diff --git a/analysis/current/analysis-report-hybrid.json b/analysis/current/analysis-report-hybrid.json index 077f43bc950eec8e8d92d54cd29fb1c8c89dd3e7..0ae892a8f9c387844e35181400554bd04512d789 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": "20260627T060016Z", - "generated_at": "2026-06-27T06:09:28Z", + "snapshot_id": "20260627T120014Z", + "generated_at": "2026-06-27T12:10:31Z", "evidence_quality": "full", "llm_enrichment": true, "meta_bugs": [ @@ -1050,117 +1050,21 @@ ] }, { - "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-46895-4", - "summary": "Cluster of 1 issues and 3 PRs centered on issue #46897.", + "cluster_id": "cluster-46894-7", + "summary": "Cluster of 3 issues and 4 PRs centered on issue #46897.", "status": "open", "confidence": 0.8, "canonical_issue_number": 46897, - "canonical_pr_number": 46895, + "canonical_pr_number": 46903, "issue_numbers": [ - 46897 + 46894, + 46897, + 46901 ], "pr_numbers": [ 46895, 46898, + 46903, 46916 ], "evidence_types": [ @@ -1172,22 +1076,33 @@ { "left_pr_number": 46895, "right_pr_number": 46898, - "code_similarity": 0.121, - "size_similarity": 0.483, + "code_similarity": 0.117, + "size_similarity": 0.467, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.164, + "patch_similarity": 0.156, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46895, + "right_pr_number": 46903, + "code_similarity": 0.24, + "size_similarity": 0.844, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.476, "shared_filenames": [], "shared_file_areas": [] }, { "left_pr_number": 46895, "right_pr_number": 46916, - "code_similarity": 0.341, - "size_similarity": 0.644, + "code_similarity": 0.335, + "size_similarity": 0.622, "file_overlap": 0.4, "area_overlap": 0.194, - "patch_similarity": 0.162, + "patch_similarity": 0.154, "shared_filenames": [ "src/transformers/models/cohere_asr/modeling_cohere_asr.py", "src/transformers/models/cohere_asr/modular_cohere_asr.py" @@ -1233,6 +1148,17 @@ } ] }, + { + "left_pr_number": 46898, + "right_pr_number": 46903, + "code_similarity": 0.125, + "size_similarity": 0.553, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.099, + "shared_filenames": [], + "shared_file_areas": [] + }, { "left_pr_number": 46898, "right_pr_number": 46916, @@ -1293,6 +1219,116 @@ ] } ] + }, + { + "left_pr_number": 46903, + "right_pr_number": 46916, + "code_similarity": 0.17, + "size_similarity": 0.737, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.154, + "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 + ] + ] + } + ] } ] }, @@ -2047,31 +2083,305 @@ ] }, { - "cluster_id": "cluster-44002-2", + "cluster_id": "cluster-45415-2", "summary": "Cluster of 2 related pull requests linked by soft_similarity.", "status": "open", "confidence": 0.5, "canonical_issue_number": null, - "canonical_pr_number": 44002, + "canonical_pr_number": 45425, "issue_numbers": [], "pr_numbers": [ - 44002, - 44029 + 45415, + 45425 ], "evidence_types": [ "soft_similarity" ], "pr_comparisons": [ { - "left_pr_number": 44002, - "right_pr_number": 44029, - "code_similarity": 0.135, - "size_similarity": 0.43, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.329, - "shared_filenames": [], - "shared_file_areas": [] + "left_pr_number": 45415, + "right_pr_number": 45425, + "code_similarity": 0.092, + "size_similarity": 0.171, + "file_overlap": 0.065, + "area_overlap": 0.053, + "patch_similarity": 0.13, + "shared_filenames": [ + "src/transformers/modeling_flash_attention_utils.py", + "src/transformers/modeling_utils.py", + "utils/check_types.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/modeling_flash_attention_utils.py", + "left_ranges": [ + [ + 12, + 18 + ], + [ + 148, + 156 + ] + ], + "right_ranges": [ + [ + 31, + 37 + ], + [ + 148, + 154 + ] + ] + }, + { + "filename": "src/transformers/modeling_utils.py", + "left_ranges": [ + [ + 29, + 35 + ], + [ + 175, + 181 + ], + [ + 187, + 193 + ], + [ + 295, + 301 + ], + [ + 449, + 458 + ], + [ + 840, + 851 + ], + [ + 1032, + 1038 + ], + [ + 1062, + 1068 + ], + [ + 1566, + 1572 + ], + [ + 1822, + 1828 + ], + [ + 1833, + 1841 + ], + [ + 2393, + 2399 + ], + [ + 2990, + 2996 + ], + [ + 3136, + 3143 + ], + [ + 3262, + 3268 + ], + [ + 3291, + 3299 + ], + [ + 3320, + 3326 + ], + [ + 4010, + 4016 + ], + [ + 4177, + 4183 + ], + [ + 4190, + 4196 + ], + [ + 4505, + 4511 + ], + [ + 4623, + 4629 + ], + [ + 4747, + 4753 + ], + [ + 4765, + 4771 + ] + ], + "right_ranges": [ + [ + 29, + 35 + ], + [ + 121, + 127 + ], + [ + 139, + 147 + ], + [ + 181, + 187 + ], + [ + 191, + 214 + ], + [ + 343, + 360 + ], + [ + 374, + 383 + ], + [ + 428, + 438 + ], + [ + 903, + 923 + ], + [ + 962, + 968 + ], + [ + 1010, + 1016 + ], + [ + 1079, + 1095 + ], + [ + 1112, + 1129 + ], + [ + 1626, + 1633 + ], + [ + 1853, + 1862 + ], + [ + 1873, + 1892 + ], + [ + 1895, + 1903 + ], + [ + 1924, + 1930 + ], + [ + 2448, + 2469 + ], + [ + 3059, + 3065 + ], + [ + 3205, + 3212 + ], + [ + 3328, + 3338 + ], + [ + 4095, + 4106 + ], + [ + 4293, + 4301 + ], + [ + 4635, + 4641 + ], + [ + 4757, + 4763 + ], + [ + 4836, + 4849 + ], + [ + 4889, + 4895 + ], + [ + 4907, + 4913 + ] + ] + }, + { + "filename": "utils/check_types.py", + "left_ranges": [ + [ + 4, + 10 + ], + [ + 15, + 28 + ], + [ + 32, + 54 + ] + ], + "right_ranges": [ + [ + 16, + 22 + ], + [ + 26, + 32 + ] + ] + } + ] } ] }, @@ -2351,7 +2661,17 @@ ] } ], - "duplicate_issues": [], + "duplicate_issues": [ + { + "cluster_id": "cluster-46894-7", + "canonical_issue_number": 46897, + "duplicate_issue_numbers": [ + 46894, + 46901 + ], + "reason": "Issues in cluster-46894-7 are treated as duplicates because they share closing_reference, shared_issue_target, soft_similarity evidence." + } + ], "duplicate_prs": [ { "cluster_id": "cluster-41211-3", @@ -2397,15 +2717,6 @@ "target_issue_number": 43979, "reason": "PRs in cluster-43979-11 are treated as duplicates because they converge on issue #43979 with closing_reference, shared_issue_target, soft_similarity evidence." }, - { - "cluster_id": "cluster-44002-2", - "canonical_pr_number": 44002, - "duplicate_pr_numbers": [ - 44029 - ], - "target_issue_number": null, - "reason": "PRs in cluster-44002-2 are treated as duplicates because they share soft_similarity evidence." - }, { "cluster_id": "cluster-44018-2", "canonical_pr_number": 44068, @@ -2424,6 +2735,15 @@ "target_issue_number": 45081, "reason": "PRs in cluster-45081-3 are treated as duplicates because they converge on issue #45081 with closing_reference, shared_issue_target, soft_similarity evidence." }, + { + "cluster_id": "cluster-45415-2", + "canonical_pr_number": 45425, + "duplicate_pr_numbers": [ + 45415 + ], + "target_issue_number": null, + "reason": "PRs in cluster-45415-2 are treated as duplicates because they share soft_similarity evidence." + }, { "cluster_id": "cluster-45561-3", "canonical_pr_number": 45645, @@ -2470,21 +2790,22 @@ "reason": "PRs in cluster-46869-3 are treated as duplicates because they converge on issue #46869 with closing_reference, shared_issue_target evidence." }, { - "cluster_id": "cluster-46895-4", - "canonical_pr_number": 46895, + "cluster_id": "cluster-46894-7", + "canonical_pr_number": 46903, "duplicate_pr_numbers": [ + 46895, 46898, 46916 ], "target_issue_number": 46897, - "reason": "PRs in cluster-46895-4 are treated as duplicates because they converge on issue #46897 with closing_reference, shared_issue_target, soft_similarity evidence." + "reason": "PRs in cluster-46894-7 are treated as duplicates because they converge on issue #46897 with closing_reference, shared_issue_target, soft_similarity evidence." } ], "best_issue": { - "cluster_id": "cluster-46895-4", + "cluster_id": "cluster-46894-7", "issue_number": 46897, - "reason": "Issue #46897 is the strongest global issue candidate because it is open, belongs to a cluster with 4 artifacts, and carries 0 discussion comments plus 9 inbound references.", - "score": 149.0 + "reason": "Issue #46897 is the strongest global issue candidate because it is open, belongs to a cluster with 7 artifacts, and carries 0 discussion comments plus 9 inbound references.", + "score": 179.0 }, "best_pr": { "cluster_id": "cluster-43979-11", diff --git a/analysis/current/analysis-report-hybrid.llm-reviews.json b/analysis/current/analysis-report-hybrid.llm-reviews.json index ce9680039f87f860885d47014df122c63b564335..0f668dca6c0bf98d8430e74e77ec3a9dadbe66d5 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": "20260627T060016Z", - "generated_at": "2026-06-27T06:09:28Z", + "snapshot_id": "20260627T120014Z", + "generated_at": "2026-06-27T12:10:31Z", "model": "gpt-5.4-mini?service_tier=flex", "reviews": [ { @@ -132,20 +132,20 @@ "issue:45237|issue:46015", "issue:43039|issue:44805" ], - "prepared_review_unit_hash": "f0d510929ef519664e5f4f6f3903a8a98b16db6eda79a20cf63f2b02096456ee", + "prepared_review_unit_hash": "2247dcdb0d75e321b0f8af8999c1e7724d5da68b9d46e8102102482278cbdf56", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 63, - "serialized_chars": 46573, + "serialized_chars": 46576, "estimated_input_tokens": 11644, "estimated_eval_tokens": 23544 }, @@ -154,351 +154,327 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a mix of unrelated bug reports. Only two soft pairs look like true duplicates: the ZeRO-3 BertModel index-out-of-bounds reports, and the torch 2.4.1 compatibility reports.", - "confidence": 0.93, - "canonical_issue_reason": "Issue 43596 is the clearest canonical for the Bert/ZeRO-3 duplicate pair: it describes the concrete index-out-of-bounds failure on a standard BertModel, and 43638 is just the non-pretrained variant of the same bug.", + "summary": "The cluster is mostly unrelated bug reports. Only two soft-edge pairs look like true duplicates: the DeepSpeed ZeRO-3 Bert index-out-of-bounds reports and the torch 2.4.1 compatibility reports. The rest share model names or subsystems but not the same concrete failure.", + "confidence": 0.83, + "canonical_issue_reason": "issue:43596 is the best canonical issue: it states the ZeRO-3/Bert index-out-of-bounds failure clearly, and issue:43638 reads as the same bug in a non-pretrained Bert variant.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43596 is the best single issue in the set to anchor triage because it has the strongest true duplicate match and a concrete, reusable failure signature.", + "best_issue_reason": "issue:43596 is the strongest representative overall because it is concrete, well-scoped, and anchors the only strong duplicate pair in the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44322", "right": "issue:45464", "accept": false, - "reason": "Different bugs: Qwen3.5 config attribute missing vs streaming chat/completions failure." + "reason": "Both are Qwen3.5-related, but the packet shows different failures: missing config attribute vs streaming chat/completions error." }, { "left": "issue:45237", "right": "issue:45685", "accept": false, - "reason": "Unrelated failures: AMD GPU support for GPT-OSS vs an MPS histogram op error." + "reason": "Different hardware/runtime failures: AMD GPU inference vs MPS histogram int support." }, { "left": "issue:43716", "right": "issue:45237", "accept": false, - "reason": "Different model/runtime issues: Mistral-3 image dtype mismatch vs GPT-OSS on AMD GPUs." + "reason": "Different bugs and paths: multimodal dtype mismatch vs AMD GPU execution failure." }, { "left": "issue:43828", "right": "issue:45237", "accept": false, - "reason": "Different bugs: autocast dtype mismatch in Phi-tiny-MoE vs GPT-OSS AMD GPU failure." + "reason": "Autocast dtype mismatch is not the same bug as the AMD GPU failure." }, { "left": "issue:45237", "right": "issue:46468", "accept": false, - "reason": "Different models and symptoms; no shared concrete code path is evident." + "reason": "Different concrete failures: AMD GPU issue vs broken official example." }, { "left": "issue:45237", "right": "issue:45507", "accept": false, - "reason": "Different issues: GPT-OSS AMD GPU support vs GraniteMoEHybrid invalid method call." + "reason": "No shared code-path problem; one is AMD GPU inference, the other an invalid-method error in a different model." }, { "left": "issue:30277", "right": "issue:35545", "accept": false, - "reason": "Different failures: DeepSpeed ZeRO-3 collective mismatch vs ONNX export error." + "reason": "Both are runtime issues, but DeepSpeed collective mismatch and ONNX export failure are different bugs." }, { "left": "issue:35545", "right": "issue:42915", "accept": false, - "reason": "Different problems: ONNX export of ModernBERT vs Qwen3Moe FP8 failure." + "reason": "Different models and different failure modes; export error vs FineGrainedFP8 issue." }, { "left": "issue:35545", "right": "issue:41720", "accept": false, - "reason": "Different code paths: export/tracing bug vs Qwen3 auto device-map cuda assert." + "reason": "ONNX export error and Qwen3 auto-device-map cuda assert are unrelated." }, { "left": "issue:45901", "right": "issue:46015", "accept": false, - "reason": "Different topics: table QA crash vs outdated RAG docs." + "reason": "Both are documentation/example problems, but they are about different docs and different breaks." }, { "left": "issue:44509", "right": "issue:44512", "accept": false, - "reason": "Both are v5 docs cleanup, but they remove different obsolete references in different documentation areas." + "reason": "Both are v5 doc cleanups, but they reference different removed features." }, { "left": "issue:35545", "right": "issue:36010", "accept": false, - "reason": "Different bugs: ONNX export error vs GenerationMixin import regression." + "reason": "Different regressions: ONNX export failure vs GenerationMixin import error." }, { "left": "issue:44918", "right": "issue:46165", "accept": false, - "reason": "Different training failures: Qwen3.5 embedding unpacking vs resume-training with trust_remote_code." + "reason": "Different bugs: Qwen3.5 embedding unpacking vs resume-training with remote code." }, { "left": "issue:46165", "right": "issue:46869", "accept": false, - "reason": "Unrelated model/runtime problems; no common underlying bug is clear." + "reason": "Different setups and symptoms: resume/trust-remote-code vs multi-GPU device-map failure." }, { "left": "issue:44805", "right": "issue:45823", "accept": false, - "reason": "Different symptoms and subsystems: mask shape/indexing mismatch vs Gemma4 device-map PLE mismatch." + "reason": "Mask/index shape mismatch is not the same as Gemma4 device mismatch in forward." }, { "left": "issue:43638", "right": "issue:44805", "accept": false, - "reason": "Both mention tensor/index errors, but the failures are different and not the same code path." + "reason": "Different failure mechanisms: ZeRO-3 Bert loading error vs mask/tensor shape mismatch." }, { "left": "issue:43638", "right": "issue:46165", "accept": false, - "reason": "Different issues: Bert ZeRO-3 index error vs resume-training with local model and remote code." + "reason": "No overlap in code path or failure mode." }, { "left": "issue:43596", "right": "issue:44805", "accept": false, - "reason": "Different failures: Bert ZeRO-3 loading bug vs mask/tensor indexing mismatch." + "reason": "Both are index-related, but the concrete bugs are different." }, { "left": "issue:44805", "right": "issue:46294", "accept": false, - "reason": "Different modalities and errors: mask shape mismatch vs video token/feature mismatch." + "reason": "Mask/index shape mismatch is unrelated to Qwen3.5 video token/template mismatch." }, { "left": "issue:44805", "right": "issue:45161", "accept": false, - "reason": "Different problems: tensor mask indexing error vs GPT-OSS MoE tensor-parallel issue." + "reason": "Mask/index shape mismatch is unrelated to GPT-OSS tensor-parallel behavior." }, { "left": "issue:46165", "right": "issue:46468", "accept": false, - "reason": "Different bugs with different models and call paths." + "reason": "Different issues: remote-code resume training vs broken official example." }, { "left": "issue:44560", "right": "issue:46165", "accept": false, - "reason": "Different issues: Qwen3-vl video StopIteration vs training resume with remote code." + "reason": "Video chat-template StopIteration is unrelated to resume-training remote-code behavior." }, { "left": "issue:34634", "right": "issue:35545", "accept": false, - "reason": "Different components: BarkProcessor voice_preset vs ModernBERT ONNX export." + "reason": "BarkProcessor voice_preset and ModernBERT ONNX export are unrelated." }, { "left": "issue:30277", "right": "issue:43638", "accept": false, - "reason": "Both involve DeepSpeed, but one is a collective mismatch and the other is a Bert ZeRO-3 index error." + "reason": "Both involve DeepSpeed, but the concrete failures differ: collective mismatch vs Bert ZeRO-3 index error." }, { "left": "issue:43596", "right": "issue:43638", "accept": true, - "reason": "Same underlying bug: both are DeepSpeed ZeRO-3 BertModel loads failing with index 0 out of bounds; only the pretrained status differs." + "reason": "Same DeepSpeed ZeRO-3/Bert index-out-of-bounds bug; 43638 is a more specific non-pretrained Bert variant." }, { "left": "issue:43039", "right": "issue:45161", "accept": false, - "reason": "Different failures: Liger Kernel cross-entropy still being called vs GPT-OSS MoE tensor-parallel behavior." + "reason": "Liger kernel cross-entropy invocation and GPT-OSS TP failure are different problems." }, { "left": "issue:43975", "right": "issue:44457", "accept": false, - "reason": "Different bugs: DeepSeek-Coder detokenization vs LoRA merge/save/reload output mismatch." + "reason": "Incorrect detokenization and LoRA merge/save/load inconsistency are different bugs." }, { "left": "issue:44292", "right": "issue:44457", "accept": false, - "reason": "Different problems: Qwen-3-8B-NVFP4 runtime error vs LoRA merge/reload inconsistency." + "reason": "Different model/runtime failures with no shared concrete code path." }, { "left": "issue:44841", "right": "issue:46661", "accept": false, - "reason": "Different issues: processor failure for Voxtral vs remote-code config_class AttributeError." + "reason": "Processor failure and remote-code config_class attribute error are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:46869", + "accept": false, + "reason": "Different multimodal/device-map failures; not the same underlying bug." + }, + { + "left": "issue:24643", + "right": "issue:30277", + "accept": false, + "reason": "Both mention DeepSpeed, but one is a weight-dimension training error and the other is a collective mismatch." + }, + { + "left": "issue:34928", + "right": "issue:44805", + "accept": false, + "reason": "Activation checkpointing/FSDP mismatch and mask-shape indexing are different code paths." }, { "left": "issue:45859", "right": "issue:45902", "accept": false, - "reason": "Same class name, but different bugs: CPU-offload forward compatibility vs uninitialized inv_freq after meta materialization." + "reason": "Same class, but different concrete bugs: CPU offload incompatibility vs uninitialized memory after meta materialization." }, { "left": "issue:45685", "right": "issue:46869", "accept": false, - "reason": "Different hardware/runtime problems: MPS Int histogram op vs YOLO multi-GPU device-map auto failure." + "reason": "MPS histogram int support and YOLO multi-GPU auto device-map are unrelated." }, { "left": "issue:43638", "right": "issue:45161", "accept": false, - "reason": "Different bugs: Bert ZeRO-3 index error vs GPT-OSS MoE tensor-parallel issue." + "reason": "Bert ZeRO-3 loading failure is not the same as GPT-OSS tensor-parallel behavior." }, { "left": "issue:34928", "right": "issue:36331", "accept": false, - "reason": "Different training failures: activation checkpointing/FSDP tensor recomputation vs CustomTrainer compute_loss signature mismatch." + "reason": "FSDP/checkpointing mismatch and CustomTrainer compute_loss signature breakage are different failures." }, { "left": "issue:43596", "right": "issue:45161", "accept": false, - "reason": "Different bugs and models; no shared concrete failure mode." + "reason": "DeepSpeed/Bert loading error and GPT-OSS tensor-parallel issue are unrelated." }, { "left": "issue:43716", "right": "issue:45507", "accept": false, - "reason": "Different model-specific issues: Mistral-3 dtype mismatch vs GraniteMoEHybrid invalid method." + "reason": "Device/dtype mismatch in Mistral-3 is not the same as GraniteMoEHybrid invalid method calls." }, { "left": "issue:44292", "right": "issue:45237", "accept": false, - "reason": "Different model/runtime failures; not the same underlying bug." + "reason": "Different model/runtime failures; no shared underlying bug." }, { "left": "issue:45874", "right": "issue:46294", "accept": false, - "reason": "Different features: reverse embedding lookup on inputs_embeds vs video chat-template feature/token mismatch." + "reason": "Different Gemma4/Qwen3.5 bugs: reverse embedding lookup vs video token/template mismatch." }, { "left": "issue:39290", "right": "issue:45081", "accept": false, - "reason": "Different breakages: Gemma3 vLLM config attribute error vs mistral regex patch tokenizer backend error." + "reason": "Different models and failures: Gemma3 config attribute vs Mistral regex patch crash." }, { "left": "issue:45800", "right": "issue:46515", "accept": true, - "reason": "Same compatibility issue: both report torch 2.4.1 incompatibility, with no meaningful distinction in the failure itself." + "reason": "Both report the same torch 2.4.1 compatibility regression; 46515 is a shorter duplicate report." }, { "left": "issue:45823", "right": "issue:45874", "accept": false, - "reason": "Different Gemma4 issues: PLE device mismatch vs expensive reverse embedding lookup." + "reason": "Different Gemma4 issues: forward device mismatch vs expensive reverse embedding lookup." }, { "left": "issue:43425", "right": "issue:44292", "accept": false, - "reason": "Different problems: torch 2.10 compatibility vs Qwen-3-8B-NVFP4 runtime error." + "reason": "Torch version incompatibility is broader than the Qwen NVFP4 runtime error; not the same bug." }, { - "left": "issue:34928", - "right": "issue:41720", - "accept": false, - "reason": "Different issues: FSDP/activation-checkpointing size mismatch vs Qwen3 auto device-map cuda assert." - }, - { - "left": "issue:43828", - "right": "issue:45685", + "left": "issue:43827", + "right": "issue:44512", "accept": false, - "reason": "Different bugs: autocast dtype mismatch vs MPS histogram op not implemented for Int." + "reason": "Both are docs after v5 removals, but they cover different removed features." }, { "left": "issue:43828", - "right": "issue:44292", - "accept": false, - "reason": "Different failures and models; no evidence they share a root cause." - }, - { - "left": "issue:43716", - "right": "issue:46468", - "accept": false, - "reason": "Different model families and symptoms." - }, - { - "left": "issue:44292", - "right": "issue:44918", - "accept": false, - "reason": "Different code paths: Qwen NVFP4 runtime error vs TRL SFT trainer input embedding unpacking." - }, - { - "left": "issue:45507", - "right": "issue:45685", - "accept": false, - "reason": "Different bugs: GraniteMoEHybrid invalid method call vs MPS histogram op unsupported for Int." - }, - { - "left": "issue:43716", - "right": "issue:44292", - "accept": false, - "reason": "Different model/runtime failures; not the same bug." - }, - { - "left": "issue:44918", - "right": "issue:45161", - "accept": false, - "reason": "Different issues: TRL SFT embedding unpacking vs GPT-OSS MoE tensor-parallel behavior." - }, - { - "left": "issue:43716", "right": "issue:45685", "accept": false, - "reason": "Different subsystems and failure modes." + "reason": "Autocast dtype mismatch and MPS histogram int support are unrelated." }, { "left": "issue:43039", - "right": "issue:44918", + "right": "issue:43828", "accept": false, - "reason": "Different issues: Liger Kernel cross-entropy invocation vs Qwen3.5 TRL embedding unpacking." + "reason": "Liger kernel cross-entropy call and autocast dtype mismatch do not share a concrete code path." }, { - "left": "issue:41762", - "right": "issue:43638", + "left": "issue:30277", + "right": "issue:36331", "accept": false, - "reason": "Both are ZeRO-3 index errors, but they affect different models and loading paths; too broad to treat as the same bug." + "reason": "DeepSpeed collective mismatch and trainer compute_loss signature breakage are unrelated." }, { - "left": "issue:43827", - "right": "issue:44512", + "left": "issue:34928", + "right": "issue:36010", "accept": false, - "reason": "Different docs removals: pipeline task references vs transformers run command references." + "reason": "FSDP/checkpointing mismatch and GenerationMixin import failure are different issues." }, { - "left": "issue:43828", - "right": "issue:45685", + "left": "issue:45237", + "right": "issue:46015", "accept": false, - "reason": "Different bugs: autocast dtype mismatch vs MPS histogram op unsupported for Int." + "reason": "AMD GPU inference failure and outdated RAG examples are not the same underlying change." }, { "left": "issue:43039", "right": "issue:44805", "accept": false, - "reason": "Different failures: Liger Kernel cross_entropy call vs mask shape/indexing mismatch." + "reason": "Liger kernel cross-entropy call and mask-shape indexing error do not share a concrete code path." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The two accepted duplicate pairs are well supported by the issue titles and descriptions, and the rejected soft pairs are phrased cautiously without overclaiming shared root cause. The canonical Bert ZeRO-3 issue choice is also defensible." + "accept": false, + "feedback": "Mostly grounded, but at least one soft-edge reason overstates the evidence: issue:44322 vs issue:45464 are both Qwen3.5-related, so saying they are from \u201cdifferent model families\u201d is not supported by the packet. The duplicate conclusions on issue:43596/issue:43638 and issue:45800/issue:46515 look plausible, but the verdict set is not fully conservative." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -571,8 +547,8 @@ "issue:43296|issue:46304", "issue:41762|issue:43596", "issue:39290|issue:41720", - "issue:43901|issue:44512", "issue:44589|issue:46304", + "issue:43901|issue:44512", "issue:45237|issue:45901", "issue:43975|issue:46015", "issue:43039|issue:44560", @@ -609,178 +585,178 @@ "issue:43572|issue:45070", "issue:45800|issue:45823" ], - "prepared_review_unit_hash": "d55e3f913db7038362b35e55a5d54edd731dafe62cbcf19ba693a4368b81a49a", + "prepared_review_unit_hash": "1424eeb5af8adcc475c972493154a68449bf031bd53696fd06029bff5c3c4b67", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 42, - "serialized_chars": 40040, - "estimated_input_tokens": 10010, - "estimated_eval_tokens": 20276 + "serialized_chars": 40042, + "estimated_input_tokens": 10011, + "estimated_eval_tokens": 20278 }, "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, with one clear duplicate pair around the v5 AutoTokenizer token-ID regression.", - "confidence": 0.76, - "canonical_issue_reason": "issue:45920 is the broadest and most representative report of the shared tokenizer regression, and it subsumes the Granite-only repro in issue:45812.", + "summary": "Most pairs are distinct model-specific or subsystem-specific issues. Two pairs look like true duplicates: the v5 AutoTokenizer wrong-token-id regression across model families, and the ZeRO-3 load-time index error during model initialization.", + "confidence": 0.68, + "canonical_issue_reason": "issue:45920 is the best canonical issue because it describes the broad v5 AutoTokenizer regression affecting multiple model families, which subsumes the narrower Granite-only report.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 best represents the actual duplicate-worthy topic in the cluster: a v5 AutoTokenizer regression affecting multiple model families.", + "best_issue_reason": "issue:45920 is the strongest global issue candidate: it is clearly framed, high-impact, and broad enough to represent the shared tokenizer regression pattern.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43866", "right": "issue:46402", "accept": false, - "reason": "Different models and failures: Ovis2 checkpoint corruption vs Sam3Model text-encoder weights not loaded." + "reason": "Different models and different failures: corrupted checkpoint vs missing text-encoder weights." }, { "left": "issue:45800", "right": "issue:46304", "accept": false, - "reason": "Both mention compatibility, but one is a generic torch/transformers version mismatch while the other is a USE_DISTRIBUTED=0 breakage." + "reason": "Unrelated compatibility regressions: torch/version mismatch vs USE_DISTRIBUTED=0 breakage." }, { "left": "issue:43296", "right": "issue:46304", "accept": false, - "reason": "PaddleOCR-VL loading in vLLM and USE_DISTRIBUTED=0 are unrelated code paths and failure modes." + "reason": "Different loading path and symptom: vLLM PaddleOCR-VL load failure vs distributed build breakage." }, { "left": "issue:41762", "right": "issue:43596", - "accept": false, - "reason": "Both hit a zero3-related IndexError, but the repros are on different model-loading paths and there is not enough evidence they are the same bug." + "accept": true, + "reason": "Both report the same ZeRO-3 model-init/load IndexError ('index 0 out of bounds' / empty tensor) and could plausibly share one fix." }, { "left": "issue:39290", "right": "issue:41720", "accept": false, - "reason": "Gemma3TextConfig missing sliding_window_pattern and Qwen3 auto device-map cuda asserts are different regressions." + "reason": "Different bugs: Gemma3 config attribute missing with vLLM vs Qwen3 auto-device-map CUDA assert." }, { - "left": "issue:43901", - "right": "issue:44512", + "left": "issue:44589", + "right": "issue:46304", "accept": false, - "reason": "Separate documentation issues in different areas, not the same underlying bug or change." + "reason": "Different low-level runtime issues: missing Float8 storage vs USE_DISTRIBUTED=0 support." }, { - "left": "issue:44589", - "right": "issue:46304", + "left": "issue:43901", + "right": "issue:44512", "accept": false, - "reason": "Float8 storage lookup failure is unrelated to the distributed-support regression in transformers v5.9.0." + "reason": "Both are docs issues, but they refer to different outdated commands and different doc sections." }, { "left": "issue:45237", "right": "issue:45901", "accept": false, - "reason": "Different subsystems and symptoms: AMD GPU support for GPT-OSS vs table-question-answering crash." + "reason": "Different task/model failures: AMD GPU support vs table-question-answering crash." }, { "left": "issue:43975", "right": "issue:46015", "accept": false, - "reason": "A DeepSeek detokenization regression and outdated RAG docs are unrelated." + "reason": "Different problems: detokenization regression vs outdated RAG examples." }, { "left": "issue:43039", "right": "issue:44560", "accept": false, - "reason": "Liger Kernel cross-entropy dispatch and Qwen3.5-vl video StopIteration are different bugs." + "reason": "Different code paths: Liger kernel cross-entropy dispatch vs Qwen3-vl video StopIteration." }, { "left": "issue:45823", "right": "issue:46294", "accept": false, - "reason": "Different models and mismatches: Gemma4 PLE device placement vs Qwen3.5 video/chat-template token mismatch." + "reason": "Different failure modes: device-map PLE mismatch vs chat-template video token/feature mismatch." }, { "left": "issue:44841", "right": "issue:45823", "accept": false, - "reason": "Voxtral processor failure and Gemma4 device mismatch are separate model-specific problems." + "reason": "Different models and subsystems: processor loading vs device-map forward mismatch." }, { "left": "issue:43866", "right": "issue:46772", "accept": false, - "reason": "Ovis2 checkpoint corruption is unrelated to DiffusionGemma NVFP4 compatibility." + "reason": "Checkpoint corruption is not the same as NVFP4 compatibility for DiffusionGemma." }, { "left": "issue:44863", "right": "issue:46015", "accept": false, - "reason": "NemotronH checkpoint loading and RAG docs being outdated are unrelated." + "reason": "NemotronH checkpoint loading is unrelated to RAG example docs drift." }, { "left": "issue:45020", "right": "issue:46304", "accept": false, - "reason": "Remote-code model breakage and USE_DISTRIBUTED=0 support regression are not the same underlying issue." + "reason": "Remote-code model breakage is a different issue from the USE_DISTRIBUTED=0 torch support regression." }, { "left": "issue:45812", "right": "issue:45920", "accept": true, - "reason": "Both report the same v5 AutoTokenizer wrong-token-ID regression; 45920 is the broader multi-model report and 45812 is a narrower Granite repro." + "reason": "Same underlying v5 AutoTokenizer regression: wrong token IDs across multiple model families, with Granite as one affected case." }, { "left": "issue:43866", "right": "issue:44863", "accept": false, - "reason": "Different model-loading failures with no shared concrete code path." + "reason": "Different model-specific loading failures with no shared symptom or fix path." }, { "left": "issue:43872", "right": "issue:46304", "accept": false, - "reason": "bitsandbytes Int8Params API incompatibility is unrelated to the distributed-build regression." + "reason": "bitsandbytes parameter-construction incompatibility is unrelated to distributed build support." }, { "left": "issue:44292", "right": "issue:46015", "accept": false, - "reason": "Qwen-3-8B-NVFP4 runtime failure and outdated RAG examples are unrelated." + "reason": "NVFP4 model runtime error is unrelated to documentation examples." }, { "left": "issue:39290", "right": "issue:43531", "accept": false, - "reason": "Both mention sliding_window, but one is a Gemma3 config attribute issue and the other is a Qwen3-MoE model-specific sliding-window bug." + "reason": "Both mention sliding-window concepts, but they are different models and different concrete failures." }, { "left": "issue:41093", "right": "issue:43638", "accept": false, - "reason": "The mask-shape mismatch and the zero3 BertModel index error are different failure signatures and likely different bugs." + "reason": "Similar zero-sized index symptoms, but the mask-shape mismatch and ZeRO-3 Bert load error are not the same bug." }, { "left": "issue:43792", "right": "issue:45901", "accept": false, - "reason": "Whisper-large-v2 runtime failure and table-question-answering crash are unrelated." + "reason": "Different domains: Whisper loading/runtime vs table-question-answering pipeline crash." }, { "left": "issue:43854", "right": "issue:43866", "accept": false, - "reason": "GLM-4.7-Flash loading in tests and Ovis2 checkpoint corruption are distinct issues." + "reason": "Different model problems: unit-test load failure vs corrupted checkpoint." }, { "left": "issue:43425", "right": "issue:45070", "accept": false, - "reason": "Both are version-related, but they affect different compatibility surfaces and are not the same regression." + "reason": "Torch version incompatibility is not the same as a PretrainedConfig pydantic-field regression." }, { "left": "issue:43975", @@ -792,91 +768,109 @@ "left": "issue:43792", "right": "issue:45070", "accept": false, - "reason": "Whisper runtime failure is unrelated to the PretrainedConfig pydantic breakage." + "reason": "Whisper runtime failure is unrelated to the pydantic config regression." }, { "left": "issue:43596", "right": "issue:45800", "accept": false, - "reason": "Zero3 BertModel IndexError and torch/transformers incompatibility are not the same bug." + "reason": "Both mention load/runtime problems, but the ZeRO-3 index error is not a torch 2.4.1 compatibility issue." }, { "left": "issue:45362", "right": "issue:45464", "accept": false, - "reason": "Both touch Qwen3.5 chat flows, but one is a chat-template crash and the other is a streaming API failure; too different to merge as duplicates." + "reason": "Same model family, but different interfaces and failure modes: transformers chat crash vs streaming chat/completions failure." + }, + { + "left": "issue:43541", + "right": "issue:46869", + "accept": false, + "reason": "Different model families and execution paths: grouped_mm tracing vs multi-GPU device-map failure." + }, + { + "left": "issue:44292", + "right": "issue:45901", + "accept": false, + "reason": "NVFP4 runtime failure and table QA crash are unrelated." + }, + { + "left": "issue:39290", + "right": "issue:45823", + "accept": false, + "reason": "Different models and different device/config issues; only broad version-related similarity." }, { "left": "issue:45464", "right": "issue:45685", "accept": false, - "reason": "Chat streaming failure and an MPS histogram op issue are unrelated." + "reason": "Chat streaming failure is unrelated to an MPS histogram kernel missing for Int." }, { "left": "issue:36010", "right": "issue:39290", "accept": false, - "reason": "GenerationMixin import error and Gemma3 sliding_window attribute error are unrelated." + "reason": "Import error for GenerationMixin is unrelated to the Gemma3TextConfig sliding-window attribute issue." }, { "left": "issue:43541", "right": "issue:45685", "accept": false, - "reason": "Mixtral torch-dynamo grouped_mm tracing and the MPS histogram issue are separate bugs." + "reason": "Different execution failures: torch dynamo grouped_mm tracing vs MPS histogram kernel unsupported." }, { "left": "issue:43541", "right": "issue:43828", "accept": false, - "reason": "Both involve model execution, but one is a grouped_mm tracing failure and the other is an autocast dtype mismatch." + "reason": "Grouped_mm tracing error and autocast dtype mismatch are different bugs." }, { "left": "issue:44841", "right": "issue:45084", "accept": false, - "reason": "Voxtral processor failure and a compile error about non-template nodes are unrelated." + "reason": "Processor loading failure is unrelated to compile-time template-node errors." }, { "left": "issue:43404", "right": "issue:43866", "accept": false, - "reason": "Untied lm_head weights in Mistral3 and Ovis2 checkpoint corruption are different issues." + "reason": "Weight-tying bug in Mistral3 is unrelated to a corrupted Ovis2 checkpoint." }, { "left": "issue:45507", "right": "issue:46869", "accept": false, - "reason": "GraniteMoEHybrid invalid method and YOLO multi-GPU auto device-map failure are unrelated." + "reason": "Different model families and different invalid-method/device-map failures." }, { "left": "issue:41093", "right": "issue:41762", "accept": false, - "reason": "Different error shapes and symptoms: mask-size mismatch vs zero3 index-out-of-bounds during Gemma3 loading." + "reason": "Both are IndexErrors, but one is a mask-length mismatch and the other is a ZeRO-3 zero-size index error." }, { "left": "issue:45399", "right": "issue:46661", "accept": false, - "reason": "Flash-attn fallback gating and remote-code model_class/config_class handling are unrelated." + "reason": "Fallback logic for flash-attn2 is unrelated to remote-code model_class config_class errors." }, { "left": "issue:43572", "right": "issue:45070", "accept": false, - "reason": "StableLm pad_token_idx regression and a PretrainedConfig pydantic breakage are different compatibility bugs." + "reason": "A missing StableLmConfig field is not the same as a broader pydantic PretrainedConfig breakage." }, { "left": "issue:45800", "right": "issue:45823", "accept": false, - "reason": "Torch/transformers version incompatibility and Gemma4 device-mapping mismatch are not the same concrete bug." + "reason": "Different regressions: torch/distributed compatibility vs Gemma4 device mismatch." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet: issue:45920 and issue:45812 are the only clearly duplicate-worthy pair, and the rest of the cluster appears heterogeneous. The soft-edge verdicts are conservative overall and do not overclaim evidence. No PR-pair mergeability concerns apply here." + "feedback": "The summary is grounded in the packet: issue:45920 is clearly the broad v5 AutoTokenizer regression, and the duplicate judgments are conservative. The one accepted ZeRO-3 pair is reasonably plausible from the shared IndexError/load-time symptom, and the rest of the soft-edge rejections stay on the safe side without overstating equivalence." }, "evaluator_used": true, "retried": false, @@ -974,234 +968,168 @@ "issue:43873|issue:45005", "issue:44530|issue:45005", "issue:44360|issue:44512", - "issue:44492|issue:44512", "issue:44863|issue:45005", + "issue:44492|issue:44512", "issue:45183|issue:46463", "issue:46489|issue:46491" ], - "prepared_review_unit_hash": "6a538354b3f80a6e75e4adc1734b0412bce501176282a4e361eb87891925c4fb", + "prepared_review_unit_hash": "d636e2d5160dac2b6fe0c5e04345b931fd1d560c565665241c3785b4be4ac469", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36887, - "estimated_input_tokens": 9222, - "estimated_eval_tokens": 18700 + "serialized_chars": 36889, + "estimated_input_tokens": 9223, + "estimated_eval_tokens": 18702 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous; most soft edges are false positives. The only plausible duplicate groups are around trust_remote_code custom-module loading, Qwen3.5 video-input token accounting, and input validation/error-message wording.", - "confidence": 0.67, - "canonical_issue_reason": "No single issue cleanly represents the whole cluster, but issue 45698 is the strongest anchor among the near-duplicates because it captures the broad custom-module misloading symptom that also covers the leaf-directory collision case.", + "summary": "Mostly unrelated issue reports. The strongest cluster anchor is the cross-model input-validation bug, and I\u2019m only accepting soft edges where the titles clearly point to the same underlying defect; the remote-code cache/module pair is too speculative and should be rejected.", + "confidence": 0.81, + "canonical_issue_reason": "Issue 45183 is the broad umbrella report for the shared input-validation bug across multiple models, which also covers the narrower T5 decoder-input case in 46463.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45698 is the most actionable and generally worded issue among the candidates, and it plausibly subsumes the closely related 45632 report.", + "best_issue_reason": "Issue 45183 is the clearest root-cause-oriented report in the set and best represents the shared validation logic defect.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43296", "right": "issue:43366", "accept": false, - "reason": "Different model/setup problems: PaddleOCR-VL load failure vs GGUF gpt-oss architecture support." + "reason": "Different products and failure modes: PaddleOCR-VL loading in vLLM vs GGUF GPT-OSS architecture support." }, { "left": "issue:44387", "right": "issue:45005", "accept": false, - "reason": "One is int4 quantization memory/OOM behavior; the other is tied-weights handling for translation models." + "reason": "Int4 quantization OOM and tied-weights behavior are unrelated bugs." }, { "left": "issue:44805", "right": "issue:44918", "accept": false, - "reason": "Both are runtime errors, but the shapes and code paths differ: mask indexing vs TRL input embedding unpacking." + "reason": "Both are shape-related errors, but they involve different model paths and different causes." }, { "left": "issue:43541", "right": "issue:45161", "accept": false, - "reason": "Both mention MoE/Mixtral/GPT-OSS, but one is torch dynamo grouped_mm tracing and the other is tensor-parallel support." + "reason": "Both mention MoE models, but grouped_mm tracing and tensor-parallel support are different code-path problems." }, { "left": "issue:43828", "right": "issue:45464", "accept": false, - "reason": "Different multimodal/dtype failures with different models and failure modes." + "reason": "Autocast dtype mismatch and streaming chat/completions failure are not the same bug." }, { "left": "issue:43366", "right": "issue:45464", "accept": false, - "reason": "Unrelated: GGUF gpt-oss architecture support vs Qwen3.5 streaming inference failure." + "reason": "Unrelated issues: GPT-OSS GGUF architecture support vs Qwen3.5 streaming inference." }, { "left": "issue:44291", "right": "issue:45070", "accept": false, - "reason": "Both are v5 regressions, but one is init_empty_weights API breakage and the other is pydantic config-field behavior." + "reason": "init_empty_weights argument handling and pydantic PretrainedConfig breakage are different compatibility issues." }, { "left": "issue:43531", "right": "issue:45070", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior is unrelated to PretrainedConfig/pydantic issues." + "reason": "Qwen3-MoE sliding-window behavior and a pydantic config regression are unrelated." }, { "left": "issue:41093", "right": "issue:43596", "accept": false, - "reason": "Similar error type only; the mask-size mismatch and zero3/BertModel index error are different bugs." + "reason": "Both are indexing errors, but they arise in different initialization/training flows and do not look like the same defect." }, { "left": "issue:45020", "right": "issue:45800", "accept": false, - "reason": "Remote-code breakage and torch/transformers version incompatibility are not the same concrete issue." + "reason": "Remote-code model breakage and a torch/transformers version incompatibility are separate compatibility reports." }, { "left": "issue:43716", "right": "issue:46294", "accept": false, - "reason": "Both involve multimodal inputs, but one is image-preprocessor dtype mismatch and the other is video-template token accounting." + "reason": "Both involve multimodal inputs, but image preprocessor dtype mismatch and Qwen3.5 video token accounting are different bugs." }, { "left": "issue:43054", "right": "issue:43994", "accept": false, - "reason": "Both are about Siglip2 quality, but one is text-embedding degradation and the other is incorrect overall outputs/loading behavior." + "reason": "Same model family, but one is a relative text-embedding quality complaint and the other is a broader bad-output/loading symptom; not clearly the same underlying bug." }, { "left": "issue:44479", "right": "issue:46294", "accept": true, - "reason": "Both describe the same Qwen3.5 video-input regression where chat-template/video-token accounting mismatches the provided video features." + "reason": "Both describe the Qwen video-input regression where video features and video tokens are miscounted; 46294 looks like a specific manifestation of the broader regression in 44479." }, { "left": "issue:46468", "right": "issue:46869", "accept": false, - "reason": "Different domains: Nemotron example breakage vs YOLO multi-GPU device-map auto failure." + "reason": "Nemotron example breakage and YOLO multi-GPU device-map auto failure are unrelated." }, { "left": "issue:45759", "right": "issue:45902", "accept": false, - "reason": "Composite-config unwrapping and uninitialized rotary-embedding memory are separate bugs." + "reason": "Composite config unwrapping and uninitialized rotary embedding memory are different issues." }, { "left": "issue:43596", "right": "issue:46165", "accept": false, - "reason": "Zero3/BertModel index error and trust_remote_code resume issues are unrelated." + "reason": "deepspeed zero3 init failure and resume-training with trust_remote_code are not the same code-path problem." }, { "left": "issue:46640", "right": "issue:46661", "accept": false, - "reason": "Flash Attention 2 usage and remote-code config_class attribute errors are different code paths." - }, - { - "left": "issue:43716", - "right": "issue:44918", - "accept": false, - "reason": "Both are multimodal-related, but image dtype mismatch is not the same as the TRL embedding-unpacking failure." - }, - { - "left": "issue:39186", - "right": "issue:44918", - "accept": false, - "reason": "FSDP 2-D weight enforcement is unrelated to the Qwen3.5/TRl tensor-shape issue." - }, - { - "left": "issue:40990", - "right": "issue:44960", - "accept": false, - "reason": "High perplexity on gpt-oss and GLM5 are unrelated model-quality reports." - }, - { - "left": "issue:43645", - "right": "issue:45399", - "accept": false, - "reason": "Custom-model notebook initialization and flash-attn2 fallback logic are unrelated." - }, - { - "left": "issue:45632", - "right": "issue:45698", - "accept": true, - "reason": "Both point to the same trust_remote_code custom-module loading/caching bug; 45632 is the specific leaf-directory collision variant of 45698's wrong-module symptom." - }, - { - "left": "issue:43404", - "right": "issue:45005", - "accept": false, - "reason": "Tied lm_head weights in Mistral3 and translation-model tied-weight issues are different model families and failures." - }, - { - "left": "issue:42915", - "right": "issue:45399", - "accept": false, - "reason": "FineGrainedFP8Config failure and flash-attn2 fallback gating are unrelated." - }, - { - "left": "issue:43873", - "right": "issue:45005", - "accept": false, - "reason": "Quantization/offloading behavior is not the same bug as tied-weight handling." - }, - { - "left": "issue:44530", - "right": "issue:45005", - "accept": false, - "reason": "PagedAttentionCache linear_attention crash and tied weights are unrelated." - }, - { - "left": "issue:44360", - "right": "issue:44512", - "accept": false, - "reason": "A DSA indexer note and a docs typo are not the same underlying change." - }, - { - "left": "issue:44492", - "right": "issue:44512", - "accept": false, - "reason": "Cache-strategy typo and removed run-command docs are unrelated documentation issues." - }, - { - "left": "issue:44863", - "right": "issue:45005", - "accept": false, - "reason": "NemotronH checkpoint loading and translation-model tied weights are different bugs." + "reason": "Flash Attention 2 issue and remote-code model_class/config_class attribute error are different failures." }, { "left": "issue:45183", "right": "issue:46463", "accept": true, - "reason": "Both are the same validation-family bug: incorrect XOR/input validation between input_ids and inputs_embeds leading to misleading error messages, with 46463 as a T5-specific manifestation." + "reason": "Both are about the same input-validation logic; 46463 is the T5-specific instance of the broader validation-message bug in 45183." }, { "left": "issue:46489", "right": "issue:46491", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different models and different round-trip behaviors." + "reason": "Both are tokenizer regressions, but they affect different models and different behaviors." + }, + { + "left": "issue:45632", + "right": "issue:45698", + "accept": false, + "reason": "They are plausibly related to remote-code/custom-module loading, but the titles do not clearly show the same concrete bug or a single mergeable fix." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is conservative and stays within what the packet supports. The accepted soft edges are plausibly related by the titles, and the rejected pairs are not over-merged. The only potentially close call is 45632/45698, but the reasoning is still cautious enough to pass." + "accept": false, + "feedback": "Overall grounded, but one soft-edge acceptance is too speculative for the packet: issue:45632 vs issue:45698 is plausibly related to remote-code/custom-module loading, yet the titles do not clearly show the same bugfix or a single mergeable PR, so that verdict is not conservative enough. The other accepted pairs look more defensible." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -1293,22 +1221,22 @@ "issue:44403|issue:44488", "issue:24643|issue:28530", "issue:42947|issue:43856", - "issue:43643|issue:45020", "issue:41762|issue:42915", + "issue:43643|issue:45020", "issue:44991|issue:45399", "issue:43824|issue:43872", "issue:43782|issue:44960", "issue:43881|issue:43883", "issue:45685|issue:45823" ], - "prepared_review_unit_hash": "9e1943f5cfb7c064917e63f085e3a01ff3815c954071dc7854a976de8481c9a4", + "prepared_review_unit_hash": "4b356b265339cf228f0d8664f5a145396faa49c49540e3fb3cae909e39728d2b", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, @@ -1323,204 +1251,204 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster mostly contains unrelated issue reports, but a few soft pairs do look like the same underlying bug: non-persistent buffer corruption in v5, StaticCache.get_seq_length returning a tensor instead of int, tokenizer regression in v5+, and the SigLIP2 implementation discrepancy. The rest are different models or code paths.", - "confidence": 0.78, - "canonical_issue_reason": "issue:43950 is the clearest root-cause regression report: it names the transformers 5.x non-persistent-buffer corruption directly and cleanly subsumes narrower manifestations like issue:44534 and likely issue:45902.", + "summary": "Most items are unrelated, but there is a clear duplicate cluster around the Transformers v5 non-persistent-buffer regression; `issue:43950` is the best representative of that bug family.", + "confidence": 0.77, + "canonical_issue_reason": "`issue:43950` most clearly states the core regression (`from_pretrained()` corrupting non-persistent buffers) and aligns with the closest duplicate reports (`issue:43644`, `issue:44534`).", "canonical_pr_reason": null, - "best_issue_reason": "issue:43950 is the best canonical anchor because it is broad, technically specific, and framed as the regression itself rather than a single model-specific symptom.", + "best_issue_reason": "It is the most general and explicit report of the shared buffer-corruption bug, with the clearest path to a single fix.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43975", "right": "issue:44292", "accept": false, - "reason": "Different failure modes and models: DeepSeek-Coder detokenization vs Qwen NVFP4 loading/runtime error." + "reason": "Different models and failure modes; both are tokenizer/loading bugs only at a broad level." }, { "left": "issue:44918", "right": "issue:45507", "accept": false, - "reason": "Different models and code paths; one is Qwen3.5 input-embedding unpacking with TRL, the other GraniteMoEHybrid method invocation." + "reason": "Different concrete problems: TRL SFT embedding unpacking vs GraniteMoEHybrid method call." }, { "left": "issue:43950", "right": "issue:44534", "accept": true, - "reason": "Same underlying bug: transformers v5 non-persistent buffer handling/corruption, with 44534 phrased as the same regression." + "reason": "Same underlying regression: non-persistent buffers being filled/corrupted in Transformers v5." }, { "left": "issue:39692", "right": "issue:43054", "accept": false, - "reason": "One is a docs/example mismatch and quantization failure; the other is a SigLIP2 quality/performance complaint." + "reason": "Docs/example errors vs model-quality discrepancy; not the same bug." }, { "left": "issue:43644", "right": "issue:43950", - "accept": false, - "reason": "Unrelated regressions: missing config fields from trust_remote_code vs non-persistent buffer corruption." + "accept": true, + "reason": "Both report the same non-persistent-buffer junk/corruption regression." }, { "left": "issue:36331", "right": "issue:41762", "accept": false, - "reason": "Different subsystems: Trainer loss signature change vs DeepSpeed ZeRO-3 model loading IndexError." + "reason": "Unrelated trainer API mismatch vs ZeRO-3 model-loading IndexError." }, { "left": "issue:43872", "right": "issue:44589", "accept": false, - "reason": "Both are quantization/backend issues, but they describe different concrete incompatibilities and code paths." + "reason": "Different quantization/storage incompatibilities; no shared code-path evidence." }, { "left": "issue:43828", "right": "issue:44928", "accept": false, - "reason": "Different models and symptoms; dtype mismatch in Phi-tiny-MoE is not the same bug as Qwen3.5 RLHF gradient explosion." + "reason": "ImportError for a class name vs Qwen3.5 attention/position-id crash." }, { "left": "issue:45987", "right": "issue:46858", "accept": true, - "reason": "Same bug family: get_seq_length returning a tensor/device tensor instead of an int, causing downstream static-cache arithmetic failures." + "reason": "Same `cache.get_seq_length()` contract bug; the later report exposes the backend-specific failure." }, { "left": "issue:45920", "right": "issue:46489", "accept": true, - "reason": "Same tokenizer regression in transformers v5+: both report incorrect token IDs/output for affected model families." + "reason": "Same tokenizer regression in v5; `46489` looks like a model-specific gap in the broader fix." }, { "left": "issue:43054", "right": "issue:45701", "accept": false, - "reason": "SigLIP2 embedding quality complaint vs a broad tokenizer-change report; not the same concrete issue." + "reason": "SigLIP2 embedding quality complaint vs unrelated tokenization change report." }, { "left": "issue:38175", "right": "issue:43054", "accept": false, - "reason": "Both mention SigLIP2, but one is zero probabilities and the other is relative embedding degradation; no clear shared root cause." + "reason": "Both involve SigLIP2, but the symptoms are different and not clearly the same bug." }, { "left": "issue:43950", "right": "issue:45902", - "accept": true, - "reason": "Same non-persistent-buffer regression, with 45902 describing a specific manifestation after meta\u2192to_empty(cuda) materialization." + "accept": false, + "reason": "`45902` is a model-specific `to_empty`/offload issue, not clearly the same bug as the general buffer-corruption regression." }, { "left": "issue:41553", "right": "issue:43883", "accept": false, - "reason": "Bad AutoTokenizer error message for Voxtral is unrelated to Molmo's missing tied-weights attribute." + "reason": "Different models and different loading errors." }, { "left": "issue:43054", "right": "issue:43493", "accept": true, - "reason": "Likely the same SigLIP2 implementation discrepancy: HF vs JAX mismatch plausibly explains the degraded text embeddings." + "reason": "Both point to the same SigLIP2 implementation discrepancy behind degraded text embeddings." }, { "left": "issue:44910", "right": "issue:45859", "accept": false, - "reason": "Both involve Qwen3.5/Qwen3_5Moe, but one is FlashAttention position_id handling and the other is CPU-offload incompatibility in rotary embedding." + "reason": "Flash-Attention packed-sequence crash vs CPU-offload rotary embedding incompatibility." }, { "left": "issue:42947", "right": "issue:44387", "accept": false, - "reason": "Different performance problems: LoRA gradient checkpointing vs int4 quantization memory bloat/OOM." + "reason": "Different memory issues with different mechanisms and settings." }, { "left": "issue:45663", "right": "issue:45902", "accept": false, - "reason": "Different model paths and failures: Gemma4 FSDP2 attention KeyError vs non-persistent buffer materialization." + "reason": "Different models and different defects; no shared concrete bug." }, { "left": "issue:41553", "right": "issue:43881", "accept": false, - "reason": "Different model-loading failures; no shared underlying bug is evident." + "reason": "Different model-loading failures." }, { "left": "issue:44403", "right": "issue:44451", "accept": false, - "reason": "Generic load-time noise vs a specific model loading failure for ScandiBERT." + "reason": "Unrelated: log noise vs a specific model load failure." }, { "left": "issue:43975", "right": "issue:45520", "accept": false, - "reason": "Tokenizer detokenization bug vs Python 3.13 flash_attn import mapping issue." + "reason": "Tokenizer detokenization bug vs flash-attn import issue." }, { "left": "issue:44403", "right": "issue:44488", "accept": false, - "reason": "Different artifacts and failures; both are loading-related but not the same bug." + "reason": "General loading noise vs a specific tokenizer/model load failure." }, { "left": "issue:24643", "right": "issue:28530", "accept": false, - "reason": "Training with DeepSpeed weight-shape error vs early-stopping metric selection; unrelated." + "reason": "Different training-time errors and unrelated root causes." }, { "left": "issue:42947", "right": "issue:43856", "accept": false, - "reason": "Different memory-related training issues with different causes and model setups." + "reason": "Both mention memory, but the bugs are different (LoRA checkpointing vs MoE training footprint)." }, { - "left": "issue:43643", - "right": "issue:45020", + "left": "issue:41762", + "right": "issue:42915", "accept": false, - "reason": "Related theme of remote_code, but one is missing config fields and the other is a broader compatibility regression; not the same concrete bug." + "reason": "Different model families and different loading/configuration failures." }, { - "left": "issue:41762", - "right": "issue:42915", + "left": "issue:43643", + "right": "issue:45020", "accept": false, - "reason": "Gemma3 ZeRO-3 loading IndexError vs Qwen3MoE FineGrainedFP8 failure; different code paths." + "reason": "Both involve remote code, but one is missing fields in AutoConfig while the other is a broader regression report." }, { "left": "issue:44991", "right": "issue:45399", "accept": false, - "reason": "Tokenizer loading failure and flash-attn fallback blocking are unrelated." + "reason": "Tokenizer load failure vs flash-attn fallback selection problem." }, { "left": "issue:43824", "right": "issue:43872", "accept": false, - "reason": "Importing a model class vs bitsandbytes Int8Params constructor incompatibility." + "reason": "Import error for a missing class vs bitsandbytes parameter incompatibility." }, { "left": "issue:43782", "right": "issue:44960", "accept": false, - "reason": "Qwen3VL weight_only load error vs a vague GLM5 issue; no shared root cause." + "reason": "Different model-specific loading/runtime failures." }, { "left": "issue:43881", "right": "issue:43883", "accept": false, - "reason": "Both are model-loading issues, but they affect different models and failure mechanisms." + "reason": "Different models and unrelated errors." }, { "left": "issue:45685", "right": "issue:45823", "accept": false, - "reason": "MPS histogram/int limitation vs Gemma4 PLE device mismatch are different backend bugs." + "reason": "Different device/backend issues with different failure signatures." } ] }, "evaluator_result": { "accept": true, - "feedback": "Accept. The canonical anchor choice and the soft-pair judgments are conservative and grounded in the packet: the accepted pairs are all clear same-bug families by title/body framing (non-persistent buffers, StaticCache.get_seq_length, tokenizer regression, SigLIP2 discrepancy), and the rejected pairs are appropriately kept separate by model/code-path differences. I don\u2019t see an obvious overstatement." + "feedback": "The summary is grounded in the packet, and the accepted soft pairs are conservative and supported by the issue titles/body context. No clear overstatement or non-mergeable PR pairing appears in the chosen cluster." }, "evaluator_used": true, "retried": false, @@ -1598,10 +1526,10 @@ "issue:43828|issue:45823", "issue:45020|issue:45356", "issue:45715|issue:46315", - "issue:43827|issue:44509", "issue:45488|issue:46489", - "issue:46894|issue:46897", + "issue:43827|issue:44509", "issue:43872|issue:46869", + "issue:46894|issue:46897", "issue:43404|issue:43950", "issue:45759|issue:46661", "issue:43872|issue:44291", @@ -1621,223 +1549,217 @@ "issue:44530|issue:44863", "issue:44863|issue:46402" ], - "prepared_review_unit_hash": "100af846bdef61f1ee147fce57474a9894ecc6623c94204688c10dec265418af", + "prepared_review_unit_hash": "b053d3cd771ec9efffc27d7954a63581b25b52b933839d960c8bd53a78824221", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 37086, - "estimated_input_tokens": 9272, - "estimated_eval_tokens": 18800 + "serialized_chars": 37083, + "estimated_input_tokens": 9271, + "estimated_eval_tokens": 18798 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The packet is mostly heterogeneous. The docs regression pair and the mask-shape IndexError pair look like true duplicates; the Whisper/fbank pair is only a weak thematic match and should not be accepted on the evidence shown.", - "confidence": 0.81, - "canonical_issue_reason": "No single issue represents the whole packet well. The clearest duplicate subclusters are 43827/44509 for the stale v5 docs references and 41093/44805 for the same mask-shape IndexError; 43827 is the cleanest docs representative.", + "summary": "Mostly a heterogeneous set of unrelated bug reports. A few soft pairs are true duplicates or close enough to share one fix: the v5 docs pipeline-removal reports, the `_is_hf_initialized` load-path regression, the Qwen3/Qwen3.5 ZeRO-3 weight-loading failures, and the two training-loss double-shift reports. Most other pairs only share a broad subsystem or similar wording.", + "confidence": 0.84, + "canonical_issue_reason": "No single canonical issue covers the whole cluster; the strongest representative of the true duplicate-like reports is issue 46897, which is a concrete, high-signal instance of the repeated training-loss double-shift bug.", "canonical_pr_reason": null, - "best_issue_reason": "43827 is the strongest overall issue representative because it cleanly states the v5 docs regression and has a clear duplicate in 44509.", + "best_issue_reason": "Issue 46897 is the cleanest standalone issue in the cluster: it is specific, clearly reproducible, and its sibling 46894 looks like the same underlying bug report with a different affected model.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45081", "right": "issue:45399", "accept": false, - "reason": "Tokenizer regex patching and flash-attn2 fallback checks are different bugs with different code paths." + "reason": "Different bugs: tokenizer regex patch crash vs flash-attn2 fallback gating." }, { "left": "issue:45020", "right": "issue:45823", "accept": false, - "reason": "Remote-code loading breakage and a device-map forward mismatch are not the same concrete failure." + "reason": "Both are remote/device-map regressions, but the failures and code paths are different." }, { "left": "issue:44661", "right": "issue:45020", "accept": false, - "reason": "add-new-model-like/tokenizer mapping logic is unrelated to the broader remote_code regression." + "reason": "One is add-new-model-like tokenizer mapping logic; the other is a broad remote_code regression." }, { "left": "issue:42491", "right": "issue:44368", "accept": false, - "reason": "LoRA compatibility across versions is not the same bug as a missing gradient-checkpointing flag warning." + "reason": "Same model family, but one is a version-compatibility load failure and the other is a warning about tie_word_embeddings." }, { "left": "issue:43828", "right": "issue:45823", "accept": false, - "reason": "Dtype mismatch under autocast and device mismatch with auto device_map are different failure modes." + "reason": "Different symptoms: autocast dtype mismatch vs device_map forward device mismatch." }, { "left": "issue:45020", "right": "issue:45356", "accept": false, - "reason": "Generic remote_code breakage and a tokenizer codec regression do not share one concrete bug path." + "reason": "Both mention tokenizer/model loading, but the root causes are distinct." }, { "left": "issue:45715", "right": "issue:46315", "accept": false, - "reason": "These are both tokenizer performance issues, but they affect opposite conversion directions and likely different code paths." - }, - { - "left": "issue:43827", - "right": "issue:44509", - "accept": true, - "reason": "Same stale v5 docs regression: both report documentation still referencing removed pipeline tasks." + "reason": "Both are tokenizer performance issues, but they affect different methods and likely different hot paths." }, { "left": "issue:45488", "right": "issue:46489", "accept": false, - "reason": "Both involve tokenizers, but the affected models and bugs are different and not shown to be the same underlying issue." + "reason": "Same model family, but one is a tokenizer-pretokenizer override regression and the other is an output-correctness issue with a different described fix gap." }, { - "left": "issue:46894", - "right": "issue:46897", - "accept": false, - "reason": "The titles suggest a similar training-loss pattern, but different models and insufficient evidence prevent treating them as the same bug." + "left": "issue:43827", + "right": "issue:44509", + "accept": true, + "reason": "Both report the same docs regression: removed pipeline tasks still referenced after v5." }, { "left": "issue:43872", "right": "issue:46869", "accept": false, - "reason": "bitsandbytes constructor incompatibility is unrelated to the YOLO multi-GPU device_map failure." + "reason": "Unexpected kwarg vs multi-GPU YOLO device-map failure; not the same bug." + }, + { + "left": "issue:46894", + "right": "issue:46897", + "accept": true, + "reason": "Nearly identical training-loss double-shift reports; same underlying bug, just different affected model." }, { "left": "issue:43404", "right": "issue:43950", "accept": false, - "reason": "lm_head tying in a specific multimodal model is unrelated to non-persistent buffer corruption in from_pretrained()." + "reason": "lm_head tying bug vs non-persistent buffer corruption are unrelated model-loading issues." }, { "left": "issue:45759", "right": "issue:46661", "accept": false, - "reason": "Different model-construction failures: text_config unwrapping versus remote-code config_class access." + "reason": "Different remote-code/config handling failures with no clear shared concrete code path." }, { "left": "issue:43872", "right": "issue:44291", - "accept": false, - "reason": "Both mention _is_hf_initialized, but the concrete failures differ: bitsandbytes Int8Params versus init_empty_weights." + "accept": true, + "reason": "Both center on the unexpected `_is_hf_initialized` argument during model loading/initialization." }, { "left": "issue:45020", "right": "issue:46661", "accept": false, - "reason": "Broad remote-code breakage and missing config_class on model_class are distinct loading failures." + "reason": "Shared remote-code theme, but the reported failures are not specific enough to treat as the same bug." }, { "left": "issue:43792", "right": "issue:44220", "accept": false, - "reason": "Both mention Whisper/fbank, but the evidence here does not establish the same concrete buggy code path, so this is only uncertain." + "reason": "Different audio/model-loading failures; not the same code-path problem." }, { "left": "issue:43257", "right": "issue:45305", "accept": false, - "reason": "Both touch DeepSpeed/Qwen training, but one is missing MoE weight conversion and the other is gradient averaging with GAS." + "reason": "Both involve DeepSpeed, but one is missing Qwen3 MOE weight conversion and the other is gradient averaging with loss kwargs." }, { "left": "issue:41553", "right": "issue:44991", "accept": false, - "reason": "Bad tokenizer error messaging for Voxtral is unrelated to the est-roberta tokenizer load failure." + "reason": "Tokenizer-loading failures, but for different models and different failure modes." }, { "left": "issue:36331", "right": "issue:41093", "accept": false, - "reason": "Unexpected compute_loss kwarg and mask/tensor shape mismatch are different trainer/runtime bugs." + "reason": "Trainer signature mismatch vs tensor mask shape mismatch; unrelated." }, { "left": "issue:42947", "right": "issue:45923", "accept": false, - "reason": "Gradient checkpointing being ineffective with PEFT LoRA is not the same as a missing supports_gradient_checkpointing flag." + "reason": "Both mention gradient checkpointing, but one is a LoRA/PEFT ineffectiveness issue and the other is a missing flag on a remote-code variant." }, { "left": "issue:43854", "right": "issue:44863", "accept": false, - "reason": "Different model-loading problems for different model families; no shared underlying bug is evident." + "reason": "Different model-specific loading failures with no shared concrete bug." }, { "left": "issue:43940", "right": "issue:45313", - "accept": false, - "reason": "Both are ZeRO-3 loading failures on Qwen variants, but the symptoms and affected components differ enough to keep them separate." + "accept": true, + "reason": "Same underlying Qwen3-family ZeRO-3 weight-loading failure, just reported on different model variants." }, { "left": "issue:41553", "right": "issue:44488", "accept": false, - "reason": "Different tokenizer-loading failures: a Voxtral error message issue versus a model loading failure for cjvt/sleng-bert." + "reason": "Bad error message vs model-loading failure for a specific checkpoint; not the same issue." }, { "left": "issue:44928", "right": "issue:45588", "accept": false, - "reason": "RLHF NaN/SDPA fallback issue is unrelated to flash_attention.py crashing on sink-less models." + "reason": "Gradient explosion in RLHF vs flash-attention crash on sink-less models are unrelated." }, { "left": "issue:41093", "right": "issue:44805", - "accept": true, - "reason": "Same mask-shape IndexError; 44805 is a later, more detailed report of the same bug." + "accept": false, + "reason": "Same generic IndexError wording, but likely different masks/tensors and different root causes." }, { "left": "issue:44479", "right": "issue:44560", "accept": false, - "reason": "Both are video regressions in Qwen-family models, but the affected models and failure signatures differ." + "reason": "Both are video-input regressions, but they affect different model families and symptoms." }, { "left": "issue:24643", "right": "issue:30064", "accept": false, - "reason": "DeepSpeed training weight-shape error is unrelated to void segmentation map processing." + "reason": "Training DeepSpeed weight-shape error vs segmentation-map processing failure; unrelated." }, { "left": "issue:44530", "right": "issue:44863", "accept": false, - "reason": "PagedAttentionCache linear_attention crash and NemotronH checkpoint loading are different model/backend failures." + "reason": "PagedAttention linear_attention crash vs checkpoint-loading failure; different bugs." }, { - "left": "issue:43872", - "right": "issue:44291", - "accept": false, - "reason": "Reiterating the shared keyword is not enough; the constructor/type mismatch contexts differ and should stay separate." - }, - { - "left": "issue:45020", - "right": "issue:46661", + "left": "issue:44863", + "right": "issue:46402", "accept": false, - "reason": "Remote-code breakage and model_class.config_class access errors are distinct problems." + "reason": "Different model-loading regressions for different model families." } ] }, "evaluator_result": { - "accept": false, - "feedback": "Mostly grounded, but one soft-edge acceptance is too strong for the packet: 43792/44220 only share a Whisper/fbank theme from titles, and the evidence shown here does not establish the same concrete bug path. The docs pair (43827/44509) and mask-shape pair (41093/44805) look fine, but the Whisper pair should be downgraded to uncertain rather than accepted." + "accept": true, + "feedback": "The summary is grounded in the packet, and the accepted soft edges are stated conservatively enough from the available evidence (mainly title-level similarity). The only caveat is that a couple of the duplicate-like calls rely on shared bug patterns rather than explicit cross-links, but that is still within a reasonable conservative read of the packet." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -1921,15 +1843,15 @@ "issue:43645|issue:43646", "issue:44560|issue:45685", "issue:43550|issue:44910", - "issue:42617|issue:43366", "issue:44683|issue:46515", + "issue:42617|issue:43366", "issue:44560|issue:44918", "issue:36010|issue:42915", "issue:45812|issue:46489", "issue:45823|issue:46515", "issue:44661|issue:45823", - "issue:38175|issue:43994", "issue:44560|issue:44805", + "issue:38175|issue:43994", "issue:45672|issue:46899", "issue:44877|issue:45030", "issue:43632|issue:45834", @@ -1941,14 +1863,14 @@ "issue:38175|issue:39692", "issue:43644|issue:44534" ], - "prepared_review_unit_hash": "6e2d7f93d566bf7e242bc3d9f595168bf62ba865e2a93038586c65a87e34cf98", + "prepared_review_unit_hash": "809b93a4ffc5c33e193f8488befbd30656ac3414fc7f1184af0634fc25adda80", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, @@ -1963,228 +1885,228 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous transformer bug reports, with only a few true duplicate pairs: the v5 non-persistent-buffer regression, the custom-model initialization regression, a SigLIP2 output bug pair, and a Gemma4 4-bit dtype/casting pair. Most other soft links are broad-similarity false positives across different models or code paths.", - "confidence": 0.86, - "canonical_issue_reason": "43644 is the clearest canonical anchor because it is a tightly scoped v5 regression with a direct duplicate in 44534 and cleanly represents one real duplicate subgroup in an otherwise mixed cluster.", + "summary": "This cluster is mostly heterogeneous: most pairs are only superficially similar, but a few look like true duplicates, especially the non-persistent-buffer regression and a couple of model-specific output/dtype regressions.", + "confidence": 0.71, + "canonical_issue_reason": "issue:43644 is the clearest canonical issue in the set: it states the root regression directly, is broad enough to subsume the near-identical follow-up report 44534, and is more precise than the more symptom-driven or model-specific issues.", "canonical_pr_reason": null, - "best_issue_reason": "43644 is the best issue representative: concrete, reproducible, and backed by an exact duplicate (44534), making it the strongest anchor for this cluster.", + "best_issue_reason": "issue:43644 is the best consolidation target because it has the cleanest bug description and the strongest duplicate overlap, while most other issues in the cluster are unrelated.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43366", "right": "issue:43828", "accept": false, - "reason": "Different bugs and different model families: GGUF architecture support vs Phi autocast dtype mismatch." + "reason": "Different bugs: GGUF gpt-oss support vs a Phi-tiny-MoE dtype/autocast failure." }, { "left": "issue:43596", "right": "issue:46515", "accept": false, - "reason": "Zero3/BertModel index error is unrelated to a general torch version incompatibility report." + "reason": "Both mention failures, but one is a deepspeed/Bert index error and the other is a torch version incompatibility report." }, { "left": "issue:45325", "right": "issue:45902", "accept": false, - "reason": "Both involve model internals, but one is a Qwen2.5-VL rope scaling bug and the other is Qwen3.5-MoE uninitialized memory after meta materialization." + "reason": "Different model families and different code paths: Qwen2.5-VL rope indexing vs Qwen3_5Moe rotary inv_freq materialization." }, { "left": "issue:43366", "right": "issue:43531", "accept": false, - "reason": "GGUF gpt-oss support and Qwen3-MoE sliding_window are unrelated issues." + "reason": "Unrelated issues: GGUF support for gpt-oss vs Qwen3-MoE sliding_window behavior." }, { "left": "issue:43638", "right": "issue:46515", "accept": false, - "reason": "Same broad area of loading/runtime failures, but the concrete bugs are different: Bert/zero3 index error vs torch 2.4.1 incompatibility." + "reason": "A similar stack is involved, but the concrete bug is different: Bert/deepspeed zero3 index error vs general torch incompatibility." }, { "left": "issue:45663", "right": "issue:45823", "accept": false, - "reason": "Both are Gemma4, but one is an FSDP2 shared_kv_states training bug and the other is a device_map forward mismatch." + "reason": "Both are Gemma4, but one is FSDP2 shared_kv_states rebuilding and the other is device_map auto PLE mismatch." }, { "left": "issue:44155", "right": "issue:45584", "accept": false, - "reason": "Both are audio-related, but one is batched inference leakage in AudioFlamingo3 and the other is Whisper empty-transcription generation failure." + "reason": "Different subsystems and symptoms: AudioFlamingo3 batching leak vs Whisper empty-transcription generation failure." }, { "left": "issue:43366", "right": "issue:44589", "accept": false, - "reason": "GGUF architecture support is unrelated to the Float8 storage-object loading error." + "reason": "No shared bug: gpt-oss GGUF support vs a Float8 storage lookup error." }, { "left": "issue:43824", "right": "issue:43957", "accept": false, - "reason": "Importing Qwen2_5_VLForConditionalGeneration and meta-device loading failures are different code paths and different symptoms." + "reason": "ImportError for a specific class is not the same bug as meta-device loading breaking some models." }, { "left": "issue:43950", "right": "issue:45834", "accept": false, - "reason": "Non-persistent buffer corruption and Kosmos2.5 OCR index errors are unrelated." + "reason": "The buffer-corruption regression is unrelated to Kosmos2.5 long-OCR indexing." }, { "left": "issue:42915", "right": "issue:43957", "accept": false, - "reason": "Qwen3-MoE FineGrainedFP8Config failure is unrelated to generic meta-device loading breakage." + "reason": "Different failures: Qwen3Moe FP8 config vs generic meta-device model loading." }, { "left": "issue:44910", "right": "issue:46858", "accept": false, - "reason": "Both involve attention/cache mechanics, but the concrete failures are different: packed 3D position_ids on Qwen3.5 vs device tensor seq_length arithmetic on non-CUDA backends." + "reason": "Distinct attention/cache bugs: Qwen3.5 flash-attention position_ids vs static-cache seq_length tensor arithmetic." }, { "left": "issue:43716", "right": "issue:44560", "accept": false, - "reason": "Mistral-3 image dtype mismatch and Qwen3-vl video StopIteration are distinct multimodal bugs." + "reason": "Different model paths and errors: Mistral-3 preprocessor dtype mismatch vs Qwen3-vl video StopIteration." }, { "left": "issue:43645", "right": "issue:43646", "accept": true, - "reason": "These are the same underlying Transformers v5 custom model initialization regression; the notebook mention is just a reproduction context." + "reason": "These describe the same underlying regression: custom model initialization breaks in Transformers 5, with one report just being the Jupyter-specific case." }, { "left": "issue:44560", "right": "issue:45685", "accept": false, - "reason": "Qwen3-vl video StopIteration is unrelated to the MPS histogram implementation error for MoE." + "reason": "Completely different issues: video embedding StopIteration vs MPS histogram on Int." }, { "left": "issue:43550", "right": "issue:44910", "accept": false, - "reason": "Different failure modes: torch.compile with SDPA on Bamba vs FlashAttention illegal memory access on Qwen3.5." + "reason": "Both are attention-related, but the concrete failures are different code paths and different root causes." }, { - "left": "issue:42617", - "right": "issue:43366", + "left": "issue:44683", + "right": "issue:46515", "accept": false, - "reason": "3d_parallel.py runtime failure and GGUF gpt-oss support are unrelated." + "reason": "Compiled flex_attention on torch >=2.9 is unrelated to a torch 2.4.1 incompatibility report." }, { - "left": "issue:44683", - "right": "issue:46515", + "left": "issue:42617", + "right": "issue:43366", "accept": false, - "reason": "Compiled flex_attention on torch >=2.9 and torch 2.4.1 incompatibility are different issues." + "reason": "A runtime script failure is not the same as GGUF gpt-oss architecture support." }, { "left": "issue:44560", "right": "issue:44918", "accept": false, - "reason": "Different Qwen3.x multimodal/training failures with no shared concrete code path." + "reason": "Different bugs: Qwen3-vl video StopIteration vs TRL embedding unpacking failure." }, { "left": "issue:36010", "right": "issue:42915", "accept": false, - "reason": "GenerationMixin import error and Qwen3-MoE FP8 failure are unrelated." + "reason": "Import path regression for GenerationMixin is unrelated to Qwen3Moe FineGrainedFP8Config failure." }, { "left": "issue:45812", "right": "issue:46489", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different model families and different tokenizer code paths." + "reason": "Both are tokenizer-related, but they concern different model families and different underlying causes." }, { "left": "issue:45823", "right": "issue:46515", "accept": false, - "reason": "Gemma4 device_map mismatch and generic torch incompatibility are not the same bug." + "reason": "Gemma4 device placement bug vs a generic torch version incompatibility." }, { "left": "issue:44661", "right": "issue:45823", "accept": false, - "reason": "Tokenizer-mapping failure and Gemma4 PLE device mismatch are unrelated." - }, - { - "left": "issue:38175", - "right": "issue:43994", - "accept": true, - "reason": "Same SigLIP2 model, same symptom class of bad/incorrect outputs; these look like the same underlying model/processor bug." + "reason": "Tokenizer mapping registration failure is unrelated to Gemma4 PLE device mismatch." }, { "left": "issue:44560", "right": "issue:44805", "accept": false, - "reason": "Qwen3-vl video StopIteration and a generic mask/tensor shape IndexError are different failures." + "reason": "Different symptoms and code paths: video StopIteration vs mask/tensor shape mismatch." + }, + { + "left": "issue:38175", + "right": "issue:43994", + "accept": true, + "reason": "Same model and same class of bad-output regression for SigLIP2; both reports point to the same inference path failing." }, { "left": "issue:45672", "right": "issue:46899", "accept": true, - "reason": "Both describe Gemma4 4-bit inference breakage caused by unsafe dtype handling in multimodal/audio-vision embedders." + "reason": "Both report the same Gemma4 multimodal 4-bit dtype-cast regression affecting audio/vision modules." }, { "left": "issue:44877", "right": "issue:45030", "accept": false, - "reason": "Config validation regressions on different model families are too broad to merge as one duplicate." + "reason": "Different model/config validation failures with no shared code path." }, { "left": "issue:43632", "right": "issue:45834", "accept": false, - "reason": "_is_hf_initialized flag breakage is unrelated to Kosmos2.5 OCR indexing." + "reason": "_is_hf_initialized flag regression is unrelated to Kosmos2.5 OCR index errors." }, { "left": "issue:43638", "right": "issue:44661", "accept": false, - "reason": "Zero3/Bert index error and add-new-model-like/tokenizer mapping failure are unrelated." + "reason": "Bert/deepspeed zero3 initialization and tokenizer mapping registration are different bugs." }, { "left": "issue:43596", "right": "issue:44661", "accept": false, - "reason": "Same as above: Bert zero3 initialization bug vs tokenizer mapping bug." + "reason": "No shared underlying bug: Bert zero3 index error vs tokenizer mapping issue." }, { "left": "issue:42491", "right": "issue:43720", "accept": false, - "reason": "LoRA hf4.x/hf5.x compatibility and BitNet packed-weight loading are different problems." + "reason": "LoRA/model-compatibility regression is unrelated to packed-weight unloading in BitNet accelerate loading." }, { "left": "issue:44661", "right": "issue:44805", "accept": false, - "reason": "Tokenizer mapping failure and generic mask-shape IndexError are unrelated." + "reason": "Tokenizer mapping registration issue is unrelated to a tensor mask shape error." }, { "left": "issue:43317", "right": "issue:44756", "accept": false, - "reason": "device_map auto dequantized-model offload failure and Strix Halo mmap OOM are different loading/resource issues." + "reason": "Both involve loading/memory, but one is dequantized offload on device_map=auto and the other is Strix Halo mmap OOM." }, { "left": "issue:38175", "right": "issue:39692", "accept": false, - "reason": "Both involve SigLIP2, but 39692 is a docs/example-specific model/processor mismatch plus quantization failure, not the same concrete runtime bug as 38175." + "reason": "Same model family, but one is bad inference outputs while the other is a docs-example/quantization failure; not the same concrete bug." }, { "left": "issue:43644", "right": "issue:44534", "accept": true, - "reason": "Exact same non-persistent-buffer regression in Transformers v5; 44534 is a later duplicate of 43644." + "reason": "Near-identical reports of the same non-persistent-buffer junk/corruption regression in Transformers 5.x." } ] }, "evaluator_result": { "accept": true, - "feedback": "The cluster summary is grounded in the packet and stays conservative: it treats most links as false positives, and the accepted soft edges are supported by closely aligned titles (especially the near-identical v5 buffer-regression pair and the custom-model init pair). The canonical 43644 anchor is well supported by its exact duplicate 44534." + "feedback": "The consolidation choice is grounded in the packet: issue 43644 and 44534 have nearly identical titles, and the other accepted pairs are likewise supported by the issue titles as same-bug follow-ups rather than merely similar symptoms. The rejected soft edges are conservative and mostly stay at the title level without overstating evidence. No mergeability concerns apply here since these are issue pairs, not PR pairs." }, "evaluator_used": true, "retried": false, @@ -2259,8 +2181,8 @@ "issue:42915|issue:44756", "issue:44451|issue:44488", "issue:44960|issue:44977", - "issue:44863|issue:46846", "issue:46459|issue:46512", + "issue:44863|issue:46846", "issue:45313|issue:45357", "issue:44488|issue:44991", "issue:45507|issue:46468", @@ -2270,8 +2192,8 @@ "issue:43257|issue:43854", "issue:43856|issue:43873", "issue:43856|issue:44387", - "issue:43854|issue:43931", "issue:43708|issue:44075", + "issue:43854|issue:43931", "issue:42491|issue:45335", "issue:44960|issue:46050", "issue:46897|issue:46901", @@ -2281,224 +2203,224 @@ "issue:44164|issue:44291", "issue:44910|issue:45588", "issue:44402|issue:44488", - "issue:39692|issue:44960", "issue:43425|issue:43957", + "issue:39692|issue:44960", "issue:41720|issue:44155" ], - "prepared_review_unit_hash": "2c1e5e71bea78e2ac069db7d107676d421e06d1a52e1391d59f9800899bbf0e3", + "prepared_review_unit_hash": "8fb1f90a10c2bfc133ef8024730424e3ee1e93da1e83354d6593fd74fe2e6368", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36099, - "estimated_input_tokens": 9025, - "estimated_eval_tokens": 18306 + "serialized_chars": 36103, + "estimated_input_tokens": 9026, + "estimated_eval_tokens": 18308 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose collection of Transformers 5.x/model-specific regressions rather than one unified bug. I only see one strong duplicate pair: the `resize_token_embeddings` / output-embedding regression across Gemma4 and T5Gemma; the rest are too model-, API-, or failure-mode-specific to merge safely.", - "confidence": 0.58, - "canonical_issue_reason": "`issue:45276` is the clearest representative of the duplicate `resize_token_embeddings` regression: it names the concrete API, the affected embedding surfaces, and the expected propagation failure.", + "summary": "Mostly a heterogeneous mix of unrelated model-loading, tokenizer, quantization, and training bugs. Only one soft edge looks like a true duplicate: the two training-loss double-shift reports. The tokenizer/model-loading pairs are similar in theme but not strong enough to merge confidently.", + "confidence": 0.68, + "canonical_issue_reason": "Issue 43873 is the broadest active bug in the set and the best umbrella representative for the recurring runtime-memory/quantization theme, even though the cluster is not a tight duplicate group.", "canonical_pr_reason": null, - "best_issue_reason": "`issue:45276` is the strongest issue to anchor the cluster because it is specific, actionable, and broad enough to explain the matched T5Gemma report as the same root problem.", + "best_issue_reason": "Issue 43873 is the best single triage anchor: it is open, clearly actionable, and closest to the common inference-memory/offloading failures that recur across several items.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42491", "right": "issue:45276", "accept": false, - "reason": "Different bugs: LoRA/qwen3_moe compatibility versus a `resize_token_embeddings` embedding-propagation regression." + "reason": "Different model families and different failures; only broad embedding/resize wording overlaps." }, { "left": "issue:41720", "right": "issue:42915", "accept": false, - "reason": "Both mention Qwen3-related failures, but one is auto device mapping/cuda assert and the other is fine-grained FP8 config." + "reason": "Different Qwen3 failures: auto device mapping cuda assert vs FineGrainedFP8Config load failure." }, { "left": "issue:42915", "right": "issue:44756", "accept": false, - "reason": "Distinct failures: Qwen3Moe FP8 config versus mmap/OOM on Strix Halo." + "reason": "Unrelated bugs: Qwen3 MoE FP8 failure vs Strix Halo mmap/OOM behavior." }, { "left": "issue:44451", "right": "issue:44488", "accept": false, - "reason": "Both are model-loading complaints, but they target different checkpoints and no shared root cause is evident." + "reason": "Both are BERT-family load failures, but the titles point to different model-specific regressions without enough evidence of the same root cause." }, { "left": "issue:44960", "right": "issue:44977", "accept": false, - "reason": "Unrelated model/runtime failures: GLM5 versus Qwen3.5 flash-attention generation." + "reason": "Different models and different flash-attention problems." }, { - "left": "issue:44863", - "right": "issue:46846", + "left": "issue:46459", + "right": "issue:46512", "accept": false, - "reason": "Different code paths: NemotronH checkpoint loading versus Qwen3.5 tensor-parallel planning." + "reason": "Different code paths: dtype ignored on composite checkpoint load vs stale config.dtype after from_config." }, { - "left": "issue:46459", - "right": "issue:46512", + "left": "issue:44863", + "right": "issue:46846", "accept": false, - "reason": "Both involve load-time state/dtype handling, but one is composite-checkpoint dtype propagation and the other is `from_config` stale config metadata." + "reason": "Different model families and different TP/load-planning failures." }, { "left": "issue:45313", "right": "issue:45357", "accept": false, - "reason": "ZeRO-3 weight loading and save_pretrained visual-key regression are separate bugs." + "reason": "Both involve Qwen3.5, but one is ZeRO-3 weight loading and the other is save_pretrained visual-key regression." }, { "left": "issue:44488", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer-load regressions, but they affect different models and likely different failure points." + "reason": "Same general tokenizer-loading theme, but different checkpoints and no clear proof of the same underlying bug." }, { "left": "issue:45507", "right": "issue:46468", "accept": false, - "reason": "Different model implementations and symptoms: invalid method call in GraniteMoEHybrid versus a broken Nemotron example." + "reason": "Different models and unrelated invalid-method vs broken-example failures." }, { "left": "issue:43644", "right": "issue:43645", "accept": false, - "reason": "Transformers 5.0 regressions, but one is buffer initialization and the other is Jupyter custom-model construction." + "reason": "Same release, but distinct bugs: non-persistent buffer junk vs Jupyter custom-model initialization." }, { "left": "issue:42491", "right": "issue:45292", "accept": false, - "reason": "LoRA/qwen3_moe incompatibility is not the same bug as `resize_token_embeddings` not updating output embeddings." + "reason": "Different issues; only a shared resize/token-embedding theme." }, { "left": "issue:39692", "right": "issue:43994", "accept": false, - "reason": "Different SigLIP/SigLIP2 model-loading and inference problems; not the same concrete defect." + "reason": "Both mention SigLIP2, but one is a docs example with quantization/model-processor mismatch and the other is nonsensical inference output." }, { "left": "issue:43257", "right": "issue:43854", "accept": false, - "reason": "Qwen3 MoE deepspeed conversion and GLM-4.7-Flash unit-test loading are unrelated." + "reason": "Different models and different failure modes; only broad loading trouble overlaps." }, { "left": "issue:43856", "right": "issue:43873", "accept": false, - "reason": "Both are quantization/memory complaints, but one is training-memory usage and the other is offloading behavior." + "reason": "Both are memory-related, but one is Qwen3 MoE training efficiency and the other is quantization/offloading behavior." }, { "left": "issue:43856", "right": "issue:44387", "accept": false, - "reason": "Same broad area, but the specific problem differs: training memory usage versus int4 quantization reserved-memory OOM." + "reason": "Related to memory pressure, but one is training inefficiency and the other is an int4 quantization regression." }, { - "left": "issue:43854", - "right": "issue:43931", + "left": "issue:43708", + "right": "issue:44075", "accept": false, - "reason": "Different models and failure modes: GLM load issue versus Qwen3-VL weight-shape mismatch." + "reason": "Completely different training scheduler vs optimizer-argument bug." }, { - "left": "issue:43708", - "right": "issue:44075", + "left": "issue:43854", + "right": "issue:43931", "accept": false, - "reason": "Trainer resume/max-steps logic and SGD argument handling are unrelated." + "reason": "Different models and different loading failures." }, { "left": "issue:42491", "right": "issue:45335", "accept": false, - "reason": "LoRA/qwen3_moe incompatibility is not the same concrete issue as T5Gemma embedding resizing." + "reason": "Both mention embeddings, but the model subclasses and broken paths differ." }, { "left": "issue:44960", "right": "issue:46050", "accept": false, - "reason": "GLM5 loading/generation versus quantized-model save/load corruption are different bugs." + "reason": "Different serialization bugs: GLM5 model issue vs sinq save/load corruption." }, { "left": "issue:46897", "right": "issue:46901", - "accept": false, - "reason": "They share a \"double-shift\" theme, but the affected models and exact training paths differ, so this is too weak to merge." + "accept": true, + "reason": "Both describe the same training-loss double-shift bug: loss is computed against shifted labels instead of the original labels." }, { "left": "issue:45488", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but one is a pre-tokenizer override bug and the other is wrong token IDs for Granite models." + "reason": "Both are tokenizer regressions, but one is a LlamaTokenizer pre-tokenizer override and the other is wrong token IDs for Granite models." }, { "left": "issue:45276", "right": "issue:45335", - "accept": true, - "reason": "Both report the same `resize_token_embeddings` propagation bug: model-specific output/decoder embeddings are not updated after resizing." + "accept": false, + "reason": "Same API surface, but likely model-specific resize_token_embeddings omissions rather than one shared bug." }, { "left": "issue:45362", "right": "issue:45406", "accept": false, - "reason": "Qwen3.5 chat crash and Gemma4Processor serve crash are unrelated failures." + "reason": "Unrelated crashes in transformers chat vs transformers serve, with different model/processor causes." }, { "left": "issue:44164", "right": "issue:44291", "accept": false, - "reason": "Save/from_pretrained extra_state handling and init_empty_weights argument compatibility are different loading paths." + "reason": "Different save/load internals: extra_state handling vs init_empty_weights argument mismatch." }, { "left": "issue:44910", "right": "issue:45588", "accept": false, - "reason": "Both are flash-attention crashes, but one is illegal memory access from 3D position IDs and the other is an AttributeError on `s_aux=None`." + "reason": "Both are flash_attention.py issues, but the bugs are different: packed-sequence misinterpretation vs s_aux=None AttributeError." }, { "left": "issue:44402", "right": "issue:44488", "accept": false, - "reason": "Tokenizer vocab-size mismatch and model-loading failure are not the same bug." + "reason": "Related tokenizer/vocab symptoms, but one is a vocab-size mismatch and the other is a load failure." }, { - "left": "issue:39692", - "right": "issue:44960", + "left": "issue:43425", + "right": "issue:43957", "accept": false, - "reason": "Completely different model families and symptoms: SigLIP2 example errors versus GLM5." + "reason": "Torch version compatibility vs meta-device model-loading breakage are different problems." }, { - "left": "issue:43425", - "right": "issue:43957", + "left": "issue:39692", + "right": "issue:44960", "accept": false, - "reason": "Torch version incompatibility and meta-device model-loading regression are separate issues." + "reason": "Unrelated SigLIP2 docs/runtime issue vs GLM5 issue." }, { "left": "issue:41720", "right": "issue:44155", "accept": false, - "reason": "Qwen3 device-mapping CUDA assert and AudioFlamingo3 batched inference leakage are unrelated." + "reason": "Different subsystems and failures: Qwen3 device mapping vs AudioFlamingo3 batched inference leakage." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative: it correctly treats most pairs as model-specific or failure-mode-specific, and the only accepted duplicate pair (issue:45276 with issue:45335) is well supported by the titles. No overstatement of evidence stood out." + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft edge, issue:46897 vs issue:46901, is well supported by the titles and descriptions as the same training-loss double-shift pattern. The other pairwise rejections are cautious rather than overstated, and the umbrella/canonical choice of issue:43873 is defensible as an open, actionable anchor." }, "evaluator_used": true, "retried": false, @@ -2573,8 +2495,8 @@ "issue:44079|issue:46829", "issue:43404|issue:43454", "issue:44291|issue:44756", - "issue:43873|issue:44387", "issue:42947|issue:45704", + "issue:43873|issue:44387", "issue:44292|issue:44912", "issue:36010|issue:39404", "issue:45663|issue:45859", @@ -2601,226 +2523,226 @@ "issue:44589|issue:44756", "issue:43425|issue:43828" ], - "prepared_review_unit_hash": "4b15f3806416f168c260cbeeed709a0526076d87446ba77e255c614a01a60181", + "prepared_review_unit_hash": "544cbb29b5bae9efff4d0eb1792a3139adff137110832aa7c9a80a435d63e111", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37127, - "estimated_input_tokens": 9282, - "estimated_eval_tokens": 18820 + "serialized_chars": 37130, + "estimated_input_tokens": 9283, + "estimated_eval_tokens": 18822 }, "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 resize_token_embeddings regression across Gemma/t5gemma variants; most other pairs are merely superficially similar and should stay separate.", - "confidence": 0.71, - "canonical_issue_reason": "issue:45292 is the best anchor for the only real duplicate family here: it states the generic resize_token_embeddings failure, while issue:45276 and issue:45335 are narrower model-specific variants of the same regression.", + "summary": "Mostly unrelated issues; only a small duplicate cluster around `resize_token_embeddings` and a likely shared training-loss double-shift bug stand out.", + "confidence": 0.81, + "canonical_issue_reason": "issue:45292 is the most generic and central `resize_token_embeddings` regression, and it has the strongest overlap with the closely related embedding-resize reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45292 is the most representative and reusable issue title in the cluster: it describes the shared bug at the right abstraction level without overfitting to one model name.", + "best_issue_reason": "Best representative of the cluster because it is the clearest, most general duplicate hub among the accepted issue pairs.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45663", "right": "issue:45759", "accept": false, - "reason": "Different subsystems and failure modes: Gemma4 FSDP2 shared_kv_states rebuilding vs Qwen multimodal config unwrapping." + "reason": "Different model paths and failures: FSDP2 shared-kv state rebuild vs. composite Qwen config unwrapping." }, { "left": "issue:44079", "right": "issue:46829", "accept": false, - "reason": "Both involve output containers, but one is a ModelOutput key assignment bug and the other is a VideoPrism hidden_states field mix-up." + "reason": "Both involve output structures, but they describe different bugs in different classes of outputs." }, { "left": "issue:43404", "right": "issue:43454", "accept": false, - "reason": "Both mention lm_head tying, but they are different model implementations and not clearly the same code-path bug." + "reason": "Both mention untied `lm_head`, but they are separate model families and not clearly the same concrete code-path bug." }, { "left": "issue:44291", "right": "issue:44756", "accept": false, - "reason": "init_empty_weights argument handling and mmap/OOM on Strix Halo are unrelated issues." + "reason": "Unrelated loading/initialization failure vs. mmap/OOM platform issue." }, { - "left": "issue:43873", - "right": "issue:44387", + "left": "issue:42947", + "right": "issue:45704", "accept": false, - "reason": "Both touch quantization memory behavior, but one is offloading and the other is int4 reserved-memory/OOM; too broad to merge." + "reason": "Gradient checkpointing with LoRA is unrelated to the Apex RMSNorm memory leak." }, { - "left": "issue:42947", - "right": "issue:45704", + "left": "issue:43873", + "right": "issue:44387", "accept": false, - "reason": "Gradient checkpointing/LoRA effectiveness and T5 apex RMSNorm memory leak are unrelated." + "reason": "Both are quantization/memory related, but one is offloading behavior and the other is int4 OOM; not the same bug." }, { "left": "issue:44292", "right": "issue:44912", "accept": false, - "reason": "Different quantization formats and symptoms; one is NVFP4 loading, the other MXFP4 fallback to bf16." + "reason": "Different quantization formats and outcomes: load failure vs. fallback to bf16." }, { "left": "issue:36010", "right": "issue:39404", "accept": false, - "reason": "ImportError for GenerationMixin and Whisper pipeline return_language are unrelated." + "reason": "Import error for `GenerationMixin` is unrelated to Whisper pipeline `return_language` behavior." }, { "left": "issue:45663", "right": "issue:45859", "accept": false, - "reason": "FSDP2 shared_kv_states KeyError and CPU-offload incompatibility in Qwen rotary embedding are different bugs." + "reason": "Different model-specific offload/forward problems with no clear shared code-path." }, { "left": "issue:44387", "right": "issue:44756", "accept": false, - "reason": "Both mention OOM, but the causes are different: int4 quantization memory growth vs disabling mmap on Strix Halo." + "reason": "Both mention OOM, but the causes are unrelated." }, { "left": "issue:42947", "right": "issue:43381", "accept": false, - "reason": "LoRA gradient checkpointing ineffectiveness is not the same as checkpointing being disallowed in eval mode." + "reason": "Both involve gradient checkpointing, but one is LoRA ineffectiveness and the other is eval-mode incompatibility." }, { "left": "issue:44155", "right": "issue:45823", "accept": false, - "reason": "AudioFlamingo3 batch leakage and Gemma4 device mismatch are unrelated model-specific forward bugs." + "reason": "Different model families and failures; no same underlying bug is evident." }, { "left": "issue:43957", "right": "issue:44756", "accept": false, - "reason": "meta-device loading regressions and Strix Halo mmap OOM are not the same issue." + "reason": "Meta-device loading issues are not the same as the Strix Halo mmap OOM bug." }, { "left": "issue:43335", "right": "issue:45663", "accept": false, - "reason": "SwitchTransformers sparse-layer construction and Gemma4 FSDP2 state rebuilding are unrelated." + "reason": "Separate bugs in different models: SwitchTransformers config construction vs. Gemma4 FSDP2 attention state." }, { "left": "issue:45276", "right": "issue:45292", "accept": true, - "reason": "Same underlying resize_token_embeddings regression; 45292 is the generic version and 45276 is a model-specific manifestation." + "reason": "Same `resize_token_embeddings` regression; 45276 is a model-specific instance of the generic output_embeddings problem in 45292." }, { "left": "issue:44863", "right": "issue:46468", "accept": false, - "reason": "NemotronH checkpoint loading and the official Nemotron example being broken are related only at the model family level, not clearly the same bug." + "reason": "Nemotron checkpoint loading and the broken official example are related by model family but not clearly the same bug." }, { "left": "issue:44479", "right": "issue:44877", "accept": false, - "reason": "Video input regression for Qwen VL models and strict config blocking granite_speech loading are unrelated." + "reason": "Different regressions in unrelated model/config loading paths." }, { "left": "issue:43381", "right": "issue:44928", "accept": false, - "reason": "Checkpointing-in-eval and RLHF gradient explosion from 3D position_ids are different problems." + "reason": "Gradient checkpointing in eval mode is unrelated to the Qwen3.5 RLHF NaN issue." }, { "left": "issue:44164", "right": "issue:44756", "accept": false, - "reason": "save/from_pretrained extra_state handling and Strix Halo mmap OOM are unrelated." + "reason": "Save/from_pretrained extra_state handling is unrelated to the Strix Halo mmap OOM bug." }, { "left": "issue:45684", "right": "issue:45698", "accept": false, - "reason": "Both involve custom-model save/load, but one is read-only file permissions during save_pretrained and the other is loading the wrong custom module." + "reason": "Both concern custom-model save/load, but one is permissions propagation and the other is wrong module loading." }, { "left": "issue:45588", "right": "issue:45663", "accept": false, - "reason": "Flash attention s_aux=None crash and Gemma4 shared_kv_states rebuilding are unrelated." + "reason": "Flash-attention sinkless-model crash is unrelated to the Gemma4 FSDP2 KeyError." }, { "left": "issue:43854", "right": "issue:45216", "accept": false, - "reason": "GLM-4.7-Flash unit-test loading and Qwen3.5 checkpoint save_pretrained regression are different model-specific issues." + "reason": "Different model-loading regressions and different failure modes." }, { "left": "issue:42796", "right": "issue:46326", "accept": false, - "reason": "Both are tokenizer-related, but BPE marker handling and continue_final_message behavior are distinct bugs." + "reason": "Tokenizer BPE-marker handling is not the same bug as `continue_final_message` on mistral_common tokenizers." }, { "left": "issue:45292", "right": "issue:45335", "accept": true, - "reason": "Same resize_token_embeddings bug family; both report embeddings not being updated after resize in different model wrappers." + "reason": "Same class of bug: `resize_token_embeddings` fails to propagate resized embeddings to model-specific embedding references." }, { "left": "issue:46894", "right": "issue:46901", - "accept": false, - "reason": "Both describe double-shift loss errors, but they are in different model implementations and not clearly one shared code-path bug." + "accept": true, + "reason": "Both describe the same training-loss double-shift bug: the loss is computed against labels shifted by one position instead of the original labels." }, { "left": "issue:43828", "right": "issue:43957", "accept": false, - "reason": "Autocast dtype mismatch and meta-device loading failures are unrelated." + "reason": "Different dtype/meta-loading regressions with no evidence of the same underlying issue." }, { "left": "issue:44402", "right": "issue:44451", "accept": false, - "reason": "Tokenizer vocab-size mismatch and ScandiBERT loading failure are not the same bug." + "reason": "Both are model-load problems, but one is vocab/token placement and the other is a separate loading failure." }, { "left": "issue:43646", "right": "issue:43708", "accept": false, - "reason": "Custom model initialization in notebooks and Trainer resume max_steps miscalculation are unrelated." + "reason": "Custom model initialization breakage and Trainer `resume_from_checkpoint` max-step miscalculation are unrelated." }, { "left": "issue:43645", "right": "issue:44164", "accept": false, - "reason": "Custom model init in Jupyter and extra_state save/load failures are different defects." + "reason": "Notebook custom-model initialization and extra_state save/load are different issues." }, { "left": "issue:44589", "right": "issue:44756", "accept": false, - "reason": "Float8 storage lookup error and mmap-related OOM are unrelated." + "reason": "Storage-object lookup failure is unrelated to the Strix Halo mmap/OOM issue." }, { "left": "issue:43425", "right": "issue:43828", "accept": false, - "reason": "Torch 2.10 compatibility and Phi-tiny-MoE autocast dtype mismatch are not the same issue." + "reason": "Torch version compatibility and autocast dtype mismatch are separate problems." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded and conservative overall. The resize_token_embeddings family is the only clearly supported duplicate cluster in the packet, and the other pairwise rejections stay appropriately cautious rather than over-claiming equivalence." + "feedback": "The analysis is grounded in the packet and stays conservative on the soft edges. The accepted duplicate pairs all look mergeable into single bugfix PRs: the resize_token_embeddings cluster (45276/45292/45335) and the training-loss double-shift pair (46894/46901). Rejections are generally cautious and not overstated." }, "evaluator_used": true, "retried": false, @@ -2894,9 +2816,9 @@ "issue:46315|issue:46491", "issue:45812|issue:46315", "issue:39404|issue:45701", - "issue:39404|issue:46665", "issue:44530|issue:45084", "issue:43611|issue:43645", + "issue:39404|issue:46665", "issue:44683|issue:45823", "issue:43425|issue:46869", "issue:43425|issue:43716", @@ -2925,232 +2847,232 @@ "issue:44488|issue:45081", "issue:43611|issue:43644" ], - "prepared_review_unit_hash": "dc9f7882e8112fc4bf12345cf5e8b69d51a6dc7ee971808871054e820c56df2b", + "prepared_review_unit_hash": "c01c773c21081d09e102b1f758de2084a71d89e1cea0150b5009665743e298c8", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37732, - "estimated_input_tokens": 9433, - "estimated_eval_tokens": 19122 + "serialized_chars": 37739, + "estimated_input_tokens": 9435, + "estimated_eval_tokens": 19126 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "A loose v5-regression cluster dominated by tokenizer/model-loading failures, with a few unrelated export/training/memory bugs mixed in. I would keep it mostly split; I don\u2019t see any soft pair that is clearly the same underlying bug.", - "confidence": 0.72, - "canonical_issue_reason": "issue:43611 is the broadest and most central v5 loading-regression report in the set; it best serves as an umbrella anchor, even though the cluster itself is heterogeneous.", + "summary": "Most items are separate regression reports across different models, tokenizer paths, or infra behaviors. The closest thematic area is v5 model-loading regressions, with issue 43611 as the best broad representative, but the soft pairs are still too heterogeneous to merge as duplicates.", + "confidence": 0.78, + "canonical_issue_reason": "Issue 43611 is the broadest and most central core regression report here: it targets Transformers 5.0.0 model-loading behavior and is closest in scope to several other loading/init complaints.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43611 is the best representative because it describes a generic Transformers 5 loading breakage rather than a single model-specific symptom.", + "best_issue_reason": "Issue 43611 is the best overall issue candidate because it is a clear, general loading regression rather than a model-specific symptom, so it best represents the cluster theme.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46315", "right": "issue:46491", "accept": false, - "reason": "Both are tokenizer regressions, but one is a slowdown in convert_tokens_to_ids and the other is a CodeLlama encode/decode space bug; different code paths and fixes." + "reason": "One is a tokenizer performance regression in convert_tokens_to_ids; the other is a CodeLlama encode/decode space-roundtrip correctness bug. Different code paths and symptoms." }, { "left": "issue:45812", "right": "issue:46315", "accept": false, - "reason": "Granite token ID corruption and slow slow-tokenizer conversion are distinct bugs; same subsystem, not the same underlying defect." + "reason": "Granite token IDs are wrong; the other is slower conversion for slow tokenizers with many added tokens. Same tokenizer area, but not the same bug." }, { "left": "issue:39404", "right": "issue:45701", "accept": false, - "reason": "Whisper pipeline return_language regression is unrelated to a general tokenization behavior change report." - }, - { - "left": "issue:39404", - "right": "issue:46665", - "accept": false, - "reason": "Pipeline return_language failure and from_pretrained-in-classes inconsistency are different API behaviors and different fixes." + "reason": "Whisper pipeline return_language regression is unrelated to a general tokenization-version-change report." }, { "left": "issue:44530", "right": "issue:45084", "accept": false, - "reason": "PagedAttentionCache invalid group type and template-node compilation failure are separate internals with no clear shared root cause." + "reason": "PagedAttentionCache/linear_attention crash on Qwen3.5 is unrelated to a template-node compilation TypeError." }, { "left": "issue:43611", "right": "issue:43645", "accept": false, - "reason": "base_model_prefix loading breakage and Jupyter custom-model initialization issues are both v5 regressions, but not the same code-path problem." + "reason": "Both involve v5 model init/loading regressions, but one is base_model_prefix loading and the other is defining custom models in notebooks; not the same concrete bug." + }, + { + "left": "issue:39404", + "right": "issue:46665", + "accept": false, + "reason": "Whisper pipeline return_language and from_pretrained inside Python classes are distinct behaviors with different failure modes." }, { "left": "issue:44683", "right": "issue:45823", "accept": false, - "reason": "Compiled flex_attention on torch>=2.9 and Gemma4 device mismatch are unrelated compatibility bugs." + "reason": "Torch >=2.9 flex_attention compilation failure is unrelated to Gemma4 device placement mismatch." }, { "left": "issue:43425", "right": "issue:46869", "accept": false, - "reason": "Torch 2.10 incompatibility is not the same as multi-GPU device_map auto failure on YOLO models." + "reason": "A torch version incompatibility report is not the same as a YOLO multi-GPU device_map bug." }, { "left": "issue:43425", "right": "issue:43716", "accept": false, - "reason": "Torch version incompatibility and Mistral-3 image-preprocessor dtype mismatch are different issues." + "reason": "Torch 2.10 incompatibility and Mistral-3 dtype mismatch are different problems." }, { "left": "issue:43425", "right": "issue:45823", "accept": false, - "reason": "General torch incompatibility does not match the Gemma4 PLE device-mapping bug." + "reason": "Library-version incompatibility is not the same as a device_map forward-pass mismatch." }, { "left": "issue:44589", "right": "issue:45084", "accept": false, - "reason": "Missing Float8 storage and non-template-node compilation errors are unrelated failures." + "reason": "Float8 storage lookup failure and non-template-node compilation are unrelated runtime errors." }, { "left": "issue:44479", "right": "issue:44805", "accept": false, - "reason": "Qwen video-input regression and a mask-shape IndexError are not the same underlying bug." + "reason": "Video-input regression for Qwen VL models and a generic mask/index shape mismatch are not the same underlying defect." }, { "left": "issue:43906", "right": "issue:44530", "accept": false, - "reason": "An isolated reproduction of a prior issue and a Qwen3.5 PagedAttentionCache crash are different problems." + "reason": "An isolated reproduction of another issue is not the same as the PagedAttentionCache linear_attention crash." }, { "left": "issue:43901", "right": "issue:44509", "accept": false, - "reason": "These are both docs regressions, but they concern different removed/changed pipeline APIs." + "reason": "Both are docs-related, but they describe different pipeline task doc regressions." }, { "left": "issue:44671", "right": "issue:44991", "accept": false, - "reason": "CamemBERT masked-LM prediction errors and est-roberta tokenizer loading failures are unrelated." + "reason": "CamemBERT masked-LM prediction regression and tokenizer loading failure for a different model are distinct." }, { "left": "issue:43994", "right": "issue:44079", "accept": false, - "reason": "SigLIP2 nonsensical outputs and ModelOutput key assignment with None are different bugs." + "reason": "SigLIP2 incorrect outputs and ModelOutput key assignment logic are different bugs in different layers." }, { "left": "issue:44265", "right": "issue:44898", "accept": false, - "reason": "torch.export failure with torch_compilable_check and Perceiver non-default resolution failure are distinct export/model issues." + "reason": "torch.export/exportability via torch_compilable_check is unrelated to Perceiver interpolation at non-default resolution." }, { "left": "issue:36032", "right": "issue:43452", "accept": false, - "reason": "T5 tokenizer method-name conflict and gguf_file load failures are different tokenizer/loading bugs." + "reason": "T5Tokenizer add_special_tokens conflict and gguf_file/from_pretrained breakage are different tokenizer/model-loading failures." }, { "left": "issue:39404", "right": "issue:44912", "accept": false, - "reason": "Whisper pipeline regression and MXFP4 quantization fallback are unrelated." + "reason": "Whisper return_language pipeline behavior is unrelated to MXFP4 quantization fallback on git-oss-20b." }, { "left": "issue:46050", "right": "issue:46326", "accept": false, - "reason": "sinq save/load breakage and continue_final_message on mistral_common are separate features and code paths." + "reason": "Quantized save/load breakage after sinq is not the same as the mistral_common continue_final_message bug." + }, + { + "left": "issue:44164", + "right": "issue:45759", + "accept": false, + "reason": "extra_state save/load handling and unwrapping text_config for composite Qwen configs are different code paths." }, { "left": "issue:43906", "right": "issue:46326", "accept": false, - "reason": "The reproduction of an older issue is not the same as the mistral_common continue_final_message bug." + "reason": "A reproduction of another issue does not match the mistral_common continue_final_message regression." }, { "left": "issue:44991", "right": "issue:45081", "accept": false, - "reason": "Tokenizer load failure for est-roberta and a Mistral regex patch backend_tokenizer error are different tokenizer issues." + "reason": "Both are tokenizer-related, but one is wrong loading behavior for a specific model and the other is a Mistral regex patch attribute error; not the same bug." }, { "left": "issue:43299", "right": "issue:43931", "accept": false, - "reason": "Both involve Qwen3-VL loading, but one is a dev0 MoE loading regression and the other is a specific weight-shape mismatch; too little evidence of one shared bug." + "reason": "Both involve Qwen3-VL loading, but one is a broad dev-version regression for MoE models and the other is a specific weight-shape mismatch; too different to merge." }, { "left": "issue:45823", "right": "issue:46899", "accept": false, - "reason": "Gemma4 device-mismatch during forward and Gemma4 4-bit multimodal embedder dtype-cast breakage are separate runtime bugs." + "reason": "Gemma4 device_map mismatch during forward and unsafe dtype casting in multimodal embedders are separate multimodal bugs." }, { "left": "issue:45479", "right": "issue:46894", "accept": false, - "reason": "Single-label classification zero-loss behavior is not the same as a training-loss double-shift issue." + "reason": "Degenerate zero loss in sequence classification is unrelated to the ASR double-shift training-loss bug." }, { "left": "issue:43856", "right": "issue:44756", "accept": false, - "reason": "Qwen3 MoE memory usage and Strix Halo mmap OOM are different resource issues." + "reason": "Qwen3 MoE training memory inefficiency and disabling mmap on Strix Halo are unrelated performance/oom reports." }, { "left": "issue:43596", "right": "issue:44683", "accept": false, - "reason": "Deepspeed ZeRO3/BertModel index error and torch>=2.9 compiled flex_attention failure are unrelated." + "reason": "Zero3/BertModel indexing failure and compiled flex_attention incompatibility are different failures." }, { "left": "issue:45479", "right": "issue:46897", "accept": false, - "reason": "Degenerate classification loss and Florence2 double-shift training loss are not the same defect." + "reason": "The sequence-classification zero-loss bug is not the same as the Florence2 training-loss double-shift regression." }, { "left": "issue:44164", "right": "issue:46821", "accept": false, - "reason": "extra_state save/load handling and TimesFM window_size attribute access are different from-pretrained/config issues." + "reason": "extra_state handling and TimesFM window_size AttributeError are unrelated issues." }, { "left": "issue:44488", "right": "issue:44987", "accept": false, - "reason": "Loading cjvt/sleng-bert and loading physical-intelligence/fast on transformers>=5.1.0 are model-specific failures with no clear shared root cause." + "reason": "Both are loading failures, but for different models and likely different root causes." }, { "left": "issue:44488", "right": "issue:45081", "accept": false, - "reason": "Generic model loading failure and Mistral regex patch backend_tokenizer crash are different problems." - }, - { - "left": "issue:44488", - "right": "issue:45081", - "accept": false, - "reason": "Same pair as above; the symptoms and code paths do not align." + "reason": "A sleng-bert load failure and the Mistral regex/backend_tokenizer error are distinct tokenizer-loading bugs." }, { "left": "issue:43611", "right": "issue:43644", "accept": false, - "reason": "Model-loading with base_model_prefix and non-persistent buffer initialization are separate v5 regressions." + "reason": "Both are v5 regressions in model handling, but one is about base_model_prefix loading and the other about non-persistent buffers; not the same concrete fix." } ] }, @@ -3228,8 +3150,8 @@ "issue:42796|issue:43404", "issue:46285|issue:46402", "issue:36032|issue:42915", - "issue:32090|issue:39404", "issue:43716|issue:45084", + "issue:32090|issue:39404", "issue:44737|issue:45859", "issue:44661|issue:44683", "issue:43643|issue:44530", @@ -3239,14 +3161,14 @@ "issue:43541|issue:45084", "issue:43644|issue:43646", "issue:43582|issue:43632", - "issue:34567|issue:40444", "issue:42175|issue:46178", + "issue:34567|issue:40444", "issue:43883|issue:44488", "issue:40444|issue:46326", "issue:44206|issue:44220", "issue:44811|issue:45584", - "issue:43452|issue:43482", "issue:36032|issue:43645", + "issue:43452|issue:43482", "issue:44987|issue:44991", "issue:43317|issue:43482", "issue:45084|issue:45507", @@ -3255,224 +3177,221 @@ "issue:43883|issue:44991", "issue:43646|issue:44534" ], - "prepared_review_unit_hash": "af9882f1d3323fb9213d212c18ff5b929da7aee986ea066b9d82de2d3a47dbf4", + "prepared_review_unit_hash": "e7ad9176e7c6e9cc6a6295fda029b3ad1c4fdd435c78ae6fc9da670be8517064", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36693, - "estimated_input_tokens": 9174, - "estimated_eval_tokens": 18604 + "serialized_chars": 36691, + "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 near-matches only by subsystem or release theme, not the same concrete bug. The strongest duplicates are the audio feature-extraction reports and the GGUF loading reports; the rest should stay separate.", - "confidence": 0.71, - "canonical_issue_reason": "Issue 43452 is the broadest, most concrete loading regression here: it describes a generic GGUF/from_pretrained break that can subsume specific model reports more cleanly than the narrower, model-specific issues.", + "summary": "This cluster is heterogeneous: most items are unrelated Transformers regressions or model-specific loading/tokenizer bugs, with only superficial overlap around v5 and initialization/loading behavior. I would not treat these as duplicates; 43646 is only the broadest representative issue, not a true canonical duplicate target.", + "confidence": 0.31, + "canonical_issue_reason": "Issue 43646 is the broadest, most generic regression report in the set ('custom model initialization' under Transformers 5.0.0), so it is the least specific representative even though the cluster is not a true duplicate group.", "canonical_pr_reason": null, - "best_issue_reason": "43452 is the best canonical issue candidate because it states the failure mode at the right abstraction level (GGUF loading via from_pretrained) while still naming the impacted APIs clearly.", + "best_issue_reason": "43646 is the best single representative because it describes a broad Transformers v5 regression affecting initialization, which is more general than the other model- or tokenizer-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44462", "right": "issue:46489", "accept": false, - "reason": "Both are tokenizer-loading regressions, but 46489 is a model-specific DeepSeek-Coder v1 bug tied to a gap in PR #44801, not clearly the same root cause as \"ignores tokenizer.json\"." + "reason": "Both concern tokenizer loading, but one is about repo tokenizer.json being ignored and the other is a DeepSeek-Coder v1 tokenizer regression tied to a specific v5+ fix gap; not the same bug." }, { "left": "issue:42796", "right": "issue:43404", "accept": false, - "reason": "Tokenizer BPE-marker handling and Mistral3 lm_head tying are unrelated code paths and symptoms." + "reason": "Tokenizer BPE-marker handling and a tied lm_head weight bug in Mistral3 are different code paths and failure modes." }, { "left": "issue:46285", "right": "issue:46402", "accept": false, - "reason": "One is a CLIP model structure inconsistency, the other is missing weight loading in Sam3Model; different model families and bugs." + "reason": "Different models and different regressions: CLIP text model structure mismatch vs Sam3 text-encoder weights not loading." }, { "left": "issue:36032", "right": "issue:42915", "accept": false, - "reason": "T5 tokenizer special-token name conflict and Qwen3Moe FP8 failure are different failures with no shared concrete code path." + "reason": "T5 tokenizer attribute conflict and Qwen3MoE FP8 config failure are unrelated tokenizer vs quantization issues." }, { - "left": "issue:32090", - "right": "issue:39404", + "left": "issue:43716", + "right": "issue:45084", "accept": false, - "reason": "Trainer GPU broadcast NoneType error and Whisper return_language pipeline regression are unrelated." + "reason": "Image preprocessor dtype mismatch is not the same as a template compilation error." }, { - "left": "issue:43716", - "right": "issue:45084", + "left": "issue:32090", + "right": "issue:39404", "accept": false, - "reason": "Image preprocessor dtype mismatch is not the same bug as template compilation failing on non-template nodes." + "reason": "Trainer GPU broadcast TypeError and Whisper pipeline return_language breakage are unrelated subsystems and bugs." }, { "left": "issue:44737", "right": "issue:45859", "accept": false, - "reason": "Both involve device placement, but XLNet CPU arange and Qwen3_Moe CPU offload are distinct implementation bugs." + "reason": "Both involve rotary embeddings / offload concerns, but one is CPU device placement and the other is a Qwen3.5MoE offload compatibility issue." }, { "left": "issue:44661", "right": "issue:44683", "accept": false, - "reason": "Tokenizer mapping/model-addition logic and compiled flex_attention on torch 2.9 are unrelated." + "reason": "Tokenizer mapping lookup failure and compiled flex_attention on torch>=2.9 are different failures." }, { "left": "issue:43643", "right": "issue:44530", "accept": false, - "reason": "Missing fields from trust_remote_code config return and PagedAttentionCache linear_attention failure are different issues." + "reason": "trust_remote_code config field loss and PagedAttentionCache linear_attention crash are separate issues." }, { "left": "issue:44451", "right": "issue:44987", "accept": false, - "reason": "Both are model-loading failures, but the affected models and likely failure points differ; too broad to merge." + "reason": "Both are loading failures for different models, but the causes are different and model-specific." }, { "left": "issue:40990", "right": "issue:43421", "accept": false, - "reason": "High perplexity on a model benchmark and runtime post-processor updates after special-token edits are not the same bug." + "reason": "High perplexity on gpt-oss and runtime post-processor updates for special tokens are not the same bug." }, { "left": "issue:42915", "right": "issue:44530", "accept": false, - "reason": "FP8 config failure for Qwen3Moe and PagedAttentionCache group-type crash are separate model/runtime problems." + "reason": "Different Qwen model families and different problems: FP8 config vs cache/group type crash." }, { "left": "issue:43541", "right": "issue:45084", "accept": false, - "reason": "Torch dynamo grouped_mm tracing failure and template compilation errors are different code paths." + "reason": "Torch dynamo tracing on Mixtral float32 grouped_mm is unrelated to template-node compilation errors." }, { "left": "issue:43644", "right": "issue:43646", "accept": false, - "reason": "Both are v5 regressions around custom model initialization, but one is specifically about non-persistent buffers being junk-filled while the other is broader; not enough to treat as the same bug." + "reason": "Both are Transformers v5 regressions, but one is junk-filled non-persistent buffers and the other is custom model initialization failure; related theme, not the same underlying bug." }, { "left": "issue:43582", "right": "issue:43632", "accept": false, - "reason": "Apple Silicon warmup TypeError and _is_hf_initialized flag breakage are unrelated regressions." + "reason": "AppleSilicon warmup TypeError and _is_hf_initialized flag breakage are distinct initialization/runtime issues." }, { - "left": "issue:34567", - "right": "issue:40444", + "left": "issue:42175", + "right": "issue:46178", "accept": false, - "reason": "TrainerState token counting and multi-image iterable dataset finetuning failures are not the same issue." + "reason": "Missing tensorflow backend in '.[torch]' install and undefined tensorflow_text backend are both backend-mapping problems, but not the same missing dependency or code path." }, { - "left": "issue:42175", - "right": "issue:46178", + "left": "issue:34567", + "right": "issue:40444", "accept": false, - "reason": "Backend packaging omission for pip extras and an unsupported tensorflow_text backend mapping error are distinct problems." + "reason": "TrainerState token counting and Qwen2.5-VL iterable dataset multi-image failure are unrelated." }, { "left": "issue:43883", "right": "issue:44488", "accept": false, - "reason": "Missing tied-weights metadata on Molmo and a specific tokenizer load failure for cjvt/sleng-bert are unrelated." + "reason": "Both mention model loading failures, but one is a missing tied-weights attribute and the other is a tokenizer-loading regression for a specific model." }, { "left": "issue:40444", "right": "issue:46326", "accept": false, - "reason": "Multi-image Qwen2.5-VL finetuning and Mistral-common continue_final_message behavior are different features and failures." + "reason": "Multi-image IterableDataset finetuning failure and mistral_common continue_final_message behavior are different multimodal/tokenizer bugs." }, { "left": "issue:44206", "right": "issue:44220", - "accept": true, - "reason": "These look like the same audio feature-extraction bug: both point at _torch_extract_fbank_features() / feature extractor crashes, with the second being the helper-function manifestation of the first." + "accept": false, + "reason": "LasrFeatureExtractor unsupported center arg and _torch_extract_fbank_features are closely related audio preprocessing issues, but the reports point to different API breakages and likely different fixes." }, { "left": "issue:44811", "right": "issue:45584", "accept": false, - "reason": "Both are Whisper-related, but one is batch_decode skip_special_tokens handling and the other is empty-transcription failure after align_special_tokens; different behavior." - }, - { - "left": "issue:43452", - "right": "issue:43482", - "accept": true, - "reason": "Both are GGUF loading regressions in transformers v5: 43452 describes the generic from_pretrained break, and 43482 is a specific Qwen2.5-GGUF manifestation of that same loading path failure." + "reason": "Both involve Whisper token handling, but batch_decode skip_special_tokens and empty-transcription generation after align_special_tokens are different behaviors." }, { "left": "issue:36032", "right": "issue:43645", "accept": false, - "reason": "T5 tokenizer method conflict and custom model init failures in notebooks are different regressions." + "reason": "T5 tokenizer method conflict and Jupyter custom model initialization are not the same bug." + }, + { + "left": "issue:43452", + "right": "issue:43482", + "accept": false, + "reason": "gguf_file breaking from_pretrained and Qwen2.5-GGUF loading failure are both GGUF-related but not the same concrete failure mode." }, { "left": "issue:44987", "right": "issue:44991", "accept": false, - "reason": "Both are version-dependent tokenizer/model load failures, but they affect different models and do not clearly share the same root cause." + "reason": "Both are loading regressions in transformers>=5, but they affect different models and likely different tokenizer/config paths." }, { "left": "issue:43317", "right": "issue:43482", "accept": false, - "reason": "GPU/CPU offload dequantized-model loading and GGUF loading on v5 are different failure modes." + "reason": "GPU+CPU offload dequantized load failure and Qwen2.5-GGUF v5 loading failure are separate loader/offload bugs." }, { "left": "issue:45084", "right": "issue:45507", "accept": false, - "reason": "Template compilation errors and GraniteMoEHybrid invalid method calls are unrelated." + "reason": "Template compilation error and GraniteMoEHybrid invalid method calls are unrelated." }, { "left": "issue:43262", "right": "issue:43377", "accept": false, - "reason": "Sampling-rate defaulting in audio chat templates and MIMI batched-vs-single padding mask mismatch are different bugs." + "reason": "Audio processor sampling-rate defaulting and MIMI padding-mask batch inconsistency are different processor/encoder bugs." }, { "left": "issue:43746", "right": "issue:46326", "accept": false, - "reason": "PEFT adapter loading for GraniteSpeech and mistral_common continue_final_message behavior are unrelated." + "reason": "PEFT adapter local checkpoint loading and mistral_common continue_final_message behavior do not share the same underlying issue." }, { "left": "issue:43883", "right": "issue:44991", "accept": false, - "reason": "Molmo tied-weights metadata and tokenizer loading for EMBEDDIA/est-roberta are not the same problem." + "reason": "Missing all_tied_weights_keys and tokenizer loading regression for EMBEDDIA/est-roberta are different failures." }, { "left": "issue:43646", "right": "issue:44534", "accept": false, - "reason": "Both are Transformers v5 regressions, but one is custom model initialization and the other is non-persistent buffers being filled with junk; related theme, not the same concrete bug." + "reason": "These are very similar on the surface, but 44534 is the generic 'non-persistent buffers junk' regression, while 43646 is specifically custom model initialization; related family, not a duplicate." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative overall. The analyst mostly rejects weak subsystem-level similarities, and the two accepted pairs are plausibly the same regressions based on the issue titles alone: the audio feature-extraction pair shares the same helper/path, and the GGUF pair is a generic vs. specific manifestation of the same loading failure. No PR-pair mergeability issue applies here." - }, - "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 }, @@ -3502,9 +3421,9 @@ "issue:43425", "issue:43531", "issue:43540", - "issue:43611", "issue:43632", "issue:43645", + "issue:43716", "issue:43742", "issue:43792", "issue:43819", @@ -3547,12 +3466,12 @@ "issue:43377|issue:43819", "issue:43404|issue:44466", "issue:44464|issue:45987", - "issue:44008|issue:46899", "issue:44987|issue:45081", + "issue:44008|issue:46899", "issue:43994|issue:44610", "issue:44451|issue:45081", - "issue:28530|issue:34567", "issue:42915|issue:44291", + "issue:28530|issue:34567", "issue:36032|issue:43742", "issue:44062|issue:44987", "issue:44265|issue:44610", @@ -3563,221 +3482,227 @@ "issue:43632|issue:46821", "issue:36032|issue:41720", "issue:45216|issue:45357", - "issue:44991|issue:46315", "issue:43531|issue:44877", + "issue:44991|issue:46315", "issue:43425|issue:44403", "issue:43296|issue:43531", "issue:44220|issue:44683", - "issue:43611|issue:44534" + "issue:43716|issue:46899" ], - "prepared_review_unit_hash": "d37f949bf763e90852afa16d45be39485f2ccb3984715c31c86b2efba4f29a4d", + "prepared_review_unit_hash": "b32089a9bd297c139f9979ca2ea916a80ee73c6009dfc57af8bdf95fa2e99e2e", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36744, - "estimated_input_tokens": 9186, - "estimated_eval_tokens": 18628 + "serialized_chars": 36729, + "estimated_input_tokens": 9183, + "estimated_eval_tokens": 18622 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose collection of Transformers regressions; only two pairs look like true duplicates. No PRs are present.", - "confidence": 0.44, - "canonical_issue_reason": "issue:45216 is the clearest duplicate anchor because issue:45357 looks like the same Qwen3.5 save_pretrained / visual-encoder-key regression continuing into a later release.", + "summary": "This cluster is mostly a mix of unrelated bugs across loading, generation, multimodal processing, and trainer internals. Only one soft edge looks like a true duplicate/follow-up pair: the Qwen3.5 save_pretrained regression in 45216/45357.", + "confidence": 0.89, + "canonical_issue_reason": "Issue 45216 is the best canonical anchor: it reports the broader Qwen3.5 save_pretrained checkpoint regression, and 45357 reads like a later narrower follow-up about the same incorrect visual-encoder key serialization. The rest of the cluster is unrelated noise.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43611 is the broadest and most central issue in the set, covering a core Transformers v5 model-loading regression theme rather than a single model-specific symptom.", + "best_issue_reason": "45216 is the strongest representative issue because it is the earlier, broader report in the only true duplicate pair and gives the clearest problem statement.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45084", "right": "issue:46869", "accept": false, - "reason": "Different subsystems and failure modes: template compilation vs YOLO multi-GPU device mapping." + "reason": "Different failures: template-node compilation vs YOLO multi-GPU device-mapping cuda assert." }, { "left": "issue:43792", "right": "issue:46869", "accept": false, - "reason": "Unrelated models and bugs; Whisper loading/can't run is not the same as YOLO device-map failures." + "reason": "Unrelated model/runtime problems: Whisper inference failure vs YOLO auto device map failure." }, { "left": "issue:44403", "right": "issue:45081", "accept": false, - "reason": "Both touch loading behavior, but one is generic loading noise and the other is a tokenizer regex crash; not the same bug." + "reason": "One is a logging/noise complaint; the other is a concrete tokenizer regex crash in Mistral loading." }, { "left": "issue:43645", "right": "issue:44534", "accept": false, - "reason": "Both are v5 regressions, but one is custom-model init in notebooks and the other is non-persistent buffer corruption." + "reason": "Both are v5 regressions, but they hit different code paths: notebook custom-model init vs non-persistent buffer initialization." }, { "left": "issue:44849", "right": "issue:45325", "accept": false, - "reason": "Same family, but different concrete symptoms: hidden-states output bug vs RoPE/position-id scaling bug." + "reason": "Different Qwen model bugs: hidden-states output handling vs rope index scaling for still images/video positions." }, { "left": "issue:43377", "right": "issue:43819", "accept": false, - "reason": "Both are audio-model issues, but they involve different models and different code paths." + "reason": "Both involve output mismatches, but they concern different models and mechanisms (padding-mask support vs missing STE)." }, { "left": "issue:43404", "right": "issue:44466", "accept": false, - "reason": "Both mention tied lm_head weights, but one is missing tying and the other is serialization varying by device; not the same concrete defect." + "reason": "Related theme of lm_head/tied weights, but one is a specific Mistral3 tying bug and the other is device-dependent serialization behavior." }, { "left": "issue:44464", "right": "issue:45987", "accept": false, - "reason": "Compiled generation inconsistency and cache sequence-length return type are different issues." + "reason": "Chunked generation inconsistency is not the same bug as StaticCache.get_seq_length returning the wrong type." }, { - "left": "issue:44008", - "right": "issue:46899", + "left": "issue:44987", + "right": "issue:45081", "accept": false, - "reason": "Different multimodal models and distinct bugs: variable shadowing vs unsafe dtype casting." + "reason": "Both are load failures, but one is a tokenizer loading regression for a specific model and the other is a Mistral regex patch crash." }, { - "left": "issue:44987", - "right": "issue:45081", + "left": "issue:44008", + "right": "issue:46899", "accept": false, - "reason": "Different tokenizer/loading failures; no clear shared code-path or defect." + "reason": "Both are multimodal dtype-related, but they affect different model families and have different root causes." }, { "left": "issue:43994", "right": "issue:44610", "accept": false, - "reason": "Different model families and symptoms: bad SigLIP outputs vs OmDet-Turbo image-size mismatch." + "reason": "Different vision issues: nonsensical SigLIP2 outputs vs OmDet-Turbo processor/model input-size mismatch." }, { "left": "issue:44451", "right": "issue:45081", "accept": false, - "reason": "Unrelated loading failure vs Mistral tokenizer regex crash." + "reason": "Both are tokenizer/model loading failures, but the models and underlying causes differ." }, { - "left": "issue:28530", - "right": "issue:34567", + "left": "issue:42915", + "right": "issue:44291", "accept": false, - "reason": "Trainer early-stopping config issue and token-count tracking bug are unrelated." + "reason": "Different initialization failures: FineGrainedFP8Config vs unexpected _is_hf_initialized argument in init_empty_weights." }, { - "left": "issue:42915", - "right": "issue:44291", + "left": "issue:28530", + "right": "issue:34567", "accept": false, - "reason": "Different problems: quantization/config handling vs init_empty_weights argument compatibility." + "reason": "Trainer early-stopping metric selection and num_input_tokens_seen updating are unrelated trainer issues." }, { "left": "issue:36032", "right": "issue:43742", "accept": false, - "reason": "Tokenizer method-name collision and MobileLLM key error are not the same bug." + "reason": "Different loading errors: T5Tokenizer add_special_tokens conflict vs MobileLLM key error." }, { "left": "issue:44062", "right": "issue:44987", "accept": false, - "reason": "Both are tokenizer-related, but one is an AddedToken argument conflict and the other is a model/tokenizer loading failure." + "reason": "AddedToken special-argument duplication is a tokenizer constructor bug, not the same as a model-specific tokenizer load failure." }, { "left": "issue:44265", "right": "issue:44610", "accept": false, - "reason": "torch.export/compilable-check failure is unrelated to an image processor size mismatch." + "reason": "torch.export/torch_compilable_check failure is unrelated to OmDet-Turbo processor image sizing." }, { "left": "issue:42915", "right": "issue:44589", "accept": false, - "reason": "Both involve version compatibility, but the concrete failures differ completely." + "reason": "FineGrainedFP8Config failure is unrelated to missing Float8 storage type during serialization/loading." }, { "left": "issue:43089", "right": "issue:45987", "accept": false, - "reason": "Generation performance overhead is not the same as cache length contract behavior." + "reason": "Generation performance overhead is not the same as a cache API contract/type bug." }, { "left": "issue:43322", "right": "issue:43540", "accept": false, - "reason": "Both are multimodal loading/runtime issues, but they affect different models and different failure points." + "reason": "Different multimodal problems on different models: Llava Next segfault vs Qwen3OmniMoe video-input ValueError." }, { "left": "issue:44479", "right": "issue:45325", - "accept": true, - "reason": "These look like the same Qwen VL RoPE/temporal-position regression in v5.3.0, with one report focusing on video inputs and the other on still-image position_ids." + "accept": false, + "reason": "Related Qwen VL/video area, but 44479 is a broad video regression while 45325 is a specific still-image temporal-position scaling bug." }, { "left": "issue:43632", "right": "issue:46821", "accept": false, - "reason": "Different regressions: HF initialization flag breakage vs TimesFM window_size attribute error." + "reason": "Transformers v5 init-flag breakage is unrelated to the TimesFM window_size AttributeError." }, { "left": "issue:36032", "right": "issue:41720", "accept": false, - "reason": "Tokenizer loading conflict and Qwen3 auto device-map CUDA assert are unrelated." + "reason": "Tokenizer-load conflict and Qwen3 auto device-map cuda assert are unrelated." }, { "left": "issue:45216", "right": "issue:45357", "accept": true, - "reason": "Both report the same Qwen3.5 save_pretrained / checkpoint serialization regression, with 45357 sounding like a later manifestation of the same bug." + "reason": "Same underlying Qwen3.5 save_pretrained regression: both report incorrect checkpoint/visual-encoder key serialization, with 45357 reading like a narrower follow-up." + }, + { + "left": "issue:43531", + "right": "issue:44877", + "accept": false, + "reason": "Qwen3-MoE sliding_window handling and granite_speech strict-config loading are unrelated." }, { "left": "issue:44991", "right": "issue:46315", "accept": false, - "reason": "Tokenizer loading failure and tokenizer conversion slowdown are different issues." + "reason": "Both involve slow tokenizers, but one is a load failure and the other is a performance regression in convert_tokens_to_ids." }, { - "left": "issue:43531", - "right": "issue:44877", + "left": "issue:43425", + "right": "issue:44403", "accept": false, - "reason": "Different model/config problems with no shared concrete bug." + "reason": "Torch version incompatibility is unrelated to the 'unnecessary noise' loading complaint." }, { "left": "issue:43296", "right": "issue:43531", "accept": false, - "reason": "vLLM PaddleOCR-VL load failure and Qwen3-MoE sliding_window behavior are unrelated." + "reason": "PaddleOCR-VL loading failure and Qwen3-MoE sliding_window behavior are unrelated." }, { "left": "issue:44220", "right": "issue:44683", "accept": false, - "reason": "Audio feature extraction and compiled flex_attention failures are unrelated." + "reason": "Audio fbank extraction and compiled flex_attention are different subsystems and failure modes." }, { - "left": "issue:43611", - "right": "issue:44534", + "left": "issue:43716", + "right": "issue:46899", "accept": false, - "reason": "Both are v5 regressions, but one is base_model_prefix loading and the other is junk-filled non-persistent buffers; not the same underlying defect." + "reason": "Both mention dtype issues in multimodal paths, but they are different models and different bugs." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet: there are no PRs, and the claim that only two issue pairs look duplicate is conservative. The accepted soft pairs are supported by closely related titles, and the rejected pairs avoid overclaiming sameness. The best-issue and canonical-issue rationales are subjective but still consistent with the available evidence." + "feedback": "The summary is well grounded in the packet and the one accepted soft edge is supported by the issue titles: 45216 and 45357 both concern the Qwen3.5 save_pretrained checkpoint/visual-encoder key serialization regression. The other soft-edge rejections are phrased conservatively and do not overclaim duplicate status from similarity alone." }, "evaluator_used": true, "retried": false, @@ -3816,8 +3741,9 @@ "issue:43525", "issue:43531", "issue:43582", + "issue:43611", "issue:43632", - "issue:43716", + "issue:43646", "issue:43746", "issue:43792", "issue:43856", @@ -3847,7 +3773,7 @@ "issue:46909" ], "soft_pairs": [ - "issue:43716|issue:46899", + "issue:43611|issue:44534", "issue:40444|issue:45325", "issue:45812|issue:46491", "issue:44403|issue:44991", @@ -3855,8 +3781,8 @@ "issue:43632|issue:44756", "issue:42796|issue:43746", "issue:43792|issue:44589", - "issue:33666|issue:40444", "issue:43404|issue:46899", + "issue:33666|issue:40444", "issue:40990|issue:44811", "issue:44291|issue:44589", "issue:42796|issue:46710", @@ -3865,220 +3791,227 @@ "issue:44079|issue:45663", "issue:41553|issue:43582", "issue:43482|issue:45538", - "issue:43421|issue:44568", "issue:43632|issue:44534", + "issue:43421|issue:44568", "issue:43531|issue:44291", "issue:42175|issue:45542", - "issue:44462|issue:45812", "issue:41720|issue:45081", + "issue:44462|issue:45812", "issue:43317|issue:43856", "issue:44079|issue:44534", "issue:44509|issue:46909", "issue:43296|issue:46821", - "issue:43377|issue:45491" + "issue:43377|issue:45491", + "issue:43611|issue:43646" ], - "prepared_review_unit_hash": "99911d182514ad6d720d6d268e78aaf4eed313eca8c95de75c7da0bc5e36b3bc", + "prepared_review_unit_hash": "3dc730cc5a3f9c96fedc0bcdb8fdd7a026852b27bca4a5b251ccd83475a87e69", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 29, - "serialized_chars": 35904, - "estimated_input_tokens": 8976, - "estimated_eval_tokens": 18208 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36775, + "estimated_input_tokens": 9194, + "estimated_eval_tokens": 18644 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is very heterogeneous: most items are separate model-specific regressions, tokenizer quirks, packaging issues, or docs bugs. I do not see any true duplicate PR/issue pair among the soft edges; at best they share broad themes like v5 regressions or tokenizer behavior.", - "confidence": 0.69, - "canonical_issue_reason": "Issue 44534 is the broadest core regression in the set and the best representative of the cluster\u2019s overall v5-era breakage theme. It is still not a duplicate of the others, just the most canonical umbrella-style report.", + "summary": "This cluster is mostly a grab-bag of unrelated bugs. The only loose common thread is Transformers v5 loading/init/tokenizer regressions, but the soft pairs still split across different concrete code paths and model families, so none look like true duplicates.", + "confidence": 0.47, + "canonical_issue_reason": "issue:43611 is the broadest, most central v5 regression here: it targets model loading behavior itself, which is the closest umbrella for several other init/load-related reports in the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "44534 has the widest scope and is closest to an underlying platform-level regression, whereas most other issues are narrower model- or subsystem-specific reports.", + "best_issue_reason": "issue:43611 is the best single representative because it captures the broad model-loading regression theme without being as narrow as the model-specific or symptom-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43716", - "right": "issue:46899", + "left": "issue:43611", + "right": "issue:44534", "accept": false, - "reason": "Both are multimodal dtype problems, but they affect different models and code paths; not the same bug." + "reason": "Both are v5 regressions, but one is model loading/base_model_prefix and the other is non-persistent buffer initialization; different code paths." }, { "left": "issue:40444", "right": "issue:45325", "accept": false, - "reason": "Both are Qwen2.5-VL, but one is a finetuning/data-pipeline failure and the other is a temporal position-id scaling bug." + "reason": "Both are Qwen2.5-VL, but one fails finetuning with multiple images per prompt and the other mis-scales still-image rope positions." }, { "left": "issue:45812", "right": "issue:46491", "accept": false, - "reason": "Both are tokenizer regressions, but Granite token IDs and CodeLlama leading-space round-tripping are different behaviors." + "reason": "Tokenizer regressions, but different models and symptoms: Granite wrong token IDs vs CodeLlama leading-space round-trip." }, { "left": "issue:44403", "right": "issue:44991", "accept": false, - "reason": "Loading-related, but one is log noise and the other is a concrete tokenizer loading failure for a specific model." + "reason": "One is noisy load logging; the other is a tokenizer loading failure for EMBEDDIA/est-roberta." }, { "left": "issue:43381", "right": "issue:45923", "accept": false, - "reason": "Both mention gradient checkpointing, but one is an eval-mode restriction and the other is a missing config flag on a trust_remote_code variant." + "reason": "Both mention gradient checkpointing, but one forbids eval-mode use and the other is a missing support flag on a trust_remote_code variant." }, { "left": "issue:43632", "right": "issue:44756", "accept": false, - "reason": "Completely different problems: an internal initialization flag regression vs mmap-related OOM avoidance." + "reason": "One is the `_is_hf_initialized` v5 init regression; the other is an mmap OOM workaround on Strix Halo." }, { "left": "issue:42796", "right": "issue:43746", "accept": false, - "reason": "Tokenizer BPE marker handling is unrelated to PEFT/local checkpoint loading." + "reason": "Tokenizer BPE marker handling vs PEFT local-checkpoint loading; unrelated concrete bugs." }, { "left": "issue:43792", "right": "issue:44589", "accept": false, - "reason": "No shared code path: Whisper runtime failure versus a float8 storage deserialization/type error." + "reason": "Whisper runtime failure vs missing Float8 storage type; unrelated." }, { - "left": "issue:33666", - "right": "issue:40444", + "left": "issue:43404", + "right": "issue:46899", "accept": false, - "reason": "Both are Qwen vision-training issues, but they describe different failures and likely different fixes." + "reason": "Mistral3 lm_head weight tying bug vs Gemma4 4-bit quantization dtype cast issue." }, { - "left": "issue:43404", - "right": "issue:46899", + "left": "issue:33666", + "right": "issue:40444", "accept": false, - "reason": "Different multimodal bugs: weight tying in Mistral3 vs unsafe dtype casting in Gemma4 embedders." + "reason": "Both involve Qwen VL training, but multi-GPU training and multi-image iterable-dataset failures are different bugs." }, { "left": "issue:40990", "right": "issue:44811", "accept": false, - "reason": "Unrelated: perplexity evaluation issue vs processor.batch_decode special-token handling." + "reason": "High perplexity on gpt-oss vs Whisper batch_decode skip_special_tokens behavior." }, { "left": "issue:44291", "right": "issue:44589", "accept": false, - "reason": "Both are v5 compatibility failures, but one is init_empty_weights argument handling and the other is float8 storage lookup." + "reason": "`init_empty_weights`/_is_hf_initialized arg issue vs Float8 storage lookup error." }, { "left": "issue:42796", "right": "issue:46710", "accept": false, - "reason": "Tokenizer marker handling and accuracy regression are different bug classes with no shared concrete code path." + "reason": "Tokenizer BPE-marker bug vs DeepSeek accuracy regression after upgrade." }, { "left": "issue:43482", "right": "issue:43525", "accept": false, - "reason": "Both are model-loading problems, but they affect different models and different missing/unsupported config fields." + "reason": "GGUF loading failure vs missing `pad_token_id` on Llama4Config." }, { "left": "issue:35707", "right": "issue:40444", "accept": false, - "reason": "Progressive 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 vs multi-image iterable-dataset finetuning." }, { "left": "issue:44079", "right": "issue:45663", "accept": false, - "reason": "ModelOutput key assignment and Gemma4 FSDP2 shared_kv_states are unrelated." + "reason": "ModelOutput key assignment bug vs Gemma4 FSDP2 `shared_kv_states` KeyError." }, { "left": "issue:41553", "right": "issue:43582", "accept": false, - "reason": "Bad tokenizer error message and AppleSilicon TypeError are unrelated." + "reason": "Bad AutoTokenizer error message for Voxtral vs Apple Silicon caching_allocator_warmup TypeError." }, { "left": "issue:43482", "right": "issue:45538", "accept": false, - "reason": "GGUF loading failure and CLIPTokenizer max length are different tokenizer/model-loading issues." + "reason": "GGUF/v5 loading failure vs CLIPTokenizer max_length bug." }, { - "left": "issue:43421", - "right": "issue:44568", + "left": "issue:43632", + "right": "issue:44534", "accept": false, - "reason": "Both involve special-token behavior, but one is runtime post-processor updates and the other is add_special_tokens not adding BOS/EOS for a specific tokenizer." + "reason": "`_is_hf_initialized` flag regression vs non-persistent buffer junk fill." }, { - "left": "issue:43632", - "right": "issue:44534", + "left": "issue:43421", + "right": "issue:44568", "accept": false, - "reason": "Different regression types: initialization flag breakage versus non-persistent buffer corruption." + "reason": "Runtime post-processor updates for special tokens vs mdeberta BOS/EOS not added." }, { "left": "issue:43531", "right": "issue:44291", "accept": false, - "reason": "Qwen3-MoE sliding-window behavior is unrelated to init_empty_weights/_is_hf_initialized." + "reason": "Qwen3-MoE sliding-window bug vs `init_empty_weights` constructor-arg regression." }, { "left": "issue:42175", "right": "issue:45542", "accept": false, - "reason": "Both touch TensorFlow packaging, but one is missing TF in the torch extra and the other is tensorboard-installed-without-TF backend handling." + "reason": "Both mention TensorFlow packaging, but one is missing tensorflow in `.[torch]` and the other is tensorboard-only backend detection; not the same failure." }, { - "left": "issue:44462", - "right": "issue:45812", + "left": "issue:41720", + "right": "issue:45081", "accept": false, - "reason": "Both are tokenizer regressions, but one is ignoring tokenizer.json and the other is wrong token IDs for Granite models." + "reason": "Qwen3 auto device mapping cudaErrorAssert vs Mistral regex patch crash while loading tokenizer." }, { - "left": "issue:41720", - "right": "issue:45081", + "left": "issue:44462", + "right": "issue:45812", "accept": false, - "reason": "Auto device mapping cuda assert and Mistral regex patch crash are unrelated." + "reason": "AutoTokenizer ignoring tokenizer.json vs Granite token-ID regression." }, { "left": "issue:43317", "right": "issue:43856", "accept": false, - "reason": "Device-map/offload load failure and Qwen3 MoE memory inefficiency are different problems." + "reason": "Dequantized model offload loading failure vs Qwen3 MoE training memory inefficiency." }, { "left": "issue:44079", "right": "issue:44534", "accept": false, - "reason": "ModelOutput key handling is unrelated to junk-filled non-persistent buffers." + "reason": "ModelOutput key assignment vs non-persistent buffer junk fill." }, { "left": "issue:44509", "right": "issue:46909", "accept": false, - "reason": "Both are docs issues, but they are different documents and different defects." + "reason": "Docs about removed pipeline tasks vs a broken DIA documentation example." }, { "left": "issue:43296", "right": "issue:46821", "accept": false, - "reason": "PaddleOCR-VL/vLLM load failure and TimesFM window_size AttributeError are unrelated." + "reason": "PaddleOCR-VL load failure in vLLM vs TimesFM `window_size` AttributeError." }, { "left": "issue:43377", "right": "issue:45491", "accept": false, - "reason": "Both mention batching/padding, but they are different models and different failure modes (encoder mismatch vs NaN embeddings)." + "reason": "MIMI padding-mask batch discrepancy vs Gemma3 NaNs with sliding-window mixed lengths." + }, + { + "left": "issue:43611", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 loading/init regressions, but one is `base_model_prefix` loading and the other is custom model initialization; related theme, different bugs." } ] }, @@ -4113,14 +4046,13 @@ "issue:42915", "issue:43262", "issue:43296", + "issue:43377", "issue:43454", "issue:43482", "issue:43493", "issue:43550", "issue:43606", - "issue:43611", "issue:43644", - "issue:43646", "issue:43701", "issue:43746", "issue:43818", @@ -4136,6 +4068,7 @@ "issue:44466", "issue:44479", "issue:44488", + "issue:44610", "issue:44829", "issue:44877", "issue:45072", @@ -4151,7 +4084,6 @@ "issue:46821" ], "soft_pairs": [ - "issue:43611|issue:43646", "issue:43818|issue:46402", "issue:44829|issue:45479", "issue:43644|issue:44079", @@ -4160,8 +4092,8 @@ "issue:44265|issue:45072", "issue:43493|issue:43994", "issue:40444|issue:45381", - "issue:45325|issue:45834", "issue:45325|issue:45381", + "issue:45325|issue:45834", "issue:43454|issue:44466", "issue:42915|issue:43482", "issue:44291|issue:44877", @@ -4173,233 +4105,231 @@ "issue:43262|issue:43746", "issue:44479|issue:45823", "issue:43881|issue:44488", - "issue:35707|issue:41720", "issue:43828|issue:46821", + "issue:35707|issue:41720", "issue:34634|issue:35707", "issue:39692|issue:40444", "issue:43454|issue:43550", "issue:43606|issue:43819", "issue:44186|issue:44265", - "issue:41720|issue:45823" + "issue:41720|issue:45823", + "issue:43377|issue:44610" ], - "prepared_review_unit_hash": "a0fb84dccc83f43d5e18157864a399b6692f95b229f3fad85e0e9deb2c2e7562", + "prepared_review_unit_hash": "bc0e84454553d1a498491c1ff9724f9e3df9d80d641825e75e8add6efa59b113", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36225, - "estimated_input_tokens": 9057, - "estimated_eval_tokens": 18370 + "serialized_chars": 36278, + "estimated_input_tokens": 9070, + "estimated_eval_tokens": 18396 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a collection of unrelated Transformers issues across different models/subsystems. Only the two SigLIP2 reports look like a plausible duplicate pair.", - "confidence": 0.74, - "canonical_issue_reason": "issue:43493 is the better canonical target for the SigLIP2 duplicate pair because it states the underlying HF-vs-JAX implementation discrepancy, which likely explains the downstream bad outputs reported in issue:43994.", + "summary": "The items are heterogeneous and do not form a single duplicate cluster. All listed soft pairs look like false positives: they share broad model/subsystem themes but not the same concrete bug or change path.", + "confidence": 0.92, + "canonical_issue_reason": "No clear canonical issue: the set spans many unrelated bugs (model loading, tokenization, quantization, device mapping, generation, and export), so selecting one as the duplicate anchor would be misleading.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 is the strongest overall standalone issue in the cluster: it has the broadest scope, the most discussion/inbound references, and a concrete reproducible failure mode around quantized offloading.", + "best_issue_reason": "No single issue is globally best as a representative duplicate target; the soft pairs are mostly broad-topic similarities rather than the same underlying defect.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43611", - "right": "issue:43646", - "accept": false, - "reason": "Different failure modes: loading with `base_model_prefix` vs. custom model initialization under Transformers 5." - }, { "left": "issue:43818", "right": "issue:46402", "accept": false, - "reason": "Both involve model internals, but one is a Video-LLaVA tower/temporal-attention bug and the other is Sam3 text-encoder weight loading." + "reason": "Different bugs in different models: one is a Video-LLaVA temporal-attention/video-tower issue, the other is Sam3Model text-encoder weight loading." }, { "left": "issue:44829", "right": "issue:45479", "accept": false, - "reason": "Both describe degenerate training, but one is tied to `flash_attention_3` and the other is a `num_labels=1` single-label classification loss issue." + "reason": "Both mention bad training/loss behavior, but the causes differ: flash_attention_3 regression vs. single-label classification with num_labels=1." }, { "left": "issue:43644", "right": "issue:44079", "accept": false, - "reason": "Non-persistent buffer serialization junk and `ModelOutput` key assignment are different code paths." + "reason": "Unrelated code paths: non-persistent buffer junk initialization vs. ModelOutput key assignment when a key was previously None." }, { "left": "issue:42175", "right": "issue:43296", "accept": false, - "reason": "Packaging/backend dependency issue versus a model-loading failure for PaddleOCR-VL." + "reason": "Package/backend installation problem vs. a vLLM model-load failure; no shared concrete bug." }, { "left": "issue:43818", "right": "issue:45357", "accept": false, - "reason": "Video tower architecture bug vs. incorrect `save_pretrained` visual-encoder key handling." + "reason": "One is a model architecture/loading defect, the other is save_pretrained writing incorrect visual encoder keys." }, { "left": "issue:44265", "right": "issue:45072", "accept": false, - "reason": "`torch.export`/`torch_compilable_check` failure is unrelated to a bfloat16 dtype mismatch during inference." + "reason": "torch.export/checking failure and bf16 dtype mismatch are different failure modes, even though both involve transformers internals." }, { "left": "issue:43493", "right": "issue:43994", - "accept": true, - "reason": "Same SigLIP2 output-correctness bug family: one reports HF-vs-JAX discrepancy, the other reports nonsensical outputs via AutoModel/pipeline, both pointing to the same model implementation path." + "accept": false, + "reason": "Same model family, but one is HF-vs-JAX implementation divergence while the other is bad AutoModel/pipeline output; not clearly the same bug." }, { "left": "issue:40444", "right": "issue:45381", "accept": false, - "reason": "Multiple-images finetuning with IterableDataset is a different bug from Qwen2.5-VL video position-id handling." + "reason": "Different Qwen2.5-VL scenarios: finetuning on multi-image prompts vs. video-input position-id regression." }, { "left": "issue:45325", - "right": "issue:45834", + "right": "issue:45381", "accept": false, - "reason": "Qwen2.5-VL rope/position-id scaling bug and Kosmos2.5 OCR index error are unrelated." + "reason": "Both involve Qwen2.5-VL position handling, but one is still-image temporal scaling and the other is video vision_position_ids; too different to treat as duplicates." }, { "left": "issue:45325", - "right": "issue:45381", + "right": "issue:45834", "accept": false, - "reason": "Both are Qwen2.5-VL multimodal regressions, but one is still-image temporal position scaling and the other is video `vision_position_ids`; not the same concrete bug." + "reason": "One is a rope-index scaling bug, the other is a long-OCR index error; shared model family is not enough." }, { "left": "issue:43454", "right": "issue:44466", "accept": false, - "reason": "One is a tied-weight generation bug in AyaVision, the other is device-dependent serialization inconsistency for `lm_head.weight`." + "reason": "Both mention lm_head/tied weights, but one is failure to tie weights causing bad generation and the other is device-dependent serialization inconsistency." }, { "left": "issue:42915", "right": "issue:43482", "accept": false, - "reason": "FineGrainedFP8Config failure and GGUF loading failure are separate model/version issues." + "reason": "FP8 configuration failure vs. GGUF loading failure in v5 are unrelated." }, { "left": "issue:44291", "right": "issue:44877", "accept": false, - "reason": "Unexpected `_is_hf_initialized` argument and strict config rejection are different loading regressions." + "reason": "init_empty_weights argument mismatch is a different problem from strict config rejection of granite_speech." }, { "left": "issue:43701", "right": "issue:45325", "accept": false, - "reason": "Checkpoint key mismatch is unrelated to Qwen2.5-VL rope-index scaling." + "reason": "Checkpoint key mismatch and Qwen2.5-VL position-id scaling are unrelated bugs." }, { "left": "issue:43873", "right": "issue:45701", "accept": false, - "reason": "Quantized offloading behavior and version-dependent tokenization changes do not share the same code path." + "reason": "Quantization/offload behavior and tokenization version changes are distinct issues." }, { "left": "issue:42796", "right": "issue:45381", "accept": false, - "reason": "Tokenizer BPE-marker handling is not the same as Qwen2.5-VL video position-id regression." + "reason": "Tokenizer BPE-marker handling and Qwen2.5-VL video position IDs are different defects." }, { "left": "issue:43606", "right": "issue:44265", "accept": false, - "reason": "CPU offload device mismatch and `torch.export` failure on `torch_compilable_check` are distinct problems." + "reason": "CPU-offload device mismatch is not the same as export failing on torch_compilable_check." }, { "left": "issue:40990", "right": "issue:45584", "accept": false, - "reason": "Perplexity regression on GPT-OSS and Whisper empty-transcription failure are unrelated model behaviors." + "reason": "High perplexity on WikiText-2 and Whisper empty-transcription failure are unrelated symptoms and code paths." }, { "left": "issue:43262", "right": "issue:43746", "accept": false, - "reason": "Audio chat-template sampling-rate defaulting and PEFT checkpoint loading are different subsystems." + "reason": "Audio chat-template sampling-rate defaulting and GraniteSpeech PEFT checkpoint loading are unrelated." }, { "left": "issue:44479", "right": "issue:45823", "accept": false, - "reason": "Qwen video-input regression and Gemma4 PLE device mismatch are unrelated model-specific bugs." + "reason": "Both are inference regressions, but for different models and different root causes: Qwen video regression vs. Gemma4 PLE device mismatch." }, { "left": "issue:43881", "right": "issue:44488", "accept": false, - "reason": "GLM-4V loading failure and sleng-bert loading failure are separate model issues." + "reason": "Different models and different load failures; not a duplicate bug." }, { - "left": "issue:35707", - "right": "issue:41720", + "left": "issue:43828", + "right": "issue:46821", "accept": false, - "reason": "Progressive generation with `inputs_embeds/past_key_values` and Qwen3 auto device-mapping CUDA asserts are different failures." + "reason": "Autocast dtype mismatch and TimesFM window_size AttributeError are unrelated." }, { - "left": "issue:43828", - "right": "issue:46821", + "left": "issue:35707", + "right": "issue:41720", "accept": false, - "reason": "Autocast dtype mismatch and TimesFM `window_size` AttributeError are unrelated." + "reason": "Progressive generation with inputs_embeds/past_key_values is a different code path from auto device mapping causing cudaErrorAssert." }, { "left": "issue:34634", "right": "issue:35707", "accept": false, - "reason": "BarkProcessor voice preset behavior and generation with `inputs_embeds/past_key_values` are unrelated." + "reason": "BarkProcessor voice_preset and progressive generation with embeddings are unrelated." }, { "left": "issue:39692", "right": "issue:40444", "accept": false, - "reason": "SigLIP2 doc-example mistakes and Qwen2.5-VL multi-image finetuning failure are different issues." + "reason": "SigLIP2 doc-example problems and Qwen2.5-VL multi-image finetuning failures do not share the same concrete bug." }, { "left": "issue:43454", "right": "issue:43550", "accept": false, - "reason": "AyaVision tied-weight loading/generation bug and Bamba SDPA compile bug do not match." + "reason": "Tied-weight serialization/generation bug vs. Bamba-9B-v2 torch.compile+SDPA failure; different root causes." }, { "left": "issue:43606", "right": "issue:43819", "accept": false, - "reason": "CPU offload device mismatch is unrelated to DAC `from_latents` vs forward-pass mismatch." + "reason": "CPU offload device mismatch and DAC forward/from_latents mismatch are unrelated." }, { "left": "issue:44186", "right": "issue:44265", "accept": false, - "reason": "Tokenizer crashes on NER/padding and `torch.export` failures are different code paths." + "reason": "Tokenizer crashes on NER/padding are not the same as torch.export failures." }, { "left": "issue:41720", "right": "issue:45823", "accept": false, - "reason": "Qwen3 auto device-mapping CUDA assert and Gemma4 PLE forward mismatch are unrelated." + "reason": "Different models and different device-related failures; no shared concrete bug." + }, + { + "left": "issue:43377", + "right": "issue:44610", + "accept": false, + "reason": "MIMI padding-mask output mismatch and OmDet processor/model input-size mismatch are distinct issues." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is well grounded in the packet. The choice of issue:43873 as the strongest standalone issue is supported by its higher discussion and inbound-reference counts, and the SigLIP2 pairing of issue:43493 with issue:43994 is a conservative, evidence-based duplicate call. The soft-edge verdicts are appropriately cautious and do not overstate similarity." - }, - "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 }, @@ -4445,7 +4375,7 @@ "issue:44466", "issue:44534", "issue:44568", - "issue:44610", + "issue:44756", "issue:44877", "issue:44898", "issue:44977", @@ -4466,7 +4396,6 @@ "issue:46821" ], "soft_pairs": [ - "issue:43377|issue:44610", "issue:46305|issue:46315", "issue:43257|issue:45923", "issue:44568|issue:45920", @@ -4490,228 +4419,229 @@ "issue:36331|issue:42915", "issue:44466|issue:45357", "issue:43450|issue:45701", - "issue:43065|issue:46402", "issue:44877|issue:44991", + "issue:43065|issue:46402", "issue:40990|issue:45701", "issue:44186|issue:45072", "issue:26413|issue:28530", "issue:41720|issue:42617", - "issue:43550|issue:43720" + "issue:43550|issue:43720", + "issue:43873|issue:44756" ], - "prepared_review_unit_hash": "d909e08fba73060cf519b69c3cc4f531d24287d4bf57e3ca343cc453cfcd59e0", + "prepared_review_unit_hash": "d4887696d7b7ea2d8a65df39955eb1a8da7b1ca3f072ecbc7715ba4c736e86ee", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37008, - "estimated_input_tokens": 9252, - "estimated_eval_tokens": 18760 + "serialized_chars": 36975, + "estimated_input_tokens": 9244, + "estimated_eval_tokens": 18744 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a heterogeneous mix of Transformers bug reports across tokenization, model loading/offload, quantization, config, and training behavior. The pairwise similarities are mostly superficial, so none of the soft pairs look like true duplicates.", - "confidence": 0.93, - "canonical_issue_reason": "No single canonical issue stands out: the cluster spans unrelated bugs and subsystems, and the apparent overlaps are mostly broad theme similarity rather than the same underlying defect.", + "summary": "This cluster is mostly heterogeneous: several unrelated bug reports were surfaced by soft text similarity, but none of the candidate pairs look like the same underlying issue. The closest themes are tokenizer regressions and device_map/offloading/quantization problems, yet each pair differs in concrete failure mode or affected path.", + "confidence": 0.78, + "canonical_issue_reason": "issue:43873 is the broadest and most central bug in the set: an open, highly discussed offloading/quantization regression with multiple nearby similarity hits, making it the best representative issue for this cluster.", "canonical_pr_reason": null, - "best_issue_reason": "No global best issue is suitable as a representative duplicate target because the items are too diverse to collapse into one concrete bug report.", + "best_issue_reason": "issue:43873 has the strongest cluster-level relevance and the widest problem scope among the issues shown, so it is the best single issue to anchor triage even though it is not a duplicate of the others.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43377", - "right": "issue:44610", - "accept": false, - "reason": "Both involve batched input handling, but one is a MIMI padding-mask discrepancy and the other is an OmDet image size mismatch; different code paths and fixes." - }, { "left": "issue:46305", "right": "issue:46315", "accept": false, - "reason": "Sliding-window off-by-one in streaming masks and slow `convert_tokens_to_ids` are unrelated bugs." + "reason": "Both are performance bugs, but one is a MoonshineStreaming masking off-by-one and the other is slow `convert_tokens_to_ids`; different code paths and symptoms." }, { "left": "issue:43257", "right": "issue:45923", "accept": false, - "reason": "Qwen3 MoE weight conversion/loading and a missing gradient-checkpointing flag are different problems." + "reason": "Qwen3 MoE weight conversion under accelerate+deepspeed is unrelated to a missing `supports_gradient_checkpointing` flag on a trust_remote_code variant." }, { "left": "issue:44568", "right": "issue:45920", "accept": false, - "reason": "Both are tokenizer regressions, but one is missing BOS/EOS insertion for a specific tokenizer while the other is wrong token IDs across many models; not the same bug." + "reason": "`add_special_tokens` not adding BOS/EOS for mdeberta-v3 is a tokenizer-special-case bug, while the other is wrong token IDs across several model families after a v5 regression." }, { "left": "issue:42796", "right": "issue:46209", "accept": false, - "reason": "Tokenizer BPE marker handling and FP8 inference failure on B200 are unrelated." + "reason": "Tokenizer BPE marker handling and FP8 inference on B200 are unrelated failures in different subsystems." }, { "left": "issue:43454", "right": "issue:45072", "accept": false, - "reason": "AyaVision weight tying and dtype mismatches in inference are separate model/runtime issues." + "reason": "Incorrect batched video processor shape and dtype mismatches in bfloat16 inference are different bugs with different affected components." }, { "left": "issue:45823", "right": "issue:46821", "accept": false, - "reason": "Device-map forward mismatch for Gemma4 PLE and a TimesFM `window_size` AttributeError are different failures." + "reason": "Gemma4 PLE device mismatch during forward is not the same as TimesFM `window_size` raising `AttributeError`; different models and error classes." }, { "left": "issue:43377", "right": "issue:44898", "accept": false, - "reason": "Both are batching-related, but one is missing padding-mask support and the other is a Perceiver resolution/interpolation bug; not the same defect." + "reason": "MIMI encoder batching/padding-mask behavior is unrelated to Perceiver image classification failing despite `interpolate_pos_encoding=True`." }, { "left": "issue:43454", "right": "issue:43720", "accept": false, - "reason": "AyaVision tie-weights failure and BitNet packed-weight unpacking during accelerate loading are unrelated." + "reason": "Video processor batching shape issues do not match BitNet packed-weights unpacking during accelerate loading." }, { "left": "issue:40990", "right": "issue:42947", "accept": false, - "reason": "High perplexity on a specific model and ineffective gradient checkpointing with LoRA are different training/inference issues." + "reason": "High perplexity on gpt-oss-20b is a model quality/regression report, while the other is a gradient checkpointing effectiveness issue with PEFT LoRA." }, { "left": "issue:42617", "right": "issue:42915", "accept": false, - "reason": "3d_parallel.py execution failure and Qwen3Moe fine-grained FP8 failure are unrelated." + "reason": "Failure to run `3d_parallel.py` is unrelated to Qwen3MoE failing with FineGrainedFP8Config." }, { "left": "issue:43749", "right": "issue:44534", "accept": false, - "reason": "Both touch loading/state handling, but FSDP CPU RAM efficient loading and non-persistent buffer junk are different concrete bugs." + "reason": "Broken FSDP CPU RAM efficient loading and non-persistent buffers filled with junk are different regression classes." }, { "left": "issue:43317", "right": "issue:43873", "accept": false, - "reason": "Both mention offload/quantization, but one is dequantized-model loading with device_map auto and the other is a broader quantization offloading issue; too different to merge." + "reason": "Both involve loading/offloading, but one is a dequantized model load failure with gpu+cpu offload and the other is a broader quantization offloading issue; not the same concrete bug." }, { "left": "issue:43572", "right": "issue:45759", "accept": false, - "reason": "Missing `pad_token_idx` in config and failure to unwrap `text_config` are different config-loading bugs." + "reason": "Missing `pad_token_idx` in StableLmConfig is a config regression, while the other is `from_config` not unwrapping `text_config` for composite Qwen multimodal configs." }, { "left": "issue:43526", "right": "issue:43883", "accept": false, - "reason": "BEiT label reduction and missing `all_tied_weights_keys` are unrelated." + "reason": "`reduce_labels` returning one label in BeitImageProcessorFast is unrelated to a missing `all_tied_weights_keys` attribute in MolmoForCausalLM." }, { "left": "issue:43322", "right": "issue:46050", "accept": false, - "reason": "LLaVA Next segmentation fault and sinq save/load breakage are unrelated loading bugs." + "reason": "Segmentation fault loading Llava Next models and broken sinq quantized save/load behavior are not the same code-path problem." }, { "left": "issue:41720", "right": "issue:43322", "accept": false, - "reason": "Qwen3 auto device-mapping CUDA assert and LLaVA Next segfault are different model/runtime failures." + "reason": "Qwen3 auto device mapping cudaErrorAssert and Llava Next segmentation fault are both load-time crashes, but they affect different models and failure mechanisms." }, { "left": "issue:43881", "right": "issue:44991", "accept": false, - "reason": "glm-4v-9b loading failure and tokenizer loading regression are unrelated." + "reason": "glm-4v-9b loading failure and tokenizer loading failure for EMBEDDIA/est-roberta are unrelated model-specific regressions." }, { "left": "issue:36331", "right": "issue:42617", "accept": false, - "reason": "`compute_loss` signature breakage and `3d_parallel.py` execution failure are unrelated trainer/script issues." + "reason": "Unexpected `num_items_in_batch` in `compute_loss` is a trainer API compatibility issue, not the same as failing to run `3d_parallel.py`." }, { "left": "issue:44387", "right": "issue:45704", "accept": false, - "reason": "Both are memory-related, but one is int4 quantization memory growth and the other is an apex RMSNorm leak; different causes and fixes." + "reason": "Int4 quantization OOM in Transformers 5.x and T5's apex FusedRMSNorm memory leak are different memory issues with different root causes." }, { "left": "issue:41720", "right": "issue:44977", "accept": false, - "reason": "Qwen3 auto device-map CUDA assert and Qwen3.5 flash-attention generation issues are not the same bug." + "reason": "Qwen3 auto device mapping crashes on A800, whereas Qwen3.5 flash-attention generation is a different generation-path regression." }, { "left": "issue:36331", "right": "issue:42915", "accept": false, - "reason": "`compute_loss` keyword mismatch and Qwen3Moe FP8 failure are unrelated." + "reason": "Trainer `compute_loss` signature incompatibility is unrelated to Qwen3MoE FineGrainedFP8Config failure." }, { "left": "issue:44466", "right": "issue:45357", "accept": false, - "reason": "Both are serialization regressions, but one concerns tied `lm_head.weight` behavior and the other incorrect visual encoder keys; different artifacts." + "reason": "Both mention Qwen3.5 and weights, but one is inconsistent lm_head serialization by device and the other is incorrect visual encoder keys on save; different bugs." }, { "left": "issue:43450", "right": "issue:45701", "accept": false, - "reason": "Batched video shape and version-dependent tokenization are unrelated." + "reason": "Incorrect batched video processor shape and version-dependent tokenization changes are different subsystems." }, { - "left": "issue:43065", - "right": "issue:46402", + "left": "issue:44877", + "right": "issue:44991", "accept": false, - "reason": "Sam3PixelDecoder dummy Conv2d and Sam3 text encoder weights not loading are different subcomponent issues." + "reason": "Strict config blocking granite_speech loading is not the same as tokenizer loading failure for EMBEDDIA/est-roberta." }, { - "left": "issue:44877", - "right": "issue:44991", + "left": "issue:43065", + "right": "issue:46402", "accept": false, - "reason": "Strict config blocking `granite_speech` and tokenizer loading regression are unrelated." + "reason": "Dummy `nn.Conv2d` in Sam3PixelDecoder and Sam3Model text encoder weights not loading are separate Sam3 issues with different affected modules." }, { "left": "issue:40990", "right": "issue:45701", "accept": false, - "reason": "Extremely high perplexity and tokenization changes across versions are not the same issue." + "reason": "Perplexity on gpt-oss-20b and tokenization changing across versions are not the same bug." }, { "left": "issue:44186", "right": "issue:45072", "accept": false, - "reason": "LayoutLMv2 tokenizer crashes and dtype mismatches in inference are unrelated." + "reason": "LayoutLMv2Tokenizer crashes on NER/batched padding and bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel are unrelated." }, { "left": "issue:26413", "right": "issue:28530", "accept": false, - "reason": "Checkpoint resume epoch-iterator failure and early-stopping metric lookup are different trainer behaviors." + "reason": "Checkpoint resume epoch iterator failure and early stopping metric lookup failure are distinct training-loop issues." }, { "left": "issue:41720", "right": "issue:42617", "accept": false, - "reason": "CUDA assert on Qwen3 device mapping and running 3d_parallel.py are unrelated." + "reason": "Qwen3 auto device map cudaErrorAssert and inability to run `3d_parallel.py` are different runtime failures." }, { "left": "issue:43550", "right": "issue:43720", "accept": false, - "reason": "Bamba torch.compile SDPA failure and BitNet packed-weight unpacking are different code paths." + "reason": "Bamba-9B-v2 torch.compile SDPA failure and BitNet packed-weight unpacking during accelerate loading are different model/backend bugs." + }, + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "Quantization offloading problems and disabling mmap on Strix Halo to avoid OOM are related only at a high level; they are not the same concrete issue." } ] }, @@ -4736,6 +4666,7 @@ "cluster_id": "cluster-10000-500", "nodes": [ "issue:28530", + "issue:30064", "issue:33666", "issue:36010", "issue:36331", @@ -4745,6 +4676,7 @@ "issue:43295", "issue:43316", "issue:43317", + "issue:43322", "issue:43404", "issue:43425", "issue:43526", @@ -4755,7 +4687,6 @@ "issue:43606", "issue:43646", "issue:43708", - "issue:43873", "issue:43881", "issue:44008", "issue:44186", @@ -4766,7 +4697,6 @@ "issue:44671", "issue:44734", "issue:44743", - "issue:44756", "issue:44898", "issue:44913", "issue:45161", @@ -4785,12 +4715,11 @@ "issue:46869" ], "soft_pairs": [ - "issue:43873|issue:44756", "issue:46612|issue:46858", "issue:43577|issue:43708", "issue:43295|issue:45701", - "issue:43550|issue:43606", "issue:45685|issue:46821", + "issue:43550|issue:43606", "issue:44186|issue:44610", "issue:43646|issue:46665", "issue:43404|issue:45357", @@ -4812,198 +4741,212 @@ "issue:44186|issue:44361", "issue:43526|issue:43881", "issue:43316|issue:44913", - "issue:43425|issue:46821" + "issue:43425|issue:46821", + "issue:42915|issue:43322", + "issue:28530|issue:30064" ], - "prepared_review_unit_hash": "ccf1e3ba3f196e5fc3e8a476e854487221cd15c0c472a4446629d5c194e4533e", + "prepared_review_unit_hash": "1a3bbb3d74fc120cd30bc5e20b91ea2123d977837ab0e2522d72af36e86ee658", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 28, - "serialized_chars": 36122, - "estimated_input_tokens": 9031, - "estimated_eval_tokens": 18318 + "soft_pair_count": 29, + "serialized_chars": 36434, + "estimated_input_tokens": 9109, + "estimated_eval_tokens": 18474 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: it mixes unrelated Transformers regressions across loading, tokenization, generation, quantization, offloading, and serve code paths. None of the soft pairs look like the same concrete bug/change, so all candidate edges are rejected. The broad custom-model initialization regression is the best representative issue in the set.", - "confidence": 0.74, - "canonical_issue_reason": "Issue 43646 is the broadest, most central regression report here: a Transformers 5.0 breakage around custom model initialization, which best fits the cluster\u2019s overall version/loading-regression theme.", - "canonical_pr_reason": null, - "best_issue_reason": "43646 is the best single representative because it is concrete, high-level, and closest to several custom-loading/version-regression reports, even though the cluster is not a tight duplicate set.", - "best_pr_reason": null, + "summary": "This cluster is highly heterogeneous: training, generation, serving, tokenizer, config, and model-specific regressions are mixed together, with no clear duplicate center and no PRs provided.", + "confidence": 0.96, + "canonical_issue_reason": "No single issue is a credible canonical duplicate target because the reports cover unrelated bugs and features across many subsystems.", + "canonical_pr_reason": "No PRs are present in this cluster, so there is no canonical PR.", + "best_issue_reason": "No issue is a strong global representative; the set is too diverse to collapse onto one underlying bug.", + "best_pr_reason": "No PRs are present, so there is no best PR candidate.", "soft_edge_verdicts": [ - { - "left": "issue:43873", - "right": "issue:44756", - "accept": false, - "reason": "Different bugs: quantization/offload behavior versus mmap/OOM on Strix Halo; no shared code path." - }, { "left": "issue:46612", "right": "issue:46858", "accept": false, - "reason": "Both involve cache/generation, but one is beam-search reorder skipping in specific models and the other is static-cache slice arithmetic on non-CUDA backends." + "reason": "Both mention cache/generation, but one is beam-search reorder skipping and the other is static-cache seq-length arithmetic on non-CUDA backends; different bugs." }, { "left": "issue:43577", "right": "issue:43708", "accept": false, - "reason": "Unrelated: dtype propagation in Blip2 loading versus Trainer max_steps calculation on checkpoint resume." + "reason": "One is dtype loading for BLIP2, the other is resume_from_checkpoint max_steps math; unrelated failures." }, { "left": "issue:43295", "right": "issue:45701", "accept": false, - "reason": "Both are version-related regressions, but one is processor/tokenizer API breakage and the other is tokenization output changes; different underlying issues." + "reason": "One is a processor.tokenizer regression, the other is a broad tokenization-version change; not the same concrete bug." }, { - "left": "issue:43550", - "right": "issue:43606", + "left": "issue:45685", + "right": "issue:46821", "accept": false, - "reason": "Different models and failures: torch.compile with SDPA on Bamba versus CPU-offload device mismatch on bark-small." + "reason": "MPS histogram/int error and TimesFM window_size AttributeError are different model/API issues." }, { - "left": "issue:45685", - "right": "issue:46821", + "left": "issue:43550", + "right": "issue:43606", "accept": false, - "reason": "MoE MPS histogram support error is unrelated to TimesFM window_size AttributeError." + "reason": "Torch.compile+SDPA on Bamba and CPU offload device mismatch on bark-small are distinct code-path failures." }, { "left": "issue:44186", "right": "issue:44610", "accept": false, - "reason": "Different model-specific preprocessing bugs: LayoutLMv2 tokenizer padding/NER crash versus OmDet-Turbo image-size mismatch." + "reason": "LayoutLMv2 tokenizer crashes on NER padding/truncation, while OmDet-Turbo has an image-size mismatch; different subsystems." }, { "left": "issue:43646", "right": "issue:46665", "accept": false, - "reason": "Related broadly to loading/custom code, but one is a Transformers 5.0 initialization regression and the other is from_pretrained behavior inside Python classes." + "reason": "Both involve loading/custom model behavior, but one is Transformers 5.0 custom initialization breakage and the other is from_pretrained inside Python classes; too broad and not the same bug." }, { "left": "issue:43404", "right": "issue:45357", "accept": false, - "reason": "Different models and failure modes: untied lm_head in Mistral3 versus wrong visual encoder keys saved for Qwen3.5." + "reason": "One is untied lm_head weights in Mistral3, the other is incorrect visual encoder keys on save_pretrained for Qwen3.5." }, { "left": "issue:44734", "right": "issue:46858", "accept": false, - "reason": "Both touch cache/generation, but one crashes in serve KV-cache continuation and the other is a static-cache device-tensor arithmetic bug." + "reason": "Both are cache-related, but one is serve KV-cache continuation indexing and the other is prefill seq-length tensor arithmetic; different failure points." }, { "left": "issue:44610", "right": "issue:44898", "accept": false, - "reason": "Both concern input sizing, but they affect different models and mechanisms; not the same bug." + "reason": "Processor output size mismatch vs Perceiver non-default-resolution interpolation failure are not the same underlying issue." }, { "left": "issue:42915", "right": "issue:43317", "accept": false, - "reason": "Both involve quantization/offload, but one is FineGrainedFP8 on Qwen3MoE and the other is dequantized loading with device_map auto." + "reason": "FineGrainedFP8Config failure for Qwen3Moe and dequantized model load with gpu+cpu offload are unrelated." }, { "left": "issue:36010", "right": "issue:42617", "accept": false, - "reason": "Completely unrelated: import error for GenerationMixin versus inability to run a 3d_parallel script." + "reason": "GenerationMixin import failure and inability to run 3d_parallel.py are unrelated." }, { "left": "issue:44423", "right": "issue:45406", "accept": false, - "reason": "Both are serve multimodal crashes, but the exceptions and root causes differ: string `.to()` misuse versus missing `_tokenizer` on Gemma4Processor." + "reason": "Both are serving crashes, but one is multimodal continuous batching string-to(). error and the other is missing _tokenizer on Gemma4Processor; different bugs." }, { "left": "issue:44008", "right": "issue:45823", "accept": false, - "reason": "Different Gemma model families and distinct forward-time failures: variable-name collision versus device mismatch with device_map auto." + "reason": "Gemma3n variable-name collision in forward and Gemma4 PLE device mismatch are different model-specific regressions." + }, + { + "left": "issue:46821", + "right": "issue:46869", + "accept": false, + "reason": "TimesFM window_size AttributeError and YOLO multi-GPU device_map failure do not share the same code path." }, { "left": "issue:45632", "right": "issue:45684", "accept": false, - "reason": "Different filesystem/save_pretrained problems: trust_remote_code cache collisions versus read-only permission propagation." + "reason": "One is a trust_remote_code cache-path collision, the other is permissions propagation in save_pretrained; unrelated." }, { "left": "issue:44464", "right": "issue:46858", "accept": false, - "reason": "Both involve compiled generation, but one is inconsistent outputs with chunked generation and the other is a cache-length tensor arithmetic failure." + "reason": "Chunked generation inconsistency under compiled forward is a different bug from static-cache seq-length arithmetic." }, { "left": "issue:41720", "right": "issue:45161", "accept": false, - "reason": "Different model-parallelism issues: cudaErrorAssert on Qwen3 auto device mapping versus TP not working on GPT-OSS MoE." + "reason": "Auto device-mapped Qwen3 cudaErrorAssert and GPT-OSS MoE TP malfunction are different parallelism/device issues." }, { "left": "issue:28530", "right": "issue:33666", "accept": false, - "reason": "Unrelated: early stopping metric resolution versus Qwen2-VL multi-GPU training." + "reason": "Early-stopping metric lookup and Qwen2-VL multi-GPU training are unrelated." }, { "left": "issue:44671", "right": "issue:44743", "accept": false, - "reason": "Different bugs in different models: CamemBERT MLM predictions versus Qwen3.5 recurrent state reset with cache." + "reason": "CamemBERT masked-LM prediction regression and Qwen3.5 recurrent-state reset with cache are different model behaviors." }, { "left": "issue:43531", "right": "issue:43572", "accept": false, - "reason": "Unrelated model-specific issues: sliding window behavior in Qwen3-MoE versus missing pad_token_idx in StableLmConfig." + "reason": "Qwen3-MoE sliding_window handling and StableLmConfig pad_token_idx omission are unrelated configuration/model issues." }, { "left": "issue:42915", "right": "issue:46291", "accept": false, - "reason": "Different problems: Qwen3MoE FP8 failure versus pipeline import failing due to kernels version/revision requirements." + "reason": "FP8 model failure and kernels import/version requirement error are separate dependency/runtime problems." }, { "left": "issue:36331", "right": "issue:41720", "accept": false, - "reason": "Trainer compute_loss signature mismatch is unrelated to Qwen3 device-map CUDA assert." + "reason": "Trainer compute_loss signature change and Qwen3 auto device mapping CUDA assert are unrelated." }, { "left": "issue:44186", "right": "issue:44361", "accept": false, - "reason": "Both are tokenizer/task crashes, but they affect different tokenizers and different error conditions." + "reason": "Different tokenizers, different models, and different crash conditions; not the same bug." }, { "left": "issue:43526", "right": "issue:43881", "accept": false, - "reason": "Different issues: BeitImageProcessorFast reduce_labels bug versus glm-4v-9b loading failure." + "reason": "Beit reduce_labels bug and glm-4v-9b loading failure are unrelated." }, { "left": "issue:43316", "right": "issue:44913", "accept": false, - "reason": "Both are config-related, but one is an API discrepancy in Gemma3TextConfig and the other is a GPTNeoX rotary_pct reload bug." + "reason": "Gemma3TextConfig API mismatch and GPTNeoXConfig rotary_pct reload regression are distinct config issues." }, { "left": "issue:43425", "right": "issue:46821", "accept": false, - "reason": "No meaningful overlap: Torch 2.10 incompatibility versus TimesFM window_size AttributeError." + "reason": "Torch 2.10 compatibility and TimesFM window_size AttributeError are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:43322", + "accept": false, + "reason": "FineGrainedFP8Config failure and Llava Next segmentation fault are different bugs." + }, + { + "left": "issue:28530", + "right": "issue:30064", + "accept": false, + "reason": "Early stopping metric resolution and void segmentation-map processing are unrelated." } ] }, @@ -5027,7 +4970,6 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-500", "nodes": [ - "issue:28530", "issue:29159", "issue:30064", "issue:33357", @@ -5037,8 +4979,6 @@ "issue:36032", "issue:41720", "issue:42490", - "issue:42915", - "issue:43322", "issue:43329", "issue:43388", "issue:43531", @@ -5061,12 +5001,15 @@ "issue:44589", "issue:44625", "issue:44898", + "issue:44987", "issue:44991", + "issue:45042", "issue:45059", "issue:45072", "issue:45357", "issue:45406", "issue:45464", + "issue:45479", "issue:45701", "issue:45704", "issue:45759", @@ -5077,17 +5020,15 @@ "issue:46858" ], "soft_pairs": [ - "issue:42915|issue:43322", - "issue:28530|issue:30064", "issue:33357|issue:35707", "issue:29159|issue:36032", "issue:43540|issue:43782", "issue:43550|issue:43653", "issue:44387|issue:45701", - "issue:43873|issue:45704", "issue:30064|issue:34634", - "issue:44492|issue:44509", + "issue:43873|issue:45704", "issue:43606|issue:43720", + "issue:44492|issue:44509", "issue:44534|issue:46665", "issue:43531|issue:45759", "issue:41720|issue:43746", @@ -5106,156 +5047,146 @@ "issue:44077|issue:44625", "issue:43644|issue:46665", "issue:44062|issue:44451", - "issue:44898|issue:45072" + "issue:44898|issue:45072", + "issue:44987|issue:45042", + "issue:44625|issue:45479" ], - "prepared_review_unit_hash": "cf41b9722903f31118a16702984d6cc895a3ce6d69d24f5eb13a3b03f7816897", + "prepared_review_unit_hash": "4679ea54ab8bb52a7c12b6f82d5ca966221e7fe72f56c1a2e617cfe9798664d1", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36754, - "estimated_input_tokens": 9189, - "estimated_eval_tokens": 18634 + "serialized_chars": 36838, + "estimated_input_tokens": 9210, + "estimated_eval_tokens": 18676 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose semantic grouping of unrelated Transformers issues. The only clearly duplicate-like theme is the v5 non-persistent-buffer regression, but none of the provided soft-edge pairs are strong enough to merge as true duplicates/changes.", - "confidence": 0.33, - "canonical_issue_reason": "Issue 43644 is the clearest concrete regression report in the cluster and is the most canonical anchor for the v5 non-persistent-buffer bug theme; the nearby 44534 report looks like a rephrase of the same problem.", + "summary": "This cluster is not a true duplicate set: the items span unrelated tokenizer, multimodal, quantization/offload, serving, and training issues. I would reject every soft edge candidate.", + "confidence": 0.93, + "canonical_issue_reason": "No single issue is a good canonical representative because the cluster is a loose similarity bucket covering many unrelated bug classes and code paths.", "canonical_pr_reason": null, - "best_issue_reason": "43644 is the best representative because it describes a specific, reproducible framework regression with a clear failure mode, while most other issues here are model-specific or unrelated.", + "best_issue_reason": "No issue is especially suitable as a global best representative for dedupe; the artifacts do not converge on one underlying bug or fix.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:42915", - "right": "issue:43322", - "accept": false, - "reason": "Different models and failures: FP8 config loading vs Llava Next segfault." - }, - { - "left": "issue:28530", - "right": "issue:30064", - "accept": false, - "reason": "Early stopping/metric selection is unrelated to void segmentation map processing." - }, { "left": "issue:33357", "right": "issue:35707", "accept": false, - "reason": "MacOS CLIP bus error is a different code path from generation with inputs_embeds/past_key_values." + "reason": "Different problems: MacOS CLIP bus error vs generation with inputs_embeds/past_key_values." }, { "left": "issue:29159", "right": "issue:36032", "accept": false, - "reason": "Both involve tokenizers, but one is slow/fast inconsistency and the other is a T5 tokenizer class-loading conflict." + "reason": "Both tokenizer-related, but one is slow/fast inconsistency while the other is a T5 tokenizer load-time method conflict." }, { "left": "issue:43540", "right": "issue:43782", "accept": false, - "reason": "Different Qwen3 multimodal bugs: video input handling vs weight_only loading." + "reason": "Both involve Qwen multimodal loading/processing, but they hit different model paths and failure modes." }, { "left": "issue:43550", "right": "issue:43653", "accept": false, - "reason": "Unrelated subsystems: torch.compile/SDPA vs BigBird tokenizer special-token registration." + "reason": "Unrelated: torch.compile/SDPA failure in Bamba vs BigBirdTokenizer special-token decode bug." }, { "left": "issue:44387", "right": "issue:45701", "accept": false, - "reason": "Quantization OOM and tokenization changes are different issues with no shared bug path." - }, - { - "left": "issue:43873", - "right": "issue:45704", - "accept": false, - "reason": "Offloading/quantization behavior is unrelated to T5's RMSNorm memory leak." + "reason": "Int4 quantization memory growth is unrelated to tokenization changes across versions." }, { "left": "issue:30064", "right": "issue:34634", "accept": false, - "reason": "Image segmentation-map processing and BarkProcessor voice_preset are unrelated." + "reason": "Different processors and symptoms: void segmentation map handling vs Bark voice_preset behavior." }, { - "left": "issue:44492", - "right": "issue:44509", + "left": "issue:43873", + "right": "issue:45704", "accept": false, - "reason": "A typo in cache strategies is not the same as stale docs mentioning removed pipeline tasks." + "reason": "Offloading/quantization mismatch is not the same as T5's apex RMSNorm memory leak." }, { "left": "issue:43606", "right": "issue:43720", "accept": false, - "reason": "Different failure modes and models: CPU offload device mismatch vs packed-weight unpacking during accelerate loading." + "reason": "Both are load/offload regressions, but they affect different models and concrete code paths." + }, + { + "left": "issue:44492", + "right": "issue:44509", + "accept": false, + "reason": "A documentation typo is not the same as stale pipeline-task docs removed in v5." }, { "left": "issue:44534", "right": "issue:46665", "accept": false, - "reason": "Non-persistent buffer corruption is unrelated to from_pretrained behavior inside Python classes." + "reason": "Non-persistent buffer initialization is unrelated to from_pretrained behavior inside Python classes." }, { "left": "issue:43531", "right": "issue:45759", "accept": false, - "reason": "Qwen3-MoE sliding_window bug and Qwen multimodal config unwrapping are different problems." + "reason": "Different Qwen config bugs: sliding_window handling vs failing to unwrap text_config in from_config." }, { "left": "issue:41720", "right": "issue:43746", "accept": false, - "reason": "Auto device-mapping CUDA assert is unrelated to GraniteSpeech PEFT local checkpoint loading." + "reason": "Qwen3 auto device-mapping CUDA assert is unrelated to GraniteSpeech PEFT checkpoint loading." }, { "left": "issue:42490", "right": "issue:43329", "accept": false, - "reason": "Trailing-slash import warning and undefined video-branch variables are different bugs." + "reason": "Local-path import warning and multimodal token counting are separate bugs." }, { "left": "issue:43388", "right": "issue:45059", "accept": false, - "reason": "Metric gathering tuple-label truncation is unrelated to SAM3 text/box behavior." + "reason": "Metric gathering truncation of tuple labels is unrelated to SAM3 text/bbox behavior." }, { "left": "issue:45406", "right": "issue:45464", "accept": false, - "reason": "Gemma4Processor serve crash and Qwen3.5 streaming API failure are distinct issues." + "reason": "Gemma4Processor missing _tokenizer in serve is a different failure than Qwen streaming inference." }, { "left": "issue:43606", "right": "issue:43653", "accept": false, - "reason": "CPU offload device mismatch is not the same as tokenizer special-token registration." + "reason": "Device-mismatch CPU offload bug and tokenizer special-token registration bug do not share a code path." }, { "left": "issue:34567", "right": "issue:35707", "accept": false, - "reason": "TrainerState token counting and progressive generation are unrelated." + "reason": "Trainer token counters and progressive generation with cache are unrelated." }, { "left": "issue:43531", "right": "issue:46821", "accept": false, - "reason": "Different models and bugs: Qwen3-MoE sliding_window vs TimesFM window_size attribute." + "reason": "Both mention windowing, but Qwen3-MoE sliding_window and TimesFM window_size are different APIs and bugs." }, { "left": "issue:34634", @@ -5267,55 +5198,67 @@ "left": "issue:43540", "right": "issue:46858", "accept": false, - "reason": "Video-input processing and static-cache seq-length tensor arithmetic are different code paths." + "reason": "Qwen3OmniMoe video-processing ValueError is not the same as static-cache slice arithmetic on non-CUDA backends." }, { "left": "issue:45759", "right": "issue:45859", "accept": false, - "reason": "Both are Qwen3.5-related, but one is config unwrapping and the other is CPU-offload rotary embedding compatibility." + "reason": "Both are Qwen3.5 config/offload issues, but one is config unwrapping while the other is embedding forward compatibility under CPU offload." }, { "left": "issue:44451", "right": "issue:44991", "accept": false, - "reason": "Model-specific tokenizer load failures are not enough to treat these as the same bug." + "reason": "Different model/tokenizer loading regressions for different checkpoints." }, { "left": "issue:45357", "right": "issue:46402", "accept": false, - "reason": "Saving incorrect visual encoder keys and missing text-encoder weights on load are different directions and symptoms." + "reason": "Both concern multimodal save/load state, but they affect different models and different weight keys." }, { "left": "issue:43531", "right": "issue:44589", "accept": false, - "reason": "Sliding-window behavior and missing Float8 storage are unrelated." + "reason": "Sliding-window handling in Qwen3-MoE is unrelated to missing Float8 storage errors during load." }, { "left": "issue:44077", "right": "issue:44625", "accept": false, - "reason": "patchtsmixer post_init policy and Qwen3.5 num_labels propagation are unrelated." + "reason": "Optional post_init validation and num_labels propagation are different configuration bugs." }, { "left": "issue:43644", "right": "issue:46665", "accept": false, - "reason": "Junk-filled non-persistent buffers are unrelated to from_pretrained behavior in user-defined classes." + "reason": "Non-persistent buffer junk in v5 is not the same as from_pretrained behavior inside user-defined Python classes." }, { "left": "issue:44062", "right": "issue:44451", "accept": false, - "reason": "AddedToken keyword conflicts and ScandiBERT loading failure are not the same concrete bug." + "reason": "AddedToken keyword duplication is a tokenizer construction bug, not a general model-loading failure." }, { "left": "issue:44898", "right": "issue:45072", "accept": false, - "reason": "Perceiver interpolation failure and bfloat16 dtype mismatch are unrelated." + "reason": "Perceiver image resolution handling and bfloat16 dtype mismatches are unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45042", + "accept": false, + "reason": "One is a model loading regression, the other is an image-processor dependency requirement bug." + }, + { + "left": "issue:44625", + "right": "issue:45479", + "accept": false, + "reason": "Config propagation to text_config is distinct from the degenerate zero-loss issue in sequence classification." } ] }, @@ -5344,9 +5287,11 @@ "issue:39692", "issue:41720", "issue:42886", + "issue:43295", "issue:43381", "issue:43421", "issue:43493", + "issue:43550", "issue:43653", "issue:43720", "issue:43746", @@ -5357,25 +5302,23 @@ "issue:43927", "issue:43994", "issue:44186", + "issue:44361", "issue:44368", "issue:44514", "issue:44545", "issue:44610", - "issue:44625", "issue:44671", "issue:44843", "issue:44918", "issue:44936", - "issue:44987", "issue:44995", - "issue:45042", "issue:45081", "issue:45356", "issue:45357", "issue:45381", - "issue:45479", "issue:45704", "issue:45759", + "issue:45834", "issue:45907", "issue:45923", "issue:45987", @@ -5389,8 +5332,6 @@ "issue:46858" ], "soft_pairs": [ - "issue:44987|issue:45042", - "issue:44625|issue:45479", "issue:43381|issue:43844", "issue:44368|issue:45357", "issue:39692|issue:44610", @@ -5409,238 +5350,240 @@ "issue:43994|issue:44671", "issue:39692|issue:43493", "issue:46032|issue:46463", - "issue:43421|issue:44186", "issue:43782|issue:45381", + "issue:43421|issue:44186", "issue:45081|issue:45356", - "issue:44995|issue:46858", "issue:30064|issue:36010", + "issue:44995|issue:46858", "issue:45356|issue:45907", "issue:39692|issue:43873", - "issue:43720|issue:43819", "issue:45987|issue:46829", + "issue:43720|issue:43819", "issue:46829|issue:46832", - "issue:43653|issue:43720" + "issue:43653|issue:43720", + "issue:43295|issue:45834", + "issue:43550|issue:44361" ], - "prepared_review_unit_hash": "49330532c7e9d13990193e435fe55c4141b47fe9c45321075041108012557ee3", + "prepared_review_unit_hash": "2fff6a4147a75e5534bed2495d51020b71d658c7a05c01542e19cd7352d497d8", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37549, - "estimated_input_tokens": 9388, - "estimated_eval_tokens": 19032 + "serialized_chars": 37467, + "estimated_input_tokens": 9367, + "estimated_eval_tokens": 18990 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated issues that were linked by loose lexical similarity. The only clear duplicate pair is the two Qwen2_5_VLProcessor batched padding=False crash reports; everything else should stay separate.", - "confidence": 0.96, - "canonical_issue_reason": "issue:44545 is the cleanest representative of the one true duplicate pair: it states the batched padding=False crash clearly and matches issue:44514 on the same underlying bug.", + "summary": "This cluster is mostly a loose semantic grouping of tokenizer/processor/model-loading bug reports, not a single duplicate set. The only clear same-issue duplicate is the Qwen2_5_VLProcessor batched apply_chat_template crash pair; the rest are related only at subsystem level and should stay separate.", + "confidence": 0.81, + "canonical_issue_reason": "Issue 44514 is the clearest exact duplicate representative: 44545 reports the same Qwen2_5_VLProcessor.apply_chat_template batched-input crash with padding=False.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44545 is the strongest canonical issue candidate because it is the clearer, more concise report of the duplicated Qwen2_5_VLProcessor.apply_chat_template failure.", + "best_issue_reason": "44514 is the best global representative because it is a concrete, exact duplicate pair with a specific reproducible crash; the other candidates are broader or only loosely related.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44987", - "right": "issue:45042", - "accept": false, - "reason": "Different failures in different subsystems: loading a specific model vs PIL backend image processors requiring torchvision." - }, - { - "left": "issue:44625", - "right": "issue:45479", - "accept": false, - "reason": "One is Qwen3.5 config propagation into text_config; the other is a generic sequence-classification loss bug across models." - }, { "left": "issue:43381", "right": "issue:43844", "accept": false, - "reason": "Eval-mode gradient checkpointing restriction is unrelated to a ZeRO-3 gradient explosion regression." + "reason": "Both touch training-time behavior, but one is gradient checkpointing in eval mode and the other is a ZeRO-3/DeepSpeed gradient anomaly; different bugs and code paths." }, { "left": "issue:44368", "right": "issue:45357", "accept": false, - "reason": "A warning about tie_word_embeddings in LoRA fine-tuning is not the same bug as incorrect visual encoder keys being saved." + "reason": "Both mention Qwen model warnings/regressions, but one is a tie_word_embeddings warning and the other is incorrect visual-encoder save keys; not the same bug." }, { "left": "issue:39692", "right": "issue:44610", "accept": false, - "reason": "Both mention vision models, but one is a SigLIP2 docs/model mismatch and the other is an OmDet-Turbo image-size mismatch." + "reason": "Both are vision/multimodal issues, but one is a SigLIP2 doc/model mismatch plus quantization failure and the other is an OmDet-Turbo image-size mismatch." }, { "left": "issue:46655", "right": "issue:46693", "accept": false, - "reason": "Video padding calculation in a processor is unrelated to a FlashAttention2 long-context performance regression." + "reason": "Video-related, but one is a padding calculation bug in the video processor and the other is a FA2 long-context performance regression." }, { "left": "issue:43421", "right": "issue:43927", "accept": false, - "reason": "Runtime special-token post-processing and save/load of custom token IDs are different tokenizer bugs." + "reason": "Both involve token handling, but one is runtime post-processor updates and the other is save/load losing custom token IDs; different failure modes." }, { "left": "issue:41720", "right": "issue:45704", "accept": false, - "reason": "Qwen3 device-mapping CUDA asserts and T5 apex RMSNorm memory leak are unrelated." + "reason": "Unrelated root causes: Qwen3 auto device mapping cudaErrorAssert versus a T5 apex fused RMSNorm memory leak." }, { "left": "issue:43746", "right": "issue:46050", "accept": false, - "reason": "PEFT/local checkpoint loading failure is a different code path from quantized save_pretrained/from_pretrained corruption." + "reason": "Both are load/save regressions, but one is PEFT adapter loading from local checkpoints and the other is sinq quantized model persistence." }, { "left": "issue:43381", "right": "issue:45704", "accept": false, - "reason": "Eval-mode checkpointing behavior is unrelated to the T5 apex memory leak." + "reason": "Gradient checkpointing in eval mode and an apex RMSNorm leak are different bugs with no shared code path." }, { "left": "issue:42886", "right": "issue:44843", "accept": false, - "reason": "Both involve offline mode, but one is cache loading and the other is an unconditional model_info() call in _patch_mistral_regex." + "reason": "Both break offline workflows, but one is generic cache loading with HF_HUB_OFFLINE and the other is a specific model_info() call in _patch_mistral_regex." }, { "left": "issue:45704", "right": "issue:45923", "accept": false, - "reason": "T5 apex memory leak and missing supports_gradient_checkpointing on a trust_remote_code model are separate issues." + "reason": "T5 RMSNorm leak versus a missing supports_gradient_checkpointing flag on a trust_remote_code model; unrelated issues." }, { "left": "issue:44918", "right": "issue:44936", "accept": false, - "reason": "Qwen3.5 embedding unpacking in TRL SFT is not the same as trainer.evaluate() failing after trainer.train()." + "reason": "Both involve training/evaluation workflows, but one is SFT embedding unpacking and the other is trainer.evaluate() failing after train(); different code paths." }, { "left": "issue:44514", "right": "issue:44545", "accept": true, - "reason": "Same bug report twice: Qwen2_5_VLProcessor.apply_chat_template crashes on batched input when padding=False." + "reason": "Same title and same reported Qwen2_5_VLProcessor.apply_chat_template crash on batched input when padding=False; clear duplicate." }, { "left": "issue:45759", "right": "issue:45923", "accept": false, - "reason": "Unwrapping text_config in composite multimodal configs is unrelated to the gradient-checkpointing flag on trust_remote_code variants." + "reason": "Composite Qwen config unwrapping and a missing gradient-checkpointing flag are separate model-initialization issues." }, { "left": "issue:43746", "right": "issue:43873", "accept": false, - "reason": "Local checkpoint loading for GraniteSpeech adapters is unrelated to quantization offloading behavior." + "reason": "Local checkpoint loading for PEFT adapters versus quantization offloading behavior; too different to merge." }, { "left": "issue:43421", "right": "issue:43653", "accept": false, - "reason": "Post-processor updates after special-token mutation are different from BigBirdTokenizer not registering mask as special." + "reason": "Runtime post-processor refresh and BigBird mask-token special-token registration are related only broadly through tokenizers." }, { "left": "issue:43994", "right": "issue:44671", "accept": false, - "reason": "SigLIP2 pipeline/model output issues and CamemBERT masked LM regressions are unrelated." + "reason": "Different models and failures: SigLIP2 nonsensical outputs versus CamemBERT masked-LM prediction issues." }, { "left": "issue:39692", "right": "issue:43493", "accept": false, - "reason": "Both are SigLIP2-related, but one is a broken documentation example and the other is an implementation discrepancy against JAX." + "reason": "Both are SigLIP2-related, but one is a doc/example mismatch and quantization issue while the other is HF/JAX implementation parity; not the same concrete bug." }, { "left": "issue:46032", "right": "issue:46463", "accept": false, - "reason": "A silent use_cache correctness bug is not the same as an incorrect error message in T5 decoder input validation." + "reason": "One is incorrect use_cache results for Mamba2Mixer, the other is just an error-message bug in T5 decoder validation." }, { - "left": "issue:43421", - "right": "issue:44186", + "left": "issue:43782", + "right": "issue:45381", "accept": false, - "reason": "Tokenizer post-processor updates and LayoutLMv2 NER/batching crashes are different bugs." + "reason": "Different Qwen VL bugs: from_pretrained weight_only loading versus incorrect vision_position_ids in video input." }, { - "left": "issue:43782", - "right": "issue:45381", + "left": "issue:43421", + "right": "issue:44186", "accept": false, - "reason": "Qwen3VL weight_only loading error and Qwen2.5-VL video position-id bug are separate multimodal issues." + "reason": "Both tokenizer-related, but one is special-token post-processor updates and the other is LayoutLMv2 NER/padding crashes; not duplicates." }, { "left": "issue:45081", "right": "issue:45356", "accept": false, - "reason": "Both touch tokenizer regex handling, but one is a crash in _patch_mistral_regex and the other is a Kimi-K2.5 codec/regression report." + "reason": "Both involve _patch_mistral_regex, but one is an AttributeError loading Mistral tokenizers and the other is Kimi-K2.5 codec/warning regression; not the same failure." }, { - "left": "issue:44995", - "right": "issue:46858", + "left": "issue:30064", + "right": "issue:36010", "accept": false, - "reason": "A stale indexer cache on second forward pass is unrelated to static-cache seq_length returning a device tensor." + "reason": "Completely unrelated: void segmentation map processing versus an ImportError for GenerationMixin." }, { - "left": "issue:30064", - "right": "issue:36010", + "left": "issue:44995", + "right": "issue:46858", "accept": false, - "reason": "Void segmentation map processing and a GenerationMixin import error have no meaningful overlap." + "reason": "Both mention cache/sequence length, but one is a stale indexer cache on second forward pass and the other is a static-cache device-tensor arithmetic bug under torch.compile." }, { "left": "issue:45356", "right": "issue:45907", "accept": false, - "reason": "Tokenizer codec/regex regression and list_repo_files revision forwarding are unrelated." + "reason": "Tokenizer regression versus a hub API revision-forwarding bug; unrelated." }, { "left": "issue:39692", "right": "issue:43873", "accept": false, - "reason": "SigLIP2 example/model issues and quantization offloading behavior are different bugs." + "reason": "SigLIP2 doc/model mismatch versus offloading with quantization; no shared underlying bug." }, { - "left": "issue:43720", - "right": "issue:43819", + "left": "issue:45987", + "right": "issue:46829", "accept": false, - "reason": "BitNet packed-weight loading and DAC from_latents/STE mismatch are unrelated code paths." + "reason": "StaticCache.get_seq_length() returning a tensor and VideoPrism output-field misassignment are unrelated." }, { - "left": "issue:45987", - "right": "issue:46829", + "left": "issue:43720", + "right": "issue:43819", "accept": false, - "reason": "StaticCache seq_length contract bug and VideoPrism output-field mismatch are unrelated." + "reason": "BitNet packed-weight loading and DAC from_latents STE mismatch are different model-specific bugs." }, { "left": "issue:46829", "right": "issue:46832", "accept": false, - "reason": "VideoPrism hidden_states misassignment is different from RT-DETR feature-level forwarding errors." + "reason": "One is an output-field bug in VideoPrism classification, the other is an RT-DETR forward-shape mismatch; different code paths." }, { "left": "issue:43653", "right": "issue:43720", "accept": false, - "reason": "BigBirdTokenizer special-token registration and BitNet accelerate-loading issues are unrelated." + "reason": "BigBird special-token decode failure and BitNet accelerate-loading unpacking are unrelated." + }, + { + "left": "issue:43295", + "right": "issue:45834", + "accept": false, + "reason": "Processor.tokenizer regression in v4.57.5 versus a Kosmos2.5 long-OCR index error; different issues." + }, + { + "left": "issue:43550", + "right": "issue:44361", + "accept": false, + "reason": "A torch.compile/SDPA crash in Bamba is unrelated to an MLukeTokenizer AttributeError." