42757", + "issue:42831", + "issue:42886", + "issue:42907", + "issue:42994", + "issue:43066", + "issue:43479", + "issue:43576", + "issue:43644", + "issue:43646", + "issue:43723", + "issue:43761", + "issue:43784", + "issue:43824", + "issue:44038", + "issue:44117", + "issue:44261", + "issue:44263", + "issue:44360", + "issue:44368", + "issue:44373", + "issue:44464", + "issue:44483", + "issue:44514", + "issue:44568", + "issue:44704", + "issue:44829", + "issue:44871", + "issue:44913", + "issue:44945", + "issue:45092", + "issue:45103", + "issue:45278", + "issue:45310", + "issue:45357", + "issue:45406", + "issue:45753", + "issue:46002", + "issue:46050", + "issue:46082", + "issue:46305" + ], + "soft_pairs": [ + "issue:44483|issue:45406", + "issue:43761|issue:46082", + "issue:43723|issue:44871", + "issue:45092|issue:45357", + "issue:44263|issue:44360", + "issue:42907|issue:46050", + "issue:30990|issue:38175", + "issue:44261|issue:45092", + "issue:44514|issue:45103", + "issue:43644|issue:46305", + "issue:44829|issue:46002", + "issue:43479|issue:44913", + "issue:42831|issue:45310", + "issue:22355|issue:28282", + "issue:44263|issue:44373", + "issue:41628|issue:43824", + "issue:42831|issue:44038", + "issue:39401|issue:44568", + "issue:42886|issue:42994", + "issue:44368|issue:46305", + "issue:42886|issue:43066", + "issue:44945|issue:45753", + "issue:42757|issue:45278", + "issue:44464|issue:46082", + "issue:44117|issue:44704", + "issue:43576|issue:43784", + "issue:43646|issue:46002", + "issue:42371|issue:44263" + ], + "prepared_review_unit_hash": "044dabe110ae4d3583be79fb8cf869850c47935d2693f323924eeb7e4dfcd5da", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 35914, + "estimated_input_tokens": 8979, + "estimated_eval_tokens": 18214 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous; most pairs share only broad subsystem overlap (imports, tokenizers, configs, saving/loading) and not the same concrete bug. No soft edge should be merged.", + "confidence": 0.92, + "canonical_issue_reason": "issue:45278 is the broadest umbrella-style report in the set, centered on widespread import regressions after a version upgrade, so it is the best single representative issue.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the most general and representative issue here; it can stand in for the import-regression side of the cluster better than the many narrow, model-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44483", + "right": "issue:45406", + "accept": false, + "reason": "Different failures: chat-completions request handling versus a Gemma4Processor attribute/import crash." + }, + { + "left": "issue:43761", + "right": "issue:46082", + "accept": false, + "reason": "Unrelated code paths: CLIP vision forward regression versus a LlamaConfig validation error." + }, + { + "left": "issue:43723", + "right": "issue:44871", + "accept": false, + "reason": "Both involve tokenizers, but one is a generic AutoTokenizer loading issue and the other is a Gemma eos_token_id config mismatch." + }, + { + "left": "issue:45092", + "right": "issue:45357", + "accept": false, + "reason": "Both touch multimodal model persistence, but one is meta-init incompatibility and the other is wrong visual encoder keys on save." + }, + { + "left": "issue:44263", + "right": "issue:44360", + "accept": false, + "reason": "Same subsystem name, but the bugs differ: torch.split handling versus a missing ReLU in the DSA indexer." + }, + { + "left": "issue:42907", + "right": "issue:46050", + "accept": false, + "reason": "Both concern quantized model save/load, but they appear to be different quantization paths and different failure modes." + }, + { + "left": "issue:30990", + "right": "issue:38175", + "accept": false, + "reason": "No same root cause: a loading hang in Sentence Transformers versus zero probabilities in a specific SigLIP2 model." + }, + { + "left": "issue:44261", + "right": "issue:45092", + "accept": false, + "reason": "Different model families and bugs: MLA norm epsilon precision error versus remote-code checkpoint/meta-init incompatibility." + }, + { + "left": "issue:44514", + "right": "issue:45103", + "accept": false, + "reason": "Processor batched chat-template crash and auto-docstring annotation handling are unrelated failures." + }, + { + "left": "issue:43644", + "right": "issue:46305", + "accept": false, + "reason": "Different implementation areas and symptoms: non-persistent buffer corruption versus a streaming mask off-by-one." + }, + { + "left": "issue:44829", + "right": "issue:46002", + "accept": false, + "reason": "Both affect training numerics, but one is flash-attention-3 specific and the other is NaNs after init/checkpointing." + }, + { + "left": "issue:43479", + "right": "issue:44913", + "accept": false, + "reason": "Both are config serialization/init issues, but they concern different config classes and different fields." + }, + { + "left": "issue:42831", + "right": "issue:45310", + "accept": false, + "reason": "Different problems: FineGrainedFP8 accuracy drift versus a Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:22355", + "right": "issue:28282", + "accept": false, + "reason": "Both are import-related, but one is a missing transformers.onnx module and the other is a missing PyTorch dependency." + }, + { + "left": "issue:44263", + "right": "issue:44373", + "accept": false, + "reason": "A DSA indexer runtime issue and a docstring bug are not the same underlying change." + }, + { + "left": "issue:41628", + "right": "issue:43824", + "accept": false, + "reason": "Both are missing import symbols, but they are different exported objects and likely separate packaging/export regressions." + }, + { + "left": "issue:42831", + "right": "issue:44038", + "accept": false, + "reason": "Too vague and model-specific to treat as the same bug; one is an accuracy issue, the other a Qwen3-VL-Moe bug report." + }, + { + "left": "issue:39401", + "right": "issue:44568", + "accept": false, + "reason": "Different tokenizer bugs affecting different models and behaviors: offset mapping versus special token insertion." + }, + { + "left": "issue:42886", + "right": "issue:42994", + "accept": false, + "reason": "Offline tokenizer cache loading and quantized model save failures are unrelated." + }, + { + "left": "issue:44368", + "right": "issue:46305", + "accept": false, + "reason": "A config warning about tied embeddings is unrelated to the MoonshineStreaming mask off-by-one." + }, + { + "left": "issue:42886", + "right": "issue:43066", + "accept": false, + "reason": "Both involve tokenizers, but one is offline cache loading and the other is a decoder type mismatch." + }, + { + "left": "issue:44945", + "right": "issue:45753", + "accept": false, + "reason": "Pipeline-parallel output corruption and image-triggered infinite looping are different runtime failures." + }, + { + "left": "issue:42757", + "right": "issue:45278", + "accept": false, + "reason": "45278 is a broad umbrella, but this specific huggingface_hub import error is not clearly the same underlying bug." + }, + { + "left": "issue:44464", + "right": "issue:46082", + "accept": false, + "reason": "Compiled-forward generation inconsistency and a LlamaConfig head_dim validation error are unrelated." + }, + { + "left": "issue:44117", + "right": "issue:44704", + "accept": false, + "reason": "Both are from_pretrained internals, but one is a tokenizer mapping assumption and the other is kwargs propagation to cached_file." + }, + { + "left": "issue:43576", + "right": "issue:43784", + "accept": false, + "reason": "A broken CLI command and an import-time NameError in sentence-transformers are different issues." + }, + { + "left": "issue:43646", + "right": "issue:46002", + "accept": false, + "reason": "Custom model initialization breakage is not the same as NaNs in classification heads after init." + }, + { + "left": "issue:42371", + "right": "issue:44263", + "accept": false, + "reason": "TF32 API guidance and a DSA indexer torch.split issue are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 41, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:28218", + "issue:30990", + "issue:33357", + "issue:36296", + "issue:39401", + "issue:41628", + "issue:42907", + "issue:43116", + "issue:43122", + "issue:43232", + "issue:43408", + "issue:43452", + "issue:43475", + "issue:43550", + "issue:43723", + "issue:43756", + "issue:43784", + "issue:43824", + "issue:43937", + "issue:44038", + "issue:44060", + "issue:44222", + "issue:44261", + "issue:44295", + "issue:44297", + "issue:44351", + "issue:44360", + "issue:44373", + "issue:44442", + "issue:44485", + "issue:44488", + "issue:44514", + "issue:44521", + "issue:44625", + "issue:44704", + "issue:44779", + "issue:44829", + "issue:44945", + "issue:44964", + "issue:45200", + "issue:45216", + "issue:45245", + "issue:45278", + "issue:45290", + "issue:45478", + "issue:45593", + "issue:45600", + "issue:46132" + ], + "soft_pairs": [ + "issue:43550|issue:46132", + "issue:30990|issue:33357", + "issue:44295|issue:44297", + "issue:43232|issue:45593", + "issue:44704|issue:45290", + "issue:42907|issue:44945", + "issue:43723|issue:44442", + "issue:44261|issue:44373", + "issue:43723|issue:44779", + "issue:42907|issue:43937", + "issue:41628|issue:45278", + "issue:43756|issue:44060", + "issue:43408|issue:43475", + "issue:43784|issue:45600", + "issue:43122|issue:44514", + "issue:42907|issue:44964", + "issue:43116|issue:45200", + "issue:44625|issue:44829", + "issue:44360|issue:44485", + "issue:43122|issue:44521", + "issue:44038|issue:44297", + "issue:43232|issue:45245", + "issue:28218|issue:30990", + "issue:44038|issue:44295", + "issue:42907|issue:43452", + "issue:43122|issue:44779", + "issue:36296|issue:43116", + "issue:44038|issue:44222", + "issue:45216|issue:45478", + "issue:39401|issue:44488", + "issue:39401|issue:44521", + "issue:43824|issue:44351" + ], + "prepared_review_unit_hash": "68e039eca2d9368d85eb0bdef569efe99e4b0ae0d05e58c48f73a42b341151da", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 36901, + "estimated_input_tokens": 9226, + "estimated_eval_tokens": 18708 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose collection of distinct transformer regressions; the strongest common thread is tokenizer/processor loading or behavior changes in v5, but most pairs are not the same underlying bug.", + "confidence": 0.41, + "canonical_issue_reason": "issue:43723 is the broadest representative of the recurring v5 tokenizer-loading regression theme, and is more central than the highly specific reports around it.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43723 best fits the cluster as a representative issue because it captures the generic AutoTokenizer v5 failure pattern seen across several neighboring reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43550", + "right": "issue:46132", + "accept": false, + "reason": "Different problems: one is a torch.compile/SDPA failure on a specific model, the other is an AttentionInterface.register behavior change." + }, + { + "left": "issue:30990", + "right": "issue:33357", + "accept": false, + "reason": "Unrelated symptoms and models: loading hang for Sentence Transformers vs MacOS bus error for a community CLIP model." + }, + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "One is about reading position_ids after buffer registration; the other is tokenizer_class metadata written by save_pretrained." + }, + { + "left": "issue:43232", + "right": "issue:45593", + "accept": false, + "reason": "Different code paths and failures: generation kwargs after sync_gpus vs missing auxiliary losses in D-FINE." + }, + { + "left": "issue:44704", + "right": "issue:45290", + "accept": false, + "reason": "Both involve processing APIs, but the bugs differ: cached_file kwargs forwarding vs chat-template crashes with tool-call assistant messages." + }, + { + "left": "issue:42907", + "right": "issue:44945", + "accept": false, + "reason": "Different subsystems: dequantized model saving vs incorrect output under pipeline parallelism." + }, + { + "left": "issue:43723", + "right": "issue:44442", + "accept": false, + "reason": "Both are AutoTokenizer load regressions, but the concrete failures and tokenizer targets are different; not the same bug." + }, + { + "left": "issue:44261", + "right": "issue:44373", + "accept": false, + "reason": "Different issues: config.rms_norm_eps precision mismatch vs a wrong docstring for position_ids." + }, + { + "left": "issue:43723", + "right": "issue:44779", + "accept": false, + "reason": "Both mention tokenizer regressions in v5, but they affect different tokenizers and appear to be separate concrete failures." + }, + { + "left": "issue:42907", + "right": "issue:43937", + "accept": false, + "reason": "Different bugs: save_pretrained/dequantization vs GenerationConfig validation for GLM-5." + }, + { + "left": "issue:41628", + "right": "issue:45278", + "accept": false, + "reason": "One missing import is not the same as a broader collection of import errors after an upgrade." + }, + { + "left": "issue:43756", + "right": "issue:44060", + "accept": false, + "reason": "Different model-specific architecture issues: RoPE layer dropping vs tied-weights warning mapping." + }, + { + "left": "issue:43408", + "right": "issue:43475", + "accept": false, + "reason": "Both are SAM 3 related, but one is a model-type warning and the other is a missing attribute in a different execution path." + }, + { + "left": "issue:43784", + "right": "issue:45600", + "accept": false, + "reason": "Different failures: NameError importing sentence-transformers vs auto_mappings.py referencing removed Sam3LiteText configs." + }, + { + "left": "issue:43122", + "right": "issue:44514", + "accept": false, + "reason": "Tokenizer-version regression vs multimodal chat-template crash; related area but not the same underlying bug." + }, + { + "left": "issue:42907", + "right": "issue:44964", + "accept": false, + "reason": "Distinct issues involving model saving/loading and a specific multimodal model load failure." + }, + { + "left": "issue:43116", + "right": "issue:45200", + "accept": false, + "reason": "Different behavior: example script returning empty classification results vs Gemma 4 fine-tuning requiring mm_token_type_ids." + }, + { + "left": "issue:44625", + "right": "issue:44829", + "accept": false, + "reason": "Different model/training bugs: Qwen3.5 num_labels propagation vs flash_attention_3 degenerating sequence classification training." + }, + { + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "Unrelated technical areas: DSA indexer ReLU discussion vs GLM-5 RoPE implementation." + }, + { + "left": "issue:43122", + "right": "issue:44521", + "accept": false, + "reason": "Both touch chat/tokenization APIs, but one is a general tokenizer regression and the other is multimodal assistant_masks being all zero." + }, + { + "left": "issue:44038", + "right": "issue:44297", + "accept": false, + "reason": "Different issues: Qwen3-VL-Moe behavior in v5 vs tokenizer_class metadata mismatch on save_pretrained." + }, + { + "left": "issue:43232", + "right": "issue:45245", + "accept": false, + "reason": "No shared code path: generation kwargs handling vs a category-count runtime limit." + }, + { + "left": "issue:28218", + "right": "issue:30990", + "accept": false, + "reason": "Tokenizer spacing bug vs a loading hang; not the same underlying issue." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "Both are MoE-related only at a high level, but one is a model behavior bug and the other is FP8 save_pretrained support." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Different load/save failures: dequantized model saving vs gguf_file breaking AutoTokenizer/AutoModelForCausalLM loading." + }, + { + "left": "issue:43122", + "right": "issue:44779", + "accept": false, + "reason": "Similar theme of tokenization regressions in v5, but the reports describe different tokenizers and different failure modes." + }, + { + "left": "issue:36296", + "right": "issue:43116", + "accept": false, + "reason": "Tensor parallel training bug is unrelated to the multi-label classification example returning empty results." + }, + { + "left": "issue:45216", + "right": "issue:45478", + "accept": false, + "reason": "Same model family, but one is save_pretrained checkpoint corruption and the other is from_pretrained loading failure." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "Different problems: Qwen3 offset_mapping regression vs a specific model load failure for cjvt/sleng-bert." + }, + { + "left": "issue:39401", + "right": "issue:44521", + "accept": false, + "reason": "Tokenizer offset_mapping bug and multimodal assistant_masks bug are separate issues." + }, + { + "left": "issue:43824", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for unrelated symbols and packages." + }, + { + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "Different subjects entirely; no shared bug or code path is evident." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 42, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:22355", + "issue:28218", + "issue:29127", + "issue:29159", + "issue:33453", + "issue:38617", + "issue:39401", + "issue:42175", + "issue:42222", + "issue:42371", + "issue:42757", + "issue:42831", + "issue:43066", + "issue:43116", + "issue:43122", + "issue:43441", + "issue:43493", + "issue:43611", + "issue:43784", + "issue:43825", + "issue:43874", + "issue:43906", + "issue:43937", + "issue:43976", + "issue:43992", + "issue:44038", + "issue:44075", + "issue:44297", + "issue:44351", + "issue:44355", + "issue:44380", + "issue:44485", + "issue:44625", + "issue:44704", + "issue:44779", + "issue:44936", + "issue:44938", + "issue:45083", + "issue:45278", + "issue:45357", + "issue:45446", + "issue:45478", + "issue:45563", + "issue:45753", + "issue:46002", + "issue:46077", + "issue:46231" + ], + "soft_pairs": [ + "issue:22355|issue:28218", + "issue:43784|issue:45446", + "issue:22355|issue:29159", + "issue:42757|issue:44351", + "issue:43122|issue:43906", + "issue:43992|issue:44704", + "issue:43066|issue:44779", + "issue:44355|issue:46231", + "issue:43937|issue:44485", + "issue:43611|issue:46002", + "issue:42371|issue:43874", + "issue:44075|issue:44936", + "issue:42222|issue:43441", + "issue:42831|issue:44297", + "issue:43976|issue:45446", + "issue:45357|issue:45478", + "issue:44038|issue:45478", + "issue:43116|issue:43825", + "issue:43493|issue:45753", + "issue:45083|issue:45446", + "issue:43116|issue:43493", + "issue:43937|issue:45563", + "issue:22355|issue:29127", + "issue:33453|issue:39401", + "issue:44625|issue:46002", + "issue:44380|issue:46077", + "issue:38617|issue:45278", + "issue:42175|issue:44938" + ], + "prepared_review_unit_hash": "becfb9322c52a52a2885f336cedf9f4174e1be0a65b109bec2139aeb901b8fbb", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 34992, + "estimated_input_tokens": 8748, + "estimated_eval_tokens": 17752 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a loose thematically similar set, not a true duplicate cluster: it mixes tokenizer regressions, import/API breakages, model-specific bugs, and a docs mismatch. None of the soft-edge pairs look like the same underlying bug/change, so I rejected them all. Among the issues, the broadest representative is the tokenizer regression report in 43122.", + "confidence": 0.77, + "canonical_issue_reason": "issue:43122 is the broadest concrete regression report in the dominant tokenizer theme: it describes behavior changing across versions for the same tokenizer, which makes it a better anchor than narrower loading/save/offset-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43122 is the best representative issue for triage because it is version-anchored, user-facing, and broad enough to sit near several other tokenizer-regression reports in this set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:22355", + "right": "issue:28218", + "accept": false, + "reason": "Different failure modes: missing module import vs tokenizer spacing behavior; no shared code path or concrete fix." + }, + { + "left": "issue:43784", + "right": "issue:45446", + "accept": false, + "reason": "Unrelated import problems in different components (sentence-transformers import vs flex_attention version check)." + }, + { + "left": "issue:22355", + "right": "issue:29159", + "accept": false, + "reason": "One is an ONNX import error, the other is slow/fast tokenizer inconsistency; too broad to be the same bug." + }, + { + "left": "issue:42757", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different symbols and different packages/APIs; not the same underlying issue." + }, + { + "left": "issue:43122", + "right": "issue:43906", + "accept": false, + "reason": "The second item is only a reproduction reference with no clear matching symptom; insufficient evidence of the same bug." + }, + { + "left": "issue:43992", + "right": "issue:44704", + "accept": false, + "reason": "Different API paths and failures: missing model weight loading vs AutoProcessor kwargs forwarding." + }, + { + "left": "issue:43066", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but one is decoder-type metadata and the other is wrong tokenization results for Deepseek; not the same concrete bug." + }, + { + "left": "issue:44355", + "right": "issue:46231", + "accept": false, + "reason": "Compiled-Python runtime errors and a documentation parameter-name mismatch are unrelated." + }, + { + "left": "issue:43937", + "right": "issue:44485", + "accept": false, + "reason": "Same model family, but one is GenerationConfig validation and the other is RoPE implementation discussion; different code paths." + }, + { + "left": "issue:43611", + "right": "issue:46002", + "accept": false, + "reason": "Loading with base_model_prefix is unrelated to NaNs in classification heads." + }, + { + "left": "issue:42371", + "right": "issue:43874", + "accept": false, + "reason": "TF32 settings guidance and a missing image-processor method are unrelated issues." + }, + { + "left": "issue:44075", + "right": "issue:44936", + "accept": false, + "reason": "Optimizer argument handling and trainer evaluate-after-train behavior are different subsystems and symptoms." + }, + { + "left": "issue:42222", + "right": "issue:43441", + "accept": false, + "reason": "Different model breakages (ViTpose model loading vs Ministral FlashAttention) with no shared underlying defect." + }, + { + "left": "issue:42831", + "right": "issue:44297", + "accept": false, + "reason": "FP8 accuracy and tokenizer save_pretrained metadata mismatch are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:45446", + "accept": false, + "reason": "Python-version compatibility regression vs a specific import/version gate bug in flex_attention; not the same issue." + }, + { + "left": "issue:45357", + "right": "issue:45478", + "accept": false, + "reason": "Both concern Qwen3.5, but one is save_pretrained key persistence and the other is from_pretrained loading; different bugs." + }, + { + "left": "issue:44038", + "right": "issue:45478", + "accept": false, + "reason": "Different Qwen model variants and different failure modes; too little to merge as one bug." + }, + { + "left": "issue:43116", + "right": "issue:43825", + "accept": false, + "reason": "Example-script classification output bug and a pipeline error-message issue are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:45753", + "accept": false, + "reason": "Different model families and symptoms: SigLIP2 implementation discrepancy vs Qwen3_5 infinite loop on a specific image." + }, + { + "left": "issue:45083", + "right": "issue:45446", + "accept": false, + "reason": "Qwen3_omni_moe feature-length helper behavior and a flex_attention import/version check are unrelated." + }, + { + "left": "issue:43116", + "right": "issue:43493", + "accept": false, + "reason": "Classification example behavior and SigLIP2 implementation parity are different problems." + }, + { + "left": "issue:43937", + "right": "issue:45563", + "accept": false, + "reason": "GenerationConfig validation and a stale warning in paged generate() are separate generation-path issues." + }, + { + "left": "issue:22355", + "right": "issue:29127", + "accept": false, + "reason": "Importing transformers.onnx and a LayoutLMv3 error-message clarity issue are unrelated." + }, + { + "left": "issue:33453", + "right": "issue:39401", + "accept": false, + "reason": "Both are tokenizer regressions, but one is loading regression and the other is wrong offset mapping for Qwen3; not clearly the same bug." + }, + { + "left": "issue:44625", + "right": "issue:46002", + "accept": false, + "reason": "Config propagation for Qwen3.5 and NaNs in classification heads are different defects." + }, + { + "left": "issue:44380", + "right": "issue:46077", + "accept": false, + "reason": "GPT-2 attention scaling under SDPA/FlashAttention and Gemma4 bidirectional attention masking are separate attention-path bugs." + }, + { + "left": "issue:38617", + "right": "issue:45278", + "accept": false, + "reason": "Both are import regressions, but they involve different missing symbols and likely different dependency/version breakages." + }, + { + "left": "issue:42175", + "right": "issue:44938", + "accept": false, + "reason": "Backend extras installation and Python 3.14 loading failure are unrelated environment/package issues." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 43, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:38617", + "issue:39401", + "issue:41628", + "issue:42222", + "issue:42371", + "issue:42673", + "issue:43064", + "issue:43066", + "issue:43122", + "issue:43381", + "issue:43408", + "issue:43421", + "issue:43450", + "issue:43673", + "issue:43742", + "issue:43827", + "issue:43937", + "issue:43986", + "issue:44075", + "issue:44117", + "issue:44261", + "issue:44291", + "issue:44410", + "issue:44466", + "issue:44483", + "issue:44484", + "issue:44496", + "issue:44779", + "issue:44829", + "issue:44877", + "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:45588", + "issue:45656", + "issue:46002", + "issue:46178", + "issue:46231", + "issue:46298" + ], + "soft_pairs": [ + "issue:44117|issue:45588", + "issue:39401|issue:44779", + "issue:43064|issue:43673", + "issue:43381|issue:44945", + "issue:43827|issue:45446", + "issue:43450|issue:43986", + "issue:43937|issue:44977", + "issue:44466|issue:46002", + "issue:45405|issue:45446", + "issue:42673|issue:45072", + "issue:44117|issue:45081", + "issue:43122|issue:46298", + "issue:44410|issue:45310", + "issue:44291|issue:45071", + "issue:44829|issue:45216", + "issue:42222|issue:46178", + "issue:38617|issue:41628", + "issue:39401|issue:43421", + "issue:43408|issue:43742", + "issue:44877|issue:46231", + "issue:44484|issue:45310", + "issue:43937|issue:44496", + "issue:43066|issue:43122", + "issue:44483|issue:45464", + "issue:44075|issue:45656", + "issue:44964|issue:45278", + "issue:42371|issue:44261" + ], + "prepared_review_unit_hash": "a33378f76a051575f4908ea140c53a7a5128e13cffb71344fec3b158a39e5fe1", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35784, + "estimated_input_tokens": 8946, + "estimated_eval_tokens": 18148 + }, + "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. The soft pairs share broad keywords like tokenization, model loading, or optimization, but they do not describe the same concrete bug or change path, so none should be merged as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44117", + "right": "issue:45588", + "accept": false, + "reason": "Different subsystems: tokenizer mapping fallback vs flash-attention crash handling for sink-less models." + }, + { + "left": "issue:39401", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer bugs, but one is Qwen3 offset mappings and the other is DeepSeek tokenization output; different concrete failures." + }, + { + "left": "issue:43064", + "right": "issue:43673", + "accept": false, + "reason": "Optimizer state sharding on FSDP2/PEFT is unrelated to GenerationMixin cache missing during chunked prefill." + }, + { + "left": "issue:43381", + "right": "issue:44945", + "accept": false, + "reason": "Eval-mode gradient checkpointing behavior is a different code path from pipeline-parallel incorrect outputs." + }, + { + "left": "issue:43827", + "right": "issue:45446", + "accept": false, + "reason": "Docs still mentioning pipeline() is unrelated to a runtime PyTorch version check bug." + }, + { + "left": "issue:43450", + "right": "issue:43986", + "accept": false, + "reason": "Batched video processor shape handling is not the same as AutoProcessor crashing without torchvision." + }, + { + "left": "issue:43937", + "right": "issue:44977", + "accept": false, + "reason": "GenerationConfig validation failure is different from Qwen3.5 flash-attention generation behavior." + }, + { + "left": "issue:44466", + "right": "issue:46002", + "accept": false, + "reason": "lm_head serialization/tied-weights issue is unrelated to NaNs in classification heads during checkout/init." + }, + { + "left": "issue:45405", + "right": "issue:45446", + "accept": false, + "reason": "A PEFT minimum-version bump is not the same bug as an AuxRequest import version check." + }, + { + "left": "issue:42673", + "right": "issue:45072", + "accept": false, + "reason": "VRAM leak in multithreaded Qwen3 inference is unrelated to dtype mismatches in CI inference tests." + }, + { + "left": "issue:44117", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer mapping None and Mistral regex patch crashes affect different failure points." + }, + { + "left": "issue:43122", + "right": "issue:46298", + "accept": false, + "reason": "Tokenization output differences and a noisy logging message are unrelated." + }, + { + "left": "issue:44410", + "right": "issue:45310", + "accept": false, + "reason": "Missing Qwen3Next projections is a model-architecture bug, not the same as Qwen3.5 MoE from_pretrained loading failure." + }, + { + "left": "issue:44291", + "right": "issue:45071", + "accept": false, + "reason": "Unexpected _is_hf_initialized argument and PretrainedConfig type-checking are different compatibility issues." + }, + { + "left": "issue:44829", + "right": "issue:45216", + "accept": false, + "reason": "Flash-attention training degeneration is not the same as save_pretrained checkpoint serialization regression." + }, + { + "left": "issue:42222", + "right": "issue:46178", + "accept": false, + "reason": "VitPose model breakage and backend-mapping errors for tensorflow_text are unrelated load-path problems." + }, + { + "left": "issue:38617", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols and different removal/compatibility paths." + }, + { + "left": "issue:39401", + "right": "issue:43421", + "accept": false, + "reason": "Wrong offset mapping and runtime post-processor refresh behavior are different tokenizer issues." + }, + { + "left": "issue:43408", + "right": "issue:43742", + "accept": false, + "reason": "Model-type mismatch warning for SAM3 trackers is unrelated to the MobileLLM config key error." + }, + { + "left": "issue:44877", + "right": "issue:46231", + "accept": false, + "reason": "Strict config loading for granite_speech is unrelated to a docs parameter-name mismatch." + }, + { + "left": "issue:44484", + "right": "issue:45310", + "accept": false, + "reason": "A question about max_shard_size has no overlap with a Qwen3.5 MoE loading error." + }, + { + "left": "issue:43937", + "right": "issue:44496", + "accept": false, + "reason": "Invalid GenerationConfig and unrecognized model/config.json issues are distinct loading failures." + }, + { + "left": "issue:43066", + "right": "issue:43122", + "accept": false, + "reason": "Wrong tokenizer decoder type and changed tokenization output across versions are not the same concrete defect." + }, + { + "left": "issue:44483", + "right": "issue:45464", + "accept": false, + "reason": "Chat/completions request rejection and streaming inference failure on Qwen3.5-0.8B are different API-path bugs." + }, + { + "left": "issue:44075", + "right": "issue:45656", + "accept": false, + "reason": "SGD args being ignored is unrelated to deepspeed calling optimizer.step() twice." + }, + { + "left": "issue:44964", + "right": "issue:45278", + "accept": false, + "reason": "A specific Phi-4 multimodal loading failure is not the same as broad import errors after upgrading transformers." + }, + { + "left": "issue:42371", + "right": "issue:44261", + "accept": false, + "reason": "TF32 configuration warnings are unrelated to MLA q_a_layernorm precision/config handling." + }, + { + "left": "issue:45278", + "right": "issue:41628", + "accept": false, + "reason": "General import errors after upgrade are too broad to equate with one specific missing import symbol." + }, + { + "left": "issue:44977", + "right": "issue:45081", + "accept": false, + "reason": "Qwen3.5 flash-attention generation problems and Mistral regex patch crashes are different tokenizer/model paths." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 44, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:38617", + "issue:41628", + "issue:42175", + "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:43577", + "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:44360", + "issue:44368", + "issue:44410", + "issue:44484", + "issue:44485", + "issue:44493", + "issue:44829", + "issue:44933", + "issue:44945", + "issue:44977", + "issue:44987", + "issue:45071", + "issue:45200", + "issue:45440", + "issue:45478", + "issue:45542", + "issue:45588", + "issue:46050", + "issue:46172", + "issue:46231" + ], + "soft_pairs": [ + "issue:42222|issue:45542", + "issue:43012|issue:43577", + "issue:44410|issue:45478", + "issue:42175|issue:42222", + "issue:38617|issue:44220", + "issue:43874|issue:44261", + "issue:43097|issue:44493", + "issue:41628|issue:45071", + "issue:44263|issue:44485", + "issue:45071|issue:45542", + "issue:42673|issue:44977", + "issue:42673|issue:43906", + "issue:43012|issue:43408", + "issue:44263|issue:46172", + "issue:44360|issue:46172", + "issue:44484|issue:45478", + "issue:43441|issue:44829", + "issue:42371|issue:42831", + "issue:43931|issue:44038", + "issue:43335|issue:45071", + "issue:43097|issue:43452", + "issue:43874|issue:44263", + "issue:44933|issue:46231", + "issue:44117|issue:44987", + "issue:43408|issue:44368", + "issue:43976|issue:44945", + "issue:43986|issue:44484", + "issue:45440|issue:46172", + "issue:43097|issue:45588", + "issue:43064|issue:44315", + "issue:43097|issue:46050", + "issue:42673|issue:45200", + "issue:38617|issue:42757", + "issue:44062|issue:44117" + ], + "prepared_review_unit_hash": "5bbef32ea2fb55d41feb3d8ba479ae302a0f7ff09ddebeb063740bbf238145e6", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37842, + "estimated_input_tokens": 9461, + "estimated_eval_tokens": 19178 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No soft edge looks like a true duplicate; most pairs are only loosely related by subsystem, model family, or symptom type. I\u2019d anchor the cluster on issue:43097 as the clearest high-signal regression report, but there is no mergeable PR here.", + "confidence": 0.61, + "canonical_issue_reason": "issue:43097 is the clearest, most broadly impactful regression report in this set, with explicit references and a concrete API-break symptom.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43097 is the strongest representative issue: it is specific, reproducible, and likely to attract the most relevant fix context.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42222", + "right": "issue:45542", + "accept": false, + "reason": "Different failures: vitpose model breakage vs missing TF backend wiring." + }, + { + "left": "issue:43012", + "right": "issue:43577", + "accept": false, + "reason": "Both involve dtype/precision, but the model paths and bugs differ." + }, + { + "left": "issue:44410", + "right": "issue:45478", + "accept": false, + "reason": "Different model families and different loader failures; not the same bug." + }, + { + "left": "issue:42175", + "right": "issue:42222", + "accept": false, + "reason": "Packaging/backend dependency issue vs a specific model regression." + }, + { + "left": "issue:38617", + "right": "issue:44220", + "accept": false, + "reason": "Unrelated import error versus audio feature extraction bug." + }, + { + "left": "issue:43874", + "right": "issue:44261", + "accept": false, + "reason": "Different GLM subcomponents and different failure modes." + }, + { + "left": "issue:43097", + "right": "issue:44493", + "accept": false, + "reason": "Config-removal warning is not the same as unexpected position-id key errors." + }, + { + "left": "issue:41628", + "right": "issue:45071", + "accept": false, + "reason": "Import failure for AutoImageProcessor vs a type-checking regression." + }, + { + "left": "issue:44263", + "right": "issue:44485", + "accept": false, + "reason": "Tokenizer/indexer split bug and RoPE implementation bug are distinct." + }, + { + "left": "issue:45071", + "right": "issue:45542", + "accept": false, + "reason": "Type checking regression is unrelated to TF backend installation problems." + }, + { + "left": "issue:42673", + "right": "issue:44977", + "accept": false, + "reason": "VRAM leak in threaded loading is not the flash-attention generation bug." + }, + { + "left": "issue:42673", + "right": "issue:43906", + "accept": false, + "reason": "No evidence they describe the same underlying bug." + }, + { + "left": "issue:43012", + "right": "issue:43408", + "accept": false, + "reason": "Different model-specific warnings with different causes." + }, + { + "left": "issue:44263", + "right": "issue:46172", + "accept": false, + "reason": "Different tensor-shape/indexing issues in different model code paths." + }, + { + "left": "issue:44360", + "right": "issue:46172", + "accept": false, + "reason": "Activation omission and dtype/transpose bug are unrelated." + }, + { + "left": "issue:44484", + "right": "issue:45478", + "accept": false, + "reason": "Shard-size question vs Qwen3.5 model-loading failure." + }, + { + "left": "issue:43441", + "right": "issue:44829", + "accept": false, + "reason": "Both mention flash attention, but the models and symptoms differ." + }, + { + "left": "issue:42371", + "right": "issue:42831", + "accept": false, + "reason": "TF32 warning/config discussion is not the same as FineGrainedFP8 accuracy loss." + }, + { + "left": "issue:43931", + "right": "issue:44038", + "accept": false, + "reason": "Related model family, but different load-time errors and code paths." + }, + { + "left": "issue:43335", + "right": "issue:45071", + "accept": false, + "reason": "Sparse-layer config bug is unrelated to PretrainedConfig type checking." + }, + { + "left": "issue:43097", + "right": "issue:43452", + "accept": false, + "reason": "API removal warning and gguf loading breakage are different issues." + }, + { + "left": "issue:43874", + "right": "issue:44263", + "accept": false, + "reason": "Different GLM bugs: missing method vs incorrect split outputs." + }, + { + "left": "issue:44933", + "right": "issue:46231", + "accept": false, + "reason": "Missing import and documentation typo are not duplicates." + }, + { + "left": "issue:44117", + "right": "issue:44987", + "accept": false, + "reason": "Tokenizer mapping assumption bug versus a specific model loading failure." + }, + { + "left": "issue:43408", + "right": "issue:44368", + "accept": false, + "reason": "Different warning text and different model/config contexts." + }, + { + "left": "issue:43976", + "right": "issue:44945", + "accept": false, + "reason": "Python version compatibility issue vs pipeline-parallelism output corruption." + }, + { + "left": "issue:43986", + "right": "issue:44484", + "accept": false, + "reason": "AutoProcessor torchvision dependency issue is unrelated to max shard size." + }, + { + "left": "issue:45440", + "right": "issue:46172", + "accept": false, + "reason": "Different DeepSeek model generations and different implementation bugs." + }, + { + "left": "issue:43097", + "right": "issue:45588", + "accept": false, + "reason": "Config-removal warning is unrelated to flash-attention AttributeError." + }, + { + "left": "issue:43064", + "right": "issue:44315", + "accept": false, + "reason": "Distributed optimizer-state bug is not the same as Liger kernel application." + }, + { + "left": "issue:43097", + "right": "issue:46050", + "accept": false, + "reason": "API removal warning and quantized save/load breakage are distinct." + }, + { + "left": "issue:42673", + "right": "issue:45200", + "accept": false, + "reason": "Threaded VRAM leak has nothing to do with Gemma 4 multimodal token IDs." + }, + { + "left": "issue:38617", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but for different symbols and modules." + }, + { + "left": "issue:44062", + "right": "issue:44117", + "accept": false, + "reason": "Tokenizer AddedToken construction bug is unrelated to tokenizer-mapping None handling." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 45, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:38617", + "issue:42617", + "issue:42890", + "issue:42907", + "issue:43012", + "issue:43066", + "issue:43097", + "issue:43208", + "issue:43502", + "issue:43550", + "issue:43792", + "issue:43825", + "issue:43873", + "issue:43906", + "issue:43931", + "issue:43986", + "issue:43992", + "issue:44038", + "issue:44117", + "issue:44162", + "issue:44222", + "issue:44261", + "issue:44351", + "issue:44368", + "issue:44410", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44514", + "issue:44655", + "issue:44704", + "issue:44938", + "issue:44964", + "issue:45103", + "issue:45290", + "issue:45341", + "issue:45372", + "issue:45375", + "issue:45478", + "issue:45561", + "issue:45588", + "issue:45712", + "issue:45753", + "issue:45859", + "issue:46123", + "issue:46172", + "issue:46178", + "issue:46298" + ], + "soft_pairs": [ + "issue:44410|issue:45375", + "issue:42890|issue:46298", + "issue:43931|issue:44410", + "issue:44038|issue:44410", + "issue:43502|issue:43992", + "issue:45341|issue:45859", + "issue:38617|issue:44351", + "issue:43097|issue:43873", + "issue:44261|issue:44485", + "issue:44964|issue:45341", + "issue:44704|issue:45103", + "issue:43986|issue:44117", + "issue:42617|issue:45341", + "issue:43986|issue:45372", + "issue:43066|issue:43906", + "issue:43986|issue:45290", + "issue:45341|issue:45561", + "issue:45341|issue:45588", + "issue:43208|issue:43550", + "issue:43986|issue:45341", + "issue:43992|issue:44484", + "issue:43012|issue:43550", + "issue:43825|issue:44655", + "issue:43012|issue:44368", + "issue:45712|issue:46298", + "issue:42907|issue:44655", + "issue:42907|issue:45753", + "issue:43097|issue:46298", + "issue:44162|issue:46172", + "issue:43792|issue:46123", + "issue:44938|issue:46178", + "issue:44483|issue:44514", + "issue:44222|issue:45478" + ], + "prepared_review_unit_hash": "6eb95c7b03f7ba1c3987155558fff7edb69417820347e94a9b840b90506edf9b", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37333, + "estimated_input_tokens": 9334, + "estimated_eval_tokens": 18924 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: the soft pairs mostly share only broad subsystem keywords (imports, loading, save_pretrained, testing, Qwen/processor paths) rather than the same concrete bug. I rejected all soft edges and found no clear canonical issue/PR within this set.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44410", + "right": "issue:45375", + "accept": false, + "reason": "Both are Qwen3-family config/model issues, but they describe different missing fields/projection problems and likely different code paths." + }, + { + "left": "issue:42890", + "right": "issue:46298", + "accept": false, + "reason": "One is a seeding-related test flake; the other is a noisy logging message. Same testing area only at a very broad level." + }, + { + "left": "issue:43931", + "right": "issue:44410", + "accept": false, + "reason": "Model weight-shape mismatch for Qwen3-VL is not the same bug as missing attention projections in qwen3next." + }, + { + "left": "issue:44038", + "right": "issue:44410", + "accept": false, + "reason": "Qwen3-VL-Moe loading breakage and qwen3next missing projections are distinct model-specific failures." + }, + { + "left": "issue:43502", + "right": "issue:43992", + "accept": false, + "reason": "local_files_only causing network requests is a loader/network bug; missing embed_tokens.weight is a model checkpoint loading mismatch." + }, + { + "left": "issue:45341", + "right": "issue:45859", + "accept": false, + "reason": "A testing_utils bug and a Qwen3.5 rotary embedding CPU-offload bug are unrelated." + }, + { + "left": "issue:38617", + "right": "issue:44351", + "accept": false, + "reason": "Both are ImportErrors, but for different symbols and likely different release/compatibility regressions." + }, + { + "left": "issue:43097", + "right": "issue:43873", + "accept": false, + "reason": "Removed config flag documentation/behavior and quantization offloading issues are different problems." + }, + { + "left": "issue:44261", + "right": "issue:44485", + "accept": false, + "reason": "Different model internals: MLA layernorm epsilon handling vs GLM-5 RoPE implementation." + }, + { + "left": "issue:44964", + "right": "issue:45341", + "accept": false, + "reason": "Phi-4 multimodal loading failure is unrelated to a testing_utils bug." + }, + { + "left": "issue:44704", + "right": "issue:45103", + "accept": false, + "reason": "AutoProcessor kwargs forwarding and an auto_docstring annotation crash are separate code paths." + }, + { + "left": "issue:43986", + "right": "issue:44117", + "accept": false, + "reason": "Video-model AutoProcessor crash without torchvision and TOKENIZER_MAPPING_NAMES returning None are different loader failures." + }, + { + "left": "issue:42617", + "right": "issue:45341", + "accept": false, + "reason": "Running 3d_parallel.py and a testing_utils bug do not describe the same underlying defect." + }, + { + "left": "issue:43986", + "right": "issue:45372", + "accept": false, + "reason": "Both affect processor loading, but one is a missing torchvision path and the other is a missing mistral_common symbol." + }, + { + "left": "issue:43066", + "right": "issue:43906", + "accept": false, + "reason": "Wrong tokenizer decoder type in v5 is not clearly the same as an isolated reproduction of another issue." + }, + { + "left": "issue:43986", + "right": "issue:45290", + "accept": false, + "reason": "Processor loading without torchvision and apply_chat_template tool-call handling are distinct failures." + }, + { + "left": "issue:45341", + "right": "issue:45561", + "accept": false, + "reason": "A helper bug in testing_utils and an xdist race on captured_info.txt are not the same bug." + }, + { + "left": "issue:45341", + "right": "issue:45588", + "accept": false, + "reason": "Testing utils vs flash_attention sinkless-model crash: different modules and failure modes." + }, + { + "left": "issue:43208", + "right": "issue:43550", + "accept": false, + "reason": "xLSTM training bugs and a Bamba SDPA torch.compile failure are unrelated model/runtime issues." + }, + { + "left": "issue:43986", + "right": "issue:45341", + "accept": false, + "reason": "Processor/torchvision loading failure is unrelated to the testing_utils issue despite both being utility-level bugs." + }, + { + "left": "issue:43992", + "right": "issue:44484", + "accept": false, + "reason": "Missing checkpoint weights and a question about save_pretrained sharding size are not the same bug." + }, + { + "left": "issue:43012", + "right": "issue:43550", + "accept": false, + "reason": "A bfloat16 compile warning and a Bamba torch.compile SDPA failure are different PyTorch behaviors." + }, + { + "left": "issue:43825", + "right": "issue:44655", + "accept": false, + "reason": "A misleading pipeline error message and inability to save Pipeline objects are separate pipeline issues." + }, + { + "left": "issue:43012", + "right": "issue:44368", + "accept": false, + "reason": "The bfloat16 precision warning is unrelated to a tie_word_embeddings warning from fine-tuning." + }, + { + "left": "issue:45712", + "right": "issue:46298", + "accept": false, + "reason": "Leftover dummy classes in dummy_pt_objects.py and a noisy logging message are not the same defect." + }, + { + "left": "issue:42907", + "right": "issue:44655", + "accept": false, + "reason": "Saving dequantized models and saving Pipeline objects both involve serialization, but they are different code paths and failures." + }, + { + "left": "issue:42907", + "right": "issue:45753", + "accept": false, + "reason": "Model save failure for dequantized Ministrals/Devstrals is unrelated to an infinite-loop inference bug." + }, + { + "left": "issue:43097", + "right": "issue:46298", + "accept": false, + "reason": "A removed config behavior and noisy logging do not point to the same underlying issue." + }, + { + "left": "issue:44162", + "right": "issue:46172", + "accept": false, + "reason": "ESM2 breakage and a DeepSeek-v4 dtype/transpose bug are unrelated model-specific regressions." + }, + { + "left": "issue:43792", + "right": "issue:46123", + "accept": false, + "reason": "Whisper-large-v2 runtime failure and MaskGenerationPipeline partial-batch loss are different pipeline/model bugs." + }, + { + "left": "issue:44938", + "right": "issue:46178", + "accept": false, + "reason": "Python 3.14 import/load failure and tensorflow_text backend mapping error are different environment/backend issues." + }, + { + "left": "issue:44483", + "right": "issue:44514", + "accept": false, + "reason": "Chat-completions request rejection and apply_chat_template batching crash are separate API-path bugs." + }, + { + "left": "issue:44222", + "right": "issue:45478", + "accept": false, + "reason": "FP8 save_pretrained for moe models and a Qwen3.5 MoE from_pretrained error are not the same concrete failure." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 46, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:36010", + "issue:36683", + "issue:38617", + "issue:42490", + "issue:42757", + "issue:42994", + "issue:43012", + "issue:43064", + "issue:43066", + "issue:43208", + "issue:43299", + "issue:43352", + "issue:43475", + "issue:43502", + "issue:43525", + "issue:43526", + "issue:43611", + "issue:43618", + "issue:43673", + "issue:43761", + "issue:43824", + "issue:43867", + "issue:43881", + "issue:43901", + "issue:43906", + "issue:43937", + "issue:44038", + "issue:44230", + "issue:44242", + "issue:44380", + "issue:44483", + "issue:44485", + "issue:44625", + "issue:44655", + "issue:44683", + "issue:45216", + "issue:45276", + "issue:45290", + "issue:45600", + "issue:45698", + "issue:45753", + "issue:46077", + "issue:46082", + "issue:46123", + "issue:46132", + "issue:46172", + "issue:46209", + "issue:46298" + ], + "soft_pairs": [ + "issue:43208|issue:43526", + "issue:43012|issue:44683", + "issue:42490|issue:43012", + "issue:42757|issue:43824", + "issue:45600|issue:46298", + "issue:38617|issue:43502", + "issue:45698|issue:46172", + "issue:44038|issue:44230", + "issue:43673|issue:46123", + "issue:44483|issue:45290", + "issue:43064|issue:44380", + "issue:43618|issue:43761", + "issue:43299|issue:43867", + "issue:43066|issue:46209", + "issue:36683|issue:43475", + "issue:42994|issue:44655", + "issue:45276|issue:46077", + "issue:43901|issue:44655", + "issue:44625|issue:46077", + "issue:43352|issue:43937", + "issue:43611|issue:44655", + "issue:36010|issue:42757", + "issue:43867|issue:45216", + "issue:46077|issue:46082", + "issue:43502|issue:44485", + "issue:36683|issue:43525", + "issue:43618|issue:45216", + "issue:36683|issue:44242", + "issue:44380|issue:46132", + "issue:36683|issue:43881", + "issue:43906|issue:45753", + "issue:43299|issue:45753" + ], + "prepared_review_unit_hash": "f12a4213e247b18bdd010f3fd71cec91345de9e9cf12b9cf7f29151db0308b30", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 36833, + "estimated_input_tokens": 9209, + "estimated_eval_tokens": 18674 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is not a true duplicate set: the items span unrelated Transformers regressions, import failures, config/schema issues, saving/loading bugs, and model-specific behavior changes. None of the soft edges look like the same concrete bug or change.", + "confidence": 0.95, + "canonical_issue_reason": "No single canonical issue cleanly represents the cluster; the topics are too heterogeneous. If forced to pick one representative, issue 45698 is the broadest active save/load regression, but it does not cover the rest.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45698 is the best representative only in a loose sense because it is an active, broadly described save/load regression with multiple references, but it is not a duplicate umbrella for the other items.