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded and conservative overall. The only clearly duplicate pair in the packet is issue:44514/issue:44545, which have near-identical titles and matching descriptions. The rest of the soft-edge rejections are appropriately cautious and tied to materially different failure modes or code paths." + "feedback": "The analysis is grounded in the packet and stays conservative. The only clearly supported duplicate pair is issue:44514 / issue:44545, which share the same Qwen2_5_VLProcessor.apply_chat_template batched-input crash. The remaining pairwise rejections are appropriately cautious based on the issue titles alone, and the summary does not overclaim broader duplication beyond subsystem-level similarity." }, "evaluator_used": true, "retried": false, @@ -5667,12 +5610,12 @@ "issue:36331", "issue:39692", "issue:41720", - "issue:43295", "issue:43335", "issue:43531", "issue:43540", "issue:43550", "issue:43596", + "issue:43606", "issue:43611", "issue:43618", "issue:43643", @@ -5687,7 +5630,6 @@ "issue:44062", "issue:44186", "issue:44220", - "issue:44361", "issue:44442", "issue:44448", "issue:44488", @@ -5698,6 +5640,7 @@ "issue:44877", "issue:44898", "issue:44991", + "issue:45072", "issue:45081", "issue:45084", "issue:45183", @@ -5711,10 +5654,8 @@ "issue:46858" ], "soft_pairs": [ - "issue:43295|issue:45834", - "issue:43550|issue:44361", - "issue:43550|issue:43927", "issue:44062|issue:44488", + "issue:43550|issue:43927", "issue:43596|issue:44220", "issue:45381|issue:46693", "issue:43540|issue:46294", @@ -5739,210 +5680,205 @@ "issue:43644|issue:43749", "issue:34634|issue:36010", "issue:45183|issue:45479", - "issue:45584|issue:45834" + "issue:45584|issue:45834", + "issue:43606|issue:45072" ], - "prepared_review_unit_hash": "59d160e69839dc23ae5a4ec954f7ca06ad2280edd4875bf5099e422609e66620", + "prepared_review_unit_hash": "6e034c7e02e035893ceed71979bff5673b37259e1a9d9144c08e7c84406b879c", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36588, - "estimated_input_tokens": 9147, - "estimated_eval_tokens": 18550 + "soft_pair_count": 28, + "serialized_chars": 36296, + "estimated_input_tokens": 9074, + "estimated_eval_tokens": 18404 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is heterogeneous: most items are distinct model-, tokenizer-, config-, or performance-specific issues rather than duplicates. None of the soft edge pairs looks like the same underlying bug/change. If one issue must serve as the anchor, #43611 is the broadest regression in the set.", - "confidence": 0.86, - "canonical_issue_reason": "Issue #43611 is the most broadly applicable regression here: it affects model loading across Transformers v5 via `base_model_prefix`, making it a better cluster anchor than the more model-specific reports.", + "summary": "This cluster is a loose collection of mostly unrelated Transformers bug reports. A few items cluster around loading/config/tokenizer regressions, but the soft pairs still point to different concrete failure modes, models, or code paths, so none should be merged as duplicates.", + "confidence": 0.87, + "canonical_issue_reason": "issue:43611 is the most broadly representative loading regression in the set: it describes a Transformers 5.0.0 model-loading breakage around `base_model_prefix`, which aligns with the cluster\u2019s strongest repeated theme of load/config failures.", "canonical_pr_reason": null, - "best_issue_reason": "#43611 is the best representative issue because it describes a cross-cutting runtime regression with wide user impact, whereas the others are narrower or isolated to specific models/code paths.", + "best_issue_reason": "issue:43611 is the best single issue to anchor this cluster because it is a clear, concrete regression with broader applicability than the more model-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43295", - "right": "issue:45834", - "accept": false, - "reason": "Different bugs: one is a processor/tokenizer regression, the other is a Kosmos2.5 long-OCR index error." - }, - { - "left": "issue:43550", - "right": "issue:44361", + "left": "issue:44062", + "right": "issue:44488", "accept": false, - "reason": "Different subsystems and failure modes: Bamba torch.compile/SDPA vs MLukeTokenizer task handling." + "reason": "Both involve tokenizer loading/special-token behavior, but one is an `AddedToken(..., special=...)` argument error and the other is a model-specific tokenizer load failure; different bugs." }, { "left": "issue:43550", "right": "issue:43927", "accept": false, - "reason": "Unrelated issues: Bamba compile failure vs DiaConfig losing custom token IDs on save/load." - }, - { - "left": "issue:44062", - "right": "issue:44488", - "accept": false, - "reason": "Both are tokenizer-related, but one is an AddedToken constructor error and the other is a model-loading failure for a specific tokenizer." + "reason": "One is a `torch.compile`/SDPA failure in Bamba, the other is a save/load config token-ID loss in Dia; unrelated code paths." }, { "left": "issue:43596", "right": "issue:44220", "accept": false, - "reason": "Different code paths: deepspeed ZeRO-3/BertModel init vs `_torch_extract_fbank_features()`." + "reason": "`BertModel` zero3 init IndexError and `_torch_extract_fbank_features()` are different subsystems and different failure modes." }, { "left": "issue:45381", "right": "issue:46693", "accept": false, - "reason": "Qwen2.5-VL video position IDs bug is not the same as the FA2 long-context performance regression." + "reason": "Qwen2.5-VL video position IDs and FA2 long-context performance regression are unrelated issues." }, { "left": "issue:43540", "right": "issue:46294", "accept": false, - "reason": "Both involve Qwen video flows, but they are different models and different mismatches/errors in the pipeline." + "reason": "Both are video-related, but one is a Qwen3OmniMoe video-processing ValueError and the other is a chat-template token/feature count mismatch in Qwen3.5-9B." }, { "left": "issue:44008", "right": "issue:46858", "accept": false, - "reason": "Gemma3n forward variable collision is unrelated to the static-cache slice arithmetic bug." + "reason": "Gemma3n variable-name collision in forward is unrelated to `_prefill` cache sequence-length tensor arithmetic on non-CUDA backends." }, { "left": "issue:43531", "right": "issue:43643", "accept": false, - "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to `trust_remote_code=True` returning incomplete configs." + "reason": "Sliding-window handling in Qwen3-MoE and missing fields from `AutoConfig.from_pretrained(..., trust_remote_code=True)` are different bugs." }, { "left": "issue:43611", "right": "issue:46665", "accept": false, - "reason": "Both concern loading behavior, but the specific regression is different: base_model_prefix loading vs from_pretrained inside Python classes." + "reason": "Both touch model loading, but one is a specific `base_model_prefix` regression and the other is inconsistent `from_pretrained` behavior inside Python classes." }, { "left": "issue:43653", "right": "issue:43819", "accept": false, - "reason": "Tokenizer special-token registration bug is unrelated to DAC.from_latents not matching forward without STE." + "reason": "BigBirdTokenizer special-token registration and DAC `from_latents`/forward mismatch are unrelated." }, { "left": "issue:44220", "right": "issue:44661", "accept": false, - "reason": "Different features entirely: fbank extraction vs add-new-model-like failing on TOKENIZER_MAPPING_NAMES." + "reason": "Audio feature extraction and `TOKENIZER_MAPPING_NAMES` model-addition logic are separate paths." }, { "left": "issue:44186", "right": "issue:44898", "accept": false, - "reason": "LayoutLMv2 tokenizer padding/NER crash is not the same bug as Perceiver image classification at non-default resolution." + "reason": "LayoutLMv2 tokenizer padding/truncation crashes and Perceiver image-classification failures are different model families and bugs." }, { "left": "issue:43618", "right": "issue:43697", "accept": false, - "reason": "CLIPOutput attentions assignment and RTDetrV2 output drift are separate model bugs." + "reason": "Both are output regressions, but CLIP attentions missing and RTDetrV2 differing outputs are not the same underlying defect." }, { "left": "issue:41720", "right": "issue:44991", "accept": false, - "reason": "Qwen3 auto device-map CUDA assert and est-roberta tokenizer loading failure are unrelated." + "reason": "Qwen3 auto device mapping CUDA assert and tokenizer loading for `EMBEDDIA/est-roberta` are unrelated." }, { "left": "issue:30064", "right": "issue:36331", "accept": false, - "reason": "Image processor segmentation-map failure and `compute_loss` signature mismatch are completely different issues." + "reason": "Void segmentation-map processing and `compute_loss(..., num_items_in_batch)` are different issues in different layers of the stack." }, { "left": "issue:43643", "right": "issue:44877", "accept": false, - "reason": "Both are config-loading problems, but one is missing fields from `trust_remote_code`, the other is strict config rejection for granite_speech." + "reason": "Both concern config loading strictness, but one is missing fields from `trust_remote_code` and the other is a `granite_speech` config rejection; different root causes." }, { "left": "issue:43550", "right": "issue:43819", "accept": false, - "reason": "Torch.compile SDPA failure in Bamba is unrelated to DAC latent decoding mismatch." + "reason": "`torch.compile`+SDPA in Bamba and DAC latent decoding mismatch are unrelated." }, { "left": "issue:43335", "right": "issue:43756", "accept": false, - "reason": "SwitchTransformers sparse-layer creation and Smollm3 RoPE-layer dropping are different model-configuration bugs." + "reason": "SwitchTransformers sparse-layer creation and Smollm3 RoPE-layer dropping are different model-specific architecture bugs." }, { "left": "issue:44843", "right": "issue:45081", "accept": false, - "reason": "Same helper name, but different root causes: offline `model_info()` call vs missing `backend_tokenizer` attribute." + "reason": "Both involve `_patch_mistral_regex`, but one is an offline `model_info()` call and the other is an AttributeError on `backend_tokenizer`; separate failure modes." }, { "left": "issue:44821", "right": "issue:45084", "accept": false, - "reason": "AutoImageProcessor URL loading and template-node compilation are unrelated failures." + "reason": "AutoImageProcessor URL loading and template compilation errors are unrelated." }, { "left": "issue:44186", "right": "issue:44442", "accept": false, - "reason": "LayoutLMv2 tokenizer crashes and AutoTokenizer failing on FastSpeech2ConformerTokenizer are distinct tokenizer bugs." + "reason": "LayoutLMv2 tokenizer input handling and FastSpeech2Conformer tokenizer loading are different tokenizer bugs." }, { "left": "issue:39692", "right": "issue:45834", "accept": false, - "reason": "Documentation/example errors for SigLIP2 are unrelated to Kosmos2.5 OCR long-input indexing." + "reason": "SigLIP2 doc-example errors and Kosmos2.5 long OCR index errors are not the same underlying bug." }, { "left": "issue:44448", "right": "issue:44671", "accept": false, - "reason": "Different models and causes: Pegasus v4/v5 output drift vs CamemBERT masked-LM prediction regression." + "reason": "Both are v5 output differences, but they affect different models and likely different code paths." }, { "left": "issue:43540", "right": "issue:44008", "accept": false, - "reason": "Qwen3OmniMoe video processing error and Gemma3n audio mask variable collision are unrelated multimodal bugs." + "reason": "Qwen3OmniMoe video-input processing and Gemma3n audio mask attribute collision are different modalities and bugs." }, { "left": "issue:43644", "right": "issue:43749", "accept": false, - "reason": "Non-persistent buffer initialization junk and FSDP CPU RAM efficient loading are different v5 regressions." + "reason": "Non-persistent-buffer junk values and FSDP CPU RAM efficient loading are both v5/loading-adjacent, but not the same defect." }, { "left": "issue:34634", "right": "issue:36010", "accept": false, - "reason": "BarkProcessor voice preset handling and GenerationMixin import failure do not describe the same defect." + "reason": "BarkProcessor voice preset and `GenerationMixin` import failure are unrelated." }, { "left": "issue:45183", "right": "issue:45479", "accept": false, - "reason": "Input validation error messages across models are unrelated to the degenerate zero-loss classification bug." + "reason": "Input validation XOR messaging and degenerate zero loss in sequence classification are different model-behavior bugs." }, { "left": "issue:45584", "right": "issue:45834", "accept": false, - "reason": "Whisper empty-transcription generation failure is not the same as Kosmos2.5 long OCR input handling." + "reason": "Whisper empty transcription after special-token alignment and Kosmos2.5 long OCR index errors are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:45072", + "accept": false, + "reason": "CPU-offload device mismatch in Bark and dtype mismatches in bfloat16 inference are distinct runtime issues." } ] }, @@ -5976,7 +5912,6 @@ "issue:43550", "issue:43577", "issue:43582", - "issue:43606", "issue:43645", "issue:43716", "issue:43720", @@ -6016,234 +5951,224 @@ "issue:46846" ], "soft_pairs": [ - "issue:43606|issue:45072", "issue:43322|issue:45685", "issue:43550|issue:45072", "issue:43388|issue:46032", "issue:44479|issue:45834", - "issue:43825|issue:43994", "issue:43782|issue:44977", + "issue:43825|issue:43994", "issue:43582|issue:46512", "issue:43577|issue:43818", "issue:44991|issue:46291", "issue:43421|issue:44811", + "issue:43950|issue:45987", "issue:44561|issue:45020", "issue:43716|issue:44008", - "issue:43950|issue:45987", "issue:43931|issue:46846", "issue:46294|issue:46463", - "issue:43825|issue:43827", - "issue:28218|issue:29159", "issue:44805|issue:44936", + "issue:28218|issue:29159", "issue:44610|issue:45072", "issue:43540|issue:45381", "issue:43720|issue:45072", "issue:43749|issue:44291", + "issue:43825|issue:43827", "issue:43577|issue:43582", "issue:43645|issue:46665", "issue:44671|issue:45701", "issue:42175|issue:45081" ], - "prepared_review_unit_hash": "667680724238723077c7d844a2931b7d5e3aaf8ae47df7cb973d6057fd5353c7", + "prepared_review_unit_hash": "5751aff18ffb06408a9d2d78faa1ef1b9a877bee07f551d210e8069e4cc2ccfe", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 27, - "serialized_chars": 36220, - "estimated_input_tokens": 9055, - "estimated_eval_tokens": 18366 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35339, + "estimated_input_tokens": 8835, + "estimated_eval_tokens": 17926 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose mix of unrelated Transformers bugs, with one clear duplicate pair around tokenizer added-token spacing/slow-vs-fast inconsistency. I would only merge the tokenizer pair; the rest are different models, code paths, or failure modes.", - "confidence": 0.84, - "canonical_issue_reason": "issue:28218 is the most concrete description of the shared tokenizer regression (extra space after an added token), and issue:29159 reads like a broader manifestation of the same problem.", - "canonical_pr_reason": null, - "best_issue_reason": "issue:28218 is the best issue anchor because it is specific, actionable, and captures the actual symptom more directly than the broader tokenizer-consistency report.", - "best_pr_reason": null, + "summary": "This cluster is highly heterogeneous: tokenizer/pipeline docs issues, multimodal/video regressions, dtype/loading bugs, trainer/cache problems, and a few API-contract issues. None of the soft pairs look like the same underlying bug or change, so they should stay separate.", + "confidence": 0.86, + "canonical_issue_reason": "No canonical issue: the cluster is a mix of unrelated tokenizer, video, loading, dtype, and trainer bugs.", + "canonical_pr_reason": "No pull requests are present in this cluster.", + "best_issue_reason": "No clear global representative; the issue set is too heterogeneous for a meaningful duplicate anchor.", + "best_pr_reason": "No pull requests are present in this cluster.", "soft_edge_verdicts": [ - { - "left": "issue:43606", - "right": "issue:45072", - "accept": false, - "reason": "Both mention model/backend incompatibility, but one is a CPU-offload device mismatch and the other is a bfloat16 dtype mismatch on different models." - }, { "left": "issue:43322", "right": "issue:45685", "accept": false, - "reason": "Different failures: Llava Next loading segfault versus an MPS histogram kernel not implemented for Int." + "reason": "Different failures: Llava Next loading crash vs an MPS histogram/int support error in moe code." }, { "left": "issue:43550", "right": "issue:45072", "accept": false, - "reason": "torch.compile+SDPA compile failure is a different code-path problem from dtype mismatch inference errors." + "reason": "Different models and root causes: torch.compile/SDPA failure vs dtype mismatch in bfloat16 inference." }, { "left": "issue:43388", "right": "issue:46032", "accept": false, - "reason": "Metric gathering drops labels in the last batch; the other is incorrect cache behavior for seq_len>1. Not the same bug." + "reason": "Trainer metric gathering drops labels; Mamba2Mixer is a cache correctness bug." }, { "left": "issue:44479", "right": "issue:45834", "accept": false, - "reason": "Both are multimodal input shape issues, but they affect different models and different preprocessing paths." + "reason": "Both are multimodal regressions, but they affect different models and different code paths." }, { - "left": "issue:43825", - "right": "issue:43994", + "left": "issue:43782", + "right": "issue:44977", "accept": false, - "reason": "One is an error-message/translation-support issue; the other is a bad model output regression with AutoModel/pipeline." + "reason": "Loading with weight_only=True is not the same as Qwen3.5 flash-attention generation failure." }, { - "left": "issue:43782", - "right": "issue:44977", + "left": "issue:43825", + "right": "issue:43994", "accept": false, - "reason": "Weight-only loading failure and flash-attention generation failure are distinct problems." + "reason": "One is an error-message regression; the other is a model/pipeline inference-quality problem." }, { "left": "issue:43582", "right": "issue:46512", "accept": false, - "reason": "Apple Silicon warmup TypeError is unrelated to stale config dtype after from_config(dtype=...)." + "reason": "Apple Silicon allocator warmup TypeError is unrelated to stale config.dtype after from_config()." }, { "left": "issue:43577", "right": "issue:43818", "accept": false, - "reason": "BLIP2 dtype propagation and Video-LLaVA temporal-attention/weight-sharing are different model-specific bugs." + "reason": "Blip2 dtype propagation and Video-LLaVA temporal-attention/weight-sharing are different bugs." }, { "left": "issue:44991", "right": "issue:46291", "accept": false, - "reason": "Tokenizer loading failure and kernels dependency/version import failure are unrelated." + "reason": "Tokenizer-loading failure and pipelines import failure are different subsystems and distinct errors." }, { "left": "issue:43421", "right": "issue:44811", "accept": false, - "reason": "Post-processor not updating after special-token changes is different from batch_decode ignoring skip_special_tokens." + "reason": "Runtime post-processor updates are not the same bug as batch_decode ignoring skip_special_tokens." + }, + { + "left": "issue:43950", + "right": "issue:45987", + "accept": false, + "reason": "Non-persistent buffer corruption and StaticCache return-type mismatch are unrelated contract bugs." }, { "left": "issue:44561", "right": "issue:45020", "accept": false, - "reason": "Both involve recent-version remote_code breakage, but the evidence is too broad to treat them as the same concrete issue." + "reason": "Both mention remote_code breakage, but the underlying removed APIs/problems are different." }, { "left": "issue:43716", "right": "issue:44008", "accept": false, - "reason": "Image preprocessor/model dtype mismatch is unrelated to a variable-name collision in Gemma3nModel.forward()." - }, - { - "left": "issue:43950", - "right": "issue:45987", - "accept": false, - "reason": "Silent corruption of non-persistent buffers is a different regression from StaticCache.get_seq_length returning the wrong type." + "reason": "dtype mismatch in a processor/model pair is unrelated to a variable name collision in forward()." }, { "left": "issue:43931", "right": "issue:46846", "accept": false, - "reason": "Qwen3-VL weight-shape loading errors and Qwen3.5 TP-plan OOM/shape issues are not the same bug." + "reason": "Qwen3-VL weight-shape mismatch and missing TP-plan entries are different loading vs tensor-parallel bugs." }, { "left": "issue:46294", "right": "issue:46463", "accept": false, - "reason": "Qwen3.5 video chat-template mismatch and T5 decoder-input validation messaging are unrelated." + "reason": "Video chat-template token/feature mismatch is unrelated to T5 decoder input validation messaging." }, { - "left": "issue:43825", - "right": "issue:43827", + "left": "issue:44805", + "right": "issue:44936", "accept": false, - "reason": "One concerns a runtime pipeline error message; the other is documentation still referencing pipeline() after removals." + "reason": "A tensor-mask shape error is not the same issue as trainer.evaluate() failing after train()." }, { "left": "issue:28218", "right": "issue:29159", - "accept": true, - "reason": "These look like the same tokenizer regression: extra spacing after added tokens fits the broader slow/fast tokenizer inconsistency report." - }, - { - "left": "issue:44805", - "right": "issue:44936", "accept": false, - "reason": "Mask/index shape mismatch at inference is unrelated to trainer.evaluate() failing after trainer.train()." + "reason": "Both are tokenizer regressions, but one is about added-token spacing while the other is a broader slow/fast inconsistency." }, { "left": "issue:44610", "right": "issue:45072", "accept": false, - "reason": "Processor image-size mismatch and dtype mismatch inference failures are different issues." + "reason": "Different models and symptoms: processor size mismatch vs dtype mismatches." }, { "left": "issue:43540", "right": "issue:45381", "accept": false, - "reason": "Both are video-input bugs, but they affect different Qwen model families and different failure modes." + "reason": "Different Qwen video-input bugs with different reported failures and likely different code paths." }, { "left": "issue:43720", "right": "issue:45072", "accept": false, - "reason": "Packed-weight loading during accelerate is unrelated to dtype mismatches in inference." + "reason": "Packed-weight unpacking during accelerate loading is unrelated to bfloat16 dtype mismatches." }, { "left": "issue:43749", "right": "issue:44291", "accept": false, - "reason": "FSDP CPU RAM-efficient loading is a different loading path from init_empty_weights rejecting an unexpected argument." + "reason": "FSDP CPU RAM efficient loading and init_empty_weights argument handling are distinct loading-path regressions." + }, + { + "left": "issue:43825", + "right": "issue:43827", + "accept": false, + "reason": "A bad v5 error message and stale documentation are related topics, but not the same underlying bug/change." }, { "left": "issue:43577", "right": "issue:43582", "accept": false, - "reason": "These are unrelated dtype/model-loading and Apple Silicon allocator problems." + "reason": "Blip2 dtype propagation and Apple Silicon allocator warmup are unrelated." }, { "left": "issue:43645", "right": "issue:46665", "accept": false, - "reason": "Jupyter notebook custom-model initialization and from_pretrained-inside-classes are separate contexts with different failure causes." + "reason": "Jupyter custom-model initialization regression is different from from_pretrained behavior inside Python classes." }, { "left": "issue:44671", "right": "issue:45701", "accept": false, - "reason": "Masked LM prediction regression and tokenizer behavior changes are too different to merge as one bug." + "reason": "CamemBERT prediction errors are not the same as a general tokenization-change report." }, { "left": "issue:42175", "right": "issue:45081", "accept": false, - "reason": "Package/backend install issues are unrelated to a Mistral tokenizer regex patch crash." + "reason": "Missing TensorFlow in an install extra is unrelated to the Mistral tokenizer regex crash." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays mostly conservative on the soft pairs. The only accepted pair, issue:28218 with issue:29159, is a plausible tokenizer duplicate based on the titles and cluster context. The other verdicts avoid overreach and correctly treat different models, code paths, or failure modes as separate." - }, - "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 }, @@ -6314,17 +6239,17 @@ "issue:45003|issue:46285", "issue:46697|issue:46728", "issue:43653|issue:44912", - "issue:43262|issue:43819", "issue:43782|issue:45325", + "issue:43262|issue:43819", "issue:45081|issue:45406", - "issue:45372|issue:46326", "issue:43582|issue:44857", + "issue:45372|issue:46326", "issue:44589|issue:45042", "issue:43550|issue:44912", "issue:43742|issue:45020", "issue:43329|issue:45600", - "issue:44060|issue:44368", "issue:45084|issue:45538", + "issue:44060|issue:44368", "issue:43452|issue:44991", "issue:43526|issue:43697", "issue:43257|issue:43329", @@ -6334,207 +6259,207 @@ "issue:43819|issue:43927", "issue:36010|issue:43645", "issue:43531|issue:45042", - "issue:43638|issue:44936", "issue:42175|issue:46291", + "issue:43638|issue:44936", "issue:42490|issue:43257", "issue:43720|issue:44610" ], - "prepared_review_unit_hash": "4b884a015de29ab89f0312a8a215e01ccef9cb2ab4407b0f7154239fde6a38f1", + "prepared_review_unit_hash": "ac181fda98c744b47840a70625f63e9ada62b896795779db17107b0c8daad113", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36524, - "estimated_input_tokens": 9131, - "estimated_eval_tokens": 18518 + "serialized_chars": 36528, + "estimated_input_tokens": 9132, + "estimated_eval_tokens": 18520 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "All suggested soft matches are false positives: the paired issues mostly share only broad themes like loading, trainers, or image processors, but they describe different models, different code paths, or different failure modes. None of these pairs look like the same underlying bug/change.", - "confidence": 0.95, - "canonical_issue_reason": null, + "summary": "The cluster is highly heterogeneous: the soft-similarity links mostly connect unrelated model-, tokenizer-, backend-, or training-specific regressions. I would not merge any of the suggested pairs; they do not look like the same underlying bug or change.", + "confidence": 0.81, + "canonical_issue_reason": "issue:45020 is the broadest and most central report in the set ('Recent transformers versions break models using `remote_code`'), so it is the best anchor if one must pick a representative issue. It is still not a duplicate of the others, but it best matches the general theme of upstream loading regressions.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "issue:45020 is the strongest cluster representative because it is a broadly scoped, user-facing regression with enough generality to sit near other loading/import failures, even though the rest of the items are not duplicates of it.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:34634", "right": "issue:36331", "accept": false, - "reason": "Unrelated subsystems: BarkProcessor voice presets vs a Trainer compute_loss signature regression." + "reason": "Different subsystems and failure modes: BarkProcessor voice_preset behavior vs CustomTrainer compute_loss signature mismatch." }, { "left": "issue:45003", "right": "issue:46285", "accept": false, - "reason": "Both touch modeling code, but one is unsafe sys.modules access and the other is CLIP text-model structure mismatch; different bugs." + "reason": "Both involve modeling internals, but one is unsafe sys.modules access and the other is a CLIP wrapper structure inconsistency; not the same bug." }, { "left": "issue:46697", "right": "issue:46728", "accept": false, - "reason": "Both are image-processor counting/padding issues, but for different models and different mechanisms." + "reason": "Both are image-processor sizing issues, but they affect different models and different calculations (padding vs patch-count logic)." }, { "left": "issue:43653", "right": "issue:44912", "accept": false, - "reason": "Tokenizer special-token registration vs MXFP4 quantization loading; no shared bug." + "reason": "Tokenizer special-token registration is unrelated to MXFP4 quantization fallback during loading." }, { - "left": "issue:43262", - "right": "issue:43819", + "left": "issue:43782", + "right": "issue:45325", "accept": false, - "reason": "Audio-related, but one is chat-template sampling-rate defaulting and the other is DAC latent reconstruction/STE behavior." + "reason": "Different VL model bugs: one is a weight_only load error, the other is a RoPE / temporal position scaling issue." }, { - "left": "issue:43782", - "right": "issue:45325", + "left": "issue:43262", + "right": "issue:43819", "accept": false, - "reason": "Both involve Qwen VL models, but one is a from_pretrained weight_only load error and the other is a rope index scaling bug." + "reason": "Audio chat-template sampling-rate defaulting is unrelated to DAC latent reconstruction / STE mismatch." }, { "left": "issue:45081", "right": "issue:45406", "accept": false, - "reason": "Both are processor/tokenizer loading problems, but they fail on different missing attributes and different model families." + "reason": "Both involve Mistral/Gemma processor loading, but one is a regex patch crash and the other is a missing _tokenizer attribute; different code paths." }, { - "left": "issue:45372", - "right": "issue:46326", + "left": "issue:43582", + "right": "issue:44857", "accept": false, - "reason": "One is an import dependency break in mistral_common; the other is continue_final_message behavior for a different tokenizer path." + "reason": "A TypeError in caching_allocator_warmup is not the same issue as LwDetrImageLoss crashing under float16 AMP." }, { - "left": "issue:43582", - "right": "issue:44857", + "left": "issue:45372", + "right": "issue:46326", "accept": false, - "reason": "A platform-specific allocator TypeError and a CUDA AMP loss crash are not the same code-path problem." + "reason": "Importing ReasoningEffort from mistral_common and continue_final_message behavior are distinct processor/tokenizer problems." }, { "left": "issue:44589", "right": "issue:45042", "accept": false, - "reason": "Float8 storage lookup failure vs PIL backend torchvision dependency regression; unrelated." + "reason": "Missing Float8 storage support and an erroneous torchvision dependency are unrelated." }, { "left": "issue:43550", "right": "issue:44912", "accept": false, - "reason": "Different model/runtime bugs: torch.compile with SDPA vs MXFP4 loading fallback." + "reason": "torch.compile with SDPA on Bamba is a different failure mode from MXFP4 quantization fallback on git-oss-20b." }, { "left": "issue:43742", "right": "issue:45020", "accept": false, - "reason": "A specific MobileLLM load key error is not the same as the broad remote_code regression." + "reason": "A key error loading MobileLLM-125M is a specific model/config issue, whereas 45020 is a broader remote_code regression; not the same concrete bug." }, { "left": "issue:43329", "right": "issue:45600", "accept": false, - "reason": "Video multimodal token counting bug vs auto_mappings referencing removed configs; different affected code." + "reason": "Undefined video-token helper variables in multimodal token counting is unrelated to stale auto_mappings references." }, { - "left": "issue:44060", - "right": "issue:44368", + "left": "issue:45084", + "right": "issue:45538", "accept": false, - "reason": "Both mention tied-weights warnings, but they concern different Qwen variants and different causes." + "reason": "A template compilation TypeError and CLIPTokenizer's model_max_length bug are unrelated." }, { - "left": "issue:45084", - "right": "issue:45538", + "left": "issue:44060", + "right": "issue:44368", "accept": false, - "reason": "Template compilation error and CLIPTokenizer max_length issue are unrelated." + "reason": "Both mention tied-weights warnings, but they involve different models, triggers, and warning behavior; not clearly the same underlying defect." }, { "left": "issue:43452", "right": "issue:44991", "accept": false, - "reason": "Both are from_pretrained loading failures, but one is gguf_file handling and the other is a tokenizer regression for a specific model." + "reason": "Both are tokenizer/model loading failures, but one concerns gguf_file handling and the other a specific tokenizer incompatibility for EMBEDDIA/est-roberta." }, { "left": "issue:43526", "right": "issue:43697", "accept": false, - "reason": "One is a BeitImageProcessorFast label-reduction bug; the other is a model output mismatch in RTDetrV2." + "reason": "BeitImageProcessorFast label reduction and RTDetrV2 output divergence are different model-specific bugs." }, { "left": "issue:43257", "right": "issue:43329", "accept": false, - "reason": "Qwen3 MoE loading with accelerate/deepspeed is unrelated to multimodal token counting in video branches." + "reason": "Qwen3 MoE weight conversion with accelerate/deepspeed is unrelated to multimodal token counting for video processors." }, { "left": "issue:43526", "right": "issue:44554", "accept": false, - "reason": "Different domains and failure modes: image-label reduction vs MPS attention correctness." + "reason": "These are unrelated: image-processor label reduction vs an MPS attention correctness issue." }, { "left": "issue:43906", "right": "issue:46209", "accept": false, - "reason": "An isolated reproduction of an older issue is unrelated to FP8 inference failure on B200." + "reason": "An isolated reproduction of an older issue is not the same as an FP8 inference failure on B200; different hardware and bug class." }, { "left": "issue:36010", "right": "issue:36331", "accept": false, - "reason": "Both are API breakages, but import path breakage and Trainer signature change are distinct regressions." + "reason": "ImportError for GenerationMixin and a Trainer signature change are separate breakages." }, { "left": "issue:43819", "right": "issue:43927", "accept": false, - "reason": "DAC latent decoding mismatch vs DiaConfig token-ID persistence; different components." + "reason": "DAC latent mismatch and DiaConfig losing custom token IDs are different save/load vs generation bugs." }, { "left": "issue:36010", "right": "issue:43645", "accept": false, - "reason": "GenerationMixin import failure is unrelated to custom-model initialization in notebooks." + "reason": "One is a transformers import regression; the other is a Jupyter custom-model initialization breakage in Transformers 5.0.0." }, { "left": "issue:43531", "right": "issue:45042", "accept": false, - "reason": "Qwen3-MoE sliding_window bug and PIL/torchvision dependency regression are unrelated." + "reason": "Qwen3-MoE sliding_window behavior is unrelated to PIL backend processors requiring torchvision." }, { - "left": "issue:43638", - "right": "issue:44936", + "left": "issue:42175", + "right": "issue:46291", "accept": false, - "reason": "Training/evaluation failures, but one is a deepspeed zero3 index error and the other is evaluate-after-train behavior; no clear same root cause." + "reason": "Package install/backend dependency omission is not the same as kernels v0.15 requiring version/revision metadata." }, { - "left": "issue:42175", - "right": "issue:46291", + "left": "issue:43638", + "right": "issue:44936", "accept": false, - "reason": "Different packaging/import failures: missing TensorFlow backend in install extras vs kernels version/revision requirement." + "reason": "A deepspeed zero3/non-pretrained Bert indexing error is unrelated to trainer.evaluate() failing after trainer.train()." }, { "left": "issue:42490", "right": "issue:43257", "accept": false, - "reason": "Local-path trailing-slash import warning and Qwen3 MoE accelerate/deepspeed loading are separate loading bugs." + "reason": "Local-path import warning with trailing slash and Qwen3 MoE weight conversion are unrelated." }, { "left": "issue:43720", "right": "issue:44610", "accept": false, - "reason": "Packed-weight unpacking during accelerate loading and OmDet-Turbo processor size mismatch are unrelated." + "reason": "Packed-weight unpacking during accelerate loading is unrelated to OmDet-Turbo processor/model input-size mismatch." } ] }, @@ -6609,9 +6534,9 @@ ], "soft_pairs": [ "issue:39692|issue:44912", - "issue:43746|issue:44811", "issue:43854|issue:44977", "issue:43643|issue:45759", + "issue:43746|issue:44811", "issue:38175|issue:44610", "issue:43257|issue:43696", "issue:43262|issue:44811", @@ -6624,8 +6549,8 @@ "issue:42915|issue:43742", "issue:43749|issue:44756", "issue:34567|issue:40990", - "issue:43650|issue:43906", "issue:44291|issue:45042", + "issue:43650|issue:43906", "issue:42915|issue:44821", "issue:29159|issue:34567", "issue:43606|issue:45859", @@ -6636,196 +6561,196 @@ "issue:43525|issue:46291", "issue:44568|issue:45356" ], - "prepared_review_unit_hash": "e2e421971c3dbc471809f2450382084f156e860c7eae227c3ab9466f61a5b4b4", + "prepared_review_unit_hash": "d27fc01cb6871579ab1f3cc4dd94254c3e2790bbc88cbcbe24ff997e5a6bf86b", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 36014, - "estimated_input_tokens": 9004, - "estimated_eval_tokens": 18264 + "serialized_chars": 36011, + "estimated_input_tokens": 9003, + "estimated_eval_tokens": 18262 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a grab bag of mostly unrelated closed issues across loading, tokenization, multimodal/video, quantization, cache, and CPU-offload regressions. None of the soft pairs look like true duplicates; they share only broad subsystem keywords, not the same concrete bug path.", - "confidence": 0.78, - "canonical_issue_reason": "issue:43749 is the strongest anchor: it has the most discussion/inbound references and describes a concrete, high-impact loading regression with a clearly reproducible scope.", + "summary": "This cluster is mostly heterogeneous; I don\u2019t see any true duplicate pairs among the soft-edge candidates. The only partial theme is multimodal/video regressions, but even there the reports point to different concrete failures.", + "confidence": 0.24, + "canonical_issue_reason": "issue:44479 is the broadest, highest-signal report in the set: a concrete video-input regression affecting multiple Qwen VL/MoE variants, making it the best anchor for the small multimodal/video subtheme.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43749 is the best representative issue for the cluster because it is well-scoped, impactful, and more central than the other candidates; it is also easier to use as a triage anchor than the narrower model-specific reports.", + "best_issue_reason": "Same as canonical: it is the clearest, most representative bug report among otherwise unrelated issues, with a specific regression title and multiple affected model families.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:39692", "right": "issue:44912", "accept": false, - "reason": "Different models and failures: SigLIP2 doc-example errors vs gpt-oss-20b MXFP4 quantization fallback/OOM behavior." - }, - { - "left": "issue:43746", - "right": "issue:44811", - "accept": false, - "reason": "Different code paths: PEFT local-checkpoint loading for GraniteSpeech vs Whisper processor batch_decode skip_special_tokens handling." + "reason": "Different failures: SigLIP2 doc example/model-processor mismatch vs GPT-OSS MXFP4 quantization fallback." }, { "left": "issue:43854", "right": "issue:44977", "accept": false, - "reason": "Different model families and bugs: GLM-4.7-Flash unit-test load failure vs Qwen3.5 flash-attention generation regression." + "reason": "Different model/code paths: GLM-4.7 loading in tests vs Qwen3.5 flash-attention generation." }, { "left": "issue:43643", "right": "issue:45759", "accept": false, - "reason": "Both involve config loading, but one is missing fields with trust_remote_code and the other is text_config unwrapping for composite multimodal configs; not the same bug." + "reason": "Both involve config/loading behavior, but one is missing fields from trust_remote_code and the other is text_config unwrapping for composite Qwen multimodal configs; not the same bug." + }, + { + "left": "issue:43746", + "right": "issue:44811", + "accept": false, + "reason": "Different components: GraniteSpeech PEFT local checkpoint loading vs Whisper batch_decode skip_special_tokens handling." }, { "left": "issue:38175", "right": "issue:44610", "accept": false, - "reason": "Different models and symptoms: siglip2 zero probabilities vs OmDet-Turbo processor/model input-size mismatch." + "reason": "Different symptoms and models: unexpected zero probabilities in SigLIP2 vs OmDet-Turbo processor/model input-size mismatch." }, { "left": "issue:43257", "right": "issue:43696", "accept": false, - "reason": "Qwen3 MOE weight conversion under accelerate+deepspeed is unrelated to GPT-oss-20b CUDA OOM." + "reason": "Qwen3 MoE weight conversion/loading issue vs GPT-oss CUDA OOM; unrelated root causes." }, { "left": "issue:43262", "right": "issue:44811", "accept": false, - "reason": "Audio template sampling-rate defaulting and Whisper decode token filtering are separate processor behaviors." + "reason": "Audio chat-template sampling-rate default vs Whisper batch_decode token-decoding behavior; different bugs." }, { "left": "issue:44387", "right": "issue:44617", "accept": false, - "reason": "Quantization memory regression in Transformers 5.x is not the same as Sam3Video model-specific OOM." + "reason": "Both are OOM-related, but one is a Transformers 5.x int4 reserved-memory regression and the other is a Sam3Video model OOM; not the same code-path problem." }, { "left": "issue:43257", "right": "issue:43931", "accept": false, - "reason": "Different Qwen-related failures: MOE weight conversion vs Qwen3-VL shape mismatch on load." + "reason": "Different Qwen model loading failures: MoE weight conversion under accelerate/deepspeed vs Qwen3-VL weight-shape mismatch." }, { "left": "issue:45459", "right": "issue:45803", "accept": false, - "reason": "Both mention hidden errors, but they are different exception-handling bugs in different components and code paths." + "reason": "Both mention bad exception handling, but they are different functions and different failure modes (protobuf decode masking vs bare except in Fuyu tensor conversion)." }, { "left": "issue:43329", "right": "issue:43450", "accept": false, - "reason": "Undefined helper/variable usage in multimodal token counting is not the same as batched video processor shape regression." + "reason": "Both involve video/multimodal processing, but one is undefined helpers in multimodal token counting and the other is batched video shape handling; insufficient evidence of the same bug." }, { "left": "issue:46421", "right": "issue:46424", "accept": false, - "reason": "Both are cache-related, but one is lazy-init recompilation and the other is max_cache_length applicability." + "reason": "Both are cache-related, but one is lazy-init recompilation and the other is max_cache_length being ignored except for Quantized Cache." }, { "left": "issue:43329", "right": "issue:44479", "accept": false, - "reason": "The first is an undefined-name bug in token counting; the second is a broader video-input regression across Qwen VL variants." + "reason": "Related to video inputs, but one is undefined helper variables in _get_num_multimodal_tokens and the other is a broader v5.3.0 regression across several Qwen VL models." }, { "left": "issue:42915", "right": "issue:43742", "accept": false, - "reason": "Different failures: FineGrainedFP8Config handling for Qwen3Moe vs MobileLLM-125M load key error." + "reason": "Different failures: FineGrainedFP8Config support in Qwen3MoE vs key error loading MobileLLM-125M." }, { "left": "issue:43749", "right": "issue:44756", "accept": false, - "reason": "Both are loading/memory issues, but one is FSDP CPU RAM efficient loading and the other is mmap behavior on Strix Halo." + "reason": "Different memory/loading issues: FSDP CPU RAM efficient loading vs mmap on Strix Halo causing OOM." }, { "left": "issue:34567", "right": "issue:40990", "accept": false, - "reason": "TrainerState token counters not updating is unrelated to GPT-oss-20b perplexity on WikiText-2." + "reason": "Unrelated training-state counter bug vs high perplexity on GPT-oss-20b." }, { - "left": "issue:43650", - "right": "issue:43906", + "left": "issue:44291", + "right": "issue:45042", "accept": false, - "reason": "No shared bug: placeholder issue text versus an isolated reproduction of a different upstream issue." + "reason": "Different unrelated regressions: init_empty_weights/_is_hf_initialized vs PIL image processors requiring torchvision." }, { - "left": "issue:44291", - "right": "issue:45042", + "left": "issue:43650", + "right": "issue:43906", "accept": false, - "reason": "Model init_empty_weights argument handling is unrelated to PIL image processor torchvision dependency regression." + "reason": "No clear overlap: one is an ambiguous 'ADD THE DATA' issue and the other is a reproduction of a different upstream bug." }, { "left": "issue:42915", "right": "issue:44821", "accept": false, - "reason": "GPT-oss-20b loading/quantization issues do not match AutoImageProcessor URL-loading failure." + "reason": "Different areas: Qwen3MoE FP8 config vs AutoImageProcessor loading from URL." }, { "left": "issue:29159", "right": "issue:34567", "accept": false, - "reason": "Tokenizer slow/fast inconsistencies are unrelated to TrainerState token-count tracking." + "reason": "Tokenizer slow/fast inconsistency is unrelated to TrainerState.num_input_tokens_seen not updating." }, { "left": "issue:43606", "right": "issue:45859", "accept": false, - "reason": "Both mention CPU offload, but they affect different models and likely different offload paths." + "reason": "Both mention CPU offload, but they affect different models/components and don\u2019t look like the same underlying fix." }, { "left": "issue:43927", "right": "issue:44912", "accept": false, - "reason": "DiaConfig save/load token-ID loss is unrelated to GPT-oss-20b MXFP4 quantization fallback." + "reason": "Different bugs: DiaConfig token IDs lost on save/load vs GPT-oss MXFP4 quantization fallback." }, { "left": "issue:43540", "right": "issue:46701", "accept": false, - "reason": "Different modalities and failures: Qwen3OmniMoe video-input ValueError vs DiffusionGemma pipeline output issue." + "reason": "Both are multimodal inference issues, but one is Qwen3OmniMoe video-input processing and the other is DiffusionGemma image-text-to-text pipeline behavior." }, { "left": "issue:28282", "right": "issue:29870", "accept": false, - "reason": "PyTorch-missing ImportError and Bart generate() decoder_attention_mask UnboundLocalError are unrelated." + "reason": "Completely different problems: missing PyTorch ImportError vs Bart generate() unbound decoder_attention_mask." }, { "left": "issue:43454", "right": "issue:43577", "accept": false, - "reason": "AyaVision weight-tying causing garbage text and Blip2 dtype propagation are different model-loading bugs." + "reason": "Different model bugs: AyaVision weight tying vs Blip2 dtype propagation." }, { "left": "issue:43525", "right": "issue:46291", "accept": false, - "reason": "Missing pad_token_id in Llama4Config is unrelated to pipeline import failing on kernels version/revision requirements." + "reason": "Unrelated: missing pad_token_id on Llama4Config vs kernels dependency/version import failure." }, { "left": "issue:44568", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer regressions, but they concern different tokenizer implementations, models, and failure modes." + "reason": "Both are tokenizer regressions, but they affect different models and different tokenization/codecs behavior." } ] }, @@ -6919,29 +6844,29 @@ "issue:42915|issue:43749", "issue:44849|issue:45701", "issue:43452|issue:43582", - "issue:27957|issue:28282", "issue:24643|issue:29159", + "issue:27957|issue:28282", "issue:44987|issue:45907", "issue:41720|issue:43540", "issue:42175|issue:44488", - "issue:44568|issue:45488", "issue:43540|issue:44560", + "issue:44568|issue:45488", "issue:44792|issue:44964" ], - "prepared_review_unit_hash": "1102b6c8863cab3a7efd83da51ccf3c9883996c7d3cb5a6ce50845665e7c6267", + "prepared_review_unit_hash": "f49754b0ed942496412476e5df6819e66bafa067b27eec344e56f4870677dd70", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36020, + "serialized_chars": 36019, "estimated_input_tokens": 9005, "estimated_eval_tokens": 18266 }, @@ -6950,174 +6875,180 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly a set of unrelated issue candidates that only share broad Transformers theme words. I found no strong duplicate PRs, and only one possible soft duplicate is still too weak to accept under strict triage.", - "confidence": 0.91, - "canonical_issue_reason": "issue:45488 is the clearest, most actionable report: it is a concrete regression with a precise title, active discussion, and inbound references, making it a good cluster representative.", + "summary": "This cluster is a broad mix of mostly unrelated Transformers issues; many are in adjacent areas like tokenization, model loading, vision/video regressions, and dtype handling, but the soft pairs do not look like true duplicates of the same concrete bug.", + "confidence": 0.88, + "canonical_issue_reason": "issue:45488 is the clearest, most central representative of the tokenizer-behavior regression theme in this cluster and has the strongest cross-model impact.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45488 is the strongest overall issue to keep as the canonical example because it describes a specific user-facing regression with enough detail to triage and verify.", + "best_issue_reason": "issue:45488 is the best global issue candidate because it is specific enough to be actionable yet broad enough to represent the recurring tokenizer/version regression pattern seen here.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44987", "right": "issue:45538", "accept": false, - "reason": "Different problems: one is a model-loading failure for a specific repo, the other is a CLIPTokenizer max-length bug." + "reason": "Different bugs: one is a model load failure for a specific repo, the other is CLIPTokenizer max_length behavior." }, { "left": "issue:28282", "right": "issue:34689", "accept": false, - "reason": "Generic missing-PyTorch import error vs a specific Llama 3.2 vision model-loading regression; not the same bug." + "reason": "Unrelated failure modes: missing PyTorch import vs a Llama Vision model-loading regression." }, { "left": "issue:45538", "right": "issue:46869", "accept": false, - "reason": "Tokenizer configuration bug vs multi-GPU device-map failure; unrelated failure modes." + "reason": "Tokenizer max_length bug vs YOLO auto device-map failure are unrelated." }, { "left": "issue:44857", "right": "issue:46832", "accept": false, - "reason": "Both are detection-model issues, but one is AMP/CUDA crash in loss and the other is a forward shape/type error; different code paths." + "reason": "Both are object-detection related, but the concrete code paths differ: loss/AMP crash vs forward shape/feature-level mismatch." }, { "left": "issue:43697", "right": "issue:45357", "accept": false, - "reason": "RT-DETR output regression vs Qwen3.5 save_pretrained key mismatch; different subsystems and symptoms." + "reason": "Different models and symptoms: RT-DETR output drift vs Qwen3.5 save_pretrained key naming." }, { "left": "issue:43425", "right": "issue:46291", "accept": false, - "reason": "Torch compatibility complaint vs kernels package revision/version requirement; unrelated dependency issues." + "reason": "Separate dependency issues: Torch compatibility vs kernels revision/version requirement." }, { "left": "issue:33357", "right": "issue:34567", "accept": false, - "reason": "MacOS bus error in CLIP model loading vs TrainerState token counter not updating; no shared underlying bug." + "reason": "Mac bus error on CLIP model load is unrelated to TrainerState token counting." }, { "left": "issue:43404", "right": "issue:45127", "accept": false, - "reason": "Both mention tied weights, but one is a missing tie in a specific model class and the other is LoRA merge corruption with extended vocab; not the same fix." + "reason": "Weight-tying bug in Mistral3 is not the same as LoRA merge collapse with extended vocab." }, { "left": "issue:42915", "right": "issue:45538", "accept": false, - "reason": "FineGrainedFP8Config failure in Qwen3Moe vs CLIPTokenizer max_length bug; unrelated." + "reason": "FP8 config failure for Qwen3Moe is unrelated to CLIPTokenizer max_length." }, { "left": "issue:45137", "right": "issue:45405", "accept": false, - "reason": "DeepSpeed ZeRO3 deque crash vs PEFT version bump/release mismatch; completely different issues." + "reason": "DeepSpeed deque error and unreleased PEFT version bump are unrelated." }, { "left": "issue:42175", "right": "issue:43824", "accept": false, - "reason": "Packaging/backend install problem vs missing Qwen2.5-VL symbol import; different dependency layers." + "reason": "Packaging/backend installation issue vs missing model class import are different problems." }, { "left": "issue:28282", "right": "issue:29127", "accept": false, - "reason": "Missing PyTorch import error vs LayoutLMv3 error-message clarity issue; not duplicates." + "reason": "Missing PyTorch import and LayoutLMv3 box-validation messaging are not the same bug." }, { "left": "issue:44062", "right": "issue:44991", "accept": false, - "reason": "These are both tokenizer-loading failures, but the evidence is too broad to confirm the same root cause from the titles alone." + "reason": "AddedToken keyword collision is not evidenced as the same issue as est-roberta tokenizer loading." }, { "left": "issue:44479", "right": "issue:46693", "accept": false, - "reason": "Video-input regression for Qwen models vs FA2 long-context performance regression; different symptoms and likely different code paths." + "reason": "Video-input regression and FA2 long-context performance regression are distinct changes." }, { "left": "issue:43742", "right": "issue:44291", "accept": false, - "reason": "MobileLLM key error vs unexpected _is_hf_initialized argument during init_empty_weights; both are load-time errors but not the same bug." + "reason": "MobileLLM key error and init_empty_weights argument mismatch are different loading failures." }, { "left": "issue:44936", "right": "issue:46821", "accept": false, - "reason": "Trainer evaluate-after-train failure vs TimesFM window_size attribute error; unrelated components." + "reason": "Trainer evaluate/train state issue is unrelated to TimesFM window_size attribute handling." }, { "left": "issue:43577", "right": "issue:46512", "accept": false, - "reason": "Model/QFormer dtype staying float32 vs stale config.dtype after from_config; related theme but different bugs and fixes." + "reason": "Both involve dtype consistency, but one is submodule weights staying float32 while the other is stale config dtype; not the same concrete bug." }, { "left": "issue:42915", "right": "issue:43749", "accept": false, - "reason": "Qwen3Moe FP8 config failure vs FSDP CPU RAM efficient loading breakage; different features." + "reason": "Qwen3Moe FP8 failure and FSDP CPU RAM efficient loading are unrelated." }, { "left": "issue:44849", "right": "issue:45701", "accept": false, - "reason": "Qwen3.5 output_hidden_states bug vs version-dependent tokenization changes; unrelated." + "reason": "Qwen3.5 hidden-state bug is not the same as version-dependent tokenization changes." }, { "left": "issue:43452", "right": "issue:43582", "accept": false, - "reason": "gguf_file loader breakage vs Apple Silicon allocator TypeError; different areas." + "reason": "gguf_file loading breakage and Apple Silicon allocator TypeError are unrelated." }, { - "left": "issue:27957", - "right": "issue:28282", + "left": "issue:24643", + "right": "issue:29159", "accept": false, - "reason": "Flash Attention 2 issue in XLMRoberta vs missing-PyTorch import error; unrelated." + "reason": "DeepSpeed weight-dimension runtime error and tokenizer slow/fast inconsistency are different subsystems." }, { - "left": "issue:24643", - "right": "issue:29159", + "left": "issue:27957", + "right": "issue:28282", "accept": false, - "reason": "DeepSpeed training RuntimeError vs tokenizer slow/fast inconsistency; no shared code-path problem." + "reason": "Flash Attention 2 with XLM-Roberta is unrelated to missing PyTorch import." }, { "left": "issue:44987", "right": "issue:45907", "accept": false, - "reason": "Model-loading regression vs list_repo_files revision-forwarding bug; not duplicates." + "reason": "Model loading failure for a specific repo and list_repo_files revision forwarding are different bugs." }, { "left": "issue:41720", "right": "issue:43540", "accept": false, - "reason": "Auto device-mapping CUDA assert vs video-input processing ValueError in Qwen3OmniMoe; distinct failures." + "reason": "Auto device-map CUDA assert for Qwen3 and video-input processing error for Qwen3OmniMoe are not the same issue." }, { "left": "issue:42175", "right": "issue:44488", "accept": false, - "reason": "Backend installation packaging issue vs a model-loading failure for a specific repository; different problems." + "reason": "Backend install packaging bug vs model-loading issue for cjvt/sleng-bert are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44560", + "accept": false, + "reason": "Both mention video, but they affect different Qwen models and failure points; too broad to merge." }, { "left": "issue:44568", "right": "issue:45488", "accept": false, - "reason": "Both are tokenizer regressions, but one is BOS/EOS insertion behavior and the other is ByteLevel pre-tokenizer being overridden; not the same bug." + "reason": "Different tokenizer regressions: special-token addition vs pre-tokenizer override causing DeepSeek breakage." }, { "left": "issue:44792", "right": "issue:44964", "accept": false, - "reason": "Janus image-generation test failure vs Phi-4 multimodal model-loading issue; different models and failure points." + "reason": "Janus test failure and Phi-4 multimodal loading failure are distinct multimodal issues." } ] }, @@ -7199,8 +7130,8 @@ "issue:44062|issue:45081", "issue:45042|issue:45081", "issue:43582|issue:46821", - "issue:43742|issue:44589", "issue:45672|issue:45674", + "issue:43742|issue:44589", "issue:43756|issue:46399", "issue:43756|issue:45325", "issue:44568|issue:46491", @@ -7215,205 +7146,205 @@ "issue:45823|issue:46463", "issue:43452|issue:44488", "issue:43697|issue:46829", - "issue:45357|issue:45920", - "issue:44625|issue:44877" + "issue:44625|issue:44877", + "issue:45357|issue:45920" ], - "prepared_review_unit_hash": "54d75f45b823f8e8382284e7b026704fb9f63638c384c64c6d9a5d9a768355a5", + "prepared_review_unit_hash": "1338f770209de7cc1ef8462118c4b0a36c595467f89931a000f7e54008e4b78c", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 27, - "serialized_chars": 35594, - "estimated_input_tokens": 8899, - "estimated_eval_tokens": 18054 + "serialized_chars": 35602, + "estimated_input_tokens": 8901, + "estimated_eval_tokens": 18058 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most pairs are clearly unrelated. The only plausible duplicate edge is the Qwen video-input regression pair (44479 \u2194 45381), which appears to describe the same underlying position-id bug across related Qwen VL variants.", - "confidence": 0.74, - "canonical_issue_reason": "issue:44479 is the best canonical issue because it is the broadest, most central report of the Qwen video-input regression and already spans multiple affected Qwen variants.", + "summary": "Only one soft pair looks like the same underlying bug: the Qwen2.5-VL/Qwen3-VL video position-id regression. The rest are mostly topic-level similarity (tokenizers, dtype, multimodal, training, loading) but appear to be different concrete failures.", + "confidence": 0.79, + "canonical_issue_reason": "issue:44479 is the best issue anchor because it describes the broad video-input regression and is the more general formulation of the only likely duplicate pair; issue:45381 looks like a narrower reproduction of the same problem.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44479 is the strongest representative issue in this cluster: it describes a concrete regression with multiple affected models and is the most likely target for consolidation.", + "best_issue_reason": "issue:44479 is the strongest representative issue in the cluster: it is broader, more user-facing, and closest to the one plausible duplicate relationship in the set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44479", "right": "issue:45381", "accept": true, - "reason": "Both report the same Qwen VL video-input regression around wrong position ids / temporal scaling in v5.3.0; 45381 looks like a narrower manifestation of the broader bug in 44479." + "reason": "Both report the same Qwen2.5-VL/Qwen3-VL video-input position-id regression; 45381 looks like a narrower reproduction of the broader 44479 bug." }, { "left": "issue:38175", "right": "issue:44898", "accept": false, - "reason": "Different models and different failure modes: unexpected zero probabilities in SigLIP2 vs Perceiver non-default-resolution failure." + "reason": "Both are vision-model correctness issues, but they involve different models and failure modes (SigLIP zero probabilities vs Perceiver resizing/interpolation)." }, { "left": "issue:44291", "right": "issue:44821", "accept": false, - "reason": "Unrelated load-time errors: unexpected _is_hf_initialized argument vs AutoImageProcessor URL loading." + "reason": "Different loading failures: one is an init_empty_weights/_is_hf_initialized argument mismatch, the other is loading AutoImageProcessor from a URL." }, { "left": "issue:43540", "right": "issue:44479", "accept": false, - "reason": "Both involve video inputs, but they affect different Qwen model families and do not look like the same underlying bug." + "reason": "Both involve video inputs, but they target different model families and likely different code paths (Qwen3OmniMoe vs Qwen2.5-VL/Qwen3-VL)." }, { "left": "issue:34567", "right": "issue:36331", "accept": false, - "reason": "TrainerState token-count tracking and CustomTrainer compute_loss signature incompatibility are separate training issues." + "reason": "Both touch Trainer behavior, but one is token accounting and the other is a compute_loss API signature break." }, { "left": "issue:43257", "right": "issue:43756", "accept": false, - "reason": "Different subsystems: DeepSpeed/Accelerate MoE weight conversion vs SmolLM3 RoPE layer count regression." + "reason": "Unrelated model-loading/weight-conversion issue versus a RoPE-layer-count bug in Smollm3." }, { "left": "issue:44062", "right": "issue:45081", "accept": false, - "reason": "Tokenizer AddedToken construction error is unrelated to the Mistral regex patch AttributeError." + "reason": "Both are tokenizer bugs, but the concrete failures differ: AddedToken keyword collision vs Mistral regex patch accessing backend_tokenizer." }, { "left": "issue:45042", "right": "issue:45081", "accept": false, - "reason": "Image processor torchvision dependency regression is unrelated to tokenizer regex patching." + "reason": "Completely different subsystems: PIL image processors requiring torchvision vs a Mistral tokenizer regex patch crash." }, { "left": "issue:43582", "right": "issue:46821", "accept": false, - "reason": "Apple Silicon allocator TypeError and TimesFM window_size AttributeError are unrelated." + "reason": "Apple Silicon allocator warmup TypeError is unrelated to the TimesFM window_size AttributeError." }, { - "left": "issue:43742", - "right": "issue:44589", + "left": "issue:45672", + "right": "issue:45674", "accept": false, - "reason": "MobileLLM key error and Float8 storage lookup failure are different loading problems." + "reason": "Both are 4-bit inference problems, but one is a uint8 torch.finfo() error in Gemma4 audio modules and the other is a BitsAndBytesConfig skip-module regression." }, { - "left": "issue:45672", - "right": "issue:45674", + "left": "issue:43742", + "right": "issue:44589", "accept": false, - "reason": "Both mention 4-bit inference, but one is a torch.finfo uint8 audio-module error and the other is a BitsAndBytesConfig skip-module exclusion bug." + "reason": "Different model-load failures: MobileLLM key error vs missing Float8 storage type." }, { "left": "issue:43756", "right": "issue:46399", "accept": false, - "reason": "SmolLM3 RoPE-layer issue and Qwen3.5 kernel-registration crash are unrelated." + "reason": "Both mention RoPE/rotary concepts, but one is a missing RoPE layers bug and the other is a kernel registration decorator crash." }, { "left": "issue:43756", "right": "issue:45325", "accept": false, - "reason": "Both mention RoPE, but the affected models and concrete failures are different." + "reason": "Different rotary-position bugs in different models: missing RoPE layers vs incorrect still-image temporal scaling in Qwen2.5-VL." }, { "left": "issue:44568", "right": "issue:46491", "accept": false, - "reason": "Tokenizer special-token insertion regression and CodeLlama round-trip spacing regression are different tokenizer bugs." + "reason": "Different tokenizer regressions: missing BOS/EOS insertion vs leading-space stripping on encode/decode." }, { "left": "issue:45405", "right": "issue:45600", "accept": false, - "reason": "Released-version requirement mismatch and removed-config reference in auto_mappings are unrelated CI/release issues." + "reason": "A dependency-version bump issue and a removed-config reference in auto_mappings are not the same bug." }, { "left": "issue:43716", "right": "issue:46701", "accept": false, - "reason": "Different models and pipelines: Mistral-3 dtype mismatch vs DiffusionGemma image-text-to-text pipeline failure." + "reason": "Both are multimodal, but one is a dtype mismatch between preprocessor and model; the other is a pipeline/output wiring failure." }, { "left": "issue:43577", "right": "issue:45072", "accept": false, - "reason": "Both are dtype-related, but they concern different models/code paths and do not look mergeable as one fix." + "reason": "Both mention dtype mismatches, but they affect different architectures and likely different conversion/loading paths." }, { "left": "issue:43531", "right": "issue:44821", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior is unrelated to AutoImageProcessor URL loading." + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to AutoImageProcessor URL loading." }, { "left": "issue:43906", "right": "issue:46468", "accept": false, - "reason": "An isolated reproduction of a prior issue is not the same as the Nemotron example failure." + "reason": "One is only an isolated reproduction of issue 38071; the other is a broken official Nemotron example. Not enough evidence they are the same underlying bug." }, { "left": "issue:24643", "right": "issue:29870", "accept": false, - "reason": "DeepSpeed weight-dimension error and BART generate decoder mask initialization error are unrelated." + "reason": "Training-with-Deepspeed RuntimeError and BART generate() UnboundLocalError are unrelated failures." }, { "left": "issue:44737", "right": "issue:46463", "accept": false, - "reason": "XLNet CPU-side arange bug and T5 decoder-input validation messaging are unrelated." + "reason": "XLNet CPU placement bug and T5 decoder-input validation error are different code paths and symptoms." }, { "left": "issue:43906", "right": "issue:46693", "accept": false, - "reason": "Prior-issue reproduction and FA2 long-context performance regression are unrelated." + "reason": "Isolated reproduction of an older issue is not the same as a FA2 long-context performance regression." }, { "left": "issue:45823", "right": "issue:46463", "accept": false, - "reason": "Gemma4 device mismatch and T5 decoder validation message are unrelated." + "reason": "Gemma4 device-map mismatch is unrelated to T5 decoder input validation." }, { "left": "issue:43452", "right": "issue:44488", "accept": false, - "reason": "gguf_file loading breakage and sleng-bert load failure are different loader problems." + "reason": "Both are model-loading complaints, but one is specifically about gguf_file support and the other is a separate sleng-bert loading failure." }, { "left": "issue:43697", "right": "issue:46829", "accept": false, - "reason": "RTDetrV2 output drift and VideoPrism hidden_states field mix-up are different model-output bugs." + "reason": "Different model families and outputs: RTDetrV2 forward differences vs VideoPrism output field misassignment." }, { - "left": "issue:45357", - "right": "issue:45920", + "left": "issue:44625", + "right": "issue:44877", "accept": false, - "reason": "Qwen3.5 save_pretrained key regression and tokenizer token-id regression affect different code paths." + "reason": "Different config issues: Qwen3.5 num_labels propagation vs strict config blocking granite_speech loading." }, { - "left": "issue:44625", - "right": "issue:44877", + "left": "issue:45357", + "right": "issue:45920", "accept": false, - "reason": "Qwen3.5 num_labels propagation and granite_speech strict-config loading are separate config issues." + "reason": "One is a Qwen3.5 save_pretrained visual-encoder key regression; the other is a tokenizer ID regression across several models." } ] }, "evaluator_result": { "accept": true, - "feedback": "The cluster summary is mostly grounded in the packet, and the only accepted soft edge (44479 \u2194 45381) is a conservative duplicate call based on strong title overlap within the same Qwen video-input regression family. I did not see evidence of overstated mergeability for any other pair." + "feedback": "The summary is grounded in the packet and stays conservative. The accepted pair (issue:44479 / issue:45381) is well-supported by the titles and scope difference, and the rejected pairs are appropriately framed as topic similarity rather than the same bug. No overstatement stands out." }, "evaluator_used": true, "retried": false, @@ -7502,8 +7433,8 @@ "issue:43653|issue:44361", "issue:41762|issue:44936", "issue:43452|issue:44843", - "issue:43688|issue:43749", "issue:43819|issue:44811", + "issue:43688|issue:43749", "issue:38175|issue:43295", "issue:43994|issue:44898", "issue:43525|issue:46701", @@ -7511,190 +7442,190 @@ "issue:45542|issue:46178", "issue:43295|issue:45356" ], - "prepared_review_unit_hash": "81b8c520fc3e77676de82ba99032f07922ccf9722ff4b9cb0e5d595da142b8ae", + "prepared_review_unit_hash": "76a9d14727f863f01a171768c685cf57f9c0e3e4349e7595346fba0bb2df2293", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 26, - "serialized_chars": 36202, - "estimated_input_tokens": 9051, - "estimated_eval_tokens": 18358 + "serialized_chars": 36196, + "estimated_input_tokens": 9049, + "estimated_eval_tokens": 18354 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab-bag of unrelated bug reports. The soft pairs share superficial vocabulary (loading, tokenizer, config, vision, etc.) but not the same concrete code path, so none look safe to merge as duplicates.", - "confidence": 0.94, - "canonical_issue_reason": "Issue 43749 is the broadest and most discussed loading/regression report here, so it is the closest thing to a cluster anchor, but the set is not a true duplicate cluster.", + "summary": "The cluster is highly heterogeneous: it mixes unrelated bugs in tokenizers, model configs, generation, quantization, distributed loading, and backend detection. None of the soft-edge pairs look like the same underlying bug or change, so all are rejected as duplicates.", + "confidence": 0.92, + "canonical_issue_reason": "Issue 45005 is the broadest and most representative entry here: it describes a v5 tied-weights regression with wider model-family impact, making it the best umbrella candidate despite the cluster\u2019s overall heterogeneity.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43749 has the strongest activity and broad applicability among the items, making it the best representative issue in this mixed cluster.", + "best_issue_reason": "Issue 45005 is the clearest and most general issue in the set, with a concrete regression description and broad applicability across translation/tied-weight models.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45663", "right": "issue:46846", "accept": false, - "reason": "Different models and failures: Gemma-4 FSDP2 KeyError vs Qwen3.5 TP-plan OOM/shape error. Only both are distributed-training themed." + "reason": "Different model families and failure modes: Gemma4 FSDP2 layer state bug vs Qwen3.5 TP plan OOM/shape error." }, { "left": "issue:43540", "right": "issue:46693", "accept": false, - "reason": "One is a video-input validation error in Qwen3OmniMoe; the other is a long-context FlashAttention performance regression. Different bug classes." + "reason": "Unrelated code paths: Qwen3OmniMoe video-input validation vs FA2 long-context performance regression." }, { "left": "issue:44913", "right": "issue:46121", "accept": false, - "reason": "Both touch RoPE-related code, but one is config reload persistence and the other is JSON-loaded list handling in validation. Not the same defect." + "reason": "Both touch RoPE/rotation config handling, but one is reload persistence of a config field and the other is validation/type conversion for a JSON-loaded list." }, { "left": "issue:44737", "right": "issue:45588", "accept": false, - "reason": "XLNet CPU placement bug and flash-attention None crash are unrelated code paths and symptoms." + "reason": "XLNet CPU device placement bug and flash_attention crash on missing s_aux are unrelated implementation defects." }, { "left": "issue:28282", "right": "issue:32090", "accept": false, - "reason": "ImportError for missing PyTorch and Trainer GPU broadcast TypeError are separate environment/runtime issues." + "reason": "Import-time PyTorch missing error vs Trainer GPU broadcast TypeError; completely different surfaces and causes." }, { "left": "issue:46463", "right": "issue:46899", "accept": false, - "reason": "T5 decoder-input error-message issue vs Gemma-4 4-bit quantization dtype cast bug. Different models and root causes." + "reason": "T5 decoder-input validation error message vs Gemma4 multimodal dtype-cast quantization bug; not the same problem." }, { "left": "issue:43697", "right": "issue:45216", "accept": false, - "reason": "Both are regressions, but one is RTDetrV2 output mismatch and the other is Qwen3.5 checkpoint saving corruption. Not the same code path." + "reason": "RTDetrV2 output drift in v5 and Qwen3.5 save_pretrained regression are different models and code paths." }, { "left": "issue:43526", "right": "issue:45479", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug is unrelated to the sequence-classification zero-loss problem." + "reason": "BeitImageProcessorFast label reduction bug is unrelated to the single-label-classification zero-loss issue." }, { "left": "issue:43720", "right": "issue:44898", "accept": false, - "reason": "BitNet packed-weight loading and Perceiver interpolation failure are different architectures and loading/inference paths." + "reason": "BitNet packed-weight loading bug vs Perceiver non-default-resolution interpolation failure; different subsystems and symptoms." }, { "left": "issue:43643", "right": "issue:45375", "accept": false, - "reason": "Both mention missing fields, but one is generic AutoConfig with trust_remote_code and the other is a specific Qwen3_5MoeVisionConfig strict-field drop." + "reason": "Generic trust_remote_code config reconstruction issue vs a specific Qwen3_5MoeVisionConfig field being dropped; related only at a very high level." }, { "left": "issue:43761", "right": "issue:46829", "accept": false, - "reason": "Both involve output fields, but CLIPVisionModel hidden_states regression and VideoPrism output mis-assignment are separate model-specific bugs." + "reason": "Both concern output fields, but CLIPVision hidden_states regression and VideoPrism hidden_states misassignment are different model implementations." }, { "left": "issue:45200", "right": "issue:46463", "accept": false, - "reason": "Gemma 4 mm_token_type_ids defaulting and T5 decoder-input validation messaging are unrelated." + "reason": "Gemma4 missing default mm_token_type_ids and T5 decoder-input error messaging are unrelated bugs." }, { "left": "issue:44466", "right": "issue:45005", "accept": false, - "reason": "Both relate loosely to tied weights, but one is device-dependent serialization and the other is translation-model tied-weight handling. Too broad to merge." + "reason": "Both mention tied weights, but one is device-dependent lm_head serialization while the other is a broader translation-model regression without the same concrete failure mode." }, { "left": "issue:44361", "right": "issue:44442", "accept": false, - "reason": "Different tokenizers and failures: MLukeTokenizer task AttributeError vs AutoTokenizer loading FastSpeech2ConformerTokenizer." + "reason": "Two different tokenizer load/task failures for different tokenizers; same broad area, not the same bug." }, { "left": "issue:44493", "right": "issue:46050", "accept": false, - "reason": "UNEXPECTED Key with position id and sinq quantized save/load breakage are distinct loading/serialization problems." + "reason": "Position-id key warnings and sinq save/load breakage are distinct model-state issues." }, { "left": "issue:43653", "right": "issue:44361", "accept": false, - "reason": "BigBirdTokenizer special-token decode bug is unrelated to MLukeTokenizer task AttributeError." + "reason": "BigBird mask-token special-token registration and MLuke task AttributeError are unrelated tokenizer defects." }, { "left": "issue:41762", "right": "issue:44936", "accept": false, - "reason": "Gemma3 ZeRO-3 load-time IndexError and trainer.evaluate() after train() are not the same failure mode or subsystem." + "reason": "Gemma3 ZeRO-3 load-time index error and trainer.evaluate() after train() failure have no shared concrete code path." }, { "left": "issue:43452", "right": "issue:44843", "accept": false, - "reason": "Both affect from_pretrained, but gguf_file support breakage and offline model_info calls are different concrete bugs." + "reason": "GGUF from_pretrained handling and an offline model_info call in mistral regex patch are different loading bugs." }, { - "left": "issue:43688", - "right": "issue:43749", + "left": "issue:43819", + "right": "issue:44811", "accept": false, - "reason": "Aux-loss normalization in OLMoE/GPT Oss is unrelated to FSDP CPU RAM-efficient loading." + "reason": "DAC latent conversion mismatch and Whisper batch_decode skip_special_tokens handling are unrelated audio/model API issues." }, { - "left": "issue:43819", - "right": "issue:44811", + "left": "issue:43688", + "right": "issue:43749", "accept": false, - "reason": "DAC latent reconstruction mismatch and Whisper skip_special_tokens decode behavior are different model APIs and bugs." + "reason": "Aux-loss normalization in MoE models vs FSDP CPU RAM efficient loading failure are different problems." }, { "left": "issue:38175", "right": "issue:43295", "accept": false, - "reason": "Siglip2 zero-probabilities and processor.tokenizer/image forwarding regression are unrelated issues." + "reason": "Siglip2 zero probabilities and processor.tokenizer regression are separate multimodal issues." }, { "left": "issue:43994", "right": "issue:44898", "accept": false, - "reason": "Both are vision-related, but Siglip2 nonsensical outputs and Perceiver interpolation failure are different models and root causes." + "reason": "Siglip2 nonsensical outputs and Perceiver interpolation failure are different model-specific regressions." }, { "left": "issue:43525", "right": "issue:46701", "accept": false, - "reason": "Llama4Config pad_token_id missing and DiffusionGemma image-text-to-text pipeline breakage are unrelated." + "reason": "Llama4Config missing pad_token_id and DiffusionGemma image-text pipeline breakage are unrelated." }, { "left": "issue:45372", "right": "issue:46170", "accept": false, - "reason": "Gemma4 processor import failure via mistral_common and Windows ROCm torch.distributed dependency issues are separate dependency problems." + "reason": "Gemma4 processor import dependency issue and Windows ROCm torch.distributed dependency issue are different environment/import failures." }, { "left": "issue:45542", "right": "issue:46178", "accept": false, - "reason": "Both concern backend availability, but tensorboard-without-TF vs tensorflow_text backend lookup are not the same bug." + "reason": "TensorFlow backend detection error and tensorflow_text backend mapping error are separate backend-registration problems." }, { "left": "issue:43295", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer regressions, but the reported behaviors and implicated code paths are different." + "reason": "Processor.tokenizer regression and Kimi-K2.5 codec/fix_mistral_regex warning regression do not share the same concrete bug." } ] }, @@ -7774,8 +7705,8 @@ "issue:43646|issue:43749", "issue:43452|issue:43577", "issue:43756|issue:46402", - "issue:43688|issue:45704", "issue:44843|issue:44991", + "issue:43688|issue:45704", "issue:43653|issue:43927", "issue:44464|issue:44743", "issue:43688|issue:45701", @@ -7789,131 +7720,131 @@ "issue:44734|issue:46612", "issue:44466|issue:46050", "issue:43504|issue:43720", - "issue:43504|issue:43550", "issue:44977|issue:46693", + "issue:43504|issue:43550", "issue:44464|issue:44977", "issue:41720|issue:44464", "issue:43643|issue:44561", "issue:43577|issue:44554" ], - "prepared_review_unit_hash": "dcec38aa04d8b0b0ac8eab02d339ccfa4559e434202d7c224374ba11a4f039b9", + "prepared_review_unit_hash": "5a5fbcedb9929434bb86ede3e43c9d1fd73d03fe8dd9c41c9ebc890ce8b0547b", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 35538, - "estimated_input_tokens": 8885, - "estimated_eval_tokens": 18026 + "serialized_chars": 35543, + "estimated_input_tokens": 8886, + "estimated_eval_tokens": 18028 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is a loose, heterogeneous set of Transformers issues: tokenization regressions, model-loading failures, generation/cache bugs, and a few unrelated performance/config problems. None of the soft pairs look like the same concrete bug or change.", - "confidence": 0.91, - "canonical_issue_reason": "If a representative issue is needed, issue 44464 is the best anchor: it is a concrete, reproducible generation regression and sits closest to several other generation/cache-related soft matches in this set.", + "summary": "This cluster is heterogeneous: the soft pairs span different models, APIs, and failure modes, so none look like true duplicates. The issues are mostly separate regressions in tokenization, loading, generation, config propagation, and attention paths.", + "confidence": 0.66, + "canonical_issue_reason": "Issue 43646 is the broadest and most representative regression in the set: a clear Transformers 5.0 model-initialization break with a concrete code path, making it the best single issue to anchor nearby reports.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44464 is the most usable representative because it is specific, actionable, and broadly similar in theme to the other generation-related items without being overly model-specific.", + "best_issue_reason": "Issue 43646 is the strongest standalone representative because it is a general v5 regression with a clearly described failure mode and broad duplicate-triage utility.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43531", "right": "issue:43742", "accept": false, - "reason": "Different bug classes and models: sliding_window/recurrent-state behavior in Qwen3-MoE vs a key error while loading MobileLLM-125M." + "reason": "Different failures: Qwen3-MoE sliding_window behavior vs a MobileLLM pretrained-load key error." }, { "left": "issue:42898", "right": "issue:46491", "accept": false, - "reason": "Both are tokenizer regressions, but 42898 is a broad v5 clean_up_tokenization_spaces change while 46491 is a CodeLlama leading-space round-trip bug; different code path and symptom." + "reason": "Both are tokenization regressions, but one is about v5 cleanup behavior and the other is a CodeLlama round-trip spacing bug." }, { "left": "issue:29159", "right": "issue:33357", "accept": false, - "reason": "Slow-vs-fast tokenizer inconsistency is unrelated to the CLIP MacOS bus error." + "reason": "Tokenizer inconsistency vs a MacOS bus error on a CLIP model; different symptom and subsystem." }, { "left": "issue:43653", "right": "issue:44568", "accept": false, - "reason": "Both concern special tokens, but one is BigBird mask-token registration and the other is mdeberta BOS/EOS insertion; not the same underlying bug." + "reason": "Both involve special tokens, but they affect different tokenizers and different concrete behaviors." }, { "left": "issue:43646", "right": "issue:43749", "accept": false, - "reason": "AutoConfig trust_remote_code missing fields is unrelated to FSDP_CPU_RAM_EFFICIENT_LOADING being broken." + "reason": "Model initialization break in v5 vs FSDP CPU RAM efficient loading; unrelated code paths." }, { "left": "issue:43452", "right": "issue:43577", "accept": false, - "reason": "gguf AutoTokenizer/AutoModel loading failure and BLIP2 dtype propagation are distinct loading issues with different root causes." + "reason": "GGUF/AutoModel loading break and BLIP2 dtype propagation are distinct loading bugs." }, { "left": "issue:43756", "right": "issue:46402", "accept": false, - "reason": "Missing RoPE layers in Smollm3 and Sam3 text-encoder weights not loading are different model-specific load bugs." + "reason": "Smollm3 RoPE-layer mismatch and Sam3Model text-encoder weight loading are different model-specific regressions." }, { - "left": "issue:43688", - "right": "issue:45704", + "left": "issue:44843", + "right": "issue:44991", "accept": false, - "reason": "Auxiliary-loss normalization and a T5 apex.FusedRMSNorm memory leak are unrelated bugs." + "reason": "Offline model_info call in a regex patch vs tokenizer loading failure for a specific repo; not the same bug." }, { - "left": "issue:44843", - "right": "issue:44991", + "left": "issue:43688", + "right": "issue:45704", "accept": false, - "reason": "Both are tokenizer-loading regressions, but one is an offline-mode model_info() call in _patch_mistral_regex and the other is a specific tokenizer load failure for EMBEDDIA/est-roberta." + "reason": "Aux-loss normalization and an apex RMSNorm memory leak are unrelated issues." }, { "left": "issue:43653", "right": "issue:43927", "accept": false, - "reason": "Both involve token handling, but 43653 is a missing special-token registration and 43927 is custom token IDs being lost across save/load." + "reason": "BigBird special-token registration and DiaConfig custom token IDs being lost are different mechanisms." }, { "left": "issue:44464", "right": "issue:44743", "accept": false, - "reason": "Both are generation/cache issues, but 44464 is compiled-forward chunked generation inconsistency while 44743 is recurrent state resetting with cache in modular_qwen3_5." + "reason": "Both involve generation/caching, but one is compiled forward chunked generation and the other is recurrent-state reset in modular_qwen3_5." }, { "left": "issue:43688", "right": "issue:45701", "accept": false, - "reason": "Aux-loss normalization is unrelated to a general version-dependent tokenization change." + "reason": "Auxiliary-loss normalization is unrelated to a tokenization version-change report." }, { "left": "issue:44484", "right": "issue:44704", "accept": false, - "reason": "save_pretrained shard-size behavior and AutoProcessor kwarg forwarding are different API issues." + "reason": "A save_pretrained shard-size question and missing kwargs in AutoProcessor loading are not the same bug." }, { "left": "issue:45491", "right": "issue:46002", "accept": false, - "reason": "Both mention NaNs, but Gemma3 mixed-length sliding-window attention on GPU is a different failure mode from NaNs in classification heads after checkout/init." + "reason": "Both mention NaNs, but one is Gemma3 sliding-window attention on GPU and the other is classification-head NaNs after checkout/init." }, { "left": "issue:44479", "right": "issue:44625", "accept": false, - "reason": "A video-input regression in Qwen VL models is unrelated to num_labels propagation in Qwen3.5 config." + "reason": "Video-input regression for Qwen variants vs num_labels propagation in Qwen3.5; different failures." }, { "left": "issue:44821", @@ -7925,67 +7856,73 @@ "left": "issue:43688", "right": "issue:45440", "accept": false, - "reason": "Aux-loss normalization and native DeepseekV3MoE divergence are different correctness issues in different code paths." + "reason": "Aux-loss normalization is not the same as DeepSeekV3 native-vs-remote divergence." }, { "left": "issue:33666", "right": "issue:36010", "accept": false, - "reason": "Qwen2-VL multi-GPU training and the GenerationMixin import error are unrelated." + "reason": "Qwen2-VL multi-GPU training and GenerationMixin import failure are different topics and code paths." }, { "left": "issue:32090", "right": "issue:33666", "accept": false, - "reason": "Trainer GPU broadcast of NoneType and Qwen2-VL multi-GPU training are not the same bug." + "reason": "Trainer GPU broadcast NoneType error is unrelated to Qwen2-VL multi-GPU training." }, { "left": "issue:44734", "right": "issue:46612", "accept": false, - "reason": "Both involve generation state, but 44734 is a KV-cache continuation indexing crash in serve while 46612 is beam-search cache reorder being skipped for specific models." + "reason": "Both touch cache/generation, but one is a server KV-cache indexing crash and the other is beam-search reorder being skipped for certain model classes." }, { "left": "issue:44466", "right": "issue:46050", "accept": false, - "reason": "Inconsistent lm_head.weight serialization and broken quantized save/load for sinq are both persistence bugs, but they are different mechanisms." + "reason": "Serialization of lm_head.weight by device and quantized sinq save/load breakage are distinct persistence bugs." }, { "left": "issue:43504", "right": "issue:43720", "accept": false, - "reason": "Beit legacy-field loading and BitNet packed-weight unpacking during accelerate loading are distinct load-time failures." + "reason": "Legacy-field preset loading and packed-weight unpacking during accelerate loading are different load-time issues." + }, + { + "left": "issue:44977", + "right": "issue:46693", + "accept": false, + "reason": "Qwen3.5 flash-attention generation correctness vs FA2 long-context performance regression; not the same underlying bug." }, { "left": "issue:43504", "right": "issue:43550", "accept": false, - "reason": "A Beit pretrained-preset load issue is unrelated to Bamba torch.compile+SDPA runtime failure." + "reason": "Beit legacy-field loading and Bamba torch.compile+SDPA failure are unrelated." }, { - "left": "issue:44977", - "right": "issue:46693", + "left": "issue:44464", + "right": "issue:44977", "accept": false, - "reason": "Both mention Qwen3.5 and FA2, but 44977 is a functional generation regression while 46693 is a performance regression after 5.2.0." + "reason": "Compiled-forward chunked generation inconsistency and Qwen3.5 flash-attention generation failure are different regressions." }, { "left": "issue:41720", "right": "issue:44464", "accept": false, - "reason": "cudaErrorAssert on auto device mapping for Qwen3 is unrelated to compiled-forward chunked generation inconsistency." + "reason": "Auto device mapping cudaErrorAssert and compiled-forward chunked generation are separate problems." }, { "left": "issue:43643", "right": "issue:44561", "accept": false, - "reason": "Both touch trust_remote_code, but 43643 is missing fields from AutoConfig and 44561 is a breakage from removing is_torch_fx_available." + "reason": "Both involve trust_remote_code, but one is missing config fields and the other is removal of is_torch_fx_available; different breakages." }, { "left": "issue:43577", "right": "issue:44554", "accept": false, - "reason": "BLIP2 dtype propagation and an MPS attention correctness issue with mismatched head dims are unrelated bugs." + "reason": "BLIP2 dtype propagation and an MPS attention correctness issue are unrelated." } ] }, @@ -8062,14 +7999,14 @@ "issue:43611|issue:46846", "issue:43720|issue:43927", "issue:43504|issue:44610", - "issue:43606|issue:43927", "issue:33666|issue:36331", - "issue:43278|issue:43688", + "issue:43606|issue:43927", "issue:43782|issue:46693", + "issue:43278|issue:43688", "issue:43262|issue:43295", "issue:43720|issue:44361", - "issue:44964|issue:45405", "issue:44514|issue:45290", + "issue:44964|issue:45405", "issue:43493|issue:43825", "issue:44554|issue:46365", "issue:43525|issue:43572", @@ -8085,219 +8022,219 @@ "issue:43278|issue:43844", "issue:43646|issue:44466", "issue:46209|issue:46710", - "issue:33357|issue:33666", "issue:44849|issue:46209", + "issue:33357|issue:33666", "issue:44568|issue:46489", "issue:43278|issue:43701" ], - "prepared_review_unit_hash": "f4f783095bedc81f5aa4b360f33463f145381a9e0599c8176c83f5b9d1524f4d", + "prepared_review_unit_hash": "d0cc6bfe47aeef879f8e3e97515ca7db5fdea046272500748d9cf3127a571260", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36794, - "estimated_input_tokens": 9199, - "estimated_eval_tokens": 18654 + "serialized_chars": 36792, + "estimated_input_tokens": 9198, + "estimated_eval_tokens": 18652 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: it mixes unrelated tokenizer, processor, model-loading, training, and performance bugs across many model families. None of the soft-edge pairs look like the same underlying fix target, so I would not merge any of them as duplicates.", - "confidence": 0.96, - "canonical_issue_reason": "No single canonical duplicate stands out; the cluster is too broad and the issues do not share one concrete bug. If forced to pick a representative, issue 43611 is the broadest v5-loading regression, but it is still not a true cluster anchor.", + "summary": "The cluster is highly heterogeneous. Most pairs only share a broad subsystem label like tokenizer, processor, config, or model loading, but they describe different concrete failures and code paths. No soft pair looks safely mergeable into a single duplicate issue.", + "confidence": 0.94, + "canonical_issue_reason": "issue:44568 is the cleanest, most self-contained tokenizer regression in the set, and the only nearby soft overlap is another tokenizer/template-related bug rather than a model-loading or training issue.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43611 is the most general, user-facing regression among the set and best represents the broad theme of v5 breakage, though it should not be treated as a duplicate hub for the whole cluster.", + "best_issue_reason": "issue:44568 is the strongest representative of the limited tokenizer-regression theme in this cluster because it is specific, reproducible, and narrowly scoped.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43611", "right": "issue:46846", "accept": false, - "reason": "Different failures: base_model_prefix loading regression vs missing linear_attn TP plan entries." + "reason": "Both mention model internals, but one is a base_model_prefix loading issue and the other is a tensor-parallel plan gap; different bugs." }, { "left": "issue:43720", "right": "issue:43927", "accept": false, - "reason": "BitNet packed-weight loading bug vs DiaConfig token-id persistence bug; unrelated code paths." + "reason": "Packed-weight loading failure and missing token-id propagation are unrelated code paths." }, { "left": "issue:43504", "right": "issue:44610", "accept": false, - "reason": "BEiT legacy-field load failure vs OmDet-Turbo processor image-size mismatch." - }, - { - "left": "issue:43606", - "right": "issue:43927", - "accept": false, - "reason": "CPU offload device mismatch in Bark vs config token-id loss in Dia; different bugs." + "reason": "Legacy field loading in BEiT and an OmDet processor input-size mismatch are different problems." }, { "left": "issue:33666", "right": "issue:36331", "accept": false, - "reason": "Multi-GPU Qwen2-VL training issue vs CustomTrainer signature mismatch; not the same problem." + "reason": "Multi-GPU training support and a custom trainer signature error are not the same defect." }, { - "left": "issue:43278", - "right": "issue:43688", + "left": "issue:43606", + "right": "issue:43927", "accept": false, - "reason": "Embedding dtype regression vs auxiliary-loss normalization bug; separate model behaviors." + "reason": "CPU offload device mismatch is unrelated to DiaConfig token-id persistence." }, { "left": "issue:43782", "right": "issue:46693", "accept": false, - "reason": "Qwen3VL weight_only loading error vs FA2 long-context performance regression." + "reason": "Qwen3VL weight-only loading and FA2 long-context performance regression are different failure modes." + }, + { + "left": "issue:43278", + "right": "issue:43688", + "accept": false, + "reason": "Embedding dtype drift and auxiliary-loss normalization are separate training-time issues." }, { "left": "issue:43262", "right": "issue:43295", "accept": false, - "reason": "Different processor regressions: sampling-rate default vs processor.tokenizer/images compatibility." + "reason": "Both involve processors, but one is a sampling-rate default and the other is processor/tokenizer API breakage." }, { "left": "issue:43720", "right": "issue:44361", "accept": false, - "reason": "Packed-weight loading bug vs MLukeTokenizer task AttributeError; unrelated." + "reason": "Accelerate packed-weight loading and MLukeTokenizer task handling are unrelated." }, { - "left": "issue:44964", - "right": "issue:45405", + "left": "issue:44514", + "right": "issue:45290", "accept": false, - "reason": "Phi-4 multimodal loading failure vs unreleased PEFT version bump; not duplicates." + "reason": "Both hit apply_chat_template, but one is batched padding=False and the other is assistant tool-call/no-content handling; different edge cases." }, { - "left": "issue:44514", - "right": "issue:45290", + "left": "issue:44964", + "right": "issue:45405", "accept": false, - "reason": "Both involve apply_chat_template, but the crashing inputs and failure modes differ." + "reason": "Phi-4 multimodal loading and a PEFT version pin are not the same bug." }, { "left": "issue:43493", "right": "issue:43825", "accept": false, - "reason": "SigLIP2 implementation discrepancy vs pipeline error-message regression; different layers." + "reason": "SigLIP2 implementation discrepancy and an incorrect pipeline error message are unrelated." }, { "left": "issue:44554", "right": "issue:46365", "accept": false, - "reason": "MPS attention correctness bug vs Trainer import conflict with datasets; unrelated." + "reason": "An MPS attention correctness issue and a Trainer import conflict with datasets are different failures." }, { "left": "issue:43525", "right": "issue:43572", "accept": false, - "reason": "Both are missing pad-token fields, but they are model-specific config issues with different attributes and fixes." + "reason": "Both are missing-pad-token config complaints, but they affect different model configs and attributes." }, { "left": "issue:44977", "right": "issue:46209", "accept": false, - "reason": "Qwen3.5 flash-attention generation bug vs FP8 inference failure on B200; different hardware/code paths." + "reason": "Qwen3.5 flash-attention generation and FP8 inference on B200 are distinct issues." }, { "left": "issue:44625", "right": "issue:45584", "accept": false, - "reason": "num_labels propagation bug vs Whisper empty-transcription generation bug." + "reason": "num_labels propagation in Qwen3.5 and Whisper empty-transcription failures are unrelated." }, { "left": "issue:43452", "right": "issue:44451", "accept": false, - "reason": "gguf_file loading regression vs ScandiBERT load failure; not the same root cause." + "reason": "GGUF loading breakage and a specific ScandiBERT loading failure are different model-loading bugs." }, { "left": "issue:28218", "right": "issue:33357", "accept": false, - "reason": "Tokenizer spacing bug vs MacOS bus error in CLIP model; unrelated." + "reason": "Tokenizer spacing after added tokens and a MacOS bus error in CLIP are unrelated." }, { "left": "issue:45823", "right": "issue:46050", "accept": false, - "reason": "Gemma4 device-map mismatch vs sinq save/load quantization breakage." + "reason": "Gemma4 device-mismatch forward errors and sinq save/load corruption are different code paths." }, { "left": "issue:42796", "right": "issue:44977", "accept": false, - "reason": "Tokenizer BPE-marker handling vs Qwen3.5 flash-attention generation; different failures." + "reason": "Ministral tokenizer BPE-marker handling and Qwen3.5 flash-attn generation are unrelated." }, { "left": "issue:43994", "right": "issue:46192", "accept": false, - "reason": "SigLIP2 output quality issue vs return_tensors being ignored; unrelated API behavior." + "reason": "SigLIP2 nonsensical outputs and silently ignored return_tensors are different API bugs." }, { "left": "issue:43493", "right": "issue:43844", "accept": false, - "reason": "SigLIP2 implementation discrepancy vs ZeRO-3 gradient increase; not the same bug." + "reason": "SigLIP2 implementation mismatch and a ZeRO-3 gradient regression do not describe the same bug." }, { "left": "issue:39692", "right": "issue:43295", "accept": false, - "reason": "Documentation example errors vs processor.tokenizer regression; different artifact types and causes." + "reason": "Documentation example errors and processor/tokenizer regression are not the same underlying issue." }, { "left": "issue:43278", "right": "issue:43844", "accept": false, - "reason": "dtype regression in evaluation vs gradient growth under ZeRO-3; separate training issues." + "reason": "Embedding dtype drift and training-gradient explosion are different training regressions." }, { "left": "issue:43646", "right": "issue:44466", "accept": false, - "reason": "Custom model initialization breakage vs device-dependent lm_head serialization inconsistency." + "reason": "Custom model initialization breakage and lm_head serialization inconsistency are separate model lifecycle bugs." }, { "left": "issue:46209", "right": "issue:46710", "accept": false, - "reason": "FP8 inference failure on B200 vs accuracy regression after upgrade; different symptoms and fixes." + "reason": "FP8 inference failure and an accuracy regression after version upgrade are not the same defect." }, { - "left": "issue:33357", - "right": "issue:33666", + "left": "issue:44849", + "right": "issue:46209", "accept": false, - "reason": "MacOS CLIP bus error vs Qwen2-VL multi-GPU training; no shared code path." + "reason": "Qwen3.5 hidden-state bug and FP8/B200 inference failure are unrelated." }, { - "left": "issue:44849", - "right": "issue:46209", + "left": "issue:33357", + "right": "issue:33666", "accept": false, - "reason": "Qwen3.5 hidden_states bug vs FP8 hardware-specific inference failure." + "reason": "MacOS bus error in CLIP and Qwen2-VL multi-GPU training are different subsystems and failure modes." }, { "left": "issue:44568", "right": "issue:46489", "accept": false, - "reason": "Special-token insertion bug in mdeberta-v3-base vs DeepSeek-Coder v1 tokenizer regression; same subsystem but not the same concrete bug." + "reason": "Both are tokenizer regressions in v5+, but they affect different models and different special-token/BPE paths; not the same concrete bug." }, { "left": "issue:43278", "right": "issue:43701", "accept": false, - "reason": "Embedding dtype regression vs checkpoint resume key mismatch; unrelated." + "reason": "Embedding dtype change and resume_from_checkpoint key mismatch are unrelated." } ] }, @@ -8371,22 +8308,22 @@ "issue:46858" ], "soft_pairs": [ - "issue:44792|issue:45405", "issue:44488|issue:44843", + "issue:44792|issue:45405", "issue:43927|issue:44361", "issue:44568|issue:45812", "issue:43824|issue:44589", "issue:43575|issue:45405", - "issue:43504|issue:43606", "issue:44849|issue:45357", + "issue:43504|issue:43606", "issue:43257|issue:43299", - "issue:45245|issue:46710", - "issue:46399|issue:46846", "issue:46821|issue:46832", - "issue:43606|issue:44112", - "issue:44493|issue:46285", + "issue:46399|issue:46846", + "issue:45245|issue:46710", "issue:44589|issue:45520", "issue:43334|issue:43572", + "issue:44493|issue:46285", + "issue:43606|issue:44112", "issue:43334|issue:45759", "issue:45072|issue:46832", "issue:43504|issue:43653", @@ -8397,205 +8334,205 @@ "issue:43844|issue:44964", "issue:45200|issue:45229", "issue:43873|issue:46050", - "issue:44869|issue:46752", - "issue:29942|issue:30333" + "issue:29942|issue:30333", + "issue:44869|issue:46752" ], - "prepared_review_unit_hash": "44437a3e15d8804bd818974deeabe902a6cda31379185866587aca655a77df7c", + "prepared_review_unit_hash": "4c9fce7a591a0371a7d3a15a55b0add0cf7a627656bc2a3293368898be025f0d", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36307, - "estimated_input_tokens": 9077, - "estimated_eval_tokens": 18410 + "serialized_chars": 36304, + "estimated_input_tokens": 9076, + "estimated_eval_tokens": 18408 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is not a duplicate set: the items span unrelated model-loading, tokenizer, cache/offload, CI, and training issues. The suggested soft matches are mostly superficial keyword overlap (e.g., \"missing\", \"load\", \"OOM\", \"pad_token\") rather than the same concrete bug.", - "confidence": 0.98, - "canonical_issue_reason": null, - "canonical_pr_reason": null, - "best_issue_reason": null, - "best_pr_reason": null, + "summary": "This cluster is a loose collection of unrelated issue reports with similar wording, not a true duplicate set. The soft edges mostly connect superficially similar regressions across different models, tokenizers, cache/offload paths, and CI failures; none look like the same underlying bug or change.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue stands out: the items span many unrelated bugs and regressions across different subsystems and models.", + "canonical_pr_reason": "No pull requests are present in this cluster.", + "best_issue_reason": "No issue is a clear global representative; several are broad regressions, but none meaningfully subsume the rest of the cluster.", + "best_pr_reason": "No pull requests are present in this cluster.", "soft_edge_verdicts": [ { - "left": "issue:44792", - "right": "issue:45405", + "left": "issue:44488", + "right": "issue:44843", "accept": false, - "reason": "Different bugs: a janus generation test failure vs an unrelated dependency version pin issue." + "reason": "Both concern loading/tokenizer behavior, but one is a model load failure for cjvt/sleng-bert and the other is an offline-mode regression in _patch_mistral_regex; different root causes." }, { - "left": "issue:44488", - "right": "issue:44843", + "left": "issue:44792", + "right": "issue:45405", "accept": false, - "reason": "Both mention loading/tokenizer behavior, but one is a model loading problem and the other is an offline hub call bug in regex patching." + "reason": "Janus test failure and an unreleased PEFT version bump are unrelated issues." }, { "left": "issue:43927", "right": "issue:44361", "accept": false, - "reason": "Different tokenizer/model failures with different symptoms and code paths; not the same underlying bug." + "reason": "Different AttributeErrors in different code paths: Dia config save/load vs MLuke tokenizer task handling." }, { "left": "issue:44568", "right": "issue:45812", "accept": false, - "reason": "Both involve tokenizer behavior, but one is missing BOS/EOS insertion for a specific model while the other is a broad Granite token-ID regression." + "reason": "Both are tokenizer regressions, but one is missing BOS/EOS on mdeberta-v3 and the other is wrong token IDs for Granite models; not the same bug." }, { "left": "issue:43824", "right": "issue:44589", "accept": false, - "reason": "One is an import error for a Qwen VL class; the other is a storage backend/type mismatch. No shared failure path." + "reason": "ImportError for a missing Qwen2.5-VL class and a Float8 storage TypeError are unrelated." }, { "left": "issue:43575", "right": "issue:45405", "accept": false, - "reason": "Qwen2-57B tensor-parallel OOM is unrelated to a PEFT minimum-version bump." + "reason": "Qwen2 OOM with tensor parallelism and a PEFT release pin issue are unrelated." }, { - "left": "issue:43504", - "right": "issue:43606", + "left": "issue:44849", + "right": "issue:45357", "accept": false, - "reason": "Legacy field loading bug in BEiT is unrelated to Bark CPU offload device-mismatch failures." + "reason": "Both are Qwen3.5-related, but hidden-state output bug and incorrect save_pretrained visual keys are different failures." }, { - "left": "issue:44849", - "right": "issue:45357", + "left": "issue:43504", + "right": "issue:43606", "accept": false, - "reason": "Different Qwen variants and different failures: output_hidden_states behavior vs incorrect visual encoder keys on save_pretrained." + "reason": "BEiT legacy-field loading and Bark CPU-offload device mismatch are different code paths and defects." }, { "left": "issue:43257", "right": "issue:43299", "accept": false, - "reason": "Both are Qwen3 MOE-related, but one is an accelerate+deepspeed conversion path and the other is a transformers dev loading regression for Qwen3VL." + "reason": "Both mention Qwen3 MOE, but one is accelerate+deepspeed weight conversion and the other is transformers-dev model loading; not the same fix." }, { - "left": "issue:45245", - "right": "issue:46710", + "left": "issue:46821", + "right": "issue:46832", "accept": false, - "reason": "Category-count RuntimeError and DeepSeek accuracy regression are distinct problems with no common code path." + "reason": "TimesFM window_size AttributeError and RT-DETR forward TypeError are unrelated." }, { "left": "issue:46399", "right": "issue:46846", "accept": false, - "reason": "Both are Qwen3.5-related, but one is a missing decorator for a kernel registration and the other is a missing TP plan entry causing OOM/shape errors." + "reason": "Both involve Qwen3.5 TP/kernels, but missing decorator registration and missing TP plan entries are different concrete bugs." }, { - "left": "issue:46821", - "right": "issue:46832", + "left": "issue:45245", + "right": "issue:46710", "accept": false, - "reason": "TimesFM window_size attribute error and RT-DETR feature-level mismatch are unrelated model-specific bugs." + "reason": "Category-cardinality limit error and an accuracy regression are not the same underlying issue." }, { - "left": "issue:43606", - "right": "issue:44112", + "left": "issue:44589", + "right": "issue:45520", "accept": false, - "reason": "Both are CI failures, but they stem from different tests and different device-handling issues." + "reason": "Float8 storage lookup failure and Python 3.13 flash_attn import keyerror are unrelated." }, { - "left": "issue:44493", - "right": "issue:46285", + "left": "issue:43334", + "right": "issue:43572", "accept": false, - "reason": "Position-id key warnings across many models vs CLIP structure inconsistency; different regressions and fixes." + "reason": "Both are missing config fields after v5, but Qwen3VL pad_token_id and StableLm pad_token_idx are different regressions." }, { - "left": "issue:44589", - "right": "issue:45520", + "left": "issue:44493", + "right": "issue:46285", "accept": false, - "reason": "Float8 storage lookup failure is unrelated to a flash_attn import-utils KeyError on Python 3.13." + "reason": "Unexpected position-id keys and CLIP text-model structure changes are unrelated." }, { - "left": "issue:43334", - "right": "issue:43572", + "left": "issue:43606", + "right": "issue:44112", "accept": false, - "reason": "Qwen3-VL missing pad_token_id and StableLmConfig pad_token_idx are separate model/config bugs." + "reason": "Both are CI/device-related, but Bark CPU offload mismatch and GraniteSpeech stale device override test are different failures." }, { "left": "issue:43334", "right": "issue:45759", "accept": false, - "reason": "Qwen3-VL pad_token_id AttributeError is unrelated to AutoModelForCausalLM.from_config not unwrapping text_config." + "reason": "Qwen3VL pad_token_id loss and composite Qwen 3.5/3.6 config unwrapping are distinct loading issues." }, { "left": "issue:45072", "right": "issue:46832", "accept": false, - "reason": "Dtype mismatch CI failures vs RT-DETR feature-level mismatch; different subsystems and symptoms." + "reason": "bf16 dtype mismatch in SwitchTransformers/TimmWrapperModel and RT-DETR feature-level mismatch are unrelated." }, { "left": "issue:43504", "right": "issue:43653", "accept": false, - "reason": "BEiT legacy-field loading bug is unrelated to BigBirdTokenizer special-token registration/empty decode behavior." + "reason": "BEiT legacy-field loading and BigBirdTokenizer special-token registration are unrelated." }, { "left": "issue:43572", "right": "issue:44291", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig and unexpected _is_hf_initialized arg in init_empty_weights are different version-regression issues." + "reason": "StableLm missing pad_token_idx and init_empty_weights unexpected argument are different regressions." }, { "left": "issue:44464", "right": "issue:46424", "accept": false, - "reason": "Chunked generation inconsistency with compiled forward is unrelated to cache length handling for quantized cache only." + "reason": "Compiled chunked generation inconsistency and max_cache_length handling for quantized cache are unrelated." }, { "left": "issue:43937", "right": "issue:46701", "accept": false, - "reason": "GenerationConfig validation failure for GLM-5 is unrelated to DiffusionGemma image-text pipeline failure." + "reason": "DiaConfig generation error and DiffusionGemma pipeline failure are separate bugs." }, { "left": "issue:46832", "right": "issue:46858", "accept": false, - "reason": "RT-DETR backbone/output mismatch and static-cache seq-length tensor bug are separate runtime errors." + "reason": "RT-DETR num_feature_levels mismatch and cache.get_seq_length device-tensor arithmetic are unrelated." }, { "left": "issue:43844", "right": "issue:44964", "accept": false, - "reason": "Gradient explosion with ZeRO-3 and a Phi-4 multimodal loading failure are unrelated problems." + "reason": "ZeRO-3 gradient-growth behavior and Phi-4 multimodal load failure do not share a concrete bug." }, { "left": "issue:45200", "right": "issue:45229", "accept": false, - "reason": "Gemma 4 fine-tuning token defaults vs multi-GPU inference OOM are not the same underlying bug." + "reason": "Gemma 4 token-type defaulting and multi-GPU OOM are different problems." }, { "left": "issue:43873", "right": "issue:46050", "accept": false, - "reason": "Quantization offload behavior and sinq save/load breakage are different quantization issues with different code paths." + "reason": "Both mention quantization/offloading, but one is an offloading bug and the other is broken save/load for sinq quantization; not mergeable as one fix." }, { - "left": "issue:44869", - "right": "issue:46752", + "left": "issue:29942", + "right": "issue:30333", "accept": false, - "reason": "Whisper timestamp decoding crash on trailing replacement char is unrelated to apply_chat_template on an empty list." + "reason": "Flash Attention 2 test failures and MLFlow reporting on failed training jobs are unrelated." }, { - "left": "issue:29942", - "right": "issue:30333", + "left": "issue:44869", + "right": "issue:46752", "accept": false, - "reason": "Flash Attention 2 test failures and MLflow job-reporting issues are entirely unrelated." + "reason": "Whisper timestamp decoding crash and empty-list chat template IndexError are different code paths and bugs." } ] }, @@ -8693,206 +8630,206 @@ "issue:44186|issue:46192", "issue:43572|issue:46821", "issue:43295|issue:44610", - "issue:45823|issue:46192", "issue:45200|issue:45823", + "issue:45823|issue:46192", "issue:29870|issue:29942" ], - "prepared_review_unit_hash": "df15d95f8ee061fd431cf0696a05a2350e9ab3a00fbccdb43c89b23e038dd6dc", + "prepared_review_unit_hash": "8b4d7554374534a84fdebf0a2c3e03e641558b37db5959384cb0665b8c8e69aa", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 35560, - "estimated_input_tokens": 8890, - "estimated_eval_tokens": 18036 + "serialized_chars": 35567, + "estimated_input_tokens": 8892, + "estimated_eval_tokens": 18040 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose mix of unrelated Transformers issues across tokenizers, generation, serving, multimodal processors, quantization, and model loading. None of the soft-similarity pairs look like the same underlying bug/change, so I would not merge any of them as duplicates.", - "confidence": 0.86, - "canonical_issue_reason": "No true duplicate hub stands out from the titles alone; if one issue must anchor the cluster, issue:43873 is the broadest active report and has the strongest inbound reference signal, so it is the closest thing to a cluster representative.", + "summary": "The items are mostly unrelated issue reports across tokenizer, generation, serving, quantization, and multimodal code paths. I rejected every soft pair because they describe different concrete failures, not the same underlying bug.", + "confidence": 0.96, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 is the best single issue representative because it is broad, currently open, and appears to be the most central runtime/memory-management report in the set.", + "best_issue_reason": "issue:45200 is the strongest standalone issue: it is specific, reproducible, and has a clear expected fix (default mm_token_type_ids to zeros for text-only fine-tuning in Gemma 4).", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45229", "right": "issue:45823", "accept": false, - "reason": "Both are Gemma4-related, but one is a multi-GPU CUDA OOM and the other is a device-map mismatch in forward; different failure modes and code paths." + "reason": "Same model family, but one is CUDA OOM and the other is a device-map mismatch during forward; different failure modes." }, { "left": "issue:43994", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 incorrect outputs vs a position-id key warning across many models; too different to be the same bug." + "reason": "SigLIP2 result-quality regression vs unexpected position-id key warnings; unrelated bugs." }, { "left": "issue:45588", "right": "issue:46050", "accept": false, - "reason": "Flash-attention sinkless-model crash vs sinq save/load corruption; unrelated subsystems and symptoms." + "reason": "Flash-attention sinkless-model crash vs quantized save/load corruption; different code paths and symptoms." }, { "left": "issue:42490", "right": "issue:43696", "accept": false, - "reason": "Local-path import warning and GPT-oss OOM are unrelated." + "reason": "Trailing-slash import warning and GPU OOM are unrelated." }, { "left": "issue:29159", "right": "issue:32090", "accept": false, - "reason": "Tokenizer slow/fast inconsistency is not the same issue as a Trainer GPU broadcast NoneType error." + "reason": "Tokenizer slow/fast inconsistency vs Trainer broadcast NoneType error; no shared underlying defect." }, { "left": "issue:44977", "right": "issue:46701", "accept": false, - "reason": "Qwen3.5 flash-attention generation failure vs a DiffusionGemma pipeline output issue; different models and execution paths." + "reason": "Qwen flash-attention generation bug vs DiffusionGemma pipeline failure; different models and subsystems." }, { "left": "issue:45084", "right": "issue:45520", "accept": false, - "reason": "Non-template-node compile error and flash_attn import KeyError are separate import/compilation problems." + "reason": "Template-node compilation error vs flash_attn import KeyError; unrelated failures." }, { "left": "issue:28218", "right": "issue:34567", "accept": false, - "reason": "Added-token spacing behavior and TrainerState token-count tracking are unrelated." + "reason": "Added-token spacing bug vs TrainerState counter not updating; different concerns." }, { "left": "issue:43065", "right": "issue:46369", "accept": false, - "reason": "A dummy Conv2d implementation and a memory overflow report on sam3-litetext are not the same defect." + "reason": "Dummy Conv2d placeholder issue vs memory overflow in sam3-litetext; not the same bug." }, { "left": "issue:44554", "right": "issue:45446", "accept": false, - "reason": "MPS attention correctness with mismatched head dims is unrelated to a PyTorch version check bug." + "reason": "MPS attention correctness issue vs PyTorch version guard for AuxRequest import; different layers." }, { "left": "issue:44423", "right": "issue:44734", "accept": false, - "reason": "Both are serve crashes, but they hit different continuations and different tensor/object errors; not the same underlying bug." + "reason": "Both are serve crashes, but one is multimodal .to() handling and the other is KV-cache continuation indexing; distinct code paths." }, { "left": "issue:15354", "right": "issue:38175", "accept": false, - "reason": "TorchScript GeneratorExp export failure and SigLIP2 zero probabilities are unrelated." + "reason": "torch.jit.script GeneratorExp export failure vs SigLIP2 zero-probability output issue; unrelated." }, { "left": "issue:43493", "right": "issue:43701", "accept": false, - "reason": "SigLIP2 HF-vs-JAX discrepancy and resume_from_checkpoint key mismatch do not share a concrete code path." + "reason": "SigLIP2 HF/JAX discrepancy vs checkpoint key mismatch; different bugs." }, { "left": "issue:41720", "right": "issue:46710", "accept": false, - "reason": "A800 cudaErrorAssert under auto device mapping is different from an accuracy regression after upgrading Transformers." + "reason": "CUDA assert under auto device mapping vs accuracy regression after upgrade; not the same defect." }, { "left": "issue:45406", "right": "issue:46121", "accept": false, - "reason": "Gemma4Processor _tokenizer crash in serve and a rope-parameter JSON list type error are unrelated." + "reason": "Gemma4Processor missing _tokenizer vs rope-validation type error; different failure surfaces." }, { "left": "issue:44977", "right": "issue:46710", "accept": false, - "reason": "Flash-attention generation breakage in Qwen3.5 is not the same as a model accuracy regression after version upgrade." + "reason": "Flash-attention generation bug vs model accuracy regression; unrelated." }, { "left": "issue:45072", "right": "issue:45198", "accept": false, - "reason": "BF16 dtype mismatches in SwitchTransformers/TimmWrapperModel and Wav2Vec2 save/tokenization failures are unrelated." + "reason": "Bfloat16 dtype mismatch in CI vs Wav2Vec2 save/tokenization failure; different models and paths." }, { "left": "issue:43295", "right": "issue:43450", "accept": false, - "reason": "Processor.tokenizer regression with image inputs and batched video shape errors are different multimodal bugs." + "reason": "Processor/tokenizer regression vs video batch shape bug; distinct issues." }, { "left": "issue:37428", "right": "issue:43957", "accept": false, - "reason": "Missing flash-support import and meta-device model loading failures are not the same issue." + "reason": "Both concern loading/imports, but one is a missing flash-attention helper import and the other is meta-device loading breakage; not the same bug." }, { "left": "issue:45588", "right": "issue:45759", "accept": false, - "reason": "Flash-attention sinkless-model crash and from_config text_config unwrapping are separate processor/config problems." + "reason": "Flash-attention None/s_aux crash vs composite Qwen config unwrapping; unrelated." }, { "left": "issue:43540", "right": "issue:45200", "accept": false, - "reason": "Video-input ValueError in Qwen3OmniMoe and missing mm_token_type_ids in Gemma4 are different multimodal preprocessing bugs." + "reason": "Video-input processing error in Qwen3OmniMoe vs Gemma4 token-type defaulting bug; different models and code paths." }, { "left": "issue:43295", "right": "issue:43873", "accept": false, - "reason": "Processor/tokenizer regression and quantization offloading behavior are far apart in both symptom and code path." + "reason": "Processor tokenizer regression vs quantization offloading problem; no shared root cause." }, { "left": "issue:44186", "right": "issue:46192", "accept": false, - "reason": "LayoutLMv2Tokenizer NER/padding crash and return_tensors being ignored are distinct tokenizer API issues." + "reason": "NER/batched-padding tokenizer crash vs return_tensors being ignored; different tokenizer behaviors." }, { "left": "issue:43572", "right": "issue:46821", "accept": false, - "reason": "StableLmConfig pad_token_idx regression and TimesFM window_size AttributeError are unrelated config bugs." + "reason": "StableLmConfig pad_token_idx missing vs TimesFM window_size AttributeError; unrelated configuration bugs." }, { "left": "issue:43295", "right": "issue:44610", "accept": false, - "reason": "Tokenizer/processor regression and OmDet-Turbo processor size mismatch are different model-specific preprocessing problems." + "reason": "Processor/tokenizer regression vs OmDet-Turbo image-size mismatch; different model-specific issues." }, { - "left": "issue:45823", - "right": "issue:46192", + "left": "issue:45200", + "right": "issue:45823", "accept": false, - "reason": "Gemma4 device-mapping mismatch and return_tensors being ignored are unrelated." + "reason": "Both are Gemma4 issues, but one is token-type defaults and the other is device-map placement; not the same bug." }, { - "left": "issue:45200", - "right": "issue:45823", + "left": "issue:45823", + "right": "issue:46192", "accept": false, - "reason": "One is a missing default mm_token_type_ids, the other is a device placement mismatch in forward; not the same bug." + "reason": "Device placement mismatch vs tokenizer return_tensors handling; unrelated." }, { "left": "issue:29870", "right": "issue:29942", "accept": false, - "reason": "Bart generate decoder_attention_mask bug and Flash Attention 2 test failures are unrelated." + "reason": "Bart decoder_attention_mask assignment bug vs Flash Attention 2 test failures; different problems." } ] }, @@ -8971,8 +8908,8 @@ "issue:43450|issue:43575", "issue:43572|issue:44987", "issue:44554|issue:45356", - "issue:42490|issue:43493", "issue:37428|issue:43824", + "issue:42490|issue:43493", "issue:43232|issue:44849", "issue:43295|issue:43493", "issue:44466|issue:45127", @@ -8993,190 +8930,190 @@ "issue:45200|issue:46899", "issue:44792|issue:45563" ], - "prepared_review_unit_hash": "12676e1e6536853b46f180aea05bc526a8306f7591ff034d825b4ec2982b20db", + "prepared_review_unit_hash": "0d5eeed72df9239567d8d922979b3c0510e8b6a5c3964ad73b83b204f755870c", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 26, - "serialized_chars": 35887, - "estimated_input_tokens": 8972, - "estimated_eval_tokens": 18200 + "serialized_chars": 35884, + "estimated_input_tokens": 8971, + "estimated_eval_tokens": 18198 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is a loose semantic mix of unrelated Transformers regressions and docs/test issues; none of the proposed soft pairs look like true duplicates, so I would not merge any of them. I\u2019d treat issue 44521 as the weakest umbrella representative only because it has the most engagement, but the group is too heterogeneous for a strong canonical duplicate target.", - "confidence": 0.9, - "canonical_issue_reason": "Issue 44521 has the strongest engagement signals (discussion and inbound references) and sits on a high-traffic API path, but the cluster is too heterogeneous to support a truly canonical duplicate issue.", + "summary": "The cluster is mostly heterogeneous: nearly all pairs are distinct bugs spanning tokenization, model config/load, multimodal processing, import errors, and training/runtime regressions. I found no duplicate PRs and no strong same-bug pair among the soft candidates.", + "confidence": 0.92, + "canonical_issue_reason": "issue:44521 is the strongest standalone representative: it is open, recently active, and has the most inbound references among the listed issues, while the rest of the set is largely unrelated.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44521 is the best single representative by impact/engagement, though only weakly so because the cluster spans many unrelated bugs.", + "best_issue_reason": "issue:44521 is the best global issue anchor in this set because it is the most active and broadly scoped current bug report, even though the cluster does not form a true duplicate group.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43550", "right": "issue:43844", "accept": false, - "reason": "Different failures: Bamba torch.compile/SDPA correctness vs DeepSpeed ZeRO-3 gradient growth." + "reason": "Different failure modes: torch.compile/SDPA on Bamba vs gradient blow-up under ZeRO-3. No shared bug or code path indicated." }, { "left": "issue:45520", "right": "issue:46178", "accept": false, - "reason": "Unrelated import/backend problems: flash_attn KeyError vs tensorflow_text missing from BACKENDS_MAPPING." + "reason": "Both are import/backend errors, but they involve different subsystems and packages (`flash_attn` vs `tensorflow_text`) with unrelated root causes." }, { "left": "issue:43450", "right": "issue:43575", "accept": false, - "reason": "Video processor batched-shape bug vs Qwen2 TP loading OOM; different code paths and symptoms." + "reason": "Unrelated issues: batched video processor shape vs tensor-parallel OOM on Qwen2. Different code paths and symptoms." }, { "left": "issue:43572", "right": "issue:44987", "accept": false, - "reason": "Both are loading/config regressions, but one is StableLm pad_token_idx and the other is a package/version load failure for a specific model." + "reason": "Both concern model loading/configuration, but the specific bugs differ and there is no evidence of the same underlying regression." }, { "left": "issue:44554", "right": "issue:45356", "accept": false, - "reason": "MPS attention correctness issue vs tokenizer codec/regression in Kimi-K2.5; not the same bug." + "reason": "Completely different areas: MPS attention correctness vs tokenizer codec handling/warning regression." }, { - "left": "issue:42490", - "right": "issue:43493", + "left": "issue:37428", + "right": "issue:43824", "accept": false, - "reason": "Trailing-slash local path warning is unrelated to the SigLIP2 HF-vs-JAX implementation discrepancy." + "reason": "Both are ImportErrors, but for different symbols and model families; this looks like separate missing-export issues." }, { - "left": "issue:37428", - "right": "issue:43824", + "left": "issue:42490", + "right": "issue:43493", "accept": false, - "reason": "Both are import errors, but they involve different missing symbols and different modules/code paths." + "reason": "Path-trailing-slash import warning is unrelated to the SigLIP2 implementation discrepancy." }, { "left": "issue:43232", "right": "issue:44849", "accept": false, - "reason": "Generation kwargs update after sync_gpus is unrelated to the Qwen3.5 output_hidden_states bug." + "reason": "Different generation bugs: sync_gpus kwargs update vs output_hidden_states behavior in Qwen3.5." }, { "left": "issue:43295", "right": "issue:43493", "accept": false, - "reason": "Processor.tokenizer/image passthrough regression is a different bug from the SigLIP2 implementation mismatch." + "reason": "Custom processor/tokenizer regression and SigLIP2 model discrepancy are unrelated failures." }, { "left": "issue:44466", "right": "issue:45127", "accept": false, - "reason": "Serialization inconsistency for lm_head.weight is not the same underlying bug as LoRA merge collapse after vocab extension." + "reason": "Both mention tied weights/lm_head, but one is serialization/device-dependent state and the other is LoRA merge collapse. Not the same concrete bug." }, { "left": "issue:39692", "right": "issue:46192", "accept": false, - "reason": "Docs/example mistakes and silently ignored return_tensors are different issues." + "reason": "Documentation/model mismatch and quantization failure are unrelated to `return_tensors` being ignored when `text_kwargs` is passed." }, { "left": "issue:44521", "right": "issue:45290", "accept": false, - "reason": "Both touch apply_chat_template, but one is zeroed assistant masks and the other is a crash on tool-call messages; not the same failure." + "reason": "Same API surface, but different bugs: all-zero assistant masks vs crash on tool-call assistant messages." }, { "left": "issue:43653", "right": "issue:44112", "accept": false, - "reason": "BigBirdTokenizer special-token registration bug vs GraniteSpeech CI device-override test flake; unrelated." + "reason": "BigBirdTokenizer special-token bug and a GraniteSpeech CI test flake are unrelated." }, { "left": "issue:45381", "right": "issue:46209", "accept": false, - "reason": "Qwen2.5-VL video position ids bug is unrelated to FP8 inference failure on B200." + "reason": "Qwen2.5-VL vision position ids and FP8 inference on B200 are separate hardware/model issues." }, { "left": "issue:44737", "right": "issue:46894", "accept": false, - "reason": "XLNet CPU tensor allocation bug vs CohereASR double-shift training-loss bug; different subsystems." + "reason": "XLNet device placement bug and CohereASR double-shift training bug are unrelated." }, { "left": "issue:28218", "right": "issue:35141", "accept": false, - "reason": "Tokenizer adds extra space after added token, while the other is output embedding reinit after resizing; distinct problems." + "reason": "Tokenizer spacing after added tokens is not the same as embedding reinitialization after resize/post_init." }, { "left": "issue:43278", "right": "issue:43856", "accept": false, - "reason": "Embedding dtype drift in evaluate is not the same as Qwen3 MoE memory inefficiency." + "reason": "Dtype drift between train/eval is unrelated to MoE memory inefficiency." }, { "left": "issue:42915", "right": "issue:43824", "accept": false, - "reason": "FineGrainedFP8Config failure for Qwen3Moe and Qwen2_5_VL import error are unrelated." + "reason": "FineGrainedFP8Config failure on Qwen3Moe and a missing Qwen2.5-VL export are unrelated." }, { "left": "issue:43950", "right": "issue:46285", "accept": false, - "reason": "Non-persistent buffer corruption in from_pretrained is a different regression from the CLIP text-model structure change." + "reason": "Non-persistent buffer corruption in `from_pretrained` is a different regression from CLIP text-model structure inconsistency." }, { "left": "issue:43824", "right": "issue:43975", "accept": false, - "reason": "ImportError for Qwen2_5_VL is unrelated to DeepSeek detokenization regression." + "reason": "Missing export ImportError and DeepSeek detokenization regression do not share an underlying fix." }, { "left": "issue:43452", "right": "issue:46512", "accept": false, - "reason": "gguf_file loading breakage and stale config.dtype after from_config are different lifecycle bugs." + "reason": "gguf_file loading failures are unrelated to stale `config.dtype` after `from_config(dtype=...)`." }, { "left": "issue:43334", "right": "issue:43525", "accept": false, - "reason": "Same symptom class (missing pad_token_id), but different models/configs and no evidence of one shared underlying fix." + "reason": "Both mention missing `pad_token_id`, but they affect different config classes/model families; no evidence of a single shared bug." }, { "left": "issue:43262", "right": "issue:43450", "accept": false, - "reason": "Audio apply_chat_template sampling-rate default bug is unrelated to batched video output shape." + "reason": "Audio chat-template sampling-rate default and video batch-shape handling are different preprocessing bugs." }, { "left": "issue:44442", "right": "issue:45198", "accept": false, - "reason": "AutoTokenizer failing FastSpeech2ConformerTokenizer is a different issue than Wav2Vec2 save/tokenization failure." + "reason": "AutoTokenizer failing for FastSpeech2Conformer and Wav2Vec2 save/tokenization failure are separate audio-tokenizer issues." }, { "left": "issue:45200", "right": "issue:46899", "accept": false, - "reason": "Missing mm_token_type_ids default and unsafe dtype casting in multimodal embedders are distinct multimodal bugs." + "reason": "Text-only Gemma4 fine-tuning mm_token_type_ids issue and 4-bit quantized multimodal dtype-cast issue are different failures." }, { "left": "issue:44792", "right": "issue:45563", "accept": false, - "reason": "Janus image-generation test failure and stale paged-generate warning are not the same code-path problem." + "reason": "Janus image-generation test failure and paged-generate warning behavior are unrelated." } ] }, @@ -9255,8 +9192,8 @@ "issue:43526|issue:46192", "issue:45200|issue:45325", "issue:43827|issue:43901", - "issue:43335|issue:43550", "issue:41720|issue:46209", + "issue:43335|issue:43550", "issue:44561|issue:46234", "issue:43388|issue:43526", "issue:43819|issue:45083", @@ -9264,8 +9201,8 @@ "issue:44610|issue:44857", "issue:46424|issue:46858", "issue:45759|issue:46285", - "issue:45520|issue:46170", "issue:46512|issue:46821", + "issue:45520|issue:46170", "issue:43606|issue:45083", "issue:38175|issue:45412", "issue:44206|issue:45584", @@ -9280,20 +9217,20 @@ "issue:44855|issue:45083", "issue:43278|issue:43825" ], - "prepared_review_unit_hash": "9e864900eb2e405c07ec049e758e6b4646bf2ed5b0d2a0ac236515f274ffbbef", + "prepared_review_unit_hash": "9fa4d0ccde49d7b9ae5156a3f7570d9fe4c0a0be971cedffa6969ea48b57f548", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36749, + "serialized_chars": 36751, "estimated_input_tokens": 9188, "estimated_eval_tokens": 18632 }, @@ -9302,186 +9239,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: most pairs share only a broad subsystem or theme, not the same concrete bug/change, so all soft edges should be rejected. I\u2019d use issue 43611 as the best representative regression in the set.", - "confidence": 0.62, - "canonical_issue_reason": "issue:43611 is the most broadly impactful concrete regression here: a v5.0 loading break tied to `base_model_prefix`, with clear user-facing impact across multiple model classes.", + "summary": "This cluster is mostly a collection of unrelated closed issues spanning model-loading errors, cache/device bugs, docs regressions, and model-specific failures. None of the soft pairs look like the same underlying bug or change, so no pair should be merged as duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "issue:43825 is the most broadly user-facing and clearly framed regression in the set, with a concrete behavior mismatch and a wider impact than the more model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "Same reason: it is a well-scoped, high-impact runtime regression rather than a narrow model-specific edge case or a docs-only issue.", + "best_issue_reason": "issue:43825 is the best representative issue here because it is specific, reproducible, and broadly applicable compared with the many narrow hardware/model/config-only cases.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43819", "right": "issue:44112", "accept": false, - "reason": "Different bugs: DAC latent reconstruction mismatch vs a GraniteSpeech CI device-override test failure." + "reason": "Different subsystems and failures: DAC latent conversion mismatch vs GraniteSpeech stale device override CI test." }, { "left": "issue:44898", "right": "issue:45412", "accept": false, - "reason": "Perceiver interpolation failure and RT-DETR memory leak are unrelated code paths and failure modes." + "reason": "One is a Perceiver image-size/interpolation bug; the other is an RT-DETR memory-release issue." }, { "left": "issue:43526", "right": "issue:46192", "accept": false, - "reason": "Both touch processors, but one is label reduction and the other is `text_kwargs`/`return_tensors` handling." + "reason": "Different code paths: BEiT fast image label reduction vs generic processor argument handling for return_tensors." }, { "left": "issue:45200", "right": "issue:45325", "accept": false, - "reason": "Different multimodal bugs: missing default `mm_token_type_ids` vs RoPE position scaling for still images." + "reason": "Both are multimodal-position related, but they affect different models and different logic (mm_token_type_ids vs rope index scaling)." }, { "left": "issue:43827", "right": "issue:43901", "accept": false, - "reason": "Both are docs issues, but they concern different APIs and different deprecations." + "reason": "Both are docs regressions after v5, but they concern different APIs and different incorrect statements." }, { - "left": "issue:43335", - "right": "issue:43550", + "left": "issue:41720", + "right": "issue:46209", "accept": false, - "reason": "SwitchTransformers config bug and Bamba torch.compile/SDPA failure are distinct problems." + "reason": "Different models and environments: Qwen3 auto device mapping on A800 vs FP8 inference on B200." }, { - "left": "issue:41720", - "right": "issue:46209", + "left": "issue:43335", + "right": "issue:43550", "accept": false, - "reason": "Different GPU/hardware failures with no shared concrete code path." + "reason": "Unrelated bugs: SwitchTransformers config construction vs Bamba torch.compile/SDPA failure." }, { "left": "issue:44561", "right": "issue:46234", "accept": false, - "reason": "Both involve missing utils in v5, but the missing symbols and breakage mechanisms differ." + "reason": "Both mention missing helpers, but they are different symbols with different affected models and import paths." }, { "left": "issue:43388", "right": "issue:43526", "accept": false, - "reason": "Metric gathering drops tuple labels; Beit image processor returns only one label. Different layers and symptoms." + "reason": "Different label-processing bugs in different components; one is metric gathering, the other is image processor label reduction." }, { "left": "issue:43819", "right": "issue:45083", "accept": false, - "reason": "Different helper/model logic: latent STE mismatch vs feature-length helper behavior." + "reason": "Different helper/functions and model families; no shared underlying defect is evident." }, { "left": "issue:44857", "right": "issue:45593", "accept": false, - "reason": "AMP crash in LwDetrImageLoss is unrelated to D-FINE's auxiliary-loss omission." + "reason": "Different failures: AMP/CUDA crash in LwDetrImageLoss vs missing auxiliary losses in D-FINE." }, { "left": "issue:44610", "right": "issue:44857", "accept": false, - "reason": "Processor image-size mismatch and AMP/CUDA loss crash are separate bugs." + "reason": "Processor input-size mismatch vs numerical crash in a different model loss path." }, { "left": "issue:46424", "right": "issue:46858", "accept": false, - "reason": "Cache-length config handling and static-cache slice arithmetic are different defects." + "reason": "Both involve cache internals, but one is about honoring max_cache_length and the other about device-tensor seq length arithmetic under torch.compile." }, { "left": "issue:45759", "right": "issue:46285", "accept": false, - "reason": "Composite Qwen config unwrapping is unrelated to CLIP structure inconsistency." + "reason": "Different model-loading structure issues: composite Qwen config unwrapping vs CLIP text-model attribute inconsistency." }, { - "left": "issue:45520", - "right": "issue:46170", + "left": "issue:46512", + "right": "issue:46821", "accept": false, - "reason": "Both are import/platform issues, but they stem from different dependencies and environments." + "reason": "Unrelated issues: stale config dtype after from_config vs TimesFM window_size AttributeError." }, { - "left": "issue:46512", - "right": "issue:46821", + "left": "issue:45520", + "right": "issue:46170", "accept": false, - "reason": "Stale dtype config state and TimesFM `window_size` AttributeError are unrelated." + "reason": "Different platform-specific import/runtime problems with different dependencies and root causes." }, { "left": "issue:43606", "right": "issue:45083", "accept": false, - "reason": "CPU offload device mismatch is unrelated to qwen3_omni_moe feature-length logic." + "reason": "A CPU offload device mismatch is unrelated to qwen3_omni_moe feature-length helper behavior." }, { "left": "issue:38175", "right": "issue:45412", "accept": false, - "reason": "SigLIP2 zero probabilities and RT-DETR memory leakage are unrelated." + "reason": "Unexpected zero probabilities in SigLIP2 is not the same bug as RT-DETR memory not being freed." }, { "left": "issue:44206", "right": "issue:45584", "accept": false, - "reason": "Unsupported `center` arg crash and Whisper empty-transcription failure are different audio bugs." + "reason": "Different audio/model pipeline bugs with different triggers and failure modes." }, { "left": "issue:43335", "right": "issue:43653", "accept": false, - "reason": "Sparse-layer creation and BigBird special-token registration are separate issues." + "reason": "Different components and symptoms: SwitchTransformers sparse-layer config vs BigBirdTokenizer special-token registration." }, { "left": "issue:37428", "right": "issue:43425", "accept": false, - "reason": "Flash-attention import-symbol failure and generic Torch 2.10 incompatibility are not the same bug." + "reason": "A missing internal import is not the same as a broad torch-version incompatibility report." }, { "left": "issue:43550", "right": "issue:44112", "accept": false, - "reason": "Bamba compile/SDPA runtime failure is unrelated to a stale GraniteSpeech CI test." + "reason": "Model compile/SDPA failure vs a stale CI test; no shared code-path bug." }, { "left": "issue:29942", "right": "issue:32090", "accept": false, - "reason": "Flash Attention test failures and Trainer `_gpu_broadcast_one` TypeError are unrelated." + "reason": "FlashAttention test failures and trainer broadcast NoneType errors are unrelated." }, { "left": "issue:28282", "right": "issue:33357", "accept": false, - "reason": "PyTorch-missing import error and MacOS bus error in CLIP are different failure modes." + "reason": "PyTorch-not-found import error vs MacOS bus error on a community CLIP model are different issues." }, { "left": "issue:43611", "right": "issue:43931", "accept": false, - "reason": "Broad v5 loading regression vs a model-specific Qwen3-VL shape mismatch." + "reason": "Different model-loading failures: base_model_prefix behavior change vs a Qwen3-VL weight-shape mismatch." }, { "left": "issue:44448", "right": "issue:44743", "accept": false, - "reason": "Pegasus v4/v5 output drift and Qwen recurrent-state reset are unrelated." + "reason": "Different regressions: Pegasus output drift across versions vs Qwen3 recurrent-state reset with cache." }, { "left": "issue:43335", "right": "issue:43504", "accept": false, - "reason": "SwitchTransformers sparse-layer config bug vs Beit pretrained-load legacy-field bug." + "reason": "SwitchTransformers config bug and Beit semantic-segmentation legacy-field load failure are unrelated." }, { "left": "issue:44855", "right": "issue:45083", "accept": false, - "reason": "Python 3.13 DebertaV2 import parsing failure is unrelated to qwen3_omni_moe helper behavior." + "reason": "Python 3.13 torch.jit parsing/IndentationError is unrelated to qwen3_omni_moe output-length helper behavior." }, { "left": "issue:43278", "right": "issue:43825", "accept": false, - "reason": "Embedding dtype drift in train/eval is unrelated to the v5 pipeline error-message regression." + "reason": "Different bugs in different areas: dtype drift in eval vs pipeline error-message regression." } ] }, @@ -9556,12 +9493,12 @@ ], "soft_pairs": [ "issue:29127|issue:32090", - "issue:43335|issue:45083", "issue:43720|issue:44112", - "issue:43335|issue:45593", - "issue:43335|issue:44112", + "issue:43335|issue:45083", "issue:37428|issue:44351", "issue:44514|issue:44521", + "issue:43335|issue:44112", + "issue:43335|issue:45593", "issue:45290|issue:46326", "issue:45127|issue:45245", "issue:45588|issue:45859", @@ -9572,227 +9509,227 @@ "issue:44448|issue:44779", "issue:43819|issue:44493", "issue:39692|issue:43526", - "issue:26413|issue:31501", "issue:43606|issue:44857", + "issue:26413|issue:31501", "issue:28218|issue:32090", "issue:45459|issue:46234", - "issue:43701|issue:45083", "issue:39692|issue:44493", "issue:44617|issue:46697", + "issue:43701|issue:45083", "issue:44206|issue:45823", "issue:32090|issue:33357", "issue:42490|issue:43575", - "issue:44360|issue:44561", - "issue:44857|issue:45072" + "issue:44857|issue:45072", + "issue:44360|issue:44561" ], - "prepared_review_unit_hash": "a543cfbcad867e46d9104f1c9badd65c4c065d01f18d0430a1ab32e0e8d2b8b1", + "prepared_review_unit_hash": "f3f7e8029975bef796e344a61cb2987f0ea96fc55b41800f249ffaa894ea6058", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36466, - "estimated_input_tokens": 9117, - "estimated_eval_tokens": 18490 + "serialized_chars": 36464, + "estimated_input_tokens": 9116, + "estimated_eval_tokens": 18488 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated bug reports. The only clear duplicate pair is the repeated Qwen3.5 Moe `from_pretrained` regression; the rest are similar only at a broad subsystem level and should stay separate.", - "confidence": 0.98, - "canonical_issue_reason": "issue:45310 is the earlier and more detailed report, and issue:45478 repeats the same Qwen3.5 Moe `from_pretrained` error verbatim.", + "summary": "This cluster is mostly unrelated issues from different subsystems; the only clear duplicate is the two identical Qwen3.5 MoE from_pretrained reports. Everything else is just broad thematic overlap, not the same underlying bug.", + "confidence": 0.96, + "canonical_issue_reason": "issue:45310 is the older of the exact duplicate pair (45310/45478) and cleanly captures the Qwen3.5 MoE from_pretrained regression.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45310 is the strongest canonical issue candidate because it predates the duplicate and has the fuller original discussion context.", + "best_issue_reason": "issue:45310 is the strongest canonical issue because it is an exact duplicate match with 45478 and is a concrete, narrowly scoped bug report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:29127", "right": "issue:32090", "accept": false, - "reason": "Different failures in different paths: LayoutLMv3 error-message clarity vs Trainer `_gpu_broadcast_one` rejecting `NoneType`." - }, - { - "left": "issue:43335", - "right": "issue:45083", - "accept": false, - "reason": "Different models and bugs: SwitchTransformers sparse-layer creation vs qwen3_omni_moe feature-length helper behavior." + "reason": "Different failures in different code paths: LayoutLMv3 error messaging vs Trainer NoneType GPU broadcast." }, { "left": "issue:43720", "right": "issue:44112", "accept": false, - "reason": "Both are CI/device-related, but one is BitNet weight unpacking during accelerate loading and the other is a stale GraniteSpeech device override test." - }, - { - "left": "issue:43335", - "right": "issue:45593", - "accept": false, - "reason": "SwitchTransformers config bug and D-FINE auxiliary-loss behavior are unrelated." + "reason": "Both involve loading/device handling, but they are model-specific CI failures with different root causes." }, { "left": "issue:43335", - "right": "issue:44112", + "right": "issue:45083", "accept": false, - "reason": "Sparse-layer config bug vs GraniteSpeech stale-device CI failure; not the same code path." + "reason": "Both touch sparse/Transformer internals, but one is SwitchTransformers config creation and the other is a qwen3_omni_moe helper bug." }, { "left": "issue:37428", "right": "issue:44351", "accept": false, - "reason": "Both are import-related, but they concern different missing symbols and different modules." + "reason": "Both are import errors, but they concern different missing symbols and different APIs." }, { "left": "issue:44514", "right": "issue:44521", "accept": false, - "reason": "Both involve `apply_chat_template`, but one is batching/padding crash while the other is incorrect assistant masks for multimodal inputs." + "reason": "Same general apply_chat_template area, but one crashes on batched padding=False input and the other produces all-zero assistant masks for multimodal inputs." + }, + { + "left": "issue:43335", + "right": "issue:44112", + "accept": false, + "reason": "Unrelated: SwitchTransformers config bug vs a GraniteSpeech stale device override CI failure." + }, + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "Different models and symptoms; sparse-layer creation is not the same as missing auxiliary losses." }, { "left": "issue:45290", "right": "issue:46326", "accept": false, - "reason": "Different chat-template bugs: tool-call assistant messages with no content vs `continue_final_message` on `mistral_common` tokenizers." + "reason": "Both concern chat templates, but they fail in different tokenizer flows and on different inputs." }, { "left": "issue:45127", "right": "issue:45245", "accept": false, - "reason": "Extended-vocabulary LoRA merge collapse is unrelated to the 2^24 category limit runtime error." + "reason": "Completely different bugs: LoRA merging/output collapse vs a category-cardinality runtime limit." }, { "left": "issue:45588", "right": "issue:45859", "accept": false, - "reason": "Different failure modes in different components: flash-attention `s_aux=None` crash vs CPU-offload incompatibility in Qwen3_5MoeTextRotaryEmbedding." + "reason": "Different code paths: flash_attention sinkless-model handling vs Qwen3_5Moe rotary embedding CPU-offload incompatibility." }, { "left": "issue:34689", "right": "issue:35141", "accept": false, - "reason": "Model-loading regression for Llama 3.2 Vision is not the same as output embedding reinitialization after resizing embeddings." + "reason": "Both involve model loading/weights, but the actual bugs are unrelated." }, { "left": "issue:42490", "right": "issue:43854", "accept": false, - "reason": "Local-path import warning and GLM-4.7-Flash loading failure are distinct issues." + "reason": "Local-path import warning vs a specific model-loading failure are not the same bug." }, { "left": "issue:46431", "right": "issue:46566", "accept": false, - "reason": "Processor missing a chat template is unrelated to DiffusionGemma weights staying on `meta` with `device_map='auto'`." + "reason": "Processor chat-template absence and DiffusionGemma meta-device loading are unrelated." }, { "left": "issue:45310", "right": "issue:45478", "accept": true, - "reason": "Same exact Qwen3.5 Moe `from_pretrained` error; `45478` is a duplicate of `45310`." + "reason": "Exact same title and same reported Qwen3.5 MoE from_pretrained failure on transformers>=5.4.0." }, { "left": "issue:44448", "right": "issue:44779", "accept": false, - "reason": "Separate version-regression reports for different models and tokenization/generation paths." + "reason": "Both are version regressions, but they affect different models/tokenizers and likely different code paths." }, { "left": "issue:43819", "right": "issue:44493", "accept": false, - "reason": "DAC latent/forward mismatch and unexpected position-id keys are different bugs." + "reason": "DAC from_latents mismatch is unrelated to unexpected position-id keys in model loading." }, { "left": "issue:39692", "right": "issue:43526", "accept": false, - "reason": "SigLIP2 documentation example errors are unrelated to BeitImageProcessorFast `reduce_labels` behavior." + "reason": "One is a docs/example issue for SigLIP2, the other is a BeitImageProcessorFast functional bug." }, { - "left": "issue:26413", - "right": "issue:31501", + "left": "issue:43606", + "right": "issue:44857", "accept": false, - "reason": "Checkpoint-resume epoch-iterator failure is not the same as FineWeb SLM training not starting." + "reason": "Different failures: CPU offload device mismatch vs AMP/CUDA crash in a loss module." }, { - "left": "issue:43606", - "right": "issue:44857", + "left": "issue:26413", + "right": "issue:31501", "accept": false, - "reason": "CPU-offload device mismatch in Bark is unrelated to LwDetrImageLoss AMP CUDA crash." + "reason": "Checkpoint resume iterator failure is not the same as a FineWeb training startup failure." }, { "left": "issue:28218", "right": "issue:32090", "accept": false, - "reason": "Tokenizer extra-space behavior is a different bug from Trainer broadcast type handling." + "reason": "Tokenizer spacing after added tokens is unrelated to Trainer GPU broadcast errors." }, { "left": "issue:45459", "right": "issue:46234", "accept": false, - "reason": "Tokenizer protobuf error masking and missing `is_falcon_mamba_ssm_available` are distinct utility-layer issues." - }, - { - "left": "issue:43701", - "right": "issue:45083", - "accept": false, - "reason": "Checkpoint key mismatch and qwen3_omni_moe helper behavior do not describe the same bug." + "reason": "Tokenizer protobuf error handling and a missing import_utils helper are unrelated." }, { "left": "issue:39692", "right": "issue:44493", "accept": false, - "reason": "Documentation example issues are unrelated to the unexpected position-id key regression." + "reason": "Different issues: SigLIP2 docs/example errors vs unexpected position-id key behavior." }, { "left": "issue:44617", "right": "issue:46697", "accept": false, - "reason": "CUDA OOM in Sam3Video and Swin2SR over-padding are unrelated." + "reason": "CUDA OOM for Sam3Video and Swin2SR image over-padding are unrelated." + }, + { + "left": "issue:43701", + "right": "issue:45083", + "accept": false, + "reason": "resume_from_checkpoint key mismatch is not the same as qwen3_omni_moe output-length helper behavior." }, { "left": "issue:44206", "right": "issue:45823", "accept": false, - "reason": "Unsupported `center` arg crash and Gemma4 PLE device mismatch are different code paths." + "reason": "Different model/runtime bugs: unsupported center arg in LasrFeatureExtractor vs Gemma4 PLE device mismatch." }, { "left": "issue:32090", "right": "issue:33357", "accept": false, - "reason": "Trainer `_gpu_broadcast_one` type error is unrelated to the MacOS CLIP bus error." + "reason": "NoneType GPU broadcast and MacOS bus error on CLIP loading do not share the same root cause." }, { "left": "issue:42490", "right": "issue:43575", "accept": false, - "reason": "Import warning for local paths is unrelated to tensor-parallel OOM on Qwen2-57B-A14B-Instruct." + "reason": "A local-path import warning is unrelated to tensor-parallel OOM on Qwen2-57B." }, { - "left": "issue:44360", - "right": "issue:44561", + "left": "issue:44857", + "right": "issue:45072", "accept": false, - "reason": "DSA indexer discussion and removal of `is_torch_fx_available` breaking trust-remote-code models are unrelated." + "reason": "Both mention numeric precision, but they are different modules and failure modes." }, { - "left": "issue:44857", - "right": "issue:45072", + "left": "issue:44360", + "right": "issue:44561", "accept": false, - "reason": "Different dtype/precision failures in different models; not the same underlying bug." + "reason": "DSA indexer activation question and v5 trust_remote_code import breakage are unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet, and the only accepted pair (issue:45310/issue:45478) is a clear duplicate based on identical titles and highly similar content. The remaining soft-edge verdicts stay conservative by treating broad subsystem similarity as insufficient for duplication." + "feedback": "The summary is grounded in the packet and stays conservative: it identifies the only clearly duplicate pair (issue:45310 / issue:45478) and does not overclaim the softer overlaps. The accepted duplicate verdict is well supported by the identical title and closely matching regression description. No PR-pair mergeability concerns apply here." }, "evaluator_used": true, "retried": false, @@ -9875,9 +9812,9 @@ "issue:10000|issue:15354", "issue:44977|issue:45229", "issue:43278|issue:43381", - "issue:29870|issue:33290", "issue:43927|issue:45698", "issue:44855|issue:46399", + "issue:29870|issue:33290", "issue:45290|issue:46655", "issue:40444|issue:43526", "issue:44857|issue:45588", @@ -9885,223 +9822,223 @@ "issue:32090|issue:33290", "issue:46421|issue:46858", "issue:44561|issue:44779", - "issue:44448|issue:44466", "issue:43873|issue:45229", - "issue:25251|issue:32090", + "issue:44448|issue:44466", "issue:43976|issue:45440", + "issue:25251|issue:32090", "issue:42490|issue:43873", "issue:44568|issue:45491", "issue:43278|issue:43575", "issue:44871|issue:44977" ], - "prepared_review_unit_hash": "5a2f54f25b95b5bdd80872bcb8bf6211efca454a03d380bbd7e7d1519a965201", + "prepared_review_unit_hash": "1c85fcbe3630875a89b5eab988b6444c73d2f7efec45c00daa7097011a137328", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 37072, - "estimated_input_tokens": 9268, - "estimated_eval_tokens": 18792 + "serialized_chars": 37075, + "estimated_input_tokens": 9269, + "estimated_eval_tokens": 18794 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These reports are largely unrelated, model-specific regressions spanning tokenizer behavior, flash-attention, cache handling, serialization, and packaging. The similarity is mostly thematic, not the same underlying bug/change, so none of the soft pairs should be merged.", - "confidence": 0.95, - "canonical_issue_reason": "issue:45920 is the broadest, most representative standalone bug here: a multi-model v5 tokenizer regression with inbound references and strong discussion activity.", + "summary": "The cluster is highly heterogeneous: most items are distinct bug reports spanning different models, subsystems, and failure modes. None of the soft-edge pairs look like true duplicates of the same underlying defect. If one issue has to represent the cluster, the broad v5 tokenizer regression (multiple model families affected) is the strongest anchor.", + "confidence": 0.74, + "canonical_issue_reason": "issue:45920 is the broadest, highest-impact report here: a v5 AutoTokenizer regression affecting multiple model families, with a concrete and widely reproducible symptom.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 has the best combination of breadth, cross-model impact, and external references, making it the strongest issue to anchor this cluster.", + "best_issue_reason": "issue:45920 is the best overall representative because it spans several models, clearly describes a regression, and is more general than the narrower single-model bugs in the set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43550", "right": "issue:45083", "accept": false, - "reason": "Different failures: Bamba torch.compile/SDPA vs qwen3_omni_moe feature-length helper bug." + "reason": "Different models and different failures: torch.compile+SDPA on Bamba vs an undefined helper-variable bug in qwen3_omni_moe." }, { "left": "issue:27957", "right": "issue:34689", "accept": false, - "reason": "Flash-Attn2 support request vs Llama 3.2 Vision model-loading regression." + "reason": "Flash Attention 2 support for XLM-Roberta and a Llama 3.2 Vision loading regression are unrelated problems." }, { "left": "issue:29942", "right": "issue:35141", "accept": false, - "reason": "Failing Flash-Attn2 tests are unrelated to embedding reinitialization after resize." + "reason": "Failing Flash Attention 2 tests and token-embedding resize/post_init reinitialization are separate code paths." }, { "left": "issue:38175", "right": "issue:44493", "accept": false, - "reason": "Zero probabilities in siglip2 is not the same as unexpected position-id keys across many models." + "reason": "Unexpected zero probabilities in siglip2 and unexpected position-id keys in many models are different symptoms and likely different bugs." }, { "left": "issue:29870", "right": "issue:32090", "accept": false, - "reason": "Generate-time decoder mask bug vs Trainer/NoneType broadcast error." + "reason": "Bart generate() decoder mask assignment and Trainer _gpu_broadcast_one NoneType support are unrelated runtime errors." }, { "left": "issue:43278", "right": "issue:43329", "accept": false, - "reason": "Embedding dtype drift on eval is unrelated to undefined vars in multimodal token counting." + "reason": "Embedding dtype drift in evaluate and a multimodal token-count helper using undefined names are distinct issues." }, { "left": "issue:44448", "right": "issue:44977", "accept": false, - "reason": "Different output regression for Pegasus vs Qwen3.5 flash-attention generation crash." + "reason": "Pegasus v4/v5 output differences and Qwen3.5 flash-attention generation failures do not describe the same defect." }, { "left": "issue:45920", "right": "issue:46192", "accept": false, - "reason": "Tokenizer ID regression vs return_tensors being ignored in text_kwargs." + "reason": "Tokenizer ID regressions across many models are not the same as silently ignoring return_tensors when text_kwargs is passed." }, { "left": "issue:45290", "right": "issue:45381", "accept": false, - "reason": "Chat-template tool-call crash and Qwen2.5-VL video position-id bug affect different paths." + "reason": "Chat-template crashes on tool-call assistant messages and Qwen2.5-VL video vision_position_ids issues are separate features." }, { "left": "issue:10000", "right": "issue:15354", "accept": false, - "reason": "German DistilBertModel issue is unrelated to torch.jit.script GeneratorExp export failure." + "reason": "German DistilBertModel issue and GeneratorExp/torch.jit.script export failure are unrelated." }, { "left": "issue:44977", "right": "issue:45229", "accept": false, - "reason": "Flash-attention generation problem is unrelated to Gemma4 multi-GPU OOM." + "reason": "Flash-attention generation behavior and multi-GPU CUDA OOM on Gemma4 are different failure modes." }, { "left": "issue:43278", "right": "issue:43381", "accept": false, - "reason": "Eval-mode gradient checkpointing is a different bug from BF16-to-FP32 dtype changes." - }, - { - "left": "issue:29870", - "right": "issue:33290", - "accept": false, - "reason": "Decoder attention-mask assignment bug is unrelated to deepspeed adafactor OOM." + "reason": "BF16-to-FP32 embedding dtype change in eval is not the same as gradient checkpointing being disallowed in eval mode." }, { "left": "issue:43927", "right": "issue:45698", "accept": false, - "reason": "Custom token IDs lost on save/load is different from loading the wrong custom module after save_pretrained." + "reason": "Saving/loading custom token IDs and loading the wrong custom module after save_pretrained are both serialization-related, but they are different bugs." }, { "left": "issue:44855", "right": "issue:46399", "accept": false, - "reason": "Python 3.13 import parsing error is unrelated to Qwen3.5 kernel registration/decorator failure." + "reason": "Python 3.13 torch.jit parsing/IndentationError and a missing decorator in Qwen3.5 kernel registration are distinct import/runtime issues." + }, + { + "left": "issue:29870", + "right": "issue:33290", + "accept": false, + "reason": "A generation-time UnboundLocalError and deepspeed+adafactor OOM are not the same underlying problem." }, { "left": "issue:45290", "right": "issue:46655", "accept": false, - "reason": "Tool-call chat-template crash is not the same as SmolVLM video padding calculation bug." + "reason": "Chat template tool-call crashes and SmolVLM video padding miscalculation affect different parts of the stack." }, { "left": "issue:40444", "right": "issue:43526", "accept": false, - "reason": "IterableDataset multi-image finetuning failure is unrelated to BeitImageProcessorFast reduce_labels." + "reason": "IterableDataset multi-image finetuning failure and BeitImageProcessorFast label reduction bug are unrelated." }, { "left": "issue:44857", "right": "issue:45588", "accept": false, - "reason": "AMP/CUDA loss crash and flash_attention s_aux=None attribute error are different code paths." + "reason": "LwDetrImageLoss AMP/CUDA crash and flash_attention.py s_aux=None AttributeError are separate code paths." }, { "left": "issue:37428", "right": "issue:42175", "accept": false, - "reason": "Missing flash-attention helper import is unrelated to torch-only pip backend packaging." + "reason": "Flash-attention utility import breakage and missing TensorFlow in a torch-only install are unrelated packaging/import issues." }, { "left": "issue:32090", "right": "issue:33290", "accept": false, - "reason": "Trainer broadcast NoneType error and quantized deepspeed OOM are different symptoms and subsystems." + "reason": "Trainer broadcast NoneType support and deepspeed optimizer OOM are different training-time failures." }, { "left": "issue:46421", "right": "issue:46858", "accept": false, - "reason": "Chunked-prefill recompilation with lazy cache init is distinct from static-cache seq_length tensor arithmetic." + "reason": "Lazy cache init causing recompilation and device-tensor seq-length arithmetic in static cache are both cache-related but not the same bug." }, { "left": "issue:44561", "right": "issue:44779", "accept": false, - "reason": "Removed is_torch_fx_available helper breakage is unrelated to DeepSeek tokenizer regression." - }, - { - "left": "issue:44448", - "right": "issue:44466", - "accept": false, - "reason": "Pegasus output differences do not match the lm_head.weight serialization/tied-weights issue." + "reason": "Removal of is_torch_fx_available breaking trust_remote_code and DeepSeek tokenizer regressions are separate regressions." }, { "left": "issue:43873", "right": "issue:45229", "accept": false, - "reason": "Quantization/offloading behavior is unrelated to Gemma4 multi-GPU CUDA OOM." + "reason": "Quantization offloading problems and Gemma4 multi-GPU OOM are both memory-related but not the same defect." }, { - "left": "issue:25251", - "right": "issue:32090", + "left": "issue:44448", + "right": "issue:44466", "accept": false, - "reason": "Pipeline top_k nesting bug is unrelated to the Trainer broadcast NoneType error." + "reason": "Pegasus output divergence across versions and lm_head.weight serialization differing by device are different issues." }, { "left": "issue:43976", "right": "issue:45440", "accept": false, - "reason": "Python-version compatibility bug is unrelated to DeepSeekV3MoE divergence from remote implementation." + "reason": "Python-version compatibility in Transformers 5.1.0 and DeepSeekV3MoE parity with remote code are unrelated." + }, + { + "left": "issue:25251", + "right": "issue:32090", + "accept": false, + "reason": "Pipeline top_k nesting behavior and Trainer broadcast NoneType errors are unrelated." }, { "left": "issue:42490", "right": "issue:43873", "accept": false, - "reason": "Trailing-slash import warning is not the same as offloading failing with quantization." + "reason": "Trailing-slash local model warning and quantization offloading malfunction are unrelated." }, { "left": "issue:44568", "right": "issue:45491", "accept": false, - "reason": "Tokenizer add_special_tokens regression is unrelated to Gemma3 sliding-window NaN embeddings." + "reason": "mDeBERTa add_special_tokens behavior and Gemma3 NaN embeddings with mixed-length batches are different tokenizer/model bugs." }, { "left": "issue:43278", "right": "issue:43575", "accept": false, - "reason": "Eval embedding dtype drift is unrelated to tensor-parallel OOM on Qwen2-57B." + "reason": "Eval dtype change and Qwen2-57B tensor-parallel OOM are not the same issue." }, { "left": "issue:44871", "right": "issue:44977", "accept": false, - "reason": "Gemma3 eos_token_id inconsistency is a different bug from Qwen3.5 flash-attention generation failure." + "reason": "Gemma eos_token_id config mismatch and Qwen3.5 flash-attention generation problems are unrelated." } ] }, @@ -10177,9 +10114,9 @@ "soft_pairs": [ "issue:43381|issue:43701", "issue:45588|issue:45823", + "issue:41720|issue:45229", "issue:33357|issue:39692", "issue:29127|issue:34567", - "issue:41720|issue:45229", "issue:44977|issue:45563", "issue:45198|issue:45803", "issue:44361|issue:45198", @@ -10208,226 +10145,226 @@ "issue:45372|issue:45459", "issue:42490|issue:43381" ], - "prepared_review_unit_hash": "71cd6015c21faf7a4f6f7e40763cf034a244b7872ef2c0f33546c3eb90daa161", + "prepared_review_unit_hash": "4a729ac2ae51120b355e85f814ade252dffb233b4778854d10a371eb95a6a4cf", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37287, - "estimated_input_tokens": 9322, - "estimated_eval_tokens": 18900 + "serialized_chars": 37284, + "estimated_input_tokens": 9321, + "estimated_eval_tokens": 18898 }, "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. The soft links are generally based on broad lexical similarity (same model family, same subsystem, or same error class) rather than the same underlying bug, so none should be merged as duplicates.", - "confidence": 0.95, - "canonical_issue_reason": "No clear canonical issue: the items span unrelated model-specific regressions, tokenizer/load bugs, training-state issues, quantization problems, and documentation/examples.", + "summary": "This cluster is heterogeneous: the soft pairs span unrelated tokenizer, generation, vision, training, export, and memory issues, so I rejected all duplicate candidates. I\u2019d use Gemma4 multi-GPU CUDA OOM as the best standalone issue anchor because it is a concrete runtime failure with substantial discussion.", + "confidence": 0.9, + "canonical_issue_reason": "issue:45229 is the clearest standalone bug in the set: it has the strongest discussion signal and a concrete, reproducible multi-GPU CUDA OOM failure mode.", "canonical_pr_reason": null, - "best_issue_reason": "No issue is representative enough to act as a reliable duplicate target for the whole cluster.", + "best_issue_reason": "issue:45229 is the most triage-worthy representative issue: specific, user-facing, and likely to be useful as a cluster anchor.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43381", "right": "issue:43701", "accept": false, - "reason": "Both involve training utilities, but one is eval-mode gradient checkpointing and the other is resume-from-checkpoint key mismatch; different code paths." + "reason": "Different bugs: eval-mode gradient checkpointing restriction vs resume-from-checkpoint key mismatch." }, { "left": "issue:45588", "right": "issue:45823", "accept": false, - "reason": "Different failures in Gemma4/flash-attention vs device-map forwarding; same broad area, not the same bug." + "reason": "Different code paths: flash-attention crash on sink-less models vs Gemma4 device-map forward mismatch." + }, + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "Both are GPU inference failures, but one is a Qwen3 device-map assert and the other is a Gemma4 OOM; not the same fix." }, { "left": "issue:33357", "right": "issue:39692", "accept": false, - "reason": "Both mention vision models, but one is a MacOS CLIP bus error and the other is a SigLIP2 documentation/example issue." + "reason": "Unrelated symptoms: MacOS CLIP bus error vs SigLIP2 documentation/example errors." }, { "left": "issue:29127", "right": "issue:34567", "accept": false, - "reason": "Unrelated problems: LayoutLMv3 error messaging vs TrainerState token-count tracking." - }, - { - "left": "issue:41720", - "right": "issue:45229", - "accept": false, - "reason": "Both are inference/OOM-like reports, but they affect different models and different failure modes (cuda assert vs CUDA OOM)." + "reason": "LayoutLMv3 box/error-message issue and TrainerState token-count update issue are unrelated." }, { "left": "issue:44977", "right": "issue:45563", "accept": false, - "reason": "Qwen3.5 flash-attention generation bug vs a stale warning in paged generate(); not the same defect." + "reason": "Flash-attention generation failure and stale warning for num_return_sequences are different bugs." }, { "left": "issue:45198", "right": "issue:45803", "accept": false, - "reason": "Different tokenizer/preprocessing paths and different bugs: save/tokenization failure vs broad exception masking in Fuyu conversion." + "reason": "Wav2Vec2 save/tokenization failure is not the same as a bare-except error-swallowing bug in Fuyu." }, { "left": "issue:44361", "right": "issue:45198", "accept": false, - "reason": "Different tokenizer families and unrelated crashes." + "reason": "Different tokenizer/model failures: MLuke AttributeError vs Wav2Vec2 save/tokenization breakage." }, { "left": "issue:44186", "right": "issue:45198", "accept": false, - "reason": "LayoutLMv2 tokenizer padding/NER crash is unrelated to Wav2Vec2 save/tokenization failure." + "reason": "LayoutLMv2 NER/padding crash and Wav2Vec2 save/tokenization failure are separate issues." }, { "left": "issue:43653", "right": "issue:44743", "accept": false, - "reason": "BigBird tokenizer special-token registration vs Qwen3.5 recurrent-state reset; no shared underlying bug." + "reason": "Tokenizer special-token registration bug and Qwen recurrent-state reset bug are unrelated." }, { "left": "issue:43611", "right": "issue:46121", "accept": false, - "reason": "Base-model-prefix loading regression is unrelated to rope-parameter validation and JSON list handling." + "reason": "Model-loading regression from base_model_prefix is unrelated to rope-validation type handling." }, { "left": "issue:29870", "right": "issue:34689", "accept": false, - "reason": "BART generation kwargs bug vs Llama 3.2 Vision model loading regression; different subsystems and fixes." + "reason": "BART generate decoder-mask bug and Llama 3.2 Vision model-loading regression are different failures." }, { "left": "issue:45372", "right": "issue:45588", "accept": false, - "reason": "Gemma4 processor import/dependency issue vs flash-attention AttributeError; same model family, different concrete failure." + "reason": "Gemma4 processor import failure is a different bug than the flash_attention s_aux=None crash." }, { "left": "issue:42915", "right": "issue:42994", "accept": false, - "reason": "Both are quantization-related, but one is FineGrainedFP8 loading and the other is quantized-model saving." + "reason": "FP8 config loading failure and quantized model saving failure are distinct operations." }, { "left": "issue:43756", "right": "issue:45440", "accept": false, - "reason": "Different model implementations and different symptoms (RoPE-layer mismatch vs DeepSeekV3MoE divergence)." + "reason": "Smollm3 RoPE-layer mismatch and DeepseekV3MoE divergence are different model-implementation bugs." }, { "left": "issue:43653", "right": "issue:44442", "accept": false, - "reason": "BigBird tokenizer special-token bug is unrelated to AutoTokenizer loading FastSpeech2ConformerTokenizer." + "reason": "BigBird special-token decode bug and AutoTokenizer loading failure are unrelated." }, { "left": "issue:29942", "right": "issue:30064", "accept": false, - "reason": "Flash Attention 2 test failures and void-segmentation-map processing are unrelated." + "reason": "Flash Attention 2 test failures and void segmentation-map processing failure are not the same bug." }, { "left": "issue:44206", "right": "issue:44479", "accept": false, - "reason": "Unsupported center argument in image feature extraction is unrelated to a video-input regression in Qwen VL models." + "reason": "Unsupported center-arg regression and video-input regression affect different code paths and models." }, { "left": "issue:45372", "right": "issue:45406", "accept": false, - "reason": "Related Gemma4 area, but one is a mistral_common import failure and the other is a missing _tokenizer attribute in serve; not the same code-path bug." + "reason": "Gemma4 processor import breakage and missing _tokenizer attribute are related only broadly; they are different failure points." }, { "left": "issue:43412", "right": "issue:45588", "accept": false, - "reason": "Executorch export guard/support issues are unrelated to flash-attention sink-less model crashes." + "reason": "Executorch export guard/erfinv problems are unrelated to the flash_attention sink-less crash." }, { "left": "issue:45459", "right": "issue:45712", "accept": false, - "reason": "Tokenizer protobuf error handling vs leftover dummy classes; no shared bug." + "reason": "Protobuf decode-error handling and leftover dummy-class cleanup are completely different issues." }, { "left": "issue:45491", "right": "issue:46693", "accept": false, - "reason": "Both mention FA2/long context, but one is NaN embeddings and the other is a performance regression." + "reason": "Both mention FA2/long-context, but one is a correctness NaN bug in Gemma3 batching and the other is a performance regression." }, { "left": "issue:44265", "right": "issue:45198", "accept": false, - "reason": "torch.export/torch_compilable_check failure is unrelated to Wav2Vec2 save/tokenization failure." + "reason": "torch.export failure with torch_compilable_check is unrelated to Wav2Vec2 save/tokenization failure." }, { "left": "issue:29127", "right": "issue:33357", "accept": false, - "reason": "LayoutLMv3 error-message clarity and CLIP MacOS bus error are unrelated." + "reason": "LayoutLMv3 box-clarity issue and CLIP MacOS bus error are unrelated." }, { "left": "issue:43065", "right": "issue:43994", "accept": false, - "reason": "Dummy Conv2d placeholder in SAM3 pixel decoder is unrelated to SigLIP2 nonsensical AutoModel/pipeline results." + "reason": "Dummy Conv2d cleanup in Sam3PixelDecoder is unrelated to SigLIP2 model/pipeline mismatch." }, { "left": "issue:43065", "right": "issue:45412", "accept": false, - "reason": "Dummy SAM3 layer cleanup is unrelated to RT-DETR memory not being released." + "reason": "Dummy-class cleanup and RT-DETR memory leak are different bugs." }, { "left": "issue:45245", "right": "issue:45563", "accept": false, - "reason": "Category-count limit error and stale generate warning are unrelated." + "reason": "Category-count runtime error and stale warning for num_return_sequences are unrelated." }, { "left": "issue:43575", "right": "issue:43701", "accept": false, - "reason": "Tensor-parallel OOM for a large Qwen model is unrelated to checkpoint key mismatch." + "reason": "Tensor-parallel OOM and checkpoint key mismatch are different failure modes." }, { "left": "issue:43066", "right": "issue:44462", "accept": false, - "reason": "Tokenizer decoder-type migration bug vs AutoTokenizer ignoring tokenizer.json; different loading paths and fixes." + "reason": "Wrong tokenizer decoder type in v5 and ignoring tokenizer.json are separate tokenizer-loading bugs." }, { "left": "issue:45440", "right": "issue:45593", "accept": false, - "reason": "DeepSeekV3MoE divergence is unrelated to D-FINE auxiliary-loss logic." + "reason": "DeepSeek implementation divergence and D-FINE auxiliary-loss behavior are unrelated model bugs." }, { "left": "issue:45372", "right": "issue:45459", "accept": false, - "reason": "Gemma4 processor import regression is unrelated to protobuf decode-error handling in tokenizers." + "reason": "Gemma4 processor import error and protobuf-related tokenizer error handling are different issues." }, { "left": "issue:42490", "right": "issue:43381", "accept": false, - "reason": "Local-path slash import warning is unrelated to eval-mode gradient checkpointing behavior." + "reason": "Local-path trailing-slash import warning and gradient-checkpointing eval-mode restriction are unrelated." } ] }, @@ -10532,214 +10469,214 @@ "issue:43262|issue:44206", "issue:42490|issue:44743" ], - "prepared_review_unit_hash": "8d1828c9158d36f878f86610b93a8cc64216a40d85b77f22cacc11c4556c931b", + "prepared_review_unit_hash": "b4bd286a654f6a491b565464c9315dca0c2914eca89713147f94e508d54dae50", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36563, - "estimated_input_tokens": 9141, - "estimated_eval_tokens": 18538 + "serialized_chars": 36569, + "estimated_input_tokens": 9143, + "estimated_eval_tokens": 18542 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "All proposed pairs look like topical false positives rather than true duplicates. They share broad keywords (e.g. Flash Attention, OOM, Qwen, audio) but not the same concrete code path or fix.", + "summary": "The cluster is mostly heterogeneous issue reports with only superficial overlap around Flash Attention, Qwen, memory/OOM, and processor bugs; no soft-edge pair looks like the same concrete underlying bug or change.", "confidence": 0.96, - "canonical_issue_reason": "No clear canonical issue emerges: the items are semantically mixed, and none of the soft pairs describe the same underlying bug/change strongly enough to anchor a duplicate cluster.", + "canonical_issue_reason": "Issue 46693 is the best representative: it is a specific, recent, clearly described FA2 regression with current activity, while the rest of the cluster is mostly unrelated single-bug reports.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a good global representative because the set spans unrelated regressions and feature requests rather than one coherent problem area.", + "best_issue_reason": "46693 is the most actionable issue in the set because it states a concrete regression, a concrete workload (long-context generation), and has active references; it is the strongest anchor for the cluster even though the cluster itself is not duplicate-heavy.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:29127", "right": "issue:35141", "accept": false, - "reason": "Different bugs: LayoutLMv3 error-message clarity vs token embedding reinitialization in post_init." + "reason": "Different subsystems and failures: LayoutLMv3 error messaging vs token-embedding reinitialization during post_init." }, { "left": "issue:43421", "right": "issue:45491", "accept": false, - "reason": "One is about runtime special-token/post-processor updates; the other is a Gemma3 NaN embedding issue." + "reason": "Tokenizer post-processor update behavior is unrelated to Gemma3 NaN embeddings on mixed-length batching." }, { "left": "issue:43994", "right": "issue:45412", "accept": false, - "reason": "Different subsystems and symptoms: Siglip2 bad outputs vs RT-DETR memory not being released." + "reason": "Different models and bug classes: nonsensical predictions vs model memory not being released." }, { "left": "issue:45412", "right": "issue:46369", "accept": false, - "reason": "Both are memory-related, but one is a GC/memory leak report and the other is model call memory overflow; not the same bug." + "reason": "Both are memory-related, but one is GC/release behavior in RT-DETR and the other is an OOM in sam3-litetext; no shared code-path is evident." }, { "left": "issue:43575", "right": "issue:44387", "accept": false, - "reason": "Both mention OOM, but the causes differ: tensor-parallel loading vs int4 quantization reserved-memory regression." + "reason": "Both mention OOM, but one is tensor-parallel model loading and the other is int4 quantization memory growth; too different to merge." }, { "left": "issue:44442", "right": "issue:45072", "accept": false, - "reason": "Tokenizer loading failure vs dtype mismatch in bfloat16 inference; unrelated failures." + "reason": "Tokenizer loading failure vs dtype mismatch in inference; different components and failure modes." }, { "left": "issue:44464", "right": "issue:45584", "accept": false, - "reason": "Chunked generation with compiled forward is unrelated to Whisper empty-transcription failure after align_special_tokens." + "reason": "Chunked generation with compiled forward is unrelated to Whisper empty-transcription handling after align_special_tokens." }, { "left": "issue:42915", "right": "issue:43931", "accept": false, - "reason": "Qwen3Moe FP8 config failure and Qwen3-VL shape-mismatch loading error are different bugs." + "reason": "Qwen3Moe FP8 config failure is unrelated to Qwen3-VL weight-shape mismatch during loading." }, { "left": "issue:43526", "right": "issue:45834", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug is unrelated to Kosmos2.5 long-OCR index errors." + "reason": "BEiT reduce_labels bug and Kosmos2.5 long-OCR index error are different processor issues." }, { "left": "issue:43526", "right": "issue:46693", "accept": false, - "reason": "Label reduction bug vs FA2 long-context performance regression; no shared code path." + "reason": "Different area and bug type: image processor label reduction vs FA2 long-context performance regression." }, { "left": "issue:43824", "right": "issue:44279", "accept": false, - "reason": "Specific missing import error vs a vague dependency issue; not the same failure." + "reason": "Specific missing import for a Qwen2.5-VL class is not the same as a generic dependency issue." }, { "left": "issue:27957", "right": "issue:29127", "accept": false, - "reason": "Flash Attention 2 support for XLM-Roberta is unrelated to LayoutLMv3 box-validation messaging." + "reason": "XLMRoberta Flash Attention 2 support and LayoutLMv3 error clarity are unrelated." }, { "left": "issue:27957", "right": "issue:33290", "accept": false, - "reason": "Flash Attention 2 support/problem vs deepspeed Adafactor OOM; different areas and fixes." + "reason": "Flash Attention 2 support and AdaFactor/DeepSpeed OOM are different bugs." }, { "left": "issue:45250", "right": "issue:46640", "accept": false, - "reason": "Too generic on both sides; they only share the Flash Attention 2 topic, not a concrete shared bug/change." + "reason": "Both are generic FA2 titles, but neither identifies the same concrete bug or code path; insufficient evidence to merge." }, { "left": "issue:43994", "right": "issue:46369", "accept": false, - "reason": "Siglip2 incorrect results and SAM3 LiteText memory overflow are unrelated model-specific problems." + "reason": "Different model families and symptoms: bad outputs vs memory overflow." }, { "left": "issue:33290", "right": "issue:34689", "accept": false, - "reason": "OOM in deepspeed Adafactor is not the same as a model-loading regression for Llama 3.2 Vision." + "reason": "OOM in DeepSpeed with AdaFactor is unrelated to a model-loading regression for Llama 3.2 Vision." }, { "left": "issue:43526", "right": "issue:44186", "accept": false, - "reason": "Different image/tokenization bugs: label reduction in BeitImageProcessorFast vs LayoutLMv2Tokenizer crashes." + "reason": "ImageProcessor reduce_labels bug and LayoutLMv2Tokenizer padding/NER crash are different components." }, { "left": "issue:33290", "right": "issue:35141", "accept": false, - "reason": "Optimizer OOM and embedding reinitialization are unrelated issues." + "reason": "Optimizer-related OOM and embedding reinitialization are unrelated." }, { "left": "issue:43334", "right": "issue:43531", "accept": false, - "reason": "Missing pad_token_id in Qwen3-VL text config is unrelated to a sliding_window bug in Qwen3-MoE." + "reason": "Missing pad_token_id on Qwen3-VL text config is unrelated to sliding_window behavior in Qwen3-MoE." }, { "left": "issue:46121", "right": "issue:46399", "accept": false, - "reason": "RoPE validation TypeError and rotary-pos-emb registration failure are distinct code paths." + "reason": "Both involve RoPE-adjacent code, but one is parameter conversion validation and the other is kernel registration/decorator failure; not the same bug." }, { "left": "issue:45588", "right": "issue:45600", "accept": false, - "reason": "Flash-attention sink-less AttributeError is unrelated to stale auto_mappings config references." + "reason": "Flash-attention sink handling and removed auto_mappings configs are separate CI/runtime issues." }, { "left": "issue:43381", "right": "issue:43873", "accept": false, - "reason": "Gradient checkpointing in eval mode and quantization offloading are different concerns." + "reason": "Gradient checkpointing in eval mode and quantization offloading are unrelated behaviors." }, { "left": "issue:44514", "right": "issue:45381", "accept": false, - "reason": "Batched apply_chat_template padding crash is not the same as wrong vision_position_ids in Qwen2.5-VL video input." + "reason": "Both are Qwen2.5-VL issues, but one is batched chat-template padding and the other is video vision_position_ids; different bugs." }, { "left": "issue:43761", "right": "issue:45859", "accept": false, - "reason": "CLIP hidden_states regression and Qwen3_5Moe CPU-offload incompatibility are unrelated." + "reason": "CLIP hidden_states regression and Qwen3.5MoE CPU-offload incompatibility are unrelated." }, { "left": "issue:43646", "right": "issue:44077", "accept": false, - "reason": "Generic custom-model initialization break is broader than the specific patchtsmixer post_init policy issue." + "reason": "Custom model initialization breakage is not the same as patchtsmixer's post_init contract issue." }, { "left": "issue:44977", "right": "issue:45245", "accept": false, - "reason": "Qwen3.5 flash-attention generation and a categories-limit runtime error are unrelated." + "reason": "Flash-attention generation issues and a categories-limit RuntimeError are unrelated." }, { "left": "issue:44743", "right": "issue:44811", "accept": false, - "reason": "Recurrent-state reset with cache is unrelated to Whisper batch_decode skip_special_tokens handling." + "reason": "Qwen3.5 recurrent-state reset and Whisper skip_special_tokens handling are different code paths." }, { "left": "issue:44442", "right": "issue:44898", "accept": false, - "reason": "AutoTokenizer loading failure and Perceiver non-default-resolution failure are different bugs." + "reason": "FastSpeech2ConformerTokenizer loading and Perceiver interpolation failure are unrelated." }, { "left": "issue:43262", "right": "issue:44206", "accept": false, - "reason": "Audio chat-template sampling-rate defaulting and an unsupported center arg in LasrFeatureExtractor are separate issues." + "reason": "Audio chat-template sample-rate defaulting and feature-extractor center-arg crashes are different audio bugs." }, { "left": "issue:42490", "right": "issue:44743", "accept": false, - "reason": "Trailing-slash local-model import warning has nothing to do with recurrent state reset in Qwen3." + "reason": "Trailing-slash import warning and Qwen3.5 recurrent-state reset are unrelated." } ] }, @@ -10817,8 +10754,8 @@ "issue:43720|issue:44442", "issue:42175|issue:46234", "issue:27957|issue:29870", - "issue:43493|issue:43827", "issue:43316|issue:43525", + "issue:43493|issue:43827", "issue:44062|issue:45907", "issue:44077|issue:44479", "issue:45325|issue:46697", @@ -10829,215 +10766,215 @@ "issue:43316|issue:45759", "issue:43761|issue:44079", "issue:44190|issue:44496", - "issue:43761|issue:44188", "issue:44987|issue:46178", - "issue:43931|issue:45216", "issue:44442|issue:45803", + "issue:43931|issue:45216", + "issue:43761|issue:44188", "issue:43232|issue:43388", - "issue:42907|issue:45698", "issue:43701|issue:46506", + "issue:42907|issue:45698", "issue:42617|issue:43575", "issue:27957|issue:32090", - "issue:43550|issue:43701", "issue:43575|issue:43742", + "issue:43550|issue:43701", "issue:43606|issue:43701" ], - "prepared_review_unit_hash": "1a00ed5b77468b8943edc35a61c3a957c8ea10ec0b0a291dfa3de6abccb653a9", + "prepared_review_unit_hash": "b47d8598c56dbc6b1b0f2183eeca55a20715f9709a93923d5c2cc40d5827fb2d", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36217, - "estimated_input_tokens": 9055, - "estimated_eval_tokens": 18366 + "serialized_chars": 36215, + "estimated_input_tokens": 9054, + "estimated_eval_tokens": 18364 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: it mixes unrelated bugs in generation, loading, configs, quantization/offloading, tokenizer behavior, docs, and distributed training. None of the soft pairs appears to be the same underlying issue, so I rejected them all.", - "confidence": 0.95, - "canonical_issue_reason": "If a single representative is needed, issue 43701 is the clearest self-contained bug report in the set, but the cluster is too diverse to justify a true canonical duplicate.", + "summary": "This cluster is not a real duplicate set; it spans unrelated issues across model loading, configs, generation, distributed training, docs, and processors. All soft edges should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": "Issue 43701 is the closest representative only because it has multiple weak loading/runtime links, but the overall cluster is too heterogeneous to treat as one duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43701 is the most broadly readable and concrete report, but only as a weak representative; it is not meaningfully connected to the rest of the cluster.", + "best_issue_reason": "Issue 43701 is the best available cluster representative due to repeated soft-linking, but it is still not a strong duplicate anchor.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45003", "right": "issue:45440", "accept": false, - "reason": "Different bug classes and code paths: unsafe sys.modules access vs divergence in DeepseekV3MoE implementation." + "reason": "Different bugs: unsafe sys.modules access vs a DeepSeekV3 model divergence. No shared code-path problem." }, { "left": "issue:43720", "right": "issue:44442", "accept": false, - "reason": "BitNet accelerate-loading failure and tokenizer-loading failure are unrelated." + "reason": "Unrelated failures: BitNet loading/packed weights vs tokenizer loading for FastSpeech2Conformer." }, { "left": "issue:42175", "right": "issue:46234", "accept": false, - "reason": "Package extra dependency coverage vs missing import utility symbol; not the same bug." + "reason": "Different backend/package issues: pip extras omitting tensorflow vs a missing import_utils availability helper." }, { "left": "issue:27957", "right": "issue:29870", "accept": false, - "reason": "FlashAttention support for XLMRoberta is unrelated to BART generate() decoder_attention_mask initialization." + "reason": "Distinct model/runtime problems: XLM-RoBERTa Flash Attention 2 vs BART generate() decoder_attention_mask bug." }, { - "left": "issue:43493", - "right": "issue:43827", + "left": "issue:43316", + "right": "issue:43525", "accept": false, - "reason": "Model implementation discrepancy vs stale docs referencing pipeline(); different issues." + "reason": "Both are config-related, but one is a Gemma3TextConfig API discrepancy and the other is a missing Llama4Config attribute; not the same bug." }, { - "left": "issue:43316", - "right": "issue:43525", + "left": "issue:43493", + "right": "issue:43827", "accept": false, - "reason": "Both are config-related, but they concern different models and different missing/changed fields." + "reason": "Completely different domains: SigLIP2 implementation mismatch vs docs still referencing removed pipeline() APIs." }, { "left": "issue:44062", "right": "issue:45907", "accept": false, - "reason": "Tokenizer AddedToken argument duplication and list_repo_files revision forwarding are unrelated." + "reason": "Tokenizer AddedToken construction error vs Hub repo file listing not forwarding revision; unrelated paths." }, { "left": "issue:44077", "right": "issue:44479", "accept": false, - "reason": "Optional post_init validation for patchtsmixer is not the same as the Qwen video-input regression." + "reason": "PatchTSMixer post_init validation issue vs Qwen video input regression; different models and failure modes." }, { "left": "issue:45325", "right": "issue:46697", "accept": false, - "reason": "Qwen2.5-VL RoPE scaling bug and Swin2SR padding bug affect different preprocessing logic." + "reason": "Different preprocessing bugs: Qwen2.5-VL rope indexing vs Swin2SR over-padding." }, { "left": "issue:43873", "right": "issue:44871", "accept": false, - "reason": "Quantization/offloading behavior and Gemma eos_token_id mismatch are not the same underlying defect." + "reason": "Quantization/offloading behavior vs Gemma-3 eos_token_id config mismatch; not the same underlying defect." }, { "left": "issue:44855", "right": "issue:45137", "accept": false, - "reason": "Python 3.13 torch.jit parsing/indentation issue is unrelated to DeepSpeed ZeRO3 deque underflow." + "reason": "Python 3.13 torch.jit parsing error vs DeepSpeed ZeRO3 deque underflow; unrelated failures." }, { "left": "issue:43643", "right": "issue:44913", "accept": false, - "reason": "Remote-code config field loss vs GPTNeoX rotary_pct persistence are different config-loading problems." + "reason": "Both touch config persistence, but one is trust_remote_code missing fields and the other is GPTNeoX rotary_pct reload behavior; not the same bug." }, { "left": "issue:43749", "right": "issue:43931", "accept": false, - "reason": "FSDP CPU RAM-efficient loading breakage and Qwen3-VL shape mismatch are distinct loading failures." + "reason": "Different loading problems: FSDP CPU RAM efficient loading vs Qwen3-VL weight shape mismatch." }, { "left": "issue:43316", "right": "issue:45759", "accept": false, - "reason": "Gemma3TextConfig API discrepancy and composite Qwen multimodal config unwrapping are different config/model-selection bugs." + "reason": "Both involve config objects, but one is an API discrepancy in Gemma3TextConfig and the other is AutoModelForCausalLM.from_config not unwrapping text_config." }, { "left": "issue:43761", "right": "issue:44079", "accept": false, - "reason": "CLIPVisionModel hidden_states regression and ModelOutput key assignment when None are separate runtime behaviors." + "reason": "CLIPVisionModel hidden_states regression and ModelOutput key assignment are related at a high level but not clearly the same concrete bug." }, { "left": "issue:44190", "right": "issue:44496", "accept": false, - "reason": "Local dataset loading in a training script is unrelated to AutoConfig failing on an unrecognized model." + "reason": "Local dataset loading failure in a training script vs unrecognized model/config loading for Olmo-Hybrid-Instruct." + }, + { + "left": "issue:44987", + "right": "issue:46178", + "accept": false, + "reason": "Different backend issues: transformers>=5.1.0 loading regression vs tensorflow_text backend mapping error." }, { "left": "issue:44442", "right": "issue:45803", "accept": false, - "reason": "Tokenizer loading failure for FastSpeech2Conformer and bare except swallowing interrupts/errors in FuyuBatchFeature are unrelated." + "reason": "Tokenizer loading failure vs FuyuBatchFeature bare except swallowing errors; unrelated code paths." }, { - "left": "issue:43232", - "right": "issue:43388", + "left": "issue:43931", + "right": "issue:45216", "accept": false, - "reason": "Generation kwargs update after sync_gpus and gather_for_metrics tuple-label dropping affect different subsystems." + "reason": "Different model-loading regressions: Qwen3-VL shape mismatch vs Qwen3.5 save_pretrained checkpoint corruption." }, { - "left": "issue:42907", - "right": "issue:45698", + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "CLIPVisionModel output regression vs attention-kernel divergence under torch.compile; not the same issue." + }, + { + "left": "issue:43232", + "right": "issue:43388", "accept": false, - "reason": "Saving dequantized Ministrals/Devstrals and loading the wrong custom module after save_pretrained are distinct save/load issues." + "reason": "Generation kwargs sync issue vs gather_for_metrics dropping tuple label elements; unrelated." }, { "left": "issue:43701", "right": "issue:46506", "accept": false, - "reason": "Checkpoint-key mismatch and CPU/CUDA output divergence are not the same bug." + "reason": "resume_from_checkpoint key mismatch is unrelated to CPU/CUDA divergence in PPDocLayoutV3." + }, + { + "left": "issue:42907", + "right": "issue:45698", + "accept": false, + "reason": "Saving dequantized models vs loading the wrong custom module after save_pretrained; both involve save/load but different bugs." }, { "left": "issue:42617", "right": "issue:43575", "accept": false, - "reason": "A script launch problem and a tensor-parallel OOM on a specific model are different failures." + "reason": "3d_parallel.py runtime failure vs tensor-parallel OOM when loading a specific Qwen2 model; not the same defect." }, { "left": "issue:27957", "right": "issue:32090", "accept": false, - "reason": "FlashAttention2 support for XLMRoberta is unrelated to Trainer GPU broadcast rejecting None." - }, - { - "left": "issue:43550", - "right": "issue:43701", - "accept": false, - "reason": "torch.compile SDPA failure in Bamba-9B-v2 is unrelated to resume_from_checkpoint key mismatch." + "reason": "Flash Attention 2 model issue vs Trainer gpu_broadcast_one NoneType error; unrelated." }, { "left": "issue:43575", "right": "issue:43742", "accept": false, - "reason": "Qwen2-57B tensor-parallel OOM and MobileLLM key error do not share a concrete code-path." - }, - { - "left": "issue:43606", - "right": "issue:43701", - "accept": false, - "reason": "CPU offload device mismatch and checkpoint key mismatch are different bugs in different flows." - }, - { - "left": "issue:43761", - "right": "issue:44188", - "accept": false, - "reason": "Hidden_states regression in CLIPVisionModel and attention-kernel divergence under allow_is_bidirectional_skip are unrelated." + "reason": "Tensor-parallel OOM for Qwen2-57B vs key error when loading MobileLLM-125M; distinct loading failures." }, { - "left": "issue:44987", - "right": "issue:46178", + "left": "issue:43550", + "right": "issue:43701", "accept": false, - "reason": "Version-specific model-loading failure and missing BACKENDS_MAPPING entry for tensorflow_text are not the same issue." + "reason": "torch.compile/SDPA failure in Bamba vs resume_from_checkpoint key mismatch; different subsystems and symptoms." }, { - "left": "issue:43931", - "right": "issue:45216", + "left": "issue:43606", + "right": "issue:43701", "accept": false, - "reason": "Shape mismatch when loading Qwen3-VL and incorrect save_pretrained output for Qwen3.5 are separate load/save bugs." + "reason": "CPU offload device mismatch for bark-small vs checkpoint key mismatch; unrelated." } ] }, @@ -11116,11 +11053,11 @@ "issue:43295|issue:43825", "issue:16998|issue:22355", "issue:43295|issue:44871", - "issue:44871|issue:45584", "issue:44464|issue:46421", + "issue:44871|issue:45584", "issue:43316|issue:43716", - "issue:44077|issue:44534", "issue:42175|issue:43316", + "issue:44077|issue:44534", "issue:44871|issue:46050", "issue:43646|issue:44913", "issue:45803|issue:46365", @@ -11135,24 +11072,24 @@ "issue:43761|issue:44479", "issue:43504|issue:45698", "issue:45542|issue:46234", - "issue:43352|issue:46640", "issue:43540|issue:44514", - "issue:43572|issue:46178" + "issue:43572|issue:46178", + "issue:43352|issue:46640" ], - "prepared_review_unit_hash": "d088350cedc2175181464014623910b4e31cfc927d685804f0da835cd54e8bcc", + "prepared_review_unit_hash": "518b60358e2439ab4c8890a005e64615beb9759b4977b7aaaffbcc84650b160e", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 35662, + "serialized_chars": 35663, "estimated_input_tokens": 8916, "estimated_eval_tokens": 18088 }, @@ -11161,174 +11098,174 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is a loose grab-bag of unrelated Transformers issues. None of the soft pairs look like the same concrete bug or change, so all are rejected.", - "confidence": 0.87, - "canonical_issue_reason": "issue:45698 is the strongest standalone bug report here: it is open, has the most discussion, and describes a concrete reproducible regression in from_pretrained/save_pretrained behavior.", + "summary": "The cluster is mostly a grab bag of unrelated Transformers bug reports, loosely unified by v5 regressions and model/loading/config edge cases. None of the soft pairs look like the same underlying bug or change, so I would not merge any of them as duplicates.", + "confidence": 0.91, + "canonical_issue_reason": "issue:43646 is the broadest umbrella for the cluster\u2019s recurring theme of Transformers v5 breaking custom model initialization/loading behavior.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45698 is the best overall issue candidate because it is specific, user-facing, and already has active investigation; the other issues are narrower or clearly unrelated to one another.", + "best_issue_reason": "issue:45698 is the most actionable representative: it is open, actively discussed, and specifically about a custom-model loading regression after save_pretrained/from_pretrained.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:25251", "right": "issue:28282", "accept": false, - "reason": "Different bugs: pipeline output shape regression vs missing PyTorch import for AutoModel." + "reason": "Different subsystems and failures: pipeline output shape vs missing PyTorch backend import." }, { "left": "issue:46234", "right": "issue:46291", "accept": false, - "reason": "One is a missing import_utils symbol; the other is a kernels dependency/version check on importing pipelines." + "reason": "One is a missing utility symbol, the other is a kernels dependency/version requirement on pipelines." }, { "left": "issue:43295", "right": "issue:43825", "accept": false, - "reason": "Both are v5 regressions, but one is about processor/tokenizer access for custom models and the other is an incorrect pipeline task message." + "reason": "Different regressions: processor/tokenizer image handling vs pipeline translation error messaging." }, { "left": "issue:16998", "right": "issue:22355", "accept": false, - "reason": "Unrelated: model_max_length handling for DeBERTa-V3 vs missing transformers.onnx module." + "reason": "Unrelated: model_max_length behavior for DeBERTa vs ONNX module import failure." }, { "left": "issue:43295", "right": "issue:44871", "accept": false, - "reason": "Different model families and failures: processor/tokenizer regression vs Gemma eos_token_id mismatch." + "reason": "Both are v5-era regressions, but they affect different models and code paths (multimodal processor vs Gemma EOS config)." }, { - "left": "issue:44871", - "right": "issue:45584", + "left": "issue:44464", + "right": "issue:46421", "accept": false, - "reason": "Gemma eos token configuration bug vs Whisper empty-transcription failure after token alignment." + "reason": "Both mention chunking/caching, but one is inconsistent generation output and the other is spurious recompilation." }, { - "left": "issue:44464", - "right": "issue:46421", + "left": "issue:44871", + "right": "issue:45584", "accept": false, - "reason": "Chunked generation inconsistency with compiled forward is not the same as spurious recompilation from cache lazy init." + "reason": "Different models and bugs: Gemma EOS config mismatch vs Whisper empty-transcription generation failure." }, { "left": "issue:43316", "right": "issue:43716", "accept": false, - "reason": "API discrepancy in Gemma3TextConfig is unrelated to Mistral-3 image preprocessor/model dtype mismatch." + "reason": "API discrepancy in config classes is not the same as image preprocessor/model dtype mismatch." }, { - "left": "issue:44077", - "right": "issue:44534", + "left": "issue:42175", + "right": "issue:43316", "accept": false, - "reason": "PatchTSMixer post_init restriction is a model-class API issue, not the non-persistent buffer junk-fill regression." + "reason": "Backend packaging/install issue vs config API inconsistency; not the same bug." }, { - "left": "issue:42175", - "right": "issue:43316", + "left": "issue:44077", + "right": "issue:44534", "accept": false, - "reason": "Backend packaging/install issue vs config API discrepancy; no shared underlying defect." + "reason": "PatchTSMixer post_init validation and non-persistent buffer initialization are separate changes." }, { "left": "issue:44871", "right": "issue:46050", "accept": false, - "reason": "Gemma eos_token_id inconsistency is distinct from quantized model save/load breakage." + "reason": "Gemma EOS token mismatch is unrelated to sinq quantization save/load breakage." }, { "left": "issue:43646", "right": "issue:44913", "accept": false, - "reason": "General custom model initialization breakage is not the same as GPTNeoX rotary_pct not persisting on reload." + "reason": "Broad custom initialization regression vs GPTNeoX config field not round-tripping on reload." }, { "left": "issue:45803", "right": "issue:46365", "accept": false, - "reason": "Bare except swallowing errors in Fuyu conversion is unrelated to Trainer import being blocked by datasets." + "reason": "Different import/runtime problems: Fuyu tensor conversion exception handling vs Trainer import breaking with datasets installed." }, { "left": "issue:44162", "right": "issue:44792", "accept": false, - "reason": "ESM2 breakage and a janus image-generation test failure are different models and failure modes." + "reason": "ESM2 model breakage and a janus image-generation test failure are not the same underlying issue." }, { "left": "issue:43452", "right": "issue:44496", "accept": false, - "reason": "Both concern model loading, but one is gguf_file handling for from_pretrained and the other is missing model_type in config.json." + "reason": "Both concern model loading, but one is gguf_file handling and the other is missing model_type in config.json." }, { "left": "issue:43404", "right": "issue:43441", "accept": false, - "reason": "lm_head tying in Mistral3ForConditionalGeneration is a different bug from a FlashAttention support failure in Ministral-3." + "reason": "Weight tying in Mistral3 and FlashAttention support errors for Ministral-3 are distinct bugs." }, { "left": "issue:44242", "right": "issue:46894", "accept": false, - "reason": "Both are loss-related, but they affect different models and mechanisms; the training-loss double-shift bug is not the same as missing load-balancing loss." + "reason": "Both are training-loss issues, but they affect different models and different loss computation bugs." }, { "left": "issue:45632", "right": "issue:46424", "accept": false, - "reason": "Cache path collision for local trust_remote_code models is unrelated to max_cache_length only applying to quantized cache." + "reason": "Cache path collision for trust_remote_code is unrelated to max_cache_length handling in quantized cache." }, { "left": "issue:45803", "right": "issue:46170", "accept": false, - "reason": "Different environment failures: Fuyu exception handling vs Windows/ROCm torch.distributed dependency." + "reason": "Bare except swallowing errors is unrelated to Windows/ROCm import failures around torch.distributed." }, { "left": "issue:44610", "right": "issue:46369", "accept": false, - "reason": "Processor input-size mismatch for OmDet-Turbo is unrelated to sam3-litetext memory overflow." + "reason": "Processor input-size mismatch and sam3-litetext memory overflow are different failure modes." }, { "left": "issue:43650", "right": "issue:46202", "accept": false, - "reason": "A malformed issue title about data is unrelated to Gemma pipeline roundtrip failures." + "reason": "Placeholder/data issue vs Gemma pipeline round-trip tokenization bug; no shared underlying defect." }, { "left": "issue:43761", "right": "issue:44479", "accept": false, - "reason": "CLIPVisionModel hidden_states regression is not the same as the Qwen video-input regression." + "reason": "Hidden-states regression in CLIPVisionModel is not the same as Qwen VL video-input regression." }, { "left": "issue:43504", "right": "issue:45698", "accept": false, - "reason": "Legacy field handling in Beit loading is a different concrete bug from wrong custom module loading after save_pretrained." + "reason": "Both involve loading saved models, but one is a legacy field in BEiT presets and the other is wrong custom-module resolution after save/load." }, { "left": "issue:45542", "right": "issue:46234", "accept": false, - "reason": "TensorBoard/TensorFlow backend packaging issue is unrelated to a missing import_utils helper." - }, - { - "left": "issue:43352", - "right": "issue:46640", - "accept": false, - "reason": "The FlashAttention-2 support complaint is too vague to equate with the NemotronHForCausalLM support error." + "reason": "TensorFlow backend packaging issue vs missing import_utils symbol; unrelated." }, { "left": "issue:43540", "right": "issue:44514", "accept": false, - "reason": "Qwen3OmniMoe video-processing ValueError and Qwen2_5_VLProcessor batched apply_chat_template crash are different paths." + "reason": "Different Qwen multimodal bugs: video processing in Qwen3OmniMoe vs batched chat-template padding crash in Qwen2.5-VL." }, { "left": "issue:43572", "right": "issue:46178", "accept": false, - "reason": "StableLmConfig pad_token_idx regression is unrelated to the tensorflow_text backend mapping error." + "reason": "StableLm config field regression and tensorflow_text backend registration failure are unrelated." + }, + { + "left": "issue:43352", + "right": "issue:46640", + "accept": false, + "reason": "FlashAttention support errors for different models are not the same code-path problem." } ] }, @@ -11402,8 +11339,8 @@ "issue:46897" ], "soft_pairs": [ - "issue:44792|issue:45600", "issue:43065|issue:44617", + "issue:44792|issue:45600", "issue:43388|issue:43688", "issue:44792|issue:44977", "issue:45127|issue:45200", @@ -11416,9 +11353,9 @@ "issue:44734|issue:45406", "issue:46082|issue:46697", "issue:45632|issue:46617", - "issue:44869|issue:46832", "issue:44855|issue:45588", "issue:45245|issue:46858", + "issue:44869|issue:46832", "issue:43827|issue:43994", "issue:43673|issue:46421", "issue:43931|issue:44060", @@ -11430,190 +11367,196 @@ "issue:25251|issue:27957", "issue:43976|issue:44855" ], - "prepared_review_unit_hash": "aed5e4635fb2ee774cfb77df6f6442aab4951b77dd104f895847af99376ab2c9", + "prepared_review_unit_hash": "75acb37c90654aa7fb4d2a4efbb4521e1b88608b9e66da2cf5e726ac7501f832", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 35949, - "estimated_input_tokens": 8988, - "estimated_eval_tokens": 18232 + "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": "The cluster is mostly heterogeneous; several items are only loosely related by subsystem, but none of the soft-edge pairs look like the same concrete bug or change. The best canonical issue among the set is the non-persistent-buffer regression report.", - "confidence": 0.45, - "canonical_issue_reason": "Issue 43644 is the clearest and earliest report of a concrete regression ('non-persistent buffers filled with junk'), and it is the most canonical-looking candidate in the set.", + "summary": "No soft-edge pair looks like the same underlying bug/change; they mostly share only broad subsystem keywords. The clearest recurring theme in the set is the v5 non-persistent-buffer regression, best represented by the broader buffer-junk report.", + "confidence": 0.88, + "canonical_issue_reason": "issue:44534 is the most canonical issue in this set: it states the shared regression most broadly and cleanly, and is the best representative of the repeated non-persistent-buffer problem.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43644 is the most specific, actionable, and representative report among the issues here; it describes a clear runtime regression rather than a broad symptom or a secondary follow-up.", + "best_issue_reason": "issue:44534 is the strongest global representative because it captures a concrete, repository-wide v5 regression without overfitting to a narrower version string.