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43208", + "right": "issue:43526", + "accept": false, + "reason": "Different bugs: xLSTM training blockers vs a BeitImageProcessorFast label-reduction bug." + }, + { + "left": "issue:43012", + "right": "issue:44683", + "accept": false, + "reason": "Different failures: a PyTorch warning during bfloat16 compilation vs a hard compile error in flex_attention on torch>=2.9." + }, + { + "left": "issue:42490", + "right": "issue:43012", + "accept": false, + "reason": "Unrelated symptoms: local-path slash import warning vs bfloat16 precision warning." + }, + { + "left": "issue:42757", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but for different symbols and modules; not the same underlying breakage." + }, + { + "left": "issue:45600", + "right": "issue:46298", + "accept": false, + "reason": "CI break from removed Sam3LiteText configs vs a noisy logging message; not the same bug." + }, + { + "left": "issue:38617", + "right": "issue:43502", + "accept": false, + "reason": "ImportError for layer_type_validation vs API requests despite local_files_only; unrelated code paths." + }, + { + "left": "issue:45698", + "right": "issue:46172", + "accept": false, + "reason": "Save/load custom module regression vs a deepseek-v4 tensor/dtype transpose issue; different code paths." + }, + { + "left": "issue:44038", + "right": "issue:44230", + "accept": false, + "reason": "Both mention Qwen3-VL-Moe, but one is a loading regression and the other is fp8 inference support; not the same bug." + }, + { + "left": "issue:43673", + "right": "issue:46123", + "accept": false, + "reason": "Generation cache missing in chunked prefill vs batch-finalization dropping results in MaskGenerationPipeline; unrelated." + }, + { + "left": "issue:44483", + "right": "issue:45290", + "accept": false, + "reason": "Both touch chat APIs, but one is a v5.3 /v1/chat/completions request rejection and the other is a chat-template crash with tool calls." + }, + { + "left": "issue:43064", + "right": "issue:44380", + "accept": false, + "reason": "Distributed optimizer-state corruption vs GPT2 attention-scaling being ignored in SDPA/FlashAttention; different problems." + }, + { + "left": "issue:43618", + "right": "issue:43761", + "accept": false, + "reason": "Both are CLIP v5 regressions, but one is missing attentions and the other missing hidden_states; too different to merge as one bug." + }, + { + "left": "issue:43299", + "right": "issue:43867", + "accept": false, + "reason": "Qwen3VL-Moe loading regression vs a state_dict-sorting load error; not the same underlying failure." + }, + { + "left": "issue:43066", + "right": "issue:46209", + "accept": false, + "reason": "Wrong tokenizer decoder type vs FP8 inference on B200; unrelated issues." + }, + { + "left": "issue:36683", + "right": "issue:43475", + "accept": false, + "reason": "Different missing attributes on different model objects: Gemma3Config.vocab_size vs Sam3VisionEncoderOutput.fpn_position_embeddings." + }, + { + "left": "issue:42994", + "right": "issue:44655", + "accept": false, + "reason": "Quantized model saving failure vs Pipeline save_pretrained serialization; related area but different concrete bugs." + }, + { + "left": "issue:45276", + "right": "issue:46077", + "accept": false, + "reason": "Gemma4 embedding resize propagation bug vs Gemma4 attention-mask construction bug; different code paths." + }, + { + "left": "issue:43901", + "right": "issue:44655", + "accept": false, + "reason": "Docs/behavior mismatch for return_all_scores vs inability to save Pipeline objects; unrelated." + }, + { + "left": "issue:44625", + "right": "issue:46077", + "accept": false, + "reason": "Qwen3.5 num_labels propagation issue vs Gemma4 attention masking; different models and bugs." + }, + { + "left": "issue:43352", + "right": "issue:43937", + "accept": false, + "reason": "Flash Attention 2 support gap in Nemotron vs invalid GenerationConfig in GLM-5; unrelated." + }, + { + "left": "issue:43611", + "right": "issue:44655", + "accept": false, + "reason": "Loading with base_model_prefix regression vs Pipeline save_pretrained failure; different lifecycle stages and failures." + }, + { + "left": "issue:36010", + "right": "issue:42757", + "accept": false, + "reason": "Different import errors from different packages and symbols; not the same duplicate." + }, + { + "left": "issue:43867", + "right": "issue:45216", + "accept": false, + "reason": "State_dict-sorting load error vs incorrect checkpoint saved by save_pretrained; related to serialization but not the same bug." + }, + { + "left": "issue:46077", + "right": "issue:46082", + "accept": false, + "reason": "Gemma4 causal-mask regression vs LlamaConfig head_dim validation; unrelated." + }, + { + "left": "issue:43502", + "right": "issue:44485", + "accept": false, + "reason": "Offline/local-files-only network leak vs GLM-5 RoPE implementation discussion; no shared concrete bug." + }, + { + "left": "issue:36683", + "right": "issue:43525", + "accept": false, + "reason": "Two different missing config attributes (vocab_size vs pad_token_id); same pattern, but not the same underlying bug." + }, + { + "left": "issue:36683", + "right": "issue:44242", + "accept": false, + "reason": "Gemma3Config attribute error vs missing load-balancing loss when output_router_logits=False; unrelated." + }, + { + "left": "issue:44380", + "right": "issue:46132", + "accept": false, + "reason": "GPT2 attention scaling ignored in backend kernels vs AttentionInterface.register changing function behavior; different features and bugs." + }, + { + "left": "issue:36683", + "right": "issue:43881", + "accept": false, + "reason": "Gemma3Config missing vocab_size vs glm-4v-9b loading failed; both are load/config problems but not the same one." + }, + { + "left": "issue:43906", + "right": "issue:45753", + "accept": false, + "reason": "An isolated reproduction of another issue vs an infinite-loop bug on a specific image; insufficient evidence of the same underlying defect." + }, + { + "left": "issue:43299", + "right": "issue:45753", + "accept": false, + "reason": "Qwen3VL-Moe loading regression vs Qwen3_5 infinite loop on one image; different symptoms and code paths." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 47, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:28282", + "issue:33453", + "issue:36296", + "issue:36683", + "issue:41669", + "issue:42548", + "issue:42890", + "issue:43064", + "issue:43097", + "issue:43352", + "issue:43441", + "issue:43502", + "issue:43575", + "issue:43576", + "issue:43618", + "issue:43867", + "issue:43976", + "issue:43992", + "issue:44062", + "issue:44112", + "issue:44117", + "issue:44222", + "issue:44246", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44380", + "issue:44492", + "issue:44655", + "issue:44749", + "issue:44855", + "issue:44857", + "issue:44938", + "issue:44964", + "issue:44977", + "issue:45125", + "issue:45310", + "issue:45357", + "issue:45375", + "issue:45440", + "issue:45478", + "issue:45479", + "issue:45561", + "issue:45600", + "issue:46132", + "issue:46202", + "issue:46285" + ], + "soft_pairs": [ + "issue:43352|issue:43976", + "issue:43992|issue:45310", + "issue:43441|issue:44380", + "issue:43352|issue:45478", + "issue:44355|issue:44855", + "issue:43867|issue:46202", + "issue:44380|issue:45479", + "issue:43867|issue:45375", + "issue:43352|issue:44380", + "issue:36683|issue:44117", + "issue:43618|issue:44380", + "issue:41669|issue:44246", + "issue:43097|issue:43352", + "issue:42890|issue:45561", + "issue:43064|issue:44492", + "issue:43976|issue:44938", + "issue:43867|issue:45357", + "issue:43992|issue:45478", + "issue:36296|issue:43575", + "issue:43064|issue:44655", + "issue:28282|issue:33453", + "issue:43618|issue:46285", + "issue:43618|issue:44857", + "issue:44655|issue:44938", + "issue:43618|issue:44977", + "issue:44360|issue:45440", + "issue:43502|issue:44062", + "issue:42890|issue:44964", + "issue:42890|issue:44112", + "issue:42890|issue:45600", + "issue:44222|issue:45125", + "issue:43576|issue:46132", + "issue:44355|issue:44749", + "issue:45125|issue:45478", + "issue:42548|issue:44351" + ], + "prepared_review_unit_hash": "c211c314d48ef32cd94d158030c34668bac6708acf565e33931c77cc453b445e", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 35, + "serialized_chars": 37289, + "estimated_input_tokens": 9323, + "estimated_eval_tokens": 18902 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is a loose collection of Transformers issue reports spanning unrelated regressions (imports, tokenizer/model loading, flash-attention, tensor parallelism, save/load, CI, and Python-version compatibility). None of the soft pairs look like the same underlying defect, so there is no clear canonical duplicate anchor.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43352", + "right": "issue:43976", + "accept": false, + "reason": "Both mention model/runtime breakage, but one is a FlashAttention support issue for NemotronH and the other is a Python-version compatibility problem for Transformers 5.1.0." + }, + { + "left": "issue:43992", + "right": "issue:45310", + "accept": false, + "reason": "Different failures: UMT5Encoder missing embed_tokens.weight vs Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:43441", + "right": "issue:44380", + "accept": false, + "reason": "Both involve attention backends, but one is Ministral-3 failing with FlashAttention while the other is GPT2 ignoring attention scaling config under SDPA/FlashAttention; different code paths." + }, + { + "left": "issue:43352", + "right": "issue:45478", + "accept": false, + "reason": "Same broad FlashAttention theme, but distinct model families and reported symptoms; not the same concrete bug." + }, + { + "left": "issue:44355", + "right": "issue:44855", + "accept": false, + "reason": "Both are import-time failures, but one is about compiled Python files and the other is a Python 3.13 torch.jit.script parsing/IndentationError issue." + }, + { + "left": "issue:43867", + "right": "issue:46202", + "accept": false, + "reason": "Both are model loading/roundtrip-related, but the specific failures differ: sorted state_dict loading vs Gemma pipeline roundtrip on the \u2581 token." + }, + { + "left": "issue:44380", + "right": "issue:45479", + "accept": false, + "reason": "Attention/ML terms overlap, but one is backend-specific config being ignored and the other is a sequence-classification loss bug." + }, + { + "left": "issue:43867", + "right": "issue:45375", + "accept": false, + "reason": "A load-model issue vs a missing vision config field; unrelated root causes." + }, + { + "left": "issue:43352", + "right": "issue:44380", + "accept": false, + "reason": "Both reference FlashAttention, but one is lack of support and the other is ignored scaling config; not mergeable as one fix." + }, + { + "left": "issue:36683", + "right": "issue:44117", + "accept": false, + "reason": "Different objects and failures: Gemma3Config missing vocab_size vs tokenizer mapping sometimes returning None." + }, + { + "left": "issue:43618", + "right": "issue:44380", + "accept": false, + "reason": "Different model API regressions: CLIPOutput attentions missing vs GPT2 attention scaling ignored under certain backends." + }, + { + "left": "issue:41669", + "right": "issue:44246", + "accept": false, + "reason": "Import-* cleanup/performance discussion vs intermittent slow import of transformers; related only at a very high level." + }, + { + "left": "issue:43097", + "right": "issue:43352", + "accept": false, + "reason": "One is an API removal/change notice, the other is a model FlashAttention support bug; not the same defect." + }, + { + "left": "issue:42890", + "right": "issue:45561", + "accept": false, + "reason": "A flaky integration test missing set_seed() vs a pytest-xdist race in captured_info.txt; both test-related but different issues." + }, + { + "left": "issue:43064", + "right": "issue:44492", + "accept": false, + "reason": "Training/FSDP optimizer-state corruption vs a typo in cache strategies; unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44938", + "accept": false, + "reason": "Both are Python-version compatibility complaints, but they target different versions and likely different breakpoints." + }, + { + "left": "issue:43867", + "right": "issue:45357", + "accept": false, + "reason": "Load-time state_dict ordering problem vs save_pretrained writing incorrect visual encoder keys." + }, + { + "left": "issue:43992", + "right": "issue:45478", + "accept": false, + "reason": "Different models and failures; no shared code path beyond both being from_pretrained reports." + }, + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Tensor-parallel training bug vs OOM when loading a large Qwen model with tp; related area, different underlying problems." + }, + { + "left": "issue:43064", + "right": "issue:44655", + "accept": false, + "reason": "Wrong optimizer state under v5/FSDP2/PEFT vs inability to save Pipeline objects; both serialization/training-adjacent but distinct." + }, + { + "left": "issue:28282", + "right": "issue:33453", + "accept": false, + "reason": "ImportError about missing PyTorch vs tokenizer loading regression; different subsystems and symptoms." + }, + { + "left": "issue:43618", + "right": "issue:46285", + "accept": false, + "reason": "Both concern CLIP model structure, but one is output attentions and the other is inconsistent text model structure; not the same bug." + }, + { + "left": "issue:43618", + "right": "issue:44857", + "accept": false, + "reason": "CLIPOutput attentions regression vs LwDetrImageLoss crashing under float16 AMP; unrelated." + }, + { + "left": "issue:44655", + "right": "issue:44938", + "accept": false, + "reason": "Saving Pipeline objects vs Python 3.14 import/load failure; different failure modes." + }, + { + "left": "issue:43618", + "right": "issue:44977", + "accept": false, + "reason": "Different attention-related reports on different models; one is missing attentions in CLIPOutput, the other is Qwen3.5 generation failing with flash-attention." + }, + { + "left": "issue:44360", + "right": "issue:45440", + "accept": false, + "reason": "A DSA/indexer discussion vs DeepseekV3MoE implementation divergence; no shared bug." + }, + { + "left": "issue:43502", + "right": "issue:44062", + "accept": false, + "reason": "Local-files-only still making API requests vs AddedToken constructor argument conflict; unrelated." + }, + { + "left": "issue:42890", + "right": "issue:44964", + "accept": false, + "reason": "Flaky test seeding issue vs inability to load a specific multimodal model on latest Transformers." + }, + { + "left": "issue:42890", + "right": "issue:44112", + "accept": false, + "reason": "Both are test failures, but one is missing set_seed coverage and the other is a stale device-override CI failure." + }, + { + "left": "issue:42890", + "right": "issue:45600", + "accept": false, + "reason": "Test flakiness vs auto_mappings referencing removed configs; different root causes." + }, + { + "left": "issue:44222", + "right": "issue:45125", + "accept": false, + "reason": "FP8 save_pretrained bug for MoE vs missing _tp_plan for tensor parallelism; same model family, different functionality." + }, + { + "left": "issue:43576", + "right": "issue:46132", + "accept": false, + "reason": "Broken transformers env command vs AttentionInterface.register behavior change; unrelated command/API issues." + }, + { + "left": "issue:44355", + "right": "issue:44749", + "accept": false, + "reason": "Compiled-file import errors vs a performance regression after upgrading Transformers; different problems." + }, + { + "left": "issue:45125", + "right": "issue:45478", + "accept": false, + "reason": "Missing tensor-parallel plan vs Qwen3.5 MoE from_pretrained error; likely adjacent but not the same defect." + }, + { + "left": "issue:42548", + "right": "issue:44351", + "accept": false, + "reason": "Top-level import errors, but different missing symbols: PreTrainedModel vs HybridCache." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 48, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:30333", + "issue:30990", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:36296", + "issue:38617", + "issue:41628", + "issue:42548", + "issue:42757", + "issue:42831", + "issue:43010", + "issue:43299", + "issue:43502", + "issue:43526", + "issue:43576", + "issue:43723", + "issue:43824", + "issue:43827", + "issue:43874", + "issue:43901", + "issue:43976", + "issue:44038", + "issue:44222", + "issue:44230", + "issue:44273", + "issue:44315", + "issue:44355", + "issue:44360", + "issue:44368", + "issue:44387", + "issue:44556", + "issue:44623", + "issue:44655", + "issue:44908", + "issue:44938", + "issue:44961", + "issue:45003", + "issue:45071", + "issue:45125", + "issue:45230", + "issue:45278", + "issue:45375", + "issue:45468", + "issue:45542", + "issue:45656", + "issue:45850", + "issue:46209" + ], + "soft_pairs": [ + "issue:44387|issue:44556", + "issue:30333|issue:33453", + "issue:45125|issue:45375", + "issue:43010|issue:45278", + "issue:30990|issue:31795", + "issue:43723|issue:44556", + "issue:43502|issue:44222", + "issue:42831|issue:43874", + "issue:43010|issue:45850", + "issue:43010|issue:44368", + "issue:43502|issue:44961", + "issue:30333|issue:36296", + "issue:44355|issue:45071", + "issue:43827|issue:44961", + "issue:44355|issue:44655", + "issue:44315|issue:44360", + "issue:33453|issue:34689", + "issue:43299|issue:44556", + "issue:42757|issue:43576", + "issue:33453|issue:36296", + "issue:43576|issue:45071", + "issue:43010|issue:45656", + "issue:43576|issue:43824", + "issue:31795|issue:38617", + "issue:44961|issue:45468", + "issue:44222|issue:44230", + "issue:43010|issue:43976", + "issue:43576|issue:44938", + "issue:44908|issue:44961", + "issue:44315|issue:46209", + "issue:36296|issue:43901", + "issue:41628|issue:42548", + "issue:36296|issue:43526", + "issue:44623|issue:45230", + "issue:44273|issue:45542", + "issue:44038|issue:45125", + "issue:44273|issue:45003" + ], + "prepared_review_unit_hash": "4d11498e82557383630a78f9e85bf2f6e27316f3038cfbfdc497b33deff10384", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 37, + "serialized_chars": 38083, + "estimated_input_tokens": 9521, + "estimated_eval_tokens": 19298 + }, + "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 issues: import/load regressions, FP8/MoE bugs, tensor-parallel problems, docs mismatches, and unrelated runtime errors. None of the soft pairs look like the same concrete underlying bug or change.", + "confidence": 0.66, + "canonical_issue_reason": "Issue 45278 is the broadest umbrella-style report here (v4.57 -> v5.5 import failures), so it is the closest canonical issue among otherwise heterogeneous reports.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45278 is the best representative issue because it is the most general v5 regression report and overlaps conceptually with several other loading/import complaints, even though the cluster is not a true duplicate set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44387", + "right": "issue:44556", + "accept": false, + "reason": "Both are v5 regressions, but one is an int4/OOM memory issue and the other is a checkpoint reload/loading failure; different bugs and code paths." + }, + { + "left": "issue:30333", + "right": "issue:33453", + "accept": false, + "reason": "MLFlow failed-job reporting and tokenizer loading regression are unrelated." + }, + { + "left": "issue:45125", + "right": "issue:45375", + "accept": false, + "reason": "Both involve Qwen3_5Moe, but one is tensor-parallel plan support and the other is a config field being dropped; not the same bug." + }, + { + "left": "issue:43010", + "right": "issue:45278", + "accept": false, + "reason": "No-grad decoration for cache/layer updates is unrelated to broad import errors after upgrading to v5.5.0." + }, + { + "left": "issue:30990", + "right": "issue:31795", + "accept": false, + "reason": "Sentence Transformers hanging on load is not the same as a documentation issue about model.forward arguments." + }, + { + "left": "issue:43723", + "right": "issue:44556", + "accept": false, + "reason": "Both are loading problems, but one is AutoTokenizer loading in v5 and the other is a checkpoint reload regression; too different to merge." + }, + { + "left": "issue:43502", + "right": "issue:44222", + "accept": false, + "reason": "Local-files-only network leakage and FP8 save_pretrained issues are unrelated." + }, + { + "left": "issue:42831", + "right": "issue:43874", + "accept": false, + "reason": "An FP8 accuracy issue is not the same as a missing image-processor method causing an AttributeError." + }, + { + "left": "issue:43010", + "right": "issue:45850", + "accept": false, + "reason": "A specific no-grad bug is not the same as a broad complaint that v5 breaks many models." + }, + { + "left": "issue:43010", + "right": "issue:44368", + "accept": false, + "reason": "Cache update gradients and a tie_word_embeddings warning are unrelated." + }, + { + "left": "issue:43502", + "right": "issue:44961", + "accept": false, + "reason": "local_files_only API requests and an unrelated 'racoon' issue do not match." + }, + { + "left": "issue:30333", + "right": "issue:36296", + "accept": false, + "reason": "MLFlow reporting and tensor-parallel training are distinct issues." + }, + { + "left": "issue:44355", + "right": "issue:45071", + "accept": false, + "reason": "Compiled Python file errors and PretrainedConfig type checking are different regressions." + }, + { + "left": "issue:43827", + "right": "issue:44961", + "accept": false, + "reason": "Docs still referencing pipeline() and an unrelated 'racoon' issue are not duplicates." + }, + { + "left": "issue:44355", + "right": "issue:44655", + "accept": false, + "reason": "Compiled-file errors and saving Pipeline objects are separate save/load problems." + }, + { + "left": "issue:44315", + "right": "issue:44360", + "accept": false, + "reason": "Liger Kernel application and a DSA indexer ReLU complaint are unrelated." + }, + { + "left": "issue:33453", + "right": "issue:34689", + "accept": false, + "reason": "Tokenizer loading regression and Llama 3.2 model-loading failure are both load-related but not the same underlying bug." + }, + { + "left": "issue:43299", + "right": "issue:44556", + "accept": false, + "reason": "Qwen3VL MoE loading breakage and generic checkpoint reload breakage are different failures." + }, + { + "left": "issue:42757", + "right": "issue:43576", + "accept": false, + "reason": "A missing huggingface_hub import and the transformers env command bug are unrelated." + }, + { + "left": "issue:33453", + "right": "issue:36296", + "accept": false, + "reason": "Tokenizer loading regression and tensor-parallel training bug do not share the same code path." + }, + { + "left": "issue:43576", + "right": "issue:45071", + "accept": false, + "reason": "Broken env CLI and PretrainedConfig type checking are unrelated." + }, + { + "left": "issue:43010", + "right": "issue:45656", + "accept": false, + "reason": "Cache/layer update no-grad handling is not the same as double optimizer stepping under deepspeed." + }, + { + "left": "issue:43576", + "right": "issue:43824", + "accept": false, + "reason": "Broken env CLI and a missing Qwen2_5_VLForConditionalGeneration import are different issues." + }, + { + "left": "issue:31795", + "right": "issue:38617", + "accept": false, + "reason": "Documentation confusion in model.forward and an ImportError for layer_type_validation are unrelated." + }, + { + "left": "issue:44961", + "right": "issue:45468", + "accept": false, + "reason": "An empty/placeholder issue and a Gemma-4 audio positional-encoding bug are not duplicates." + }, + { + "left": "issue:44222", + "right": "issue:44230", + "accept": false, + "reason": "Both mention FP8/MoE, but one is save_pretrained behavior and the other is inference support; different concrete bugs." + }, + { + "left": "issue:43010", + "right": "issue:43976", + "accept": false, + "reason": "Cache update no-grad behavior is unrelated to Python 3.9/3.10 compatibility." + }, + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Lazy loading problems and a TensorBoard-without-TensorFlow backend error are not the same bug." + }, + { + "left": "issue:44038", + "right": "issue:45125", + "accept": false, + "reason": "Both are Qwen3-Moe-related, but one is a broad v5/Qwen3-VL-Moe bug and the other is a missing tensor-parallel plan; not enough to call them duplicates." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Lazy loading issues and sys.modules access in modeling_utils may be related conceptually, but they are not clearly the same concrete bug." + }, + { + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "processor.save_pretrained missing files and a generic bug report are not duplicates." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Same as above: possible relation to import/lazy-loading, but not a clear same-bug match." + }, + { + "left": "issue:44315", + "right": "issue:46209", + "accept": false, + "reason": "Liger Kernel application and FP8 inference on B200 are separate issues." + }, + { + "left": "issue:36296", + "right": "issue:43901", + "accept": false, + "reason": "Tensor-parallel training and docs about return_all_scores are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:42548", + "accept": false, + "reason": "Import errors for AutoImageProcessor and PreTrainedModel are both namespace issues, but not the same symbol or failure mode." + }, + { + "left": "issue:36296", + "right": "issue:43526", + "accept": false, + "reason": "Tensor-parallel training and BeitImageProcessorFast reduce_labels behavior are unrelated." + }, + { + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "Processor save_pretrained file omissions and a generic bug report are not the same issue." + }, + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Lazy loading and TensorBoard/TensorFlow backend configuration are different problems." + }, + { + "left": "issue:44038", + "right": "issue:45125", + "accept": false, + "reason": "Qwen3-VL-Moe breakage and missing _tp_plan are related only at a high level; not the same underlying defect." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Lazy loading and unsafe sys.modules access are too loosely related to merge as duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 49, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:31795", + "issue:33453", + "issue:41669", + "issue:42371", + "issue:42890", + "issue:43352", + "issue:43408", + "issue:43525", + "issue:43572", + "issue:43576", + "issue:43673", + "issue:43827", + "issue:43874", + "issue:43976", + "issue:44077", + "issue:44230", + "issue:44246", + "issue:44273", + "issue:44295", + "issue:44297", + "issue:44315", + "issue:44351", + "issue:44360", + "issue:44373", + "issue:44393", + "issue:44492", + "issue:44556", + "issue:44655", + "issue:44704", + "issue:44829", + "issue:44861", + "issue:44908", + "issue:44961", + "issue:45071", + "issue:45092", + "issue:45230", + "issue:45278", + "issue:45310", + "issue:45341", + "issue:45362", + "issue:45468", + "issue:45478", + "issue:45753", + "issue:45850", + "issue:46095", + "issue:46143", + "issue:46144", + "issue:46218" + ], + "soft_pairs": [ + "issue:44393|issue:45071", + "issue:44230|issue:46095", + "issue:44373|issue:44961", + "issue:44393|issue:45753", + "issue:43576|issue:44556", + "issue:41669|issue:43673", + "issue:43525|issue:45850", + "issue:44556|issue:45092", + "issue:43572|issue:45850", + "issue:45362|issue:45850", + "issue:41669|issue:46218", + "issue:42890|issue:44908", + "issue:44246|issue:44273", + "issue:43408|issue:44315", + "issue:44704|issue:44908", + "issue:43352|issue:44393", + "issue:44297|issue:45230", + "issue:43976|issue:44393", + "issue:44295|issue:45230", + "issue:43576|issue:44908", + "issue:44908|issue:45071", + "issue:41669|issue:44351", + "issue:41669|issue:45341", + "issue:45278|issue:45850", + "issue:43673|issue:44492", + "issue:46143|issue:46144", + "issue:43874|issue:45850", + "issue:44393|issue:45478", + "issue:44077|issue:44861", + "issue:43976|issue:46218", + "issue:43827|issue:44908", + "issue:31795|issue:33453", + "issue:41669|issue:44492", + "issue:44315|issue:44829", + "issue:42371|issue:44492", + "issue:43576|issue:44273", + "issue:44655|issue:44861", + "issue:44360|issue:45468", + "issue:45230|issue:45310", + "issue:44273|issue:44908", + "issue:43576|issue:46218", + "issue:43827|issue:44492" + ], + "prepared_review_unit_hash": "ed8235708529f42e87927a2891745a401a0bdea062e7d56ad15d4d95a17105b0", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 42, + "serialized_chars": 39625, + "estimated_input_tokens": 9907, + "estimated_eval_tokens": 20070 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly closed/open issue reports about Transformers v5 regressions: import/lazy-loading slowdowns, model-loading/config compatibility breakages, and API changes. No PRs are present in the cluster.", + "confidence": 0.78, + "canonical_issue_reason": "issue:45850 is the broadest umbrella report in the set: it explicitly frames a wide v5 regression affecting many models, making it the best single issue to anchor the cluster even though it is vague.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45850 is the best global representative because it is broad, current, and covers the same general family of v5 breakage complaints seen across the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44393", + "right": "issue:45071", + "accept": false, + "reason": "Different problems: Qwen3-VL multimodal output failure vs PretrainedConfig type checking regression." + }, + { + "left": "issue:44230", + "right": "issue:46095", + "accept": false, + "reason": "Unrelated model/features: fp8 inference support for Qwen3-VL/Qwen3.5 MoE vs DeepSeekV4 weight conversion question." + }, + { + "left": "issue:44373", + "right": "issue:44961", + "accept": false, + "reason": "Docstring bug vs a completely unrelated issue titled 'racoon'." + }, + { + "left": "issue:44393", + "right": "issue:45753", + "accept": false, + "reason": "Both are Qwen-family vision issues, but the failure modes differ materially: bbox hallucination/error vs infinite loop on a specific image." + }, + { + "left": "issue:43576", + "right": "issue:44556", + "accept": false, + "reason": "Both are v5 regressions, but one is the `transformers env` command and the other is checkpoint reload compatibility; not the same code path." + }, + { + "left": "issue:41669", + "right": "issue:43673", + "accept": false, + "reason": "Import-star cleanup/performance issue vs missing generation cache during chunked prefill; different subsystems and fixes." + }, + { + "left": "issue:43525", + "right": "issue:45850", + "accept": false, + "reason": "Specific `pad_token_id`/config attribute error vs a broad complaint about many models breaking in v5." + }, + { + "left": "issue:44556", + "right": "issue:45092", + "accept": false, + "reason": "Both concern v5 checkpoint loading, but one is generic reload failure and the other is remote-code/meta-init incompatibility; too different to treat as duplicates." + }, + { + "left": "issue:43572", + "right": "issue:45850", + "accept": false, + "reason": "Specific StableLmConfig `pad_token_idx` regression vs a broad v5 breakage umbrella issue." + }, + { + "left": "issue:45362", + "right": "issue:45850", + "accept": false, + "reason": "Qwen3.5 chat crash is a concrete model bug, not the same as a general v5 breakage report." + }, + { + "left": "issue:45278", + "right": "issue:45850", + "accept": false, + "reason": "Broad overlap in release complaints, but one is about import errors after upgrade and the other is a generic 'too many models' regression; not a single concrete bug." + }, + { + "left": "issue:44273", + "right": "issue:44908", + "accept": false, + "reason": "Lazy loading/import behavior vs inverse_sqrt scheduler kwargs; unrelated code paths." + }, + { + "left": "issue:44297", + "right": "issue:45230", + "accept": false, + "reason": "Tokenizer save_pretrained metadata mismatch vs an empty generic bug report; no evidence they are the same defect." + }, + { + "left": "issue:46143", + "right": "issue:46144", + "accept": false, + "reason": "Both involve RoFormer, but one is kwargs propagation and the other is attention-interface usage; related area, not the same bug." + }, + { + "left": "issue:43827", + "right": "issue:44492", + "accept": false, + "reason": "Documentation reference to removed pipelines vs a typo in cache strategies; unrelated." + }, + { + "left": "issue:44295", + "right": "issue:45230", + "accept": false, + "reason": "position_ids buffer access error vs generic bug report; no shared concrete failure." + }, + { + "left": "issue:44655", + "right": "issue:44908", + "accept": false, + "reason": "Pipeline save/load issue vs scheduler kwargs handling; unrelated." + }, + { + "left": "issue:41669", + "right": "issue:46218", + "accept": false, + "reason": "Import performance cleanup vs an import error for a specific Qwen3_5 symbol; same broad import area, but not the same bug." + }, + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "DSA indexer/ReLU discussion vs Gemma4Audio positional encoding bug; different models and failures." + }, + { + "left": "issue:45230", + "right": "issue:45310", + "accept": false, + "reason": "A generic bug report vs a specific Qwen3.5 MoE from_pretrained error; no concrete match." + }, + { + "left": "issue:44246", + "right": "issue:44273", + "accept": true, + "reason": "Both point at the same underlying import/lazy-loading regression: import transformers becoming slow and lazy loading not functioning properly." + }, + { + "left": "issue:43874", + "right": "issue:45850", + "accept": false, + "reason": "Specific GLM46V multimodal token counting failure vs a broad v5 breakage umbrella issue." + }, + { + "left": "issue:44077", + "right": "issue:44861", + "accept": false, + "reason": "Different defects in different utilities: patchtsmixer post_init handling vs tied-weight key attribute crash." + }, + { + "left": "issue:43976", + "right": "issue:46218", + "accept": false, + "reason": "Python-version compatibility complaint vs a specific missing import symbol; not the same issue." + }, + { + "left": "issue:43827", + "right": "issue:44908", + "accept": false, + "reason": "Docs referencing removed pipeline APIs vs scheduler kwargs handling; unrelated." + }, + { + "left": "issue:31795", + "right": "issue:33453", + "accept": false, + "reason": "Documentation confusion vs tokenizer loading regression; not the same underlying bug." + }, + { + "left": "issue:41669", + "right": "issue:44492", + "accept": false, + "reason": "Import-star performance regression vs a typo in cache-strategy docs; unrelated." + }, + { + "left": "issue:44315", + "right": "issue:44829", + "accept": false, + "reason": "Liger Kernel application bug vs degenerate training with flash_attention_3; different behavior and fixes." + }, + { + "left": "issue:42371", + "right": "issue:44492", + "accept": false, + "reason": "TF32 settings migration issue vs cache-strategy typo; unrelated." + }, + { + "left": "issue:43576", + "right": "issue:44273", + "accept": false, + "reason": "Broken `transformers env` command vs lazy loading not working properly; both v5 issues, but different code paths." + }, + { + "left": "issue:44655", + "right": "issue:44861", + "accept": false, + "reason": "Saving Pipeline objects vs `_tied_weight_keys` attribute crash; unrelated." + }, + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "Different problem domains and no shared code path." + }, + { + "left": "issue:45230", + "right": "issue:45310", + "accept": false, + "reason": "Generic report vs specific Qwen3.5 MoE loading failure; not duplicates." + }, + { + "left": "issue:44273", + "right": "issue:44908", + "accept": false, + "reason": "Lazy loading regression vs scheduler kwargs bug; unrelated." + }, + { + "left": "issue:43576", + "right": "issue:46218", + "accept": false, + "reason": "Broken env command vs missing Qwen3_5 import; not the same issue." + }, + { + "left": "issue:43827", + "right": "issue:44492", + "accept": false, + "reason": "Docs about removed pipelines vs cache-strategy typo; unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: the cluster is dominated by closed/open Transformers v5 regression issues, and issue:45850 is a reasonable broad umbrella representative. The soft-edge verdicts are conservative overall, with one plausible accept on the import/lazy-loading pair and the rest appropriately rejected based on distinct titles/code paths. No PRs are present." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 18, + "review_unit_index": 50, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-441", + "nodes": [ + "issue:30333", + "issue:31515", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:36296", + "issue:37934", + "issue:38617", + "issue:41669", + "issue:42371", + "issue:42913", + "issue:42936", + "issue:43066", + "issue:43240", + "issue:43298", + "issue:43352", + "issue:43502", + "issue:43519", + "issue:43525", + "issue:43572", + "issue:43606", + "issue:43668", + "issue:43704", + "issue:43717", + "issue:43813", + "issue:44031", + "issue:44230", + "issue:44261", + "issue:44263", + "issue:44351", + "issue:44485", + "issue:44623", + "issue:44861", + "issue:44908", + "issue:44998", + "issue:45290", + "issue:45310", + "issue:45468", + "issue:45850", + "issue:46018", + "issue:46095", + "issue:46144", + "issue:46172", + "issue:46218" + ], + "soft_pairs": [ + "issue:46018|issue:46172", + "issue:43519|issue:44485", + "issue:43519|issue:44908", + "issue:41669|issue:43704", + "issue:31795|issue:34689", + "issue:31515|issue:31795", + "issue:44485|issue:45468", + "issue:43704|issue:44485", + "issue:43298|issue:46095", + "issue:45850|issue:46218", + "issue:43502|issue:43519", + "issue:42371|issue:43704", + "issue:43519|issue:44623", + "issue:43519|issue:43704", + "issue:43606|issue:45468", + "issue:44263|issue:45468", + "issue:44261|issue:45468", + "issue:44351|issue:46218", + "issue:43352|issue:46218", + "issue:42913|issue:43066", + "issue:44908|issue:44998", + "issue:44861|issue:45290", + "issue:43525|issue:44861", + "issue:43704|issue:44998", + "issue:43572|issue:44861", + "issue:41669|issue:44998", + "issue:43066|issue:44998", + "issue:31515|issue:38617", + "issue:44230|issue:44623", + "issue:44998|issue:46218", + "issue:30333|issue:31515", + "issue:31515|issue:36296", + "issue:44230|issue:45310", + "issue:31515|issue:33453", + "issue:43298|issue:44230", + "issue:43717|issue:46144", + "issue:42936|issue:46018", + "issue:43668|issue:44031", + "issue:42936|issue:43240", + "issue:43240|issue:46144", + "issue:42936|issue:44031", + "issue:42936|issue:46144", + "issue:42936|issue:43813", + "issue:43240|issue:43668", + "issue:43668|issue:43813", + "issue:37934|issue:43668", + "issue:42936|issue:43668" + ], + "prepared_review_unit_hash": "04ff93f98507054db59ea683e1f2a01d441f814f00ef1ec4da933e5355fd3870", + "component_budget": { + "node_count": 441, + "item_count": 441, + "soft_pair_count": 1620, + "serialized_chars": 840341, + "estimated_input_tokens": 210086, + "estimated_eval_tokens": 420428 + }, + "budget": { + "node_count": 44, + "item_count": 44, + "soft_pair_count": 47, + "serialized_chars": 38701, + "estimated_input_tokens": 9676, + "estimated_eval_tokens": 19608 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a loose Transformers v5 regression/loading/tokenizer cluster with many superficially similar but distinct issues. I found no true duplicate soft edges; the candidates mostly share only broad subsystem or version-regression wording. No PRs are present.", + "confidence": 0.89, + "canonical_issue_reason": "issue:45850 is the broadest umbrella report for the v5 breakage/regression theme and can serve as the cluster anchor better than the more specific failures.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45850 has the widest scope and strongest discussion signal, so it is the best representative issue even though the cluster is too heterogeneous for true deduping.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46018", + "right": "issue:46172", + "accept": false, + "reason": "Different model families and bugs: DeepSeek-V4 gating vs a deepseek-v4 dtype/transpose problem." + }, + { + "left": "issue:43519", + "right": "issue:44485", + "accept": false, + "reason": "Qwen3VL timestamp math is unrelated to GLM-5 RoPE implementation." + }, + { + "left": "issue:43519", + "right": "issue:44908", + "accept": false, + "reason": "Processor timestamp calculation and scheduler kwargs handling are unrelated code paths." + }, + { + "left": "issue:41669", + "right": "issue:43704", + "accept": false, + "reason": "Import-star performance/footprint cleanup is unrelated to a VRAM leak in dataloader threads." + }, + { + "left": "issue:31795", + "right": "issue:34689", + "accept": false, + "reason": "Documentation confusion in forward args is not the same as a model-loading regression." + }, + { + "left": "issue:31515", + "right": "issue:31795", + "accept": false, + "reason": "Slow checkpoint loading and forward-doc confusion are different issues." + }, + { + "left": "issue:44485", + "right": "issue:45468", + "accept": false, + "reason": "Both touch positional/attention-style internals, but they are different models and different failures." + }, + { + "left": "issue:43704", + "right": "issue:44485", + "accept": false, + "reason": "VRAM leak in threaded loading is unrelated to GLM-5 RoPE math." + }, + { + "left": "issue:43298", + "right": "issue:46095", + "accept": false, + "reason": "Audio continuous batching support and DeepSeek-V4 weight conversion are different feature requests." + }, + { + "left": "issue:45850", + "right": "issue:46218", + "accept": false, + "reason": "Broad v5 breakage versus a specific import failure for Qwen3_5ForCausalLM." + }, + { + "left": "issue:43502", + "right": "issue:43519", + "accept": false, + "reason": "Offline file-only network calls are unrelated to Qwen3VL timestamp calculation." + }, + { + "left": "issue:42371", + "right": "issue:43704", + "accept": false, + "reason": "TF32 settings guidance and a dataloader-thread VRAM leak are unrelated." + }, + { + "left": "issue:43519", + "right": "issue:44623", + "accept": false, + "reason": "Processor timestamp math and processor.save_pretrained missing files are different bugs." + }, + { + "left": "issue:43519", + "right": "issue:43704", + "accept": false, + "reason": "Qwen3VL timestamp math and multi-thread VRAM leakage do not share the same failure." + }, + { + "left": "issue:43606", + "right": "issue:45468", + "accept": false, + "reason": "CPU offload device mismatch is unrelated to Gemma4 audio positional encoding." + }, + { + "left": "issue:44263", + "right": "issue:45468", + "accept": false, + "reason": "torch.split return-value handling in GLM-MoE is unrelated to Gemma4 audio positional encoding." + }, + { + "left": "issue:44261", + "right": "issue:45468", + "accept": false, + "reason": "Missing rms_norm_eps in MLA q_a_layernorm is a different model/config bug." + }, + { + "left": "issue:44351", + "right": "issue:46218", + "accept": false, + "reason": "Different missing imports for different symbols/modules." + }, + { + "left": "issue:43352", + "right": "issue:46218", + "accept": false, + "reason": "Flash Attention support gating and a Qwen3_5 import failure are unrelated." + }, + { + "left": "issue:42913", + "right": "issue:43066", + "accept": false, + "reason": "Both are tokenizer/v5 related, but one is behavior drift and the other is a decoder type bug." + }, + { + "left": "issue:44908", + "right": "issue:44998", + "accept": false, + "reason": "Scheduler kwargs bug and an off-topic issue are unrelated." + }, + { + "left": "issue:44861", + "right": "issue:45290", + "accept": false, + "reason": "Tied-weight key handling and chat-template tool-call crashing are different code paths." + }, + { + "left": "issue:43525", + "right": "issue:44861", + "accept": false, + "reason": "Missing pad_token_id in config is unrelated to tied-weight key extraction." + }, + { + "left": "issue:43704", + "right": "issue:44998", + "accept": false, + "reason": "VRAM leak and off-topic issue are unrelated." + }, + { + "left": "issue:43572", + "right": "issue:44861", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to tied-weight key crashes." + }, + { + "left": "issue:41669", + "right": "issue:44998", + "accept": false, + "reason": "Import-star cleanup and off-topic issue are unrelated." + }, + { + "left": "issue:43066", + "right": "issue:44998", + "accept": false, + "reason": "Tokenizer decoder-type regression and off-topic issue are unrelated." + }, + { + "left": "issue:31515", + "right": "issue:38617", + "accept": false, + "reason": "Checkpoint load slowness is unrelated to an ImportError for layer_type_validation." + }, + { + "left": "issue:44230", + "right": "issue:44623", + "accept": false, + "reason": "FP8 support request is unrelated to missing processor save files." + }, + { + "left": "issue:44998", + "right": "issue:46218", + "accept": false, + "reason": "Off-topic issue and a Qwen3_5 import failure are unrelated." + }, + { + "left": "issue:30333", + "right": "issue:31515", + "accept": false, + "reason": "MLflow reporting after failed training and slow from_pretrained checkpoint loading are unrelated." + }, + { + "left": "issue:31515", + "right": "issue:36296", + "accept": false, + "reason": "Checkpoint load slowness and tensor-parallel training bug are different failures." + }, + { + "left": "issue:44230", + "right": "issue:45310", + "accept": false, + "reason": "FP8 inference support and Qwen3.5 MoE from_pretrained error are not the same bug." + }, + { + "left": "issue:31515", + "right": "issue:33453", + "accept": false, + "reason": "Slow checkpoint loading and tokenizer-loading regression are distinct issues." + }, + { + "left": "issue:43298", + "right": "issue:44230", + "accept": false, + "reason": "Audio continuous batching support and FP8 inference support are unrelated." + }, + { + "left": "issue:43717", + "right": "issue:46144", + "accept": false, + "reason": "Mamba init_weights behavior is unrelated to RoFormer attention interface usage." + }, + { + "left": "issue:42936", + "right": "issue:46018", + "accept": false, + "reason": "Mask2former ignore_value handling is unrelated to DeepSeek-V4 shared-expert gating." + }, + { + "left": "issue:43668", + "right": "issue:44031", + "accept": false, + "reason": "A type-hint issue in ModernBERTConfig is unrelated to tokenizer regex warnings." + }, + { + "left": "issue:42936", + "right": "issue:43240", + "accept": false, + "reason": "Ignore_value usage in Mask2former and loss-kwargs forwarding are unrelated." + }, + { + "left": "issue:43240", + "right": "issue:46144", + "accept": false, + "reason": "Loss kwargs forwarding and RoFormer attention interface are unrelated." + }, + { + "left": "issue:42936", + "right": "issue:44031", + "accept": false, + "reason": "Mask2former ignore_value and tokenizer regex warning are unrelated." + }, + { + "left": "issue:42936", + "right": "issue:46144", + "accept": false, + "reason": "Mask2former ignore_value and RoFormer attention interface are unrelated." + }, + { + "left": "issue:42936", + "right": "issue:43813", + "accept": false, + "reason": "Mask2former ignore_value and a quantization-operation typo are unrelated." + }, + { + "left": "issue:43240", + "right": "issue:43668", + "accept": false, + "reason": "Loss kwargs forwarding and a config type-hint typo are unrelated." + }, + { + "left": "issue:43668", + "right": "issue:43813", + "accept": false, + "reason": "Config type-hint correctness and a quantization-operation typo are different issues." + }, + { + "left": "issue:37934", + "right": "issue:43668", + "accept": false, + "reason": "Llama4TextL2Norm semantics and ModernBERTConfig type hints are unrelated." + }, + { + "left": "issue:42936", + "right": "issue:43668", + "accept": false, + "reason": "Mask2former ignore_value and a ModernBERTConfig type hint 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": 18, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10519-3", + "nodes": [ + "issue:10519", + "issue:11005", + "issue:11307" + ], + "soft_pairs": [ + "issue:10519|issue:11307", + "issue:10519|issue:11005", + "issue:11005|issue:11307" + ], + "prepared_review_unit_hash": "c96d02fdde0fc3e7748f502effef9f622ce56178adb91d01270bf86cad702203", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 2877, + "estimated_input_tokens": 720, + "estimated_eval_tokens": 1696 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 2877, + "estimated_input_tokens": 720, + "estimated_eval_tokens": 1696 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These three issues are about unrelated areas: token truncation behavior, learning-rate scheduler functionality, and Wav2Vec2 word time offsets. They do not describe the same underlying bug or feature request, so the cluster should not be merged as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:10519", + "right": "issue:11307", + "accept": false, + "reason": "Different feature areas and requests: truncation strategy in tokenization vs word start/end timestamps in Wav2Vec2. No shared code-path or concrete duplicate symptom." + }, + { + "left": "issue:10519", + "right": "issue:11005", + "accept": false, + "reason": "Unrelated topics: token truncation direction vs ReduceLROnPlateau-like scheduler behavior. They do not appear to be the same bug or change." + }, + { + "left": "issue:11005", + "right": "issue:11307", + "accept": false, + "reason": "Different ML features: learning-rate scheduling vs extracting token time offsets from speech models. No plausible single fix or merged PR target." + } + ] + }, + "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": 18, + "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": "2e45e3e60be5be66820c62f40ee2c17efa7ef14551655c89acd275263076d6b7", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2478, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2478, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster does not look like a single bug family: one issue is about TAPAS tokenization differing from TensorFlow code, one about a private-model download/login error, and one about failing to load bert-base-uncased. The soft links are likely superficial Hugging Face model-related similarity rather than the same underlying problem.", + "confidence": 0.89, + "canonical_issue_reason": "issue:16618 is the most concrete and broadly representative user-facing failure in the set (a model loading bug), whereas the others concern unrelated tokenization behavior and private-model authentication messaging.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:16618 is the best standalone issue because it describes a specific, actionable load failure with a common model name and is easier to triage than the more specialized or ancillary reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:13244", + "right": "issue:16618", + "accept": false, + "reason": "Tokenization mismatch in TAPAS and inability to load bert-base-uncased are different code paths and different failure modes; they do not appear to be the same underlying bug." + }, + { + "left": "issue:15980", + "right": "issue:16618", + "accept": false, + "reason": "Private-model download/login error is an authentication/message issue, while bert-base-uncased load failure is a separate model-loading problem; no clear shared 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": 4, + "component_count": 18, + "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": "2c9b694aa1bef3b70a6daeefa69a15e923d7d3fa679b71e776bb8c94d07b709f", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1589, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1589, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two separate community contribution requests: one for Flash Attention 2 support across more architectures, and one for GGUF support across more architectures. They share a broad theme of architecture support, but not the same underlying bug or change.", + "confidence": 0.98, + "canonical_issue_reason": "No canonical issue: the issues track different feature requests for different code paths and are not duplicates.", + "canonical_pr_reason": null, + "best_issue_reason": "No single best issue for deduplication; each issue represents a distinct enhancement area rather than a duplicate of the other.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:26350 (Community contribution: Adding Flash Attention 2 support for more architectures)", + "right": "issue:33260 (Community contribution: Adding GGUF support for more architectures)", + "accept": false, + "reason": "Different capabilities and likely different implementation paths; they share only a broad \"support more architectures\" theme, so they are not the same underlying request." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 5, + "component_count": 18, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-27088-8", + "nodes": [ + "issue:27088", + "issue:28936", + "issue:38435", + "issue:43834", + "issue:44365", + "issue:44370", + "issue:46222", + "issue:46226" + ], + "soft_pairs": [ + "issue:27088|issue:44370", + "issue:27088|issue:44365", + "issue:27088|issue:43834", + "issue:27088|issue:38435", + "issue:27088|issue:28936", + "issue:28936|issue:43834", + "issue:28936|issue:44365", + "issue:28936|issue:38435", + "issue:28936|issue:44370", + "issue:38435|issue:44370", + "issue:38435|issue:44365", + "issue:38435|issue:43834", + "issue:38435|issue:46222", + "issue:38435|issue:46226", + "issue:43834|issue:44370", + "issue:43834|issue:44365", + "issue:43834|issue:46222", + "issue:43834|issue:46226", + "issue:44365|issue:44370", + "issue:44365|issue:46222", + "issue:44365|issue:46226", + "issue:44370|issue:46222", + "issue:44370|issue:46226", + "issue:46222|issue:46226" + ], + "prepared_review_unit_hash": "48debc4ac5c4516ca7ae7ae2544714a2d73ea8679f6fef2a0396b60e5b960593", + "component_budget": { + "node_count": 8, + "item_count": 8, + "soft_pair_count": 24, + "serialized_chars": 13528, + "estimated_input_tokens": 3382, + "estimated_eval_tokens": 7020 + }, + "budget": { + "node_count": 8, + "item_count": 8, + "soft_pair_count": 24, + "serialized_chars": 13528, + "estimated_input_tokens": 3382, + "estimated_eval_tokens": 7020 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Mixed i18n cluster: Turkish/Spanish/Romanian translation requests are distinct, while the three generic placeholder issues form one duplicate subgroup and the two Thai-text issues form another exact duplicate pair. No single issue cleanly canonicalizes the whole set.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue fits the full cluster because it contains separate language-specific requests plus two independent duplicate subgroups.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43834 is the best representative of the placeholder-template duplicate subgroup, but it is not a good global representative for the language-specific requests or the Thai duplicate pair.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:27088", + "right": "issue:44370", + "accept": false, + "reason": "Different target language/request; Turkish vs generic placeholder text." + }, + { + "left": "issue:27088", + "right": "issue:44365", + "accept": false, + "reason": "Different target language/request; Turkish vs generic placeholder text." + }, + { + "left": "issue:27088", + "right": "issue:43834", + "accept": false, + "reason": "Turkish translation request is separate from the generic placeholder issue." + }, + { + "left": "issue:27088", + "right": "issue:38435", + "accept": false, + "reason": "Different language requests: Turkish vs Romanian." + }, + { + "left": "issue:27088", + "right": "issue:28936", + "accept": false, + "reason": "Different language requests: Turkish vs Spanish." + }, + { + "left": "issue:28936", + "right": "issue:43834", + "accept": false, + "reason": "Spanish translation request is separate from the generic placeholder issue." + }, + { + "left": "issue:28936", + "right": "issue:44365", + "accept": false, + "reason": "Spanish translation request is separate from the generic placeholder issue." + }, + { + "left": "issue:28936", + "right": "issue:38435", + "accept": false, + "reason": "Different language requests: Spanish vs Romanian." + }, + { + "left": "issue:28936", + "right": "issue:44370", + "accept": false, + "reason": "Spanish translation request is separate from the generic placeholder issue." + }, + { + "left": "issue:38435", + "right": "issue:44370", + "accept": false, + "reason": "Romanian translation request is separate from the generic placeholder issue." + }, + { + "left": "issue:38435", + "right": "issue:44365", + "accept": false, + "reason": "Romanian translation request is separate from the generic placeholder issue." + }, + { + "left": "issue:38435", + "right": "issue:43834", + "accept": false, + "reason": "Romanian translation request is separate from the generic placeholder issue." + }, + { + "left": "issue:38435", + "right": "issue:46222", + "accept": false, + "reason": "Different content: Romanian translation request vs Thai-text issue." + }, + { + "left": "issue:38435", + "right": "issue:46226", + "accept": false, + "reason": "Different content: Romanian translation request vs Thai-text issue." + }, + { + "left": "issue:43834", + "right": "issue:44370", + "accept": true, + "reason": "Same generic placeholder translation issue; looks like a duplicate copy of the same template artifact." + }, + { + "left": "issue:43834", + "right": "issue:44365", + "accept": true, + "reason": "Same generic placeholder translation issue; looks like a duplicate copy of the same template artifact." + }, + { + "left": "issue:43834", + "right": "issue:46222", + "accept": false, + "reason": "Generic placeholder issue vs distinct Thai-text issue; not the same underlying request." + }, + { + "left": "issue:43834", + "right": "issue:46226", + "accept": false, + "reason": "Generic placeholder issue vs distinct Thai-text issue; not the same underlying request." + }, + { + "left": "issue:44365", + "right": "issue:44370", + "accept": true, + "reason": "Same generic placeholder translation issue; looks like a duplicate copy of the same template artifact." + }, + { + "left": "issue:44365", + "right": "issue:46222", + "accept": false, + "reason": "Generic placeholder issue vs distinct Thai-text issue; not the same underlying request." + }, + { + "left": "issue:44365", + "right": "issue:46226", + "accept": false, + "reason": "Generic placeholder issue vs distinct Thai-text issue; not the same underlying request." + }, + { + "left": "issue:44370", + "right": "issue:46222", + "accept": false, + "reason": "Generic placeholder issue vs distinct Thai-text issue; not the same underlying request." + }, + { + "left": "issue:44370", + "right": "issue:46226", + "accept": false, + "reason": "Generic placeholder issue vs distinct Thai-text issue; not the same underlying request." + }, + { + "left": "issue:46222", + "right": "issue:46226", + "accept": true, + "reason": "Exact duplicate Thai-text issue; same title/content and same underlying artifact." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: the three language-specific translation requests are reasonably treated as separate, the three placeholder-template issues form a coherent duplicate subgroup, and the two Thai-title issues are an exact duplicate pair. The soft-edge verdicts are conservative overall, with no clear overstatement of evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 6, + "component_count": 18, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-36817-16", + "nodes": [ + "issue:36817", + "issue:36879", + "issue:37015", + "issue:41115", + "issue:41211", + "issue:42503", + "issue:42738", + "issue:43671", + "issue:43837", + "issue:43909", + "issue:44121", + "issue:44205", + "issue:45027", + "issue:45177", + "issue:45322", + "issue:46133" + ], + "soft_pairs": [ + "issue:36817|issue:41211", + "issue:36817|issue:43671", + "issue:36817|issue:36879", + "issue:36879|issue:41211", + "issue:36879|issue:43909", + "issue:36879|issue:44121", + "issue:36879|issue:43837", + "issue:36879|issue:43671", + "issue:37015|issue:43837", + "issue:37015|issue:44121", + "issue:37015|issue:43909", + "issue:36879|issue:37015", + "issue:37015|issue:42503", + "issue:41115|issue:43837", + "issue:41115|issue:44121", + "issue:41115|issue:42503", + "issue:41115|issue:45027", + "issue:41211|issue:43671", + "issue:42503|issue:43837", + "issue:42503|issue:43671", + "issue:42503|issue:42738", + "issue:42738|issue:45027", + "issue:43671|issue:43837", + "issue:43671|issue:45027", + "issue:43671|issue:45177", + "issue:43671|issue:44121", + "issue:43837|issue:43909", + "issue:43837|issue:45322", + "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", + "issue:45322|issue:46133" + ], + "prepared_review_unit_hash": "907f945b8a32ec29c695556c788f027f25f7cb421aae0585240a85f07e97c890", + "component_budget": { + "node_count": 16, + "item_count": 16, + "soft_pair_count": 36, + "serialized_chars": 21661, + "estimated_input_tokens": 5416, + "estimated_eval_tokens": 11088 + }, + "budget": { + "node_count": 16, + "item_count": 16, + "soft_pair_count": 36, + "serialized_chars": 21661, + "estimated_input_tokens": 5416, + "estimated_eval_tokens": 11088 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are all distinct model-request issues that share a generic 'add model/support' pattern, but they target different models and code paths. No soft edge looks like a true duplicate or same underlying change.", + "confidence": 0.93, + "canonical_issue_reason": "issue:43671 is the most central and broadly representative request in this cluster, with the widest set of soft links and a generic model-support framing.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43671 is the best umbrella representative for the cluster because it is broad, central, and least tied to a single narrow model variant.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36817", + "right": "issue:41211", + "accept": false, + "reason": "Different requested models and architectures; only the 'add model' wording overlaps." + }, + { + "left": "issue:36817", + "right": "issue:43671", + "accept": false, + "reason": "EuroBert config vs Qwen3-TTS support are unrelated model additions." + }, + { + "left": "issue:36817", + "right": "issue:36879", + "accept": false, + "reason": "EuroBert config and RF-DETR target different model families and code paths." + }, + { + "left": "issue:36879", + "right": "issue:41211", + "accept": false, + "reason": "RF-DETR and DEIMv2 are distinct detection models, not the same change." + }, + { + "left": "issue:36879", + "right": "issue:43909", + "accept": false, + "reason": "Object detection support vs an audio model request are unrelated." + }, + { + "left": "issue:36879", + "right": "issue:44121", + "accept": false, + "reason": "RF-DETR and weight-sparse transformer support are different model changes." + }, + { + "left": "issue:36879", + "right": "issue:43837", + "accept": false, + "reason": "RF-DETR and Qwen3-ASR are different tasks and model families." + }, + { + "left": "issue:36879", + "right": "issue:43671", + "accept": false, + "reason": "RF-DETR and Qwen3-TTS are unrelated requests beyond generic model support." + }, + { + "left": "issue:37015", + "right": "issue:43837", + "accept": false, + "reason": "NeoBERT and Qwen3-ASR are different model families and capabilities." + }, + { + "left": "issue:37015", + "right": "issue:44121", + "accept": false, + "reason": "NeoBERT addition is not the same as sparse-transformer support." + }, + { + "left": "issue:37015", + "right": "issue:43909", + "accept": false, + "reason": "A BERT-style text model and an audio model are unrelated." + }, + { + "left": "issue:36879", + "right": "issue:37015", + "accept": false, + "reason": "RF-DETR and NeoBERT are distinct requests with no shared concrete bug/change." + }, + { + "left": "issue:37015", + "right": "issue:42503", + "accept": false, + "reason": "NeoBERT and ModernVBERT are different model-specific requests, not a duplicate fix." + }, + { + "left": "issue:41115", + "right": "issue:43837", + "accept": false, + "reason": "MiniCPM3 architecture support and Qwen3-ASR support are separate model integrations." + }, + { + "left": "issue:41115", + "right": "issue:44121", + "accept": false, + "reason": "MiniCPM3 and a weight-sparse transformer request do not describe the same change." + }, + { + "left": "issue:41115", + "right": "issue:42503", + "accept": false, + "reason": "MiniCPM3 and ModernVBERT are different model additions." + }, + { + "left": "issue:41115", + "right": "issue:45027", + "accept": false, + "reason": "MiniCPM3 and Voxtral TTS target different model families and tasks." + }, + { + "left": "issue:41211", + "right": "issue:43671", + "accept": false, + "reason": "DEIMv2 detection support and Qwen3-TTS support are unrelated." + }, + { + "left": "issue:42503", + "right": "issue:43837", + "accept": false, + "reason": "ModernVBERT models and Qwen3-ASR are different requests." + }, + { + "left": "issue:42503", + "right": "issue:43671", + "accept": false, + "reason": "ModernVBERT and Qwen3-TTS are not the same underlying change." + }, + { + "left": "issue:42503", + "right": "issue:42738", + "accept": false, + "reason": "Both are BERT-themed, but one is a specific model request and the other is a general RoPE discussion; not a duplicate." + }, + { + "left": "issue:42738", + "right": "issue:45027", + "accept": false, + "reason": "A general BERT-like RoPE request and Voxtral TTS support are unrelated." + }, + { + "left": "issue:43671", + "right": "issue:43837", + "accept": false, + "reason": "TTS and ASR are different capabilities; same family, but not the same concrete change." + }, + { + "left": "issue:43671", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-TTS and Voxtral TTS are different model integrations." + }, + { + "left": "issue:43671", + "right": "issue:45177", + "accept": false, + "reason": "Audio TTS support and DeepSeek-OCR2 are unrelated model requests." + }, + { + "left": "issue:43671", + "right": "issue:44121", + "accept": false, + "reason": "Qwen3-TTS and sparse-transformer support are different code paths and model families." + }, + { + "left": "issue:43837", + "right": "issue:43909", + "accept": false, + "reason": "Qwen3-ASR and LFM2.5 Audio are both audio-related, but they are distinct model requests." + }, + { + "left": "issue:43837", + "right": "issue:45322", + "accept": false, + "reason": "ASR support and EUPE vision-language encoder support are unrelated." + }, + { + "left": "issue:43909", + "right": "issue:44205", + "accept": false, + "reason": "Audio model support and SAM3-LiteText are different model domains." + }, + { + "left": "issue:43671", + "right": "issue:43909", + "accept": false, + "reason": "Qwen3-TTS and LFM2.5 Audio 1.5B are different audio model integrations." + }, + { + "left": "issue:44121", + "right": "issue:45322", + "accept": false, + "reason": "Sparse transformer support and EUPE are different model requests." + }, + { + "left": "issue:43837", + "right": "issue:44121", + "accept": false, + "reason": "ASR support and weight-sparse transformer support are unrelated." + }, + { + "left": "issue:43837", + "right": "issue:45027", + "accept": false, + "reason": "Both are audio-related, but ASR and TTS requests are distinct model changes." + }, + { + "left": "issue:37015", + "right": "issue:45027", + "accept": false, + "reason": "NeoBERT and Voxtral-4B-TTS are unrelated model additions." + }, + { + "left": "issue:43837", + "right": "issue:45177", + "accept": false, + "reason": "ASR support and OCR model support are different features." + }, + { + "left": "issue:45322", + "right": "issue:46133", + "accept": false, + "reason": "EUPE and TIPSv2 are separate vision-language model requests, not the same change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 7, + "component_count": 18, + "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": "e940da8bb581782bf6d4b3ef3f8b36c70f45fb8a583a2bfe387241a705d44e1c", + "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": "The two issues are not duplicates: one is about block decoding, the other is about adding DeepSeek v3.2 support. They describe different features/problems with no shared target or code path.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41084", + "right": "issue:43037", + "accept": false, + "reason": "Different topics and likely different code paths; block decoding is unrelated to DeepSeek model support." + } + ] + }, + "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": 18, + "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": "3b73fc8e8f885e0913655ff4a86d78ff1ced2108635afee613a6dc41ea6aa593", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1584, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1584, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two closed issues both mention Trainer customization, but they describe different problems: one is about `remove_unused_column` behavior, the other about FLOPS estimation for custom trainers. They should not be treated as duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "No strong canonical issue emerges because the topics are distinct rather than overlapping reports of the same bug.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #43598 is slightly more specific and actionable around a concrete Trainer setting, so it is the better representative if one must be chosen.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43598", + "right": "issue:43600", + "accept": false, + "reason": "Different underlying concerns: one is about column-removal customizability in Trainer, the other about FLOPS accounting for custom trainers. Shared 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": 18, + "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": "3c603234892ba3ead0b7c30c3952d891c4d5084dd7c129512050878b5cc25296", + "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": "The two issues are related only at the very broad level of adding AutoModel support, but they target different model families and different auto classes, so they are not duplicates.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43684 is the broader request: it covers Qwen3-Omni registration for both AutoModel and AutoModelForConditionalGeneration, whereas 44405 is a narrower, separate request for sequence classification on Qwen3.5.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43684", + "right": "issue:44405", + "accept": false, + "reason": "Different underlying features/models: Qwen3-Omni auto-registration vs Qwen3.5 sequence-classification support. They do not describe the same bug or change path and would not be merged into one fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 10, + "component_count": 18, + "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": "3461e2b8c6ce78d0b69667ade54a7017554095fb126ad3695b9c3001cc0f9e11", + "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": "The cluster appears to contain one unrelated issue ('Fett') and a duplicate trio of near-identical 'huggingface ' issues. With no substantive bodies, the safest canonical choice is the earliest huggingface issue.", + "confidence": 0.92, + "canonical_issue_reason": "issue:43845 is the earliest of the three identical 'huggingface ' issues and is the best representative for that duplicate set; issue:43835 ('Fett') looks unrelated.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43845 is the strongest global issue representative because it matches the other two huggingface issues exactly in title and is earliest, while the only other title in the cluster is unrelated.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43835", + "right": "issue:43845", + "accept": false, + "reason": "Different titles with no shared substantive details; nothing indicates the same underlying report." + }, + { + "left": "issue:43835", + "right": "issue:43846", + "accept": false, + "reason": "Different titles with no shared substantive details; nothing indicates the same underlying report." + }, + { + "left": "issue:43835", + "right": "issue:43859", + "accept": false, + "reason": "Different titles with no shared substantive details; nothing indicates the same underlying report." + }, + { + "left": "issue:43845", + "right": "issue:43846", + "accept": true, + "reason": "Same normalized title ('huggingface ') and close creation times, with no evidence of distinct topics." + }, + { + "left": "issue:43845", + "right": "issue:43859", + "accept": true, + "reason": "Same normalized title ('huggingface ') and close creation times; likely duplicate submissions of the same issue." + }, + { + "left": "issue:43846", + "right": "issue:43859", + "accept": true, + "reason": "Same normalized title ('huggingface ') and close creation times, consistent with duplicates of one report." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet: the three \"huggingface \" issues are title-identical and closely timed, while \"Fett\" is distinct. The canonical choice of the earliest huggingface issue is conservative, and the soft-edge verdicts do not overstate evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 11, + "component_count": 18, + "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": "d53f1092e43db99dd1fa31b3ae01c9423542bf97eae17fc59f5a890c967f15ba", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 9, + "serialized_chars": 5794, + "estimated_input_tokens": 1449, + "estimated_eval_tokens": 3154 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 9, + "serialized_chars": 5794, + "estimated_input_tokens": 1449, + "estimated_eval_tokens": 3154 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is dominated by near-empty/spam-like issues with no concrete bug or change details. Issue 44418 is the only somewhat substantive topic, but there is still not enough evidence that any pair is a true duplicate.", + "confidence": 0.34, + "canonical_issue_reason": "issue:44418 has the most meaningful, non-spam title and is the only item that looks like a real triage target; the others are generic spam placeholders.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44418 is the best representative of the cluster because it is the only issue with an identifiable subject matter and some potential triage value.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44371", + "right": "issue:44572", + "accept": false, + "reason": "Both are just '' with no body detail; identical placeholder titles are not enough to prove the same underlying issue." + }, + { + "left": "issue:44371", + "right": "issue:44573", + "accept": false, + "reason": "Both are just '' with no actionable description, so there is no concrete evidence they report the same problem." + }, + { + "left": "issue:44371", + "right": "issue:44418", + "accept": false, + "reason": "One is spam-like and the other is a generic documentation request; the titles do not indicate the same underlying bug or change." + }, + { + "left": "issue:44371", + "right": "issue:44937", + "accept": false, + "reason": "The placeholder spam title and the specific 'Google LLC' title are unrelated on their face, with no body evidence tying them together." + }, + { + "left": "issue:44418", + "right": "issue:44572", + "accept": false, + "reason": "A documentation enhancement request and a spam placeholder do not describe the same issue; there is no shared concrete problem statement." + }, + { + "left": "issue:44418", + "right": "issue:44573", + "accept": false, + "reason": "The documentation suggestion is too generic, and the other item is empty spam; no basis to merge as duplicates." + }, + { + "left": "issue:44572", + "right": "issue:44573", + "accept": false, + "reason": "Matching spam placeholders alone do not establish a duplicate; there is no description or target to confirm the same underlying report." + }, + { + "left": "issue:44572", + "right": "issue:44937", + "accept": false, + "reason": "A spam placeholder and a specific entity-name title are not the same reported problem, and there is no supporting detail." + }, + { + "left": "issue:44573", + "right": "issue:44937", + "accept": false, + "reason": "These titles are not describing the same concrete bug or request, and the bodies provide no evidence of a shared 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": 12, + "component_count": 18, + "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": "a64c90beeda203331664a7c1862ba6065c02a171e9900ce7eeb28bc31b169da5", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1645, + "estimated_input_tokens": 412, + "estimated_eval_tokens": 1080 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1645, + "estimated_input_tokens": 412, + "estimated_eval_tokens": 1080 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two closed issues share an identical title and some textual overlap, but the evidence is not strong enough to call them near-identical duplicates or to consolidate them as the same report.", + "confidence": 0.9, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue #45145 is the slightly better cluster representative because it is earlier and already has an inbound reference, but this is only a mild preference, not a duplicate determination.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45145", + "right": "issue:45151", + "accept": false, + "reason": "Identical titles and moderate similarity suggest related reports, but the packet does not show sufficiently similar bodies or evidence of the same underlying issue to justify duplicate consolidation." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The pairing is only supported as a soft similarity match with identical titles and moderate Jaccard similarity; the packet does not show the bodies are \"highly similar\" or that the issues are \"near-identical duplicates.\" The consolidation/duplicate language is a bit too strong for the evidence, so this should be rejected as overstated." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 13, + "component_count": 18, + "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": 18, + "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": "0e6650b004bf55a238b3d2999a96a9ab90dcb286490a30050c01223489a249c6", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1670, + "estimated_input_tokens": 418, + "estimated_eval_tokens": 1092 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1670, + "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 involve 4-bit inference), but they describe different bugs: one is a torch.finfo() failure on uint8 audio-module weights, the other is a BitsAndBytesConfig skip-module regression that drops the default lm_head exclusion and triggers an AssertionError.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 45674 is the better canonical representative because it is the more general, currently open regression with clearer reproduction around BitsAndBytesConfig behavior and 4-bit inference.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45674 is the best overall issue in the cluster: it is broader, more actionable, and likely to gather the most useful triage attention for the 4-bit inference path.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45672", + "right": "issue:45674", + "accept": false, + "reason": "Reject: these appear to be distinct bugs with different symptoms and code paths. 45672 is a torch.finfo() TypeError on uint8 audio-module weights, while 45674 is a BitsAndBytesConfig skip-module handling regression that removes the default lm_head exclusion and causes an AssertionError." + } + ] + }, + "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": 18, + "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": "aeb748a30b21ab9a3edbacf594a998ee1d4142181cd086af6f8393d45addcc32", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1476, + "estimated_input_tokens": 369, + "estimated_eval_tokens": 994 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1476, + "estimated_input_tokens": 369, + "estimated_eval_tokens": 994 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are not duplicates of the same underlying bug; they appear to be separate name/date entries with no technical context. The soft similarity is superficial only.", + "confidence": 0.93, + "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": "Titles share a similar personal-name/date pattern, but there is no evidence they describe the same problem or request; they look like distinct entries." + } + ] + }, + "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": 18, + "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": "f6a653532046d023650df77efcfc20b583ac910ec692a97d99e6cc5806e017a2", + "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 two closed issues only share a loose textual similarity in personal-name/date formatting and do not show any evidence of the same underlying bug or request.