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44792", - "right": "issue:45600", + "left": "issue:43065", + "right": "issue:44617", "accept": false, - "reason": "Different bug surfaces: a janus test failure versus a removed Sam3LiteText config reference." + "reason": "Same model family (Sam3) but different failures: a dummy Conv2d implementation versus CUDA OOM in video inference." }, { - "left": "issue:43065", - "right": "issue:44617", + "left": "issue:44792", + "right": "issue:45600", "accept": false, - "reason": "Unrelated symptoms: a dummy Conv2d implementation detail versus a CUDA OOM report." + "reason": "Different code paths: janus image-generation test failure versus CI breakage from removed Sam3LiteText config mappings." }, { "left": "issue:43388", "right": "issue:43688", "accept": false, - "reason": "Different training/eval bugs: metrics label truncation versus auxiliary loss normalization." + "reason": "One is a dataloader/gather bug dropping labels; the other is an MoE auxiliary-loss normalization issue." }, { "left": "issue:44792", "right": "issue:44977", "accept": false, - "reason": "Same general area of generation, but different models and code paths; not the same bug." + "reason": "Unrelated model-specific failures: janus test generation issue versus Qwen3.5 flash-attention generation behavior." }, { "left": "issue:45127", "right": "issue:45200", "accept": false, - "reason": "One is about LoRA merge/tied embeddings, the other about mm_token_type_ids defaults; separate issues." + "reason": "LoRA merge / tied-embedding collapse is a different bug from Gemma 4 mm_token_type_ids defaulting in text-only fine-tuning." }, { "left": "issue:43673", "right": "issue:44743", "accept": false, - "reason": "Both involve cache-like behavior, but one is missing cache in chunked prefill and the other resets recurrent states." + "reason": "Both involve cache/state handling, but one is missing GenerationMixin cache in chunked prefill while the other resets recurrent states for Qwen3.5 modular code." }, { "left": "issue:41950", "right": "issue:43475", "accept": false, - "reason": "Different SAM3 failures: pipeline processor lookup versus a missing vision encoder attribute." + "reason": "Video-classification pipeline processor lookup is unrelated to a SAM 3 VisionEncoderOutput attribute error." }, { "left": "issue:44190", "right": "issue:44964", "accept": false, - "reason": "Different loading paths: local dataset script failure versus a specific multimodal model load failure." + "reason": "Local dataset loading in a training script is different from a multimodal model load failure in latest transformers." }, { "left": "issue:43525", "right": "issue:43531", "accept": false, - "reason": "Unrelated model config/attention bugs: missing pad_token_id versus sliding_window behavior." + "reason": "LlamaConfig pad_token_id handling is unrelated to the Qwen3-MoE sliding_window bug." }, { "left": "issue:44534", "right": "issue:44913", "accept": false, - "reason": "Both are configuration/state issues, but non-persistent buffers corruption is not the same as rotary_pct reload behavior." + "reason": "A v5 buffer-initialization regression is unrelated to GPTNeoXConfig rotary_pct not persisting on reload." }, { "left": "issue:35141", "right": "issue:42490", "accept": false, - "reason": "Different problems: embedding reinitialization versus a path-with-slash import warning." + "reason": "Resized embeddings being reinitialized is a model-weight bug, not the local-path trailing-slash import warning." }, { "left": "issue:44734", "right": "issue:45406", "accept": false, - "reason": "Separate serve/model failures: KV-cache continuation indexing versus a missing processor attribute." + "reason": "KV-cache continuation indexing failure in serve is unrelated to Gemma4Processor missing _tokenizer." }, { "left": "issue:46082", "right": "issue:46697", "accept": false, - "reason": "Completely different domains: head_dim validation versus image padding behavior." + "reason": "LlamaConfig head_dim validation is a config constraint issue, not an image processor padding bug." }, { "left": "issue:45632", "right": "issue:46617", "accept": false, - "reason": "Both touch trust_remote_code caching, but one is a leaf-directory collision and the other is a FileNotFoundError; not the same concrete bug." - }, - { - "left": "issue:44869", - "right": "issue:46832", - "accept": false, - "reason": "Unrelated model paths: Whisper timestamp decoding versus RT-DETR feature-level mismatch." + "reason": "Both touch trust_remote_code caching, but one is a leaf-directory cache collision and the other is a cache load FileNotFoundError; not the same bug." }, { "left": "issue:44855", "right": "issue:45588", "accept": false, - "reason": "Different failure modes and code paths: Python 3.13 JIT parsing versus flash_attention handling of s_aux=None." + "reason": "Python 3.13 torch.jit parsing/IndentationError is unrelated to flash-attention crashing on s_aux=None." }, { "left": "issue:45245", "right": "issue:46858", "accept": false, - "reason": "A category-count limit error is unrelated to a device-tensor slice arithmetic bug in cache prefill." + "reason": "Category cardinality overflow is unrelated to cache.get_seq_length() returning a device tensor in static-cache prefill." + }, + { + "left": "issue:44869", + "right": "issue:46832", + "accept": false, + "reason": "Whisper timestamp decoding crash and RT-DETR feature-level mismatch are different model-specific bugs." }, { "left": "issue:43827", "right": "issue:43994", "accept": false, - "reason": "One is a docs-removal mismatch, the other is a model/pipeline inference bug; not the same issue." + "reason": "Stale docs referencing pipeline() are not the same as a model producing nonsense with AutoModel and pipeline." }, { "left": "issue:43673", "right": "issue:46421", "accept": false, - "reason": "Related chunked-prefill/cache area, but missing cache and lazy-init recompilation are distinct concrete regressions." + "reason": "Missing cache in chunked prefill and cache lazy initialization causing recompilation are related only at a high level, not the same code-path bug." }, { "left": "issue:43931", "right": "issue:44060", "accept": false, - "reason": "Different Qwen model issues: weight-shape mismatch versus tied-weights warning contamination." + "reason": "Qwen3-VL weight-shape mismatch is a load-time checkpoint issue, not the tied-weights warning in Qwen3-Next." }, { "left": "issue:44242", "right": "issue:46897", "accept": false, - "reason": "Both concern loss computation, but one is missing load-balancing loss and the other is a double-shift training bug." + "reason": "Load-balancing loss omission and Florence2 double-shift training loss are different MoE/training bugs." }, { "left": "issue:29870", "right": "issue:35141", "accept": false, - "reason": "Unrelated bugs: BART generate-time decoder mask initialization versus embedding reinitialization after resize." + "reason": "Bart decoder_attention_mask initialization error is unrelated to post_init reinitializing output embeddings after resize." }, { "left": "issue:44617", "right": "issue:45584", "accept": false, - "reason": "Different models and failures: SAM3 video OOM versus Whisper empty-transcription decoding." + "reason": "Sam3Video OOM and Whisper empty-transcription generation failure are unrelated failures." }, { "left": "issue:43644", "right": "issue:44913", "accept": false, - "reason": "Non-persistent buffer corruption and rotary_pct reload are separate persistence/config bugs." + "reason": "A v5 non-persistent-buffer regression is unrelated to GPTNeoXConfig rotary_pct persistence." }, { "left": "issue:44933", "right": "issue:45362", "accept": false, - "reason": "Different domains entirely: image_utils import cleanup versus a Qwen3.5 chat crash." + "reason": "An import from image_utils is a library API break, not the Qwen3.5 chat crash." + }, + { + "left": "issue:25251", + "right": "issue:27957", + "accept": false, + "reason": "top_k pipeline output nesting and XLMRoberta Flash Attention 2 are unrelated issues." }, { "left": "issue:43976", "right": "issue:44855", "accept": false, - "reason": "Separate version-compatibility issues: Python 3.9/3.10 support versus a Python 3.13 IndentationError." + "reason": "Python-version packaging incompatibility is not the same as the DebertaV2Model torch.jit parsing error." } ] }, @@ -11690,8 +11633,8 @@ "issue:43550|issue:44190", "issue:45381|issue:46369", "issue:44077|issue:44877", - "issue:44479|issue:44734", "issue:41720|issue:45245", + "issue:44479|issue:44734", "issue:43335|issue:44190", "issue:43065|issue:44610", "issue:44464|issue:45563", @@ -11706,8 +11649,8 @@ "issue:45290|issue:46752", "issue:44462|issue:45907", "issue:45276|issue:45584", - "issue:44737|issue:45803", "issue:43540|issue:44734", + "issue:44737|issue:45803", "issue:44938|issue:45542", "issue:25251|issue:29870", "issue:44521|issue:45381", @@ -11715,20 +11658,20 @@ "issue:43854|issue:44964", "issue:45632|issue:46234" ], - "prepared_review_unit_hash": "929d7d544870e43fcf896e34039c982d97152a521ef5a36aa80d6c8763764f66", + "prepared_review_unit_hash": "27bb6137e0ac92ccc387cd96c140244638c98ae80c7e82ea6f30d74306de031c", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 35695, + "serialized_chars": 35696, "estimated_input_tokens": 8924, "estimated_eval_tokens": 18104 }, @@ -11737,180 +11680,180 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a grab bag of mostly unrelated single-issue bugs across generation, tokenization, multimodal processing, configs, and CI. None of the soft-edge pairs look like true duplicates.", - "confidence": 0.94, - "canonical_issue_reason": "No clear duplicate family emerges from the cluster, so there is no strong canonical issue. If a representative must be picked, issue 45584 is the cleanest self-contained bug report with a specific failure mode and reproducible symptoms.", + "summary": "This cluster is a grab-bag of unrelated Transformers bug reports spanning different models, generation paths, processors, configs, and environment issues. None of the soft pairs look like the same underlying bug or a mergeable duplicate PR.", + "confidence": 0.83, + "canonical_issue_reason": "No clear canonical issue: the items are mostly unrelated bugs/regressions across different subsystems, and none of the soft pairs describe the same underlying defect.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45584 is the best representative issue in this set: it has a focused title, a concrete user-facing crash, and enough detail to stand in as a canonical example of the cluster\u2019s overall bug style.", + "best_issue_reason": "If a representative issue must be chosen, issue:44521 is the most active/open and broadly user-facing, but it is not a true cluster canonical.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45563", "right": "issue:45584", "accept": false, - "reason": "Different generation bugs: a stale warning for num_return_sequences vs an empty-transcription crash after align_special_tokens." + "reason": "Both are generation-related, but one is a stale warning and the other is an empty-transcription crash; different failure modes and code paths." }, { "left": "issue:43550", "right": "issue:44190", "accept": false, - "reason": "Unrelated failures: Bamba torch.compile/SDPA assert vs a dataset-loading script error." + "reason": "A torch.compile SDPA failure and a local-dataset loading error in an example script are unrelated." }, { "left": "issue:45381", "right": "issue:46369", "accept": false, - "reason": "Both touch multimodal models, but one is qwen video vision_position_ids and the other is Sam3 LiteText memory overflow." + "reason": "Qwen video position-id corruption and Sam3LiteText memory overflow are different model-specific problems." }, { "left": "issue:44077", "right": "issue:44877", "accept": false, - "reason": "Different config problems: optional post_init handling vs strict config loading for granite_speech." + "reason": "One is about optional post_init behavior, the other about strict config loading; same area, not the same bug." }, { - "left": "issue:44479", - "right": "issue:44734", + "left": "issue:41720", + "right": "issue:45245", "accept": false, - "reason": "Different code paths: video-input regression in Qwen VL variants vs serve KV-cache continuation indexing." + "reason": "cudaErrorAssert in Qwen3 auto device mapping is unrelated to the 2^24 category-limit runtime error." }, { - "left": "issue:41720", - "right": "issue:45245", + "left": "issue:44479", + "right": "issue:44734", "accept": false, - "reason": "CUDA auto device-mapping assert and category-count overflow are unrelated bugs." + "reason": "A video-input regression in Qwen VL models and a serve KV-cache indexing crash are distinct bugs." }, { "left": "issue:43335", "right": "issue:44190", "accept": false, - "reason": "SwitchTransformers sparse-layer construction is unrelated to the local dataset-loading script issue." + "reason": "Sparse-layer construction in SwitchTransformers is unrelated to the image-classification dataset loading issue." }, { "left": "issue:43065", "right": "issue:44610", "accept": false, - "reason": "Sam3PixelDecoder dummy Conv2d and OmDet-Turbo input-size mismatch are different model/processor bugs." + "reason": "Dummy Conv2d in Sam3PixelDecoder and a 640px-vs-224px processor mismatch are different code paths." }, { "left": "issue:44464", "right": "issue:45563", "accept": false, - "reason": "Different generation issues: inconsistent compiled chunked generation vs a stale warning for num_return_sequences." + "reason": "Chunked generation inconsistencies with compiled forward are not the same as a stale warning for num_return_sequences." }, { "left": "issue:39692", "right": "issue:44190", "accept": false, - "reason": "Documentation example errors are unrelated to the image-classification script failing on local datasets." + "reason": "SigLIP2 documentation/example errors and local dataset loading failures are unrelated." }, { "left": "issue:45127", "right": "issue:45357", "accept": false, - "reason": "LoRA/extended-vocab collapse and incorrect visual encoder keys on save_pretrained are separate save/load problems." + "reason": "LoRA/vocab-collapse after merge and incorrect visual-encoder keys in save_pretrained are separate model-save/merge bugs." }, { "left": "issue:45381", "right": "issue:46655", "accept": false, - "reason": "Both are video-related, but they affect different models and different failure modes." + "reason": "Both touch video processing, but one is wrong vision_position_ids and the other is incorrect padding calculation." }, { "left": "issue:44112", "right": "issue:45600", "accept": false, - "reason": "Both mention CI, but one is a stale device override test and the other is a broken auto_mappings reference." + "reason": "A stale CI test and broken auto_mappings references are unrelated maintenance issues." }, { "left": "issue:44779", "right": "issue:46489", "accept": false, - "reason": "Related DeepSeek family, but not the same bug: generic tokenizer regression vs a DeepSeek-Coder v1 gap in a prior fix." + "reason": "Both concern DeepSeek tokenizers, but they describe different regressions and cannot be safely merged as one bug." }, { "left": "issue:43295", "right": "issue:45276", "accept": false, - "reason": "Processor.tokenizer regression and resize_token_embeddings propagation are different mechanisms." + "reason": "A processor/tokenizer regression and a Gemma4 embedding-resize bug are not the same defect." }, { "left": "issue:44617", "right": "issue:46369", "accept": false, - "reason": "Sam3Video OOM and Sam3-LiteText memory overflow are different resource issues in different variants." + "reason": "Sam3Video OOM and Sam3-LiteText memory overflow are different variants with different symptoms." }, { "left": "issue:43526", "right": "issue:46697", "accept": false, - "reason": "ImageProcessorFast label reduction and Swin2SR over-padding are unrelated preprocessing bugs." + "reason": "Label reduction in BeitImageProcessorFast and Swin2SR over-padding are unrelated image-processing bugs." }, { "left": "issue:45290", "right": "issue:46752", "accept": false, - "reason": "apply_chat_template crashes in different scenarios: tool-call assistant messages vs an empty list." + "reason": "Both are chat-template crashes, but one is tool-call-specific and the other is an empty-list IndexError; too different to merge." }, { "left": "issue:44462", "right": "issue:45907", "accept": false, - "reason": "AutoTokenizer ignoring tokenizer.json is unrelated to list_repo_files missing the revision argument." + "reason": "AutoTokenizer ignoring tokenizer.json and list_repo_files forwarding revision are unrelated Hub behaviors." }, { "left": "issue:45276", "right": "issue:45584", "accept": false, - "reason": "Embedding resize propagation and Whisper empty transcription are unrelated bugs." + "reason": "Gemma4 embedding resizing and Whisper empty-transcription handling are unrelated bugs." }, { - "left": "issue:44737", - "right": "issue:45803", + "left": "issue:43540", + "right": "issue:44734", "accept": false, - "reason": "XLNet CPU tensor allocation and Fuyu bare except swallowing errors are different issues." + "reason": "Qwen3OmniMoe video-input validation and serve KV-cache continuation indexing are different code paths." }, { - "left": "issue:43540", - "right": "issue:44734", + "left": "issue:44737", + "right": "issue:45803", "accept": false, - "reason": "Qwen3OmniMoe video-input ValueError and serve KV-cache continuation indexing are unrelated." + "reason": "XLNet CPU tensor allocation and Fuyu bare-except error handling are unrelated implementation issues." }, { "left": "issue:44938", "right": "issue:45542", "accept": false, - "reason": "Python 3.14 import failure and the tensorboard/TensorFlow backend issue do not share a code path." + "reason": "Python 3.14 import/load failure and undefined tf backend due to tensorboard-only install are different environment issues." }, { "left": "issue:25251", "right": "issue:29870", "accept": false, - "reason": "Pipeline top_k nesting and Bart decoder_attention_mask initialization are unrelated bugs." + "reason": "Pipeline top_k nesting behavior and Bart generate decoder_attention_mask unbound local error are unrelated generation bugs." }, { "left": "issue:44521", "right": "issue:45381", "accept": false, - "reason": "Both are multimodal issues, but assistant_masks being all zeros is not the same as incorrect vision_position_ids." + "reason": "All-zero assistant masks and wrong video vision_position_ids are both multimodal, but they are distinct bugs." }, { "left": "issue:43673", "right": "issue:46858", "accept": false, - "reason": "Chunked_prefill cache missing and static-cache seq_length tensor arithmetic are different cache bugs." + "reason": "Missing GenerationMixin cache in chunked_prefill and device-tensor seq_length arithmetic are separate cache/compile issues." }, { "left": "issue:43854", "right": "issue:44964", "accept": false, - "reason": "Two separate model-loading failures for different models and different underlying causes." + "reason": "Model-load failures for GLM-4.7-Flash and Phi-4 multimodal are different reports with different failure contexts." }, { "left": "issue:45632", "right": "issue:46234", "accept": false, - "reason": "Cache-path collision for local models is unrelated to a missing import_utils helper symbol." + "reason": "A trust_remote_code cache-path collision and a missing import_utils symbol are unrelated bugs." } ] }, @@ -11988,8 +11931,8 @@ "issue:44964|issue:45600", "issue:43643|issue:44496", "issue:44112|issue:44964", - "issue:27957|issue:35141", "issue:42175|issue:44279", + "issue:27957|issue:35141", "issue:46082|issue:46121", "issue:36246|issue:41628", "issue:42796|issue:44521", @@ -12007,8 +11950,8 @@ "issue:45083|issue:46506", "issue:42907|issue:43441", "issue:44683|issue:45446", - "issue:43526|issue:43901", "issue:43976|issue:44279", + "issue:43526|issue:43901", "issue:43381|issue:45593", "issue:43479|issue:45375", "issue:44496|issue:45042", @@ -12016,220 +11959,220 @@ "issue:43673|issue:44336", "issue:44792|issue:46424" ], - "prepared_review_unit_hash": "f693f288b874fc3bd32465a300d4f84d656ee87b4d206492211ca77b29fb743d", + "prepared_review_unit_hash": "c19925f4c81758a65165a9f2496a60e7bd45907e65666c4d7a1ad97f4682a25c", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36931, - "estimated_input_tokens": 9233, - "estimated_eval_tokens": 18722 + "serialized_chars": 36934, + "estimated_input_tokens": 9234, + "estimated_eval_tokens": 18724 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose grab bag of Transformers issues with only superficial topical overlap (imports, configs, multimodal models, caches, etc.). None of the proposed soft pairs look like the same underlying bug or change, so all soft edges should be rejected.", - "confidence": 0.84, - "canonical_issue_reason": "issue:44964 is the most concrete, reproducible, and triage-friendly report in the set: it names a specific model, a specific regression point, and a clear load failure mode.", + "summary": "This cluster is mostly a set of unrelated issues with only superficial lexical overlap (imports, configs, cache, multimodal, etc.). None of the soft pairs look like the same underlying bug or change, so no merges are justified. The strongest standalone issue here is the Phi-4 multimodal model-loading regression because it is concrete and reproducible.", + "confidence": 0.78, + "canonical_issue_reason": "issue:44964 is the most concrete, user-facing regression in the set: it names a specific model, a clear failure mode, and a current-transformers incompatibility, making it a reasonable representative issue for this cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44964 is the best single issue anchor because it is narrowly scoped and actionable, unlike the broader packaging/import/doc or vague regression reports elsewhere in the cluster.", + "best_issue_reason": "issue:44964 is the best issue candidate because it has a specific model, a specific broken workflow, and a clear reproduction target; it is more actionable than the broader import/config reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44246", "right": "issue:45003", "accept": false, - "reason": "Import-performance complaint vs unsafe sys.modules access: related area, but different bugs and different fixes." + "reason": "Both mention import-related behavior, but one is about slow imports while the other is about unsafe sys.modules access; these are different bugs and code paths." }, { "left": "issue:44964", "right": "issue:45600", "accept": false, - "reason": "Different model families and failure modes: Phi-4 multimodal load regression vs removed Sam3LiteText config refs breaking CI." + "reason": "Phi-4 multimodal loading failure and removed Sam3LiteText config references are unrelated problems in different parts of the library." }, { "left": "issue:43643", "right": "issue:44496", "accept": false, - "reason": "Both are model/config loading errors, but one is missing fields under trust_remote_code and the other is missing model_type for a specific repo." + "reason": "One is a trust_remote_code config field loss issue; the other is an unrecognized model/model_type error. Similar theme, different root cause." }, { "left": "issue:44112", "right": "issue:44964", "accept": false, - "reason": "CI test flake in GraniteSpeech is unrelated to the Phi-4 multimodal loading regression." + "reason": "A stale CI test for GraniteSpeech and a Phi-4 model load regression are unrelated." }, { - "left": "issue:27957", - "right": "issue:35141", + "left": "issue:42175", + "right": "issue:44279", "accept": false, - "reason": "FlashAttention support for XLM-R is unrelated to embedding reinitialization when resizing token embeddings." + "reason": "Missing TensorFlow in the torch extra and a generic dependency issue are not the same concrete failure." }, { - "left": "issue:42175", - "right": "issue:44279", + "left": "issue:27957", + "right": "issue:35141", "accept": false, - "reason": "Both mention dependencies, but one is a pip extras/backend packaging complaint and the other is a separate general dependency issue." + "reason": "FlashAttention support for XLMRoberta and token embedding reinitialization after resize are separate model behaviors." }, { "left": "issue:46082", "right": "issue:46121", "accept": false, - "reason": "Both are config-related, but they hit different validation paths and different inputs." + "reason": "LlamaConfig head_dim validation and rope-parameter list handling are different configuration bugs." }, { "left": "issue:36246", "right": "issue:41628", "accept": false, - "reason": "Different missing symbols: Qwen2_5_VLImageProcessor vs AutoImageProcessor." + "reason": "Both are import errors, but for different symbols and modules; no evidence they share the same underlying fix." }, { "left": "issue:42796", "right": "issue:44521", "accept": false, - "reason": "Tokenizer BPE-marker handling and multimodal assistant_mask generation are different code paths and bugs." + "reason": "Tokenizer BPE-marker handling and assistant-mask generation for multimodal chat templates are distinct issues." }, { "left": "issue:36246", "right": "issue:38617", "accept": false, - "reason": "Both are import errors, but for different removed/missing names with no shared concrete fix." + "reason": "Different missing imports in different modules; they are not the same bug." }, { "left": "issue:43404", "right": "issue:45216", "accept": false, - "reason": "Weight-tying bug in a model class is unrelated to a save_pretrained checkpoint regression." + "reason": "lm_head weight tying in Mistral3 and Qwen3.5 checkpoint saving are separate model-specific regressions." }, { "left": "issue:32090", "right": "issue:35141", "accept": false, - "reason": "Trainer broadcast NoneType error is unrelated to post_init reinitializing output embeddings." + "reason": "Trainer GPU broadcast NoneType failure and post_init reinitializing output embeddings are unrelated." }, { "left": "issue:44295", "right": "issue:46506", "accept": false, - "reason": "Buffer access error and CPU/CUDA output divergence are different problems with no clear shared root cause." + "reason": "A position_ids buffer read error and CPU/CUDA output divergence are different failure modes." }, { "left": "issue:41950", "right": "issue:44016", "accept": false, - "reason": "Video-classification pipeline image-processor lookup is unrelated to a notebook syntax error." + "reason": "A pipeline using image processors and a notebook syntax error are not the same problem." }, { "left": "issue:43531", "right": "issue:45600", "accept": false, - "reason": "Qwen3-MoE sliding-window behavior and auto_mappings CI breakage are unrelated." + "reason": "Qwen3-MoE sliding_window behavior and removed config references in auto_mappings.py are unrelated." }, { "left": "issue:43650", "right": "issue:44220", "accept": false, - "reason": "The titles do not describe the same bug; one is too vague to tie to the audio feature extraction issue." + "reason": "The placeholder 'ADD THE DATA' issue does not correspond to the fbank feature extraction bug." }, { "left": "issue:43673", "right": "issue:46424", "accept": false, - "reason": "Chunked-prefill cache regression and max_cache_length being honored only for Quantized Cache are separate cache behaviors." + "reason": "A missing GenerationMixin cache in chunked_prefill and cache max-length handling for quantized cache are related only at a high level, not as the same concrete bug." }, { "left": "issue:43976", "right": "issue:44987", "accept": false, - "reason": "Python-version compatibility failure is unrelated to loading a specific model repo." + "reason": "Python version compatibility for Transformers 5.1.0 and failing to load a specific model are different issues." }, { "left": "issue:43295", "right": "issue:45335", "accept": false, - "reason": "Processor/tokenizer regression and t5gemma embedding resize bug are different model pipeline issues." + "reason": "Processor/tokenizer regression and t5gemma decoder embedding resize behavior are separate code-path bugs." }, { "left": "issue:44016", "right": "issue:44220", "accept": false, - "reason": "Notebook syntax error is unrelated to the fbank feature extraction bug." + "reason": "Notebook syntax error and feature-extraction helper behavior are unrelated." }, { "left": "issue:45083", "right": "issue:46506", "accept": false, - "reason": "Helper output-length logic in qwen3_omni_moe is not the same as a CPU/CUDA parity discrepancy." + "reason": "A helper-function behavior bug in qwen3_omni_moe and cross-device result differences are not the same issue." }, { "left": "issue:42907", "right": "issue:43441", "accept": false, - "reason": "Dequantized checkpoint saving and FlashAttention runtime failures in the Ministral family are distinct bugs." + "reason": "Saving dequantized Ministrals/Devstrals and FlashAttention failures in Ministral-3 are different problems, even if they touch the same model family." }, { "left": "issue:44683", "right": "issue:45446", "accept": false, - "reason": "Compiled flex_attention failure on torch>=2.9 and an incorrect version check for AuxRequest import are different flex_attention issues." + "reason": "Compiled flex_attention failing on torch >= 2.9 and an incorrect version check for AuxRequest import are separate flex_attention issues." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Package Python-version compatibility and a generic dependency issue are not the same bug." }, { "left": "issue:43526", "right": "issue:43901", "accept": false, - "reason": "Image-processor label reduction bug and pipeline docs mismatch are unrelated." + "reason": "BeitImageProcessorFast reduce_labels behavior and TextClassificationPipeline docs mismatch are unrelated." }, { "left": "issue:43381", "right": "issue:45593", "accept": false, - "reason": "Gradient checkpointing eval-mode restriction has nothing to do with D-FINE auxiliary losses." + "reason": "Gradient checkpointing in eval mode and D-FINE auxiliary losses are different model/training concerns." }, { "left": "issue:43479", "right": "issue:45375", "accept": false, - "reason": "Both are multimodal config issues, but they concern different models and different missing/default fields." + "reason": "Different multimodal config classes, different missing/defaulted fields, and different model families." }, { "left": "issue:44496", "right": "issue:45042", "accept": false, - "reason": "Unrecognized model/config loading and PIL backend torchvision requirements are different bugs." + "reason": "Unrecognized model/model_type loading and PIL image processors incorrectly requiring torchvision are unrelated bugs." }, { "left": "issue:43901", "right": "issue:44496", "accept": false, - "reason": "Docs behavior mismatch is unrelated to model-type/config import failure." + "reason": "Pipeline documentation wording and model loading failure are not the same underlying issue." }, { "left": "issue:43673", "right": "issue:44336", "accept": false, - "reason": "Cache handling during generation and ANSI codes in loading_report are unrelated." + "reason": "Cache handling and ANSI escape output from loading_report are completely different concerns." }, { "left": "issue:44792", "right": "issue:46424", "accept": false, - "reason": "Janus image-generation test failure is unrelated to cache length configuration semantics." - }, - { - "left": "issue:44964", - "right": "issue:45003", - "accept": false, - "reason": "Phi-4 model loading regression and unsafe sys.modules access are different code-path problems with different fixes." + "reason": "A failing Janus image-generation test and cache max-length semantics are unrelated." } ] }, @@ -12305,8 +12248,8 @@ "soft_pairs": [ "issue:44016|issue:44062", "issue:44496|issue:45759", - "issue:43824|issue:44933", "issue:43441|issue:45588", + "issue:43824|issue:44933", "issue:46123|issue:46369", "issue:44327|issue:46752", "issue:44521|issue:46693", @@ -12317,8 +12260,8 @@ "issue:46082|issue:46399", "issue:43295|issue:44779", "issue:44075|issue:44368", - "issue:39692|issue:43901", "issue:43976|issue:45405", + "issue:39692|issue:43901", "issue:43577|issue:45216", "issue:44351|issue:45278", "issue:46082|issue:46512", @@ -12332,20 +12275,20 @@ "issue:41950|issue:45823", "issue:44242|issue:45200" ], - "prepared_review_unit_hash": "e85eb4cab8d6019c04994f248c245c0bac98b304af495ac512d62f1251643973", + "prepared_review_unit_hash": "2b8f867207fee7b1603edabf832676bbec941c7058c41ad306046147829a8518", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36213, + "serialized_chars": 36214, "estimated_input_tokens": 9054, "estimated_eval_tokens": 18364 }, @@ -12354,174 +12297,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is highly heterogeneous: most pairs are only loosely similar by topic (tokenizers, multimodal processors, save/load regressions, pipelines, imports, etc.) but describe different bugs and code paths. No soft edge looks like a true duplicate merge candidate.", - "confidence": 0.33, - "canonical_issue_reason": "If one issue must anchor the cluster, issue 44521 is the strongest candidate: it is a concrete, reproducible multimodal API regression, has the most discussion/inbound references, and is broader than the other items. That said, the set does not form a real duplicate cluster.", + "summary": "This cluster is heterogeneous: it mixes tokenizer, model-loading, pipeline, config, and documentation regressions, but none of the soft-pair candidates look like the same underlying bug/change. The closest thing to an umbrella is a broad upgrade/import-regression report, not a true duplicate set.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 45278 is the broadest umbrella-style report here ('many import errors after update from 4.57.0 to 5.5.0'), so it is the closest anchor among otherwise unrelated bugs, though it is not a true duplicate of the others.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44521 is the best standalone issue in the set because it is active, well-scoped, and has clear downstream impact on multimodal chat-template behavior. It is still not a duplicate of the others.", + "best_issue_reason": "Issue 45278 is the best representative of the cluster only because it is the most general post-upgrade breakage report; the rest are narrower, concrete failures in different code paths.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44016", "right": "issue:44062", "accept": false, - "reason": "Both involve notebook/tokenizer problems, but one is a notebook syntax error and the other is an AddedToken keyword conflict; different failure modes and fix paths." + "reason": "One is a notebook syntax error; the other is an AddedToken keyword conflict. Different failure modes and code paths." }, { "left": "issue:44496", "right": "issue:45759", "accept": false, - "reason": "Both are model-loading/config issues, but one is an unrecognized model/config problem and the other is about unwraping text_config in AutoModelForCausalLM.from_config; not the same bug." + "reason": "Both concern model loading, but one is unknown-model/config typing and the other is from_config failing to unwrap text_config for composite Qwen multimodal configs." }, { - "left": "issue:43824", - "right": "issue:44933", + "left": "issue:43441", + "right": "issue:45588", "accept": false, - "reason": "Both are import errors, but they concern different symbols/modules entirely (Qwen2_5_VLForConditionalGeneration vs image_utils)." + "reason": "Both mention FlashAttention, but one is a Ministral model-specific crash and the other is an AttributeError in flash_attention.py for sink-less models." }, { - "left": "issue:43441", - "right": "issue:45588", + "left": "issue:43824", + "right": "issue:44933", "accept": false, - "reason": "Both touch flash attention, but they hit different code paths: Ministral FlashAttention failure vs flash_attention.py crashing on s_aux=None for sink-less models." + "reason": "Both are import errors, but they target different missing symbols in different modules." }, { "left": "issue:46123", "right": "issue:46369", "accept": false, - "reason": "One is a pipeline batching bug that drops the last partial batch; the other is a model memory overflow. Different layers of the stack." + "reason": "A batching/pipeline bug vs a model memory overflow; not the same underlying issue." }, { "left": "issue:44327", "right": "issue:46752", "accept": false, - "reason": "Both are runtime errors in pipeline utilities, but one is top_k bounds handling in QA span decoding and the other is empty-list handling in apply_chat_template." + "reason": "Both crash on edge cases, but one is QA span top-k bounds and the other is apply_chat_template on an empty list." }, { "left": "issue:44521", "right": "issue:46693", "accept": false, - "reason": "Multimodal assistant mask generation and FA2 long-context performance regression are unrelated bugs, even if both affect generation workflows." + "reason": "Assistant-mask generation for multimodal chat templates is unrelated to the FA2 long-context performance regression." }, { "left": "issue:44279", "right": "issue:46515", "accept": false, - "reason": "These are both dependency/environment complaints, but one is a vague transformer dependency issue and the other is a specific torch 2.4.1 incompatibility." + "reason": "A generic dependency issue vs a torch 2.4.1 incompatibility; different root causes." }, { "left": "issue:45335", "right": "issue:45584", "accept": false, - "reason": "Both involve model/token embedding behavior, but one is resize_token_embeddings not propagating in t5gemma while the other is Whisper generation failing after align_special_tokens." + "reason": "One is resize_token_embeddings not propagating in t5gemma; the other is Whisper generation failing after align_special_tokens." }, { "left": "issue:43299", "right": "issue:45216", "accept": false, - "reason": "Both are Qwen3.5 regressions, but one is a loading failure for MoE models and the other is incorrect save_pretrained checkpoint contents." + "reason": "Qwen3VL loading failure vs Qwen3.5 save_pretrained serialization regression; related model family, but different concrete bugs." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "Dependency/install trouble vs a bumped MIN_PEFT_VERSION pointing to an unreleased package version." }, { "left": "issue:46082", "right": "issue:46399", "accept": false, - "reason": "A config validation issue for Llama head_dim is unrelated to Qwen3.5 kernel registration failures in use_kernels=True." + "reason": "Config validation around head_dim is unrelated to a missing decorator for apply_rotary_pos_emb kernel registration." }, { "left": "issue:43295", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer/processor regressions, but one is about processor.tokenizer/images handling and the other is Deepseek tokenizer output correctness." + "reason": "A regression in custom model code using processor.tokenizer/images vs Deepseek tokenizer output corruption; different APIs and symptoms." }, { "left": "issue:44075", "right": "issue:44368", "accept": false, - "reason": "SGD args being ignored and a tie_word_embeddings warning are unrelated training/config issues." + "reason": "SGD args being ignored and a tie_word_embeddings warning are unrelated." }, { - "left": "issue:39692", - "right": "issue:43901", + "left": "issue:43976", + "right": "issue:45405", "accept": false, - "reason": "Both mention documentation, but they cover different pipelines and different problems: SigLIP2 example mismatches vs TextClassificationPipeline return_all_scores docs drift." + "reason": "Python version support mismatch vs an unreleased PEFT minimum-version bump; different dependency problems." }, { - "left": "issue:43976", - "right": "issue:45405", + "left": "issue:39692", + "right": "issue:43901", "accept": false, - "reason": "Python version support failure and a PEFT minimum-version bump are different dependency-management issues." + "reason": "Different docs issues: SigLIP2 example errors vs TextClassificationPipeline docs/behavior mismatch." }, { "left": "issue:43577", "right": "issue:45216", "accept": false, - "reason": "Blip2 dtype propagation and Qwen3.5 save_pretrained serialization are different regressions with different affected code paths." + "reason": "BLIP2 dtype propagation bug vs Qwen3.5 checkpoint serialization regression; not the same bug." }, { "left": "issue:44351", "right": "issue:45278", "accept": false, - "reason": "A missing HybridCache import and a broad set of import errors after upgrade are related only at a very high level; not the same concrete issue." + "reason": "Both are import failures after updates, but one is HybridCache specifically and the other is a broad set of import errors; not the same concrete defect." }, { "left": "issue:46082", "right": "issue:46512", "accept": false, - "reason": "Both involve config attributes, but one rejects explicit head_dim while the other leaves model.config.dtype stale after from_config(dtype=...)." + "reason": "Both involve config handling, but one rejects a head_dim value while the other leaves model.config.dtype stale after from_config." }, { "left": "issue:42796", "right": "issue:43650", "accept": false, - "reason": "Tokenizer BPE marker handling and a nonspecific 'ADD THE DATA' report are not the same bug, and the latter is too vague to match." + "reason": "Tokenizer BPE marker handling vs an empty/low-signal 'ADD THE DATA' issue; no shared bug." }, { "left": "issue:44779", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but they concern different model families and different symptoms (Deepseek incorrect results vs Granite wrong token IDs)." + "reason": "Both are tokenizer regressions, but they affect different model families and different token-ID/encoding logic." }, { "left": "issue:41950", "right": "issue:45020", "accept": false, - "reason": "A video-classification processor lookup bug and broad remote_code breakage are different compatibility issues." + "reason": "A video-classification processor lookup bug is not the same as a broad remote_code model-loading regression." }, { "left": "issue:43299", "right": "issue:45357", "accept": false, - "reason": "Both concern Qwen3.5 save/load behavior, but one is a loading breakage and the other is incorrect visual encoder keys on save_pretrained." + "reason": "Qwen3VL loading vs Qwen3.5 save_pretrained still saving wrong visual-encoder keys; distinct regression points." }, { "left": "issue:44466", "right": "issue:45276", "accept": false, - "reason": "Both involve weight/embedding serialization, but one is device-dependent lm_head.weight serialization while the other is resize_token_embeddings not updating gemma4 submodules." + "reason": "lm_head.weight serialization differs by device, whereas this is about resize_token_embeddings not updating embeddings." }, { "left": "issue:43299", "right": "issue:43854", "accept": false, - "reason": "Different multimodal model families and different failures: Qwen3VL MoE loading vs GLM-4.7-Flash unit-test loading." + "reason": "Different models and failure sites: Qwen3VL loading vs GLM-4.7-Flash unit-test loading." }, { "left": "issue:43653", "right": "issue:45593", "accept": false, - "reason": "BigBirdTokenizer special-token registration and D-FINE auxiliary-loss gating are completely unrelated bugs." + "reason": "BigBirdTokenizer special-token registration is unrelated to D-FINE losing auxiliary losses when denoising is off." }, { "left": "issue:41950", "right": "issue:45823", "accept": false, - "reason": "The first is a pipeline processor-selection bug; the second is a Gemma4 device placement mismatch during forward. Different subsystems." + "reason": "Pipeline processor mismatch vs Gemma4 PLE device placement mismatch; different subsystems and symptoms." }, { "left": "issue:44242", "right": "issue:45200", "accept": false, - "reason": "Load-balancing loss gating and Gemma4 mm_token_type_ids defaults are different multimodal training issues with different code paths." + "reason": "Load-balancing loss omission in MoE routing vs Gemma4 requiring mm_token_type_ids to default to zeros." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "Dependency issue vs unreleased PEFT minimum-version bump; not the same underlying bug." } ] }, @@ -12606,9 +12561,9 @@ "issue:43441|issue:45834", "issue:42994|issue:43317", "issue:44623|issue:44749", - "issue:44336|issue:46851", "issue:44261|issue:44263", "issue:45103|issue:45372", + "issue:44336|issue:46851", "issue:45600|issue:45712", "issue:45859|issue:46851", "issue:42175|issue:43976", @@ -12623,20 +12578,20 @@ "issue:43525|issue:43881", "issue:46291|issue:46399" ], - "prepared_review_unit_hash": "d068d17fb9dd3a5ff0b4f6231d8c000d3bc575c31981429d01e47d250462d336", + "prepared_review_unit_hash": "bb35b3bf788fe5a0c52de8dfee5a882c3e429fd426850c5b2c712c5b697c5506", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 35701, + "serialized_chars": 35703, "estimated_input_tokens": 8926, "estimated_eval_tokens": 18108 }, @@ -12645,174 +12600,174 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: the soft pairs mostly share only superficial subsystem overlap (tokenizers, loading, generation, multimodal, Deepspeed, etc.) but do not describe the same concrete bug. I would keep a broad, well-discussed issue like import/performance regressions as the best representative issue, but there is no PR in the set.", - "confidence": 0.87, - "canonical_issue_reason": "issue:44246 is the best representative issue here: it has broad user impact, clear reproduction around import-time performance, and higher discussion/inbound activity than most other candidates.", + "summary": "This cluster is mostly a set of unrelated bugs/regressions that only look superficially similar by shared subsystem or wording. I would not merge any of the soft pairs as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": "No true duplicate cluster is present; if one issue must serve as an anchor, #44246 is the broadest/highest-engagement report and the least specific of the set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44246 is the strongest general-purpose issue candidate for triage because it is clearly described, has broader scope than model-specific failures, and appears more central than the narrower one-off regressions.", + "best_issue_reason": "#44246 is the best triage anchor because it is a general, package-level performance issue with strong discussion activity, but it should not absorb the other reports as duplicates.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43650", "right": "issue:46209", "accept": false, - "reason": "Unrelated: a vague \"ADD THE DATA\" issue is not the same as an FP8 inference failure on B200." + "reason": "'ADD THE DATA' is unrelated to FP8 inference on B200; no shared bug or code path." }, { "left": "issue:44297", "right": "issue:44462", "accept": false, - "reason": "Different tokenizer bugs: save_pretrained metadata mismatch vs AutoTokenizer ignoring repository tokenizer.json." + "reason": "Tokenizer save mismatch is different from AutoTokenizer ignoring repository tokenizer.json; both involve tokenizers but not the same failure." }, { "left": "issue:43479", "right": "issue:43525", "accept": false, - "reason": "Config initialization defaults for multimodal components is unrelated to Llama4 missing pad_token_id." + "reason": "Different models and bugs: Phi4Multimodal default sub-config initialization vs Llama4Config missing pad_token_id." }, { "left": "issue:44683", "right": "issue:45542", "accept": false, - "reason": "Different areas and failure modes: torch 2.9 flex_attention vs missing TensorFlow backend installation." + "reason": "Compiled flex_attention on torch>=2.9 is unrelated to missing TensorFlow backend installation." }, { "left": "issue:44016", "right": "issue:44291", "accept": false, - "reason": "Notebook syntax error is unrelated to init_empty_weights passing an unexpected _is_hf_initialized argument." + "reason": "Notebook syntax error and init_empty_weights TypeError are unrelated issues." }, { "left": "issue:44361", "right": "issue:45593", "accept": false, - "reason": "Tokenizer AttributeError on tasks is not the same as D-FINE auxiliary-loss behavior." + "reason": "MLLukeTokenizer task AttributeError is distinct from D-FINE auxiliary loss behavior." }, { "left": "issue:45081", "right": "issue:45907", "accept": false, - "reason": "Mistral regex patch crash and list_repo_files revision forwarding are different code paths and bugs." + "reason": "Mistral regex patch crash is unrelated to list_repo_files revision forwarding." }, { "left": "issue:33290", "right": "issue:45656", "accept": false, - "reason": "Both involve Deepspeed/optimizer, but one is OOM with Adafactor and the other is double optimizer stepping." + "reason": "AdamFactor OOM in deepspeed and optimizer step called twice are different DeepSpeed optimizer problems." }, { "left": "issue:43441", "right": "issue:45834", "accept": false, - "reason": "Different models and issues: FlashAttention failure for Ministral-3 vs OCR index error in Kosmos2.5." + "reason": "FlashAttention failure for Ministral-3 is unrelated to Kosmos2.5 long OCR index errors." }, { "left": "issue:42994", "right": "issue:43317", "accept": false, - "reason": "Quantized model saving failure is distinct from loading a dequantized model with gpu+cpu offload." + "reason": "Quantized model save failure and dequantized model load with cpu offload are different quantization lifecycle bugs." }, { "left": "issue:44623", "right": "issue:44749", "accept": false, - "reason": "Processor save_pretrained missing files is unrelated to a Chinese report about inference/data filtering slowdown." - }, - { - "left": "issue:44336", - "right": "issue:46851", - "accept": false, - "reason": "ANSI-code output in loading_report has nothing to do with Qwen3.5 variable-length position_ids." + "reason": "Missing processor save files and slow filtering after upgrade are unrelated." }, { "left": "issue:44261", "right": "issue:44263", "accept": false, - "reason": "Numerical precision issue in MLA q_a_layernorm is unrelated to torch.split return values in GlmMoeDsaIndexer." + "reason": "MLA q_a_layernorm epsilon precision issue is not the same as torch.split return values in GlmMoeDsaIndexer." }, { "left": "issue:45103", "right": "issue:45372", "accept": false, - "reason": "Auto-docstring future-annotations crash and ReasoningEffort import failure are different bugs in different modules." + "reason": "auto_docstring future-annotations crash is unrelated to mistral_common ReasoningEffort import failure." + }, + { + "left": "issue:44336", + "right": "issue:46851", + "accept": false, + "reason": "ANSI codes in loading_report and Qwen3.5 variable-length position_ids are separate bugs." }, { "left": "issue:45600", "right": "issue:45712", "accept": false, - "reason": "Broken auto_mappings references and leftover dummy_pt_objects cleanup are separate maintenance issues." + "reason": "auto_mappings references removed configs, while dummy_pt_objects leaks dummy classes; shared repo area only, not same bug." }, { "left": "issue:45859", "right": "issue:46851", "accept": false, - "reason": "CPU offload incompatibility in rotary embedding is not the same as variable-length input handling." + "reason": "CPU offload rotary embedding incompatibility is different from variable-length position_ids handling." }, { "left": "issue:42175", "right": "issue:43976", "accept": false, - "reason": "Packaging TensorFlow omission is unrelated to a Python 3.9+/3.10+ compatibility break." + "reason": "Backend dependency omission with pip extras is unrelated to Python version support regression." }, { "left": "issue:44623", "right": "issue:45397", "accept": false, - "reason": "Missing files on processor save_pretrained is not the same as gemma-4 zero3 from_pretrained loading." + "reason": "Processor save missing files is not the same as gemma-4 zero3 from_pretrained loading behavior." }, { "left": "issue:44671", "right": "issue:44779", "accept": false, - "reason": "Incorrect masked LM predictions in CamemBERT and incorrect Deepseek tokenizer outputs are separate model/tokenizer regressions." + "reason": "CamemBERT masked LM predictions and Deepseek tokenizer incorrect results are different model/tokenizer regressions." }, { "left": "issue:43723", "right": "issue:44162", "accept": false, - "reason": "AutoTokenizer loading failure and ESM2 breakage are different model-loading problems." + "reason": "Tokenizer loading failure in v5 and ESM2 breakage are separate model-family issues." }, { "left": "issue:43901", "right": "issue:44246", "accept": false, - "reason": "Docs mismatch for return_all_scores is unrelated to intermittent slow imports." + "reason": "Docs mentioning return_all_scores is a documentation mismatch, not the same as import time performance variance." }, { "left": "issue:43673", "right": "issue:43761", "accept": false, - "reason": "GenerationMixin cache missing during chunked_prefill and CLIPVision hidden_states regression are different components and symptoms." + "reason": "Chunked prefill cache missing and CLIPVision hidden_states regression are different code paths." }, { "left": "issue:43824", "right": "issue:43976", "accept": false, - "reason": "Missing Qwen2_5_VL import and Python version support issues are not the same bug." + "reason": "Qwen2.5-VL import failure is unrelated to Python version compatibility." }, { "left": "issue:45542", "right": "issue:45800", "accept": false, - "reason": "Missing TensorFlow backend from a torch install is distinct from a torch/transformers version incompatibility." + "reason": "TensorBoard/TensorFlow backend packaging issue is unrelated to torch/transformers version incompatibility." }, { "left": "issue:42796", "right": "issue:43441", "accept": false, - "reason": "Tokenizer BPE marker handling and FlashAttention failure are different classes of bugs." + "reason": "Ministral tokenizer BPE marker handling and FlashAttention failure are separate tokenizer vs attention bugs." }, { "left": "issue:43525", "right": "issue:43881", "accept": false, - "reason": "Missing pad_token_id on Llama4Config is unrelated to glm-4v-9b loading failure." + "reason": "Llama4Config missing pad_token_id and glm-4v-9b loading failure are distinct model config/loading issues." }, { "left": "issue:46291", "right": "issue:46399", "accept": false, - "reason": "kernels v0.15 version/revision requirement and Qwen3.5 rotary embedding registration crash are different integration failures." + "reason": "kerns v0.15 revision/version requirement is unrelated to Qwen3.5 rotary embedding registration crash." } ] }, @@ -12887,48 +12842,48 @@ "soft_pairs": [ "issue:44188|issue:44336", "issue:44016|issue:45823", - "issue:44336|issue:44373", "issue:44589|issue:45362", + "issue:44336|issue:44373", "issue:43525|issue:43643", "issue:46655|issue:46726", "issue:45325|issue:46851", "issue:41628|issue:43976", "issue:44297|issue:45397", "issue:43784|issue:45542", - "issue:45083|issue:46851", "issue:44246|issue:44496", + "issue:45083|issue:46851", "issue:46399|issue:46851", "issue:45561|issue:45563", "issue:45103|issue:46399", "issue:43526|issue:45003", - "issue:46614|issue:46665", "issue:43540|issue:46752", "issue:45072|issue:46829", + "issue:46614|issue:46665", "issue:44295|issue:45397", "issue:43531|issue:45362", "issue:42491|issue:43299", "issue:44246|issue:44871", - "issue:44336|issue:44485", "issue:44242|issue:45593", "issue:45588|issue:46851", + "issue:44336|issue:44485", "issue:45003|issue:46821", "issue:42886|issue:46132", "issue:44451|issue:46752" ], - "prepared_review_unit_hash": "a9476120dcbdbde30727f0907ed815a5295a9a36b9a7af73fed4b5397b3ca748", + "prepared_review_unit_hash": "5862de650ef5cbf4d9b5e6be7d67ab24b916a18f90bd8306599f68dc35198438", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 35841, + "serialized_chars": 35843, "estimated_input_tokens": 8961, "estimated_eval_tokens": 18178 }, @@ -12937,174 +12892,174 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose collection of unrelated Transformers issues that only share superficial keywords (models, config, loading, position_ids, etc.). I do not see any true duplicate pairs among the soft edges.", - "confidence": 0.95, - "canonical_issue_reason": "Issue 46851 is the best representative of the cluster only in the sense that it is a concrete, reproducible runtime regression with a clear code path. But the cluster is not a duplicate set, so no single issue is a true canonical for the whole group.", + "summary": "This cluster is mostly a grab bag of unrelated Transformers issues spanning Qwen/Gemma/other model regressions, tokenizer/processor edge cases, import/config problems, and a few docs/testing bugs. The apparent similarity is mostly lexical (shared model names or API terms), not evidence of true duplicates.", + "confidence": 0.67, + "canonical_issue_reason": "Issue 46851 is the most concrete and actionable report among the closest subtheme here (Qwen3.5 / position_ids / variable-length inference). No stronger true-duplicate anchor is evident from the titles alone.", "canonical_pr_reason": null, - "best_issue_reason": "46851 has the clearest, most actionable bug report and a specific model-family regression; it is the strongest stand-alone issue in this mixed cluster.", + "best_issue_reason": "Issue 46851 is the best representative issue to anchor this cluster because it is narrowly scoped, user-facing, and clearly describes a reproducible runtime failure.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44188", "right": "issue:44336", "accept": false, - "reason": "Completely different subsystems: attention kernel behavior under torch.compile vs terminal ANSI handling in loading_report." + "reason": "Different subsystems and failure modes: torch.compile attention-kernel divergence vs ANSI codes in loading_report." }, { "left": "issue:44016", "right": "issue:45823", "accept": false, - "reason": "Notebook syntax error and Gemma4 device_map forward mismatch are unrelated bug classes." + "reason": "Notebook syntax error is unrelated to a Gemma4 forward device-mismatch bug." }, { - "left": "issue:44336", - "right": "issue:44373", + "left": "issue:44589", + "right": "issue:45362", "accept": false, - "reason": "ANSI escaping in loading_report and a wrong docstring for position_ids are not the same underlying issue." + "reason": "Float8 storage lookup failure and a Qwen3.5 chat crash are distinct issues." }, { - "left": "issue:44589", - "right": "issue:45362", + "left": "issue:44336", + "right": "issue:44373", "accept": false, - "reason": "Float8 storage lookup failure is unrelated to a Qwen3.5 chat crash." + "reason": "ANSI output in a loading report is not the same as a wrong docstring for position_ids." }, { "left": "issue:43525", "right": "issue:43643", "accept": false, - "reason": "Both involve config fields, but one is a missing pad_token_id attribute and the other is trust_remote_code returning incomplete AutoConfig fields; different failures and fixes." + "reason": "Both involve config fields, but one is Llama4Config pad_token_id and the other is trust_remote_code returning incomplete AutoConfig fields; different code paths." }, { "left": "issue:46655", "right": "issue:46726", "accept": false, - "reason": "Video padding calculation bug vs missing vision backend reporting; same area, but distinct code paths and symptoms." + "reason": "Both are vision-related, but one is video padding math in SmolVLM and the other is missing vision-backend handling in AutoProcessor." }, { "left": "issue:45325", "right": "issue:46851", "accept": false, - "reason": "Both mention position_ids, but they affect different Qwen model families and different logic (RoPE scaling vs variable-length input handling)." + "reason": "Same broad family name, but different bugs: temporal scaling in Qwen2.5-VL vs variable-length position_ids in Qwen3.5." }, { "left": "issue:41628", "right": "issue:43976", "accept": false, - "reason": "Importing AutoImageProcessor and Python version compatibility are unrelated." + "reason": "AutoImageProcessor import error and Python version compatibility failure are unrelated." }, { "left": "issue:44297", "right": "issue:45397", "accept": false, - "reason": "Tokenizer save_pretrained metadata mismatch is unrelated to Gemma4 zero3 from_pretrained behavior." + "reason": "Tokenizer config save/load mismatch is not the same as a Gemma4 zero3 from_pretrained failure." }, { "left": "issue:43784", "right": "issue:45542", "accept": false, - "reason": "Both are import-time backend issues, but one is missing nn for sentence-transformers and the other is undefined tf backend when only tensorboard is installed; not the same bug." + "reason": "Sentence-transformers import NameError and missing TensorFlow backend due to tensorboard-only install are different import/runtime problems." }, { - "left": "issue:45083", - "right": "issue:46851", + "left": "issue:44246", + "right": "issue:44496", "accept": false, - "reason": "Different Qwen3 submodels and different helper paths; not the same position_ids bug." + "reason": "Import latency and unrecognized model config are not the same bug." }, { - "left": "issue:44246", - "right": "issue:44496", + "left": "issue:45083", + "right": "issue:46851", "accept": false, - "reason": "Moe load-balancing loss gating and unrecognized model_type are unrelated." + "reason": "Both mention Qwen position-related behavior, but one is a helper length calculation and the other is variable-length input handling; not the same underlying defect." }, { "left": "issue:46399", "right": "issue:46851", "accept": false, - "reason": "Kernel registration decorator crash is a separate issue from variable-length position_ids handling." + "reason": "Kernel registration decorator crash and variable-length position_ids are unrelated code paths." }, { "left": "issue:45561", "right": "issue:45563", "accept": false, - "reason": "A pytest-xdist file race and a stale generation warning are unrelated." + "reason": "A pytest-xdist race in testing utils is unrelated to a stale generate() warning." }, { "left": "issue:45103", "right": "issue:46399", "accept": false, - "reason": "Auto-docstring annotation parsing crash and Qwen kernel registration are unrelated tooling/runtime bugs." + "reason": "Auto-docstring annotation handling and attention-kernel registration are separate issues." }, { "left": "issue:43526", "right": "issue:45003", "accept": false, - "reason": "BEiT reduce_labels behavior and unsafe sys.modules access are unrelated." - }, - { - "left": "issue:46614", - "right": "issue:46665", - "accept": false, - "reason": "Stale phonemizer language state and inconsistent from_pretrained behavior are different bugs." + "reason": "BeitImageProcessorFast label reduction and unsafe sys.modules access are unrelated." }, { "left": "issue:43540", "right": "issue:46752", "accept": false, - "reason": "Qwen3OmniMoe video input validation and empty-list chat template handling are unrelated." + "reason": "Qwen3OmniMoe video-input processing and empty chat-template input handling are different bugs." }, { "left": "issue:45072", "right": "issue:46829", "accept": false, - "reason": "bfloat16 dtype mismatch issues are unrelated to VideoPrism hidden_states/output packaging." + "reason": "dtype mismatch in CI inference is unrelated to VideoPrism returning the wrong hidden_states tensor." + }, + { + "left": "issue:46614", + "right": "issue:46665", + "accept": false, + "reason": "Stale phonemizer language state and inconsistent from_pretrained behavior inside Python classes do not describe the same defect." }, { "left": "issue:44295", "right": "issue:45397", "accept": false, - "reason": "Reading a bufferized position_ids tensor and Gemma4 zero3 loading are different failure modes." + "reason": "Reading a registered position_ids buffer incorrectly is not the same as a Gemma4 zero3 loading issue." }, { "left": "issue:43531", "right": "issue:45362", "accept": false, - "reason": "Both are Qwen-family regressions, but one is sliding_window handling in Qwen3-MoE and the other is chat behavior in Qwen3.5; not the same concrete bug." + "reason": "Sliding_window behavior in Qwen3-MoE is distinct from a Qwen3.5 chat crash." }, { "left": "issue:42491", "right": "issue:43299", "accept": false, - "reason": "Both involve Qwen3 MoE compatibility, but one is LoRA reuse across major versions and the other is model loading breakage in dev0; different root causes." + "reason": "Both are version-related Qwen/MoE regressions, but they concern different models and failure modes, so they should not be merged." }, { "left": "issue:44246", "right": "issue:44871", "accept": false, - "reason": "Load-balancing loss gating and Gemma-3 eos_token_id inconsistency are unrelated." - }, - { - "left": "issue:44336", - "right": "issue:44485", - "accept": false, - "reason": "Terminal ANSI output bug and GLM-5 RoPE implementation discussion are unrelated." + "reason": "Import slowness and Gemma-3 eos_token_id inconsistency are unrelated." }, { "left": "issue:44242", "right": "issue:45593", "accept": false, - "reason": "Both mention auxiliary losses, but they are different models and different gating conditions; too broad to merge." + "reason": "Both mention missing auxiliary losses under flags, but they are different models and loss implementations; too broad to treat as duplicates." }, { "left": "issue:45588", "right": "issue:46851", "accept": false, - "reason": "flash_attention sink-less model crash is unrelated to Qwen3.5 variable-length position_ids." + "reason": "flash_attention s_aux handling and variable-length position_ids are unrelated bugs." + }, + { + "left": "issue:44336", + "right": "issue:44485", + "accept": false, + "reason": "ANSI codes in a loading report and GLM-5 RoPE implementation are unrelated." }, { "left": "issue:45003", "right": "issue:46821", "accept": false, - "reason": "Unsafe sys.modules access and TimesFM window_size AttributeError are unrelated." + "reason": "Unsafe sys.modules access and a TimesFM window_size AttributeError are different code paths." }, { "left": "issue:42886", @@ -13116,7 +13071,7 @@ "left": "issue:44451", "right": "issue:46752", "accept": false, - "reason": "ScandiBERT loading failure and empty-list chat template crash are unrelated." + "reason": "Model loading failure for ScandiBERT is unrelated to apply_chat_template on an empty list." } ] }, @@ -13202,8 +13157,8 @@ "issue:43540|issue:46655", "issue:44162|issue:45200", "issue:45375|issue:45759", - "issue:43262|issue:44246", "issue:44479|issue:46655", + "issue:43262|issue:44246", "issue:44295|issue:44749", "issue:43475|issue:46369", "issue:45859|issue:46399", @@ -13212,216 +13167,216 @@ "issue:44877|issue:45375", "issue:43295|issue:45292", "issue:45292|issue:45584", - "issue:44466|issue:45335", "issue:44623|issue:46726", + "issue:44466|issue:45335", "issue:44242|issue:44857", "issue:43526|issue:46897", "issue:44038|issue:46506", "issue:44987|issue:45459", "issue:29870|issue:30333" ], - "prepared_review_unit_hash": "878fac6693f8e40a18b5696fb830cb80d80b6407edfc4b04ade152691b94bacb", + "prepared_review_unit_hash": "639e7ddbde4d1a8b7d7da9db914fd7200fe30648d22e0c1f23051f6eaa3fe6a1", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36559, - "estimated_input_tokens": 9140, - "estimated_eval_tokens": 18536 + "serialized_chars": 36555, + "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 highly heterogeneous: most items are distinct bug reports across unrelated subsystems, and none of the soft pairs are strong enough to treat as duplicates. The closest themes are video-processing regressions and strict/config handling, but they still point to different root causes and code paths.", - "confidence": 0.22, - "canonical_issue_reason": "issue:44479 is the broadest and most representative issue here: it describes a concrete regression affecting multiple Qwen video models in v5.3.0, which is the closest thing to a shared bug theme in the set.", + "summary": "This cluster is heterogeneous: it mixes unrelated bugs across model loading, processors, configs, generation, deepspeed, docs, and CI. None of the soft pairs appear to be the same underlying issue, so all soft edges should be rejected.", + "confidence": 0.96, + "canonical_issue_reason": "No clear canonical issue for deduplication: the items do not form one duplicate family and span many unrelated subsystems/models.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44479 is the best issue candidate because it covers a widespread, user-visible regression with multiple affected models and a clear version-specific breakage; the others are narrower or unrelated.", + "best_issue_reason": "If a representative issue is needed, issue:44479 is the broadest concrete regression in the set, but it is not a true cluster canonical.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43550", "right": "issue:46506", "accept": false, - "reason": "Different models, different failure modes, and different subsystems; both mention runtime issues but not the same concrete bug." + "reason": "Different failures in different models/subsystems: Bamba SDPA torch.compile bug vs PPDocLayoutV3 CPU/CUDA mismatch." }, { "left": "issue:44779", "right": "issue:45292", "accept": false, - "reason": "Tokenizer correctness regression and resize_token_embeddings behavior are unrelated code paths." + "reason": "Tokenizer corruption in DeepSeek is unrelated to resize_token_embeddings not updating output embeddings." }, { "left": "issue:36246", "right": "issue:42757", "accept": false, - "reason": "Import errors from different symbols/modules with no shared underlying defect." + "reason": "Different missing imports from different packages: Qwen2.5 VL image processor vs huggingface_hub is_offline_mode." }, { "left": "issue:36246", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for different APIs and causes; not the same bug." + "reason": "Different missing symbols and code paths: Qwen2.5 VL image processor vs HybridCache import." }, { "left": "issue:41950", "right": "issue:44220", "accept": false, - "reason": "Video-classification processor lookup and fbank feature extraction are separate processing bugs." + "reason": "Video pipeline image-processor lookup bug is unrelated to audio fbank extraction issues." }, { "left": "issue:43901", "right": "issue:45003", "accept": false, - "reason": "Documentation mismatch for a pipeline setting is unrelated to unsafe sys.modules access." + "reason": "Docs mismatch for return_all_scores is unrelated to unsafe sys.modules access in modeling_utils." }, { "left": "issue:43257", "right": "issue:45137", "accept": false, - "reason": "Both involve DeepSpeed, but one is MoE weight conversion and the other is a ZeRO3 deque/indexing crash." + "reason": "Both involve DeepSpeed, but one is MOE weight conversion and the other is an empty deque crash; different bugs." }, { "left": "issue:45003", "right": "issue:45561", "accept": false, - "reason": "Unsafe module lookup and pytest-xdist file race are unrelated." + "reason": "Unsafe sys.modules access is unrelated to pytest-xdist captured_info.txt race conditions." }, { "left": "issue:44496", "right": "issue:45003", "accept": false, - "reason": "Model config loading failure and sys.modules access are not the same issue." + "reason": "Unrecognized model config loading is a different problem from sys.modules access." }, { "left": "issue:43540", "right": "issue:46655", "accept": false, - "reason": "Both are video-input bugs, but in different models with different failure mechanisms." + "reason": "Both are video-related, but Qwen3OmniMoe ValueError and SmolVLM padding bug have different model paths and root causes." }, { "left": "issue:44162", "right": "issue:45200", "accept": false, - "reason": "ESM2 breakage and Gemma 4 token-type defaults are unrelated." + "reason": "ESM2 breakage is unrelated to Gemma 4 mm_token_type_ids default behavior." }, { "left": "issue:45375", "right": "issue:45759", "accept": false, - "reason": "Both are Qwen multimodal config issues, but one is strict-field dropping and the other is AutoModelForCausalLM text_config unwrapping." + "reason": "Both mention Qwen multimodal configs, but one drops a missing field under strict validation and the other fails to unwrap text_config; not the same bug." }, { - "left": "issue:43262", - "right": "issue:44246", + "left": "issue:44479", + "right": "issue:46655", "accept": false, - "reason": "Audio chat-template sampling-rate defaulting and import latency are unrelated." + "reason": "Different models and code paths: Qwen VL video regression vs SmolVLM video processor padding." }, { - "left": "issue:44479", - "right": "issue:46655", + "left": "issue:43262", + "right": "issue:44246", "accept": false, - "reason": "Both are video regressions, but they affect different model families and appear to stem from different processor bugs." + "reason": "Audio sampling-rate defaults are unrelated to import latency." }, { "left": "issue:44295", "right": "issue:44749", "accept": false, - "reason": "A runtime position_ids error and a performance regression during filtering are not the same bug." + "reason": "Position_ids buffer access error is unrelated to a transformers version slowdown." }, { "left": "issue:43475", "right": "issue:46369", "accept": false, - "reason": "SAM 3 video encoder attribute error and SAM 3 LiteText memory overflow are distinct failures." + "reason": "SAM3 video missing attribute and SAM3LiteText memory overflow are distinct issues." }, { "left": "issue:45859", "right": "issue:46399", "accept": false, - "reason": "Both mention Qwen3.5 rotary/embed-related code, but one is CPU offload incompatibility and the other is a kernel registration decorator crash." + "reason": "Both are Qwen3.5-related, but CPU offload incompatibility and missing registration decorator are different concrete failures." }, { "left": "issue:45137", "right": "issue:45600", "accept": false, - "reason": "DeepSpeed ZeRO3 deque error and auto_mappings CI breakage are unrelated." + "reason": "ZeRO3 deque crash is unrelated to auto_mappings referencing removed Sam3LiteText configs." }, { "left": "issue:36246", "right": "issue:43596", "accept": false, - "reason": "Different import/runtime errors with no shared root cause." + "reason": "Missing Qwen2.5 VL import and DeepSpeed/BertModel index error are unrelated." }, { "left": "issue:44877", "right": "issue:45375", "accept": false, - "reason": "Both involve strict config handling, but they concern different models and different outcomes." + "reason": "Strict config rejection for granite_speech is unrelated to Qwen3_5MoeVisionConfig field dropping." }, { "left": "issue:43295", "right": "issue:45292", "accept": false, - "reason": "Processor API regression and token-embedding resize behavior are unrelated." + "reason": "Processor/tokenizer regression is unrelated to resize_token_embeddings not updating output embeddings." }, { "left": "issue:45292", "right": "issue:45584", "accept": false, - "reason": "Token embedding resizing and Whisper empty-transcription generation are different problems." + "reason": "Embedding resize bug is unrelated to Whisper empty transcription after align_special_tokens." }, { - "left": "issue:44466", - "right": "issue:45335", + "left": "issue:44623", + "right": "issue:46726", "accept": false, - "reason": "Serialization inconsistency for lm_head.weight and decoder.embed_tokens resize behavior are not the same underlying bug." + "reason": "Missing files on processor.save_pretrained is unrelated to AutoProcessor complaining about a missing vision backend." }, { - "left": "issue:44623", - "right": "issue:46726", + "left": "issue:44466", + "right": "issue:45335", "accept": false, - "reason": "Missing files on save_pretrained and unrecognized image processor on missing backend are different processor issues." + "reason": "Serialization of lm_head.weight is unrelated to t5gemma decoder.embed_tokens not updating on resize." }, { "left": "issue:44242", "right": "issue:44857", "accept": false, - "reason": "MoE load-balancing loss omission and AMP/CUDA crash in image loss are unrelated." + "reason": "MoE load-balancing loss omission is unrelated to LwDetrImageLoss AMP/CUDA crashes." }, { "left": "issue:43526", "right": "issue:46897", "accept": false, - "reason": "Label reduction bug in BeitImageProcessorFast and Florence2 training-loss shift issue are not the same change." + "reason": "BeitImageProcessorFast label reduction bug is unrelated to Florence2 training-loss double-shift." }, { "left": "issue:44038", "right": "issue:46506", "accept": false, - "reason": "Qwen3-VL-Moe bug and PPDocLayoutV3 CPU/CUDA discrepancy are unrelated." + "reason": "Qwen3-VL-Moe regression and PPDocLayoutV3 CPU/CUDA mismatch are different bugs." }, { "left": "issue:44987", "right": "issue:45459", "accept": false, - "reason": "Model loading failure for a specific repo and tokenizer protobuf error masking are different bugs." + "reason": "Loading physical-intelligence/fast and hidden tokenizer errors from protobuf absence are unrelated." }, { "left": "issue:29870", "right": "issue:30333", "accept": false, - "reason": "Bart generate() UnboundLocalError and MLflow job-status reporting are unrelated." + "reason": "Bart generate UnboundLocalError is unrelated to MLflow not being reported after failed training jobs." } ] }, @@ -13506,8 +13461,8 @@ "issue:45397|issue:45656", "issue:43065|issue:43408", "issue:43475|issue:43540", - "issue:44877|issue:46170", "issue:46170|issue:46832", + "issue:44877|issue:46170", "issue:44368|issue:45563", "issue:43723|issue:46710", "issue:44995|issue:45200", @@ -13524,208 +13479,208 @@ "issue:43065|issue:46123", "issue:44297|issue:44623" ], - "prepared_review_unit_hash": "7f378e615ded3dc5f9655bdb81d40ef945a63c31f18af692ec24323c99fd2721", + "prepared_review_unit_hash": "7e84ec6cc689ec9025e5e3ef6744f949f816a0ada7b1834d5fb0edc8b61693ed", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 35960, - "estimated_input_tokens": 8990, - "estimated_eval_tokens": 18236 + "serialized_chars": 35966, + "estimated_input_tokens": 8992, + "estimated_eval_tokens": 18240 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous; the soft pairs are mostly superficially similar but not the same underlying bug or change. No PRs are present, so the best anchor is an issue, not a PR.", - "confidence": 0.41, - "canonical_issue_reason": "issue:46858 is the best single anchor: it is a concrete, code-path-specific bug with a clear failure mode and is the closest thing here to a reusable bug family (cache length / static cache behavior). The rest of the cluster is largely unrelated.", + "summary": "This cluster is mostly heterogeneous: tokenizer/loading, multimodal model config, cache/compile, and generation bugs. Only the two StaticCache.get_seq_length reports look like the same underlying defect; the rest are distinct issues that should stay separate.", + "confidence": 0.91, + "canonical_issue_reason": "issue:46858 is the best canonical because it states the root cause and concrete failure mode most clearly: get_seq_length() returning a device tensor breaks slice arithmetic under static cache + torch.compile on non-CUDA backends. issue:45987 is the narrower contract complaint that this subsumes.", "canonical_pr_reason": null, - "best_issue_reason": "issue:46858 is the strongest issue candidate because it is explicit, actionable, and narrowly scoped, whereas most other items are broader feature regressions, warnings, or unrelated model-specific failures.", + "best_issue_reason": "issue:46858 is the strongest representative of the only true duplicate pair in this set: it is specific, actionable, and covers the broader downstream impact of the same get_seq_length bug.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43352", "right": "issue:45250", "accept": false, - "reason": "Both mention Flash Attention 2.0, but one is a model-specific unsupported-path error and the other is a generic topic issue; not the same bug." + "reason": "Both mention Flash Attention 2.0, but one is a model-support error for NemotronH and the other is a generic topic stub; not the same bug or change." }, { "left": "issue:42199", "right": "issue:46851", "accept": false, - "reason": "DET R cardinality/background-class handling and Qwen3.5 variable-length position_ids are unrelated code paths and symptoms." + "reason": "Different models and failures: DETR cardinality/background-class handling vs Qwen3.5 variable-length position_ids." }, { "left": "issue:43526", "right": "issue:46894", "accept": false, - "reason": "BEiT label reduction and CohereASR loss shifting are different model families and different failure modes." + "reason": "Unrelated bugs in different processors/models: BeitImageProcessorFast label reduction vs CohereASR training-loss double shift." }, { "left": "issue:44060", "right": "issue:44077", "accept": false, - "reason": "One is a tied-weights warning in Qwen3-Next; the other is a patchtsmixer post_init policy issue. Same broad theme only at a high level." + "reason": "Different subsystems and symptoms: Qwen3-Next tied-weights warning vs patchtsmixer post_init policy." }, { "left": "issue:44368", "right": "issue:44977", "accept": false, - "reason": "A tied-embeddings warning is not the same as a flash-attention generation regression in Qwen3.5." + "reason": "Same family name (Qwen3.5) but distinct problems: tied-word-embeddings warning vs flash-attention generation failure." }, { "left": "issue:43653", "right": "issue:44162", "accept": false, - "reason": "Tokenizer special-token registration and ESM2 breakage are different subsystems and different root causes." + "reason": "Tokenizer special-token registration bug vs ESM2 regression; no shared code path." }, { "left": "issue:45412", "right": "issue:46123", "accept": false, - "reason": "RT-DETR memory cleanup and MaskGenerationPipeline batching logic are unrelated bugs." + "reason": "RT-DETR memory cleanup vs MaskGenerationPipeline partial-batch handling; different behavior and code paths." }, { "left": "issue:44479", "right": "issue:46170", "accept": false, - "reason": "Video-input regression in Qwen models and Windows/ROCm import dependency failures are unrelated." + "reason": "Video-input regression in Qwen VL models and a Windows ROCm/torch.distributed import problem are unrelated." }, { "left": "issue:42886", "right": "issue:45229", "accept": false, - "reason": "Offline tokenizer cache loading and Gemma4 multi-GPU CUDA OOM are different classes of failure." + "reason": "Offline tokenizer cache loading and Gemma4 multi-GPU CUDA OOM are unrelated issues." }, { "left": "issue:45397", "right": "issue:45656", "accept": false, - "reason": "Gemma-4 ZeRO3 loading and double optimizer stepping under deepspeed are not the same concrete bug." + "reason": "Gemma-4 ZeRO-3 loading vs DeepSpeed optimizer-step duplication are different training/runtime bugs." }, { "left": "issue:43065", "right": "issue:43408", "accept": false, - "reason": "SAM3 pixel-decoder implementation detail and sam3_video vs sam3_tracker model-type warning are distinct issues." + "reason": "Both are SAM3-related, but one is a dummy Conv2d in the pixel decoder and the other is a model-type mismatch warning." }, { "left": "issue:43475", "right": "issue:43540", "accept": false, - "reason": "SAM3 vision encoder output attribute error and Qwen3OmniMoe video-input ValueError are unrelated." + "reason": "SAM3 video encoder missing attribute vs Qwen3OmniMoe video-input ValueError; distinct failures." }, { - "left": "issue:44877", - "right": "issue:46170", + "left": "issue:46170", + "right": "issue:46832", "accept": false, - "reason": "Strict config loading for granite_speech and ROCm/Windows distributed import errors do not share the same bug." + "reason": "Windows ROCm import/type dependency issue vs RT-DETR forward shape/type error; unrelated." }, { - "left": "issue:46170", - "right": "issue:46832", + "left": "issue:44877", + "right": "issue:46170", "accept": false, - "reason": "A platform/dependency import issue is not the same as an RT-DETR feature-level mismatch in forward." + "reason": "Strict granite_speech config loading and Windows ROCm dependency failures are not the same bug." }, { "left": "issue:44368", "right": "issue:45563", "accept": false, - "reason": "These are both warnings, but they concern different generation settings and different code paths." + "reason": "Both are warnings, but one is a tied-embeddings config warning and the other is a stale num_return_sequences warning in paged generate." }, { "left": "issue:43723", "right": "issue:46710", "accept": false, - "reason": "Tokenizer loading in v5 and an accuracy regression after upgrade are too different to be duplicates." + "reason": "Tokenizer loading in v5 and an accuracy regression after upgrade are different classes of issue." }, { "left": "issue:44995", "right": "issue:45200", "accept": false, - "reason": "GlmMoeDsa stale cache crash and Gemma 4 mm_token_type_ids defaults are different model behaviors." + "reason": "GlmMoeDsa second-forward cache bug vs Gemma 4 mm_token_type_ids defaulting; unrelated." }, { "left": "issue:43931", "right": "issue:44368", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch and tied-embeddings warnings are unrelated." + "reason": "Qwen3-VL weight-shape loading mismatch vs Qwen3.5 tied-embeddings warning; different code paths." }, { "left": "issue:44188", "right": "issue:46202", "accept": false, - "reason": "torch.compile attention-kernel divergence and Gemma token roundtrip failure are not the same issue." + "reason": "torch.compile branching in attention kernels and Gemma token roundtrip on \u2581 are unrelated." }, { "left": "issue:44483", "right": "issue:46202", "accept": false, - "reason": "Chat-completions API rejection and Gemma pipeline roundtrip failure are different regressions." + "reason": "Chat/completions API rejection vs tokenizer roundtrip failure; no shared underlying defect." }, { "left": "issue:46829", "right": "issue:46858", "accept": false, - "reason": "VideoPrism output field mix-up and cache.get_seq_length returning a device tensor are different bugs." + "reason": "VideoPrism output field mix-up is a model-output bug, not the StaticCache get_seq_length contract problem." }, { "left": "issue:45987", "right": "issue:46082", "accept": false, - "reason": "A cache return-type contract bug and a LlamaConfig head_dim validation issue are unrelated." + "reason": "StaticCache sequence-length return type and LlamaConfig head_dim validation are different issues." }, { "left": "issue:43906", "right": "issue:46202", "accept": false, - "reason": "An isolated reproduction of another issue is not enough to treat it as the same bug as the Gemma roundtrip failure." + "reason": "Isolated reproduction of another issue vs Gemma roundtrip failure; not the same bug." }, { "left": "issue:44995", "right": "issue:45588", "accept": false, - "reason": "A stale cache crash in GlmMoeDsa and a flash_attention.py None-handling crash are separate code paths." + "reason": "Both are runtime bugs, but one is stale indexer cache and the other is flash_attention s_aux=None handling." }, { "left": "issue:44188", "right": "issue:46421", "accept": false, - "reason": "Both mention compile/cache behavior, but one is attention-kernel divergence and the other is lazy cache init causing recompiles; not the same fix." + "reason": "Both mention compile/cache behavior, but one is kernel divergence under torch.compile and the other is lazy cache init causing recompilation." }, { "left": "issue:45685", "right": "issue:46431", "accept": false, - "reason": "MPS histogram integer support and missing chat template validation are unrelated." + "reason": "MPS histogram on Int vs missing chat template in a processor are unrelated." }, { "left": "issue:45103", "right": "issue:46858", "accept": false, - "reason": "auto_docstring annotation handling and cache length arithmetic are completely different concerns." + "reason": "Docstring annotation handling and StaticCache get_seq_length return type are separate bugs." }, { "left": "issue:43065", "right": "issue:46123", "accept": false, - "reason": "SAM3 pixel-decoder internals and partial-batch masking logic do not describe the same defect." + "reason": "SAM3 pixel-decoder dummy layer and MaskGenerationPipeline partial-batch handling are unrelated." }, { "left": "issue:44297", "right": "issue:44623", "accept": false, - "reason": "Tokenizer save_pretrained metadata mismatch and processor save_pretrained missing files are both persistence-related, but they are different failure modes and artifacts." + "reason": "Both involve save_pretrained, but tokenizer_class metadata mismatch and processor missing files are different serialization bugs." } ] }, @@ -13807,18 +13762,18 @@ "issue:44623|issue:45656", "issue:43646|issue:44162", "issue:43441|issue:46752", - "issue:45375|issue:45478", "issue:42994|issue:43482", + "issue:45375|issue:45478", "issue:43475|issue:43643", "issue:43232|issue:43673", "issue:44188|issue:46851", "issue:28282|issue:30990", - "issue:41628|issue:44933", "issue:43475|issue:46821", - "issue:45230|issue:45397", - "issue:44964|issue:45137", + "issue:41628|issue:44933", "issue:45310|issue:45375", + "issue:45230|issue:45397", "issue:43986|issue:46726", + "issue:44964|issue:45137", "issue:44869|issue:45356", "issue:44038|issue:44623", "issue:39692|issue:46726", @@ -13829,20 +13784,20 @@ "issue:43723|issue:46326", "issue:45406|issue:46202" ], - "prepared_review_unit_hash": "f8827e362c7661cdd0dcc697950e8ba2fdb86000c6f81eeef3ed42af8b8a0730", + "prepared_review_unit_hash": "75c1f2ebdb055c603ba94d11ddb851d40df412e844b261505250fc6903d40ff8", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36055, + "serialized_chars": 36056, "estimated_input_tokens": 9014, "estimated_eval_tokens": 18284 }, @@ -13851,186 +13806,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose grab-bag of Transformers v5 regressions and model-specific loading/behavior issues. The soft links are generally only broad topical similarity, not the same underlying bug or change, so they should stay separate.", - "confidence": 0.38, - "canonical_issue_reason": "Issue 43646 is the broadest, most representative regression report in the set: it frames a general Transformers 5.0 compatibility break around custom model initialization, which matches the cluster\u2019s overall theme better than the narrower model-specific reports.", + "summary": "This set is a grab bag of unrelated Transformers issues; none of the soft-edge pairs look like the same concrete bug or fix. Most similarities are broad subsystem/model-family overlap rather than duplicate reports.", + "confidence": 0.91, + "canonical_issue_reason": "issue:44162 is the strongest canonical issue candidate because it describes a severe, clearly stated regression with broad user impact; the other issues are narrower, more model-specific, or more ambiguous.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43646 is the best single issue representative because it is general, clearly regression-oriented, and spans the same v5 compatibility class as many of the other reports.", + "best_issue_reason": "issue:44162 is the best overall issue for triage quality and impact: it is concrete, high-severity, and well-scoped compared with the rest.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44361", "right": "issue:44749", "accept": false, - "reason": "Different failures: MLukeTokenizer task-time AttributeError vs a performance regression after upgrading Transformers. Same library, but not the same bug path." + "reason": "Different bugs: tokenizer/task AttributeError vs a performance regression after upgrading Transformers." }, { "left": "issue:43994", "right": "issue:46726", "accept": false, - "reason": "Both involve vision/image tooling, but one reports bad outputs from AutoModel/pipeline and the other is a missing-backend error in AutoProcessor. Different problems and fixes." + "reason": "Both involve vision/AutoProcessor paths, but one is wrong SigLIP2 outputs and the other is a missing-backend error message." }, { "left": "issue:44749", "right": "issue:45656", "accept": false, - "reason": "A data-processing slowdown after upgrade is unrelated to DeepSpeed causing optimizer.step to run twice." + "reason": "Unrelated problems: data filtering slowdown versus DeepSpeed optimizer stepping twice." }, { "left": "issue:43701", "right": "issue:44749", "accept": false, - "reason": "Checkpoint key mismatch and slowdown after upgrade are unrelated symptoms with different code paths." + "reason": "Checkpoint key mismatch is a separate training/resume issue, not the same as the performance regression." }, { "left": "issue:44871", "right": "issue:46132", "accept": false, - "reason": "Gemma eos_token_id inconsistency is unrelated to AttentionInterface.register changing callable behavior." + "reason": "Different code paths: Gemma EOS token config inconsistency vs AttentionInterface registration behavior." }, { "left": "issue:44623", "right": "issue:45656", "accept": false, - "reason": "Missing files from processor.save_pretrained is a serialization bug, not a DeepSpeed optimizer-step issue." + "reason": "Saving processor files is unrelated to the DeepSpeed optimizer-step bug." }, { "left": "issue:43646", "right": "issue:44162", "accept": false, - "reason": "Custom model initialization breakage and ESM2 being broken are both broad regressions, but they are distinct model-specific failures." + "reason": "AutoConfig trust_remote_code field loss is a different bug from the ESM2 regression." }, { "left": "issue:43441", "right": "issue:46752", "accept": false, - "reason": "FlashAttention failure for Ministral-3 is unrelated to apply_chat_template crashing on an empty list." + "reason": "FlashAttention/Ministral failure and empty-chat-template IndexError are unrelated." }, { - "left": "issue:45375", - "right": "issue:45478", + "left": "issue:42994", + "right": "issue:43482", "accept": false, - "reason": "Both are Qwen3.5/MoE related, but one is a missing config field under @strict while the other is a generic from_pretrained regression; not clearly the same concrete bug." + "reason": "Both are quantization/model-loading themed, but one is saving a quantized model and the other is loading a GGUF model." }, { - "left": "issue:42994", - "right": "issue:43482", + "left": "issue:45375", + "right": "issue:45478", "accept": false, - "reason": "Quantized model saving failure and GGUF loading failure are different quantization pathways with different failure modes." + "reason": "Same model family, but one is a missing config field dropped by strict validation and the other is a from_pretrained error; not clearly the same bug." }, { "left": "issue:43475", "right": "issue:43643", "accept": false, - "reason": "SAM 3 Video missing attribute and AutoConfig trust_remote_code field loss are unrelated issues." + "reason": "SAM video encoder attribute failure is unrelated to AutoConfig remote-code field handling." }, { "left": "issue:43232", "right": "issue:43673", "accept": false, - "reason": "Both are generation-related, but sync_gpus kwarg handling and chunked_prefill cache availability are different bugs." + "reason": "Both are generation-related, but they describe different failures in different parts of generation state handling." }, { "left": "issue:44188", "right": "issue:46851", "accept": false, - "reason": "Attention kernel divergence under torch.compile is unrelated to Qwen3.5 variable-length position_ids handling." - }, - { - "left": "issue:28282", - "right": "issue:30990", - "accept": false, - "reason": "PyTorch-not-found import error and Sentence Transformers hanging on load are different issues." + "reason": "Attention kernel divergence under torch.compile is not the same as variable-length position_ids handling in Qwen3.5." }, { - "left": "issue:41628", - "right": "issue:44933", + "left": "issue:28282", + "right": "issue:30990", "accept": false, - "reason": "Both are import/export problems, but they concern different missing symbols and different modules; too coarse to merge." + "reason": "Missing PyTorch import error is unrelated to Sentence Transformers hanging during load." }, { "left": "issue:43475", "right": "issue:46821", "accept": false, - "reason": "SAM 3 Video attribute error and TimesFM window_size AttributeError are different model-specific failures." + "reason": "Different models and failures: SAM3 vision encoder attribute error vs TimesFM window_size attribute error." }, { - "left": "issue:45230", - "right": "issue:45397", + "left": "issue:41628", + "right": "issue:44933", "accept": false, - "reason": "A generic bug report and Gemma-4 zero3 from_pretrained failure are unrelated." + "reason": "Both mention import/export issues, but the specific missing symbols and failure modes are not clearly the same." }, { - "left": "issue:44964", - "right": "issue:45137", + "left": "issue:45310", + "right": "issue:45375", "accept": false, - "reason": "Phi-4 multimodal loading failure and DeepSpeed ZeRO3 deque underflow are unrelated subsystems." + "reason": "Related Qwen3.5 topics, but one is a from_pretrained error and the other is a strict-validation field-drop bug." }, { - "left": "issue:45310", - "right": "issue:45375", + "left": "issue:45230", + "right": "issue:45397", "accept": false, - "reason": "Both touch Qwen3.5/MoE, but one is a generic from_pretrained error and the other is a specific missing deepstack_visual_indexes field; not enough evidence they are the same underlying defect." + "reason": "Generic bug report is not the same concrete issue as Gemma-4 ZeRO3 from_pretrained failure." }, { "left": "issue:43986", "right": "issue:46726", "accept": false, - "reason": "AutoProcessor crash without torchvision and AutoProcessor reporting an unrecognized image processor are different missing-dependency / error-message behaviors." + "reason": "Both involve missing vision dependencies, but one is a video-model AutoProcessor crash and the other is an image-processor recognition error." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Multimodal model loading failure is unrelated to DeepSpeed ZeRO3 deque underflow." }, { "left": "issue:44869", "right": "issue:45356", "accept": false, - "reason": "Whisper timestamp decode crash and Kimi-K2.5 tokenizer codec regression are unrelated tokenizer issues." + "reason": "Whisper decode crash from replacement character is a different tokenizer bug than Kimi-K2.5 codec handling regression." }, { "left": "issue:44038", "right": "issue:44623", "accept": false, - "reason": "Qwen3-VL-Moe from_pretrained failures and processor.save_pretrained missing files are different operations and failure modes." + "reason": "Qwen3-VL-Moe loading error and processor.save_pretrained missing files are distinct failures." }, { "left": "issue:39692", "right": "issue:46726", "accept": false, - "reason": "Documentation example errors for SigLIP2 and a runtime missing-vision-backend crash are not the same bug." + "reason": "Documentation example errors are not the same as the AutoProcessor missing-backend crash." }, { "left": "issue:46463", "right": "issue:46877", "accept": false, - "reason": "Incorrect T5 decoder-input validation messaging and auto_docstring emitting diagnostics at import time are unrelated docstring/lint behaviors." + "reason": "Incorrect T5 validation error message is unrelated to auto_docstring printing diagnostics at import time." }, { "left": "issue:44373", "right": "issue:46851", "accept": false, - "reason": "A wrong docstring for position_ids is unrelated to a runtime bug in variable-length inputs using position_ids." + "reason": "A wrong docstring is not the same as a functional bug in position_ids handling." }, { "left": "issue:42994", "right": "issue:43525", "accept": false, - "reason": "Quantized model saving failure and Llama4Config missing pad_token_id are different issues." + "reason": "Quantized model saving failure and missing pad_token_id on Llama4Config are separate issues." }, { "left": "issue:42371", "right": "issue:45092", "accept": false, - "reason": "TF32 settings guidance and InternVL2 meta-initialization incompatibility are unrelated." + "reason": "TF32 settings guidance is unrelated to InternVL2 meta-initialization incompatibility." }, { "left": "issue:43723", "right": "issue:46326", "accept": false, - "reason": "Tokenizer loading in v5 and continue_final_message behavior for mistral_common are different tokenizer bugs." + "reason": "AutoTokenizer loading regression and continue_final_message behavior for mistral_common tokenizers are different bugs." }, { "left": "issue:45406", "right": "issue:46202", "accept": false, - "reason": "Gemma4Processor missing _tokenizer and Gemma pipeline roundtrip failure on \u2581 are different bugs in different layers." + "reason": "Gemma4Processor missing _tokenizer and Gemma pipeline roundtrip failure are not the same underlying bug." } ] }, @@ -14112,16 +14067,16 @@ "issue:45593|issue:46877", "issue:42673|issue:43704", "issue:44038|issue:45310", - "issue:36010|issue:41628", "issue:44222|issue:45310", - "issue:43278|issue:45137", + "issue:36010|issue:41628", "issue:46202|issue:46701", "issue:42994|issue:43824", + "issue:43278|issue:45137", "issue:45081|issue:45103", "issue:42831|issue:44222", "issue:43232|issue:44792", - "issue:43504|issue:45561", "issue:42175|issue:44933", + "issue:43504|issue:45561", "issue:45561|issue:45600", "issue:43976|issue:44933", "issue:44496|issue:45561", @@ -14135,223 +14090,223 @@ "issue:44623|issue:45310", "issue:42673|issue:46693" ], - "prepared_review_unit_hash": "950da8a2926f00589bfc9e633d9d5b7ebb65c3add4a1fe373138a7466d403e88", + "prepared_review_unit_hash": "fb74c9f6b450fa7ae2564345acba5b29eb6b1558545bab7f0a3393c14df2dd97", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36539, - "estimated_input_tokens": 9135, - "estimated_eval_tokens": 18526 + "serialized_chars": 36536, + "estimated_input_tokens": 9134, + "estimated_eval_tokens": 18524 }, "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 clearest duplicate pair is the repeated Qwen3ForCausalLM VRAM leak report (#42673/#43704); the rest do not look like the same underlying bug.", - "confidence": 0.95, - "canonical_issue_reason": "#42673 is the best canonical issue because it is the older report in the clearest duplicate pair with #43704, and it has more discussion activity plus the same title/symptom.", + "summary": "This cluster is mostly heterogeneous and does not form a single coherent duplicate set. The only clear duplicate-looking pair is the repeated Qwen3ForCausalLM VRAM leak report (42673/43704).", + "confidence": 0.74, + "canonical_issue_reason": "Issue 42673 is the earlier, fuller report of the exact duplicate pair with 43704 (same Qwen3ForCausalLM VRAM leak title and symptom), so it is the best canonical issue candidate.", "canonical_pr_reason": null, - "best_issue_reason": "#42673 is the strongest representative issue in this cluster: it matches #43704 on the Qwen3ForCausalLM VRAM leak symptom and has the richer discussion context.", + "best_issue_reason": "42673 is the clearest standalone representative in this noisy cluster because it matches 43704 exactly and has more discussion/context than the later duplicate.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43873", "right": "issue:44945", "accept": false, - "reason": "Quantization/offloading failure vs pipeline-parallel output correctness; different code paths and symptoms." + "reason": "Both are generation/runtime complaints, but one is quantization/offloading and the other is pipeline-parallel output corruption; different bugs." }, { "left": "issue:44945", "right": "issue:45753", "accept": false, - "reason": "Pipeline-parallel output issue vs image-specific infinite loop; not the same bug." + "reason": "Different failure modes: pipeline parallelism output issue vs Qwen3_5 image-specific infinite loop." }, { "left": "issue:43784", "right": "issue:46399", "accept": false, - "reason": "Import-time NameError in sentence-transformers integration vs Qwen3.5 kernel registration crash; unrelated failures." + "reason": "Import NameError in sentence-transformers vs Qwen3.5 kernel registration ValueError; unrelated code paths." }, { "left": "issue:41628", "right": "issue:44351", "accept": false, - "reason": "Different missing symbols (`AutoImageProcessor` vs `HybridCache`) and likely separate API breakages." + "reason": "Both are import errors, but for different symbols/classes and likely different breakages." }, { "left": "issue:44857", "right": "issue:46877", "accept": false, - "reason": "CUDA/AMP loss crash vs docstring lint logging at import time; no shared bug." + "reason": "CUDA AMP loss crash vs auto_docstring import-time logging; unrelated." }, { "left": "issue:45593", "right": "issue:46877", "accept": false, - "reason": "Training loss/auxiliary-loss behavior vs auto_docstring stdout noise; unrelated." + "reason": "Auxiliary-loss behavior in D-FINE vs docstring diagnostics printing; unrelated." }, { "left": "issue:42673", "right": "issue:43704", "accept": true, - "reason": "Same title and same symptom: Qwen3ForCausalLM VRAM leak when used in multiple dataloader threads; clearly the same underlying bug." + "reason": "Same title and same concrete symptom: Qwen3ForCausalLM VRAM leak when used in multiple dataloader threads; later report is a duplicate." }, { "left": "issue:44038", "right": "issue:45310", "accept": false, - "reason": "Different Qwen MoE models, different version constraints, and different failure modes." - }, - { - "left": "issue:36010", - "right": "issue:41628", - "accept": false, - "reason": "Separate import errors for different exported symbols; not one bug." + "reason": "Both mention Qwen3/VL-Moe loading, but the model/version context and concrete failure differ." }, { "left": "issue:44222", "right": "issue:45310", "accept": false, - "reason": "FP8 save_pretrained failure vs MoE from_pretrained error; different concrete code paths." + "reason": "FP8 save_pretrained failure vs Qwen3.5 from_pretrained load error; different issues." }, { - "left": "issue:43278", - "right": "issue:45137", + "left": "issue:36010", + "right": "issue:41628", "accept": false, - "reason": "Embedding dtype drift during eval vs DeepSpeed ZeRO3 deque IndexError; unrelated." + "reason": "Import errors for different missing names (GenerationMixin vs AutoImageProcessor)." }, { "left": "issue:46202", "right": "issue:46701", "accept": false, - "reason": "Gemma token roundtrip issue vs DiffusionGemma pipeline failure; too different to merge." + "reason": "Both are Gemma-related, but one is token roundtrip on \u2581 and the other is image-text pipeline failure; different bugs." }, { "left": "issue:42994", "right": "issue:43824", "accept": false, - "reason": "Quantized model saving failure vs missing Qwen2_5_VL class import; unrelated." + "reason": "Quantized save_pretrained failure vs missing Qwen2_5_VLForConditionalGeneration import." + }, + { + "left": "issue:43278", + "right": "issue:45137", + "accept": false, + "reason": "Embedding dtype drift vs ZeRO3 deque IndexError; no overlap." }, { "left": "issue:45081", "right": "issue:45103", "accept": false, - "reason": "Tokenizer regex patch crash vs auto_docstring annotation handling crash; different subsystems." + "reason": "Tokenizer regex crash vs docstring annotation AttributeError; unrelated." }, { "left": "issue:42831", "right": "issue:44222", "accept": false, - "reason": "FP8 accuracy regression vs FP8 model saving bug; same broad area but not the same bug." + "reason": "Both touch FP8, but one is an accuracy regression and the other is a MoE save_pretrained bug." }, { "left": "issue:43232", "right": "issue:44792", "accept": false, - "reason": "Generation kwargs/sync_gpus bug vs Janus image-generation test failure; not the same failure." + "reason": "Generation kwargs bug after sync_gpus vs janus image generation test failure; different code paths." }, { - "left": "issue:43504", - "right": "issue:45561", + "left": "issue:42175", + "right": "issue:44933", "accept": false, - "reason": "Legacy-field model loading issue vs pytest-xdist race on a temp file; unrelated." + "reason": "Backend extras packaging issue vs missing image_utils import; unrelated." }, { - "left": "issue:42175", - "right": "issue:44933", + "left": "issue:43504", + "right": "issue:45561", "accept": false, - "reason": "Backend extra packaging problem vs missing image_utils import; different packaging/API issues." + "reason": "Pretrained model loading legacy-field bug vs xdist race in testing utils." }, { "left": "issue:45561", "right": "issue:45600", "accept": false, - "reason": "File-race in patched testing utils vs stale auto_mappings references; separate CI/test problems." + "reason": "Captured_info.txt race vs auto_mappings stale config references; different subsystems." }, { "left": "issue:43976", "right": "issue:44933", "accept": false, - "reason": "Python version compatibility issue vs missing import from image_utils; unrelated." + "reason": "Python version support regression vs missing image_utils import; unrelated." }, { "left": "issue:44496", "right": "issue:45561", "accept": false, - "reason": "Unrecognized model/config loading failure vs xdist file race; no shared code path." + "reason": "Unrecognized model/config loading vs xdist race in patched testing utils." }, { "left": "issue:30333", "right": "issue:33290", "accept": false, - "reason": "MLflow reporting failure vs DeepSpeed OOM; different runtime issues." + "reason": "MLflow reporting failure vs DeepSpeed/Adafactor OOM; different problems." }, { "left": "issue:45405", "right": "issue:46877", "accept": false, - "reason": "Dependency version bump complaint vs docstring import-time output; unrelated." + "reason": "MIN_PEFT_VERSION packaging issue vs docstring stdout/stderr logging." }, { "left": "issue:30990", "right": "issue:35141", "accept": false, - "reason": "Loading hang in Sentence Transformers vs embedding reinitialization after resize; different bugs." + "reason": "Loading hang in sentence-transformers vs post_init embedding reinitialization; different behaviors." }, { "left": "issue:42898", "right": "issue:44568", "accept": false, - "reason": "Tokenization cleanup behavior change vs add_special_tokens/BOS-EOS regression; related domain but different behavior." + "reason": "Tokenizer cleanup behavior change vs add_special_tokens/BOS-EOS regression; different tokenizer bugs." }, { "left": "issue:42175", "right": "issue:45712", "accept": false, - "reason": "Torch extra packaging issue vs leftover dummy classes leaking into dir(); different maintenance problems." + "reason": "TensorFlow backend packaging issue vs leftover dummy classes/check_repo failure." }, { "left": "issue:43479", "right": "issue:46829", "accept": false, - "reason": "Multimodal config default-init bug vs hidden_states field mismatch in video classification output; unrelated." + "reason": "Default multimodal config initialization bug vs hidden_states output bug in video classification." }, { "left": "issue:46077", "right": "issue:46123", "accept": false, - "reason": "Bidirectional attention mask bug vs final partial batch dropping results; different pipelines." + "reason": "Attention mask construction vs pipeline batching state flag; unrelated." }, { "left": "issue:44623", "right": "issue:45310", "accept": false, - "reason": "processor.save_pretrained missing files vs Qwen3.5 MoE from_pretrained error; not the same bug." + "reason": "Processor save_pretrained missing files vs Qwen3.5 from_pretrained error; unrelated." }, { "left": "issue:42673", "right": "issue:46693", "accept": false, - "reason": "VRAM leak in multi-threaded Qwen3ForCausalLM loading vs FA2 long-context performance regression; unrelated performance problems." + "reason": "VRAM leak in dataloader threads vs FA2 long-context performance regression; different symptoms and code paths." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The duplicate-pair judgment is mostly conservative, but the canonical issue reason overstates the packet in two places: it claims this is the \"only true duplicate pair\" and says #42673 has \"more references,\" neither of which is directly supported by the packet (both issues show 0 inbound references). Please soften those claims to what is actually evidenced, e.g. older issue with more discussion activity and matching title/symptom." + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted duplicate pair, issue 42673/43704, is well supported by identical titles and matching VRAM-leak symptom, with 42673 having more discussion/context. The remaining soft-edge rejections avoid overclaiming and are phrased cautiously." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -14420,8 +14375,8 @@ "soft_pairs": [ "issue:43479|issue:44869", "issue:43723|issue:44568", - "issue:44016|issue:46231", "issue:42898|issue:45812", + "issue:44016|issue:46231", "issue:44625|issue:46877", "issue:42898|issue:44779", "issue:39692|issue:46909", @@ -14430,8 +14385,8 @@ "issue:43784|issue:43824", "issue:27957|issue:30333", "issue:44734|issue:45103", - "issue:30333|issue:34689", "issue:42222|issue:45200", + "issue:30333|issue:34689", "issue:42898|issue:43295", "issue:42898|issue:45356", "issue:44295|issue:45656", @@ -14442,27 +14397,27 @@ "issue:22355|issue:28282", "issue:44060|issue:45125", "issue:44075|issue:46231", - "issue:44360|issue:46762", "issue:44933|issue:45712", + "issue:44360|issue:46762", "issue:42831|issue:44623", "issue:44297|issue:45656", "issue:42796|issue:43723", "issue:44222|issue:44623" ], - "prepared_review_unit_hash": "fae5594fb13d5113b23540c680f434d8fbb0e67d0c6a75ff44b6b66cdff312fd", + "prepared_review_unit_hash": "b02bf6580a1bb1e7f010f72f973b253b5700f20fec028230956dc62303bfb43b", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36263, + "serialized_chars": 36261, "estimated_input_tokens": 9066, "estimated_eval_tokens": 18388 }, @@ -14471,72 +14426,72 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No strong duplicate group emerges here: the items are mostly unrelated Transformers issues spanning tokenizers, multimodal configs, docs, model loading, and training utilities. The closest theme is general v5/tokenizer regression, but the soft pairs are still too different to merge as duplicates.", - "confidence": 0.86, - "canonical_issue_reason": "issue:42898 is the most umbrella-like representative of the recurring tokenizer/v5 regression theme, with a broad user-facing behavior change rather than a narrow model-specific symptom.", + "summary": "The cluster is mostly heterogeneous; the soft matches are topic-adjacent but not true duplicates. The closest anchor is a tokenizer regression report, but I would not merge any of the soft pairs.", + "confidence": 0.58, + "canonical_issue_reason": "Issue 42898 is the most broadly representative regression report in the set: it concerns a versioned tokenizer behavior change and is the clearest anchor among otherwise unrelated issues.", "canonical_pr_reason": null, - "best_issue_reason": "issue:42898 is the best single issue to anchor this noisy cluster because it is broad, clearly stated, and closest to the shared regression theme across several nearby candidates.", + "best_issue_reason": "Issue 42898 has the highest general triage value because it describes a concrete user-facing v5 regression with likely broad impact, even though the cluster does not form a true duplicate set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43479", "right": "issue:44869", "accept": false, - "reason": "Different bugs: default multimodal config initialization vs a Whisper word-timestamp decode crash." + "reason": "Different bugs: multimodal config default initialization vs Whisper word-timestamp decode crash." }, { "left": "issue:43723", "right": "issue:44568", "accept": false, - "reason": "Both are v5/tokenizer-related, but one is an AutoTokenizer loading failure and the other is missing BOS/EOS insertion for a specific tokenizer." + "reason": "Both involve tokenizers, but one is a tokenizer loading/import issue and the other is a specific add_special_tokens regression on mdeberta-v3." }, { - "left": "issue:44016", - "right": "issue:46231", + "left": "issue:42898", + "right": "issue:45812", "accept": false, - "reason": "A notebook syntax error and a docs parameter-name mismatch are unrelated documentation issues." + "reason": "Both are tokenizer regressions in v5, but they affect different code paths: cleanup behavior vs Granite token IDs." }, { - "left": "issue:42898", - "right": "issue:45812", + "left": "issue:44016", + "right": "issue:46231", "accept": false, - "reason": "Both involve tokenizers, but one is a general clean_up_tokenization_spaces v5 behavior change and the other is Granite token IDs being wrong." + "reason": "Notebook syntax error and doc parameter mismatch are unrelated documentation issues." }, { "left": "issue:44625", "right": "issue:46877", "accept": false, - "reason": "Qwen3.5 label propagation and auto_docstring stdout logging are unrelated subsystems." + "reason": "Qwen config propagation bug and auto_docstring stdout logging are unrelated." }, { "left": "issue:42898", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different models and different failure modes." + "reason": "Both are tokenizer regressions, but the affected models and failure mechanisms differ." }, { "left": "issue:39692", "right": "issue:46909", "accept": false, - "reason": "Different broken documentation examples for different models; not the same underlying change." + "reason": "Both are broken documentation examples, but they are different examples with different errors." }, { "left": "issue:43479", "right": "issue:44079", "accept": false, - "reason": "Multimodal config defaults and ModelOutput key assignment are unrelated code paths." + "reason": "Multimodal config initialization and ModelOutput key assignment are different code paths." }, { "left": "issue:43874", "right": "issue:45092", "accept": false, - "reason": "A missing image-processor method and remote-code/meta-init incompatibility are distinct multimodal loading issues." + "reason": "Vision processor API missing method vs meta-initialization incompatibility for old remote-code checkpoints." }, { "left": "issue:43784", "right": "issue:43824", "accept": false, - "reason": "Different import errors from different causes: missing nn import vs missing Qwen2.5-VL export." + "reason": "Distinct import failures with different missing symbols and causes." }, { "left": "issue:27957", @@ -14548,115 +14503,115 @@ "left": "issue:44734", "right": "issue:45103", "accept": false, - "reason": "Serve/KV-cache indexing crash and auto_docstring annotation handling are unrelated." + "reason": "Serving KV-cache indexing crash and auto_docstring annotations crash are unrelated." }, { - "left": "issue:30333", - "right": "issue:34689", + "left": "issue:42222", + "right": "issue:45200", "accept": false, - "reason": "MLflow failure reporting and Llama model loading regression are different problems." + "reason": "Both are multimodal-model issues, but they concern different models and different failures." }, { - "left": "issue:42222", - "right": "issue:45200", + "left": "issue:30333", + "right": "issue:34689", "accept": false, - "reason": "VitPose model breakage and Gemma 4 mm_token_type_ids defaults are different model-specific bugs." + "reason": "MLflow failed-job reporting and model loading regression are unrelated." }, { "left": "issue:42898", "right": "issue:43295", "accept": false, - "reason": "Both are v5 regressions involving tokenizer/processor behavior, but the concrete failures are different." + "reason": "Both are transformer regressions, but one is tokenizer cleanup behavior and the other is processor/tokenizer image-handling compatibility." }, { "left": "issue:42898", "right": "issue:45356", "accept": false, - "reason": "Tokenizer regressions in v5, but one is clean_up_tokenization_spaces behavior and the other is Kimi codec/regex handling." + "reason": "Different tokenizer regressions with different symptoms and code paths." }, { "left": "issue:44295", "right": "issue:45656", "accept": false, - "reason": "Reading a registered buffer and deepspeed double optimizer steps are unrelated." + "reason": "Position_ids buffer access bug and deepspeed optimizer double-step are unrelated." }, { "left": "issue:43116", "right": "issue:44190", "accept": false, - "reason": "Two different example scripts with different failures; not the same bug." + "reason": "Different example-script failures: empty multi-label outputs vs local dataset loading." }, { "left": "issue:30333", "right": "issue:35141", "accept": false, - "reason": "Failed MLflow reporting and embedding reinitialization after resize are unrelated." + "reason": "MLflow reporting on training failure and embedding reinitialization on resize are unrelated." }, { "left": "issue:45405", "right": "issue:45712", "accept": false, - "reason": "A dependency version bump problem and leftover dummy class cleanup are not the same issue." + "reason": "A dependency/version bump issue and leftover dummy-class cleanup are unrelated." }, { "left": "issue:44222", "right": "issue:44297", "accept": false, - "reason": "Both mention save_pretrained, but one is FP8 MoE model saving and the other is tokenizer config metadata." + "reason": "Different save_pretrained bugs: FP8 MoE export vs tokenizer_class mismatch." }, { "left": "issue:22355", "right": "issue:28282", "accept": false, - "reason": "Import errors, but one is missing transformers.onnx and the other is missing PyTorch entirely." + "reason": "Both are import/dependency errors, but they stem from different missing components and are not the same bug." }, { "left": "issue:44060", "right": "issue:45125", "accept": false, - "reason": "Different Qwen model bugs: a wrong tied-weights warning versus missing tensor-parallel plan metadata." + "reason": "Both are Qwen3-family issues, but one is a tied-weights warning and the other is missing tensor-parallel planning metadata." }, { "left": "issue:44075", "right": "issue:46231", "accept": false, - "reason": "Optimizer behavior and documentation wording mismatch are unrelated." + "reason": "Optimizer argument usage bug and documentation parameter-name mismatch are unrelated." }, { - "left": "issue:44360", - "right": "issue:46762", + "left": "issue:44933", + "right": "issue:45712", "accept": false, - "reason": "A DSA indexer implementation concern and Minimax attention numerical differences are different algorithmic issues." + "reason": "Import exposure in image_utils and leftover dummy classes failing checks are different hygiene issues." }, { - "left": "issue:44933", - "right": "issue:45712", + "left": "issue:44360", + "right": "issue:46762", "accept": false, - "reason": "Both touch exports/symbol visibility, but one is a nonexistent import and the other is leftover dummy classes in a different module." + "reason": "DSA indexer ReLU discussion and Minimax numerical differences are unrelated math/model issues." }, { "left": "issue:42831", "right": "issue:44623", "accept": false, - "reason": "Fine-grained FP8 accuracy and processor.save_pretrained file omission are unrelated." + "reason": "FP8 accuracy issue and processor.save_pretrained missing files are unrelated." }, { "left": "issue:44297", "right": "issue:45656", "accept": false, - "reason": "Tokenizer save metadata and deepspeed optimizer stepping are unrelated code paths." + "reason": "Tokenizer save metadata mismatch and optimizer double-step are unrelated." }, { "left": "issue:42796", "right": "issue:43723", "accept": false, - "reason": "Both are tokenizer-related, but one is BPE marker handling for a specific model and the other is a generic AutoTokenizer loading issue." + "reason": "Both concern tokenizers, but Ministral BPE-marker handling and tokenizer loading/import failure are different bugs." }, { "left": "issue:44222", "right": "issue:44623", "accept": false, - "reason": "Both involve persistence APIs, but one is FP8 MoE model saving and the other is processor file output; not the same bug." + "reason": "Different save_pretrained problems affecting different objects and failure modes." } ] }, @@ -14733,8 +14688,8 @@ "issue:45245|issue:46132", "issue:43479|issue:43644", "issue:41628|issue:42757", - "issue:44373|issue:44485", "issue:44297|issue:45310", + "issue:44373|issue:44485", "issue:46132|issue:46710", "issue:45071|issue:46082", "issue:44222|issue:44484", @@ -14746,9 +14701,9 @@ "issue:43232|issue:45834", "issue:44295|issue:44623", "issue:43475|issue:46305", - "issue:44263|issue:44360", "issue:46209|issue:46431", "issue:46305|issue:46655", + "issue:44263|issue:44360", "issue:44263|issue:44373", "issue:30990|issue:38175", "issue:42831|issue:45310", @@ -14760,214 +14715,214 @@ "issue:44261|issue:45092", "issue:44483|issue:45406" ], - "prepared_review_unit_hash": "fcb8986b7140d2eae39939c643fc793a6d2518b1657b67aaeeb3924c1c3f4555", + "prepared_review_unit_hash": "167b51dcb1d9227b801fb6a7f1dc8dc62d2b15834e879a4fead0216822bb7ea0", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 35671, - "estimated_input_tokens": 8918, - "estimated_eval_tokens": 18092 + "serialized_chars": 35675, + "estimated_input_tokens": 8919, + "estimated_eval_tokens": 18094 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is a loose collection of mostly unrelated Transformers issues. The soft edges are driven by surface-level similarity around v5 regressions, config/save/load problems, and model-specific failures, but the underlying bugs are distinct in almost every pair.", - "confidence": 0.92, - "canonical_issue_reason": "issue:45071 is the best cluster anchor because it is the broadest library-level regression here (core PretrainedConfig/type-checking behavior in v5), whereas most other issues are narrower model- or feature-specific failures.", + "summary": "This cluster is a loose set of Transformer-related issues with some superficial textual similarity, but none of the suggested pairs look like true duplicates. No PRs are present.", + "confidence": 0.66, + "canonical_issue_reason": "issue:46132 is the most central and broad framework-level regression candidate in the set, appearing in multiple soft-similarity links and covering a shared API behavior change.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45071 is the strongest representative issue for the cluster\u2019s apparent theme of v5 regression breakage: it is user-facing, cross-cutting, and more general than the model-specific or one-off API errors in the rest of the set.", + "best_issue_reason": "issue:46132 is the best representative issue because it is the broadest, most reusable bug report among the candidates; the others are narrower model- or feature-specific failures.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44448", "right": "issue:46132", "accept": false, - "reason": "Both mention v4/v5 behavior, but one is Pegasus output drift and the other is AttentionInterface.register semantics; different code paths and fixes." + "reason": "Different bugs: model output drift vs AttentionInterface.register behavior." }, { "left": "issue:45245", "right": "issue:46132", "accept": false, - "reason": "Category-count overflow is a data/encoding limit issue, not an attention-interface behavior change." + "reason": "Unrelated: category cardinality limit vs attention registration semantics." }, { "left": "issue:43479", "right": "issue:43644", "accept": false, - "reason": "Both involve config/init behavior, but one is multimodal config defaults and the other is non-persistent buffer initialization; not the same bug." + "reason": "Different model/config init bug vs non-persistent buffer corruption." }, { "left": "issue:41628", "right": "issue:42757", "accept": false, - "reason": "Two separate import errors from different packages/symbols; only superficially similar as dependency breakage." + "reason": "Both are import errors, but for different missing symbols/dependencies." }, { - "left": "issue:44373", - "right": "issue:44485", + "left": "issue:44297", + "right": "issue:45310", "accept": false, - "reason": "Wrong docstring for position_ids is documentation-only; GLM-5 RoPE is a real implementation issue." + "reason": "Tokenizer save metadata mismatch vs Qwen3.5 from_pretrained failure." }, { - "left": "issue:44297", - "right": "issue:45310", + "left": "issue:44373", + "right": "issue:44485", "accept": false, - "reason": "One is tokenizer save_pretrained metadata mismatch, the other is a from_pretrained loading error; related area but different failures." + "reason": "Docstring issue vs RoPE implementation bug." }, { "left": "issue:46132", "right": "issue:46710", "accept": false, - "reason": "Both are regressions, but one is an attention registration behavior change and the other is an accuracy regression in a specific model after upgrade." + "reason": "API behavior regression vs model accuracy regression; not the same code path." }, { "left": "issue:45071", "right": "issue:46082", "accept": false, - "reason": "Both touch config validation, but one is type-checking behavior and the other is explicit head_dim acceptance rules; not the same defect." + "reason": "Config type-checking regression vs head_dim validation bug." }, { "left": "issue:44222", "right": "issue:44484", "accept": false, - "reason": "FP8 moe save_pretrained corruption and max_shard_size defaults are unrelated save_pretrained concerns." + "reason": "FP8 save_pretrained bug vs unrelated shard-size default question." }, { "left": "issue:43723", "right": "issue:44977", "accept": false, - "reason": "Tokenizer loading failure and flash-attention generation failure affect different subsystems and different models." + "reason": "Tokenizer loading issue vs flash-attention generation bug." }, { "left": "issue:43122", "right": "issue:46209", "accept": false, - "reason": "Tokenization drift across versions and FP8 inference failure on B200 are unrelated issues." + "reason": "Tokenization change across versions vs FP8 B200 inference failure." }, { "left": "issue:44263", "right": "issue:45092", "accept": false, - "reason": "torch.split return values in a DSA indexer are unrelated to remote-code checkpoint incompatibility with meta initialization." + "reason": "Indexer tensor-split bug vs remote-code/meta-init incompatibility." }, { "left": "issue:44190", "right": "issue:46082", "accept": false, - "reason": "A local dataset loading script failure is not the same bug as LlamaConfig head_dim validation." + "reason": "Local dataset loading script issue vs LlamaConfig validation." }, { "left": "issue:43232", "right": "issue:46424", "accept": false, - "reason": "Generation kwargs update after sync_gpus and cache max length handling are different generation/cache behaviors." + "reason": "Generation kwargs sync bug vs cache-length config bug." }, { "left": "issue:43232", "right": "issue:45834", "accept": false, - "reason": "The generation bug and the Kosmos2.5 OCR index error are unrelated." + "reason": "Generation state update bug vs Kosmos OCR index error." }, { "left": "issue:44295", "right": "issue:44623", "accept": false, - "reason": "A position_ids buffer access error is not the same as processor.save_pretrained missing files." + "reason": "Buffer access error vs processor.save_pretrained file omission." }, { "left": "issue:43475", "right": "issue:46305", "accept": false, - "reason": "SAM3 missing an attribute and MoonshineStreaming off-by-one masking are different model-specific bugs." - }, - { - "left": "issue:44263", - "right": "issue:44360", - "accept": false, - "reason": "Both mention the DSA indexer, but one is about split return values and the other about a missing ReLU; different fixes." + "reason": "SAM3 missing attribute vs Moonshine off-by-one masking." }, { "left": "issue:46209", "right": "issue:46431", "accept": false, - "reason": "FP8 inference on B200 and missing chat template are unrelated issues." + "reason": "FP8 inference on B200 vs missing chat template error." }, { "left": "issue:46305", "right": "issue:46655", "accept": false, - "reason": "Both are video-processing bugs, but one is a sliding-window off-by-one and the other is a padding calculation error; not the same defect." + "reason": "Different video/window bugs with unrelated calculations." + }, + { + "left": "issue:44263", + "right": "issue:44360", + "accept": false, + "reason": "torch.split return bug vs missing ReLU in another indexer." }, { "left": "issue:44263", "right": "issue:44373", "accept": false, - "reason": "Indexer split behavior and a wrong docstring for position_ids are unrelated." + "reason": "Runtime indexing bug vs docstring typo." }, { "left": "issue:30990", "right": "issue:38175", "accept": false, - "reason": "Sentence Transformers loading hangs and SigLIP2 zero probabilities are different model-level problems." + "reason": "Unrelated symptoms: loading hang vs zero-probability outputs." }, { "left": "issue:42831", "right": "issue:45310", "accept": false, - "reason": "FP8 accuracy regression and Qwen3.5 from_pretrained failure are distinct issues." + "reason": "FP8 accuracy regression vs Qwen3.5 from_pretrained failure." }, { "left": "issue:45092", "right": "issue:45357", "accept": false, - "reason": "InternVL2 meta-init incompatibility and Qwen3.5 save_pretrained key mismatch involve different models and different failure modes." + "reason": "Checkpoint meta-init incompatibility vs incorrect save_pretrained keys." }, { "left": "issue:46369", "right": "issue:46909", "accept": false, - "reason": "SAM3 memory overflow and a broken DIA documentation example are unrelated." + "reason": "Memory overflow in a model vs broken documentation example." }, { "left": "issue:42907", "right": "issue:46050", "accept": false, - "reason": "Both involve save/load of quantized models, but one is dequantized Ministrals/Devstrals saving and the other is sinq round-trip breakage; different quantization paths." + "reason": "Different serialization issues: dequantized save failure vs quantized round-trip breakage." }, { "left": "issue:44279", "right": "issue:46328", "accept": false, - "reason": "A dependency problem and a NumPy-to-list performance issue are unrelated." + "reason": "Generic dependency issue vs pad performance regression." }, { "left": "issue:45381", "right": "issue:46305", "accept": false, - "reason": "Both concern video position/masking, but they affect different models and code paths; not the same underlying bug." + "reason": "Video position_ids bug vs streaming mask off-by-one." }, { "left": "issue:44261", "right": "issue:45092", "accept": false, - "reason": "MLA q_a_layernorm epsilon/config precision issue is unrelated to InternVL2 meta-initialization incompatibility." + "reason": "Layernorm epsilon bug vs checkpoint/meta-init incompatibility." }, { "left": "issue:44483", "right": "issue:45406", "accept": false, - "reason": "A chat/completions request rejection and a Gemma4Processor missing _tokenizer attribute are different serving failures." + "reason": "Chat-completions request handling vs Gemma4Processor missing tokenizer." } ] }, @@ -15042,8 +14997,8 @@ ], "soft_pairs": [ "issue:22355|issue:27957", - "issue:44829|issue:46002", "issue:43644|issue:46305", + "issue:44829|issue:46002", "issue:42831|issue:44038", "issue:41628|issue:43824", "issue:43408|issue:46402", @@ -15063,23 +15018,23 @@ "issue:43540|issue:46909", "issue:30990|issue:33357", "issue:42907|issue:44945", - "issue:43576|issue:43784", "issue:44295|issue:44297", + "issue:43576|issue:43784", "issue:42371|issue:44263", "issue:39401|issue:44488", "issue:42907|issue:44964", - "issue:46172|issue:46762", "issue:43976|issue:44945", + "issue:46172|issue:46762", "issue:42907|issue:43937" ], - "prepared_review_unit_hash": "64bc82c6a86de145aab6f5b4c345da8a7d8f30c26aeec00acdc0db8852a7f9ed", + "prepared_review_unit_hash": "fe26b91339a52622e2db01298c59f19706406c48024580ef4c683520048083c2", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, @@ -15094,108 +15049,114 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: most items are distinct regressions or feature-specific bugs, with a few superficially similar import/tokenizer issues but no clear duplicate pair. The best umbrella issue is the broad version-upgrade import breakage report.", - "confidence": 0.82, - "canonical_issue_reason": "issue:45278 is the broadest and most representative issue in the set, covering multiple import failures after upgrading transformers, unlike the more narrow or unrelated reports.", + "summary": "The cluster is very heterogeneous: most soft pairs are clearly different bugs, with only a few broad import/tokenizer regressions showing superficial similarity. No pair is strong enough to merge as the same duplicate issue under a strict triage standard.", + "confidence": 0.93, + "canonical_issue_reason": "issue:45278 is the broadest umbrella-style report in the set (multiple import errors after upgrading to 5.5.0), so it is the best candidate for a canonical issue among these mostly unrelated reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the most useful global issue candidate because it describes a broader, user-visible regression pattern that can subsume several import-related symptoms.", + "best_issue_reason": "issue:45278 is the most general and centrally framed issue here, making it the least-specific representative of the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44829", - "right": "issue:46002", + "left": "issue:22355", + "right": "issue:27957", "accept": false, - "reason": "Different problems: flash_attention_3 training collapse vs NaNs after checkout/init." + "reason": "Different failures: missing transformers.onnx module vs XLMRoberta/Flash Attention 2 behavior." }, { "left": "issue:43644", "right": "issue:46305", "accept": false, - "reason": "Unrelated bugs: non-persistent buffer junk values vs a sliding-window off-by-one in audio masking." + "reason": "Unrelated bugs: non-persistent buffer corruption vs MoonshineStreaming window off-by-one." + }, + { + "left": "issue:44829", + "right": "issue:46002", + "accept": false, + "reason": "Different code paths: flash_attention_3 training collapse vs NaNs in classification head init/checkout." }, { "left": "issue:42831", "right": "issue:44038", "accept": false, - "reason": "Different symptoms and code paths: FineGrainedFP8 accuracy issue vs Qwen3-VL-Moe bug." + "reason": "Different problems: FineGrainedFP8 accuracy regression vs Qwen3-VL-Moe model bug." }, { "left": "issue:41628", "right": "issue:43824", "accept": false, - "reason": "Both are import errors, but for different missing symbols and likely different export/regression causes." + "reason": "Both are import errors, but for different missing symbols and likely different fixes." }, { "left": "issue:43408", "right": "issue:46402", "accept": false, - "reason": "Different SAM3-related issues: model-type warning vs missing text encoder weights." + "reason": "Different SAM3 issues: wrong model-type warning vs text encoder weights not loading." }, { "left": "issue:42831", "right": "issue:45753", "accept": false, - "reason": "Accuracy regression in FP8 is not the same as an infinite-loop bug on a specific image." + "reason": "Different model families and symptoms: FP8 accuracy vs Qwen3_5 image loop." }, { "left": "issue:22355", "right": "issue:28218", "accept": false, - "reason": "Module import failure vs tokenizer spacing behavior; no shared bug." + "reason": "Missing module import vs tokenizer spacing after added token are unrelated." }, { "left": "issue:39401", "right": "issue:44568", "accept": false, - "reason": "Both tokenizer bugs, but one is offset_mapping for Qwen3 and the other is BOS/EOS insertion for mdeberta-v3-base." + "reason": "Different tokenizer bugs: wrong offset mapping vs missing BOS/EOS on add_special_tokens." }, { "left": "issue:42886", "right": "issue:43066", "accept": false, - "reason": "Offline cache loading failure is unrelated to a wrong tokenizer decoder type." + "reason": "Offline cache loading failure is not the same as wrong decoder type selection in v5." }, { "left": "issue:42886", "right": "issue:42994", "accept": false, - "reason": "Tokenizer cache/offline loading and quantized model saving are different failure modes." + "reason": "Different areas: offline tokenizer cache loading vs quantized model saving failure." }, { "left": "issue:42757", "right": "issue:45278", "accept": false, - "reason": "44257-like single missing import is not the same concrete bug as the broader multi-import regression; too broad to merge as duplicates." + "reason": "42257 is one specific missing import; 45278 is a broad import-regression report, not the same concrete bug." }, { "left": "issue:27957", "right": "issue:30990", "accept": false, - "reason": "Flash Attention 2 on XLMRoberta and Sentence Transformers loading hang are unrelated." + "reason": "Unrelated: Flash Attention 2 issue vs Sentence-Transformers loading hang." }, { "left": "issue:44188", "right": "issue:44373", "accept": false, - "reason": "Attention-kernel divergence and a wrong docstring are different kinds of issues." + "reason": "Attention kernel divergence is unrelated to a docstring typo." }, { "left": "issue:43408", "right": "issue:46369", "accept": false, - "reason": "SAM3 model-type warning is unrelated to memory overflow in sam3-litetext." + "reason": "Both involve SAM3, but one is a type warning and the other is a memory overflow on a different path." }, { "left": "issue:43232", "right": "issue:45593", "accept": false, - "reason": "Generation kwargs handling after sync_gpus is unrelated to D-FINE auxiliary-loss behavior." + "reason": "Generation kwargs handling is unrelated to D-FINE auxiliary-loss behavior." }, { "left": "issue:42796", "right": "issue:43122", "accept": false, - "reason": "Ministral BPE marker handling differs from a version-to-version tokenization drift report." + "reason": "Similar theme, but different bugs: BPE marker handling vs version-to-version tokenization changes." }, { "left": "issue:43646", @@ -15207,73 +15168,73 @@ "left": "issue:45381", "right": "issue:46909", "accept": false, - "reason": "Qwen2.5-VL video position IDs and a broken documentation example for DIA are unrelated." + "reason": "Different model-specific issues: Qwen2.5-VL video position IDs vs a DIA documentation example." }, { "left": "issue:43540", "right": "issue:46909", "accept": false, - "reason": "Video-input ValueError in Qwen3OmniMoe is unrelated to a documentation example bug." + "reason": "Video-input ValueError in Qwen3OmniMoe is unrelated to a broken docs example." }, { "left": "issue:30990", "right": "issue:33357", "accept": false, - "reason": "Loading hang for Sentence Transformers and a MacOS bus error in CLIP are different failures." + "reason": "Loading hang and MacOS bus error are distinct failures." }, { "left": "issue:42907", "right": "issue:44945", "accept": false, - "reason": "Dequantized Ministral save failure is unrelated to incorrect outputs under pipeline parallelism." + "reason": "Quantized saving failure is unrelated to incorrect pipeline-parallel LLM output." }, { - "left": "issue:43576", - "right": "issue:43784", + "left": "issue:44295", + "right": "issue:44297", "accept": false, - "reason": "Broken env command and a NameError importing sentence-transformers are different issues." + "reason": "Buffer registration/read error is unrelated to tokenizer_config class mismatch." }, { - "left": "issue:44295", - "right": "issue:44297", + "left": "issue:43576", + "right": "issue:43784", "accept": false, - "reason": "Position_ids buffer read error is unrelated to tokenizer_class mismatch on save_pretrained." + "reason": "Broken env CLI and sentence-transformers import NameError are different regressions." }, { "left": "issue:42371", "right": "issue:44263", "accept": false, - "reason": "TF32 settings warning and torch.split return values in GlmMoeDsaIndexer do not share a bug." + "reason": "TF32 settings guidance is unrelated to torch.split return behavior." }, { "left": "issue:39401", "right": "issue:44488", "accept": false, - "reason": "Qwen3 offset_mapping bug and failure to load cjvt/sleng-bert are different tokenizer/model issues." + "reason": "Wrong offset_mapping and failure to load a specific model are different tokenizer/model-loading bugs." }, { "left": "issue:42907", "right": "issue:44964", "accept": false, - "reason": "Quantized model saving failure is unrelated to a multimodal model load failure." + "reason": "Saving dequantized Ministrals/Devstrals is unrelated to loading Phi-4 multimodal instruct." }, { - "left": "issue:46172", - "right": "issue:46762", + "left": "issue:43976", + "right": "issue:44945", "accept": false, - "reason": "DeepSeek-V4 comb transpose bug and Minimax attention numerical drift are not the same code-path problem." + "reason": "Python version incompatibility is unrelated to pipeline-parallel output corruption." }, { - "left": "issue:43976", - "right": "issue:44945", + "left": "issue:46172", + "right": "issue:46762", "accept": false, - "reason": "Python version compatibility issue is unrelated to pipeline-parallelism output corruption." + "reason": "Different numerical issues in different models/code paths." }, { "left": "issue:42907", "right": "issue:43937", "accept": false, - "reason": "Quantized save failure is unrelated to invalid GenerationConfig handling for GLM-5." + "reason": "Quantized saving failure is unrelated to GLM-5 GenerationConfig validation." } ] }, @@ -15371,221 +15332,224 @@ "issue:42757|issue:44351", "issue:36296|issue:43116", "issue:43278|issue:44945", - "issue:22355|issue:29127", "issue:42175|issue:44938", + "issue:22355|issue:29127", "issue:43493|issue:45753", "issue:46231|issue:46877", "issue:44188|issue:46328" ], - "prepared_review_unit_hash": "d8638ad84b6d7b2e8de3643f450eb3f02de8483ca5b73400b546873670c8db49", + "prepared_review_unit_hash": "76f41f6ad899fc48ee9098e477a1886a1a3cbe9c876c147d176c50da82f31162", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36237, - "estimated_input_tokens": 9060, - "estimated_eval_tokens": 18376 + "serialized_chars": 36242, + "estimated_input_tokens": 9061, + "estimated_eval_tokens": 18378 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab bag of unrelated Transformers regressions and import/behavior issues. None of the soft pairs look like the same underlying bug or change, so all are rejected. The broadest representative issue is the general import-regression report.", - "confidence": 0.82, - "canonical_issue_reason": "issue:45278 is the best umbrella-style representative because it describes a broad upgrade-related import regression, which is the closest thing to a cluster-wide theme here.", + "summary": "Most pairs are related only by broad subsystem or symptom similarity and should stay separate. The strongest duplicate-like pair is the Qwen3.5 save/load regression (`save_pretrained` producing a bad checkpoint and `from_pretrained` failing on it).", + "confidence": 0.71, + "canonical_issue_reason": "issue:45216 is the clearest root-cause description for the Qwen3.5 serialization regression, and it plausibly covers the downstream `from_pretrained` failure in issue:45478.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the most general and high-signal issue in the set, but it still only covers one narrow family of failures rather than the whole cluster.", + "best_issue_reason": "issue:45216 is the best representative issue because it states the regression and version boundary more concretely than the related load-time symptom issue.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43756", "right": "issue:44060", "accept": false, - "reason": "Different models and failure modes: RoPE layer counting vs a tied-weights warning in Qwen3-Next." + "reason": "Both are model-specific architecture bugs, but they concern different models and different failure modes (RoPE layer behavior vs tied-weights warning)." }, { "left": "issue:41628", "right": "issue:45278", "accept": false, - "reason": "Both are import problems, but they are different symbols and likely different causes; too broad to merge." + "reason": "Both are import-related, but one is a single missing export and the other is a broad upgrade-wide import breakage; not the same bug." }, { "left": "issue:44117", "right": "issue:44704", "accept": false, - "reason": "One is TOKENIZER_MAPPING_NAMES returning None; the other is AutoProcessor not forwarding kwargs. Different code paths." + "reason": "Tokenizer mapping fallback and `AutoProcessor` kwarg forwarding are unrelated code paths with different observed failures." }, { "left": "issue:43784", "right": "issue:45600", "accept": false, - "reason": "Sentence-transformers import NameError is unrelated to stale auto_mappings entries for Sam3LiteText." + "reason": "Both mention import-time breakage, but they are caused by different missing symbols/config references in different integrations." }, { "left": "issue:39401", "right": "issue:44521", "accept": false, - "reason": "Tokenizer offset_mapping bug vs multimodal assistant_mask generation bug; not the same underlying defect." + "reason": "Both involve tokenization outputs, but one is offset mapping for Qwen3 and the other is assistant mask handling for multimodal chat templates." }, { "left": "issue:44704", "right": "issue:45290", "accept": false, - "reason": "Cached-file kwargs forwarding and chat-template tool-call crashing are separate API paths." + "reason": "Different APIs and failure modes: cached-file kwargs propagation versus chat template crashes on tool-call messages." }, { "left": "issue:44625", "right": "issue:44829", "accept": false, - "reason": "Config propagation for num_labels is unrelated to a flash_attention_3 training degeneration bug." + "reason": "One is config propagation for Qwen3.5 labels; the other is a training-quality regression tied to flash attention 3." }, { "left": "issue:44360", "right": "issue:44485", "accept": false, - "reason": "DSA indexer/ReLU discussion and GLM-5 RoPE implementation are different model/kernel issues." + "reason": "Unrelated model/math implementation issues: DSA indexer activation vs GLM-5 RoPE behavior." }, { "left": "issue:43408", "right": "issue:43475", "accept": false, - "reason": "Both mention SAM3, but one is a model-type warning and the other is a missing output attribute." + "reason": "Both are SAM3-related, but one is a model-type mismatch warning and the other is a missing attribute runtime error." }, { "left": "issue:43122", "right": "issue:44521", "accept": false, - "reason": "Tokenizer version drift is not the same as assistant_masks becoming all zero for multimodal inputs." + "reason": "Both are tokenizer/template regressions, but they affect different features and outputs." }, { "left": "issue:44038", "right": "issue:44297", "accept": false, - "reason": "Qwen3-VL-Moe import/compatibility issue vs tokenizer save_pretrained metadata mismatch; different bugs." + "reason": "One is a Qwen3-VL-Moe compatibility issue; the other is tokenizer save metadata mismatch. Different concrete bugs." }, { "left": "issue:43122", "right": "issue:46326", "accept": false, - "reason": "Different tokenizer regressions affecting different behaviors and APIs." + "reason": "Both are tokenizer-related regressions, but they concern different APIs and symptoms." }, { "left": "issue:42907", "right": "issue:43452", "accept": false, - "reason": "Saving dequantized model weights and gguf_file loading breakage are distinct workflows." + "reason": "Different save/load failures in different model paths; not the same underlying bug." }, { "left": "issue:44038", "right": "issue:44295", "accept": false, - "reason": "Qwen3-VL-Moe regression and position_ids buffer access error do not describe the same bug." + "reason": "General Qwen3-VL-Moe issue vs a position_ids buffer read error; no concrete overlap." }, { "left": "issue:45216", "right": "issue:45478", - "accept": false, - "reason": "Related family and model name, but save_pretrained regression and from_pretrained error are different concrete failures." + "accept": true, + "reason": "Same Qwen3.5 MoE save/load regression around the 5.4.0 change: a bad checkpoint saved by `save_pretrained` can plausibly surface as `from_pretrained` failure." }, { "left": "issue:22355", "right": "issue:29159", "accept": false, - "reason": "Missing transformers.onnx import is unrelated to slow/fast tokenizer inconsistency." + "reason": "A missing `transformers.onnx` module import is unrelated to slow/fast tokenizer inconsistency." }, { "left": "issue:39401", "right": "issue:46614", "accept": false, - "reason": "Qwen3 offset_mapping and stale phonemizer language state are unrelated tokenization features." + "reason": "Different tokenizer subsystems and different bugs: offset mapping vs stale phonemizer language state." }, { "left": "issue:44038", "right": "issue:44222", "accept": false, - "reason": "Both involve Qwen3-VL-Moe, but one is import/model wiring and the other is FP8 save_pretrained behavior." + "reason": "Both mention MoE, but one is a model/version bug and the other is FP8 `save_pretrained` behavior; too broad to merge." }, { "left": "issue:43824", "right": "issue:44351", "accept": false, - "reason": "Different missing imports from different modules and symbols." + "reason": "Both are import errors, but they are for different exported classes and different dependency/version contexts." }, { "left": "issue:43116", "right": "issue:45200", "accept": false, - "reason": "Example-script multi-label output bug and Gemma 4 token-type defaults are separate issues." + "reason": "Different application paths: example classification script output vs Gemma 4 multimodal training defaults." }, { "left": "issue:43122", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer regressions, but for different models and symptoms; not mergeable as one fix." + "reason": "Both are tokenizer regressions, but they are model-specific and describe different incorrect behaviors." }, { "left": "issue:42757", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but one is huggingface_hub and the other is transformers; different broken exports." + "reason": "Different missing imports from different packages; same symptom class only." }, { "left": "issue:36296", "right": "issue:43116", "accept": false, - "reason": "Tensor parallel training and multi-label classification output are unrelated." + "reason": "Tensor-parallel training and classification example behavior are unrelated." }, { "left": "issue:43278", "right": "issue:44945", "accept": false, - "reason": "Dtype mismatch during evaluate is unrelated to incorrect pipeline-parallel outputs." + "reason": "Different distributed/eval bugs with no shared concrete code path." }, { - "left": "issue:22355", - "right": "issue:29127", + "left": "issue:42175", + "right": "issue:44938", "accept": false, - "reason": "Missing module import and LayoutLMv3 error-message clarity are unrelated." + "reason": "Backend-install dependency issue versus Python 3.14 compatibility failure; not the same problem." }, { - "left": "issue:42175", - "right": "issue:44938", + "left": "issue:22355", + "right": "issue:29127", "accept": false, - "reason": "Backend packaging omission and Python 3.14 load failure are different installation/runtime problems." + "reason": "Different issue types and different subsystems: missing ONNX module vs LayoutLMv3 error messaging." }, { "left": "issue:43493", "right": "issue:45753", "accept": false, - "reason": "SigLIP2 implementation discrepancy and Qwen3_5 image loop are unrelated model-specific bugs." + "reason": "Different model families and different failures: SigLIP2 implementation discrepancy vs Qwen3_5 infinite loop on a specific image." }, { "left": "issue:46231", "right": "issue:46877", "accept": false, - "reason": "Documentation parameter-name mismatch and import-time doc-lint stdout output are separate docs tooling issues." + "reason": "Documentation parameter-name mismatch is unrelated to import-time docstring lint output." }, { "left": "issue:44188", "right": "issue:46328", "accept": false, - "reason": "torch.compile attention-kernel branching and a tokenizer phonemizer padding issue are unrelated." + "reason": "Attention-kernel divergence under `torch.compile` is unrelated to feature-extractor padding inefficiency." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The only accepted soft pair, issue:45216 vs issue:45478, is plausibly the same Qwen3.5 save/load regression, and the other verdicts are appropriately cautious by distinguishing different model families, APIs, or failure modes." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -15671,30 +15635,30 @@ "issue:43992|issue:44704", "issue:44075|issue:44936", "issue:43116|issue:43825", - "issue:28348|issue:46640", "issue:43066|issue:46710", - "issue:44263|issue:46762", + "issue:28348|issue:46640", "issue:46614|issue:46752", + "issue:44263|issue:46762", "issue:44995|issue:46697", "issue:45092|issue:46762", "issue:42222|issue:43441", "issue:43827|issue:45446", "issue:33453|issue:39401" ], - "prepared_review_unit_hash": "fac701cfef7f999166e3036fc9d2fe69d6e223945ca70344007483add8b44e17", + "prepared_review_unit_hash": "88c7488e776d16dcda7c5725398641a0508a392162150b45373a894f9f8acd10", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 35637, + "serialized_chars": 35638, "estimated_input_tokens": 8910, "estimated_eval_tokens": 18076 }, @@ -15703,180 +15667,180 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "All soft-edge candidates look like false positives: they share broad themes (tokenizers, Qwen/GLM, FlashAttention, upgrade regressions) but point to different concrete bugs, models, or code paths. No pull requests are present.", - "confidence": 0.88, - "canonical_issue_reason": "issue:45278 is the best umbrella-style anchor: it describes a broad, reproducible import regression after upgrade with multiple failing imports, making it the most representative issue in this set.", + "summary": "This cluster is mostly a loose theme match, not a true duplicate set. The soft pairs generally share broad subsystem language (tokenizers, Qwen, v5 regressions, flash attention) but describe different code paths or different symptoms, so I rejected them all.", + "confidence": 0.89, + "canonical_issue_reason": "issue:45278 is the broadest umbrella-style report in the set: it describes multiple import failures after a version upgrade, making it the most plausible canonical target if any of these are related.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the strongest global issue candidate because it is a clear, high-signal regression report with broad impact and an obvious upgrade trigger, so it is the most useful triage anchor.", + "best_issue_reason": "issue:45278 is the most suitable global issue to anchor review because it is the most general and could subsume specific import-regression reports; the rest are mostly narrower or unrelated symptoms.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43937", "right": "issue:44485", "accept": false, - "reason": "Same GLM-5 family, but one is GenerationConfig validation and the other is RoPE implementation; different failure points and not the same bug." + "reason": "Both are GLM-5 related, but one is a GenerationConfig validation failure and the other is a RoPE implementation discussion; different bugs/code paths." }, { "left": "issue:38617", "right": "issue:45278", "accept": false, - "reason": "Both are import-related, but 38617 is a specific missing symbol while 45278 is a broad upgrade regression with many import errors; not the same underlying issue." + "reason": "Both mention import problems, but one is a specific missing symbol import and the other is a broad multi-import regression report; not the same concrete failure." }, { "left": "issue:42898", "right": "issue:46614", "accept": false, - "reason": "Tokenization-space behavior change and stale phonemizer backend state are unrelated code paths." + "reason": "Tokenizer cleanup behavior and a stale phonemizer backend are unrelated regressions." }, { "left": "issue:43066", "right": "issue:44779", "accept": false, - "reason": "Both involve tokenizer regressions, but one is decoder-type selection and the other is DeepSeek output correctness; different concrete bugs." + "reason": "Both are tokenizer regressions, but one is about decoder type in v5 and the other about DeepSeek output correctness; too different to merge." }, { "left": "issue:43611", "right": "issue:46002", "accept": false, - "reason": "Model loading/base_model_prefix breakage and NaNs in classification heads are distinct problems." + "reason": "Base model prefix loading breakage and NaNs in classification heads are different failures with different symptoms and code paths." }, { "left": "issue:44038", "right": "issue:45478", "accept": false, - "reason": "They involve different Qwen3 variants and different failures: VL-Moe bug vs from_pretrained error for Qwen3.5 MoE." + "reason": "Both involve Qwen/MoE and Transformers v5, but they concern different model variants and different loading errors." }, { "left": "issue:43976", "right": "issue:45446", "accept": false, - "reason": "Python version compatibility regression and a wrong PyTorch version gate for AuxRequest import are not the same issue." + "reason": "A Python version compatibility issue is unrelated to a PyTorch version gate in flex_attention." }, { "left": "issue:44355", "right": "issue:46231", "accept": false, - "reason": "Compiled Python file errors and a docs parameter-name mismatch are unrelated." + "reason": "Compiled Python file errors and a documentation parameter-name mismatch are not the same bug." }, { "left": "issue:43116", "right": "issue:43493", "accept": false, - "reason": "Example script multi-label output bug and SigLIP2 implementation discrepancy are different components and symptoms." + "reason": "Example-script multi-label classification output failure is unrelated to a SigLIP2 implementation discrepancy." }, { "left": "issue:42673", "right": "issue:45072", "accept": false, - "reason": "VRAM leak across dataloader threads and bfloat16 dtype mismatches are separate runtime failures." + "reason": "VRAM leakage in Qwen3ForCausalLM and dtype mismatches in SwitchTransformers/TimmWrapperModel are different runtime issues." }, { "left": "issue:45357", "right": "issue:45478", "accept": false, - "reason": "Both are Qwen3.5 MoE-related, but one is save_pretrained key serialization and the other is from_pretrained loading; not the same concrete code-path problem." + "reason": "Both are Qwen3.5-related, but one is save_pretrained key serialization and the other is from_pretrained loading failure; not the same bug." }, { "left": "issue:44625", "right": "issue:46002", "accept": false, - "reason": "Propagating num_labels into text config and NaNs in classification heads are different defects." + "reason": "num_labels propagation in Qwen3.5 is unrelated to NaNs in classification heads." }, { "left": "issue:42371", "right": "issue:43874", "accept": false, - "reason": "TF32 API guidance/regression and a missing image-processor method are unrelated." + "reason": "TF32 API guidance and a missing image-patch helper in GLM46VImageProcessorFast are unrelated." }, { "left": "issue:45083", "right": "issue:45446", "accept": false, - "reason": "Helper-length computation behavior and a PyTorch version check bug are not the same underlying change." + "reason": "A feature-extraction length helper bug and a flex_attention import/version check are unrelated." }, { "left": "issue:44610", "right": "issue:46909", "accept": false, - "reason": "Processor over-padding and a broken documentation example are different kinds of issues." + "reason": "OmDet-Turbo preprocessing dimension mismatch and a broken DIA documentation example are not the same issue." }, { "left": "issue:43618", "right": "issue:46829", "accept": false, - "reason": "Both are output-field problems, but they affect different model outputs and different fields; not mergeable as one bug." + "reason": "Both involve output objects, but they affect different models and different misassigned fields." }, { "left": "issue:43992", "right": "issue:44704", "accept": false, - "reason": "Missing embed_tokens.weight and incomplete kwargs forwarding to cached_file are separate loader issues." + "reason": "A missing weight in UMT5Encoder loading is unrelated to AutoProcessor not forwarding kwargs to cached_file." }, { "left": "issue:44075", "right": "issue:44936", "accept": false, - "reason": "SGD argument handling and trainer.evaluate failing after train are unrelated." + "reason": "SGD arguments being ignored and trainer.evaluate() failing after train() are different trainer/optimizer bugs." }, { "left": "issue:43116", "right": "issue:43825", "accept": false, - "reason": "Classification example behavior and pipeline translation error messaging are different subsystems." - }, - { - "left": "issue:28348", - "right": "issue:46640", - "accept": false, - "reason": "Both mention FlashAttention 2, but one is GPT2LMHeadModel support and the other is a generic flash-attention issue; different model/code paths." + "reason": "An example classification script bug is unrelated to a pipeline() error-message regression." }, { "left": "issue:43066", "right": "issue:46710", "accept": false, - "reason": "Tokenizer decoder-type regression and DeepSeek accuracy regression after upgrade are not the same bug." + "reason": "Tokenizer decoder-type regression and a DeepSeek accuracy regression may both be version-related, but they are not the same underlying defect." }, { - "left": "issue:44263", - "right": "issue:46762", + "left": "issue:28348", + "right": "issue:46640", "accept": false, - "reason": "A torch.split return-value bug in GlmMoeDsaIndexer is different from Minimax MSA attention numerical differences." + "reason": "Flash Attention 2 support for GPT2LMHeadModel is a feature request; the other is a separate flash-attention issue." }, { "left": "issue:46614", "right": "issue:46752", "accept": false, - "reason": "Stale phonemizer backend state and an empty-list crash in apply_chat_template are unrelated." + "reason": "A stale phonemizer language/backend bug is unrelated to an empty-list apply_chat_template crash." + }, + { + "left": "issue:44263", + "right": "issue:46762", + "accept": false, + "reason": "torch.split return-value handling in one GLM MoE component and numerical differences in Minimax MSA attention are different bugs." }, { "left": "issue:44995", "right": "issue:46697", "accept": false, - "reason": "Second-forward stale cache crash and image over-padding are different code paths." + "reason": "A stale indexer cache crash and over-padding in an image processor are unrelated." }, { "left": "issue:45092", "right": "issue:46762", "accept": false, - "reason": "Old InternVL2 meta-init incompatibility and Minimax attention numerical discrepancies are unrelated." + "reason": "Meta-initialization incompatibility for old remote-code checkpoints is unrelated to Minimax attention numerical drift." }, { "left": "issue:42222", "right": "issue:43441", "accept": false, - "reason": "VitPose model breakage and Ministral-3 FlashAttention failure are different model-specific regressions." + "reason": "VitPose model breakage and a FlashAttention failure in Ministral-3 are different models and different failure modes." }, { "left": "issue:43827", "right": "issue:45446", "accept": false, - "reason": "Documentation still referencing pipeline() and a PyTorch version check for flex_attention are not the same issue." + "reason": "Documentation still mentioning pipeline() and a PyTorch import/version check in flex_attention are unrelated." }, { "left": "issue:33453", "right": "issue:39401", "accept": false, - "reason": "Generic tokenizer loading regression and Qwen3 offset_mapping are both tokenizer-related but describe different failures." + "reason": "A general tokenizer loading regression and a Qwen3 offset_mapping bug are not the same concrete bug." } ] }, @@ -15953,19 +15917,19 @@ "issue:42673|issue:42796", "issue:43937|issue:45563", "issue:44380|issue:46077", - "issue:43673|issue:46328", "issue:45356|issue:46614", + "issue:43673|issue:46328", "issue:42673|issue:46655", "issue:42222|issue:46752", "issue:44291|issue:45071", "issue:45405|issue:45446", "issue:44117|issue:45588", "issue:44466|issue:46002", - "issue:44410|issue:45310", "issue:39401|issue:44779", + "issue:44410|issue:45310", "issue:43937|issue:44977", - "issue:45341|issue:46328", "issue:43450|issue:43986", + "issue:45341|issue:46328", "issue:45341|issue:45561", "issue:44829|issue:45216", "issue:45071|issue:46512", @@ -15979,208 +15943,208 @@ "issue:44484|issue:45310", "issue:44877|issue:46231" ], - "prepared_review_unit_hash": "50670bdb5f784d2654cf074465b15fd3ab392faeb023185430ab437c1a922e2f", + "prepared_review_unit_hash": "2fdae60072f1b997bc60f611e52a57431edbf2319fd3169ae08e73822ce047bf", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 35864, - "estimated_input_tokens": 8966, - "estimated_eval_tokens": 18188 + "serialized_chars": 35866, + "estimated_input_tokens": 8967, + "estimated_eval_tokens": 18190 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Loose cluster of mostly unrelated transformers issues that only share broad theme similarity (v5 regressions, tokenization, generation, model-loading, and training). None of the soft pairs look like the same underlying bug/change, so they should stay separate.", - "confidence": 0.82, - "canonical_issue_reason": "If forced to pick one issue as the umbrella representative, issue #43122 is the broadest and most central: it describes a cross-version tokenizer regression with a clear repro, and it is closer to the other tokenizer-focused reports than the model-specific or utility-specific bugs. That said, this is still a noisy mixed cluster, not a true duplicate set.", + "summary": "This cluster is heterogeneous; the soft pairs mostly share only broad subsystem similarity (tokenizers, config/loading, video processors, testing utils) rather than the same underlying bug. I would not merge any of the proposed soft edges.", + "confidence": 0.92, + "canonical_issue_reason": "Issue 43122 is the most general and representative version-regression report in the set (tokenization behavior changed from 4.57.3 to 5.0). That said, the cluster is too mixed to treat as one true duplicate group.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #43122 is the best single issue to retain because it is the most general high-signal regression in the set and the most plausible umbrella for nearby tokenizer complaints; the rest are narrower, unrelated, or touch different code paths.", + "best_issue_reason": "43122 is the best single issue to anchor this cluster because it is broad, clearly regression-oriented, and closest to the recurring tokenization/version-change theme seen in several nearby issues.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43064", "right": "issue:43673", "accept": false, - "reason": "Different failures: optimizer-state mismatch under FSDP2+PEFT+cpu_ram_efficient_loading vs missing GenerationMixin cache during chunked prefill." + "reason": "Training optimizer-state corruption and generation cache missing are unrelated code paths and bugs." }, { "left": "issue:42673", "right": "issue:42796", "accept": false, - "reason": "One is a VRAM leak in multi-threaded dataloader use; the other is tokenizer BPE-marker handling. Same broad area, different bug." + "reason": "VRAM leak in Qwen3ForCausalLM is unrelated to a tokenizer BPE-marker bug." }, { "left": "issue:43937", "right": "issue:45563", "accept": false, - "reason": "GenerationConfig validation error vs stale warning text for num_return_sequences in paged generate; not the same defect." + "reason": "Invalid GenerationConfig and a stale paging warning are different failure modes." }, { "left": "issue:44380", "right": "issue:46077", "accept": false, - "reason": "Different attention bugs: GPT2 attention scaling ignored under SDPA/FlashAttention vs Gemma4 building causal masks when bidirectional attention is requested." + "reason": "Both touch attention behavior, but they are model-specific and not the same concrete bug." }, { - "left": "issue:43673", - "right": "issue:46328", + "left": "issue:45356", + "right": "issue:46614", "accept": false, - "reason": "Generation cache missing in chunked prefill vs SequenceFeatureExtractor.pad inefficiency; unrelated code paths." + "reason": "Tokenizer codec-handling regression and stale phonemizer backend state are different subsystems and code paths." }, { - "left": "issue:45356", - "right": "issue:46614", + "left": "issue:43673", + "right": "issue:46328", "accept": false, - "reason": "Tokenizer codec/BPE regression vs stale phonemizer backend after language switch; different tokenizer subsystems and symptoms." + "reason": "Generation cache handling and SequenceFeatureExtractor.pad performance are unrelated." }, { "left": "issue:42673", "right": "issue:46655", "accept": false, - "reason": "VRAM leak in Qwen3ForCausalLM threads vs video processor padding calculation error; not the same bug." + "reason": "VRAM leak in model threading and video padding miscalculation are not the same bug." }, { "left": "issue:42222", "right": "issue:46752", "accept": false, - "reason": "VitPose model breakage vs apply_chat_template crashing on empty list; unrelated failures." + "reason": "Vitpose model breakage and empty-list chat templating crash are unrelated." }, { "left": "issue:44291", "right": "issue:45071", "accept": false, - "reason": "Both are v5 regressions, but one is an unexpected _is_hf_initialized argument during init_empty_weights and the other is PretrainedConfig type checking. Different issues." + "reason": "Both involve v5 API changes, but unexpected init args and PretrainedConfig type checking are different defects." }, { "left": "issue:45405", "right": "issue:45446", "accept": false, - "reason": "PEFT minimum-version bump metadata problem vs incorrect torch version guard for AuxRequest import; unrelated." + "reason": "A dependency minimum bump and a PyTorch version guard bug are unrelated." }, { "left": "issue:44117", "right": "issue:45588", "accept": false, - "reason": "TOKENIZER_MAPPING_NAMES None handling vs flash_attention crash on s_aux=None; different layers of the stack." + "reason": "Tokenizer mapping None handling and flash_attention sink-less crash are unrelated." }, { "left": "issue:44466", "right": "issue:46002", "accept": false, - "reason": "Serialization of lm_head.weight depending on device vs NaNs in classification heads after checkout/init; different failure modes." + "reason": "State dict serialization of tied weights and NaNs in classification heads are different problems." }, { - "left": "issue:44410", - "right": "issue:45310", + "left": "issue:39401", + "right": "issue:44779", "accept": false, - "reason": "Missing qwen3next projections and Qwen3.5 from_pretrained errors are both model-family issues, but not the same concrete code-path bug." + "reason": "Both are tokenizer regressions, but they affect different models and appear to be separate bugs." }, { - "left": "issue:39401", - "right": "issue:44779", + "left": "issue:44410", + "right": "issue:45310", "accept": false, - "reason": "Both concern tokenizer regressions, but one is wrong offset_mapping and the other is incorrect DeepSeek tokenization results. Similar theme, not the same bug." + "reason": "Missing layer projections and a from_pretrained loading error are not the same concrete issue." }, { "left": "issue:43937", "right": "issue:44977", "accept": false, - "reason": "GLM-5 invalid GenerationConfig and Qwen3.5 flash-attention generation problems are distinct model-specific issues." + "reason": "GenerationConfig validation and flash-attention generation failure are different bugs." }, { - "left": "issue:45341", - "right": "issue:46328", + "left": "issue:43450", + "right": "issue:43986", "accept": false, - "reason": "A tiny testing_utils bug vs a padding performance issue; no shared underlying defect." + "reason": "Incorrect batched video processor shapes and missing-torchvision crash are separate issues." }, { - "left": "issue:43450", - "right": "issue:43986", + "left": "issue:45341", + "right": "issue:46328", "accept": false, - "reason": "Video processor shape bug vs confusing crash loading a video model without torchvision; different root causes." + "reason": "A testing_utils bug and a SequenceFeatureExtractor.pad inefficiency are unrelated." }, { "left": "issue:45341", "right": "issue:45561", "accept": false, - "reason": "Testing_utils bug vs pytest-xdist race on captured_info.txt; different concurrency and file-handling problems." + "reason": "Generic testing_utils bug and xdist race on captured_info.txt are not the same underlying failure." }, { "left": "issue:44829", "right": "issue:45216", "accept": false, - "reason": "Degenerate training with flash_attention_3 vs Qwen3.5 save_pretrained checkpoint regression; different code paths." + "reason": "Flash-attention training degeneration and Qwen3.5 checkpoint save regression are different code paths." }, { "left": "issue:45071", "right": "issue:46512", "accept": false, - "reason": "PreTrainedConfig type-checking breakage vs stale model.config.dtype after from_config(dtype=...); related only by version timing." + "reason": "Type-checking breakage and stale config.dtype after from_config are related only loosely, not the same bug." }, { "left": "issue:44117", "right": "issue:45081", "accept": false, - "reason": "Tokenizer mapping fallback bug vs _patch_mistral_regex backend_tokenizer AttributeError; different tokenizer plumbing bugs." + "reason": "Missing tokenizer mapping entries and backend_tokenizer AttributeError in regex patching are distinct failures." }, { "left": "issue:43408", "right": "issue:43742", "accept": false, - "reason": "Model-type warning mismatch vs key error loading facebook/MobileLLM-125M; not the same defect." + "reason": "Model-type warning and MobileLLM load key error are unrelated." }, { "left": "issue:38617", "right": "issue:41628", "accept": false, - "reason": "Both are import errors, but for different missing exports and different call sites; not duplicates." + "reason": "These are different import errors involving different symbols and modules." }, { "left": "issue:43122", "right": "issue:46298", "accept": false, - "reason": "Tokenization regression vs noisy logging message; no shared underlying bug." + "reason": "Tokenization regression and noisy logging are unrelated." }, { "left": "issue:42617", "right": "issue:45341", "accept": false, - "reason": "Script execution failure vs a small testing_utils bug; unrelated." + "reason": "A script execution failure and a testing_utils bug are unrelated." }, { "left": "issue:45362", "right": "issue:46328", "accept": false, - "reason": "Qwen3.5 chat crash vs SequenceFeatureExtractor.pad inefficiency; different components." + "reason": "Qwen3.5 chat crash and a SequenceFeatureExtractor.pad inefficiency are unrelated." }, { "left": "issue:42222", "right": "issue:46178", "accept": false, - "reason": "VitPose model breakage vs backend mapping rejecting tensorflow_text; unrelated." + "reason": "Vitpose model breakage and unsupported backend mapping error are unrelated." }, { "left": "issue:44484", "right": "issue:45310", "accept": false, - "reason": "Question about shard size default vs Qwen3.5 MoE from_pretrained error; not the same bug." + "reason": "save_pretrained shard-size question and Qwen3.5 loading error are unrelated." }, { "left": "issue:44877", "right": "issue:46231", "accept": false, - "reason": "Strict config loading failure vs a documentation parameter-name typo; not duplicates." + "reason": "Strict config loading and a documentation parameter-name mismatch are not the same bug." } ] }, @@ -16266,8 +16230,8 @@ "issue:43066|issue:43122", "issue:44263|issue:44485", "issue:43097|issue:44493", - "issue:43673|issue:45341", "issue:43931|issue:44410", + "issue:43673|issue:45341", "issue:41628|issue:45071", "issue:42222|issue:45542", "issue:45071|issue:45542", @@ -16277,237 +16241,237 @@ "issue:44263|issue:46172", "issue:44360|issue:46172", "issue:27957|issue:43352", - "issue:43931|issue:44038", "issue:43012|issue:43577", + "issue:43931|issue:44038", "issue:43441|issue:44829", "issue:43097|issue:43452", "issue:45003|issue:45341", "issue:43986|issue:45341", "issue:44261|issue:44485" ], - "prepared_review_unit_hash": "c81712ba752605d86db3ab09da09478366df739be8919be1f3d5e016bf2128d4", + "prepared_review_unit_hash": "f960eb6e45a71d1ecc15691f3494b70846ab10a123c34bd2bed8fa10538254f2", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 31, - "serialized_chars": 35867, - "estimated_input_tokens": 8967, - "estimated_eval_tokens": 18190 + "serialized_chars": 35864, + "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 mostly a grab bag of unrelated Transformers regression reports, with a few loose v5/tokenizer/import similarities. The broadest issue is the generic upgrade/import-regression report; only the tokenizer-v5 pair looks like a plausible duplicate.", - "confidence": 0.42, - "canonical_issue_reason": "issue:45278 is the broadest, most umbrella-style report in the set: it describes widespread import failures after upgrading to v5.5.0, which best matches the cluster\u2019s overall regression theme better than the many model-specific bugs.", + "summary": "This cluster is mostly a set of unrelated Transformers regressions and model-specific bugs. The only plausible duplicate pair is issue:44483 and issue:45464, which both point to a Qwen3.5 chat/completions streaming failure, but the packet only supports a likely duplicate relationship rather than a confirmed one.", + "confidence": 0.79, + "canonical_issue_reason": "issue:45464 is the best canonical candidate for the likely Qwen3.5 chat/completions streaming bug because it is the more detailed report. The duplicate relationship to issue:44483 looks plausible from the symptom match, but the packet does not provide explicit confirmation.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the best single representative issue because it is the most general and most likely to subsume other compatibility/import regressions in this cluster.", + "best_issue_reason": "issue:45464 is the strongest issue to anchor this cluster because it contains the most complete problem description and appears to describe the same underlying streaming chat/completions failure as issue:44483. That said, the evidence is only soft similarity, not hard confirmation.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42371", "right": "issue:44261", "accept": false, - "reason": "Different problems: TF32 configuration behavior vs an MLA layernorm epsilon precision bug." + "reason": "Both mention precision-related behavior, but one is about TF32 API control and the other about an MLA layernorm config/epsilon mismatch. Different bugs." }, { "left": "issue:43937", "right": "issue:44496", "accept": false, - "reason": "One is a GenerationConfig validation error; the other is an unrecognized model/config loading failure." + "reason": "GenerationConfig validation and unrecognized model/config loading are separate failure modes, not the same root cause." }, { "left": "issue:39401", "right": "issue:46209", "accept": false, - "reason": "Tokenizer offset-mapping bug and FP8 inference failure are unrelated symptoms and code paths." + "reason": "Tokenizer offset mapping and FP8 inference on B200 are unrelated symptoms and code paths." }, { "left": "issue:44410", "right": "issue:45478", "accept": false, - "reason": "Both mention Qwen3-family models, but one is missing projections in a layer while the other is a from_pretrained loading error." + "reason": "Both involve Qwen family models, but one is missing projections in qwen3next while the other is a Qwen3.5-MoE from_pretrained error. Different defects." }, { "left": "issue:42371", "right": "issue:46617", "accept": false, - "reason": "TF32 settings are unrelated to cache-loading FileNotFoundError behavior." + "reason": "TF32 API behavior and trust_remote_code cache loading are unrelated." }, { "left": "issue:44261", "right": "issue:46762", "accept": false, - "reason": "Layernorm epsilon precision mismatch is not the same underlying issue as Minimax attention numerical differences." + "reason": "Numerical precision concerns are too broad here; these are different model implementations and different bugs." }, { "left": "issue:43208", "right": "issue:46894", "accept": false, - "reason": "Both are training bugs, but they involve different models and different failure mechanisms." + "reason": "Both discuss training issues, but they affect different models and different failure mechanisms." }, { "left": "issue:44933", "right": "issue:45278", "accept": false, - "reason": "Both are import-related, but one is a specific missing export and the other is a broad upgrade regression report; not the same bug." + "reason": "The latter is a broad report of many import regressions after upgrade; the former is one specific import failure. Not enough evidence they are the same underlying bug." }, { "left": "issue:38617", "right": "issue:44220", "accept": false, - "reason": "An ImportError for a missing symbol is not the same as the fbank feature extraction issue." + "reason": "ImportError for layer_type_validation is unrelated to the _torch_extract_fbank_features issue." }, { "left": "issue:44075", "right": "issue:45656", "accept": false, - "reason": "SGD argument handling and double optimizer stepping under deepspeed are different optimizer bugs." + "reason": "Both are optimizer-related, but one says SGD args are ignored while the other says the optimizer step is called twice under deepspeed." }, { "left": "issue:43066", "right": "issue:43122", - "accept": true, - "reason": "Both point to the same v5 tokenizer regression: changed decoder/tokenization behavior after the 4.57.x -> 5.0 transition." + "accept": false, + "reason": "Both are tokenizer/version-regression reports, but one is about decoder type and the other about output tokenization differences. Too loose to merge." }, { "left": "issue:44263", "right": "issue:44485", "accept": false, - "reason": "One is a torch.split/indexing bug in a MoE indexer; the other is a RoPE implementation issue." + "reason": "A torch.split return-value issue and a RoPE implementation discussion in GLM-5 are different concrete bugs." }, { "left": "issue:43097", "right": "issue:44493", "accept": false, - "reason": "Removing tie_embeddings_and_encoder_decoder and unexpected position-id keys are different configuration/model-loading regressions." + "reason": "Removed tie_embeddings_and_encoder_decoder and unexpected position-id keys do not point to the same root cause." }, { - "left": "issue:43673", - "right": "issue:45341", + "left": "issue:43931", + "right": "issue:44410", "accept": false, - "reason": "Chunked-prefill GenerationMixin cache behavior and a testing_utils bug do not share the same failure mode." + "reason": "Different Qwen3-VL loading/model-shape problems; model family overlap is not enough." }, { - "left": "issue:43931", - "right": "issue:44410", + "left": "issue:43673", + "right": "issue:45341", "accept": false, - "reason": "Different Qwen3-related model-loading failures: weight-shape mismatch vs missing attention projections." + "reason": "Chunked prefill cache behavior and a testing_utils bug are unrelated." }, { "left": "issue:41628", "right": "issue:45071", "accept": false, - "reason": "Missing AutoImageProcessor import and PreTrainedConfig type-checking breakage are unrelated API regressions." + "reason": "AutoImageProcessor import failure and PreTrainedConfig type-checking breakage are separate regressions." }, { "left": "issue:42222", "right": "issue:45542", "accept": false, - "reason": "Vitpose model breakage and tensorboard/TensorFlow backend detection are unrelated." + "reason": "Vitpose model breakage and TensorFlow/tensorboard backend detection are different issues." }, { "left": "issue:45071", "right": "issue:45542", "accept": false, - "reason": "These are different subsystems: config type checking versus backend initialization." + "reason": "Type checking regression is unrelated to undefined tf backend errors." }, { "left": "issue:44483", "right": "issue:45464", - "accept": false, - "reason": "Both involve chat/completions, but one is request acceptance in v5.3 and the other is streaming inference on a specific model." + "accept": true, + "reason": "These appear to describe the same Qwen3.5 chat/completions streaming failure; 45464 is the more detailed report. The match is based on symptom similarity, not explicit confirmation in the packet." }, { "left": "issue:44484", "right": "issue:45478", "accept": false, - "reason": "Saving shard size defaults and Qwen3.5 from_pretrained loading errors are unrelated." + "reason": "max_shard_size behavior and a Qwen3.5-MoE from_pretrained error are unrelated." }, { "left": "issue:43874", "right": "issue:44261", "accept": false, - "reason": "Missing image-processor method and MLA layernorm epsilon precision are different bugs." + "reason": "Missing image patch-count method and MLA q_a_layernorm epsilon/config behavior are different model bugs." }, { "left": "issue:44263", "right": "issue:46172", "accept": false, - "reason": "Torch split return handling in GlmMoeDsaIndexer is unrelated to deepseek-v4 transpose behavior." + "reason": "torch.split output handling in GLM-MoE-DSA is not the same as the deepseek-v4 dtype/transpose issue." }, { "left": "issue:44360", "right": "issue:46172", "accept": false, - "reason": "A missing ReLU in the DSA indexer and a dtype/transpose issue are not the same bug." + "reason": "A missing ReLU in the DSA indexer and a DeepSeek-V4 tensor conversion/transpose issue are distinct." }, { "left": "issue:27957", "right": "issue:43352", "accept": false, - "reason": "Both mention Flash Attention 2, but one is a model support gap and the other is a model-specific incompatibility." + "reason": "Flash Attention support for XLM-Roberta and NemotronHForCausalLM are different model support gaps." }, { - "left": "issue:43931", - "right": "issue:44038", + "left": "issue:43012", + "right": "issue:43577", "accept": false, - "reason": "Both are Qwen3-VL-family reports, but the concrete failures differ: shape mismatch versus a separate model bug." + "reason": "PyTorch warning around bfloat16 compilation and BLIP2 dtype propagation are related only at a very high level." }, { - "left": "issue:43012", - "right": "issue:43577", + "left": "issue:43931", + "right": "issue:44038", "accept": false, - "reason": "A compile-time precision warning and a load-time dtype mismatch are related only loosely, not the same code-path bug." + "reason": "Both are Qwen3-VL-related, but one is a weight-shape mismatch and the other is a Qwen3-VL-Moe loading problem. Different concrete failures." }, { "left": "issue:43441", "right": "issue:44829", "accept": false, - "reason": "FlashAttention 2/3 issues in different models and different symptoms are not the same fix." + "reason": "FlashAttention failures for Ministral-3 and degenerate training with flash_attention_3 are not the same code-path problem." }, { "left": "issue:43097", "right": "issue:43452", "accept": false, - "reason": "A removed config option and a gguf_file from_pretrained regression are separate issues." + "reason": "Removed tie_embeddings/encoder-decoder behavior and gguf_file handling for from_pretrained are unrelated." }, { "left": "issue:45003", "right": "issue:45341", "accept": false, - "reason": "Unsafe sys.modules access and a testing_utils bug are unrelated." + "reason": "Unsafe sys.modules access in modeling_utils and a testing_utils bug do not match." }, { "left": "issue:43986", "right": "issue:45341", "accept": false, - "reason": "Video AutoProcessor crashing without torchvision and a testing_utils bug do not share the same underlying cause." + "reason": "AutoProcessor crash without torchvision and a testing_utils bug are unrelated." }, { "left": "issue:44261", "right": "issue:44485", "accept": false, - "reason": "Layernorm epsilon precision and GLM-5 RoPE implementation are different numerical issues, not one bug." + "reason": "Precision/epsilon concerns are too generic; these are different GLM-related bugs with different symptoms." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The evaluation is grounded in the packet and stays conservative. Using issue:45278 as the umbrella representative for the broad import-regression cluster is reasonable, and the only accepted soft pair (43066/43122) is well supported by the v5 tokenizer regression theme. The rejected pairs are mostly defensible as distinct bugs or at least not clearly mergeable into a single fix." + "accept": false, + "feedback": "The cluster summary is mostly cautious, but it overstates the evidence for issue:44483 \u2194 issue:45464. The packet only shows soft similarity and no explicit link or other hard confirmation, so calling it the \"only confirmed duplicate pair\" and saying 45464 is the clearer duplicate is stronger than the evidence supports. Please downgrade that language to a tentative/likely duplicate and avoid asserting confirmation or reproducibility from the packet alone." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -16581,9 +16545,9 @@ "issue:43531|issue:45341", "issue:43673|issue:46123", "issue:43097|issue:45588", - "issue:38617|issue:42757", - "issue:44038|issue:44410", "issue:44410|issue:45375", + "issue:44038|issue:44410", + "issue:38617|issue:42757", "issue:43208|issue:46897", "issue:43064|issue:44315", "issue:43986|issue:45372", @@ -16603,14 +16567,14 @@ "issue:44704|issue:45103", "issue:44380|issue:46421" ], - "prepared_review_unit_hash": "50e9ffb6950379da37eaf38b4682bcc445ee31f3fe550c56264a748e5ca1982e", + "prepared_review_unit_hash": "3513f0a0be88f3861cc68588d9e167366059f4f4c5ee1f794c484e0acaa06a97", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, @@ -16625,180 +16589,180 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "All soft pairs look like false positives: they share broad subsystems or similar error wording, but not the same concrete bug or change. No pull requests are present in the set.", - "confidence": 0.81, - "canonical_issue_reason": "Issue 43873 is the strongest anchor: it is still open, has the most discussion/inbound references, and describes a substantive user-impacting bug with a clear reproduction surface.", + "summary": "The soft pairs are mostly superficial overlaps (same model family, same API surface, or both are import/load errors) rather than the same underlying bug. I would not merge any of the candidate pairs. Among the listed issues, issue 43097 is the strongest canonical anchor because it is a clearly scoped v5 regression with broad impact and inbound references.", + "confidence": 0.84, + "canonical_issue_reason": "Issue 43097 is the best anchor: it describes a concrete, broad v5 regression with a clear user-visible API break and multiple inbound references.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43873 is the best overall issue candidate because it is detailed, high-signal, and central enough to attract follow-up reports.", + "best_issue_reason": "Issue 43097 is the strongest standalone issue in this set because it is specific, actionable, and likely to attract duplicates around the same breaking change.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45753", "right": "issue:46369", "accept": false, - "reason": "Different failures and models: infinite looping on a specific image vs memory overflow in sam3-litetext." + "reason": "Same general area of model inference, but one is an infinite loop in Qwen3_5 on a specific image and the other is a SAM memory overflow; different models and failure modes." }, { "left": "issue:44117", "right": "issue:44987", "accept": false, - "reason": "One is a generic tokenizer mapping assumption bug; the other is a specific model-loading failure. Similar loading context, but not the same defect." + "reason": "Both involve from_pretrained/loading, but one is a tokenizer mapping None bug and the other is a specific model load failure; no clear shared underlying defect." }, { "left": "issue:43874", "right": "issue:44263", "accept": false, - "reason": "Multimodal token counting AttributeError and a torch.split return-value issue are separate code paths." + "reason": "Both are GLM-related, but one is a missing image-patch helper in the image processor and the other is a torch.split indexing issue; different code paths." }, { "left": "issue:45440", "right": "issue:46172", "accept": false, - "reason": "Both are DeepSeek-related, but one is native-vs-remote divergence and the other is a specific dtype/transpose bug." + "reason": "Different DeepSeek model families and different bugs: remote-vs-native implementation divergence versus a dtype/transpose issue in deepseek-v4." }, { "left": "issue:43531", "right": "issue:45341", "accept": false, - "reason": "Qwen3-MoE sliding-window behavior and a testing utility bug are unrelated." + "reason": "Unrelated: a Qwen3-MoE sliding_window bug versus a testing_utils bug." }, { "left": "issue:43673", "right": "issue:46123", "accept": false, - "reason": "Cache initialization during chunked prefill and a final-batch pipeline flag bug are different behaviors." + "reason": "Both are runtime behavior issues, but one is GenerationMixin cache/chunked_prefill and the other is MaskGenerationPipeline batch bookkeeping; not the same bug." }, { "left": "issue:43097", "right": "issue:45588", "accept": false, - "reason": "A removed config option and a flash-attention crash on sink-less models are distinct issues." + "reason": "Different subsystems and symptoms: a removed config option versus a flash-attention crash on s_aux=None." }, { - "left": "issue:38617", - "right": "issue:42757", + "left": "issue:44410", + "right": "issue:45375", "accept": false, - "reason": "Both are import errors, but they involve different missing symbols from different modules and likely different dependency breaks." + "reason": "Both are Qwen3-family issues, but one is missing attention projections in a layer and the other is a missing config field silently dropped by strict deserialization." }, { "left": "issue:44038", "right": "issue:44410", "accept": false, - "reason": "Same broad model family, but one is a load-time bug and the other is missing projections in layer 0." + "reason": "Both concern Qwen3 variants, but one is a version-specific loading bug and the other is a missing projections bug in a different architecture." }, { - "left": "issue:44410", - "right": "issue:45375", + "left": "issue:38617", + "right": "issue:42757", "accept": false, - "reason": "Missing attention projections and a missing config field are separate defects in different parts of the model stack." + "reason": "Both are import errors, but they involve different modules and different missing symbols; too little evidence of a single underlying change." }, { "left": "issue:43208", "right": "issue:46897", "accept": false, - "reason": "xLSTM training blockers and a Florence2 loss-shift bug do not describe the same underlying problem." + "reason": "Both mention training bugs, but they are different models and different failure patterns; no shared code-path bug is evident." }, { "left": "issue:43064", "right": "issue:44315", "accept": false, - "reason": "Optimizer state corruption under FSDP2/PEFT is unrelated to Liger Kernel not being applied with model_init." + "reason": "Both touch training setup, but one is a wrong optimizer-state bug with FSDP2/PEFT/cpu_ram_efficient_loading and the other is Liger Kernel not being applied with model_init." }, { "left": "issue:43986", "right": "issue:45372", "accept": false, - "reason": "Different dependency failures: missing torchvision vs missing ReasoningEffort from mistral_common." + "reason": "Both are processor-loading failures, but caused by different missing dependencies/imports and different processor stacks." }, { "left": "issue:43097", "right": "issue:46050", "accept": false, - "reason": "One concerns a removed embedding-tie option; the other is save/load corruption for quantized models." + "reason": "A removed API/config option and a quantization save/load corruption bug are unrelated." }, { "left": "issue:43986", "right": "issue:44484", "accept": false, - "reason": "Processor import crash and max_shard_size semantics are unrelated." + "reason": "Processor loading without torchvision and max_shard_size defaults in save_pretrained are unrelated." }, { "left": "issue:42890", "right": "issue:46298", "accept": false, - "reason": "A flaky test due to missing set_seed and a noisy logging message are not the same bug." + "reason": "A flaky test-seeding issue and a logging-message issue are not the same underlying bug." }, { "left": "issue:46655", "right": "issue:46664", "accept": false, - "reason": "Both hit SmolVLM video processing, but one is padding calculation and the other is num_frames semantics; related area, different bug." + "reason": "Same SmolVLM video processor area, but one is padding calculation and the other is num_frames semantics; distinct defects." }, { "left": "issue:46172", "right": "issue:46710", "accept": false, - "reason": "A specific tensor transformation bug in DeepSeek-v4 is not the same as a broader accuracy regression after upgrade." + "reason": "Different model families and different symptoms: a low-level tensor operation bug versus an accuracy regression after upgrade." }, { "left": "issue:38617", "right": "issue:44351", "accept": false, - "reason": "Both are import-name failures, but they refer to different symbols and modules, so they are separate breakages." + "reason": "Both are import-name errors from transformers, but they involve different symbols and different exported APIs." }, { "left": "issue:46298", "right": "issue:46752", "accept": false, - "reason": "Logging noise and an empty-list IndexError in apply_chat_template are unrelated." + "reason": "A noisy logging message and an empty-list IndexError in apply_chat_template are unrelated." }, { "left": "issue:44062", "right": "issue:44117", "accept": false, - "reason": "AddedToken keyword duplication and TOKENIZER_MAPPING_NAMES returning None are different tokenizer issues." + "reason": "One is a tokenizers AddedToken constructor bug; the other is a tokenizer mapping/loading assumption bug." }, { "left": "issue:43097", "right": "issue:43873", "accept": false, - "reason": "A removed config field and a quantization/offloading malfunction do not share the same code-path problem." + "reason": "Removed config/API behavior and quantization offloading behavior are different issues." }, { "left": "issue:43012", "right": "issue:46821", "accept": false, - "reason": "A bfloat16 warning during compilation and a window_size AttributeError are unrelated." + "reason": "A PyTorch precision warning during bfloat16 compilation and a TimesFM window_size AttributeError are unrelated." }, { "left": "issue:44117", "right": "issue:46701", "accept": false, - "reason": "Tokenizer-mapping assumptions and a DiffusionGemma pipeline failure are different bugs." + "reason": "Different areas and failure modes: tokenizer mapping/loading versus an image-text-to-text pipeline failure." }, { "left": "issue:43502", "right": "issue:43992", "accept": false, - "reason": "local_files_only network leakage and a missing embed_tokens weight are separate from-pretrained problems." + "reason": "One is unintended API requests despite local_files_only=True; the other is a missing embed_tokens weight during UMT5Encoder loading." }, { "left": "issue:44936", "right": "issue:46365", "accept": false, - "reason": "trainer.evaluate() after train() and importing Trainer conflicting with datasets are unrelated." + "reason": "A train-then-evaluate lifecycle bug and an import conflict preventing Trainer import are different problems." }, { "left": "issue:44704", "right": "issue:45103", "accept": false, - "reason": "cached_file kwarg forwarding and auto_docstring crashing under future annotations are different subsystems." + "reason": "AutoProcessor kwargs forwarding and auto_docstring annotation handling are unrelated code paths." }, { "left": "issue:44380", "right": "issue:46421", "accept": false, - "reason": "Attention scaling being ignored and lazy cache initialization causing recompilation are distinct issues." + "reason": "GPT2 attention scaling with SDPA/FlashAttention and chunked-prefill recompilation from cache lazy init are different bugs." } ] }, @@ -16904,20 +16868,20 @@ "issue:44038|issue:44230", "issue:43756|issue:44485" ], - "prepared_review_unit_hash": "2d423037086e011d074020447f48145678ad2da55ddae6092b940622724de9f0", + "prepared_review_unit_hash": "c8f4d1cc0cf4d399fbd95fe99ab3cc0c7aa92bd38022277317f07fd3e32f7413", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36744, + "serialized_chars": 36743, "estimated_input_tokens": 9186, "estimated_eval_tokens": 18628 }, @@ -16926,96 +16890,96 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: the soft pairs mostly share only broad Transformers surface areas, not the same concrete bug. I rejected every soft edge and picked a single nominal issue representative from the most actionable load-path regression.", - "confidence": 0.42, - "canonical_issue_reason": "Issue 43502 is the most concrete, user-facing regression in the set, with a narrow loading/offline code path and a clear reproduction. The cluster itself is not a true duplicate group, so this is only a nominal representative.", + "summary": "This cluster is highly heterogeneous: most pairs share only broad surface keywords (imports, save/load, warnings, attention, backend) but not the same underlying bug. I would not merge any of the listed soft edges as duplicates.", + "confidence": 0.89, + "canonical_issue_reason": "issue:44380 is the closest thing to a representative issue because it is a concrete backend/config bug with the most plausible related edges, but the cluster is too mixed for it to serve as a true duplicate root.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43502 is the strongest standalone report here: clear symptom, specific API path, and a plausible localized fix surface. It is the best fit if one issue must represent the set.", + "best_issue_reason": "issue:44380 is the best available issue representative, but only for a narrow attention-backend subtheme; it is not a strong global canonical for this cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44938", "right": "issue:46178", "accept": false, - "reason": "Both are import/loading-related, but one is a Python 3.14 load failure and the other is an invalid backend mapping error for tensorflow_text; different bugs." + "reason": "Different failures: Python 3.14 import/load compatibility vs an invalid BACKENDS_MAPPING backend name." }, { "left": "issue:43986", "right": "issue:44117", "accept": false, - "reason": "Different failure modes: missing torchvision handling in AutoProcessor versus TOKENIZER_MAPPING_NAMES returning None in from_pretrained." + "reason": "Both are import-related, but one is a missing torchvision dependency path and the other is a tokenizer mapping assumption bug." }, { "left": "issue:42907", "right": "issue:44655", "accept": false, - "reason": "One is about failing to save dequantized Ministral/Devstral models; the other is pipeline object save_pretrained behavior. Not the same bug." + "reason": "Both mention save/pretrained behavior, but one is a dequantized-model save failure and the other is Pipeline object serialization." }, { "left": "issue:42907", "right": "issue:45753", "accept": false, - "reason": "Save/dequantization failure versus an infinite loop on a specific image in Qwen3_5; unrelated symptoms and code paths." + "reason": "Completely different problems: model saving for dequantized Ministrals/Devstrals vs an inference infinite-loop bug on Qwen3_5." }, { "left": "issue:43825", "right": "issue:44655", "accept": false, - "reason": "Pipeline translation-task warning and pipeline save_pretrained are separate issues; same subsystem only at a very broad level." + "reason": "One is an error-message/task-support issue in pipeline(), the other is failure to save Pipeline objects; different code paths." }, { "left": "issue:43012", "right": "issue:43550", "accept": false, - "reason": "One is a compile-time precision warning for bfloat16 models; the other is a Bamba SDPA torch.compile failure. Different concrete bugs." + "reason": "A compiler warning about bfloat16 precision is not the same as a torch.compile SDPA failure on Bamba." }, { "left": "issue:45712", "right": "issue:46298", "accept": false, - "reason": "Repo hygiene around dummy classes leaking without torch is unrelated to a noisy logging message." + "reason": "Repo cleanup/dummy-class leak vs noisy logging; unrelated maintenance vs logging behavior." }, { "left": "issue:43208", "right": "issue:43550", "accept": false, - "reason": "xLSTM training bugs and Bamba torch.compile/SDPA failure are distinct model-specific problems." + "reason": "xLSTM training bugs and Bamba torch.compile SDPA failure are different model-specific problems." }, { "left": "issue:43012", "right": "issue:44368", "accept": false, - "reason": "A PyTorch precision warning on compile is not the same as a tie_word_embeddings warning during Qwen3.5 LoRA fine-tuning." + "reason": "Both are warning-style issues, but one is PyTorch precision logging and the other is a tie_word_embeddings config warning." }, { "left": "issue:43992", "right": "issue:44484", "accept": false, - "reason": "Missing embed_tokens.weight in UMT5Encoder.from_pretrained is unrelated to the default max_shard_size question." + "reason": "Missing embed_tokens weight during load is unrelated to a question about max_shard_size in save_pretrained()." }, { "left": "issue:43352", "right": "issue:43976", "accept": false, - "reason": "Flash Attention support in NemotronH and Python-version compatibility for Transformers 5.1.0 are unrelated." + "reason": "Flash Attention support on NemotronH vs Python-version compatibility for Transformers 5.1.0 are unrelated." }, { "left": "issue:43352", "right": "issue:44380", "accept": false, - "reason": "Nemotron Flash Attention support and GPT2 attention scaling being ignored are different model/backend bugs." + "reason": "Both involve attention, but one is unsupported Flash Attention on a specific model and the other is GPT2 attention scaling being ignored; different bugs." }, { "left": "issue:43066", "right": "issue:43906", "accept": false, - "reason": "Wrong tokenizer decoder type in v5 is not the same as an isolated reproduction of another issue." + "reason": "Tokenizer decoder type regression and an isolated reproduction of another issue are not enough to establish the same concrete bug." }, { "left": "issue:44162", "right": "issue:46172", "accept": false, - "reason": "ESM2 breakage and a deepseek-v4 dtype/transpose bug are different model-specific failures." + "reason": "ESM2 model breakage and a DeepSeek tensor operation bug are unrelated model-specific failures." }, { "left": "issue:43097", @@ -17027,97 +16991,97 @@ "left": "issue:44380", "right": "issue:46132", "accept": false, - "reason": "GPT2 SDPA/FlashAttention scaling is unrelated to AttentionInterface.register changing the registered function behavior." + "reason": "Both touch attention behavior, but one is GPT2 attention scaling being ignored and the other is AttentionInterface.register changing callback behavior; not the same code-path bug." }, { "left": "issue:45584", "right": "issue:46365", "accept": false, - "reason": "Whisper empty-transcription generation failure and an import conflict with datasets are unrelated." + "reason": "Whisper empty-transcription generation failure and inability to import Trainer due to datasets are unrelated." }, { "left": "issue:44077", "right": "issue:46365", "accept": false, - "reason": "patchtsmixer post_init validation and Trainer import issues caused by datasets are not the same bug." + "reason": "A patchtsmixer post_init validation issue has nothing to do with a Trainer import conflict caused by datasets." }, { "left": "issue:44222", "right": "issue:45478", "accept": false, - "reason": "FP8 save_pretrained problems for moe models are distinct from a Qwen3.5 MoE from_pretrained error." + "reason": "Both are MoE/FP8 flavored, but one is save_pretrained for FP8 and the other is from_pretrained failure on Qwen3.5 MoE; different operations and symptoms." }, { "left": "issue:36683", "right": "issue:43582", "accept": false, - "reason": "Gemma3Config missing vocab_size and an Apple Silicon caching_allocator_warmup TypeError are unrelated." + "reason": "Gemma3Config missing vocab_size and an Apple Silicon caching_allocator TypeError are unrelated." }, { "left": "issue:44380", "right": "issue:45479", "accept": false, - "reason": "GPT2 attention scaling under SDPA/FlashAttention is a different bug from zero-loss classification with num_labels=1." + "reason": "Attention scaling being ignored is unrelated to the degenerate zero-loss classification bug." }, { "left": "issue:42175", "right": "issue:46365", "accept": false, - "reason": "TensorFlow not being included in the torch extra and Trainer import failing with datasets are distinct dependency issues." + "reason": "TensorFlow backend installation mix-up and a Trainer import problem involving datasets are unrelated." }, { "left": "issue:42757", "right": "issue:43824", "accept": false, - "reason": "Missing is_offline_mode import and missing Qwen2_5_VLForConditionalGeneration export are different import regressions." + "reason": "Both are import errors, but they involve different missing symbols from different modules and different root causes." }, { "left": "issue:43064", "right": "issue:44380", "accept": false, - "reason": "FSDP2/PEFT optimizer state corruption and GPT2 attention scaling are unrelated training/backend issues." + "reason": "FSDP2 optimizer-state corruption and GPT2 attention scaling behavior are unrelated code paths." }, { "left": "issue:42490", "right": "issue:43012", "accept": false, - "reason": "Trailing-slash local model loading warning is unrelated to the bfloat16 compile precision warning." + "reason": "Local-path trailing-slash warning and bfloat16 precision warning are unrelated." }, { "left": "issue:43208", "right": "issue:43526", "accept": false, - "reason": "xLSTM training bugs and BeitImageProcessorFast reduce_labels returning one label are different code paths." + "reason": "xLSTM training blockers and BeitImageProcessorFast reduce_labels behavior are unrelated." }, { "left": "issue:44483", "right": "issue:44514", "accept": false, - "reason": "A v5 chat/completions request acceptance problem is not the same as a batched apply_chat_template crash." + "reason": "An OpenAI-compatible chat endpoint rejection and a batched chat-template crash are different API paths." }, { "left": "issue:43618", "right": "issue:44380", "accept": false, - "reason": "CLIPOutput attentions no longer assigned and GPT2 attention scaling ignored are separate model behavior bugs." + "reason": "CLIPOutput attentions assignment and GPT2 attention scaling are both attention-related, but they concern different components and failures." }, { "left": "issue:38617", "right": "issue:43502", "accept": false, - "reason": "Importing layer_type_validation and making API requests despite local_files_only=True are both load-path symptoms, but they are clearly different defects." + "reason": "Missing layer_type_validation import and local_files_only still making API requests are unrelated import vs network behavior bugs." }, { "left": "issue:44038", "right": "issue:44230", "accept": false, - "reason": "Qwen3-VL-Moe breakage and fp8 support for Qwen3-VL/Qwen3.5 MoE inference are different feature/compatibility issues." + "reason": "Both are Qwen3-family/MoE-related, but one is a model compatibility bug and the other is fp8 inference support; not the same concrete issue." }, { "left": "issue:43756", "right": "issue:44485", "accept": false, - "reason": "Smollm3 RoPE layer mismatch and GLM-5 RoPE implementation discussion are not the same concrete bug." + "reason": "Smollm3 RoPE-layer mismatch and GLM-5 RoPE implementation discussion are different model-specific RoPE problems." } ] }, @@ -17178,6 +17142,7 @@ "issue:45216", "issue:45276", "issue:45278", + "issue:45290", "issue:45310", "issue:45375", "issue:45561", @@ -17185,7 +17150,6 @@ "issue:46082", "issue:46202", "issue:46365", - "issue:46369", "issue:46829", "issue:46851", "issue:46877" @@ -17199,8 +17163,8 @@ "issue:45276|issue:46077", "issue:44355|issue:46877", "issue:43352|issue:43937", - "issue:44483|issue:46369", "issue:36683|issue:43475", + "issue:44483|issue:45290", "issue:43611|issue:44655", "issue:43901|issue:44655", "issue:43867|issue:45216", @@ -17226,238 +17190,238 @@ "issue:43867|issue:45375", "issue:42890|issue:45561" ], - "prepared_review_unit_hash": "3f6f3809204bc985243ec8b22433072cc9bcf6afed97c77f091751b0ee3b7e5e", + "prepared_review_unit_hash": "ad5b501bfcf3964dc6a8c1a2b24a2b1259313d102225097f7ca59a1b3ab50123", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 34, - "serialized_chars": 37584, - "estimated_input_tokens": 9396, - "estimated_eval_tokens": 19048 + "serialized_chars": 37630, + "estimated_input_tokens": 9408, + "estimated_eval_tokens": 19072 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly a set of unrelated regression/compatibility issues across different models, loaders, and utilities. None of the soft-edge pairs look like true duplicates under a strict same-bug standard, though a few are superficially adjacent (especially CLIP and Gemma/Qwen model-family regressions).", - "confidence": 0.83, - "canonical_issue_reason": "issue:43618 is the clearest, most concrete regression report in the set: it names a specific model/output field and points to a focused code path, making it a strong canonical issue anchor.", + "summary": "This cluster is broad and heterogeneous: import/load regressions, config bugs, training issues, docs/test fixes, and model-specific failures. None of the soft pairs look like the same underlying bug or change, so no duplicates were accepted.", + "confidence": 0.95, + "canonical_issue_reason": "No canonical issue: the items span unrelated bugs across different models, APIs, and failure modes, and none of the soft pairs are true duplicates.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43618 is the best standalone issue candidate because it is narrowly scoped, actionable, and has a clear symptom-to-code-path mapping compared with the broader or more ambiguous reports.", + "best_issue_reason": "No single best issue: the cluster is too heterogeneous to nominate one representative duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42994", "right": "issue:44655", "accept": false, - "reason": "Both mention saving, but one is quantized model saving and the other is pipeline save_pretrained; different objects and likely different fixes." + "reason": "Both mention saving, but one is quantized model saving while the other is Pipeline.save_pretrained; different object types and failure modes." }, { "left": "issue:43618", "right": "issue:43761", "accept": false, - "reason": "Same CLIP family, but one is missing attentions and the other is missing hidden_states; related regression theme, not clearly the same bug." + "reason": "Both are CLIP regressions, but one is missing attentions in the output object and the other is hidden_states being None; different bugs." }, { "left": "issue:43299", "right": "issue:43867", "accept": false, - "reason": "Both concern model loading regressions, but they affect different Qwen/serialization scenarios and do not point to the same code path." + "reason": "Both concern loading, but one is a Qwen3VL model-loading regression and the other is a state_dict sorting load error; not the same root cause." }, { "left": "issue:36010", "right": "issue:42757", "accept": false, - "reason": "Different import failures from different dependencies; no shared underlying defect is evident." + "reason": "Both are import errors, but they involve different missing symbols from different packages and unrelated breakages." }, { "left": "issue:44625", "right": "issue:46365", "accept": false, - "reason": "A Qwen config propagation bug versus a Trainer import problem; unrelated subsystems and failure modes." + "reason": "One is about num_labels propagation in Qwen3.5, the other is importing Trainer being blocked by datasets; unrelated issues." }, { "left": "issue:45276", "right": "issue:46077", "accept": false, - "reason": "Both are Gemma4 issues, but resizing embeddings and bidirectional attention masks are separate behaviors and code paths." + "reason": "Both touch Gemma4, but one is embedding resize propagation and the other is attention-mask construction; different code paths." }, { "left": "issue:44355", "right": "issue:46877", "accept": false, - "reason": "Compiled-file import errors versus import-time docstring lint output; unrelated." + "reason": "Both are import-time problems, but one is compiled Python file handling and the other is auto_docstring printing diagnostics; distinct bugs." }, { "left": "issue:43352", "right": "issue:43937", "accept": false, - "reason": "Different model-specific generation/attention failures; not the same concrete bug." + "reason": "Different models and different failures: unsupported Flash Attention 2 vs invalid GenerationConfig." }, { - "left": "issue:44483", - "right": "issue:46369", + "left": "issue:36683", + "right": "issue:43475", "accept": false, - "reason": "API request validation versus memory overflow in a model forward path; clearly different problems." + "reason": "Both are AttributeErrors, but they hit different model configs/outputs and different missing attributes." }, { - "left": "issue:36683", - "right": "issue:43475", + "left": "issue:44483", + "right": "issue:45290", "accept": false, - "reason": "Different missing-attribute errors on different model families (Gemma3 vs SAM 3 Video)." + "reason": "One is a chat/completions request rejection and the other is apply_chat_template crashing on tool-call messages; different layers and causes." }, { "left": "issue:43611", "right": "issue:44655", "accept": false, - "reason": "Base_model_prefix loading regression and pipeline save_pretrained failure are separate serialization issues." + "reason": "Loading with base_model_prefix and saving Pipeline objects are unrelated failures." }, { "left": "issue:43901", "right": "issue:44655", "accept": false, - "reason": "Docs mismatch for TextClassificationPipeline versus pipeline persistence; not the same bug." + "reason": "Docs mentioning return_all_scores is a documentation mismatch, while save_pretrained on Pipeline objects is a runtime bug." }, { "left": "issue:43867", "right": "issue:45216", "accept": false, - "reason": "One is loading failure with sorted state_dict, the other is incorrect checkpoint saving; related area but different directions and likely different fixes." + "reason": "Both involve model persistence, but one is load failure with sorted state_dict and the other is an incorrect saved Qwen3.5 checkpoint." }, { "left": "issue:43976", "right": "issue:44938", "accept": false, - "reason": "Both are Python-version compatibility issues, but for different interpreter versions and different failure modes." + "reason": "Both are Python-version compatibility reports, but they affect different versions and likely different incompatibilities." }, { "left": "issue:36683", "right": "issue:44077", "accept": false, - "reason": "Different model/config attribute bug versus optimizer argument handling; unrelated." + "reason": "Missing vocab_size on Gemma3Config and ignored SGD args are completely unrelated." }, { "left": "issue:44625", "right": "issue:46077", "accept": false, - "reason": "Qwen config propagation and Gemma4 attention-mask construction are unrelated." + "reason": "One is Qwen3.5 num_labels propagation, the other is Gemma4 bidirectional attention masking; unrelated." }, { "left": "issue:46077", "right": "issue:46082", "accept": false, - "reason": "Gemma4 attention masking versus LlamaConfig head_dim validation; different code paths." + "reason": "Different configs and different bugs: Gemma4 attention-mask logic vs LlamaConfig head_dim validation." }, { "left": "issue:43618", "right": "issue:45216", "accept": false, - "reason": "CLIP output regression versus Qwen3.5 save_pretrained regression; different models and bug classes." + "reason": "CLIP attentions regression and Qwen3.5 save_pretrained regression are unrelated." }, { "left": "issue:43208", "right": "issue:46829", "accept": false, - "reason": "xLSTM training blockers versus VideoPrism output formatting; no shared bug." + "reason": "xLSTM training bugs and VideoPrism output field mix-up are unrelated model-specific issues." }, { "left": "issue:36683", "right": "issue:43525", "accept": false, - "reason": "Missing config fields in different model families; not the same defect." + "reason": "Both are missing config attributes, but they are different models and different attributes, so not the same bug." }, { "left": "issue:36683", "right": "issue:44625", "accept": false, - "reason": "Both are config-related, but they affect different models and different missing fields." + "reason": "Different missing/propagated config fields in different model families; not a duplicate." }, { "left": "issue:29870", "right": "issue:36296", "accept": false, - "reason": "Bart generate-time UnboundLocalError versus tensor-parallel training bug; unrelated." + "reason": "Bart generate() decoder_attention_mask error and tensor-parallel training bug are unrelated." }, { "left": "issue:43992", "right": "issue:45310", "accept": false, - "reason": "Both involve Qwen3.5 from_pretrained, but one is a missing encoder weight and the other is a broader model loading error; not clearly the same issue." + "reason": "Both are from_pretrained-style failures, but they affect different models and different missing/loading mechanics." }, { "left": "issue:43064", "right": "issue:44492", "accept": false, - "reason": "Optimizer-state corruption in distributed training versus a typo in cache strategies; unrelated." + "reason": "Trainer/FSDP2 optimizer-state corruption and a typo in cache strategies are unrelated." }, { "left": "issue:43010", "right": "issue:45278", "accept": false, - "reason": "no_grad decoration for cache updates versus broad import errors after upgrade; different fixes and code paths." + "reason": "Cache update no_grad behavior and broad import errors after an update are different problem classes." }, { "left": "issue:44075", "right": "issue:46365", "accept": false, - "reason": "Optimizer argument handling versus Trainer import failure; unrelated." + "reason": "Optimizer arguments not being used and Trainer import failing due to datasets are unrelated." }, { "left": "issue:44908", "right": "issue:46851", "accept": false, - "reason": "Scheduler kwargs bug versus variable-length position_ids handling; different components." + "reason": "Scheduler kwargs being ignored and Qwen3.5 variable-length position_ids are unrelated bugs." }, { "left": "issue:29870", "right": "issue:33453", "accept": false, - "reason": "Bart generation mask bug versus tokenizer loading regression; different subsystems." + "reason": "Bart generate() decoder_attention_mask error and tokenizer loading regression do not share the same failure mode." }, { "left": "issue:44938", "right": "issue:45071", "accept": false, - "reason": "Python 3.14 load failure versus PretrainedConfig type-checking regression; different issues." + "reason": "Both are version regressions, but one is Python 3.14 load failure and the other is PretrainedConfig type checking; not the same bug." }, { "left": "issue:44355", "right": "issue:44855", "accept": false, - "reason": "Compiled Python import errors versus a Python 3.13 @torch.jit.script parsing problem; unrelated." + "reason": "Compiled-file import errors and a Python 3.13 IndentationError from torch.jit.script parsing are distinct causes." }, { "left": "issue:43867", "right": "issue:46202", "accept": false, - "reason": "State_dict loading failure versus Gemma pipeline roundtrip on a specific token; different failure modes." + "reason": "Model load failure with sorted state_dict and Gemma pipeline roundtrip failure are unrelated." }, { "left": "issue:36683", "right": "issue:43064", "accept": false, - "reason": "A config attribute bug versus distributed optimizer state corruption; unrelated." + "reason": "Gemma3Config missing vocab_size and FSDP2 optimizer state corruption are unrelated." }, { "left": "issue:43867", "right": "issue:45375", "accept": false, - "reason": "Different Qwen/Gemma serialization issues with different missing fields and paths." + "reason": "A load-model error and a missing deepstack_visual_indexes config field are different bugs." }, { "left": "issue:42890", "right": "issue:45561", "accept": false, - "reason": "Missing set_seed in tests versus xdist race on a captured info file; test stability issues, but not the same bug." + "reason": "One is missing set_seed in integration tests, the other is a pytest-xdist race on captured_info.txt; different test infra issues." } ] }, @@ -17561,236 +17525,242 @@ "issue:42548|issue:44351", "issue:45125|issue:45375", "issue:46664|issue:46726", + "issue:43502|issue:44222", "issue:43010|issue:43232", - "issue:43502|issue:44961", - "issue:43502|issue:44222" + "issue:43502|issue:44961" ], - "prepared_review_unit_hash": "c4565b6fd063ed56f5fbfd45c72718c85d2e246c36114b58ac4eb80e260733ef", + "prepared_review_unit_hash": "7a589fb6ec8a07858e484a10a956a55ad5a3f5421568dc14cb3f2e8d01f92348", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37181, - "estimated_input_tokens": 9296, - "estimated_eval_tokens": 18848 + "serialized_chars": 37185, + "estimated_input_tokens": 9297, + "estimated_eval_tokens": 18850 }, "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: import errors, tokenizer/model loading regressions, tensor-parallel problems, and model-specific bugs. The only clear soft duplicate is the import-time slowdown pair around `import transformers` / `import *` in models.", - "confidence": 0.83, - "canonical_issue_reason": "Issue 41669 is the clearest, most actionable root-cause report: it identifies a specific import-time regression from `import *` in models and matches the only near-duplicate slow-import report.", + "summary": "This cluster is mostly heterogeneous and title-similar rather than true duplicates. Nearly all soft edges are false positives across unrelated bugs; the only plausible duplicate pair is the Qwen3.5 MoE / from_pretrained tensor-parallel regression.", + "confidence": 0.63, + "canonical_issue_reason": "issue:45125 is the most specific root-cause report in the set: it identifies the concrete missing _tp_plan problem in the Qwen3.5 MoE path, which cleanly anchors the only near-duplicate-looking pair here.", "canonical_pr_reason": null, - "best_issue_reason": "Best overall representative because it is concrete, scoped to one code path, and describes a fixable performance regression rather than a vague symptom or unrelated model-specific bug.", + "best_issue_reason": "issue:45125 is the best representative issue because it describes an actionable, narrowly scoped code-path failure rather than a broad symptom or unrelated regression.