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45764", + "right": "issue:45765", + "accept": false, + "reason": "The titles refer to different people and different dates; there is no issue body, linkage, or code context indicating a shared defect or duplicate report." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 17, + "component_count": 18, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45810-2", + "nodes": [ + "issue:45810", + "issue:46295" + ], + "soft_pairs": [ + "issue:45810|issue:46295" + ], + "prepared_review_unit_hash": "3e8d6bf0b1330236a40e752aede04a703a87cef620dbf1666a0e1b32ebe3b08e", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1541, + "estimated_input_tokens": 386, + "estimated_eval_tokens": 1028 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1541, + "estimated_input_tokens": 386, + "estimated_eval_tokens": 1028 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are not duplicates: one asks for a new token-classification class for Qwen3.5, while the other requests FP8 training support. They touch different features and code paths.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:46295 is the broader, more general feature request, but it is still unrelated to issue:45810 and should not be treated as a duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45810", + "right": "issue:46295", + "accept": false, + "reason": "Different underlying requests: model-specific token classification support versus FP8 training support. They do not describe the same bug or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 18, + "component_count": 18, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45869-5", + "nodes": [ + "issue:45869", + "issue:45870", + "issue:45871", + "issue:45872", + "issue:45873" + ], + "soft_pairs": [ + "issue:45869|issue:45870", + "issue:45869|issue:45871", + "issue:45870|issue:45871", + "issue:45870|issue:45872", + "issue:45870|issue:45873", + "issue:45871|issue:45872", + "issue:45871|issue:45873", + "issue:45872|issue:45873" + ], + "prepared_review_unit_hash": "ac47b6e4fb64cd07a9076b6f05ff817a1fd77993cdbdc64e724c4f5f14b55528", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5725, + "estimated_input_tokens": 1432, + "estimated_eval_tokens": 3120 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5725, + "estimated_input_tokens": 1432, + "estimated_eval_tokens": 3120 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is not a single duplicate set: it contains two exact-duplicate issue pairs about different DataForge examples, plus one unrelated dataset promotion issue. Accept the two exact-title duplicates and reject the cross-topic soft similarities.", + "confidence": 0.96, + "canonical_issue_reason": "If forced to pick one representative, issue #45870 is the cleanest canonical for the exact duplicate pair #45870/#45871; however, the cluster actually splits into two separate issue topics, so a single true canonical does not exist.", + "canonical_pr_reason": null, + "best_issue_reason": "#45870 is the best issue candidate because it is the clearest, most general example issue and has an exact duplicate partner. Still, it should not absorb #45872/#45873 since those describe a different example/change.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45869", + "right": "issue:45870", + "accept": false, + "reason": "Dataset bundle promo and a DataForge example are related only by theme, not the same underlying issue." + }, + { + "left": "issue:45869", + "right": "issue:45871", + "accept": false, + "reason": "Same as above; no evidence they describe the same bug or change." + }, + { + "left": "issue:45870", + "right": "issue:45871", + "accept": true, + "reason": "Exact same title and clearly the same example issue; this is a straightforward duplicate." + }, + { + "left": "issue:45870", + "right": "issue:45872", + "accept": false, + "reason": "Both are DataForge-related, but one is an example issue and the other is a separate time-series + scraping service topic." + }, + { + "left": "issue:45870", + "right": "issue:45873", + "accept": false, + "reason": "Different concrete topic despite similar keywords; not the same underlying change." + }, + { + "left": "issue:45871", + "right": "issue:45872", + "accept": false, + "reason": "These are two distinct issue topics, not one duplicate bug/report." + }, + { + "left": "issue:45871", + "right": "issue:45873", + "accept": false, + "reason": "Same as above; similarity is only at the broad subsystem/topic level." + }, + { + "left": "issue:45872", + "right": "issue:45873", + "accept": true, + "reason": "Exact same title and evidently the same issue; strong duplicate match." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: 45870/45871 are identical, 45872/45873 are identical, and 45869 is a distinct dataset-promo item. The canonicalization note is conservative, and the soft-edge verdicts do not overclaim beyond the available evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 1, + "component_count": 30, + "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-39895-15", + "nodes": [ + "pull_request:39895", + "pull_request:41855", + "pull_request:42668", + "pull_request:43251", + "pull_request:43254", + "pull_request:43424", + "pull_request:43823", + "pull_request:43995", + "pull_request:44827", + "pull_request:45170", + "pull_request:45213", + "pull_request:45655", + "pull_request:46048", + "pull_request:46267", + "pull_request:46268" + ], + "soft_pairs": [ + "pull_request:39895|pull_request:43424", + "pull_request:41855|pull_request:43823", + "pull_request:41855|pull_request:43424", + "pull_request:42668|pull_request:43823", + "pull_request:43251|pull_request:43823", + "pull_request:43251|pull_request:44827", + "pull_request:43424|pull_request:44827", + "pull_request:43424|pull_request:43823", + "pull_request:43823|pull_request:45655", + "pull_request:43823|pull_request:43995", + "pull_request:43995|pull_request:45213", + "pull_request:43995|pull_request:44827", + "pull_request:44827|pull_request:45170", + "pull_request:44827|pull_request:46267", + "pull_request:44827|pull_request:46048", + "pull_request:44827|pull_request:45655", + "pull_request:44827|pull_request:45213", + "pull_request:43823|pull_request:45213", + "pull_request:46048|pull_request:46268", + "pull_request:43251|pull_request:43254" + ], + "prepared_review_unit_hash": "4f3877877ddca4d741ef42f29ca2e42b06a6962e958c70b022fd4daf238ebb5c", + "component_budget": { + "node_count": 15, + "item_count": 15, + "soft_pair_count": 20, + "serialized_chars": 40304, + "estimated_input_tokens": 10076, + "estimated_eval_tokens": 20408 + }, + "budget": { + "node_count": 15, + "item_count": 15, + "soft_pair_count": 20, + "serialized_chars": 40304, + "estimated_input_tokens": 10076, + "estimated_eval_tokens": 20408 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "One clear duplicate cluster exists around `fixed_cross_entropy` kwargs forwarding (`43251` and `43254`). The rest are distinct model additions, tests, documentation, or unrelated infra changes and should remain separate.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "`pull_request:43251` is the cleaner, more complete fix for the loss kwargs bug: it explicitly forwards `weight` and `label_smoothing` to `nn.functional.cross_entropy` and is linked to the tracked issue.", + "best_issue_reason": null, + "best_pr_reason": "`pull_request:43251` is the best representative of the cluster because it directly addresses the concrete cross-entropy kwargs omission and is the more polished implementation compared with `43254`.", + "soft_edge_verdicts": [ + { + "left": "pull_request:39895", + "right": "pull_request:43424", + "accept": false, + "reason": "VideoPrism model addition and an executorch dynamic-shapes test are unrelated changes; same broad theme of testing/exportability is not enough." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43823", + "accept": false, + "reason": "Mistral tokenizer method additions and a new MobileLLM model implementation are unrelated code paths." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43424", + "accept": false, + "reason": "Tokenizer API additions and executorch export tests fix different problems in different subsystems." + }, + { + "left": "pull_request:42668", + "right": "pull_request:43823", + "accept": false, + "reason": "Processor auto-loading robustness and adding MobileLLM are different changes; no shared concrete bug." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43823", + "accept": false, + "reason": "Cross-entropy kwargs forwarding and a new model implementation are unrelated." + }, + { + "left": "pull_request:43251", + "right": "pull_request:44827", + "accept": false, + "reason": "The loss helper fix and Mistral4 test/kernel changes do not target the same bug or code path." + }, + { + "left": "pull_request:43424", + "right": "pull_request:44827", + "accept": false, + "reason": "Executorch dynamic-shape testing and Mistral4 tests are unrelated test concerns, not the same underlying issue." + }, + { + "left": "pull_request:43424", + "right": "pull_request:43823", + "accept": false, + "reason": "Dynamic-shape export tests and MobileLLM implementation are unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45655", + "accept": false, + "reason": "A new model implementation and fixing `cls.config` resolution are unrelated changes." + }, + { + "left": "pull_request:43823", + "right": "pull_request:43995", + "accept": false, + "reason": "MobileLLM implementation and Falcon output-interface refactoring are different model-specific changes." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45213", + "accept": false, + "reason": "Falcon refactoring and a model-creation skill are not the same bug/change." + }, + { + "left": "pull_request:43995", + "right": "pull_request:44827", + "accept": false, + "reason": "Falcon output-collection refactor and Mistral4 test fixes do not fix the same concrete issue." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45170", + "accept": false, + "reason": "Mistral4 test fixes and `layrnorm` spelling cleanup across CLIP-like models are unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46267", + "accept": false, + "reason": "Mistral4 test fixes and chat response parsing updates are completely different areas." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46048", + "accept": false, + "reason": "Mistral4 test/integration changes and TDT loss kernel integration are unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45655", + "accept": false, + "reason": "Mistral4 test changes and config-resolution order fixes are unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45213", + "accept": false, + "reason": "Model-specific test fixes and the model-creation skill draft are not the same underlying change." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45213", + "accept": false, + "reason": "Adding MobileLLM and creating a model-generation skill are unrelated." + }, + { + "left": "pull_request:46048", + "right": "pull_request:46268", + "accept": false, + "reason": "Both touch hub-kernel plumbing, but they target different kernels and different behaviors; they are not the same concrete change." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43254", + "accept": true, + "reason": "Both PRs fix the same `fixed_cross_entropy` kwargs-forwarding bug in the same file, and one could plausibly merge the other\u2019s intent into a single PR." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is well grounded in the packet: `43251` and `43254` are the only clear same-bug pair, both touching `src/transformers/loss/loss_utils.py` and sharing issue `43240`. The remaining soft-edge rejections are conservative and consistent with the diff previews. The only slightly subjective phrasing is the preference for `43251` as the cleaner representative, but it is still supported by the explicit kwargs-forwarding change." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 2, + "component_count": 30, + "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": "c195102c5d8f4c12bd6a08e2fd3b0c0ecd5130c20ee5dc408aebab8cfdf44754", + "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 PRs target VibeVoice support, but they are adjacent, not duplicate, changes: #40546 implements core VibeVoice integration, while #42781 adds a separate realtime variant plus an acoustic tokenizer and related updates.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "No true duplicate PR is evident. If one must serve as the representative, #42781 is the more complete and newer VibeVoice integration, but it extends the earlier work rather than fixing the same concrete code path.", + "best_issue_reason": null, + "best_pr_reason": "#42781 is the strongest cluster representative because it covers the broader, newer VibeVoice integration surface and includes the extra realtime/tokenizer components, whereas #40546 is the initial core implementation.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40546", + "right": "pull_request:42781", + "accept": false, + "reason": "They overlap on VibeVoice support and several shared files, but #42781 adds realtime and acoustic-tokenizer work, so it is not the same concrete change and would not be merged as one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 3, + "component_count": 30, + "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": "5fe4d439564b8f4da6ab04fdc8d78efe1df762eb02fb46c6ee560e811fd4c8e9", + "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 unrelated: one adds TimesFM covariate forecasting support, the other adds DINOv3 image-classification support. They touch different model families, code paths, and filenames, so they should not be clustered as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:40755", + "right": "pull_request:41224", + "accept": false, + "reason": "Different underlying changes: TimesFM forecasting with covariates versus DINOv3 classification support. No shared bug or mergeable code-path overlap; they belong to separate PRs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 4, + "component_count": 30, + "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": 30, + "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": "b098863a567bbd3fbbc235ed1496d7fefa0a89f3a5280b0d8815f4746edd17b7", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10436, + "estimated_input_tokens": 2609, + "estimated_eval_tokens": 5474 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10436, + "estimated_input_tokens": 2609, + "estimated_eval_tokens": 5474 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "All three PRs are similar in structure because they add new model integrations, docs, auto-mappings, and tests, but they are for different models (DeepSeek OCR, Isaac, Kimi2-6) and do not fix the same underlying change.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical duplicate PR here: each PR introduces a distinct model integration with its own files and model-specific code paths.", + "best_issue_reason": null, + "best_pr_reason": "PR #45630 is the best representative only because it is the most comprehensive and recent of the three model-addition PRs, but it should not be treated as a duplicate of the others.", + "soft_edge_verdicts": [ + { + "left": "pull_request:41797", + "right": "pull_request:45186", + "accept": false, + "reason": "Different target models and codebases: DeepSeek OCR vs Isaac. The overlap is just the standard new-model integration template, not the same concrete change." + }, + { + "left": "pull_request:45186", + "right": "pull_request:45630", + "accept": false, + "reason": "Different models and different implementation targets: Isaac vs Kimi2-6. Shared auto/docs/test scaffolding is too broad to treat as the same PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 6, + "component_count": 30, + "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": "9eb3835338f5f389a9d8c4c3841dd9b6adcb2de9b4bd6446cba5d3f8ee98df48", + "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 upgrades common GitHub Actions for Node 24 compatibility, the other bumps docker/build-push-action across Docker image workflows. They overlap in broad CI upkeep but target different actions and different failure modes.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "No clear duplicate pair here; if one PR must be treated as the representative, PR #42975 is the broader GitHub Actions compatibility update across many workflows and the more central maintenance change.", + "best_issue_reason": null, + "best_pr_reason": "PR #42975 is the better representative because it touches more workflow files and addresses a concrete platform compatibility goal, while PR #42976 is a narrower docker-action version bump.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42975", + "right": "pull_request:42976", + "accept": false, + "reason": "Both are GitHub Actions version updates, but they modify different actions (checkout/cache/upload-artifact vs docker/build-push-action) and fix different concerns. They could be merged independently and do not look like 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": 7, + "component_count": 30, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-42978-4", + "nodes": [ + "pull_request:42978", + "pull_request:43448", + "pull_request:43451", + "pull_request:45490" + ], + "soft_pairs": [ + "pull_request:42978|pull_request:43448", + "pull_request:42978|pull_request:43451", + "pull_request:42978|pull_request:45490", + "pull_request:43448|pull_request:43451", + "pull_request:43448|pull_request:45490", + "pull_request:43451|pull_request:45490" + ], + "prepared_review_unit_hash": "6da1e23d73484be7f7f11082c1355628621c9111a22617bc65278764412d43d6", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 14184, + "estimated_input_tokens": 3546, + "estimated_eval_tokens": 7348 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 14184, + "estimated_input_tokens": 3546, + "estimated_eval_tokens": 7348 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "All four PRs are independent model-addition efforts. They share generic Transformers boilerplate (docs, auto mappings, toctree) but target different architectures and code paths, so none should be clustered as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #43451 (Molmo2) is the strongest representative of the cluster by scope and discussion volume, with the broadest implementation surface across modeling, processing, video support, auto-mappings, and tests.", + "best_issue_reason": null, + "best_pr_reason": "PR #43451 is the best single representative because it is the most substantial and complete model-addition change in the set.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42978", + "right": "pull_request:43448", + "accept": false, + "reason": "ViT NEPA and Molmo are different model families with separate configs, model code, and tests; shared docs/auto-registration boilerplate is not enough to indicate the same underlying change." + }, + { + "left": "pull_request:42978", + "right": "pull_request:43451", + "accept": false, + "reason": "ViT NEPA and Molmo2 are unrelated model additions affecting different architectures and code paths despite both touching Transformer registry/docs files." + }, + { + "left": "pull_request:42978", + "right": "pull_request:45490", + "accept": false, + "reason": "ViT NEPA and CTSM are distinct model implementations for different domains, and the overlap is only in generic integration files." + }, + { + "left": "pull_request:43448", + "right": "pull_request:43451", + "accept": false, + "reason": "Molmo and Molmo2 are separate model variants with different processing and modeling stacks; they are not the same concrete bug or change and would not plausibly merge into one PR." + }, + { + "left": "pull_request:43448", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo and CTSM are unrelated model additions; the shared filenames are just registry/docs updates common to many PRs." + }, + { + "left": "pull_request:43451", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo2 and CTSM target different model architectures and implementation paths, so this is only broad subsystem overlap, not a duplicate fix/change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 8, + "component_count": 30, + "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": "76545cff41088254b3faa6d885cec3ebf3f6bb097f38c2143471c382e5b5c332", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7093, + "estimated_input_tokens": 1774, + "estimated_eval_tokens": 3804 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7093, + "estimated_input_tokens": 1774, + "estimated_eval_tokens": 3804 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These three PRs are not duplicates of one another: one targets Whisper decoding/tests, one is just unrelated CLIP/ViT test edits, and one is a real FP8 dequant/model-loading fix. The shared similarity is only superficial.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #44815 is the only substantive code fix here; it addresses a concrete FP8 dequant/loading path and is much more representative than the test-only or test-noise PRs.", + "best_issue_reason": null, + "best_pr_reason": "PR #44815 is the best standalone PR in the set because it contains an actual implementation change with a clear bug fix, whereas #43291 and #43665 are test-focused and appear unrelated to the same underlying problem.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43291", + "right": "pull_request:43665", + "accept": false, + "reason": "Both touch tests, but they target different models and different problems: Whisper tokenizer/decode tests vs CLIP/ViT test assertions. No shared concrete bug or change path." + }, + { + "left": "pull_request:43291", + "right": "pull_request:44815", + "accept": false, + "reason": "Whisper tokenization/test changes are unrelated to FP8 dequantization and model-loading logic. Different subsystem, different bug, no plausible single merged fix." + }, + { + "left": "pull_request:43665", + "right": "pull_request:44815", + "accept": false, + "reason": "CLIP/ViT test-only edits do not match the FP8 dequant/loading implementation fix. 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": 9, + "component_count": 30, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43488-5", + "nodes": [ + "pull_request:43488", + "pull_request:43532", + "pull_request:45932", + "pull_request:46100", + "pull_request:46114" + ], + "soft_pairs": [ + "pull_request:43488|pull_request:46114", + "pull_request:43488|pull_request:43532", + "pull_request:43532|pull_request:46114", + "pull_request:45932|pull_request:46100", + "pull_request:46100|pull_request:46114" + ], + "prepared_review_unit_hash": "5c8d59a11221d973edb8b9fff8c86e97513d815c6ed79f307b54267859bd05b5", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 5, + "serialized_chars": 12263, + "estimated_input_tokens": 3066, + "estimated_eval_tokens": 6388 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 5, + "serialized_chars": 12263, + "estimated_input_tokens": 3066, + "estimated_eval_tokens": 6388 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are mostly unrelated synthetic \"do not merge\" PRs used to test bot/CI behavior. The only mildly similar pair (45932, 46100) still targets different test-fixture goals, so I do not treat any pair as true duplicates.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46114 is the broadest representative of the cluster\u2019s synthetic test/migration theme, with CI, model, and test-file changes that best summarize the fixture-like nature of the set.", + "best_issue_reason": null, + "best_pr_reason": "PR 46114 is the best global representative because it spans the widest mix of CI and test scaffolding among the dummy PRs, making it the clearest canonical example of the cluster.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43488", + "right": "pull_request:46114", + "accept": false, + "reason": "Both are synthetic test PRs, but they touch different scaffolding: 43488 is a repo-bot/version/setup tweak around ViT tests, while 46114 is a CI/OTEL migration fixture. Not the same underlying change." + }, + { + "left": "pull_request:43488", + "right": "pull_request:43532", + "accept": false, + "reason": "Shared Vit test file is incidental; 43488 is setup/version plumbing, while 43532 rewires comment-driven workflow and notification logic. Different dummy changes." + }, + { + "left": "pull_request:43532", + "right": "pull_request:46114", + "accept": false, + "reason": "Both are bot/CI-related, but one is a self-comment workflow rewrite and the other is a forked migration/OTEL test PR. They do not fix the same code-path problem." + }, + { + "left": "pull_request:45932", + "right": "pull_request:46100", + "accept": false, + "reason": "Both are test-oriented, but 45932 changes multiple CI/model files for a Grafana/OTEL-style fixture, whereas 46100 only injects failing assertions into Vit tests. Not mergeable as one PR." + }, + { + "left": "pull_request:46100", + "right": "pull_request:46114", + "accept": false, + "reason": "Shared test-labelling is too broad; 46100 is a narrow Vit test-migration stub, while 46114 is a wider CI/model/test migration fixture. Different underlying changes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 10, + "component_count": 30, + "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": "a1d6914a2f6a41f8edc1e78e6a77d9a421ce987b54491c07fe3b03917d2f44bf", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6082, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6082, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related but distinct pull requests for different segmentation pipelines: one adds Promptable Concept Segmentation, the other adds Promptable Visual Segmentation. Their overlap is mostly shared documentation and registration plumbing, not the same underlying bug or change.