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43010", "right": "issue:45850", "accept": false, - "reason": "Different problems: `@torch.no_grad` on cache/layer updates vs a broad complaint that v5 breaks many models." + "reason": "Broad transformer regressions and both mention v5-era behavior, but one is about cache update no_grad decoration and the other is a generic complaint about many broken models." }, { "left": "issue:41669", "right": "issue:44246", - "accept": true, - "reason": "Both describe the same underlying issue: slow `transformers` import. 41669 gives the specific root cause (`import *` in models), while 44246 is the generic symptom report." + "accept": false, + "reason": "Both touch import performance, but 41669 is a specific import-* cleanup regression while 44246 is a general intermittent slow-import report with no shared code path shown." }, { "left": "issue:36296", "right": "issue:43575", "accept": false, - "reason": "Both mention tensor parallelism, but one is a training bug and the other is an OOM while loading a specific model; too different to treat as the same bug." + "reason": "Tensor-parallel is the only overlap; one is a training bug and the other is an OOM on loading a very large model, which are different failure modes and fixes." }, { "left": "issue:43010", "right": "issue:43976", "accept": false, - "reason": "Unrelated: generation/cache mutation behavior vs Python version compatibility." + "reason": "These are unrelated: cache/layer no_grad decoration versus a Python-version compatibility issue." }, { "left": "issue:43618", "right": "issue:44077", "accept": false, - "reason": "Different model issues with different symptoms and code paths." + "reason": "Different models and different bugs: CLIP attentions assignment versus patchtsmixer post_init policy." }, { "left": "issue:43992", "right": "issue:45478", "accept": false, - "reason": "Both are `from_pretrained` failures, but they involve different models and different failure modes." + "reason": "Both involve loading, but the reported model families and failure causes differ; missing embed_tokens.weight is not the same as a Qwen3.5 MoE from_pretrained error." }, { "left": "issue:43867", "right": "issue:45357", "accept": false, - "reason": "Different regression classes: state-dict loading order issue vs incorrect visual encoder keys during save." + "reason": "One is a sorted state_dict load problem, the other is incorrect save_pretrained key mapping for a different model family." }, { "left": "issue:28282", "right": "issue:33453", "accept": false, - "reason": "Completely different issues: missing PyTorch dependency vs tokenizer loading regression." + "reason": "ImportError for missing PyTorch is unrelated to a tokenizer-loading regression." }, { "left": "issue:43618", "right": "issue:46285", "accept": false, - "reason": "`CLIPOutput.attentions` assignment bug is unrelated to the CLIP text model structural inconsistency." + "reason": "These are different CLIP-related issues with distinct symptoms and code paths." }, { "left": "issue:44360", "right": "issue:45440", "accept": false, - "reason": "Different architectures and failures: DSA indexer activation issue vs DeepSeekV3 implementation divergence." + "reason": "Completely different areas: a DSA indexer discussion versus a DeepSeekV3 implementation divergence." }, { "left": "issue:42890", "right": "issue:44964", "accept": false, - "reason": "Test flakiness due to missing `set_seed()` is not the same as a model load failure for Phi-4 multimodal." + "reason": "Test seeding flakiness in SamHQ is unrelated to loading Phi-4 multimodal." }, { "left": "issue:42890", "right": "issue:44112", "accept": false, - "reason": "Both involve tests/CI, but the failures are different and not the same underlying bug." + "reason": "Both are CI/test failures, but they affect different tests and root causes; no shared bug is evident." }, { "left": "issue:42890", "right": "issue:45600", "accept": false, - "reason": "Test seed instability and broken auto-mappings reference are unrelated." + "reason": "SamHQ test seeding and auto_mappings stale config references are unrelated maintenance issues." }, { "left": "issue:42898", "right": "issue:46664", "accept": false, - "reason": "Tokenizer cleanup behavior change and video processor `num_frames` semantics are unrelated." + "reason": "Tokenization cleanup behavior and SmolVLM frame-count semantics are separate behavior changes in different components." }, { "left": "issue:43502", "right": "issue:44062", "accept": false, - "reason": "`local_files_only=True` still making requests is unrelated to `AddedToken(... special=...)` argument duplication." + "reason": "local_files_only network leakage is unrelated to the AddedToken special-argument TypeError." }, { "left": "issue:44734", "right": "issue:46617", "accept": false, - "reason": "KV-cache continuation crash in serve is unrelated to cache-loading `FileNotFoundError` for trust-remote-code models." + "reason": "One is a KV-cache continuation indexing crash in serving; the other is a cache FileNotFoundError for trust_remote_code models." }, { "left": "issue:43723", "right": "issue:44556", "accept": false, - "reason": "Related only at a very high level; one is tokenizer loading and the other is checkpoint reload after version upgrade, not clearly the same bug." + "reason": "Both are loading regressions, but one is tokenizer loading and the other is checkpoint reload across versions; the concrete failure paths differ." }, { "left": "issue:44222", "right": "issue:45125", "accept": false, - "reason": "FP8 save-pretrained behavior and missing `_tp_plan` for tensor parallelism are different issues." + "reason": "FP8 save_pretrained for MoE and missing _tp_plan for tensor parallelism are different bugs in different APIs." }, { "left": "issue:44387", "right": "issue:44556", "accept": false, - "reason": "Int4 memory regression/OOM and checkpoint reload compatibility are different failure modes." + "reason": "Int4 memory growth/OOM is unrelated to checkpoint reload compatibility." }, { "left": "issue:43824", "right": "issue:46617", "accept": false, - "reason": "Missing exported model class vs cache `FileNotFoundError` are unrelated." + "reason": "Missing exported Qwen2_5_VL class import is unrelated to a trust_remote_code cache FileNotFoundError." }, { "left": "issue:43502", "right": "issue:44485", "accept": false, - "reason": "`local_files_only` network calls and GLM-5 RoPE implementation are unrelated." + "reason": "local_files_only request leakage and GLM-5 RoPE implementation are not the same issue." }, { "left": "issue:43874", "right": "issue:46617", "accept": false, - "reason": "Missing image-processor method causing multimodal token counting failure is unrelated to cache file lookup failure." + "reason": "A missing image-processor method in GLM46V is unrelated to remote-code cache loading." }, { "left": "issue:43576", "right": "issue:46132", "accept": false, - "reason": "Broken CLI `transformers env` and `AttentionInterface.register` behavior change are different issues." + "reason": "Broken env CLI output and AttentionInterface.register behavior are distinct subsystems and fixes." + }, + { + "left": "issue:43010", + "right": "issue:43065", + "accept": false, + "reason": "Cache/layer update decoration and a dummy Conv2d in Sam3PixelDecoder are unrelated model implementation issues." }, { "left": "issue:45125", "right": "issue:45478", - "accept": false, - "reason": "Both concern Qwen3.5 MoE, but one is tensor-parallel metadata and the other is a broader `from_pretrained` error; not enough to merge as one bug." + "accept": true, + "reason": "These look like the same Qwen3.5 MoE tensor-parallel/from_pretrained regression: one states the concrete missing _tp_plan root cause, the other reports the resulting from_pretrained failure." }, { "left": "issue:30333", "right": "issue:33453", "accept": false, - "reason": "MLflow job-reporting failure and tokenizer loading regression are unrelated." + "reason": "MLflow job reporting and tokenizer loading regression have no shared code path." }, { "left": "issue:44355", "right": "issue:44749", "accept": false, - "reason": "Compiled Python-file errors and a performance regression after version upgrade are unrelated." + "reason": "Compiled Python file errors and tokenizer-processing slowdowns after upgrade are different problems." }, { "left": "issue:42548", "right": "issue:44351", "accept": false, - "reason": "Different import errors for different symbols; same package, but not the same underlying bug." + "reason": "cannot import PreTrainedModel and cannot import HybridCache are different missing-symbol/import-export regressions." }, { "left": "issue:45125", "right": "issue:45375", "accept": false, - "reason": "Missing tensor-parallel plan and missing vision config field are different Qwen3.5-MoE issues." + "reason": "Both are Qwen3.5-related, but one is missing _tp_plan and the other is a vision-config field silently dropped; different bugs." }, { "left": "issue:46664", "right": "issue:46726", "accept": false, - "reason": "Video processor `num_frames` semantics and AutoProcessor backend-missing error are unrelated." + "reason": "Video num_frames semantics and missing vision-backend handling in AutoProcessor are unrelated multimodal issues." }, { - "left": "issue:43010", - "right": "issue:43232", + "left": "issue:43502", + "right": "issue:44222", "accept": false, - "reason": "Cache/layer `no_grad` concern is unrelated to generation kwargs update after `sync_gpus`." + "reason": "Offline loading/network behavior and FP8 MoE save_pretrained are unrelated." }, { - "left": "issue:43502", - "right": "issue:44961", + "left": "issue:43010", + "right": "issue:43232", "accept": false, - "reason": "`local_files_only` network requests and the unrelated `racoon` issue are not duplicates." + "reason": "Both mention generation/cache, but one is a no_grad decoration issue and the other is a sync_gpus update-model-kwargs bug." }, { "left": "issue:43502", - "right": "issue:44222", + "right": "issue:44961", "accept": false, - "reason": "Network requests despite `local_files_only` is unrelated to FP8 MoE save/pretrained behavior." + "reason": "local_files_only API leakage and a one-word issue title are not the same bug." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet, and the one accepted soft duplicate (issue:41669 vs issue:44246) is conservative and well supported by the titles. The remaining soft-edge rejections are also appropriately cautious; no overstatement or mergeability issue is apparent." + "feedback": "The cluster summary is grounded in the packet and stays conservative overall. The only accepted soft edge, issue:45125 \u2194 issue:45478, is plausibly the same Qwen3.5 MoE tensor-parallel/from_pretrained bug, while the other pairwise rejections are supported by distinct titles and failure modes. The summary does not overclaim beyond the evidence shown." }, "evaluator_used": true, "retried": false, @@ -17896,238 +17866,238 @@ "issue:44393|issue:45753", "issue:43576|issue:44556" ], - "prepared_review_unit_hash": "61fdb544773a2c6576db7c378a82481f9c0003e5c4bb77143bc1c4dd4ac1c301", + "prepared_review_unit_hash": "75dfac5f79e41c56fb7f11070583e1a4a7ddda6670847d21728766e341c7d596", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 34, - "serialized_chars": 36924, - "estimated_input_tokens": 9231, - "estimated_eval_tokens": 18718 + "serialized_chars": 36918, + "estimated_input_tokens": 9230, + "estimated_eval_tokens": 18716 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The items are mostly a heterogeneous mix of unrelated regressions, import errors, docs issues, and model-specific bugs. None of the soft-similarity pairs look like the same underlying defect/change, so all candidate duplicate merges should be rejected.", - "confidence": 0.94, - "canonical_issue_reason": null, + "summary": "All soft edges are false positives: the cluster mixes unrelated bugs, docs issues, import errors, and model-specific regressions. No pair looks like the same underlying change or fixable in one merged PR.", + "confidence": 0.95, + "canonical_issue_reason": "No single canonical issue fits this cluster; the items are too heterogeneous and the soft similarities are only lexical.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43576 is the clearest concrete regression report in the set and the most plausible representative issue, but the cluster is too heterogeneous to treat it as a true canonical duplicate target.", + "best_issue_reason": "If forced to pick an anchor, issue:43576 is the most central/referenced concrete v5 regression, but it still does not represent the other issues well.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:30990", "right": "issue:31795", "accept": false, - "reason": "Different problems: a loading hang vs documentation confusion in `model.forward`." + "reason": "Sentence-Transformers loading hang vs documentation confusion about input_ids/past_key_values; different bug types." }, { "left": "issue:43827", "right": "issue:44961", "accept": false, - "reason": "One is a docs cleanup after pipeline removals; the other is an unrelated issue with no matching bug signal." + "reason": "Docs about v5 pipeline removals vs an unrelated one-word issue; no shared failure mode." }, { "left": "issue:30333", "right": "issue:36296", "accept": false, - "reason": "MLFlow reporting on training failure is unrelated to tensor-parallel training behavior." + "reason": "MLflow job-status reporting vs tensor-parallel training bug; unrelated subsystems and symptoms." }, { "left": "issue:43299", "right": "issue:44556", "accept": false, - "reason": "Both involve v5 regressions, but one is Qwen3VL MoE loading and the other is checkpoint reload compatibility; not the same bug." + "reason": "Different model-loading regressions with different version contexts and model families; not the same bug." }, { "left": "issue:29870", "right": "issue:31795", "accept": false, - "reason": "A BART generation `UnboundLocalError` is not the same as a documentation inconsistency." + "reason": "Generation-time UnboundLocalError vs forward-signature documentation confusion; not the same underlying issue." }, { "left": "issue:44355", "right": "issue:45071", "accept": false, - "reason": "Compiled-file execution errors are different from `PretrainedConfig` type-checking breakage." + "reason": "Compiled-Python import/runtime errors are not the same as PretrainedConfig type-checking breakage." }, { "left": "issue:33453", "right": "issue:34689", "accept": false, - "reason": "Both are loading regressions, but they affect different model-loading paths and likely different causes." + "reason": "Both are loading regressions, but one is tokenizer loading and the other is a specific Llama 3.2 vision model-load breakage." }, { "left": "issue:42757", "right": "issue:43576", "accept": false, - "reason": "Importing `is_offline_mode` from `huggingface_hub` is a separate import regression from the broken `transformers env` command." + "reason": "ImportError from huggingface_hub vs broken transformers env command; different components and failure modes." }, { "left": "issue:43576", "right": "issue:45071", "accept": false, - "reason": "A CLI command regression and a config type-checking regression are different failure modes." + "reason": "Broken CLI command vs type-checking regression in PretrainedConfig; not the same code path." }, { "left": "issue:44315", "right": "issue:44360", "accept": false, - "reason": "`model_init` losing Liger Kernel application and a DSA/ReLU discussion are unrelated." + "reason": "Liger Kernel not applied in model_init vs DSA indexer lacking ReLU; unrelated training vs debug-tooling behavior." }, { "left": "issue:31795", "right": "issue:38617", "accept": false, - "reason": "Docs about `input_ids`/`past_key_values` do not match an `ImportError` for `layer_type_validation`." + "reason": "Doc clarification for model.forward vs ImportError for layer_type_validation; no common bug." }, { "left": "issue:43576", "right": "issue:43824", "accept": false, - "reason": "Broken environment CLI and missing `Qwen2_5_VLForConditionalGeneration` export are distinct issues." + "reason": "CLI/environment command failure vs missing Qwen2.5-VL export; different areas and fixes." }, { "left": "issue:44961", "right": "issue:45468", "accept": false, - "reason": "The placeholder 'racoon' issue is not the same as the Gemma-4 audio positional encoding bug." + "reason": "A placeholder title vs Gemma-4 audio positional encoding bug; clearly unrelated." }, { "left": "issue:44485", "right": "issue:44961", "accept": false, - "reason": "GLM-5 RoPE implementation discussion is unrelated to the placeholder issue." + "reason": "GLM-5 RoPE discussion vs placeholder issue; no substantive overlap." }, { "left": "issue:44351", "right": "issue:46617", "accept": false, - "reason": "A missing `HybridCache` import is different from a cache-file `FileNotFoundError` when loading from cache." + "reason": "Missing top-level import vs file-not-found when loading cache; both import-related but not the same failure." }, { "left": "issue:44908", "right": "issue:44961", "accept": false, - "reason": "Scheduler kwargs being ignored has nothing to do with the placeholder issue." + "reason": "Scheduler kwargs ignored vs placeholder issue; no duplicate signal." }, { "left": "issue:43576", "right": "issue:44938", "accept": false, - "reason": "Both mention v5 breakage, but one is the `transformers env` command and the other is Python 3.14 load failure." + "reason": "Broken env command in v5 vs Python 3.14 loading failure; different root causes." }, { "left": "issue:36296", "right": "issue:43901", "accept": false, - "reason": "Tensor-parallel training bugs are unrelated to `return_all_scores` documentation mismatch." + "reason": "Tensor parallel training bug vs docs mismatch for TextClassificationPipeline; unrelated." }, { "left": "issue:46143", "right": "issue:46144", "accept": false, - "reason": "Missing `**kwargs` forwarding and RoFormer attention-interface usage are related broadly, but not the same concrete bug." + "reason": "Missing kwargs passthrough in RoFormer methods vs attention-interface implementation issue; related family, different concrete bug." }, { "left": "issue:43475", "right": "issue:46617", "accept": false, - "reason": "A missing SAM3 vision output attribute is unrelated to trust-remote-code cache loading failures." + "reason": "SAM 3 Video attribute error vs trust_remote_code cache FileNotFoundError; unrelated models and code paths." }, { "left": "issue:44222", "right": "issue:44230", "accept": false, - "reason": "One is about FP8 `save_pretrained`, the other about FP8 inference support; different code paths." + "reason": "FP8 save_pretrained problem vs FP8 inference support request; one is saving, the other is inference capability." }, { "left": "issue:41628", "right": "issue:42548", "accept": false, - "reason": "`AutoImageProcessor` and `PreTrainedModel` import errors are separate symbol/export problems." + "reason": "Missing AutoImageProcessor export vs missing PreTrainedModel export; similar import symptom but different symbols and likely fixes." }, { "left": "issue:44315", "right": "issue:46209", "accept": false, - "reason": "Liger Kernel application with `model_init` is unrelated to FP8 inference on B200." + "reason": "model_init/Liger Kernel application bug vs FP8 inference on B200 hardware; not the same code path." }, { "left": "issue:44485", "right": "issue:46664", "accept": false, - "reason": "GLM-5 RoPE implementation and SmolVLM `num_frames` semantics are different model-specific issues." + "reason": "GLM-5 RoPE implementation discussion vs SmolVLM num_frames semantics; different models and bugs." }, { "left": "issue:43673", "right": "issue:44492", "accept": false, - "reason": "A functional cache-missing bug is not the same as a typo in cache strategy text." + "reason": "GenerationMixin cache missing during chunked_prefill vs cache-strategy typo; one is functional regression, the other is wording." }, { "left": "issue:45362", "right": "issue:45850", "accept": false, - "reason": "A Qwen3.5-35B chat crash is much more specific than a broad complaint that v5 breaks many models." + "reason": "Qwen3.5 chat crash vs broad complaint that v5 breaks many models; different granularity and symptoms." }, { "left": "issue:44273", "right": "issue:45542", "accept": false, - "reason": "Lazy loading problems and TensorBoard/TensorFlow backend detection are different defects." + "reason": "Lazy loading malfunction vs TensorBoard-without-TensorFlow backend error; unrelated initialization issues." }, { "left": "issue:44623", "right": "issue:45230", "accept": false, - "reason": "Missing files from `processor.save_pretrained()` is not the same as a generic bug report." + "reason": "Missing processor.save_pretrained files vs generic bug report; no shared underlying defect." }, { "left": "issue:44360", "right": "issue:46664", "accept": false, - "reason": "A DSA/ReLU discussion is unrelated to SmolVLM video-frame semantics." + "reason": "DSA indexer ReLU discussion vs SmolVLM video-frame semantics; unrelated subsystems." }, { "left": "issue:44038", "right": "issue:45125", "accept": false, - "reason": "A Qwen3-VL loading bug and missing tensor-parallel plan for Qwen3_5Moe are separate model-family issues." + "reason": "Qwen3-VL / model loading report vs missing _tp_plan for tensor parallelism; different bugs." }, { "left": "issue:44273", "right": "issue:45003", "accept": false, - "reason": "Lazy loading not working and unsafe `sys.modules` access are different loading-path bugs." + "reason": "Lazy loading issue vs unsafe sys.modules access in modeling_utils; distinct root causes." }, { "left": "issue:44393", "right": "issue:45071", "accept": false, - "reason": "Qwen3-VL 2D bbox output errors are unrelated to `PretrainedConfig` type-checking failures." + "reason": "Qwen3-VL bounding-box output error vs PretrainedConfig type-checking breakage; different models and failure modes." }, { "left": "issue:44393", "right": "issue:45753", "accept": false, - "reason": "Both touch Qwen3-VL/Qwen3.5, but the symptoms differ: bbox output errors vs an infinite loop on a specific image." + "reason": "Bounding-box hallucination/error vs infinite loop on a specific image; both Qwen3.5-VL but different concrete behaviors." }, { "left": "issue:43576", "right": "issue:44556", "accept": false, - "reason": "The v5 `transformers env` CLI break and checkpoint reload incompatibility are distinct regressions." + "reason": "Broken env command vs checkpoint reload regression after upgrade; unrelated symptoms and fixes." } ] }, @@ -18241,268 +18211,274 @@ "issue:41669|issue:43704", "issue:31795|issue:34689" ], - "prepared_review_unit_hash": "b256177e7a3e7fb4bafee74583a503cc8fb6994ec6c762550612f040eeb298a5", + "prepared_review_unit_hash": "4b2d0c81a3d6ddfc37ca664e41f054f4550e86b7c12ed4c3796c7665aed45b9a", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 40, - "serialized_chars": 38535, - "estimated_input_tokens": 9634, - "estimated_eval_tokens": 19524 + "serialized_chars": 38543, + "estimated_input_tokens": 9636, + "estimated_eval_tokens": 19528 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose collection of Transformers v5 regression reports around model loading, import behavior, and config/API compatibility. The soft pairs are semantically related in places, but none are strong enough to treat as the same underlying bug/change.", - "confidence": 0.78, - "canonical_issue_reason": "45850 is the broadest umbrella in the set: it summarizes the general v5 breakage theme affecting many models and code paths.", + "summary": "This cluster is a loose semantic grab-bag of transformer regressions and model-specific bugs, with only superficial overlap around v5 upgrades, loading/import issues, and lazy-loading/performance complaints. I do not see any true duplicate pairs among the soft edges.", + "confidence": 0.9, + "canonical_issue_reason": "Issue 45850 is the closest umbrella-style report here because it broadly summarizes the v5 breakage theme, but the cluster is not a tight duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "45278 is the clearest representative of the recurring concrete pattern here: upgrade-related import/loading regressions across multiple models.", + "best_issue_reason": "Issue 45850 is the best representative issue for the cluster since it is the most general and central to the shared 'v5 breaks things' pattern, even though most members are not actual duplicates.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41669", "right": "issue:43673", "accept": false, - "reason": "Both relate to import/loading behavior, but one is about `import *` slowing imports and the other is a missing cache object during chunked prefill; different bugs." + "reason": "Different problems: import-star / import-time slowdown versus missing GenerationMixin cache during chunked prefill." }, { "left": "issue:43525", "right": "issue:45850", "accept": false, - "reason": "`pad_token_id` missing in one config class is a specific attribute regression; the other is a broad complaint about many models breaking after v5." + "reason": "One is a specific Llama4Config attribute error; the other is a broad complaint about many models breaking in v5." }, { "left": "issue:44556", "right": "issue:45092", "accept": false, - "reason": "Both are v5 loading incompatibilities, but one is checkpoint reload after upgrade and the other is remote-code/meta-init incompatibility for a specific model family." + "reason": "Both involve upgrade incompatibilities, but they describe different failure modes and model families." }, { "left": "issue:43572", "right": "issue:45850", "accept": false, - "reason": "A missing `pad_token_idx` in one config is not the same underlying issue as a general 'many models break' regression report." + "reason": "StableLmConfig pad_token_idx regression is a specific config bug, not the same as the generic v5 breakage report." }, { "left": "issue:44492", "right": "issue:45850", "accept": false, - "reason": "A typo in cache-strategy docs/labels is unrelated to the broad v5 model-breakage report." + "reason": "A cache-strategy typo is unrelated to a broad model-loading regression." }, { "left": "issue:45278", "right": "issue:45850", "accept": false, - "reason": "Related theme, but one is a concrete import-error regression after upgrade while the other is a generic umbrella complaint; not the same bug." + "reason": "Both are post-v5 regressions, but one is import errors and the other is a vague many-model breakage complaint." }, { "left": "issue:44230", "right": "issue:46095", "accept": false, - "reason": "Different requests entirely: FP8 inference support for Qwen models versus a DeepSeekV4 weight-conversion question." + "reason": "fp8 support for Qwen3-VL/Qwen3.5 differs completely from a DeepSeek weight-conversion question." }, { "left": "issue:42371", "right": "issue:44492", "accept": false, - "reason": "TF32 API guidance and a cache-strategy typo are unrelated issues." + "reason": "TF32 API guidance and cache-strategy typo are unrelated topics." }, { "left": "issue:41669", "right": "issue:46218", "accept": false, - "reason": "`import *` performance complaints are not the same as a missing `Qwen3_5ForCausalLM` import/export error." + "reason": "Import-star cleanup/performance and a missing Qwen3_5 import error are not the same bug." }, { "left": "issue:43408", "right": "issue:44315", "accept": false, - "reason": "Different model/config problems: a wrong model-type warning versus Liger Kernel not being applied with `model_init`." + "reason": "SAM model-type mismatch is unrelated to Liger Kernel not being applied under model_init." }, { "left": "issue:42890", "right": "issue:44908", "accept": false, - "reason": "One is missing `set_seed()` in tests; the other is a scheduler ignoring kwargs. No shared underlying bug." + "reason": "A flaky integration test due to missing seeds is unrelated to an inverse_sqrt scheduler kwargs bug." }, { "left": "issue:44246", "right": "issue:44273", "accept": false, - "reason": "Both mention loading/lazy-loading, but one is intermittent import slowness and the other is a general lazy-loading failure; too broad to merge." + "reason": "Both mention import/lazy-loading behavior, but one is slow import timing and the other is a functional lazy-loading failure; too weak to merge as duplicates." }, { "left": "issue:43576", "right": "issue:44908", "accept": false, - "reason": "`transformers env` is broken in v5, while the other is a scheduler keyword-argument bug; different code paths." + "reason": "A broken transformers env command is unrelated to scheduler kwargs handling." }, { "left": "issue:43352", "right": "issue:44393", "accept": false, - "reason": "Flash Attention 2 support gating for Nemotron is unrelated to a Qwen3-VL bounding-box output bug." + "reason": "Nemotron Flash Attention 2 support and Qwen3-VL bounding-box errors are different model/code paths." }, { "left": "issue:43976", "right": "issue:44393", "accept": false, - "reason": "A Python version compatibility issue is not the same as a Qwen3-VL hallucination/output problem." + "reason": "Python-version compatibility and Qwen3-VL output errors are unrelated." }, { "left": "issue:44908", "right": "issue:45071", "accept": false, - "reason": "A scheduler kwargs bug and a `PretrainedConfig` type-checking regression are separate problems." + "reason": "Missing scheduler kwargs and PretrainedConfig type-checking are different regressions." }, { "left": "issue:44295", "right": "issue:45230", "accept": false, - "reason": "Reading `position_ids` after buffer registration is a different failure mode than an unspecified bug report." + "reason": "A position_ids buffer read error is too generic to equate with an empty 'bug report' placeholder." }, { "left": "issue:44297", "right": "issue:45230", "accept": false, - "reason": "Tokenizer save metadata mismatch is a specific serialization bug, not the same as the generic bug report." + "reason": "Tokenizer save_pretrained class mismatch is unrelated to the placeholder bug report." }, { "left": "issue:44704", "right": "issue:44908", "accept": false, - "reason": "`AutoProcessor.from_pretrained` not forwarding kwargs to `cached_file` is unrelated to an inverse-sqrt scheduler parameter bug." + "reason": "AutoProcessor kwargs forwarding is unrelated to inverse_sqrt scheduler kwargs." }, { "left": "issue:44492", "right": "issue:46328", "accept": false, - "reason": "A typo in cache strategies is not the same as `SequenceFeatureExtractor.pad` converting arrays inefficiently." + "reason": "Cache-strategy typo and SequenceFeatureExtractor.pad list conversion are different issues." }, { "left": "issue:41669", "right": "issue:44351", "accept": false, - "reason": "Import-star cleanup and a `HybridCache` import failure are distinct import-related problems." + "reason": "Import-star/lazy loading cleanup is not the same as a missing HybridCache export/import error." }, { "left": "issue:41669", "right": "issue:45341", "accept": false, - "reason": "A broad import-performance issue and a small testing-utils bug are not the same underlying change." + "reason": "Import performance cleanup and a testing_utils bug are unrelated." }, { "left": "issue:43976", "right": "issue:45850", "accept": false, - "reason": "Python-version incompatibility is more specific than the broad 'many models break' v5 complaint, and they do not point to the same defect." + "reason": "Python support incompatibility and a broad v5 breakage complaint are not the same underlying bug." }, { "left": "issue:44393", "right": "issue:45478", "accept": false, - "reason": "Both are Qwen3.5/Qwen3-VL-related, but the symptoms are different model errors with no evidence of a shared root cause." + "reason": "Both are Qwen-related, but one is a 2D bbox hallucination/error and the other is a Qwen3.5 MoE from_pretrained failure." }, { "left": "issue:43976", "right": "issue:46218", "accept": false, - "reason": "A Python support regression and a missing module import are unrelated." + "reason": "Python-version compatibility is unrelated to a specific missing Qwen3_5 import." }, { "left": "issue:44077", "right": "issue:44861", "accept": false, - "reason": "One is an API/validation rule change for `post_init`; the other is an AttributeError in tied-weight key handling." + "reason": "Optional post_init enforcement and a tied-weights AttributeError are different code paths." }, { "left": "issue:31795", "right": "issue:33453", "accept": false, - "reason": "Documentation confusion about forward args is unrelated to a tokenizer-loading regression." + "reason": "Documentation confusion around forward arguments is unrelated to tokenizer loading regression." }, { "left": "issue:43576", "right": "issue:44273", "accept": false, - "reason": "A broken CLI environment command and a lazy-loading failure are too different to be duplicates." + "reason": "Broken env command and lazy-loading behavior are not the same bug." }, { "left": "issue:44315", "right": "issue:44829", "accept": false, - "reason": "Liger Kernel integration not applied with `model_init` is a separate issue from degenerate training under flash_attention_3." + "reason": "Liger Kernel application under model_init is unrelated to flash_attention_3 training degeneration." }, { "left": "issue:44360", "right": "issue:45468", "accept": false, - "reason": "A missing ReLU in the DSA indexer and Gemma-4 positional encoding are unrelated model-implementation issues." + "reason": "A DSA/indexer discussion and a Gemma-4 audio positional encoding bug are unrelated." }, { "left": "issue:43519", "right": "issue:46655", "accept": false, - "reason": "Both are processor math bugs, but one is timestamp calculation and the other is video padding calculation; different code paths." + "reason": "Incorrect timestamp calculation and incorrect padding calculation affect different processors and modalities." }, { "left": "issue:44273", "right": "issue:44908", "accept": false, - "reason": "Lazy loading failures and scheduler kwargs handling are unrelated." + "reason": "Lazy-loading problems and scheduler kwargs handling are unrelated." }, { "left": "issue:44655", "right": "issue:44861", "accept": false, - "reason": "Saving pipeline objects and tied-weight key crashes are separate serialization/internal-state bugs." + "reason": "Pipeline save_pretrained support and tied-weight key handling are distinct issues." }, { "left": "issue:45230", "right": "issue:45310", "accept": false, - "reason": "An empty bug report does not identify the same issue as the Qwen3.5 MoE from_pretrained error." + "reason": "A placeholder bug report is not a duplicate of the Qwen3.5 MoE from_pretrained error." }, { "left": "issue:43576", "right": "issue:46218", "accept": false, - "reason": "A broken environment CLI command and a missing Qwen3_5 model import are distinct failures." + "reason": "Broken env CLI and missing Qwen3_5 import are unrelated." }, { "left": "issue:43519", "right": "issue:44485", "accept": false, - "reason": "Qwen3-VL timestamp math and GLM-5 RoPE implementation are both model-specific, but not the same bug." + "reason": "Qwen3VL timestamp calculation and GLM-5 RoPE implementation are different model-specific bugs." }, { "left": "issue:45468", "right": "issue:46894", "accept": false, - "reason": "Gemma-4 audio positional encoding and CohereASR training-loss shifting are different model/training bugs." + "reason": "Gemma-4 audio positional encoding and CohereASR training-loss double shift are different bugs." + }, + { + "left": "issue:43519", + "right": "issue:44908", + "accept": false, + "reason": "Video timestamp calculation and scheduler kwargs handling are unrelated." }, { "left": "issue:41669", "right": "issue:43704", "accept": false, - "reason": "Import-star cleanup and a VRAM leak in multi-dataloader-thread use are unrelated." + "reason": "Import cleanup/performance and a VRAM leak in multi-threaded dataloaders are different issues." }, { "left": "issue:31795", "right": "issue:34689", "accept": false, - "reason": "Forward-API documentation confusion is not the same as a model-loading failure for Llama 3.2 Vision." + "reason": "Forward-doc confusion and a Llama 3.2 Vision model-loading breakage are not duplicates." } ] }, @@ -18577,8 +18553,8 @@ "issue:46018|issue:46172", "issue:43704|issue:44485", "issue:44485|issue:45468", - "issue:42371|issue:43704", "issue:43502|issue:43519", + "issue:42371|issue:43704", "issue:45850|issue:46218", "issue:43519|issue:44623", "issue:44263|issue:45468", @@ -18598,8 +18574,8 @@ "issue:31515|issue:38617", "issue:43572|issue:44861", "issue:43525|issue:44861", - "issue:28348|issue:43352", "issue:44230|issue:44623", + "issue:28348|issue:43352", "issue:29870|issue:31515", "issue:30333|issue:31515", "issue:31515|issue:36296", @@ -18619,20 +18595,20 @@ "issue:37934|issue:43668", "issue:42936|issue:43668" ], - "prepared_review_unit_hash": "d608f88c678707decacc54b86967c736f1ba99820bec13aacec1a26fcd2f60ee", + "prepared_review_unit_hash": "b05f307253f5c209236b43df5d122ff62485dae1e9221a2e16dcd213192165be", "component_budget": { "node_count": 500, "item_count": 500, "soft_pair_count": 1839, - "serialized_chars": 953882, - "estimated_input_tokens": 238471, - "estimated_eval_tokens": 477198 + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 }, "budget": { "node_count": 46, "item_count": 46, "soft_pair_count": 44, - "serialized_chars": 38877, + "serialized_chars": 38880, "estimated_input_tokens": 9720, "estimated_eval_tokens": 19696 }, @@ -18641,54 +18617,54 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is noisy and does not contain any clear duplicate issue/PR groupings. The soft pairs mostly share broad vocabulary (v5, tokenizer, attention, import, loading) but describe different models, subsystems, and failure modes. I rejected all soft edges and picked the most representative concrete issue as a weak anchor.", - "confidence": 0.58, - "canonical_issue_reason": "issue:31515 is the most concrete and broadly framed user-facing regression in the set, and it has the strongest soft connectivity, but there is still no true duplicate cluster here.", + "summary": "This is a loose cluster of mostly unrelated transformer/model regression issues, with many false-positive similarity hits. No soft pair looks like the same concrete bug/change, and there are no PRs in the set.", + "confidence": 0.42, + "canonical_issue_reason": "issue:45850 is the closest thing to an umbrella report for broad v5 breakage/regression complaints, so it is the best single representative even though it is very vague.", "canonical_pr_reason": null, - "best_issue_reason": "issue:31515 is the best single issue anchor because it is specific, actionable, and closest to a general underlying runtime/loading problem, even though it does not truly duplicate the others.", + "best_issue_reason": "issue:45850 best fits as a global representative because it spans the broadest failure theme seen here (many models breaking after 5.0), but it is still too generic to be an ideal canonical duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46018", "right": "issue:46172", "accept": false, - "reason": "Different models and bugs: DeepSeek-V4 gating vs RoFormer attention implementation." + "reason": "Same model family, but different bugs: shared-expert gating vs a DeepSeek-V4 weight conversion/transpose issue." }, { "left": "issue:43704", "right": "issue:44485", "accept": false, - "reason": "VRAM leak in dataloader threads is unrelated to GLM-5 RoPE math." + "reason": "Completely different models and failures: Qwen3 VRAM leak vs GLM-5 RoPE implementation." }, { "left": "issue:44485", "right": "issue:45468", "accept": false, - "reason": "Both are model-specific, but the bugs are different: RoPE implementation vs audio positional encoding." + "reason": "Different architectures and symptoms: GLM-5 RoPE vs Gemma-4 audio positional encoding." }, { - "left": "issue:42371", - "right": "issue:43704", + "left": "issue:43502", + "right": "issue:43519", "accept": false, - "reason": "TF32 configuration API behavior is unrelated to a multi-thread VRAM leak." + "reason": "One is unintended network access with local_files_only, the other is a timestamp calculation bug in a processor." }, { - "left": "issue:43502", - "right": "issue:43519", + "left": "issue:42371", + "right": "issue:43704", "accept": false, - "reason": "local_files_only network requests and Qwen3VL timestamp calculation are different code paths." + "reason": "TF32 configuration guidance is unrelated to a Qwen3 VRAM leak in dataloader threads." }, { "left": "issue:45850", "right": "issue:46218", "accept": false, - "reason": "A vague version-wide breakage report is not the same bug as a specific Qwen3_5 import failure." + "reason": "A broad v5 breakage complaint is not the same concrete issue as a Qwen3_5 import failure." }, { "left": "issue:43519", "right": "issue:44623", "accept": false, - "reason": "Processor timestamp math and save_pretrained missing files are unrelated." + "reason": "Different code paths: Qwen3VL timestamp math vs missing files in processor.save_pretrained()." }, { "left": "issue:44263", @@ -18700,175 +18676,169 @@ "left": "issue:44261", "right": "issue:45468", "accept": false, - "reason": "Missing rms_norm_eps in MLA q_a_layernorm does not match Gemma-4 audio positional encoding." + "reason": "RMS norm epsilon precision issue is unrelated to Gemma-4 audio positional encoding." }, { "left": "issue:43298", "right": "issue:46095", "accept": false, - "reason": "Continuous batching for audio models and DeepSeek-V4 weight conversion are unrelated." + "reason": "Audio continuous batching support and DeepSeek-V4 weight conversion are different feature requests." }, { "left": "issue:43519", "right": "issue:43704", "accept": false, - "reason": "Timestamp calculation and VRAM leak in dataloader threads are different issues." + "reason": "Timestamp calculation bug and VRAM leak are different failures in different models." }, { "left": "issue:44351", "right": "issue:46218", "accept": false, - "reason": "Both are import errors, but for different missing symbols/modules and different underlying causes." + "reason": "Both are import errors, but for different symbols and likely different root causes." }, { "left": "issue:43352", "right": "issue:46218", "accept": false, - "reason": "Flash Attention support for NemotronH is unrelated to importing Qwen3_5ForCausalLM." + "reason": "Flash Attention support for NemotronH is unrelated to a Qwen3_5 module import failure." }, { "left": "issue:42913", "right": "issue:43066", "accept": false, - "reason": "Both mention tokenizer v5, but one is a broad behavior regression and the other is a specific decoder-type bug." + "reason": "Both involve tokenizer/v5 behavior, but one is a broad behavior regression and the other is a specific decoder-type bug." }, { "left": "issue:28348", "right": "issue:45250", "accept": false, - "reason": "Both mention Flash Attention 2.0, but they target different models and different requests." + "reason": "GPT2-specific Flash Attention 2.0 support request is not the same as a generic Flash Attention 2.0 issue." }, { "left": "issue:44908", "right": "issue:44998", "accept": false, - "reason": "Scheduler kwargs handling is unrelated to the nonsensical 'Unemployment' issue." + "reason": "Scheduler kwargs handling has no relation to the unrelated 'Unemployment' issue." }, { "left": "issue:44492", "right": "issue:44998", "accept": false, - "reason": "A cache-strategy typo is not related to the unrelated 'Unemployment' issue." + "reason": "Cache-strategy typo and the unrelated title clearly do not describe the same bug." }, { "left": "issue:43618", "right": "issue:44998", "accept": false, - "reason": "CLIPOutput attentions regression is unrelated to the unrelated 'Unemployment' issue." - }, - { - "left": "issue:44998", - "right": "issue:45850", - "accept": false, - "reason": "No relation between the unrelated 'Unemployment' issue and a broad v5 breakage report." + "reason": "CLIPOutput attentions regression is unrelated to the unrelated title." }, { "left": "issue:44938", "right": "issue:44998", "accept": false, - "reason": "Python 3.14 loading failure is unrelated to the unrelated 'Unemployment' issue." + "reason": "Python 3.14 import/load failure is unrelated to the unrelated title." }, { "left": "issue:44861", "right": "issue:45290", "accept": false, - "reason": "Tied-weight key crash and chat template tool-call crash are different failures." + "reason": "Different crashes in different subsystems: tied-weights helper vs chat template tool-call handling." }, { "left": "issue:31515", "right": "issue:38617", "accept": false, - "reason": "Slow from_pretrained loading and missing layer_type_validation import are unrelated." + "reason": "Slow from_pretrained checkpoint loading is unrelated to a missing import symbol." }, { "left": "issue:43572", "right": "issue:44861", "accept": false, - "reason": "Missing pad_token_idx in config is unrelated to tied-weight key handling." + "reason": "Missing pad_token_idx in config is unrelated to _get_tied_weight_keys crashing on list input." }, { "left": "issue:43525", "right": "issue:44861", "accept": false, - "reason": "Missing pad_token_id in Llama4Config is unrelated to tied-weight key handling." + "reason": "Llama4Config pad_token_id missing and tied-weights handling are different issues." }, { - "left": "issue:28348", - "right": "issue:43352", + "left": "issue:44230", + "right": "issue:44623", "accept": false, - "reason": "Both concern Flash Attention 2.0, but they are different models and different kinds of support requests." + "reason": "FP8 inference support request is unrelated to missing files in processor.save_pretrained()." }, { - "left": "issue:44230", - "right": "issue:44623", + "left": "issue:28348", + "right": "issue:43352", "accept": false, - "reason": "FP8 support for Qwen3-VL/Qwen3.5 MoE is unrelated to processor.save_pretrained missing files." + "reason": "Both mention Flash Attention 2.0, but for different models and different support gaps." }, { "left": "issue:29870", "right": "issue:31515", "accept": false, - "reason": "Bart generate decoder mask crash and from_pretrained slowdown are unrelated." + "reason": "BART generate decoder_attention_mask bug is unrelated to slow checkpoint loading." }, { "left": "issue:30333", "right": "issue:31515", "accept": false, - "reason": "MLflow reporting on failed jobs is unrelated to checkpoint loading slowness." + "reason": "MLflow reporting on training failure is unrelated to from_pretrained load performance." }, { "left": "issue:31515", "right": "issue:36296", "accept": false, - "reason": "from_pretrained performance and tensor-parallel training bug are different problems." + "reason": "Slow checkpoint loading and tensor-parallel training bugs are different problems." }, { "left": "issue:31515", "right": "issue:31795", "accept": false, - "reason": "Loading performance and documentation confusion are not the same issue." + "reason": "Load performance is unrelated to documentation confusion about forward inputs." }, { "left": "issue:44230", "right": "issue:45310", "accept": false, - "reason": "FP8 support and a from_pretrained error for Qwen3.5 MoE are not the same concrete bug." + "reason": "FP8 inference support and Qwen3.5 from_pretrained import errors are different issues." }, { "left": "issue:43298", "right": "issue:44230", "accept": false, - "reason": "Continuous batching for audio models is unrelated to FP8 support for Qwen3-VL/Qwen3.5 MoE." + "reason": "Audio continuous batching support and FP8 inference support are different feature requests." }, { "left": "issue:43717", "right": "issue:46144", "accept": false, - "reason": "Mamba init_weights usage is unrelated to RoFormer not using the attention interface." + "reason": "Mamba init_weights behavior is unrelated to RoFormer not using the attention interface." }, { "left": "issue:42936", "right": "issue:46018", "accept": false, - "reason": "Mask2former ignore_value handling is unrelated to DeepSeek-V4 gating." + "reason": "Mask2former ignore_value handling is unrelated to DeepSeek-V4 shared-expert gating." }, { "left": "issue:43668", "right": "issue:44031", "accept": false, - "reason": "A config type hint bug is unrelated to tokenizer regex warning spam." + "reason": "A type-hint problem in ModernBERTConfig is unrelated to tokenizer regex warnings." }, { "left": "issue:42936", "right": "issue:43240", "accept": false, - "reason": "Ignored mask value handling and kwargs propagation to loss are different bugs." + "reason": "Mask2former ignore_value and kwargs not passed to loss are different code paths." }, { "left": "issue:43240", "right": "issue:46144", "accept": false, - "reason": "Loss kwargs propagation is unrelated to RoFormer attention interface use." + "reason": "Loss kwargs propagation is unrelated to RoFormer attention interface usage." }, { "left": "issue:42936", @@ -18880,37 +18850,37 @@ "left": "issue:42936", "right": "issue:46144", "accept": false, - "reason": "Mask2former ignore_value handling is unrelated to RoFormer attention interface usage." + "reason": "Mask2former ignore_value and RoFormer attention interface are unrelated bugs." }, { "left": "issue:42936", "right": "issue:43813", "accept": false, - "reason": "A missing ignore_value use site is unrelated to a typo in quantization_operation." + "reason": "Ignore_value handling is unrelated to a typo in quantization_operation." }, { "left": "issue:43240", "right": "issue:43668", "accept": false, - "reason": "Loss kwargs propagation and a config type-hint bug are unrelated." + "reason": "Passing kwargs to loss and a config type-hint mismatch are different issues." }, { "left": "issue:43668", "right": "issue:43813", "accept": false, - "reason": "Incorrect type hint and typo in quantization_operation are unrelated." + "reason": "ModernBERT norm_eps typing and a quantization typo are unrelated." }, { "left": "issue:37934", "right": "issue:43668", "accept": false, - "reason": "Llama4TextL2Norm/RMS norm question is unrelated to a ModernBERT type-hint issue." + "reason": "Questioning whether Llama4TextL2Norm is RMS norm is unrelated to a ModernBERT type-hint issue." }, { "left": "issue:42936", "right": "issue:43668", "accept": false, - "reason": "Mask2former ignore_value and ModernBERT norm_eps typing are unrelated." + "reason": "Mask2former ignore_value handling is unrelated to a ModernBERT norm_eps typing bug." } ] }, @@ -18943,52 +18913,52 @@ "issue:10519|issue:11005", "issue:11005|issue:11307" ], - "prepared_review_unit_hash": "1c745a834526bfbbd579ab48506844f5b1d94c4370237f758623c5e5f215598c", + "prepared_review_unit_hash": "5ba1e4116af6d658a42b901e6fc66417465227b9805dc083af462984337500f8", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 2875, - "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": 2875, - "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 not duplicates: one asks for a truncation-from-beginning option, one asks for ReduceLROnPlateau-like behavior, and one asks for Wav2Vec2 word boundary timestamps. They target different features and code paths.", - "confidence": 0.96, - "canonical_issue_reason": "No canonical issue: the items are about unrelated requests, so there is no single underlying bug or feature to consolidate.", + "summary": "False-positive cluster: the three issues are unrelated feature requests in different areas (tokenizer truncation behavior, learning-rate scheduling, and Wav2Vec2 word boundary offsets).", + "confidence": 0.98, + "canonical_issue_reason": "No canonical issue: there is no shared underlying bug or enhancement across these tickets.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #10519 is the most concrete and actionable of the set, with a clear feature request and specific truncation behavior, even though it is not duplicate of the others.", + "best_issue_reason": "None of the issues is a good cluster representative because they address different code paths and user needs; if forced, 10519 is the most concrete standalone request, but it is not a duplicate of the others.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:10519", "right": "issue:11307", "accept": false, - "reason": "Different domains: token truncation behavior vs Wav2Vec2 word offset extraction. No shared underlying bug or change." + "reason": "Different features in different subsystems: tokenizer truncation vs Wav2Vec2 word offset extraction; no shared bug or fix path." }, { "left": "issue:10519", "right": "issue:11005", "accept": false, - "reason": "One is about tokenizer truncation direction; the other is about learning-rate scheduling functionality. These are unrelated feature requests." + "reason": "Unrelated requests: truncation strategy for tokenizers vs ReduceLROnPlateau-like scheduler behavior." }, { "left": "issue:11005", "right": "issue:11307", "accept": false, - "reason": "ReduceLROnPlateau-like training control and Wav2Vec2 timing offsets are unrelated and would not be merged as one issue." + "reason": "Distinct domains and problems: learning-rate scheduling vs speech model word timing offsets." } ] }, @@ -19020,12 +18990,12 @@ "issue:13244|issue:16618", "issue:15980|issue:16618" ], - "prepared_review_unit_hash": "1c7c6c2b77cdc567735227dcb3a7a69e721f7082446520adda4dfb574dc6744f", + "prepared_review_unit_hash": "1eaa8d29dc06c082ed75dc9d4e81e5090c54881f1547c2376391c13db944c9c2", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 2477, + "serialized_chars": 2478, "estimated_input_tokens": 620, "estimated_eval_tokens": 1496 }, @@ -19033,7 +19003,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 2477, + "serialized_chars": 2478, "estimated_input_tokens": 620, "estimated_eval_tokens": 1496 }, @@ -19042,24 +19012,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly a loose semantic group around model/tokenization topics, but the Tapas tokenization issue is unrelated to the model-loading reports. The two model-loading issues are only superficially similar and do not look like the same underlying bug.", - "confidence": 0.72, - "canonical_issue_reason": "Issue #16618 is the broadest and most representative of the only plausible overlap in the cluster, but the duplication evidence is weak.", + "summary": "The three issues are only loosely related through model/tokenizer loading, but they describe different problems: Tapas tokenization mismatch, a private-model auth/error-message issue, and a bert-base-uncased loading failure. I would not merge any of the soft pairs as duplicates.", + "confidence": 0.88, + "canonical_issue_reason": "issue #16618 is the best representative of the cluster because it is the most general concrete model-loading failure and has the strongest apparent overlap with the other two, even though the underlying problems are still distinct.", "canonical_pr_reason": null, - "best_issue_reason": "#16618 is the best issue candidate only because it is the closest to the cluster's shared model-loading theme; #15980 is a narrower auth/error-message report and #13244 is unrelated.", + "best_issue_reason": "#16618 is the clearest central issue to keep as the cluster anchor: it is the most user-facing loading failure and the closest semantic match to the other titles, but not a true duplicate of either.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:13244", "right": "issue:16618", "accept": false, - "reason": "Tapas tokenization vs. loading bert-base-uncased are different code paths and different user problems; this is not the same bug." + "reason": "Reject: one is about Tapas tokenization diverging from TensorFlow code, the other is a failure to load bert-base-uncased. They involve different models and a different class of bug." }, { "left": "issue:15980", "right": "issue:16618", "accept": false, - "reason": "One is specifically about the error message when downloading a private model while logged out; the other is a generic inability to load bert-base-uncased. Same general area, but not the same underlying issue." + "reason": "Reject: #15980 is specifically about a bad error message when downloading a private model without login, while #16618 reports inability to load bert-base-uncased. Same broad area, but not the same underlying issue." } ] }, @@ -19089,7 +19059,7 @@ "soft_pairs": [ "issue:26350|issue:33260" ], - "prepared_review_unit_hash": "75c2f88b8a6c6012032f577cfdb1814882bcf520a49bef3c26f30f166547807f", + "prepared_review_unit_hash": "fe1eddfd054fe5f129dade87e2db858dac6f1fb1bcc9aaf1648d9445199ea9ad", "component_budget": { "node_count": 2, "item_count": 2, @@ -19111,18 +19081,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates: one requests Flash Attention 2 support for more architectures, the other requests GGUF support for more architectures. They target different features and code paths.", - "confidence": 0.97, - "canonical_issue_reason": "No single canonical issue fits here; the two issues are distinct feature requests with different underlying implementation work.", + "summary": "These are two distinct community contribution feature requests: one for Flash Attention 2 support and one for GGUF support. They target different capabilities and architectures, so they should not be merged as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": "Issue #33260 is the better representative of the cluster only in a loose triage sense because it is newer and has more inbound references, but it is not a duplicate of #26350.", "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a better canonical for the other, since they do not describe the same bug or change.", + "best_issue_reason": "Issue #33260 is the strongest standalone issue because it has higher visible engagement and is phrased as a clear support request, but it addresses a different feature than #26350.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:26350", "right": "issue:33260", "accept": false, - "reason": "Different enhancement requests (Flash Attention 2 vs GGUF support); not the same underlying change or bug." + "reason": "Different underlying feature requests: Flash Attention 2 support vs GGUF support. They concern separate code paths and architectures, not the same bug or change." } ] }, @@ -19144,7 +19114,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-27088-8", + "cluster_id": "cluster-27088-9", "nodes": [ "issue:27088", "issue:28936", @@ -19153,7 +19123,8 @@ "issue:44365", "issue:44370", "issue:46222", - "issue:46226" + "issue:46226", + "issue:46931" ], "soft_pairs": [ "issue:27088|issue:44370", @@ -19174,194 +19145,229 @@ "issue:43834|issue:44365", "issue:43834|issue:46222", "issue:43834|issue:46226", + "issue:43834|issue:46931", "issue:44365|issue:44370", "issue:44365|issue:46222", "issue:44365|issue:46226", + "issue:44365|issue:46931", "issue:44370|issue:46222", "issue:44370|issue:46226", - "issue:46222|issue:46226" + "issue:44370|issue:46931", + "issue:46222|issue:46226", + "issue:46222|issue:46931", + "issue:46226|issue:46931" ], - "prepared_review_unit_hash": "1a401dd5788e7fb5821128907bd5c740e2fa01cf59b774f1a476c09d6211d8c3", + "prepared_review_unit_hash": "d0cf7f5ebf59053edb1fee650d26714b78066996893f8c602fa8503caef81443", "component_budget": { - "node_count": 8, - "item_count": 8, - "soft_pair_count": 24, - "serialized_chars": 13529, - "estimated_input_tokens": 3383, - "estimated_eval_tokens": 7022 + "node_count": 9, + "item_count": 9, + "soft_pair_count": 29, + "serialized_chars": 15983, + "estimated_input_tokens": 3996, + "estimated_eval_tokens": 8248 }, "budget": { - "node_count": 8, - "item_count": 8, - "soft_pair_count": 24, - "serialized_chars": 13529, - "estimated_input_tokens": 3383, - "estimated_eval_tokens": 7022 + "node_count": 9, + "item_count": 9, + "soft_pair_count": 29, + "serialized_chars": 15983, + "estimated_input_tokens": 3996, + "estimated_eval_tokens": 8248 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is not one true duplicate set: most issues are separate language-specific documentation translation requests, while only the repeated placeholder-template issues and the identical Thai pair look like real duplicates.", - "confidence": 0.77, - "canonical_issue_reason": "issue:43834 is the best representative of the duplicate-like subgroup because it is the earliest of the near-identical placeholder-title issues and sits in the densest duplicate neighborhood.", + "summary": "The cluster is heterogeneous overall: the Turkish, Spanish, and Romanian translation requests look like separate language-specific issues, while there are two clearer duplicate pockets inside it \u2014 the generic [i18n-] template trio and the identical Thai-titled pair. I would only collapse those pockets internally, not the full cluster.", + "confidence": 0.74, + "canonical_issue_reason": "issue:43834 is the best canonical within the generic placeholder/template pocket: it is the clearest representative and the earliest of the near-identical [i18n-] issues.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43834 is the strongest global representative for the cluster\u2019s duplicate core, but only for the placeholder-template subgroup, not for the distinct language-specific requests.", + "best_issue_reason": "As a standalone issue, issue:43834 is the strongest anchor for one duplicate pocket because it matches the near-identical template-style issues most closely; the language-specific translation requests should remain separate.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:27088", "right": "issue:44370", "accept": false, - "reason": "Different language request (Turkish vs generic placeholder template); similar body structure is not enough to treat as the same underlying issue." + "reason": "Different language-specific translation request; same broad theme, but not the same issue." }, { "left": "issue:27088", "right": "issue:44365", "accept": false, - "reason": "Turkish translation request is distinct from the generic placeholder-template issue." + "reason": "Different language-specific request; not a duplicate." }, { "left": "issue:27088", "right": "issue:43834", "accept": false, - "reason": "Different concrete translation request; same broad docs/i18n theme only." + "reason": "Turkish translation request vs generic template-style issue; different concrete request." }, { "left": "issue:27088", "right": "issue:38435", "accept": false, - "reason": "Turkish and Romanian translation requests are separate language tasks, not the same underlying issue." + "reason": "Separate language-specific translation requests." }, { "left": "issue:27088", "right": "issue:28936", "accept": false, - "reason": "Turkish and Spanish translation requests are distinct by target language." + "reason": "Turkish and Spanish translation requests are distinct." }, { "left": "issue:28936", "right": "issue:43834", "accept": false, - "reason": "Spanish translation request is not the same issue as the generic placeholder-template ticket." + "reason": "Spanish translation request is different from the generic placeholder/template issue." }, { "left": "issue:28936", "right": "issue:44365", "accept": false, - "reason": "Different target language; not a duplicate underlying change/request." + "reason": "Different language-specific request; not the same underlying issue." }, { "left": "issue:28936", "right": "issue:38435", "accept": false, - "reason": "Spanish and Romanian translation requests are separate items." + "reason": "Spanish vs Romanian translation requests are separate." }, { "left": "issue:28936", "right": "issue:44370", "accept": false, - "reason": "Spanish translation request does not match the generic placeholder-template issue." + "reason": "Different language-specific request; not a duplicate." }, { "left": "issue:38435", "right": "issue:44370", "accept": false, - "reason": "Romanian translation request is separate from the placeholder-template ticket." + "reason": "Romanian translation request is distinct from the template-style duplicate pocket." }, { "left": "issue:38435", "right": "issue:44365", "accept": false, - "reason": "Different language-specific request; same subsystem is too broad to merge." + "reason": "Different target language and separate request." }, { "left": "issue:38435", "right": "issue:43834", "accept": false, - "reason": "Romanian request and generic placeholder-template issue are not the same underlying issue." + "reason": "Romanian translation issue vs generic template-style issue; not the same request." }, { "left": "issue:38435", "right": "issue:46222", "accept": false, - "reason": "Romanian translation request is unrelated to the Thai-titled issue." + "reason": "Different topic and different apparent request; only superficial similarity." }, { "left": "issue:38435", "right": "issue:46226", "accept": false, - "reason": "Romanian translation request is unrelated to the Thai-titled issue." + "reason": "Different topic and different apparent request; not safely mergeable." }, { "left": "issue:43834", "right": "issue:44370", "accept": true, - "reason": "Same generic placeholder title and near-identical issue shape; these look like duplicated copies of the same template-based issue." + "reason": "Near-identical template-style i18n issue; same underlying request." }, { "left": "issue:43834", "right": "issue:44365", "accept": true, - "reason": "Same placeholder-template issue with no meaningful distinguishing information; likely duplicate." + "reason": "Near-identical template-style i18n issue; same underlying request." }, { "left": "issue:43834", "right": "issue:46222", "accept": false, - "reason": "Placeholder-template issue and Thai-text issue do not look like the same underlying request." + "reason": "Different title and apparent request; not the same duplicate pocket." }, { "left": "issue:43834", "right": "issue:46226", "accept": false, - "reason": "Placeholder-template issue and Thai-text issue are not the same concrete issue." + "reason": "Different title and apparent request; not a duplicate." + }, + { + "left": "issue:43834", + "right": "issue:46931", + "accept": false, + "reason": "Different Thai title and no clear shared underlying request." }, { "left": "issue:44365", "right": "issue:44370", "accept": true, - "reason": "Same placeholder title and extremely similar body profile; plausible duplicate of one template issue." + "reason": "Near-identical template-style i18n issue; same underlying request." }, { "left": "issue:44365", "right": "issue:46222", "accept": false, - "reason": "Generic placeholder-template issue vs Thai-text issue; not the same request." + "reason": "Different title/topic; similarity alone is not enough." }, { "left": "issue:44365", "right": "issue:46226", "accept": false, - "reason": "Generic placeholder-template issue vs Thai-text issue; not a duplicate." + "reason": "Different title/topic; not the same issue." + }, + { + "left": "issue:44365", + "right": "issue:46931", + "accept": false, + "reason": "Different apparent request and title." }, { "left": "issue:44370", "right": "issue:46222", "accept": false, - "reason": "Generic placeholder-template issue and Thai-text issue appear unrelated." + "reason": "Different title/topic; not the same duplicate pocket." }, { "left": "issue:44370", "right": "issue:46226", "accept": false, - "reason": "Generic placeholder-template issue and Thai-text issue appear unrelated." + "reason": "Different title/topic; not a duplicate." + }, + { + "left": "issue:44370", + "right": "issue:46931", + "accept": false, + "reason": "Different apparent request." }, { "left": "issue:46222", "right": "issue:46226", "accept": true, - "reason": "These have the same title and appear to be exact copies of the same issue." + "reason": "Identical Thai-titled issue with effectively identical content." + }, + { + "left": "issue:46222", + "right": "issue:46931", + "accept": false, + "reason": "Similar-looking body text is not enough; titles and apparent requests differ." + }, + { + "left": "issue:46226", + "right": "issue:46931", + "accept": false, + "reason": "Different issue title and no clear same underlying request." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary stays grounded in the packet: it distinguishes the clearly different language-specific translation requests from the near-identical placeholder-template issues and the identical Thai pair. The soft-edge verdicts are conservative and do not overclaim duplicates beyond what the titles/structure support." + "accept": false, + "feedback": "The broad split between the language-specific translation issues and the placeholder-like issues is grounded, but the writeup is a bit too assertive in places. In particular, the packet supports at least two duplicate-looking subgroups: the [i18n-] trio and the identical Thai-titled pair 46222/46226. So saying 43834 is the anchor for the \"only clearly duplicate subset\" is not accurate, and \"exact duplicate\"/\"spam\" language goes beyond what the packet alone proves. A more conservative summary would note multiple near-duplicate pockets without claiming exactness." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -19441,12 +19447,12 @@ "issue:42503|issue:46653", "issue:45027|issue:46653" ], - "prepared_review_unit_hash": "0927151adf127f16702dd2d443533d3dc2fd11e8886287482f7d722bff4b1a7e", + "prepared_review_unit_hash": "2ff544196ca0abe0f75a64aaf8f1bc340de03b75454ebdae721f34574ca7fd6d", "component_budget": { "node_count": 19, "item_count": 19, "soft_pair_count": 40, - "serialized_chars": 24676, + "serialized_chars": 24674, "estimated_input_tokens": 6169, "estimated_eval_tokens": 12594 }, @@ -19454,7 +19460,7 @@ "node_count": 19, "item_count": 19, "soft_pair_count": 40, - "serialized_chars": 24676, + "serialized_chars": 24674, "estimated_input_tokens": 6169, "estimated_eval_tokens": 12594 }, @@ -19463,252 +19469,252 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly a pile of unrelated model-addition requests. The similarity signal is driven by generic wording like \"Add/Support model\" rather than a shared underlying bug or code change, so none of the soft edges look mergeable as duplicates.", - "confidence": 0.97, - "canonical_issue_reason": "Issue 43671 is the best representative anchor here because it is a clear, active model-support request and sits near the densest soft-similarity area, but the overall set is too heterogeneous to justify true duplicate canonization.", + "summary": "This is a loose topic cluster of unrelated model-addition/support requests across text, vision, audio, and architectural changes. The soft links are mostly title-similarity false positives; none of the paired issues look like true duplicates.", + "confidence": 0.88, + "canonical_issue_reason": "Issue 43671 is the closest thing to a cluster center because it is a concrete, well-discussed support request with several adjacent audio/model soft matches. That said, the set is heterogeneous and there is no true duplicate canonical issue.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43671 is the most useful global representative: it is concrete, active, and thematically closest to the only obvious mini-cluster (Qwen3 voice support). Even so, it should not be treated as a duplicate of the other requests.", + "best_issue_reason": "43671 is the best representative of the overall pattern of 'add support for a specific model' requests. It has active discussion and is concrete enough to stand in for the cluster, even though the cluster itself is not a duplicate set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36718", "right": "issue:41115", "accept": false, - "reason": "Different requested artifacts: a RoMa keypoint matcher versus a MiniCPM3 model architecture. Shared \"add model\" phrasing only." + "reason": "RoMa keypoint matcher and MiniCPM3 architecture are different vision/model requests with no shared concrete code path." }, { "left": "issue:36817", "right": "issue:41211", "accept": false, - "reason": "Both are model/config requests, but EuroBert config and DEIMv2 are unrelated model additions with no shared code-path change." + "reason": "EuroBert config and DEIMv2 are unrelated model additions." }, { "left": "issue:36817", "right": "issue:43671", "accept": false, - "reason": "EuroBert config work is unrelated to Qwen3-TTS support; same broad request pattern, different model and modality." + "reason": "EuroBert config is a text-model config request; Qwen3-TTS is an audio support request." }, { "left": "issue:36817", "right": "issue:36879", "accept": false, - "reason": "EuroBert config and RF-DETR are distinct model-support requests with different targets and implementation work." + "reason": "EuroBert config and RF-DETR target different model families and changes." }, { "left": "issue:36879", "right": "issue:41211", "accept": false, - "reason": "RF-DETR and DEIMv2 are separate detection-model additions, not the same underlying bug or change." + "reason": "RF-DETR and DEIMv2 are distinct detector/model requests." }, { "left": "issue:36879", "right": "issue:43909", "accept": false, - "reason": "RF-DETR is a detection model; LFM2.5 Audio 1.5B is an audio model. Different feature requests." + "reason": "RF-DETR is a vision detector request; LFM2.5 Audio 1.5B is an audio model request." }, { "left": "issue:36879", "right": "issue:44121", "accept": false, - "reason": "RF-DETR and a weight-sparse transformer request are different model families and different implementation work." + "reason": "RF-DETR model support and weight-sparse transformer support are unrelated changes." }, { "left": "issue:36879", "right": "issue:43837", "accept": false, - "reason": "RF-DETR versus Qwen3-ASR are unrelated model-support requests across different domains." + "reason": "RF-DETR and Qwen3-ASR are different domains and code paths." }, { "left": "issue:36879", "right": "issue:43671", "accept": false, - "reason": "RF-DETR is a vision detector request, not the same change as Qwen3-TTS support." + "reason": "RF-DETR vision support and Qwen3-TTS audio support are unrelated requests." }, { "left": "issue:37015", "right": "issue:43837", "accept": false, - "reason": "NeoBERT and Qwen3-ASR are unrelated models; no shared code-path problem." + "reason": "NeoBERT and Qwen3-ASR are unrelated model requests." }, { "left": "issue:37015", "right": "issue:44121", "accept": false, - "reason": "NeoBERT support and a weight-sparse transformer are different model additions, not duplicates." + "reason": "NeoBERT addition and weight-sparse transformer support are different changes." }, { "left": "issue:37015", "right": "issue:43909", "accept": false, - "reason": "NeoBERT and LFM2.5 Audio 1.5B are unrelated requests." + "reason": "NeoBERT is a text model request; LFM2.5 Audio is an audio model request." }, { "left": "issue:36879", "right": "issue:37015", "accept": false, - "reason": "RF-DETR and NeoBERT are different model-support issues in different domains." + "reason": "RF-DETR and NeoBERT are unrelated model additions." }, { "left": "issue:37015", "right": "issue:42503", "accept": false, - "reason": "NeoBERT and ModernVBERT models are separate model-addition requests, not one concrete change." + "reason": "NeoBERT and ModernVBERT are different BERT-like model requests, not the same change." }, { "left": "issue:41115", "right": "issue:43837", "accept": false, - "reason": "MiniCPM3 architecture support and Qwen3-ASR support are different model requests." + "reason": "MiniCPM3 architecture and Qwen3-ASR support are different model domains." }, { "left": "issue:41115", "right": "issue:44121", "accept": false, - "reason": "MiniCPM3 architecture work is not the same as adding a weight-sparse transformer model." + "reason": "MiniCPM3 support and weight-sparse transformer support are distinct requests." }, { "left": "issue:41115", "right": "issue:42503", "accept": false, - "reason": "MiniCPM3 and ModernVBERT are distinct models with different integration work." + "reason": "MiniCPM3 and ModernVBERT are different architecture/model requests." }, { "left": "issue:41115", "right": "issue:45027", "accept": false, - "reason": "MiniCPM3 versus Voxtral-4B-TTS support are unrelated model requests." + "reason": "MiniCPM3 and Voxtral TTS target different model families and modalities." }, { "left": "issue:41211", "right": "issue:43671", "accept": false, - "reason": "DEIMv2 and Qwen3-TTS support are different model additions, not the same change." + "reason": "DEIMv2 and Qwen3-TTS are unrelated model-support requests." }, { "left": "issue:42503", "right": "issue:43837", "accept": false, - "reason": "ModernVBERT and Qwen3-ASR are separate model-support tasks." + "reason": "ModernVBERT models and Qwen3-ASR are unrelated." }, { "left": "issue:42503", "right": "issue:43671", "accept": false, - "reason": "ModernVBERT models are unrelated to Qwen3-TTS support." + "reason": "ModernVBERT and Qwen3-TTS are different model families/domains." }, { "left": "issue:42503", "right": "issue:42738", "accept": false, - "reason": "ModernVBERT and BERT-like models with RoPE are broadly related in wording only; they are not the same concrete request." + "reason": "Both mention BERT-like models, but one is a specific model set request and the other is a broad RoPE architecture request; not the same concrete change." }, { "left": "issue:42738", "right": "issue:45027", "accept": false, - "reason": "RoPE-related BERT-like models and Voxtral TTS support are unrelated." + "reason": "BERT-like models with RoPE and Voxtral TTS are unrelated requests." }, { "left": "issue:42738", "right": "issue:46653", "accept": false, - "reason": "The second item is not even a model request, so these cannot be duplicates." + "reason": "BERT-like model architecture request and the 'Mirko Privitera' issue are unrelated." }, { "left": "issue:43671", "right": "issue:43837", "accept": false, - "reason": "Both are Qwen3 support requests, but TTS and ASR are different modalities and would not plausibly merge into one PR." + "reason": "Qwen3-TTS and Qwen3-ASR are different concrete support requests, not one mergeable fix." }, { "left": "issue:43671", "right": "issue:45027", "accept": false, - "reason": "Qwen3-TTS and Voxtral-4B-TTS are different model integrations, not the same underlying change." + "reason": "Both are TTS-related, but they are different models and likely different implementations." }, { "left": "issue:43671", "right": "issue:45177", "accept": false, - "reason": "Qwen3-TTS support and DeepSeek-OCR2 are unrelated model requests." + "reason": "Qwen3-TTS and DeepSeek-OCR2 are different modalities and requests." }, { "left": "issue:43671", "right": "issue:44121", "accept": false, - "reason": "Qwen3-TTS support is not the same change as adding a weight-sparse transformer model." + "reason": "Qwen3-TTS support and weight-sparse transformer support are unrelated." }, { "left": "issue:43837", "right": "issue:43909", "accept": false, - "reason": "Qwen3-ASR and LFM2.5 Audio 1.5B are both audio-related, but they are different model support requests." + "reason": "Qwen3-ASR and LFM2.5 Audio are both speech-related, but they are distinct model requests." }, { "left": "issue:43837", "right": "issue:45322", "accept": false, - "reason": "Qwen3-ASR support and EUPE are unrelated model integrations." + "reason": "Qwen3-ASR and EUPE are different model families with no shared change." }, { "left": "issue:43837", "right": "issue:44121", "accept": false, - "reason": "Qwen3-ASR support is unrelated to the weight-sparse transformer request." + "reason": "Qwen3-ASR and weight-sparse transformer are unrelated changes." }, { "left": "issue:43909", "right": "issue:44205", "accept": false, - "reason": "LFM2.5 Audio 1.5B and SAM3-LiteText are different model families and domains." + "reason": "LFM2.5 Audio and SAM3-LiteText are different model-support requests." }, { "left": "issue:43909", "right": "issue:46645", "accept": false, - "reason": "Audio model support and a Granite SWA+sink option are not the same code-path change." + "reason": "Audio model support and Granite SWA+sink option are different code-path changes." }, { "left": "issue:44121", "right": "issue:45322", "accept": false, - "reason": "Weight-sparse transformer support and EUPE are separate model requests with no concrete overlap." + "reason": "Weight-sparse transformer support and EUPE are unrelated model requests." }, { "left": "issue:43837", "right": "issue:45027", "accept": false, - "reason": "Qwen3-ASR and Voxtral-4B-TTS are different speech-model support requests." + "reason": "Qwen3-ASR and Voxtral TTS are both speech-related, but they are not the same model or fix." }, { "left": "issue:37015", "right": "issue:45027", "accept": false, - "reason": "NeoBERT and Voxtral TTS support are unrelated requests." + "reason": "NeoBERT and Voxtral TTS are unrelated model requests." }, { "left": "issue:43837", "right": "issue:45177", "accept": false, - "reason": "Qwen3-ASR and DeepSeek-OCR2 are different model integrations." + "reason": "Qwen3-ASR and DeepSeek-OCR2 target different modalities." }, { "left": "issue:45322", "right": "issue:46133", "accept": false, - "reason": "EUPE and TIPSv2 are distinct vision-language model requests; same broad area, not the same change." + "reason": "EUPE and TIPSv2 are different vision-language models and different additions." }, { "left": "issue:42503", "right": "issue:46653", "accept": false, - "reason": "ModernVBERT models and \"Mirko Privitera\" are not even comparable as the same artifact." + "reason": "ModernVBERT models and a non-technical person-name issue are unrelated." }, { "left": "issue:45027", "right": "issue:46653", "accept": false, - "reason": "Voxtral-4B-TTS support and \"Mirko Privitera\" are unrelated; no duplicate relationship." + "reason": "Voxtral TTS support and a non-model person-name issue are unrelated." } ] }, @@ -19738,7 +19744,7 @@ "soft_pairs": [ "issue:38740|issue:46443" ], - "prepared_review_unit_hash": "3cdb9c76e66faae3acf6dd34157f4a7f78e469e1a970923dd8ab3257dd93b582", + "prepared_review_unit_hash": "b6982293cca2028da48ee209d8117de862593459a8daa3c7e71b41a90a566fef", "component_budget": { "node_count": 2, "item_count": 2, @@ -19760,18 +19766,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are unrelated: one is a docs request to add Pruna as an optimization framework, the other is a Vision RoPE refactor. There is no duplicate relationship or shared code path.", - "confidence": 0.98, - "canonical_issue_reason": null, + "summary": "These two issues are not duplicates: one is a docs request to add `pruna` as an optimization framework, the other is a Vision RoPE refactor. They target different areas and different kinds of work.", + "confidence": 0.99, + "canonical_issue_reason": "Issue 46443 is the better representative only in the sense that it is the more concrete engineering task; issue 38740 is a separate docs-oriented request. There is no shared underlying bug/change.", "canonical_pr_reason": null, - "best_issue_reason": "No canonical issue is appropriate here; the issues describe different topics and different parts of the project.", + "best_issue_reason": "Issue 46443 is the stronger standalone issue for clustering purposes because it describes a specific code refactor, while 38740 is an unrelated documentation enhancement.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:38740", "right": "issue:46443", "accept": false, - "reason": "Not the same underlying bug or change: one is a documentation request about Pruna, the other is a Vision RoPE refactor in model code." + "reason": "Different topics and work types: docs addition for `pruna` vs Vision RoPE refactor. No shared code path or underlying change." } ] }, @@ -19801,7 +19807,7 @@ "soft_pairs": [ "issue:41084|issue:43037" ], - "prepared_review_unit_hash": "31b3d78ee4bc5e7a83396d91ede3d3fff2e984a35754c6a68f9d5705070ae2af", + "prepared_review_unit_hash": "61f45c00cabeb3d0f26d98fc7ddfa1b6e16346699f6cad2ee807dd76d8cc1afe", "component_budget": { "node_count": 2, "item_count": 2, @@ -19823,18 +19829,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are not duplicates: one is about block decoding behavior, the other is about DeepSeek v3.2 model support.", + "summary": "The two issues are unrelated: one is about block decoding behavior, the other is a feature request for DeepSeek v3.2 support. They do not describe the same underlying bug or change.", "confidence": 0.98, - "canonical_issue_reason": "No canonical issue: the topics are unrelated and do not describe the same underlying bug or feature request.", + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "No single best issue for this cluster, because the pair is heterogeneous rather than a duplicate set.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41084", "right": "issue:43037", "accept": false, - "reason": "Different requests: block decoding vs. DeepSeek v3.2 support. They do not point to the same code-path or change." + "reason": "Different topics and likely different workstreams; no shared concrete bug, code path, or feature target." } ] }, @@ -19864,40 +19870,40 @@ "soft_pairs": [ "issue:41306|issue:41826" ], - "prepared_review_unit_hash": "3ba5a29fad91e31c7cd5dd19c6889b497cdce05790702486a446d0f61a583285", + "prepared_review_unit_hash": "cb986bf3e46a9b7f18d5effcc5fda71527082b42a0b92553b515896f1c592004", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1584, - "estimated_input_tokens": 396, - "estimated_eval_tokens": 1048 + "serialized_chars": 1585, + "estimated_input_tokens": 397, + "estimated_eval_tokens": 1050 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1584, - "estimated_input_tokens": 396, - "estimated_eval_tokens": 1048 + "serialized_chars": 1585, + "estimated_input_tokens": 397, + "estimated_eval_tokens": 1050 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates: one is a specific API/flag behavior request for `from_pretrained()` with TiledFusedLogitsLoss/FLCE, while the other is a separate feature request to integrate TiledMLP for lower memory usage.", - "confidence": 0.95, - "canonical_issue_reason": "Issue 41306 is the more concrete, narrowly scoped report and better represents the cluster than the separate TiledMLP feature request.", + "summary": "The two issues are about different features: one requests a `from_pretrained()` flag for `TiledFusedLogitsLoss`/FLCE, while the other asks to integrate `TiledMLP` for lower memory usage. They share the word \u201cTiled\u201d but not the same bug or change request.", + "confidence": 0.96, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "41306 is the best issue to surface because it describes a specific, actionable behavior change with a clear target API.", + "best_issue_reason": "Issue 41306 is the more concrete, narrowly scoped request and is the better representative of this cluster if one must be chosen, but it is not a duplicate of 41826.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41306", "right": "issue:41826", "accept": false, - "reason": "Different underlying requests: one concerns a FLCE flag in `from_pretrained()`, the other concerns integrating TiledMLP for memory savings. They do not appear to be the same bug or change." + "reason": "Different underlying changes: FLCE `from_pretrained()` flag support versus adding TiledMLP integration for memory reduction. No shared code path or same fix target is evident." } ] }, @@ -19927,12 +19933,12 @@ "soft_pairs": [ "issue:43598|issue:43600" ], - "prepared_review_unit_hash": "0549078a92ee910dc9ef85b71802bbf8f4046f0836e07e1641b8402ef0c7cb36", + "prepared_review_unit_hash": "b7801ea712b835a4d707614157fea621cee7378057f44167475e88160db76ca3", "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 }, @@ -19940,7 +19946,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1583, + "serialized_chars": 1584, "estimated_input_tokens": 396, "estimated_eval_tokens": 1048 }, @@ -19949,18 +19955,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are thematically related around custom trainers and Trainer behavior, but they describe different underlying problems: one is about `remove_unused_column` and customizability, the other is about FLOPS estimation for custom trainers. I would not cluster them as duplicates.", - "confidence": 0.86, - "canonical_issue_reason": "Issue 43598 is the better representative because it states a concrete Trainer behavior change (`remove_unused_column`) and a broader customizability concern, while the other issue is a separate FLOPS-estimation edge case.", + "summary": "These two closed issues are in the same Trainer/customization area but 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": 0.9, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "43598 is the more central and actionable issue in this pair; 43600 is narrower and concerns a different code path.", + "best_issue_reason": "No clear canonical issue here; if forced, 43598 is the more concrete code-path request, but it does not subsume 43600.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43598", "right": "issue:43600", "accept": false, - "reason": "Related theme, but not the same bug/change: one is about column removal/customization in Trainer, the other about FLOPS accounting for custom trainers. They could be fixed separately and would not naturally merge into one issue." + "reason": "They share a broad Trainer/custom trainer theme, but the underlying bugs are different: customizability of `remove_unused_column` vs FLOPS counting for custom trainers. They would not plausibly be one issue." } ] }, @@ -19990,12 +19996,12 @@ "soft_pairs": [ "issue:43684|issue:44405" ], - "prepared_review_unit_hash": "ccc0f405441b4d4f9fe12e5dabfafe05510b4637cda319e1d7c6b6a834be9540", + "prepared_review_unit_hash": "591f232131158e896b60c92bb058044a2dcabcdaa8e5342c8319f433a0947d7e", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1611, + "serialized_chars": 1612, "estimated_input_tokens": 403, "estimated_eval_tokens": 1062 }, @@ -20003,7 +20009,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1611, + "serialized_chars": 1612, "estimated_input_tokens": 403, "estimated_eval_tokens": 1062 }, @@ -20012,18 +20018,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are about different Qwen model families and different AutoModel entry points, so they do not look like duplicates.", - "confidence": 0.96, - "canonical_issue_reason": "Issue #43684 is the broader registration request for Qwen3-Omni across AutoModel and AutoModelForConditionalGeneration, so it is the more central issue in this pair.", + "summary": "These two are related only at a broad theme level (Qwen model support), but they target different models and different AutoModel heads, so they are not duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "Issue #43684 is the broader registration problem and is the better anchor for this cluster, but it still does not subsume #44405 because the latter is a separate Qwen3.5 sequence-classification support request.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #43684 is the better representative because it targets a core model-registration gap rather than a single task-specific head.", + "best_issue_reason": "#43684 is the closest thing to a canonical item because it addresses model registration at the AutoModel/AutoModelForConditionalGeneration level, which is a more general code-path than the specific head support asked for in #44405.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43684", "right": "issue:44405", "accept": false, - "reason": "Different underlying requests: Qwen3-Omni model registration for AutoModel/AutoModelForConditionalGeneration vs. AutoModelForSequenceClassification support for Qwen3.5. They concern different configs and different model heads, so they should not be merged as the same bug." + "reason": "Different underlying changes: #43684 asks for Qwen3-Omni registration in AutoModel and AutoModelForConditionalGeneration, while #44405 asks for AutoModelForSequenceClassification support for Qwen3.5/Qwen3_5Config. Same ecosystem, but not the same bug or mergeable fix." } ] }, @@ -20060,80 +20066,77 @@ "issue:43845|issue:43859", "issue:43846|issue:43859" ], - "prepared_review_unit_hash": "a2a96ad7f6b8a935fe151f73e45175f342d4388cf7fa7d412839d3455ba059f8", + "prepared_review_unit_hash": "1d7cb5e0889869cd1d78a2057de4aceaebe321eca79e572d030f992542eae1a0", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4266, - "estimated_input_tokens": 1067, - "estimated_eval_tokens": 2390 + "serialized_chars": 4262, + "estimated_input_tokens": 1066, + "estimated_eval_tokens": 2388 }, "budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4266, - "estimated_input_tokens": 1067, - "estimated_eval_tokens": 2390 + "serialized_chars": 4262, + "estimated_input_tokens": 1066, + "estimated_eval_tokens": 2388 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Three issues share the same Hugging Face title and look like duplicate placeholder reports; the \"Fett\" issue appears unrelated.", - "confidence": 0.66, - "canonical_issue_reason": "issue #43845 is the earliest of the near-identical Hugging Face issues and is the best stand-in for the duplicate set.", + "summary": "The cluster is mostly low-information issue spam/duplicates with two title groups: one unrelated \"Fett\" issue and three near-identical \"huggingface\" issues. There isn\u2019t enough concrete content to safely merge any pair as the same underlying bug.", + "confidence": 0.42, + "canonical_issue_reason": "If this cluster were deduped, issue 43845 is the best representative because it is the earliest of the repeated \"huggingface\" reports and appears to be the central duplicate target among the similar titles.", "canonical_pr_reason": null, - "best_issue_reason": "issue #43845 is the cleanest representative of the repeated Hugging Face report cluster; the other matching issues can fold into it.", + "best_issue_reason": "43845 is the best overall issue to keep as the representative because it is the earliest repeated \"huggingface\" entry and has the strongest similarity to the other repeated title issues.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43835", "right": "issue:43845", "accept": false, - "reason": "\"Fett\" does not match the Hugging Face report and there is no evidence they describe the same issue." + "reason": "Titles are unrelated and there is no body/detail evidence tying \"Fett\" to the \"huggingface\" issues." }, { "left": "issue:43835", "right": "issue:43846", "accept": false, - "reason": "\"Fett\" looks unrelated to the Hugging Face issue cluster." + "reason": "Titles are unrelated and there is no body/detail evidence tying \"Fett\" to the \"huggingface\" issues." }, { "left": "issue:43835", "right": "issue:43859", "accept": false, - "reason": "No shared topic beyond generic similarity; this does not look like the same underlying issue." + "reason": "Titles are unrelated and there is no body/detail evidence tying \"Fett\" to the \"huggingface\" issues." }, { "left": "issue:43845", "right": "issue:43846", - "accept": true, - "reason": "Same title and very similar metadata/body size; these look like duplicate reports of the same Hugging Face issue." + "accept": false, + "reason": "The titles match closely, but the bodies are not available and there is not enough concrete evidence to confirm the same underlying bug." }, { "left": "issue:43845", "right": "issue:43859", - "accept": true, - "reason": "All are the same Hugging Face report topic, and this pair appears to be duplicate issue filing." + "accept": false, + "reason": "The titles match closely, but the bodies are not available and there is not enough concrete evidence to confirm the same underlying bug." }, { "left": "issue:43846", "right": "issue:43859", - "accept": true, - "reason": "Same repeated Hugging Face issue title and closely related report content; likely duplicates." + "accept": false, + "reason": "The titles match closely, but the bodies are not available and there is not enough concrete evidence to confirm the same underlying bug." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is supported by the packet: 43845/43846/43859 share the same title and have relatively high soft-similarity, while 43835 is notably less similar and looks separate. The canonical-choice rationale is also grounded in creation order. Soft-edge verdicts are conservative enough overall." - }, - "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 }, @@ -20167,7 +20170,7 @@ "issue:44572|issue:44937", "issue:44573|issue:44937" ], - "prepared_review_unit_hash": "c08bac667a3ecb1bbee2ab87c35add7f46ee6f6c5b4ce93827498346d48e5c92", + "prepared_review_unit_hash": "8b2cfcc8eb592d2c3d219eaf806910440f3ad3993982b0e26ebc3ca6513be7a7", "component_budget": { "node_count": 5, "item_count": 5, @@ -20189,66 +20192,66 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster looks like low-signal spam / unrelated placeholder issues, with no evidence of a shared underlying bug or change. The similar titles are not enough to treat them as duplicates.", - "confidence": 0.96, - "canonical_issue_reason": null, + "summary": "This cluster looks like low-signal noise rather than one real duplicated bug/change set: three generic spam issues, one documentation-template issue, and one unrelated promotional title. No pair has concrete evidence of the same underlying problem.", + "confidence": 0.91, + "canonical_issue_reason": "issue:44371 is the cleanest representative of the cluster because it is the earliest and most generic \"\" entry, with no extra wording tying it to a different topic.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44418 is the least obviously spammy and the only one that reads like a real user-facing report, but there is still no evidence it is a duplicate of the others.", + "best_issue_reason": "issue:44371 is the best global issue representative because it is the most neutral placeholder/noise item and does not introduce a separate topic like documentation or promotion.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44371", "right": "issue:44572", "accept": false, - "reason": "Both are just '' placeholders with no body or targets; similarity is not enough to infer the same bug." + "reason": "Both are spam-like placeholders, but there is no concrete bug report or shared code-path/change to justify treating them as the same underlying issue." }, { "left": "issue:44371", "right": "issue:44573", "accept": false, - "reason": "Both are generic spam placeholders, but there is no concrete issue content tying them to the same underlying problem." + "reason": "Same generic spam title is not enough; these could be separate spam submissions and there is no substantive issue content to merge." }, { "left": "issue:44371", "right": "issue:44418", "accept": false, - "reason": "One is spam and the other is a documentation suggestion; the titles are not evidence of the same issue." + "reason": "\"\" and a documentation enhancement suggestion are clearly different topics with no evidence of a shared bug or fix." }, { "left": "issue:44371", "right": "issue:44937", "accept": false, - "reason": "A spam placeholder and a titled issue about 'Google LLC' are not the same underlying bug or change." + "reason": "A spam placeholder and a promotional/title-like issue do not indicate the same underlying problem." }, { "left": "issue:44418", "right": "issue:44572", "accept": false, - "reason": "Documentation enhancement and spam placeholder are not a plausible duplicate pair." + "reason": "Documentation enhancement versus spam placeholder: unrelated and not the same underlying change request." }, { "left": "issue:44418", "right": "issue:44573", "accept": false, - "reason": "No concrete overlap beyond weak text similarity; these do not look like the same issue." + "reason": "Documentation enhancement versus spam placeholder: no shared bug, feature, or target area." }, { "left": "issue:44572", "right": "issue:44573", "accept": false, - "reason": "Both are spam placeholders, but there is no substantive evidence they report the same problem." + "reason": "Although the titles match, there is no actual issue detail showing they describe the same bug or change; they are just duplicate spam-like submissions." }, { "left": "issue:44572", "right": "issue:44937", "accept": false, - "reason": "A spam placeholder and a titled issue about 'Google LLC' are unrelated." + "reason": "Spam placeholder and unrelated promotional title do not describe a common underlying issue." }, { "left": "issue:44573", "right": "issue:44937", "accept": false, - "reason": "The titles and lack of body details do not support treating these as duplicates." + "reason": "Spam placeholder and unrelated promotional title are not the same concrete problem." } ] }, @@ -20278,12 +20281,12 @@ "soft_pairs": [ "issue:45145|issue:45151" ], - "prepared_review_unit_hash": "41cc2a898d4f3def52174adb836e789492e43e0df040d18800ee3d85791aad0e", + "prepared_review_unit_hash": "8befd5f326187f06668bb0b83440dab5fd92458b168429a4030783631d9f03f9", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1645, + "serialized_chars": 1646, "estimated_input_tokens": 412, "estimated_eval_tokens": 1080 }, @@ -20291,7 +20294,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1645, + "serialized_chars": 1646, "estimated_input_tokens": 412, "estimated_eval_tokens": 1080 }, @@ -20300,28 +20303,28 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two closed issues appear to be duplicate reports of the same energy-reduction proposal, with matching titles and very similar content.", - "confidence": 0.98, - "canonical_issue_reason": "Issue #45145 is the better canonical issue because it was created first and already has an inbound reference, so it is slightly more established as the primary thread.", + "summary": "The two closed issues have identical titles and very similar metadata, so #45145 is the best representative and likely duplicate target. However, the packet does not expose the bodies, so the soft match is not strong enough to confirm they are definitively the same issue.", + "confidence": 0.84, + "canonical_issue_reason": "#45145 is earlier and has the same title as #45151, making it the safest canonical duplicate target among the two.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #45145 is the best issue representative for this cluster for the same reasons: earliest creation and a reference signal, while #45151 appears to be a near-identical duplicate.", + "best_issue_reason": "#45145 is the best issue representative because it is the earliest report and the visible metadata strongly suggests the same topic as #45151.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45145", "right": "issue:45151", - "accept": true, - "reason": "Same title, same apparent topic, and highly similar body size/content indicate the same underlying issue rather than merely related reports." + "accept": false, + "reason": "The evidence only shows identical titles, similar body lengths, and a soft-similarity score. Without the issue bodies or clearer scope/goal evidence, this should be treated as a likely duplicate rather than confirmed as the same underlying issue." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The analyst output is grounded in the packet: the two issues share the exact title, have very similar body lengths, and the packet provides soft_similarity evidence. The canonical/best-issue reasoning is conservative and supported by the earlier creation time plus the inbound reference on #45145. No unsupported mergeability claims are made." + "accept": false, + "feedback": "The canonical/representative choice is grounded by chronology and identical titles, but the soft-edge acceptance overstates the evidence. The packet only shows identical titles, close body lengths, and a soft-similarity signal; it does not expose the issue bodies or confirm that the scope/goal is the same. So this should be phrased as a likely duplicate or potential duplicate, not accepted as definitively the same issue." }, "evaluator_used": true, - "retried": false, - "accepted_nontrivial_soft_edge": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -20398,7 +20401,7 @@ "soft_pairs": [ "issue:45762|issue:45763" ], - "prepared_review_unit_hash": "d4155d6c0f7c2028b26ea423ee47ae1700d44f1845642cd94b2bc8c4ac76ee83", + "prepared_review_unit_hash": "d645e636d0c8d984fc2a68d0d9e90d3df7c4d510d989162153a46e2ae228ce6d", "component_budget": { "node_count": 2, "item_count": 2, @@ -20420,18 +20423,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two closed issues are not duplicates; they are unrelated, title-only posts that appear to be personal names/dates rather than the same bug or request.", - "confidence": 0.98, + "summary": "These are two empty, contentless issues that look like separate spam/placeholder posts, not duplicates of a real bug report or change request.", + "confidence": 0.97, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Neither issue is a good canonical candidate: both have no body, no references, and no actionable problem description.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45762", "right": "issue:45763", "accept": false, - "reason": "No evidence of a shared bug, request, or code path; the titles are different names with different dates and there is no body or linkage tying them together." + "reason": "The only similarity is the surname pattern in the titles; there is no shared bug, repro, subsystem, or issue target indicating the same underlying problem." } ] }, @@ -20461,40 +20464,40 @@ "soft_pairs": [ "issue:45764|issue:45765" ], - "prepared_review_unit_hash": "c81581221d08c355e208dbce7ab99bdfb342c0a088cbdcdd915ab2972c665334", + "prepared_review_unit_hash": "68c550abe73c04bea659dc3dbac54bac9d8aec6808af5c7b777d79d992f61a18", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1481, - "estimated_input_tokens": 371, - "estimated_eval_tokens": 998 + "serialized_chars": 1480, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1481, - "estimated_input_tokens": 371, - "estimated_eval_tokens": 998 + "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 closed issues appear to be unrelated, bare-name entries with different people and dates; there is no indication of a shared bug, feature, or code change.", + "summary": "The two closed issues are just title-only personal-name/date entries with empty bodies and no shared context, so they do not appear to describe the same underlying bug or change.", "confidence": 0.98, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a good canonical candidate because both are similarly content-free and do not describe an actionable product problem.", + "best_issue_reason": "No strong canonical issue exists here; if forced to pick one anchor, issue 45764 is only marginally earlier, but the pair does not look duplicate.