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR as a duplicate target: the two PRs implement different pipelines with different model/support files and different concrete behavior.", + "best_issue_reason": null, + "best_pr_reason": "PR 43613 is the broader representative of the pair because it adds the more expansive Promptable Visual Segmentation pipeline and touches more related model/tests/docs, but it is still a separate change from 43612.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43612", + "right": "pull_request:43613", + "accept": false, + "reason": "Reject: they target different pipeline features and different code paths/models. Shared docs, pipeline registration, and metadata updates are incidental and not enough to treat them as one mergeable PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 11, + "component_count": 30, + "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-43649-2", + "nodes": [ + "pull_request:43649", + "pull_request:46301" + ], + "soft_pairs": [ + "pull_request:43649|pull_request:46301" + ], + "prepared_review_unit_hash": "fa56513447ac7cdc943a0f517b73ee28337e9a3979c00f844c23feb3d0720a26", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4550, + "estimated_input_tokens": 1138, + "estimated_eval_tokens": 2532 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4550, + "estimated_input_tokens": 1138, + "estimated_eval_tokens": 2532 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two pull requests are not duplicates: one is about CI/self-comment failure reporting workflows, while the other changes CircleCI test selection and tensor-parallel test handling. They touch different code paths and would not plausibly be merged into one PR.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: the changes address unrelated areas (workflow/reporting automation vs. CircleCI marker/test behavior) and do not fix the same concrete bug or implement the same change.", + "best_issue_reason": null, + "best_pr_reason": "PR 43649 is the better representative only because it has a clearer, coherent title and a more self-contained CI/reporting theme; PR 46301 is very generic ('check') and mixes CircleCI config with tensor-parallel test adjustments.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43649", + "right": "pull_request:46301", + "accept": false, + "reason": "Different underlying changes: 43649 modifies workflow/comment-based failure reporting, while 46301 adjusts CircleCI job markers and tensor-parallel test execution. They are not the same bug/change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 12, + "component_count": 30, + "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": "5e68d8c7ae6bf657bb26c8831a86307bebedba11bc8cd0b5e75b0a7076a5b27e", + "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 independent Trainer enhancements in the same file: one changes gradient-accumulation loss scaling, the other factors out signature-column filtering for custom trainers/PEFT. They do not look like the same bug or mergeable change.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "No issue is present. Among the PRs, 43651 is the more canonical representative because it changes a core training-path behavior, has more discussion activity, and is the more substantive trainer semantics change.", + "best_issue_reason": null, + "best_pr_reason": "43651 is the best representative PR for the cluster because it touches the main training_step loss-scaling path and has broader behavioral impact than the column-filtering refactor in 43663.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43651", + "right": "pull_request:43663", + "accept": false, + "reason": "Different changes in the same subsystem but not the same underlying problem: 43651 is about gradient accumulation loss scaling, while 43663 is about extracting signature-column filtering for custom trainers/PEFT. They would not plausibly be merged into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 13, + "component_count": 30, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43656-2", + "nodes": [ + "pull_request:43656", + "pull_request:43842" + ], + "soft_pairs": [ + "pull_request:43656|pull_request:43842" + ], + "prepared_review_unit_hash": "7e972a73e3f88f477ca7ecb3ef80ec3135f5e95b7f3f7811792a32a5478b9d38", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 3663, + "estimated_input_tokens": 916, + "estimated_eval_tokens": 2088 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 3663, + "estimated_input_tokens": 916, + "estimated_eval_tokens": 2088 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two PRs touch the same file and both mention a TypeAdapter NameError in the CLI, but they appear to address different underlying fixes rather than the same exact bug/change.", + "confidence": 0.82, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43656 is the broader, more central fix in src/transformers/cli/serve.py and appears to address the annotation-evaluation NameError directly, with wider typing cleanup. PR 43842 is a narrower follow-up focused on the optional pydantic import path.", + "best_issue_reason": null, + "best_pr_reason": "PR 43656 is the best representative of the cluster because it is the more complete change to the CLI typing/runtime-annotation problem, while PR 43842 is a targeted variant for a specific dependency scenario.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43656", + "right": "pull_request:43842", + "accept": false, + "reason": "Same file and similar symptom, but the fixes are not the same concrete change: 43656 adds future-annotations/typing cleanup, while 43842 explicitly moves TypeAdapter under TYPE_CHECKING and quotes the annotation to handle the pydantic-absent case. These are related but distinct solutions and not clearly 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": 14, + "component_count": 30, + "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": 15, + "component_count": 30, + "review_unit_index": 1, + "review_unit_count": 2, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43996-28", + "nodes": [ + "pull_request:43996", + "pull_request:43997", + "pull_request:44007", + "pull_request:44010", + "pull_request:44013", + "pull_request:44017", + "pull_request:44018", + "pull_request:44019", + "pull_request:44044", + "pull_request:44056", + "pull_request:44059", + "pull_request:44066", + "pull_request:44068", + "pull_request:44071", + "pull_request:44072", + "pull_request:44073", + "pull_request:44074", + "pull_request:44076", + "pull_request:44085", + "pull_request:44086", + "pull_request:44098", + "pull_request:44101", + "pull_request:44116", + "pull_request:44129", + "pull_request:44154", + "pull_request:44161", + "pull_request:44722" + ], + "soft_pairs": [ + "pull_request:43996|pull_request:44085", + "pull_request:43996|pull_request:44044", + "pull_request:44066|pull_request:44085", + "pull_request:44007|pull_request:44072", + "pull_request:44072|pull_request:44722", + "pull_request:44066|pull_request:44072", + "pull_request:44013|pull_request:44044", + "pull_request:44066|pull_request:44086", + "pull_request:44066|pull_request:44071", + "pull_request:44018|pull_request:44068", + "pull_request:44018|pull_request:44066", + "pull_request:44066|pull_request:44068", + "pull_request:44019|pull_request:44722", + "pull_request:44086|pull_request:44722", + "pull_request:44019|pull_request:44085", + "pull_request:44018|pull_request:44019", + "pull_request:44019|pull_request:44071", + "pull_request:44068|pull_request:44116", + "pull_request:44018|pull_request:44116", + "pull_request:44068|pull_request:44722", + "pull_request:44071|pull_request:44722", + "pull_request:44018|pull_request:44085", + "pull_request:44018|pull_request:44086", + "pull_request:44068|pull_request:44086", + "pull_request:44071|pull_request:44085", + "pull_request:44059|pull_request:44086", + "pull_request:44085|pull_request:44086", + "pull_request:44098|pull_request:44101", + "pull_request:44059|pull_request:44722", + "pull_request:44018|pull_request:44071", + "pull_request:44017|pull_request:44722", + "pull_request:44017|pull_request:44161", + "pull_request:44017|pull_request:44073", + "pull_request:44068|pull_request:44071", + "pull_request:43997|pull_request:44073", + "pull_request:44071|pull_request:44086", + "pull_request:44017|pull_request:44071", + "pull_request:44073|pull_request:44101", + "pull_request:44059|pull_request:44071", + "pull_request:44018|pull_request:44059", + "pull_request:44073|pull_request:44161", + "pull_request:44056|pull_request:44101", + "pull_request:44059|pull_request:44068", + "pull_request:44101|pull_request:44161", + "pull_request:44017|pull_request:44018", + "pull_request:43997|pull_request:44056", + "pull_request:44071|pull_request:44076", + "pull_request:44074|pull_request:44101", + "pull_request:43997|pull_request:44018", + "pull_request:44056|pull_request:44161", + "pull_request:43997|pull_request:44161", + "pull_request:44074|pull_request:44161", + "pull_request:44018|pull_request:44161", + "pull_request:43997|pull_request:44071", + "pull_request:44010|pull_request:44071", + "pull_request:44018|pull_request:44129", + "pull_request:44010|pull_request:44073", + "pull_request:43997|pull_request:44074", + "pull_request:44071|pull_request:44129", + "pull_request:44010|pull_request:44018", + "pull_request:43997|pull_request:44129", + "pull_request:43996|pull_request:44161", + "pull_request:43996|pull_request:44017", + "pull_request:44010|pull_request:44044", + "pull_request:44044|pull_request:44074", + "pull_request:44066|pull_request:44722", + "pull_request:44085|pull_request:44722", + "pull_request:44085|pull_request:44129", + "pull_request:43996|pull_request:44007", + "pull_request:43996|pull_request:44129", + "pull_request:44044|pull_request:44129", + "pull_request:44044|pull_request:44154" + ], + "prepared_review_unit_hash": "2de69f7eaaed316090d1bf8248f1c5198acfd6a4c2ad347fe7bc15d1d760ad83", + "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": 75095, + "estimated_input_tokens": 18774, + "estimated_eval_tokens": 37804 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is mostly a thematically similar wave of output-tracing refactors across many different models, not a single duplicate PR. The clearest true duplicates are the GPT-J/CodeGen revisions and the GPT-Neo revision pair; the rest are model-specific changes that should stay separate.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44722 is the cleanest representative of the main true-duplicate subcluster: it targets the same GPT-J/CodeGen output-tracing refactor, has an explicit tracking issue, and is the broadest/latest version of that concrete change.", + "best_issue_reason": null, + "best_pr_reason": "PR 44722 is the best anchor for deduping the GPT-J/CodeGen subcluster because it addresses the same concrete code-path change with the most complete scope among the duplicate candidates.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43996", + "right": "pull_request:44085", + "accept": false, + "reason": "Different models and code paths (CVT/FNet vs GPT-J); same issue target alone is not enough." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44044", + "accept": false, + "reason": "Different model families and refactor scopes (CVT/FNet vs DeBERTa)." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44085", + "accept": true, + "reason": "Same GPT-J output-tracing refactor in the same model file; these are overlapping revisions of the same concrete change." + }, + { + "left": "pull_request:44007", + "right": "pull_request:44072", + "accept": false, + "reason": "Both are output-tracing refactors, but for different CNN architectures and separate code paths." + }, + { + "left": "pull_request:44072", + "right": "pull_request:44722", + "accept": false, + "reason": "EfficientNet and GPT-J are unrelated model code paths; broad theme overlap is not a duplicate." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44072", + "accept": false, + "reason": "GPT-J/CodeGen refactor vs EfficientNet refactor; different underlying change." + }, + { + "left": "pull_request:44013", + "right": "pull_request:44044", + "accept": false, + "reason": "MobileNetV2 and DeBERTa are unrelated implementations." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J/CodeGen output tracing vs MGP-STR output tracing; different model-specific fixes." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-J/CodeGen and MPT are distinct model code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44068", + "accept": true, + "reason": "Same GPT-Neo output-tracing refactor in the same model file; looks like duplicate revisions of one change." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44066", + "accept": false, + "reason": "GPT-Neo vs GPT-J/CodeGen; same refactor pattern but different models." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-J/CodeGen vs GPT-Neo; related theme, not the same code-path change." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44722", + "accept": false, + "reason": "ResNet refactor vs GPT-J/CodeGen refactor; different model code paths." + }, + { + "left": "pull_request:44086", + "right": "pull_request:44722", + "accept": false, + "reason": "MGP-STR and GPT-J/CodeGen are unrelated model 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 code paths." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44071", + "accept": false, + "reason": "ResNet and MPT are different models; only the general refactor theme overlaps." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo vs Flaubert; different model-specific output tracing changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert are unrelated implementations." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-Neo and GPT-J/CodeGen are separate model code paths." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44722", + "accept": false, + "reason": "MPT and GPT-J/CodeGen are different models." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo and GPT-J are different model-specific refactors." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo vs MGP-STR; not the same underlying bug or change." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo and MGP-STR are unrelated code paths." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44086", + "accept": false, + "reason": "MPT and MGP-STR are distinct model implementations." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-2 and MGP-STR are unrelated model code paths." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J and MGP-STR are different models; same output-tracing theme is too broad." + }, + { + "left": "pull_request:44098", + "right": "pull_request:44101", + "accept": false, + "reason": "ViLT and Flaubert are different model families." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-2 vs GPT-J/CodeGen are separate code paths." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44722", + "accept": false, + "reason": "SegFormer and GPT-J/CodeGen are unrelated model changes." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44161", + "accept": false, + "reason": "SegFormer and LongT5 are different model-specific refactors." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44073", + "accept": false, + "reason": "SegFormer and VisualBERT are unrelated code paths." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44056", + "accept": false, + "reason": "RegNet and MPNet are different model implementations." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44161", + "accept": false, + "reason": "RegNet and LongT5 are unrelated model-specific changes." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44161", + "accept": false, + "reason": "MPNet and LongT5 are different code paths." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44017", + "accept": false, + "reason": "RegNet and SegFormer are distinct models." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J/CodeGen output-tracing refactor with the same concrete files; 44722 is a later, broader revision of the same change." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J output-tracing change in the same code path; 44722 subsumes this narrower revision." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-J and SpeechT5 are unrelated model code paths." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44007", + "accept": false, + "reason": "CVT/FNet vs RegNet/ResNet; same refactor theme but not the same underlying change." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44129", + "accept": false, + "reason": "CVT/FNet and SpeechT5 are different model implementations." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44129", + "accept": false, + "reason": "DeBERTa and SpeechT5 are unrelated model code paths." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44154", + "accept": false, + "reason": "DeBERTa and VITS are different models; not mergeable as one duplicate PR." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative about the non-duplicate pairs. The accepted duplicate pairs are plausibly mergeable as revisions of the same GPT-J/CodeGen and GPT-Neo fixes, and the rest are correctly treated as model-specific changes rather than one shared bugfix. The only minor caveat is that phrases like \"most complete scope\" are somewhat interpretive, but not overstated enough to reject." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 15, + "component_count": 30, + "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": "0a504e473fc0ff7034db79e7708bc1ed08a1ca619939031f4b5f0f75436fea71", + "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": 17122, + "estimated_input_tokens": 4281, + "estimated_eval_tokens": 8818 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a set of model-specific refactors to standardized output tracing/capture_outputs across different architectures, not duplicate fixes of the same underlying bug or change. They share a theme but target different code paths and would not be merged as one PR.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single PR is a true duplicate canonical for the cluster; the closest representative is the most complete standardized-output refactor among the model-specific changes.", + "best_issue_reason": null, + "best_pr_reason": "PR 44073 is the clearest representative of the shared pattern because it explicitly wires output-capture metadata and migrates a model to the standardized tracing decorators in a self-contained way.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44024", + "right": "pull_request:44076", + "accept": false, + "reason": "Different models (FocalNet vs ImageGPT) and different implementation details. Same refactor theme, but not the same concrete code-path change." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44074", + "accept": false, + "reason": "Both are output-tracing refactors, but they affect different architectures and separate model logic; not mergeable as one PR." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44073", + "accept": false, + "reason": "MPNet and VisualBert are distinct model implementations. Shared capture_outputs plumbing is a pattern, not a duplicate change." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44074", + "accept": false, + "reason": "Different models and different forward/output handling. Similar modernization, but not the same bug or change." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44010", + "accept": false, + "reason": "RegNet and SqueezeBert are unrelated model code paths. These are separate migrations to standardized output handling, not duplicates." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44129", + "accept": false, + "reason": "SqueezeBert vs SpeechT5: distinct implementations and refactor scopes. They share output-capture modernization but are not the same underlying change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 16, + "component_count": 30, + "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: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": "df9f3158a8bec9efd4d9df3e885aa0ae47d0f018e66d7daaf1ce8596550417f7", + "component_budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 39, + "serialized_chars": 39793, + "estimated_input_tokens": 9949, + "estimated_eval_tokens": 20154 + }, + "budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 39, + "serialized_chars": 39793, + "estimated_input_tokens": 9949, + "estimated_eval_tokens": 20154 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The cluster is a set of model-specific 'refactor output tracing' pull requests. They are thematically similar but target different architectures and code paths, so they should not be treated as duplicate fixes. The best representative PR is 44025 because it is a clear, multi-file example of the shared refactor pattern.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single PR is a true canonical duplicate for the cluster; 44025 is the closest representative because it is a concrete output-tracing refactor spanning two closely related files in one model family.", + "best_issue_reason": null, + "best_pr_reason": "44025 is the best global representative because it shows the shared pattern cleanly and covers two related depth-anything implementations, while still being a specific, mergeable code change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43998", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model families (timm_backbone vs upernet); same refactor theme, but not the same concrete code-path change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44000", + "accept": false, + "reason": "Different architectures (timm_backbone vs vision_text_dual_encoder); similar tracing refactor, not a duplicate fix." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44001", + "accept": false, + "reason": "Unrelated model implementations; both refactor output handling, but they are separate code paths." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44003", + "accept": false, + "reason": "timm_backbone and mamba/falcon_mamba are unrelated model families; not the same underlying bug or change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44029", + "accept": false, + "reason": "Different model families and output mechanics; only the high-level refactor motif matches." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44001", + "accept": false, + "reason": "MobileNetV1 and UnivNet are unrelated models; the overlap is only the generic output-tracing refactor pattern." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44002", + "accept": false, + "reason": "Different subsystems and model code paths; not a single mergeable fix." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44025", + "accept": false, + "reason": "MobileNetV1 vs depth_anything are separate model families; same style of refactor, not duplicate behavior." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44026", + "accept": false, + "reason": "MobileNetV1 and vision_encoder_decoder do not share the same concrete code path." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44027", + "accept": false, + "reason": "MobileNetV1 and speech_encoder_decoder are unrelated; only the generic tracing refactor overlaps." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44001", + "accept": false, + "reason": "vision_text_dual_encoder and univnet are different models with different forward paths; not the same change." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model families; both adjust output tracing, but not the same underlying fix." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44003", + "accept": false, + "reason": "vision_text_dual_encoder and mamba/codegen are unrelated code paths." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44025", + "accept": false, + "reason": "No shared concrete bug: one is vision-text dual encoder, the other is depth-anything refactoring." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44002", + "accept": false, + "reason": "Unrelated model families; the similarity is only in output tracing cleanup." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44003", + "accept": false, + "reason": "UnivNet and mamba/codegen do not appear to fix the same code-path problem." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44025", + "accept": false, + "reason": "Different model families and different forward implementations; not mergeable as one PR." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44029", + "accept": false, + "reason": "UperNet and RWKV are unrelated architectures; same refactor motif, not duplicate behavior." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44003", + "accept": false, + "reason": "Different subsystems and fixes; not the same concrete change." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44004", + "accept": false, + "reason": "Mamba/falcon_mamba versus CodeGen are unrelated code paths; the overlap is only the output-tracing refactor theme." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44004", + "accept": false, + "reason": "timm_backbone and codegen are different model families with different forward logic." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "UnivNet and CodeGen are unrelated; similar refactor pattern is not enough to duplicate them." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44004", + "accept": false, + "reason": "UperNet and CodeGen are separate implementations; not the same underlying bug." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44004", + "accept": false, + "reason": "vision_text_dual_encoder and codegen are different code paths and different model domains." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44026", + "accept": false, + "reason": "DepthAnything and vision_encoder_decoder are different model families; same refactor style only." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44027", + "accept": false, + "reason": "DepthAnything and speech_encoder_decoder are unrelated code paths." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44028", + "accept": false, + "reason": "DepthAnything and SuperPoint are different models; not a single concrete fix." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44029", + "accept": false, + "reason": "DepthAnything and RWKV do not share the same implementation issue." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44030", + "accept": false, + "reason": "DepthAnything and DPR are unrelated architectures; same refactor motif is too broad." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44029", + "accept": false, + "reason": "VisionEncoderDecoder and RWKV are different model families with different output paths." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44027", + "accept": false, + "reason": "VisionEncoderDecoder and SpeechEncoderDecoder are analogous wrapper models, but still separate code paths and not one mergeable fix." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44028", + "accept": false, + "reason": "Different model implementations; no shared concrete bug beyond tracing cleanup." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44030", + "accept": false, + "reason": "VisionEncoderDecoder and DPR are unrelated; not the same change." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44028", + "accept": false, + "reason": "SpeechEncoderDecoder and SuperPoint are unrelated models; only the generic output-tracing pattern matches." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44029", + "accept": false, + "reason": "SpeechEncoderDecoder and RWKV are different code paths and architectures." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44030", + "accept": false, + "reason": "SpeechEncoderDecoder and DPR do not appear to fix the same concrete issue." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44029", + "accept": false, + "reason": "SuperPoint and RWKV are unrelated model families." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44030", + "accept": false, + "reason": "SuperPoint and DPR are different code paths; not duplicate changes." + }, + { + "left": "pull_request:44029", + "right": "pull_request:44030", + "accept": false, + "reason": "RWKV and DPR are unrelated architectures, so this is not the same underlying fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 17, + "component_count": 30, + "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": "728e7b17fb2371a5cd141f8b2b6d7c46730ba48586e2465d16504f187cb9a760", + "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": "All three pull requests are separate model-addition implementations, not duplicates. They share generic Transformers scaffolding (auto registration, docs, package wiring), but each targets a different model family and code path: PenguinVL, DeepSeek-OCR-2, and Voxtral TTS.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR for a duplicate cluster: the three PRs introduce different models and do not appear to fix or implement the same concrete change.", + "best_issue_reason": null, + "best_pr_reason": "PR 45075 is the strongest representative only by maturity/activity, but it is still a distinct DeepSeek-OCR-2 model-addition PR rather than a duplicate of the others.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44662", + "right": "pull_request:45401", + "accept": false, + "reason": "Different model additions and different problem domains: PenguinVL (vision-language) vs Voxtral TTS (text-to-speech). Shared auto/docs wiring is generic and not evidence of the same underlying change." + }, + { + "left": "pull_request:44662", + "right": "pull_request:45075", + "accept": false, + "reason": "PenguinVL and DeepSeek-OCR-2 are separate model implementations with different architectures and assets; overlap is only in framework registration/docs scaffolding, so they are not the same code-path fix." + }, + { + "left": "pull_request:45075", + "right": "pull_request:45401", + "accept": false, + "reason": "DeepSeek-OCR-2 and Voxtral TTS address unrelated model families and tasks. They may touch common auto-mapping files, but they are not plausibly mergeable as one PR for the same 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": 18, + "component_count": 30, + "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": "9f481e346682d77dd3ef7bdba072e53e771a4bc0b3046a6d67d4d0b5ece73b8d", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4408, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4408, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are related to module/kernel fusion, but they are not duplicates: one defines a general Module Fusion API, while the other implements n-to-1 kernel fusion through KernelConfig and model/runtime integration.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #44979 is the broader, more foundational change: it introduces the new Module Fusion API and core module_fusion implementation, which is the best representative of this cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR #44979 is the best single PR to represent the cluster because it centers the underlying fusion abstraction rather than one specific KernelConfig application path.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44979", + "right": "pull_request:45363", + "accept": false, + "reason": "They share the general fusion theme, but they address different concrete changes: #44979 adds the module fusion API, while #45363 wires n-to-1 kernel fusion through KernelConfig and related integrations. They would not plausibly be merged as one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 19, + "component_count": 30, + "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": "8b47e36ac54c6c8679ab15a45d64bd382586bfadde51d65496ef1ce64d151cab", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4414, + "estimated_input_tokens": 1104, + "estimated_eval_tokens": 2464 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4414, + "estimated_input_tokens": 1104, + "estimated_eval_tokens": 2464 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both pull requests fix the same Mistral regex patch bug in `tokenization_utils_tokenizers.py` by switching from `tokenizer.backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer`; 45317 is the more complete version because it adds tests.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "45317 is the better canonical PR because it implements the same fix as 45086 and also includes regression test coverage.", + "best_issue_reason": null, + "best_pr_reason": "45317 is the best PR candidate: same concrete fix, plus tests, making it the most complete mergeable representation of the change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45086", + "right": "pull_request:45317", + "accept": true, + "reason": "Same underlying bug fix, same file, same code-path change, and one PR is essentially a test-augmented version of the other." + } + ] + }, + "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": 30, + "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": "b3d9772d1cf179d5e9b1a7f299fc68fa7042e74f91d6d7cfbeee89861baabd21", + "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": "The two PRs are related to NemotronH, but they tackle different changes: one is a conversion/refactor toward GraniteMoeHybrid-compatible naming, while the other splits NemotronH into dense/sparse variants with docs and tests. Shared files reflect the same subsystem, not the same underlying fix.", + "confidence": 0.88, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "PR 45569 is the more complete and user-facing change: it adds the dense/sparse model split, auto-mapping, documentation, and tests, so it is the better representative of this cluster.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45115", + "right": "pull_request:45569", + "accept": false, + "reason": "Reject: both touch NemotronH internals, but 45115 is a narrow structural/conversion refactor, while 45569 is a broader model-family split with docs/tests. They are not the same concrete change and would not plausibly merge as one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 21, + "component_count": 30, + "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": "4f4eaba59426177f515be4cc2c2cc2577223bc91721ab50fed4fbd43db7a7580", + "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 are related but not duplicates: PR #45333 adds the heterogeneous per-layer config plumbing, while PR #45332 builds on that to add heterogeneous model/cache/masking support. They target different layers of the same feature and would not be merged as one change set.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45332 is the broader umbrella change: it includes the modeling/cache integration work and depends on the config machinery added elsewhere, so it is the better representative of the overall feature cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR #45333 is the cleanest standalone representative for the foundational config API because it is narrowly scoped to per-layer configuration plumbing and tests.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45332", + "right": "pull_request:45333", + "accept": false, + "reason": "Not the same concrete change: #45333 introduces heterogeneous config support, while #45332 adds model/cache behavior that consumes that config. They are complementary feature pieces rather than duplicate fixes and would not be plausibly merged into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 22, + "component_count": 30, + "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": "c8a66d8ccc7babff6aa2ba8fb35a4d3eb7aed5f6e55f58463b5eb68c0e890050", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5513, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5513, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs in the typing/type-checking area, but they address different changes: one is a broad pass to add type checking across many `src/transformers/*.py` files, while the other introduces typing protocols/helpers and modeling-related typing fixes. They overlap in a few files, but not the same concrete bug/change.", + "confidence": 0.87, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45425 is the more concrete, self-contained typing refactor with specific helper protocol additions and targeted modeling-related fixes; it better represents a distinct change than the broad type-checking sweep in 45415.", + "best_issue_reason": null, + "best_pr_reason": "PR 45425 is the best representative of the cluster because it has a focused technical goal and clearer code-level scope, whereas 45415 is a broad, cross-file type-checking cleanup.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45415", + "right": "pull_request:45425", + "accept": false, + "reason": "Reject: both are typing-related and share a few files, but 45415 is a broad type-checking sweep while 45425 adds specific typing protocols and modeling fixes; they are not the same underlying change and would not plausibly merge as one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 23, + "component_count": 30, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45513-2", + "nodes": [ + "pull_request:45513", + "pull_request:46084" + ], + "soft_pairs": [ + "pull_request:45513|pull_request:46084" + ], + "prepared_review_unit_hash": "8c3bff436b63fa6c2eba44d9eb09e8d0e39604eb6d09b246a1ad8a7d4b3aef78", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4857, + "estimated_input_tokens": 1215, + "estimated_eval_tokens": 2686 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4857, + "estimated_input_tokens": 1215, + "estimated_eval_tokens": 2686 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both pull requests address cached-forward bugs when `seq_len > 1`, but in different model families and codepaths: OLMo/Qwen hybrid linear attention vs. Mamba2 CUDA kernels. They are thematically similar, not the same underlying change, so they should not be clustered as duplicates.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46084 is the better representative because it has a concrete linked issue, a narrowly scoped fix, and a clear bug statement about `use_cache` with `seq_len > 1` in one specific implementation.", + "best_issue_reason": null, + "best_pr_reason": "PR 46084 is the stronger canonical PR for the cluster since it is explicit, targeted, and less broad than the other PR; PR 45513 spans multiple model variants and a broader refactor-like fix across several files.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45513", + "right": "pull_request:46084", + "accept": false, + "reason": "Both involve cached multi-token forward behavior, but they fix different concrete implementations and likely different bugs: OLMo/Qwen hybrid linear attention versus Mamba2 state-space CUDA kernels. The overlap is only high-level theme, not a shared 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": 24, + "component_count": 30, + "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": "77464b2b886a37677f5e97fe4ae356f789e6eaf7613a25e921dc9ced185d2b14", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4507, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4507, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two PRs are related by expert-parallelism, but they target different bugs and should not be merged as duplicates: one fixes EP/ZeRO-3 loading and config serialization, the other fixes expert parameters being overwritten by rank-0 broadcast in EP+FSDP2.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45662 is the better representative of the cluster because it addresses a more concrete runtime correctness bug in the EP loading path, with changes across the actual code paths that overwrite expert parameters.", + "best_issue_reason": null, + "best_pr_reason": "PR 45662 is the strongest standalone candidate: it fixes a specific expert-overwrite failure mode and has broader, coherent code-path coverage than 45548.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45548", + "right": "pull_request:45662", + "accept": false, + "reason": "Both mention EP, but they fix different concrete problems: 45548 is about EP + DeepSpeed ZeRO-3 loading via accelerate and config handling, while 45662 is about experts being silently overwritten by rank-0 broadcast in EP + FSDP2. They are not the same underlying change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 25, + "component_count": 30, + "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": "e8aa6ef632066fe7e69cba204e2b10af3e56f847474b4b8948e811b3cca35836", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4589, + "estimated_input_tokens": 1148, + "estimated_eval_tokens": 2552 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4589, + "estimated_input_tokens": 1148, + "estimated_eval_tokens": 2552 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs address the same xdist-related captured_info collision in testing_utils; 45645 is the more complete, cluster-representative fix because it also updates CI log collection for the new per-worker files.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45645 is the broader, more complete implementation of the same underlying bug: it makes captured_info output xdist-safe and updates downstream CI handling to aggregate multiple files.", + "best_issue_reason": null, + "best_pr_reason": "PR 45645 is the best representative of the cluster because it covers the core code-path fix plus the necessary CI-side adjustment, whereas 45639 is the narrower code-only slice.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45639", + "right": "pull_request:45645", + "accept": true, + "reason": "Same underlying xdist captured_info collision fix; 45645 extends the same code-path change with compatible CI log aggregation, so they could plausibly be merged into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 26, + "component_count": 30, + "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-45720-3", + "nodes": [ + "pull_request:45720", + "pull_request:46023", + "pull_request:46259" + ], + "soft_pairs": [ + "pull_request:45720|pull_request:46259", + "pull_request:46023|pull_request:46259" + ], + "prepared_review_unit_hash": "ac459517ab879e8296a6f755af7cb937a2aa374f3602a511ed324c4a67764004", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10912, + "estimated_input_tokens": 2728, + "estimated_eval_tokens": 5712 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10912, + "estimated_input_tokens": 2728, + "estimated_eval_tokens": 5712 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are three separate linter-rule PRs, not duplicates: one implements TRF003 for generation outputs, one fixes TRF016 for processor boolean flags, and one adds a different TRF018 modeling rule. The shared docs/setup files and overlapping model areas are too broad to treat any pair as the same change.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46023 is the clearest standalone change: it targets a specific linter rule (TRF016), has a focused description, and the touched files align with that rule\u2019s processor-override checks.", + "best_issue_reason": null, + "best_pr_reason": "PR 46023 is the best representative of the cluster because it is the most self-contained and concretely described implementation of a single rule, whereas the others are separate rule implementations.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45720", + "right": "pull_request:46259", + "accept": false, + "reason": "Different underlying rules and code paths: TRF003 changes model forward behavior for tuple returns, while TRF018 is a separate modeling rule. Shared linter/docs infrastructure and overlapping filenames are not enough to merge them." + }, + { + "left": "pull_request:46023", + "right": "pull_request:46259", + "accept": false, + "reason": "Both are linter-rule PRs, but they implement different rules (TRF016 vs TRF018) with different target classes and fixes. They are related infrastructure work, not the same bug/change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 27, + "component_count": 30, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45737-2", + "nodes": [ + "pull_request:45737", + "pull_request:45738" + ], + "soft_pairs": [ + "pull_request:45737|pull_request:45738" + ], + "prepared_review_unit_hash": "cc2ade73a49ec6a8883a7a24c433988645588bcd049e8aad2bce56b1b24daa1c", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4131, + "estimated_input_tokens": 1033, + "estimated_eval_tokens": 2322 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4131, + "estimated_input_tokens": 1033, + "estimated_eval_tokens": 2322 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs address the same MusicgenMelody cache initialization bug in `forward()`: replacing `EncoderDecoderCache(DynamicCache, DynamicCache)` with a single `DynamicCache`. PR #45737 is the more complete duplicate because it bundles the fix with a regression test, while #45738 is the fix-only variant.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45737 is the best canonical PR because it contains the same code fix as #45738 plus a regression test, making it the most complete representation of the bug fix.", + "best_issue_reason": null, + "best_pr_reason": "PR #45737 is the strongest overall candidate: it fixes the concrete cache bug and adds a targeted regression test, whereas #45738 only changes the implementation.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45737", + "right": "pull_request:45738", + "accept": true, + "reason": "Same underlying code-path bug in `musicgen_melody` cache setup; the diffs match on the core fix and the test-only addition in #45737 could plausibly live in the same PR as #45738." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The assessment is grounded in the packet: both PRs make the same `musicgen_melody` cache initialization change, and #45737 only adds a regression test on top of that fix. The soft-edge accept is conservative enough because the diffs are mergeable into one PR for the same bugfix." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 28, + "component_count": 30, + "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-46073-3", + "nodes": [ + "pull_request:46073", + "pull_request:46154", + "pull_request:46196" + ], + "soft_pairs": [ + "pull_request:46073|pull_request:46154", + "pull_request:46073|pull_request:46196", + "pull_request:46154|pull_request:46196" + ], + "prepared_review_unit_hash": "b55001d1a33e05900e9b52d10dbb02dcf58fc8408fa8e49c5f353f1ed902272a", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 8081, + "estimated_input_tokens": 2021, + "estimated_eval_tokens": 4298 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 8081, + "estimated_input_tokens": 2021, + "estimated_eval_tokens": 4298 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Three open PRs all add SDPA dispatch support, but they target different model families and code paths. They are related by theme, not duplicates of the same fix.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #46196 is the broadest representative of the SDPA-dispatch theme, covering multiple model implementations and test updates, so it best represents the cluster even though it is not a duplicate of the others.", + "best_issue_reason": null, + "best_pr_reason": "PR #46196 is the strongest standalone PR in the set because it spans the most models and includes the most complete implementation/test coverage for the same feature area.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46073", + "right": "pull_request:46154", + "accept": false, + "reason": "Both add SDPA dispatch support, but one is for CTRL and the other for T5; they touch different model code paths and are not the same concrete bug or mergeable change." + }, + { + "left": "pull_request:46073", + "right": "pull_request:46196", + "accept": false, + "reason": "These are model-specific SDPA enablement PRs for different architectures (CTRL vs wav2vec2/seamless variants); same broad feature goal, but not the same underlying fix." + }, + { + "left": "pull_request:46154", + "right": "pull_request:46196", + "accept": false, + "reason": "T5 SDPA dispatch and the speech/multimodal SDPA work in #46196 are separate implementation efforts in different code paths, so they should not be merged as duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 29, + "component_count": 30, + "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-46126-2", + "nodes": [ + "pull_request:46126", + "pull_request:46269" + ], + "soft_pairs": [ + "pull_request:46126|pull_request:46269" + ], + "prepared_review_unit_hash": "91a6873f4f72ffde6168aef1f96d4074f6deb790013b30abc2255177000a4da1", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5570, + "estimated_input_tokens": 1393, + "estimated_eval_tokens": 3042 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5570, + "estimated_input_tokens": 1393, + "estimated_eval_tokens": 3042 + }, + "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 focused EP/Trainer integration change, while the other is a broad distributed branch base with many unrelated edits. They overlap in a few distributed files, but not the same concrete code-path fix.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46126 is the more specific, self-contained change centered on EP routing and Trainer integration; PR 46269 is a broad branch-base PR touching many distributed areas and is not a clean canonical duplicate target.", + "best_issue_reason": null, + "best_pr_reason": "PR 46126 best represents the cluster because it has a clear, narrow purpose and concrete implementation details, whereas PR 46269 is too broad and mixed-purpose to serve as a good representative.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46126", + "right": "pull_request:46269", + "accept": false, + "reason": "Reject: shared distributed filenames are not enough here. 46126 is about EP router masking and trainer integration; 46269 is a large distributed branch-base PR with many additional changes, so they do not appear to be the same underlying bug/change or 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": 30, + "component_count": 30, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46152-2", + "nodes": [ + "pull_request:46152", + "pull_request:46207" + ], + "soft_pairs": [ + "pull_request:46152|pull_request:46207" + ], + "prepared_review_unit_hash": "c8db8088063215f3b854ec41f80a0e28754df8c6713d624e3c246501634ae36b", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4276, + "estimated_input_tokens": 1069, + "estimated_eval_tokens": 2394 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4276, + "estimated_input_tokens": 1069, + "estimated_eval_tokens": 2394 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two PRs share only a broad theme of avoiding direct file-open source inspection, but they target different code paths and fixes. They should not be treated as duplicates or merged as one change.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR here: PR 46152 updates import_utils source/traversable resolution, while PR 46207 changes _can_set_attn_implementation in modeling_utils. They address different behaviors and are not the same fix.", + "best_issue_reason": null, + "best_pr_reason": "Neither PR is a good global representative of the cluster as a duplicate target; the overlap is only thematic, not a shared concrete bug or change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46152", + "right": "pull_request:46207", + "accept": false, + "reason": "Reject: both avoid file-open-based source reading, but they fix different code paths in different modules with different mechanics. They do not look mergeable into one PR for the same underlying bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + } + ] +} diff --git a/snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/manifest.json b/snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..3376eb211a78de99f738e2f8ffe921407482b0cd --- /dev/null +++ b/snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/manifest.json @@ -0,0 +1,14 @@ +{ + "analysis_id": "hybrid-model-20260601t060039z", + "artifacts": { + "hybrid": "snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "model": null, + "published_at": "2026-06-01T06:09:00Z", + "repo": "huggingface/transformers", + "schema_version": 1, + "snapshot_id": "20260601T060039Z", + "variant": "hybrid" +} diff --git a/snapshots/20260601T060039Z/manifest.json b/snapshots/20260601T060039Z/manifest.json index 5db073be1f78ae00c99c98f572fb82ad0f806a91..e9a2b67600884f99329abec8f1dca54cf2b24f97 100644 --- a/snapshots/20260601T060039Z/manifest.json +++ b/snapshots/20260601T060039Z/manifest.json @@ -32,6 +32,24 @@ "timeline_events": 269 }, "extracted_at": "2026-06-01T06:00:39Z", + "published_analysis": { + "canonical_analysis_id": "hybrid-model-20260601t060039z", + "runs": { + "hybrid-model-20260601t060039z": { + "analysis_id": "hybrid-model-20260601t060039z", + "artifacts": { + "hybrid": "snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "manifest_path": "snapshots/20260601T060039Z/analysis-runs/hybrid-model-20260601t060039z/manifest.json", + "model": null, + "published_at": "2026-06-01T06:09:00Z", + "variant": "hybrid" + } + }, + "schema_version": 1 + }, "repo": "huggingface/transformers", "snapshot_id": "20260601T060039Z", "watermark": {