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45764", "right": "issue:45765", "accept": false, - "reason": "These are just two different personal-name/date entries with no shared technical subject, so they do not look like duplicate reports of the same underlying issue." + "reason": "The titles are unrelated personal names with dates, both bodies are empty, and there is no shared target or code-context suggesting the same underlying issue." } ] }, @@ -20524,7 +20527,7 @@ "soft_pairs": [ "issue:45810|issue:46295" ], - "prepared_review_unit_hash": "4baa76b9aed350d90ff64f26f96ba08b8bfcb20cbb8cd3f42b194070f0d0c3d3", + "prepared_review_unit_hash": "2acacd0df53ef27bc1f27237e705a628803a680b9150c2e34bbb6a509101da4b", "component_budget": { "node_count": 2, "item_count": 2, @@ -20546,18 +20549,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are unrelated feature requests: one asks for Qwen3_5 token classification support, the other asks for FP8 training support. They should not be clustered as duplicates.", + "summary": "The two issues are unrelated: one requests a new token-classification/value-model class for Qwen3_5, while the other asks for FP8 training support. No duplicate relationship is present.", "confidence": 0.98, - "canonical_issue_reason": "Issue 45810 is the more concrete and actionable request, with a specific model class named in the title.", + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue 45810 is the best representative because it is narrowly scoped and clearly describes the missing functionality.", + "best_issue_reason": "Neither issue is a duplicate of the other; if one had to be prioritized as a standalone triage item, 46295 is the broader infrastructure request, while 45810 is a specific model-class feature request.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45810", "right": "issue:46295", "accept": false, - "reason": "Different feature areas and code paths: model-class support versus FP8 training support. They do not describe the same underlying bug or change." + "reason": "Different underlying requests and code paths: model-class addition for token classification vs FP8 training support. They do not describe the same bug or change." } ] }, @@ -20597,88 +20600,88 @@ "issue:45871|issue:45873", "issue:45872|issue:45873" ], - "prepared_review_unit_hash": "ffcdecb03bf3e0c570233bb2220bbc9a49b883508a690d480a621d79264b82ac", + "prepared_review_unit_hash": "27dae74efab0f3b559fccc88d47e8c46b714c83f79be748c0bb702378e6b048e", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5726, - "estimated_input_tokens": 1432, - "estimated_eval_tokens": 3120 + "serialized_chars": 5716, + "estimated_input_tokens": 1429, + "estimated_eval_tokens": 3114 }, "budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5726, - "estimated_input_tokens": 1432, - "estimated_eval_tokens": 3120 + "serialized_chars": 5716, + "estimated_input_tokens": 1429, + "estimated_eval_tokens": 3114 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This is not one duplicate cluster but two separate exact-duplicate pairs: 45870/45871 and 45872/45873. Issue 45869 is a related dataset marketing/request item but not the same underlying ask as the others.", + "summary": "Two exact duplicate issue pairs stand out: 45870\u219445871 and 45872\u219445873. Issue 45869 is a separate dataset listing and should not be merged with the example/service requests.", "confidence": 0.97, - "canonical_issue_reason": "45872 is the best single representative if one must be chosen: it has an exact duplicate (45873) and is the most concrete, actionable DataForge request. The other exact-dupe pair (45870/45871) is a separate issue, and 45869 is a different dataset-bundle request.", + "canonical_issue_reason": "No single canonical issue for the whole set; there are two distinct duplicate groups, while 45869 is unrelated.", "canonical_pr_reason": null, - "best_issue_reason": "45872 is the strongest anchor for this set because it is an exact duplicate pair with a more specific feature-oriented title ('financial time series + scraping service').", + "best_issue_reason": "45872 is the strongest representative because it is specific and actionable, and it has an exact duplicate in 45873.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45869", "right": "issue:45870", "accept": false, - "reason": "Both mention financial data/time-series themes, but one is a dataset bundle offer and the other is a DataForge example request; not the same underlying issue." + "reason": "Different requests: a financial dataset bundle vs an example about time series prediction." }, { "left": "issue:45869", "right": "issue:45871", "accept": false, - "reason": "Same as 45869/45870: related topic, but different artifact type and request scope." + "reason": "Different requests: a financial dataset bundle vs an example about time series prediction." }, { "left": "issue:45870", "right": "issue:45871", "accept": true, - "reason": "Exact title match and same apparent example request; these look like duplicate issues." + "reason": "Exact same title and clearly the same example request." }, { "left": "issue:45870", "right": "issue:45872", "accept": false, - "reason": "Both are DataForge/financial-time-series themed, but 45870 is an example while 45872 asks for a scraping service; not the same concrete change." + "reason": "Both are finance-related, but one is an example request and the other is a time-series plus scraping-service request." }, { "left": "issue:45870", "right": "issue:45873", "accept": false, - "reason": "Same reasoning as 45870/45872; related subject, but different underlying request." + "reason": "Both are finance-related, but one is an example request and the other is a time-series plus scraping-service request." }, { "left": "issue:45871", "right": "issue:45872", "accept": false, - "reason": "Exact duplicate of 45870, but still a different request from the scraping-service issue." + "reason": "Both are finance-related, but one is an example request and the other is a time-series plus scraping-service request." }, { "left": "issue:45871", "right": "issue:45873", "accept": false, - "reason": "Different underlying request despite overlapping domain terminology." + "reason": "Both are finance-related, but one is an example request and the other is a time-series plus scraping-service request." }, { "left": "issue:45872", "right": "issue:45873", "accept": true, - "reason": "Exact title match and same feature request; these are clear duplicates." + "reason": "Exact same title and clearly the same feature request." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet: 45870/45871 and 45872/45873 are exact title matches, while 45869 is clearly a different dataset-bundle request. The non-duplicate verdicts are appropriately conservative given only soft similarity evidence." + "feedback": "The summary is well grounded in the packet: 45870\u219445871 and 45872\u219445873 are supported as duplicate pairs by identical titles/body-length matches and perfect similarity, and 45869 is clearly a separate dataset-style issue. The soft-edge verdicts are also conservative, with no overreach beyond the evidence provided." }, "evaluator_used": true, "retried": false, @@ -20705,40 +20708,40 @@ "soft_pairs": [ "issue:46439|issue:46441" ], - "prepared_review_unit_hash": "e0011f7dd5e2206b7cb3b2e4207ae2ccadff8fbb842738e7fe711dfdcde82118", + "prepared_review_unit_hash": "90fa1c913ccc2ee65b31dbb6f489ea7cd497fb916cef1df20b0d003fc2f5d7b8", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1636, - "estimated_input_tokens": 409, - "estimated_eval_tokens": 1074 + "serialized_chars": 1637, + "estimated_input_tokens": 410, + "estimated_eval_tokens": 1076 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1636, - "estimated_input_tokens": 409, - "estimated_eval_tokens": 1074 + "serialized_chars": 1637, + "estimated_input_tokens": 410, + "estimated_eval_tokens": 1076 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are related to static-cache handling for linear attention, but the reported failures differ enough that they should not be merged as duplicates without more evidence.", - "confidence": 0.66, - "canonical_issue_reason": "Issue 46439 is the more general/root-cause-looking report: it describes static-cache early initialization corrupting linear-attention layers and producing shape errors, which could encompass narrower crash variants.", + "summary": "Both issues involve `cache_implementation=\"static\"` and `linear_attention`, but they describe different failure modes: one is a shape corruption / `RuntimeError` during early initialization, the other is a `KeyError: 'linear_attention'` in Qwen3.5. I would not merge them as duplicates without stronger evidence of the same code path.", + "confidence": 0.78, + "canonical_issue_reason": "Issue 46439 is the better canonical issue because it describes the broader, more generic failure mode around static-cache early initialization affecting linear-attention layers, which can subsume narrower model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 46439 is the best representative because it names the suspected mechanism and the broader failure mode, making it more useful as the primary tracking issue.", + "best_issue_reason": "Issue 46439 is the best representative of the cluster because it states the underlying bug at the subsystem level rather than a single-model symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46439", "right": "issue:46441", "accept": false, - "reason": "Both mention static cache and linear attention, but one reports shape corruption during early initialization while the other reports a Qwen3.5 KeyError for 'linear_attention'; they may be related, but they do not clearly describe the same concrete bug." + "reason": "Related area, but not clearly the same bug: one report is a shape mismatch/RuntimeError from early initialization, the other is a KeyError for `linear_attention` in Qwen3.5." } ] }, @@ -20768,7 +20771,7 @@ "soft_pairs": [ "issue:46551|issue:46552" ], - "prepared_review_unit_hash": "ae20f63216fbbe7cb3a90b3cb740641f9051c6eec93da73cd39884d1aa704961", + "prepared_review_unit_hash": "c5346631926601bd1aa98b04cf68fb38b7d5c538f9e4f2bc7c5731d2d1531f31", "component_budget": { "node_count": 2, "item_count": 2, @@ -20790,27 +20793,27 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues describe the same Wav2Vec2CTCTokenizer v5 regression verbatim. #46552 is the better canonical tracker because it is open and already has inbound references; #46551 appears to be an earlier overlapping report, but the packet does not explicitly confirm the closure reason.", - "confidence": 0.96, - "canonical_issue_reason": "#46552 is the better canonical issue because it is still open and has inbound references, while #46551 is the same bug report verbatim and appears to be the older overlapping record.", + "summary": "These two issues are effectively the same report: identical title and body about Wav2Vec2CTCTokenizer v5 incorrectly treating the word delimiter as a special token. The open issue is the better canonical representative.", + "confidence": 0.99, + "canonical_issue_reason": "Issue #46552 is the better canonical issue because it is open and already has inbound references, while #46551 is closed and appears to be the earlier duplicate copy.", "canonical_pr_reason": null, - "best_issue_reason": "#46552 is the best issue to keep as the primary tracker: it is open, already referenced, and matches #46551\u2019s title/body exactly.", + "best_issue_reason": "#46552 is the best issue to keep because it is active, open, and has the stronger likelihood of being the one that should collect discussion and fixes.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46551", "right": "issue:46552", "accept": true, - "reason": "Same exact title and same underlying tokenizer bug description; they are effectively the same issue, with no contrary evidence in the packet." + "reason": "Same title, same body, and same underlying tokenizer bug; this is a clear duplicate pair." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The pairing is well grounded as a strong duplicate-like match, but the writeup overstates the evidence by calling #46551 a \"closed duplicate.\" The packet only shows that #46551 is closed and #46552 is open with inbound references; it does not explicitly prove the closure reason. Keep the canonical recommendation, but phrase the duplicate/closure status more conservatively." + "accept": true, + "feedback": "The summary is well grounded in the packet: the two issues have identical titles and body lengths, and the open issue has inbound references, so treating #46552 as the canonical representative is reasonable. The soft-edge duplicate verdict is conservative and supported by the evidence available." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -20834,7 +20837,7 @@ "soft_pairs": [ "issue:46657|issue:46692" ], - "prepared_review_unit_hash": "a5ac32533ba3db2c6aac740cfed3a70322dabd37c21ebf3e2df84cac217200a6", + "prepared_review_unit_hash": "e3f96b716482df0cb0cd54cfb8100cad1a436cefef971ef37219c7e3563944b7", "component_budget": { "node_count": 2, "item_count": 2, @@ -20856,18 +20859,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two separate training issues in the same general area, but they describe different failure modes: one is a tensor shape mismatch during object detection training, the other is an out-of-memory problem after several epochs with RT-DETR v2. I would not merge them.", - "confidence": 0.94, - "canonical_issue_reason": "Issue #46657 is the more concrete bug report: it names a specific failure mode (tensor size mismatch) and a broader object-detection training context, making it a better representative of the cluster if forced to choose one issue.", + "summary": "These two issues are both about training object detection models, but they describe different failure modes: one is a tensor shape mismatch, the other is an out-of-memory problem occurring after several epochs in RT-DETR v2. They do not look like the same underlying bug.", + "confidence": 0.93, + "canonical_issue_reason": "Issue #46657 is the clearer standalone bug report: it names a specific runtime error (tensor size mismatch) and is more directly actionable than the broader OOM report.", "canonical_pr_reason": null, - "best_issue_reason": "#46657 is the best issue candidate because it describes an actionable code-path problem rather than a resource-pressure symptom, and its title is more likely to map to a single underlying defect.", + "best_issue_reason": "Issue #46657 is the best representative issue for the cluster because it is more concrete and specific, whereas #46692 appears to be a separate training stability/resource problem.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46657", "right": "issue:46692", "accept": false, - "reason": "They share a training-related theme, but the underlying problems differ: tensor dimension mismatch versus OOM during later epochs. These are not the same bug and would not plausibly be fixed by one change." + "reason": "The reports involve different concrete problems and failure signatures: tensor dimension mismatch vs memory exhaustion after training for some epochs. Shared object-detection context is too broad to treat as one bug." } ] }, @@ -20897,7 +20900,7 @@ "soft_pairs": [ "issue:46822|issue:46823" ], - "prepared_review_unit_hash": "6d78b03b27273ebd16d91ce5482824b1cdebe6c83a90e4df6536f20bbda4753b", + "prepared_review_unit_hash": "24e09325fbea200ff8ee3f472011ce1b45f117d97a8de040b934e330ab82bc77", "component_budget": { "node_count": 2, "item_count": 2, @@ -20919,18 +20922,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are both about model loading/device placement, but they describe different failure modes in different code paths, so they should not be clustered as duplicates.", - "confidence": 0.95, - "canonical_issue_reason": "Issue 46822 is the stronger anchor: it has more discussion and inbound references, and it describes a concrete ZeRO-3 zero.Init multi-GPU OOM scenario with a specific composite model submodule.", + "summary": "These two issues both mention model-loading and memory, but they describe different failure modes in different code paths: ZeRO-3 composite module partitioning vs 4-bit device_map/max_memory dispatch. They should not be clustered as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": "No clear canonical issue for a duplicate cluster here; the reports target different bugs and should remain separate.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 46822 is the best representative issue for the cluster because it is the more developed report and the symptom/cause are more specific and actionable.", + "best_issue_reason": "Issue 46823 is the more self-contained bug report with a clear repro and concrete expected/actual behavior, but it is not a duplicate of 46822.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46822", "right": "issue:46823", "accept": false, - "reason": "Both involve loading models onto devices, but 46822 is a ZeRO-3 partitioning/OOM bug during zero.Init on a composite submodule, while 46823 is a 4-bit + device_map/max_memory routing bug that incorrectly dispatches to CPU and raises ValueError. Different mechanisms, symptoms, and likely fixes." + "reason": "Different underlying bugs: ZeRO-3 zero.Init failing to partition a composite submodule and causing OOM versus 4-bit device_map='auto' with max_memory incorrectly dispatching modules to CPU and raising ValueError. They do not look mergeable into one fix." } ] }, @@ -20960,7 +20963,7 @@ "soft_pairs": [ "pull_request:40546|pull_request:42781" ], - "prepared_review_unit_hash": "cdd9e8ed7426e434ef423c6bb012b3959945c1c4e4d6dafb66c294041f567d27", + "prepared_review_unit_hash": "dc968de77a880f1bc6352046630b4b135c7ea09a4a1e1c76cb94cbe9d0121330", "component_budget": { "node_count": 2, "item_count": 2, @@ -20982,18 +20985,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open PRs target the VibeVoice family, but they represent different layers of work: the first is the initial VibeVoice integration, while the second adds VibeVoice Realtime and related model/processor changes.", - "confidence": 0.91, + "summary": "Both pull requests are about VibeVoice, but they are not the same change: PR 40546 adds the base VibeVoice model integration, while PR 42781 adds a separate realtime variant plus acoustic tokenizer work and generation changes. They overlap on shared architecture files, but the concrete scope is different enough that they should not be deduplicated.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #40546 is the better canonical PR because it is the broader, foundational VibeVoice implementation and has substantially more review activity, making it the primary representative of this feature set.", + "canonical_pr_reason": "PR 40546 is the better canonical PR for the cluster because it is the foundational VibeVoice implementation and broader in scope, while PR 42781 looks like a follow-on feature branch for realtime support.", "best_issue_reason": null, - "best_pr_reason": "PR #40546 is the best overall representative for the cluster since it introduces the main VibeVoice support; #42781 is a later, more specialized follow-on for realtime functionality rather than the same mergeable change.", + "best_pr_reason": "PR 40546 best represents the underlying VibeVoice integration work; PR 42781 is a narrower extension for realtime functionality and tokenizer support, not the same mergeable change.", "soft_edge_verdicts": [ { "left": "pull_request:40546", "right": "pull_request:42781", "accept": false, - "reason": "Reject: despite overlapping VibeVoice files, #40546 is the base VibeVoice implementation and #42781 is a separate VibeVoice Realtime extension with additional feature-extractor/processing and generation changes. They are related, but not the same concrete code-path fix/change and would not plausibly be merged as one PR." + "reason": "Shared VibeVoice files indicate related work, but 40546 is the base implementation and 42781 is a distinct realtime/tokenizer extension with additional generation and auto-mapping changes. They are related, not duplicate fixes or the same concrete PR." } ] }, @@ -21023,12 +21026,12 @@ "soft_pairs": [ "pull_request:40755|pull_request:41224" ], - "prepared_review_unit_hash": "c9d06f029a67398b6900ae3f9ca79a82b46b9389ade3a6fd44591ac7d500d1ef", + "prepared_review_unit_hash": "75b7127fca093b69f4410becb91f66026c512e0bebf9de38974ddb0da194ebff", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4676, + "serialized_chars": 4675, "estimated_input_tokens": 1169, "estimated_eval_tokens": 2594 }, @@ -21036,7 +21039,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4676, + "serialized_chars": 4675, "estimated_input_tokens": 1169, "estimated_eval_tokens": 2594 }, @@ -21045,18 +21048,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The cluster contains two unrelated pull requests: one adds TimesFM covariate forecasting, the other adds DINOv3 image-classification support. They do not appear to fix the same underlying bug or implement the same change.", - "confidence": 0.97, + "summary": "These PRs are unrelated: one adds TimesFM covariate forecasting, the other adds DINOv3 image classification support. They do not fix the same bug or implement the same change, so they should not be clustered as duplicates.", + "confidence": 0.99, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "No clear canonical PR: the two pull requests target different models and different features, so neither serves as a duplicate representative for the cluster.", "best_issue_reason": null, - "best_pr_reason": null, + "best_pr_reason": "Neither PR is a good global canonical choice because they address separate code paths and independent feature additions.", "soft_edge_verdicts": [ { "left": "pull_request:40755", "right": "pull_request:41224", "accept": false, - "reason": "Different model families and different feature areas; TimesFM covariate forecasting is unrelated to DINOv3 image-classification support." + "reason": "Different subsystems and different changes: TimesFM forecasting-with-covariates versus DINOv3 image classification support. They do not appear to be the same underlying bug or mergeable change." } ] }, @@ -21187,208 +21190,196 @@ "pull_request:43823|pull_request:46911", "pull_request:43251|pull_request:43254" ], - "prepared_review_unit_hash": "474094e2db7289b57fa7fce08a3971ef3abdddc72fe56555d6e0ce9b5f84477b", + "prepared_review_unit_hash": "d4d2dab884c1b13a28137e4aa864ac7c7f8b94fe9941e075c3021a118cfbd1a6", "component_budget": { "node_count": 15, "item_count": 15, "soft_pair_count": 26, - "serialized_chars": 42243, - "estimated_input_tokens": 10561, - "estimated_eval_tokens": 21378 + "serialized_chars": 42250, + "estimated_input_tokens": 10563, + "estimated_eval_tokens": 21382 }, "budget": { "node_count": 15, "item_count": 15, "soft_pair_count": 26, - "serialized_chars": 42243, - "estimated_input_tokens": 10561, - "estimated_eval_tokens": 21378 + "serialized_chars": 42250, + "estimated_input_tokens": 10563, + "estimated_eval_tokens": 21382 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly a grab bag of unrelated PRs. The only true duplicate-like pair is the two `fixed_cross_entropy` kwargs fixes; everything else targets different models, utilities, or tests and should stay separate.", - "confidence": 0.97, + "summary": "This cluster is mostly noise from loose topical similarity. The only clear duplicate is the pair of `fixed_cross_entropy` PRs; the rest touch unrelated models, docs, infra, or tests and should stay separate.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43251 is the best canonical representative for the duplicate pair because it directly references the tracked bug (43240) and cleanly implements the `cross_entropy` kwargs passthrough in `fixed_cross_entropy`.", + "canonical_pr_reason": "`pull_request:43251` is the strongest canonical anchor for the one true duplicate pair: it targets the same `cross_entropy` kwargs bug, has the explicit issue link, and is the more developed change.", "best_issue_reason": null, - "best_pr_reason": "PR 43251 is the strongest global representative of the cluster\u2019s only real duplicate candidate; it is narrowly scoped, bug-driven, and the closest match to the underlying fix.", + "best_pr_reason": "`pull_request:43251` is the best representative of the real duplicate pair because it directly fixes the `fixed_cross_entropy` kwargs gap and is better supported than `43254`; it is not a good canonical for the whole cluster since the other PRs are unrelated.", "soft_edge_verdicts": [ { "left": "pull_request:41855", "right": "pull_request:43823", "accept": false, - "reason": "Tokenizer missing-methods work for Mistral is unrelated to adding the MobileLLM model." + "reason": "Different problems: Mistral tokenizer API gaps vs adding a new MobileLLM model." }, { "left": "pull_request:41855", "right": "pull_request:43424", "accept": false, - "reason": "Mistral tokenizer API additions and an executorch export test address different code paths and different symptoms." + "reason": "Unrelated: tokenizer methods vs executorch dynamic-shape export testing." }, { "left": "pull_request:42668", "right": "pull_request:43823", "accept": false, - "reason": "Processor auto-loading robustness changes are unrelated to introducing MobileLLM." + "reason": "Processor auto-loading robustness is unrelated to adding MobileLLM." }, { "left": "pull_request:43251", "right": "pull_request:43823", "accept": false, - "reason": "A loss-helper kwargs fix is unrelated to MobileLLM model addition." + "reason": "Cross-entropy kwargs fix and MobileLLM model addition are unrelated." }, { "left": "pull_request:43251", "right": "pull_request:44827", "accept": false, - "reason": "`fixed_cross_entropy` kwargs support and Mistral4 test fixes are not the same underlying bug." + "reason": "Same file path is incidental; the bugs are different (`cross_entropy` kwargs vs Mistral4 test fixes)." }, { "left": "pull_request:43424", "right": "pull_request:46911", "accept": false, - "reason": "Executorch dynamic-shape export testing and Olmo3 RoPE handling are different features in different subsystems." + "reason": "Executorch export tests do not match Olmo3 RoPE-layer behavior changes." }, { "left": "pull_request:43424", "right": "pull_request:46707", "accept": false, - "reason": "A dynamic-shape export test is unrelated to adding native FSDP2 support." + "reason": "Dynamic-shape export testing is unrelated to native FSDP2 migration." }, { "left": "pull_request:43424", "right": "pull_request:44827", "accept": false, - "reason": "Executorch export testing and Mistral4 test/model fixes are separate concerns." + "reason": "Executorch export coverage and Mistral4 fixes are different changes." }, { "left": "pull_request:43424", "right": "pull_request:43823", "accept": false, - "reason": "Executorch export tests and MobileLLM model scaffolding are unrelated." + "reason": "Executorch export tests and MobileLLM model addition are unrelated." }, { "left": "pull_request:43823", "right": "pull_request:45655", "accept": false, - "reason": "MobileLLM model creation is unrelated to resolving `cls.config` annotation/attribute order." + "reason": "Adding MobileLLM is unrelated to config resolution order fixes." }, { "left": "pull_request:43823", "right": "pull_request:43995", "accept": false, - "reason": "MobileLLM addition and Falcon output-collection refactoring do not fix the same code path." - }, - { - "left": "pull_request:43823", - "right": "pull_request:45213", - "accept": false, - "reason": "A model implementation PR is unrelated to the new-model skill/doc/tooling PR." - }, - { - "left": "pull_request:43823", - "right": "pull_request:46707", - "accept": false, - "reason": "MobileLLM model code and FSDP2 migration are unrelated changes." - }, - { - "left": "pull_request:43823", - "right": "pull_request:46911", - "accept": false, - "reason": "MobileLLM addition and Olmo3 RoPE-per-layer support are different model implementations." + "reason": "New MobileLLM model code and Falcon output-interface refactoring are different changes." }, { "left": "pull_request:43995", "right": "pull_request:45213", "accept": false, - "reason": "Falcon output-capture refactoring and the model-creation skill are not the same bug or change." + "reason": "Falcon model refactor is unrelated to the model-creation skill PR." }, { "left": "pull_request:43995", "right": "pull_request:44827", "accept": false, - "reason": "Falcon standardized output collection and Mistral4 test fixes are unrelated." + "reason": "Falcon output-capture refactor and Mistral4 test fixes are not the same bug." }, { "left": "pull_request:44827", - "right": "pull_request:45170", + "right": "pull_request:46911", "accept": false, - "reason": "Mistral4 test fixes and the `layrnorm` typo renaming in CLIP-like models are different changes." + "reason": "Mistral4 test fixes and Olmo3 per-layer RoPE are unrelated model changes." }, { "left": "pull_request:44827", - "right": "pull_request:45213", + "right": "pull_request:46707", "accept": false, - "reason": "A model test-fix PR is unrelated to the model-creation skill repository changes." + "reason": "Mistral4 test/debug fixes do not match native FSDP2 migration." }, { "left": "pull_request:44827", - "right": "pull_request:45655", + "right": "pull_request:45170", "accept": false, - "reason": "Mistral4 test updates and config-resolution order fixes are unrelated." + "reason": "Mistral4 tests and `layrnorm` rename/conversion mapping are different issues." }, { "left": "pull_request:44827", - "right": "pull_request:46048", + "right": "pull_request:46267", "accept": false, - "reason": "Mistral4 test fixes and TDT loss-kernel integration are separate code paths." + "reason": "Mistral4 test fixes are unrelated to chat parsing/serve updates." }, { - "left": "pull_request:44827", - "right": "pull_request:46267", + "left": "pull_request:45170", + "right": "pull_request:46707", "accept": false, - "reason": "Mistral4 test fixes and chat-response parsing updates are unrelated." + "reason": "Typo/weight-renaming cleanup is unrelated to FSDP2 migration." }, { "left": "pull_request:44827", - "right": "pull_request:46707", + "right": "pull_request:45213", "accept": false, - "reason": "Mistral4 test/model changes and native FSDP2 migration are not the same underlying issue." + "reason": "Model test fixes are unrelated to the model-creation skill/meta PR." }, { - "left": "pull_request:44827", - "right": "pull_request:46911", + "left": "pull_request:43823", + "right": "pull_request:45213", "accept": false, - "reason": "Mistral4 test fixes and Olmo3 RoPE changes are unrelated model-specific work." + "reason": "A new model implementation and the model-creation skill are different artifacts." }, { - "left": "pull_request:45170", - "right": "pull_request:46707", + "left": "pull_request:44827", + "right": "pull_request:45655", "accept": false, - "reason": "A LayerNorm naming/weight-renaming fix in CLIP-like models is unrelated to FSDP2 migration." + "reason": "Mistral4 tests and config resolution ordering fixes are unrelated." }, { "left": "pull_request:46048", "right": "pull_request:46707", "accept": false, - "reason": "TDT loss-kernel integration and FSDP2 migration do not address the same bug or implementation change." + "reason": "TDT loss kernel integration is unrelated to FSDP2 module migration." }, { "left": "pull_request:44827", - "right": "pull_request:45213", + "right": "pull_request:46048", "accept": false, - "reason": "Mistral4 test/model work is unrelated to the model-creation skill PR." + "reason": "Mistral4 test fixes and TDT loss kernel integration are different changes." }, { "left": "pull_request:43823", - "right": "pull_request:45213", + "right": "pull_request:46707", + "accept": false, + "reason": "Adding MobileLLM and FSDP2 migration are unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:46911", "accept": false, - "reason": "MobileLLM model addition is unrelated to the model-creation skill and scaffolding changes." + "reason": "MobileLLM model addition is unrelated to Olmo3 RoPE logic changes." }, { "left": "pull_request:43251", "right": "pull_request:43254", "accept": true, - "reason": "Both PRs change `fixed_cross_entropy` in the same file to pass supported kwargs through to `nn.functional.cross_entropy`; they fix the same concrete loss-helper bug and could be merged into one PR." + "reason": "Both PRs fix the same `fixed_cross_entropy` kwargs support gap in `loss_utils.py`; they are near-duplicate implementations of the same concrete bug." } ] }, "evaluator_result": { "accept": true, - "feedback": "The assessment is well grounded in the packet. The only accepted pair, 43251/43254, is a conservative match: both touch `fixed_cross_entropy` in the same file, point to the same issue target (43240), and are plausibly mergeable into one bugfix PR. The rest of the pairwise rejections are appropriately cautious and do not overstate similarity." + "feedback": "The analysis is grounded in the packet and stays conservative on the soft edges. The only accepted duplicate pair, `pull_request:43251` / `pull_request:43254`, is clearly mergeable as the same `fixed_cross_entropy` kwargs bug in the same file and with the same issue target. The rest of the rejected pairs are reasonably treated as topical similarities rather than evidence of the same bug." }, "evaluator_used": true, "retried": false, @@ -21415,12 +21406,12 @@ "soft_pairs": [ "pull_request:42975|pull_request:42976" ], - "prepared_review_unit_hash": "26826399be55932318e38a3b55498429ddf195d36278f6a692bde16f602ca9b5", + "prepared_review_unit_hash": "70767d93f058c1ba4a1fa1f60e6a98f31e469f68b468ac812084043cd49ae74c", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5935, + "serialized_chars": 5934, "estimated_input_tokens": 1484, "estimated_eval_tokens": 3224 }, @@ -21428,7 +21419,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5935, + "serialized_chars": 5934, "estimated_input_tokens": 1484, "estimated_eval_tokens": 3224 }, @@ -21437,18 +21428,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open PRs both upgrade GitHub Actions, but they target different action families and change sets: one is a broad Node 24 compatibility pass, the other is a Docker build action version bump. They are related but not duplicates.", - "confidence": 0.92, + "summary": "Two open PRs both do GitHub Actions dependency bumps, but they target different actions and different workflow sets. They are related maintenance, not the same concrete change, so they should not be deduplicated.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #42975 is the better canonical representative because it has the clearest, broadest scope: upgrading multiple workflow actions for Node 24 compatibility across many CI workflows.", + "canonical_pr_reason": "PR #42975 is the broader workflow-runtime upgrade: it updates checkout/cache/upload-artifact across many workflows for Node 24 compatibility, making it the more representative maintenance PR in this cluster.", "best_issue_reason": null, - "best_pr_reason": "PR #42975 is the strongest representative of the cluster because its title and file coverage summarize the overarching workflow-action upgrade theme more completely than the narrower Docker-action bump in #42976.", + "best_pr_reason": "PR #42975 is the best overall representative because it addresses the larger cross-workflow GitHub Actions compatibility update, while #42976 is a narrower docker-action version bump.", "soft_edge_verdicts": [ { "left": "pull_request:42975", "right": "pull_request:42976", "accept": false, - "reason": "Both are workflow dependency upgrades, but they affect different underlying actions and concrete code paths (checkout/cache/upload-artifact vs docker/build-push-action). They could be part of a larger maintenance sweep, but not the same fix or mergeable as one duplicate PR." + "reason": "Both are dependency refreshes in CI workflows, but they update different actions (core workflow actions vs docker/build-push-action) and different concrete paths. They could be part of a maintenance sweep, but they are not the same underlying bug/change and should not be merged as duplicates." } ] }, @@ -21489,7 +21480,7 @@ "pull_request:45490|pull_request:46562", "pull_request:46562|pull_request:46658" ], - "prepared_review_unit_hash": "d41c076ac823ac20841c19dac202ffd2266d4266abbeaf52bd1a3d72d4f228f2", + "prepared_review_unit_hash": "47118af5c3c40cb0617d92c5bea787d4b0bfb8d81120f89a6a5257991fd06033", "component_budget": { "node_count": 6, "item_count": 6, @@ -21511,60 +21502,60 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Mostly a bundle of unrelated model-addition PRs. The only close-looking pair is Step3p7-related, but they still differ in scope: one integrates the model into Transformers, the other is a WIP scripts/mini-model conversion repo. No true duplicate pair stands out.", - "confidence": 0.9, + "summary": "This cluster is mostly a set of unrelated model-addition PRs. The only potentially close pair is Step3p7 Flash vs a WIP Step3 7 Flash conversion/script PR, but their scopes differ enough that they should not be treated as duplicates.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46562 is the strongest representative of the only near-duplicate area: a full Step 3.7 Flash Transformers integration with docs, auto-mappings, and tests. The other Step3p7 PR is a separate WIP script/conversion effort, not the same change.", + "canonical_pr_reason": "No true duplicate PR stands out in the cluster; 46562 is the best representative only because it is the core Step3p7 model implementation PR, but it is not a duplicate of the others.", "best_issue_reason": null, - "best_pr_reason": "PR 46562 is the best overall PR to anchor this cluster because it is the most complete, self-contained model integration and the only one that plausibly overlaps with another item in the set. The rest are different model additions.", + "best_pr_reason": "46562 is the most substantial and self-contained model implementation PR in the cluster. It has the strongest technical completeness and the clearest model-addition scope.", "soft_edge_verdicts": [ { "left": "pull_request:42978", "right": "pull_request:43448", "accept": false, - "reason": "Different model families (ViT NEPA vs Molmo). Shared docs/auto-registration files are just boilerplate for model additions, not the same underlying change." + "reason": "Different model families: ViT NEPA vs Molmo. Shared docs/auto-registry edits are incidental, not the same underlying change." }, { "left": "pull_request:42978", "right": "pull_request:43451", "accept": false, - "reason": "Different model families (ViT NEPA vs Molmo2). Similar scaffolding changes do not make this the same bug or change." + "reason": "Different models: ViT NEPA and Molmo2. No shared bug or code-path problem." }, { "left": "pull_request:42978", "right": "pull_request:45490", "accept": false, - "reason": "Different model families (ViT NEPA vs CTSM). They touch shared auto/docs plumbing but implement unrelated additions." + "reason": "Completely different additions: ViT NEPA vs CTSM." }, { "left": "pull_request:43448", "right": "pull_request:43451", "accept": false, - "reason": "Molmo and Molmo2 are related by naming but are distinct model integrations with different files and code paths." + "reason": "Molmo and Molmo2 are related names, but these are distinct model integrations with different files and behavior." }, { "left": "pull_request:43448", "right": "pull_request:45490", "accept": false, - "reason": "Molmo vs CTSM are unrelated model additions; overlapping framework files are not enough to treat them as duplicates." + "reason": "Molmo vs CTSM are separate model additions, not the same bug or feature." }, { "left": "pull_request:43451", "right": "pull_request:45490", "accept": false, - "reason": "Molmo2 vs CTSM are separate model integrations with different implementations and test coverage." + "reason": "Molmo2 and CTSM are unrelated model PRs despite both touching auto-mapping and docs." }, { "left": "pull_request:45490", "right": "pull_request:46562", "accept": false, - "reason": "CTSM and Step 3.7 Flash are unrelated model additions; the shared auto/doc changes are generic plumbing, not one underlying fix." + "reason": "CTSM and Step3p7 Flash are unrelated model implementations; overlap in registry/docs does not make them duplicates." }, { "left": "pull_request:46562", "right": "pull_request:46658", "accept": false, - "reason": "Both mention Step3p7, but 46562 adds the Transformers model while 46658 is a WIP mini-model/script conversion repo. They are different code paths and not plausibly merged as one PR." + "reason": "Both reference Step 3.7 Flash, but 46562 is the actual Transformers model implementation while 46658 is a WIP script/mini-model conversion artifact set. They are related support work, not the same mergeable code-path change." } ] }, @@ -21586,14 +21577,13 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-43291-8", + "cluster_id": "cluster-43291-7", "nodes": [ "pull_request:43291", "pull_request:43665", "pull_request:44815", "pull_request:46380", "pull_request:46539", - "pull_request:46870", "pull_request:46886", "pull_request:46922" ], @@ -21603,7 +21593,6 @@ "pull_request:43291|pull_request:46380", "pull_request:43291|pull_request:46922", "pull_request:43665|pull_request:46539", - "pull_request:43665|pull_request:46870", "pull_request:43665|pull_request:44815", "pull_request:43665|pull_request:46380", "pull_request:43665|pull_request:46922", @@ -21615,139 +21604,133 @@ "pull_request:46539|pull_request:46922", "pull_request:46886|pull_request:46922" ], - "prepared_review_unit_hash": "137c06b664034453c1b095ccb0d3aacf26c0c77f44f9c3c8d6b56f618668591d", + "prepared_review_unit_hash": "24a8026c812ab78609bab2024b12888fce515a40edc2f22f588195e11478e502", "component_budget": { - "node_count": 8, - "item_count": 8, - "soft_pair_count": 16, - "serialized_chars": 21660, - "estimated_input_tokens": 5415, - "estimated_eval_tokens": 11086 + "node_count": 7, + "item_count": 7, + "soft_pair_count": 15, + "serialized_chars": 19595, + "estimated_input_tokens": 4899, + "estimated_eval_tokens": 10054 }, "budget": { - "node_count": 8, - "item_count": 8, - "soft_pair_count": 16, - "serialized_chars": 21660, - "estimated_input_tokens": 5415, - "estimated_eval_tokens": 11086 + "node_count": 7, + "item_count": 7, + "soft_pair_count": 15, + "serialized_chars": 19595, + "estimated_input_tokens": 4899, + "estimated_eval_tokens": 10054 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Only one soft-edge pair looks duplicate-like: the two YOLOS device-mismatch fixes, where 46922 is a fuller version of 46886. The rest are different models, workflows, or unrelated code paths and should remain separate.", - "confidence": 0.97, + "summary": "Most items are unrelated PRs that only share vague titles or the word \u201cfix.\u201d The only strong duplicate signal is the pair of YOLOS PRs, which both target the same device_map=\"auto\" mismatch in the same model file.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46922 is the best canonical PR because it contains the more complete YOLOS device-placement fix, including the mid-position embedding device move and the no_split_modules adjustment, whereas 46886 is the narrower earlier variant.", + "canonical_pr_reason": "PR 46886 is the clearest anchor: it has the specific YOLOS device_map=\"auto\" mismatch title, an explicit issue target, and a focused one-file fix.", "best_issue_reason": null, - "best_pr_reason": "PR 46922 is the strongest representative of the only true duplicate-like pair in the cluster; it addresses the same YOLOS auto-device-map bug as 46886 and includes the broader fix surface.", + "best_pr_reason": "PR 46886 is the best representative of the cluster because it most clearly describes the concrete bug and the intended fix scope; the other PRs are either unrelated or much less specific.", "soft_edge_verdicts": [ { "left": "pull_request:43291", "right": "pull_request:43665", "accept": false, - "reason": "Both are test-oriented, but they target different model suites and different failure modes (Whisper vs CLIP/ViT)." + "reason": "Both are generic test-fix PRs, but they touch different models and different test paths; not the same underlying bug." }, { "left": "pull_request:43291", "right": "pull_request:44815", "accept": false, - "reason": "Whisper tokenizer/test changes are unrelated to FP8 dequantization and model-loading logic." + "reason": "Whisper tokenizer/test changes and FP8 dequant changes are unrelated code paths." }, { "left": "pull_request:43291", "right": "pull_request:46380", "accept": false, - "reason": "Whisper decoding/tests and generation-config validation are different subsystems with different bugs." - }, - { - "left": "pull_request:43291", - "right": "pull_request:46922", - "accept": false, - "reason": "Whisper test/tokenizer work is unrelated to the YOLOS device-placement bug." + "reason": "Whisper test cleanup is unrelated to generation-config watermarking handling and model-specific generation tests." }, { - "left": "pull_request:43665", - "right": "pull_request:46539", + "left": "pull_request:43291", + "right": "pull_request:46922", "accept": false, - "reason": "Generic CLIP/ViT test edits do not match the Phimoe CI memory-sharing fix." + "reason": "Whisper test/tokenizer work and YOLOS device mismatch fixes are different bugs in different subsystems." }, { "left": "pull_request:43665", - "right": "pull_request:46870", + "right": "pull_request:46539", "accept": false, - "reason": "Different areas: model test edits versus ASR pipeline classification changes." + "reason": "These are broad test edits in different areas (CLIP/ViT vs Phimoe integration); no shared concrete bug." }, { "left": "pull_request:43665", "right": "pull_request:44815", "accept": false, - "reason": "These are not the same change or bug; one is model tests, the other is FP8/model-loading work." + "reason": "CLIP/ViT test edits do not match the FP8 dequant/model-loading change." }, { "left": "pull_request:43665", "right": "pull_request:46380", "accept": false, - "reason": "Both touch broad model code/testing, but they address distinct issues and are not mergeable as one fix." + "reason": "Model test tweaks are unrelated to watermarking_config validation and generation behavior." }, { "left": "pull_request:43665", "right": "pull_request:46922", "accept": false, - "reason": "CLIP/ViT test edits are unrelated to YOLOS device mismatch." + "reason": "Generic model test fixes are unrelated to the YOLOS device placement bug." }, { "left": "pull_request:44815", "right": "pull_request:46539", "accept": false, - "reason": "FP8 dequantization/model-loading changes are unrelated to CI runner memory limits and integration-test setup." + "reason": "FP8 dequant/core loading changes are unrelated to the K8S/Phimoe test environment fix." }, { "left": "pull_request:44815", "right": "pull_request:46380", "accept": false, - "reason": "FP8 kernel/loading fixes and generation watermarking validation are different code paths and bugs." + "reason": "Different concrete changes: FP8 kernel/padding logic versus generation watermarking validation." }, { "left": "pull_request:44815", "right": "pull_request:46922", "accept": false, - "reason": "FP8 dequantization work is unrelated to YOLOS device placement." + "reason": "FP8 dequant fixes and YOLOS device mismatch fixes are distinct bugs." }, { "left": "pull_request:46380", "right": "pull_request:46539", "accept": false, - "reason": "Generation-config validation and Phimoe CI environment fixes are separate problems." + "reason": "Generation watermarking validation and Phimoe integration-test environment adjustments do not fix the same issue." }, { "left": "pull_request:46380", "right": "pull_request:46922", "accept": false, - "reason": "Watermarking/config validation changes are unrelated to YOLOS device mismatch." + "reason": "Unrelated: generation config validation versus YOLOS device placement." }, { "left": "pull_request:46539", "right": "pull_request:46922", "accept": false, - "reason": "CI instance-sharing fixes and YOLOS model device-placement fixes are not the same underlying issue." + "reason": "CI/Phimoe test infrastructure changes are unrelated to the YOLOS model bug." }, { "left": "pull_request:46886", "right": "pull_request:46922", "accept": true, - "reason": "Both target the same YOLOS device-mismatch bug with device_map=\"auto\"; 46922 is the broader version that also fixes the forward-path embedding placement." + "reason": "Both PRs target the same YOLOS device_map=\"auto\" device-mismatch problem in the same file; the fixes are alternative pieces of the same underlying bug and could plausibly be merged together." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The summary is mostly grounded, but several soft-edge verdicts are not conservative and are internally inconsistent: e.g. `43665` vs `44815`, `46380`, and `46922` are marked `accept: true` even though the reasons explicitly say they are different subsystems / not duplicates. That overstates evidence. The only well-supported duplicate-like pair in the packet is `46886`/`46922`, which is plausibly mergeable as the same YOLOS bugfix." + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only clearly mergeable same-bug pair is the YOLOS duo, which is supported by the shared issue target and shared file. The other rejects are reasonable given the distinct subsystems and fixes shown in the packet." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -21774,52 +21757,52 @@ "pull_request:43488|pull_request:43532", "pull_request:43532|pull_request:46747" ], - "prepared_review_unit_hash": "efa2f9436b91922ea3bbfae1e580a08be1193e208ff953122ddefdc78a09ae8f", + "prepared_review_unit_hash": "12879f44029a5a64348ccf622283775db7b9840b82256359fe129c84cc3e4ec3", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 6348, - "estimated_input_tokens": 1587, - "estimated_eval_tokens": 3430 + "serialized_chars": 6350, + "estimated_input_tokens": 1588, + "estimated_eval_tokens": 3432 }, "budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 6348, - "estimated_input_tokens": 1587, - "estimated_eval_tokens": 3430 + "serialized_chars": 6350, + "estimated_input_tokens": 1588, + "estimated_eval_tokens": 3432 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three unrelated '[don't merge]' repository-bot / workflow test PRs with only superficial similarity in title and purpose. They touch different code paths: packaging/setup, comment/workflow automation, and security-gate workflow behavior. No duplicate pair is a strong same-change match.", + "summary": "All three pull requests are separate \"don't merge\"/test-workflow probes with different code paths and purposes; they are not true duplicates of one another.", "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "PR 46747 is the most focused and self-contained change, touching only a single workflow file, so it's the least misleading representative of the cluster.", "best_issue_reason": null, - "best_pr_reason": "PR 43532 is the broadest representative of the cluster if one must be picked: it modifies multiple workflow files plus testing utilities and test code, whereas the others are narrower and target different areas.", + "best_pr_reason": "There is no strong single canonical duplicate here, but 46747 is the cleanest anchor because it is the narrowest change; 43488 and 43532 are broader and less representative.", "soft_edge_verdicts": [ { "left": "pull_request:43488", "right": "pull_request:46747", "accept": false, - "reason": "Different changes: 43488 is a setup/version and dependency tweak plus a test change, while 46747 only switches the PR CI security gate workflow trigger from pull_request_target to pull_request. Similar '[don't merge]' framing is not enough to make them the same underlying fix." + "reason": "Different change sets: 43488 is a setup/version/dependency test PR, while 46747 only flips a workflow trigger. They do not fix the same bug or code path." }, { "left": "pull_request:43488", "right": "pull_request:43532", "accept": false, - "reason": "43488 is about packaging/versioning and adding pytest-random-order, whereas 43532 rewrites comment-driven CI/workflow automation and testing utilities. They do not fix the same concrete bug and would not plausibly merge into one PR." + "reason": "43488 changes packaging/versioning and one test, whereas 43532 edits several workflows plus test utilities. The overlap is superficial, not the same underlying change." }, { "left": "pull_request:43532", "right": "pull_request:46747", "accept": false, - "reason": "Both touch workflows, but the actual code paths differ: 43532 updates self-comment CI, scheduled/slack workflows, testing helpers, and a vit test; 46747 only changes the security gate workflow trigger. This is too broad to treat as one duplicate change." + "reason": "43532 is a broad CI/workflow test PR with multiple files; 46747 is a narrow security-gate workflow tweak. Different intent and implementation." } ] }, @@ -21849,12 +21832,12 @@ "soft_pairs": [ "pull_request:43612|pull_request:43613" ], - "prepared_review_unit_hash": "0c04026f2ad3c05ed44a2cf158e0232a2a95daead3a20cf6ce7d2f9a2628f927", + "prepared_review_unit_hash": "dd3b83a3d2c95699f853bae34a3c72da6d2eafbaaf1122401ae37d8cfb7a3af3", "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 }, @@ -21862,7 +21845,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6083, + "serialized_chars": 6082, "estimated_input_tokens": 1521, "estimated_eval_tokens": 3298 }, @@ -21871,18 +21854,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two separate feature PRs: one adds a Promptable Concept Segmentation pipeline, the other adds a Promptable Visual Segmentation pipeline. They share documentation/registry plumbing, but they implement different pipelines and model integrations, so they are not duplicates.", + "summary": "These are two separate feature PRs: one adds a Promptable Concept Segmentation pipeline, the other adds a Promptable Visual Segmentation pipeline. They share documentation and pipeline scaffolding, but the underlying tasks, code paths, and test/model coverage differ, so they are not duplicates.", "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "No single canonical PR: these are distinct feature additions with different pipeline names, model hooks, and tests, despite overlapping docs/metadata updates.", + "canonical_pr_reason": "No single canonical PR: the two PRs implement different segmentation pipelines rather than the same change.", "best_issue_reason": null, - "best_pr_reason": "Neither PR is a global duplicate canonical; if a representative must be chosen, PR #43613 is slightly broader in model coverage, but it is still a separate feature from #43612.", + "best_pr_reason": "PR #43613 is the broader and more representative feature addition, with wider model/test coverage and more files touched.", "soft_edge_verdicts": [ { "left": "pull_request:43612", "right": "pull_request:43613", "accept": false, - "reason": "Shared docs and registration files are incidental. One adds Promptable Concept Segmentation, the other adds Promptable Visual Segmentation; they target different tasks and code paths, so they should not be merged as the same change." + "reason": "Reject: same broad segmentation area, but different concrete pipelines and model integrations (concept vs visual segmentation). Shared docs/scaffolding alone is not enough to treat them as one change." } ] }, @@ -21912,7 +21895,7 @@ "soft_pairs": [ "pull_request:43651|pull_request:43663" ], - "prepared_review_unit_hash": "39f1a9a98253c83e682041f8254d7b47def04c8f42bb72b29eabdf731be9b5ec", + "prepared_review_unit_hash": "28bfb5bd1b1e001ea18974f98f42d1c6844a89f7bf2bf249e1ad49c712247198", "component_budget": { "node_count": 2, "item_count": 2, @@ -21934,18 +21917,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two open PRs both touch `trainer.py`, but they implement different trainer extension hooks: one changes gradient-accumulation loss scaling, the other refactors signature-column filtering. They are not the same underlying change and should not be merged as duplicates.", - "confidence": 0.98, + "summary": "Two open trainer.py PRs, but they address different trainer extension points: gradient-accumulation loss scaling vs. signature-column filtering. They are not duplicates.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the patches address different code paths and different custom-trainer extensibility points.", + "canonical_pr_reason": "No single canonical PR: 43651 changes loss scaling behavior in training_step, while 43663 adds a separate helper for signature-column filtering in trainer input preparation.", "best_issue_reason": null, - "best_pr_reason": "PR #43651 is the clearer self-contained behavior change, but it is still unrelated to PR #43663; it is not a duplicate candidate for the cluster.", + "best_pr_reason": "43651 is the more concrete bug-fix-style PR and is the better anchor for the cluster, but it does not subsume the unrelated column-filtering refactor in 43663.", "soft_edge_verdicts": [ { "left": "pull_request:43651", "right": "pull_request:43663", "accept": false, - "reason": "Different functionality in the same file: #43651 changes gradient accumulation loss scaling, while #43663 adds a helper for model signature column filtering. They do not fix the same bug or implement the same change." + "reason": "Different underlying trainer changes: 43651 adjusts gradient-accumulation loss scaling, while 43663 refactors column filtering for model signatures. They do not fix the same code-path problem and would not plausibly merge into one PR." } ] }, @@ -21975,7 +21958,7 @@ "soft_pairs": [ "pull_request:43656|pull_request:43842" ], - "prepared_review_unit_hash": "217d61badd7b399b69bbedfc98d80aaf3b96473363c652bf420d4e64509118ab", + "prepared_review_unit_hash": "e4bde49293e53bf41234504cd19cf64502a41ec271ee8af61738ef55fbc1c51f", "component_budget": { "node_count": 2, "item_count": 2, @@ -21997,24 +21980,27 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two open PRs touch the same file and both mention TypeAdapter-related NameErrors, but they address different root causes: one is a broader annotation/runtime typing cleanup in the CLI, the other is a narrow fix for optional pydantic imports. They are related, but not duplicates.", - "confidence": 0.91, + "summary": "Both PRs target runtime annotation/import problems in `src/transformers/cli/serve.py`, but they use different fixes and do not clearly resolve the exact same bug. They are related, yet the evidence is not strong enough to treat them as one mergeable change.", + "confidence": 0.73, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43656 is the broader CLI typing fix and likely the better canonical representative for this cluster; it addresses runtime annotation evaluation in serve.py rather than only the pydantic-import edge case.", + "canonical_pr_reason": "PR #43656 is the broader representative of the cluster because it applies a file-wide annotation/runtime-safety fix in `serve.py`, whereas PR #43842 is a narrower `TypeAdapter` typing workaround.", "best_issue_reason": null, - "best_pr_reason": "PR 43656 is the strongest single representative because it appears to fix the underlying annotation NameError pattern more generally in the CLI code path, while PR 43842 is a narrower follow-up about TypeAdapter being unavailable when pydantic is missing.", + "best_pr_reason": "PR #43656 best represents the cluster\u2019s main theme: fixing annotation-related runtime failures in the CLI. It is the more general change and is less dependent on a specific import workaround.", "soft_edge_verdicts": [ { "left": "pull_request:43656", "right": "pull_request:43842", "accept": false, - "reason": "Reject: both are in serve.py and both mention NameError, but the diffs point to different bugs\u201443656 changes annotation evaluation with __future__ annotations and typing cleanup, while 43842 guards a pydantic TypeAdapter import behind TYPE_CHECKING. They are related but not the same underlying fix." + "reason": "Related but not the same underlying fix: one uses `from __future__ import annotations`, the other uses `TYPE_CHECKING` plus a stringized `TypeAdapter` annotation. That is similar in area, but not enough to treat them as one mergeable bugfix." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, + "evaluator_result": { + "accept": false, + "feedback": "The packet only supports a conservative similarity judgment: both PRs touch `src/transformers/cli/serve.py` and are about annotation/runtime import issues, but they use different fix strategies (`from __future__ import annotations` vs. `TYPE_CHECKING` import plus stringized annotation). That is enough to say they are related, not enough to confidently say they are the same underlying `TypeAdapter` NameError or that they are a single mergeable bugfix. The soft-pair acceptance is too strong, and the summary slightly overstates the evidence." + }, + "evaluator_used": true, + "retried": true, "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null @@ -22201,20 +22187,20 @@ "pull_request:44044|pull_request:44154", "pull_request:44024|pull_request:44076" ], - "prepared_review_unit_hash": "c687705031019804ae71c272450db66ee25afa8a0a1a6838cf2eeb99a4e8f792", + "prepared_review_unit_hash": "dea8d8ade4c6d900e3bc8fcaeea99658361b0d7c20b7fd4ebf7101997c44b3a4", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 77, - "serialized_chars": 84505, - "estimated_input_tokens": 21127, - "estimated_eval_tokens": 42510 + "serialized_chars": 84500, + "estimated_input_tokens": 21125, + "estimated_eval_tokens": 42506 }, "budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 72, - "serialized_chars": 76936, + "serialized_chars": 76935, "estimated_input_tokens": 19234, "estimated_eval_tokens": 38724 }, @@ -22223,114 +22209,114 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a batch of model-specific refactors for standardized output tracing/capture_outputs. The only true duplicate-like overlaps are within the same concrete model file or a broader PR that subsumes the same file-level change; cross-model \u201csame issue target\u201d pairs should stay separate.", - "confidence": 0.92, + "summary": "This cluster is mostly a broad set of model-specific output-tracing refactors that are not duplicates of each other. The only clear duplicate families are within GPT-J/CodeGen and GPT-Neo; the rest touch different models and code paths despite similar wording and shared issue references.", + "confidence": 0.88, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44722 is the best canonical PR for the GPT-J/CodeGen duplicate subset: it is the latest and most complete variant, covers the shared GPT-J tracing refactor plus the CodeGen companion change, and aligns directly with the tracking issue.", + "canonical_pr_reason": "PR 44722 is the strongest representative of the real duplicate family: it targets the same output-tracing refactor as earlier GPT-J variants, explicitly references #43979, and is the broadest/latest concrete implementation in that subset.", "best_issue_reason": null, - "best_pr_reason": "PR 44722 is the strongest representative of the duplicate subset because it consolidates the shared GPT-J/CodeGen output-tracing change rather than a narrower or earlier variant.", + "best_pr_reason": "44722 is the best single PR to keep because it covers the genuine GPT-J/CodeGen duplicate thread, is explicit about the tracking issue, and looks like the most complete version of that change.", "soft_edge_verdicts": [ { "left": "pull_request:43996", "right": "pull_request:44085", "accept": false, - "reason": "Different model families (CVT/FNet vs GPT-J); same broad output-tracing theme is not enough." + "reason": "Different model families and different code paths; same issue target alone is not enough." }, { "left": "pull_request:43996", "right": "pull_request:44044", "accept": false, - "reason": "Different models and code paths (CVT/FNet vs DeBERTa v2)." + "reason": "Unrelated models (CVT/FNet vs DeBERTa); only the generic output-tracing theme matches." }, { "left": "pull_request:44066", "right": "pull_request:44085", "accept": true, - "reason": "Both target the same GPT-J output-tracing refactor in the same file; 44085 looks like a narrower variant of the same change." + "reason": "Same GPT-J output-tracing refactor family; 44085 looks like a narrower variant of the same code-path change." }, { "left": "pull_request:44007", "right": "pull_request:44072", "accept": false, - "reason": "Both are output-tracing refactors, but they are different model implementations (ResNet/RegNet/RT-DETR vs EfficientNet)." + "reason": "Both are standardized-output refactors, but for different backbones and different implementations." }, { "left": "pull_request:44072", "right": "pull_request:44722", "accept": false, - "reason": "EfficientNet and GPT-J/CodeGen are unrelated code paths despite similar refactor style." + "reason": "EfficientNet/FocalNet-style output tracing is unrelated to the GPT-J/CodeGen change in 44722." }, { "left": "pull_request:44066", "right": "pull_request:44072", "accept": false, - "reason": "GPT-J/CodeGen vs EfficientNet; shared issue tracking does not make them duplicates." + "reason": "Different model families and different concrete forward paths." }, { "left": "pull_request:44013", "right": "pull_request:44044", "accept": false, - "reason": "MobileNetV2 and DeBERTa v2 are unrelated implementations." + "reason": "MobileNetV2 and DeBERTa are unrelated code paths." }, { "left": "pull_request:44066", "right": "pull_request:44086", "accept": false, - "reason": "GPT-J/CodeGen versus MGP-STR; different model internals and fixes." + "reason": "GPT-J/CodeGen vs MGP-STR; same theme but not the same underlying change." }, { "left": "pull_request:44018", "right": "pull_request:44068", "accept": true, - "reason": "Same model/file family (GPT-Neo) and the diffs are the same output-tracing refactor shape." + "reason": "Same GPT-Neo output-tracing refactor on the same file family; these look like duplicate iterations of one change." }, { "left": "pull_request:44066", "right": "pull_request:44071", "accept": false, - "reason": "GPT-J/CodeGen vs MPT are different model code paths." + "reason": "GPT-J/CodeGen and MPT are different model implementations." }, { "left": "pull_request:44066", "right": "pull_request:44068", "accept": false, - "reason": "GPT-J/CodeGen versus GPT-Neo; similar refactor pattern but not the same change." + "reason": "GPT-J/CodeGen vs GPT-Neo; similar refactor pattern but distinct code paths." }, { "left": "pull_request:44018", "right": "pull_request:44066", "accept": false, - "reason": "GPT-Neo and GPT-J/CodeGen are separate model-specific refactors." + "reason": "Different models and no shared concrete bug/change." }, { "left": "pull_request:44019", "right": "pull_request:44722", "accept": false, - "reason": "ResNet output tracing and GPT-J/CodeGen are unrelated changes." + "reason": "ResNet output tracing is unrelated to GPT-J/CodeGen output tracing." }, { "left": "pull_request:44086", "right": "pull_request:44722", "accept": false, - "reason": "MGP-STR and GPT-J/CodeGen are different models and different concrete fixes." + "reason": "MGP-STR and GPT-J/CodeGen are different implementations." }, { "left": "pull_request:44019", "right": "pull_request:44085", "accept": false, - "reason": "ResNet versus GPT-J; same issue tag is not enough." + "reason": "ResNet and GPT-J are different code paths." }, { "left": "pull_request:44018", "right": "pull_request:44019", "accept": false, - "reason": "GPT-Neo and ResNet are different code paths." + "reason": "GPT-Neo and ResNet are unrelated models." }, { "left": "pull_request:44019", "right": "pull_request:44071", "accept": false, - "reason": "ResNet versus MPT are unrelated implementations." + "reason": "ResNet vs MPT; only the generic tracing refactor theme overlaps." }, { "left": "pull_request:44068", @@ -22342,283 +22328,259 @@ "left": "pull_request:44068", "right": "pull_request:44116", "accept": false, - "reason": "GPT-Neo and Flaubert are unrelated model-specific refactors." + "reason": "GPT-Neo and Flaubert are unrelated code paths." }, { "left": "pull_request:44018", "right": "pull_request:44116", "accept": false, - "reason": "GPT-Neo and Flaubert do not share the same concrete bug/change." + "reason": "GPT-Neo vs Flaubert; same tracing motif, different implementation." }, { "left": "pull_request:44071", "right": "pull_request:44722", "accept": false, - "reason": "MPT and GPT-J/CodeGen are different code paths." - }, - { - "left": "pull_request:44018", - "right": "pull_request:44085", - "accept": false, - "reason": "GPT-Neo versus GPT-J; same pattern, different model fix." + "reason": "MPT and GPT-J/CodeGen are separate model changes." }, { "left": "pull_request:44018", - "right": "pull_request:44086", - "accept": false, - "reason": "GPT-Neo and MGP-STR are unrelated." - }, - { - "left": "pull_request:44068", - "right": "pull_request:44086", - "accept": false, - "reason": "GPT-Neo and MGP-STR are unrelated." - }, - { - "left": "pull_request:44071", "right": "pull_request:44085", "accept": false, - "reason": "MPT and GPT-J are different model internals." + "reason": "GPT-Neo and GPT-J are different code paths." }, { "left": "pull_request:44085", "right": "pull_request:44086", "accept": false, - "reason": "GPT-J and MGP-STR are different changes." + "reason": "Different models and no single concrete bug shared." }, { "left": "pull_request:44059", "right": "pull_request:44086", "accept": false, - "reason": "GPT-2 and MGP-STR are unrelated model-specific refactors." + "reason": "GPT-2 and MGP-STR are unrelated implementations." }, { "left": "pull_request:44098", "right": "pull_request:44101", "accept": false, - "reason": "ViLT versus Flaubert/XLM are different model families." + "reason": "ViLT and Flaubert/XLM are different model families." }, { "left": "pull_request:44059", - "right": "pull_request:44071", + "right": "pull_request:44722", "accept": false, - "reason": "GPT-2 and MPT are separate implementations." + "reason": "GPT-2 vs GPT-J/CodeGen; similar refactor language but distinct code paths." }, { - "left": "pull_request:43997", - "right": "pull_request:44017", + "left": "pull_request:44018", + "right": "pull_request:44071", "accept": false, - "reason": "RegNet and Segformer are unrelated model code paths." + "reason": "GPT-Neo and MPT are different backbones." }, { - "left": "pull_request:43997", - "right": "pull_request:44056", + "left": "pull_request:44017", + "right": "pull_request:44161", "accept": false, - "reason": "RegNet and MPNet are different model implementations." + "reason": "SegFormer and LongT5 are unrelated model implementations." }, { - "left": "pull_request:44071", - "right": "pull_request:44076", + "left": "pull_request:44017", + "right": "pull_request:44722", "accept": false, - "reason": "MPT and ImageGPT are unrelated changes." + "reason": "SegFormer output tracing is not the same change as GPT-J/CodeGen tracing." }, { - "left": "pull_request:44074", - "right": "pull_request:44101", + "left": "pull_request:44017", + "right": "pull_request:44073", "accept": false, - "reason": "TextNet and Flaubert/XLM are different model-specific refactors." + "reason": "SegFormer and VisualBert are different code paths." }, { - "left": "pull_request:43997", - "right": "pull_request:44018", + "left": "pull_request:44073", + "right": "pull_request:44101", "accept": false, - "reason": "RegNet and GPT-Neo are different code paths." + "reason": "VisualBert vs Flaubert/XLM; only general refactor similarity." }, { - "left": "pull_request:44056", + "left": "pull_request:44073", "right": "pull_request:44161", "accept": false, - "reason": "MPNet and LongT5 are unrelated model fixes." + "reason": "VisualBert and LongT5 are unrelated." }, { - "left": "pull_request:43997", - "right": "pull_request:44161", + "left": "pull_request:44059", + "right": "pull_request:44071", "accept": false, - "reason": "RegNet and LongT5 are different model families." + "reason": "GPT-2 and MPT are different model families." }, { - "left": "pull_request:44074", - "right": "pull_request:44161", + "left": "pull_request:44056", + "right": "pull_request:44101", "accept": false, - "reason": "TextNet and LongT5 are unrelated." + "reason": "MPNet and Flaubert/XLM are unrelated code paths." }, { "left": "pull_request:44018", - "right": "pull_request:44161", - "accept": false, - "reason": "GPT-Neo and LongT5 are different model-specific changes." - }, - { - "left": "pull_request:44010", - "right": "pull_request:44071", + "right": "pull_request:44059", "accept": false, - "reason": "SqueezeBert and MPT are unrelated implementations." + "reason": "GPT-Neo and GPT-2 are distinct model implementations." }, { - "left": "pull_request:44010", - "right": "pull_request:44073", + "left": "pull_request:44059", + "right": "pull_request:44068", "accept": false, - "reason": "SqueezeBert and VisualBert are different model code paths." + "reason": "GPT-2 vs GPT-Neo; same broad goal, different concrete changes." }, { - "left": "pull_request:44018", - "right": "pull_request:44129", + "left": "pull_request:44101", + "right": "pull_request:44161", "accept": false, - "reason": "GPT-Neo and SpeechT5 are unrelated model families." + "reason": "Flaubert/XLM and LongT5 are different model families." }, { "left": "pull_request:43997", - "right": "pull_request:44074", + "right": "pull_request:44017", "accept": false, - "reason": "RegNet and TextNet are different model-specific refactors." + "reason": "RegNet and SegFormer are unrelated models." }, { - "left": "pull_request:44056", - "right": "pull_request:44101", + "left": "pull_request:43997", + "right": "pull_request:44056", "accept": false, - "reason": "MPNet and Flaubert/XLM are unrelated." + "reason": "RegNet and MPNet are unrelated code paths." }, { "left": "pull_request:44071", "right": "pull_request:44076", "accept": false, - "reason": "MPT and ImageGPT are unrelated changes." + "reason": "MPT and ImageGPT are different implementations." }, { - "left": "pull_request:44018", - "right": "pull_request:44059", + "left": "pull_request:44074", + "right": "pull_request:44101", "accept": false, - "reason": "GPT-Neo and GPT-2 are different model code paths." + "reason": "TextNet and Flaubert/XLM are unrelated changes." }, { - "left": "pull_request:44059", - "right": "pull_request:44068", + "left": "pull_request:43997", + "right": "pull_request:44018", "accept": false, - "reason": "GPT-2 and GPT-Neo are separate refactors." + "reason": "RegNet and GPT-Neo are different model families." }, { - "left": "pull_request:44068", - "right": "pull_request:44722", + "left": "pull_request:44056", + "right": "pull_request:44161", "accept": false, - "reason": "GPT-Neo and GPT-J/CodeGen are not the same concrete change." + "reason": "MPNet and LongT5 are unrelated." }, { - "left": "pull_request:44071", - "right": "pull_request:44086", + "left": "pull_request:43997", + "right": "pull_request:44161", "accept": false, - "reason": "MPT and MGP-STR are unrelated." + "reason": "RegNet and LongT5 are not the same bug or change." }, { - "left": "pull_request:44017", - "right": "pull_request:44073", + "left": "pull_request:44074", + "right": "pull_request:44161", "accept": false, - "reason": "Segformer and VisualBert are different model implementations." + "reason": "TextNet and LongT5 are different code paths." }, { - "left": "pull_request:44010", - "right": "pull_request:44018", + "left": "pull_request:44018", + "right": "pull_request:44044", "accept": false, - "reason": "SqueezeBert and GPT-Neo are unrelated." + "reason": "GPT-Neo and DeBERTa are unrelated implementations." }, { - "left": "pull_request:43997", - "right": "pull_request:44129", + "left": "pull_request:44010", + "right": "pull_request:44071", "accept": false, - "reason": "RegNet and SpeechT5 are different model code paths." + "reason": "SqueezeBert and MPT are different models." }, { - "left": "pull_request:44071", - "right": "pull_request:44129", + "left": "pull_request:44010", + "right": "pull_request:44073", "accept": false, - "reason": "MPT and SpeechT5 are unrelated changes." + "reason": "SqueezeBert and VisualBert are unrelated." }, { - "left": "pull_request:43996", - "right": "pull_request:44161", + "left": "pull_request:44018", + "right": "pull_request:44129", "accept": false, - "reason": "CVT/FNet and LongT5 are unrelated model-specific refactors." + "reason": "GPT-Neo and SpeechT5 are different code paths." }, { - "left": "pull_request:43996", - "right": "pull_request:44017", + "left": "pull_request:43997", + "right": "pull_request:44018", "accept": false, - "reason": "CVT/FNet and Segformer are different model families." + "reason": "RegNet and GPT-Neo are unrelated." }, { "left": "pull_request:44010", "right": "pull_request:44044", "accept": false, - "reason": "SqueezeBert and DeBERTa v2 are unrelated." + "reason": "SqueezeBert and DeBERTa are different model families." }, { "left": "pull_request:44044", "right": "pull_request:44074", "accept": false, - "reason": "DeBERTa v2 and TextNet are different code paths." + "reason": "DeBERTa and TextNet are unrelated implementations." }, { - "left": "pull_request:44024", - "right": "pull_request:44076", - "accept": false, - "reason": "FocalNet and ImageGPT are unrelated model implementations." + "left": "pull_request:44066", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J/CodeGen output-tracing refactor family; 44722 is the broader later version." }, { "left": "pull_request:44085", "right": "pull_request:44722", "accept": true, - "reason": "Same GPT-J output-tracing change; 44722 is the broader later consolidation that subsumes this file-level fix." + "reason": "Both are GPT-J output-tracing refactors on the same file family, and 44722 subsumes the narrower version." }, { "left": "pull_request:44085", "right": "pull_request:44129", "accept": false, - "reason": "GPT-J and SpeechT5 are unrelated model-specific refactors." + "reason": "GPT-J output tracing and SpeechT5 output tracing are unrelated." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44007", + "accept": false, + "reason": "CVT/FNet and RegNet/ResNet are different model code paths." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44129", + "accept": false, + "reason": "CVT/FNet and SpeechT5 are unrelated changes." }, { "left": "pull_request:44044", "right": "pull_request:44129", "accept": false, - "reason": "DeBERTa v2 and SpeechT5 are different code paths." + "reason": "DeBERTa and SpeechT5 are different model implementations." }, { "left": "pull_request:44044", "right": "pull_request:44154", "accept": false, - "reason": "DeBERTa v2 and Vits are unrelated changes." + "reason": "DeBERTa and Vits are unrelated code paths." }, { "left": "pull_request:44024", "right": "pull_request:44076", "accept": false, - "reason": "FocalNet and ImageGPT are different model families." - }, - { - "left": "pull_request:44007", - "right": "pull_request:44019", - "accept": true, - "reason": "Both include the ResNet output-tracing refactor in the same concrete file; 44007 is the broader superset." - }, - { - "left": "pull_request:44101", - "right": "pull_request:44116", - "accept": true, - "reason": "Both cover the same Flaubert output-tracing change in the same file; 44101 additionally includes XLM." + "reason": "FocalNet and ImageGPT are different models; same refactor pattern only." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative about cross-model overlaps. The accepted soft pairs are all supported by either the same concrete file or a clear broader PR that subsumes the narrower file-level change, so they look mergeable into a single bugfix PR. The rejected cross-model pairs are handled cautiously and are consistent with the evidence." + "feedback": "The summary is grounded in the packet and stays conservative: it only groups the clearly overlapping GPT-J/CodeGen and GPT-Neo variants, while treating the other output-tracing PRs as model-specific changes. The accepted soft-edge pairs appear mergeable as single bugfix iterations within the same family, and the justification for keeping 44722 as the representative GPT-J/CodeGen PR is supported by the diff and issue links." }, "evaluator_used": true, "retried": false, @@ -22653,64 +22615,64 @@ "pull_request:43997|pull_request:44010", "pull_request:44010|pull_request:44129" ], - "prepared_review_unit_hash": "06ad026ccb0114dbbf145150e503fdc2df6607ea7f4cd1319e30f78f11f50ad2", + "prepared_review_unit_hash": "eb263695585382e64c4e445527e1e59ec6b1cc810296a1099196a2d021ba4c24", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 77, - "serialized_chars": 84505, - "estimated_input_tokens": 21127, - "estimated_eval_tokens": 42510 + "serialized_chars": 84500, + "estimated_input_tokens": 21125, + "estimated_eval_tokens": 42506 }, "budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 5, - "serialized_chars": 13069, - "estimated_input_tokens": 3268, - "estimated_eval_tokens": 6792 + "serialized_chars": 13065, + "estimated_input_tokens": 3267, + "estimated_eval_tokens": 6790 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "All six PRs are model-specific refactors around standardized output tracing/capture, but they affect different architectures and code paths, so they are not duplicates of the same change.", - "confidence": 0.97, + "summary": "These PRs are all in the same broad theme\u2014migrating model code to standardized output tracing/capture\u2014but they touch different model implementations and code paths, so they should not be deduplicated into one change.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44129 is the strongest representative because it is the only one tied to an explicit issue and makes a concrete SpeechT5 tracing refactor rather than a purely generic migration.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR 44129 has the clearest scope and traceability, with an explicit issue link and a concrete code-path change in SpeechT5.", + "best_pr_reason": "PR 44129 is the strongest standalone candidate because it has an explicit issue target and a focused SpeechT5-specific refactor, making its scope more concrete than the more generic model migration PRs.", "soft_edge_verdicts": [ { "left": "pull_request:44073", "right": "pull_request:44074", "accept": false, - "reason": "Both are output-tracing refactors, but one is for VisualBert and the other for TextNet; different model code paths mean they are not the same change." + "reason": "Same refactor pattern, but different models (VisualBert vs TextNet) and different files/tests; they are related migrations, not the same underlying code change." }, { "left": "pull_request:44056", "right": "pull_request:44073", "accept": false, - "reason": "These touch different models (MPNet vs VisualBert) and different tracing implementations, so they are not mergeable as one duplicate fix." + "reason": "Both standardize output tracing, but they modify different model internals and are not the same concrete bug fix or mergeable into one PR." }, { "left": "pull_request:44056", "right": "pull_request:44074", "accept": false, - "reason": "MPNet and TextNet are separate model implementations; this is the same refactor theme, not the same underlying bug or code change." + "reason": "Shared output-capture refactor theme only; MPNet and TextNet are separate implementations with distinct forward paths." }, { "left": "pull_request:43997", "right": "pull_request:44010", "accept": false, - "reason": "RegNet and SqueezeBert are unrelated model files with independent output-handling updates, so this is not a duplicate PR pair." + "reason": "RegNet and SqueezeBert are unrelated model code paths; similar decorator migration does not make them duplicates." }, { "left": "pull_request:44010", "right": "pull_request:44129", "accept": false, - "reason": "SqueezeBert output collection and SpeechT5 output tracing/caching are different concrete code paths; shared tooling does not make them duplicates." + "reason": "Both touch output tracing, but SpeechT5 includes a model-specific forward signature change and explicit issue linkage, so it is not the same change as SqueezeBert." } ] }, @@ -22790,274 +22752,274 @@ "pull_request:44028|pull_request:44030", "pull_request:44029|pull_request:44030" ], - "prepared_review_unit_hash": "5a20ed1d0ec6145ebdde55438fc49d476b5a425ec49115ef9e0ba92692bf18e5", + "prepared_review_unit_hash": "eb233522de54da195a69271af531a66139b0f1c8d0ccfd8cca0a539f859e611d", "component_budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 40, - "serialized_chars": 40181, - "estimated_input_tokens": 10046, - "estimated_eval_tokens": 20348 + "serialized_chars": 40171, + "estimated_input_tokens": 10043, + "estimated_eval_tokens": 20342 }, "budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 40, - "serialized_chars": 40181, - "estimated_input_tokens": 10046, - "estimated_eval_tokens": 20348 + "serialized_chars": 40171, + "estimated_input_tokens": 10043, + "estimated_eval_tokens": 20342 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are highly similar textually, but they are independent, model-specific output-tracing refactors across different architectures. They should not be deduplicated as the same change unless they touch the same concrete code path, which they do not.", - "confidence": 0.97, + "summary": "This cluster is a set of open PRs that all refactor output tracing / `can_return_tuple` behavior, but each targets a different model or code path. They look thematically related, not duplicate changes.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "If one PR must represent the cluster, 44025 is the best anchor: it is a self-contained refactor across two closely related depth-anything models and includes tests, making it the most complete example of the pattern.", + "canonical_pr_reason": "PR 44003 is the strongest representative of the theme: it makes a larger tracing refactor across related Mamba codepaths and adds the output-capturing plumbing, so it best captures the cluster\u2019s core change pattern.", "best_issue_reason": null, - "best_pr_reason": "44025 is the strongest representative of the family because it is broader than the single-file refactors and has accompanying test coverage, while still staying within one coherent model family.", + "best_pr_reason": "PR 44003 is the most substantial and representative example of the output-tracing refactor work, with the broadest implementation scope among the candidates.", "soft_edge_verdicts": [ { "left": "pull_request:43998", "right": "pull_request:44002", "accept": false, - "reason": "Different model families and code paths; same output-tracing refactor theme is too broad." + "reason": "Different models and forward paths (`timm_backbone` vs `upernet`); same refactor style, but not the same bug/change." }, { "left": "pull_request:43998", "right": "pull_request:44000", "accept": false, - "reason": "Timm backbone vs vision-text dual encoder are unrelated implementations." + "reason": "Separate model families (`timm_backbone` vs `vision_text_dual_encoder`) with distinct tracing behavior." }, { "left": "pull_request:43998", "right": "pull_request:44001", "accept": false, - "reason": "Different architectures; no shared concrete bug or mergeable code path." + "reason": "Both are output-tracing refactors, but they touch unrelated model codepaths (`timm_backbone` vs `univnet`)." }, { "left": "pull_request:43998", "right": "pull_request:44003", "accept": false, - "reason": "Timm backbone refactor is unrelated to Falcon/Mamba state/output handling." + "reason": "`timm_backbone` and `mamba` are unrelated implementations; this is a shared refactor pattern, not one duplicate fix." }, { "left": "pull_request:43998", "right": "pull_request:44029", "accept": false, - "reason": "Different model code paths; only a broad tracing-refactor similarity." + "reason": "Different architectures (`timm_backbone` vs `rwkv`) and different changes to output handling." }, { "left": "pull_request:43999", "right": "pull_request:44001", "accept": false, - "reason": "MobileNetV1 and UnivNet are separate models with separate output semantics." + "reason": "`mobilenet_v1` and `univnet` are unrelated models; both refactor output tracing but not the same change." }, { "left": "pull_request:43999", "right": "pull_request:44002", "accept": false, - "reason": "MobileNetV1 and UperNet are unrelated code paths." + "reason": "Different models and code paths (`mobilenet_v1` vs `upernet`); no shared concrete bug." }, { "left": "pull_request:43999", "right": "pull_request:44025", "accept": false, - "reason": "Different model families; no same underlying bug or change." + "reason": "These are separate depth/vision model refactors, not one mergeable PR." }, { "left": "pull_request:43999", "right": "pull_request:44026", "accept": false, - "reason": "MobileNetV1 refactor is unrelated to vision encoder-decoder wrapper changes." + "reason": "`mobilenet_v1` and `vision_encoder_decoder` affect different forward flows and output contracts." }, { "left": "pull_request:43999", "right": "pull_request:44027", "accept": false, - "reason": "Different architectures and concrete implementations." + "reason": "Distinct model families (`mobilenet_v1` vs `speech_encoder_decoder`) despite similar tracing cleanup." }, { "left": "pull_request:44000", "right": "pull_request:44001", "accept": false, - "reason": "Vision-text dual encoder and UnivNet are unrelated model paths." + "reason": "`vision_text_dual_encoder` and `univnet` are unrelated; the overlap is only the refactor pattern." }, { "left": "pull_request:44000", "right": "pull_request:44002", "accept": false, - "reason": "Different models; same refactor motif is not enough to merge." + "reason": "Different model implementations and return signatures; not the same underlying issue." }, { "left": "pull_request:44000", "right": "pull_request:44003", "accept": false, - "reason": "Vision-text dual encoder vs Falcon/Mamba output tracing are different changes." - }, - { - "left": "pull_request:44000", - "right": "pull_request:44004", - "accept": false, - "reason": "Different architectures and code paths; not the same fix." + "reason": "`vision_text_dual_encoder` and `mamba` change different code paths and output plumbing." }, { "left": "pull_request:44000", "right": "pull_request:44025", "accept": false, - "reason": "Vision-text dual encoder is unrelated to depth-anything model tracing." + "reason": "Separate models with different output APIs; similar tracing cleanup is too broad to treat as duplicate." }, { "left": "pull_request:44001", "right": "pull_request:44002", "accept": false, - "reason": "UnivNet and UperNet are separate model families." + "reason": "`univnet` and `upernet` are unrelated model families, so these are not the same fix." }, { "left": "pull_request:44001", "right": "pull_request:44003", "accept": false, - "reason": "Different concrete output/caching logic in different models." + "reason": "Different architectures and implementation details; no concrete shared bug." }, { "left": "pull_request:44001", "right": "pull_request:44025", "accept": false, - "reason": "UnivNet refactor is unrelated to depth-anything refactor." + "reason": "These target different model stacks and output contracts, not one duplicate change." }, { "left": "pull_request:44002", - "right": "pull_request:44003", + "right": "pull_request:44029", "accept": false, - "reason": "UperNet segmentation and Falcon/Mamba generation paths are unrelated." + "reason": "`upernet` and `rwkv` are unrelated; same general refactor theme only." }, { "left": "pull_request:44002", - "right": "pull_request:44004", + "right": "pull_request:44003", "accept": false, - "reason": "Different models and output contracts; not mergeable as one fix." + "reason": "Different model families and distinct tracing implementations; not mergeable as one PR." }, { "left": "pull_request:44003", "right": "pull_request:44028", "accept": false, - "reason": "Falcon/Mamba state tracing and SuperPoint feature extraction are unrelated." + "reason": "`mamba`/`falcon_mamba` output tracing is unrelated to `superpoint`; different code paths and APIs." }, { "left": "pull_request:44003", "right": "pull_request:44004", "accept": false, - "reason": "Different concrete code paths: Mamba/Falcon state handling vs CodeGen cache handling." + "reason": "`mamba` and `codegen` have different forward semantics and output shapes; not the same underlying bug." }, { "left": "pull_request:43998", "right": "pull_request:44004", "accept": false, - "reason": "Timm backbone and CodeGen are unrelated model implementations." + "reason": "`timm_backbone` and `codegen` are unrelated codepaths; shared refactor wording is not enough." }, { "left": "pull_request:44001", "right": "pull_request:44004", "accept": false, - "reason": "UnivNet and CodeGen do not share the same bug or change." + "reason": "Different model families and different output-tracing changes." }, { "left": "pull_request:44002", "right": "pull_request:44004", "accept": false, - "reason": "UperNet and CodeGen are different code paths." + "reason": "`upernet` and `codegen` do not share a concrete fix target." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44004", + "accept": false, + "reason": "These are separate model implementations with unrelated output handling." }, { "left": "pull_request:44025", "right": "pull_request:44026", "accept": false, - "reason": "Depth-anything models vs vision encoder-decoder wrappers are different concrete changes." + "reason": "Both are tracing refactors, but `depth_anything` and `vision_encoder_decoder` are different models and APIs." }, { "left": "pull_request:44025", "right": "pull_request:44027", "accept": false, - "reason": "Depth-anything and speech encoder-decoder are unrelated model families." + "reason": "`depth_anything` and `speech_encoder_decoder` are unrelated code paths." }, { "left": "pull_request:44025", "right": "pull_request:44028", "accept": false, - "reason": "Depth-anything refactor is unrelated to SuperPoint output tracing." + "reason": "Different model families and output structures; not a duplicate fix." }, { "left": "pull_request:44025", "right": "pull_request:44029", "accept": false, - "reason": "Depth-anything and RWKV are different architectures and output contracts." + "reason": "`depth_anything` and `rwkv` touch different architectures and tracing mechanisms." }, { "left": "pull_request:44025", "right": "pull_request:44030", "accept": false, - "reason": "Depth-anything refactor is unrelated to DPR encoder output handling." + "reason": "`depth_anything` and `dpr` are distinct model stacks; same broad theme only." }, { "left": "pull_request:44026", - "right": "pull_request:44027", + "right": "pull_request:44029", "accept": false, - "reason": "Vision encoder-decoder and speech encoder-decoder are separate wrapper implementations." + "reason": "`vision_encoder_decoder` and `rwkv` modify unrelated forward/output code." }, { "left": "pull_request:44026", - "right": "pull_request:44028", + "right": "pull_request:44027", "accept": false, - "reason": "Different model families; not the same underlying bug or change." + "reason": "`vision_encoder_decoder` and `speech_encoder_decoder` are separate wrappers with different output handling." }, { "left": "pull_request:44026", - "right": "pull_request:44029", + "right": "pull_request:44028", "accept": false, - "reason": "Vision encoder-decoder wrapper and RWKV generation path are unrelated." + "reason": "Different model families (`vision_encoder_decoder` vs `superpoint`) and different return semantics." }, { "left": "pull_request:44026", "right": "pull_request:44030", "accept": false, - "reason": "Different models and code paths." + "reason": "These are unrelated model implementations with only superficial similarity in refactoring style." }, { "left": "pull_request:44027", "right": "pull_request:44028", "accept": false, - "reason": "Speech encoder-decoder and SuperPoint are unrelated implementations." + "reason": "`speech_encoder_decoder` and `superpoint` are different subsystems and not the same concrete change." }, { "left": "pull_request:44027", "right": "pull_request:44029", "accept": false, - "reason": "Speech encoder-decoder wrapper changes do not match RWKV output tracing." + "reason": "Different architectures and output plumbing; no duplicate fix here." }, { "left": "pull_request:44027", "right": "pull_request:44030", "accept": false, - "reason": "Different architectures and concrete fixes." + "reason": "`speech_encoder_decoder` and `dpr` are distinct models with distinct changes." }, { "left": "pull_request:44028", "right": "pull_request:44029", "accept": false, - "reason": "SuperPoint feature tracing and RWKV sequence modeling are unrelated." + "reason": "`superpoint` and `rwkv` are unrelated code paths, despite both touching output handling." }, { "left": "pull_request:44028", "right": "pull_request:44030", "accept": false, - "reason": "Different model families and output structures." + "reason": "Different models and different forward contracts; not the same underlying bug/change." }, { "left": "pull_request:44029", "right": "pull_request:44030", "accept": false, - "reason": "RWKV and DPR are separate code paths with different output behavior." + "reason": "`rwkv` and `dpr` are separate model families with distinct refactors." } ] }, @@ -23087,12 +23049,12 @@ "soft_pairs": [ "pull_request:44662|pull_request:45401" ], - "prepared_review_unit_hash": "c9358f835f79bd06fb8233b38d0a62a9cbe91c1937b4fbe0c0460159464c1825", + "prepared_review_unit_hash": "bf69bb1bb9f0dfc13896bf2e2cc1d882c939b713c2e77de41eb3a002eddf73ee", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6552, + "serialized_chars": 6551, "estimated_input_tokens": 1638, "estimated_eval_tokens": 3532 }, @@ -23100,7 +23062,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6552, + "serialized_chars": 6551, "estimated_input_tokens": 1638, "estimated_eval_tokens": 3532 }, @@ -23109,18 +23071,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two unrelated model-addition PRs: one adds PenguinVL, the other adds Voxtral TTS. They share some generic integration files, but they target different models, different model codepaths, and only one links a specific issue. No duplicate relationship is present.", - "confidence": 0.98, + "summary": "These are two unrelated model-implementation PRs: one adds PenguinVL, the other adds Voxtral TTS. They share some standard transformers scaffolding, but they do not fix the same code path or represent one underlying change.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR for deduplication: the PRs implement different models and should not be collapsed into one change.", + "canonical_pr_reason": "No canonical PR: the two pull requests introduce different models and have different goals, so they are not duplicates of one another.", "best_issue_reason": null, - "best_pr_reason": "PR #44662 is the better representative artifact only by maturity and review activity, but it is not a duplicate of #45401.", + "best_pr_reason": "PR 45401 is the more self-contained candidate only because it has an explicit issue target and a narrower model-specific scope, but it is not a better duplicate match to 44662.", "soft_edge_verdicts": [ { "left": "pull_request:44662", "right": "pull_request:45401", "accept": false, - "reason": "Reject: PenguinVL and Voxtral TTS are distinct model implementations with different titles, filenames, and likely different codepaths. Overlap is limited to shared Transformers registration/docs plumbing, which is too broad to indicate the same underlying change." + "reason": "Reject: PenguinVL and Voxtral TTS are distinct model additions with different files, configs, and tests; shared auto-registration/docs infrastructure is generic and not evidence of the same underlying bug or change." } ] }, @@ -23150,7 +23112,7 @@ "soft_pairs": [ "pull_request:44965|pull_request:46845" ], - "prepared_review_unit_hash": "8866b115fbc0df34b9b0278ac2586f8ad455be36235585fcab6bf425c4750c64", + "prepared_review_unit_hash": "e8bb4d101a0300222c0f7c554169c0414b1c76f5861dc2915f1be254a214cbfc", "component_budget": { "node_count": 2, "item_count": 2, @@ -23172,18 +23134,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two pull requests are not duplicates: one is a workflow change to route PR CI to a smaller runner/branch target, while the other is an unrelated attempt to convert PRs to draft via a documentation workflow. They do not target the same code path or same underlying fix.", - "confidence": 0.97, + "summary": "These two PRs are not duplicates: one is about changing a PR documentation workflow / converting PRs to draft, while the other is about switching CI to a smaller runner group and includes an unrelated code tweak in `modeling_utils.py`. They do not fix the same concrete problem and would not plausibly be merged into one PR.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46845 is the more coherent and concrete change set, but it is not a duplicate of PR 44965; the two PRs address different automation paths and should remain separate.", + "canonical_pr_reason": "PR 46845 is the better representative because its title and diff are more specific and it contains a coherent CI-related change, whereas PR 44965 is titled generically and appears to be an unrelated workflow experiment.", "best_issue_reason": null, - "best_pr_reason": "PR 46845 is the better representative of the cluster because it has a clearer, self-contained CI-related purpose and an explicit branch/workflow adjustment, whereas PR 44965 is a different draft-conversion workflow change.", + "best_pr_reason": "PR 46845 is the stronger cluster representative due to clearer intent and more substantive, identifiable changes across CI config and supporting files.", "soft_edge_verdicts": [ { "left": "pull_request:44965", "right": "pull_request:46845", "accept": false, - "reason": "Different underlying automation changes: 44965 touches PR documentation/draft conversion workflow, while 46845 changes PR CI routing and adds unrelated code/test churn. They are not the same bug or mergeable as one PR." + "reason": "Different underlying changes: 44965 alters the PR documentation/draft workflow, while 46845 changes CI routing to a smaller group and touches `modeling_utils.py`. They are not the same bug or change." } ] }, @@ -23213,12 +23175,12 @@ "soft_pairs": [ "pull_request:44979|pull_request:45363" ], - "prepared_review_unit_hash": "55c4f258eca2e1b525148b1d9d7a30eb13b24b8e6db87fe21fcb35c42c094f24", + "prepared_review_unit_hash": "617920c8a8addd1247bc2c82f1400e7b258c06d8bbec74716dada39108defe65", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4408, + "serialized_chars": 4407, "estimated_input_tokens": 1102, "estimated_eval_tokens": 2460 }, @@ -23226,7 +23188,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4408, + "serialized_chars": 4407, "estimated_input_tokens": 1102, "estimated_eval_tokens": 2460 }, @@ -23235,18 +23197,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two PRs are related to kernel/module fusion infrastructure, but they implement different layers of the stack and do not look like the same mergeable change. PR #44979 adds a standalone Module Fusion API and tests; PR #45363 adds n-to-1 kernel fusion wiring via KernelConfig across integrations/modeling utilities.", - "confidence": 0.94, + "summary": "These are related kernel-fusion efforts but not the same change: PR #44979 introduces a generic module fusion API, while PR #45363 implements n-to-1 kernel fusion via `KernelConfig` and integration plumbing. They share theme and terminology, but the code paths and deliverables are different enough that they should not be deduplicated as one PR.", + "confidence": 0.88, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45363 is the better canonical representative for this cluster because it is the broader, end-user-facing fusion change and touches the runtime integration path rather than just introducing a lower-level API scaffold.", + "canonical_pr_reason": "PR #45363 is the better canonical PR for this cluster because it is the more concrete, user-facing kernel-fusion implementation (`KernelConfig`, integration hooks, example), whereas #44979 is a lower-level generic module-fusion API foundation.", "best_issue_reason": null, - "best_pr_reason": "PR #45363 is the best overall fit: it has the clearest concrete behavior change (KernelConfig-driven fusion in model loading/execution) and is more likely to anchor follow-up discussion than the smaller API-only foundation PR.", + "best_pr_reason": "PR #45363 is the best representative of the cluster\u2019s apparent intent: enabling kernel fusion behavior through `KernelConfig` and related runtime integration. PR #44979 is adjacent infrastructure, but not the same concrete feature change.", "soft_edge_verdicts": [ { "left": "pull_request:44979", "right": "pull_request:45363", "accept": false, - "reason": "Different code paths and scopes: #44979 introduces a general Module Fusion API in a new module, while #45363 implements n-to-1 kernel fusion through KernelConfig and integration hooks. Related, but not the same underlying change and not plausibly one merged PR." + "reason": "Related area, but not the same underlying change: #44979 adds a generic module fusion API, while #45363 wires up n-to-1 kernel fusion through `KernelConfig` and model integration. They could coexist in a larger roadmap, but they are not duplicates and would not plausibly be merged as one PR." } ] }, @@ -23276,7 +23238,7 @@ "soft_pairs": [ "pull_request:45086|pull_request:45317" ], - "prepared_review_unit_hash": "b56b14de3cfd1956a4b112c7dfaa72472769978f8a25a25342556eefe297eed7", + "prepared_review_unit_hash": "328e30ad130ec62e1168e432302d04cbf9e104d8da910eb75234d419646c0094", "component_budget": { "node_count": 2, "item_count": 2, @@ -23298,18 +23260,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs patch the same AttributeError in `_patch_mistral_regex` by switching from `tokenizer.backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer` and updating the sequence assignment accordingly; one PR just adds tests and a slightly broader diff around the same fix.", - "confidence": 0.99, + "summary": "Both PRs fix the same Mistral regex pre-tokenizer AttributeError in `tokenization_utils_tokenizers.py`; `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 contains the same core fix as #45086 plus test coverage, making it the more complete representation of the change.", + "canonical_pr_reason": "`45317` is the stronger canonical PR: it makes the same code-path fix as `45086` and also includes regression tests, so it is the more complete representation of the change.", "best_issue_reason": null, - "best_pr_reason": "PR #45317 is the best representative because it fixes the same concrete code-path problem as #45086 and includes validation via tests, so it is the strongest standalone record of the change.", + "best_pr_reason": "`45317` best captures the cluster\u2019s final fix because it addresses the same AttributeError and includes test coverage, making it the most suitable PR to keep.", "soft_edge_verdicts": [ { "left": "pull_request:45086", "right": "pull_request:45317", "accept": true, - "reason": "Accept: these PRs make the same exact code change in the same file/function to fix the same AttributeError, and the right PR simply adds tests around the same fix." + "reason": "Same underlying bug fix: both change the same `_patch_mistral_regex` pre-tokenizer logic to avoid the AttributeError. `45317` only adds tests, so this is a true duplicate/superset pair." } ] }, @@ -23331,48 +23293,56 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-45115-2", + "cluster_id": "cluster-45115-3", "nodes": [ "pull_request:45115", - "pull_request:45569" + "pull_request:45569", + "pull_request:46565" ], "soft_pairs": [ - "pull_request:45115|pull_request:45569" + "pull_request:45115|pull_request:45569", + "pull_request:45569|pull_request:46565" ], - "prepared_review_unit_hash": "f0e6332fac5f2e9bbab1fc838db6bde1d843efec31d5488637067b433f01637e", + "prepared_review_unit_hash": "63271d45ee6da55a94ecee8cf34ff238036080f459bcd6b5e80b64a2f08631a4", "component_budget": { - "node_count": 2, - "item_count": 2, - "soft_pair_count": 1, - "serialized_chars": 6066, - "estimated_input_tokens": 1517, - "estimated_eval_tokens": 3290 + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 9767, + "estimated_input_tokens": 2442, + "estimated_eval_tokens": 5140 }, "budget": { - "node_count": 2, - "item_count": 2, - "soft_pair_count": 1, - "serialized_chars": 6066, - "estimated_input_tokens": 1517, - "estimated_eval_tokens": 3290 + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 9767, + "estimated_input_tokens": 2442, + "estimated_eval_tokens": 5140 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both items are NemotronH-related PRs, but they address different changes: one refactors conversion/mapping compatibility, while the other splits NemotronH into dense/sparse model variants and updates docs/auto-mapping.", - "confidence": 0.91, + "summary": "These are three distinct PRs around Nemotron-related model support. PR 45569 is the broadest and most complete Nemotron H architecture split/refactor, while 45115 is a narrower conversion/inheritance refactor and 46565 is an unrelated Nemotron 3.5 ASR feature PR. I do not see a true duplicate pair here.", + "confidence": 0.78, "canonical_issue_reason": null, - "canonical_pr_reason": "No clear canonical PR for duplication: #45115 is a conversion-mapping refactor, while #45569 is a model-architecture split and documentation/API update.", + "canonical_pr_reason": "PR 45569 is the best canonical PR because it is the most substantial and complete Nemotron H change set, covering the architecture split, auto-mapping updates, docs, and tests; the other Nemotron H PR is narrower and looks like an incremental precursor rather than the same final change.", "best_issue_reason": null, - "best_pr_reason": "#45569 is the stronger representative of the main product change because it introduces the new dense/sparse NemotronH model structure and corresponding integration work.", + "best_pr_reason": "PR 45569 is the strongest representative of the cluster: it has the broadest implementation scope, the most discussion activity, and the clearest end-to-end model integration work.", "soft_edge_verdicts": [ { "left": "pull_request:45115", "right": "pull_request:45569", "accept": false, - "reason": "They share the NemotronH subsystem, but not the same underlying fix. #45115 only adds conversion-key renames for checkpoint loading, whereas #45569 reworks the model/package layout into dense and sparse variants plus docs/tests. These could not plausibly be merged as one PR without mixing distinct work." + "reason": "They overlap on Nemotron H internals, but 45115 is a focused refactor/conversion-mapping and inheritance change, while 45569 is a broader split into NemotronH dense/sparse packages with docs and tests. They are related, but not the same concrete change and would not cleanly merge as one PR." + }, + { + "left": "pull_request:45569", + "right": "pull_request:46565", + "accept": false, + "reason": "These target different models and problems: Nemotron H architecture refactoring versus Nemotron 3.5 ASR support. Shared auto/docs files are too broad to indicate a duplicate." } ] }, @@ -23402,7 +23372,7 @@ "soft_pairs": [ "pull_request:45332|pull_request:45333" ], - "prepared_review_unit_hash": "c6bc5fe1c029dde9faaf12eb3df4567fbeeb2378b17d5727a4995521a961a653", + "prepared_review_unit_hash": "9f64ac059a920aebb1e7f189e49c169cefbada8e7bdea7c3b20a432cf6c370e3", "component_budget": { "node_count": 2, "item_count": 2, @@ -23424,18 +23394,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related feature PRs for heterogeneous model support, but they are not duplicates: #45333 is a config-only slice, while #45332 includes config plus broader modeling/runtime integration. I would keep #45332 as the representative PR for the cluster and reject the soft-duplicate link.", - "confidence": 0.91, + "summary": "These are closely related feature PRs, but not duplicates: #45333 appears to be the configuration-only subset, while #45332 adds the broader heterogeneous modeling/runtime support on top of that. I would keep them separate and use #45332 as the cluster representative.", + "confidence": 0.83, "canonical_issue_reason": null, - "canonical_pr_reason": "#45332 is the broader, more complete implementation: it covers per-layer config plus modeling/runtime changes across cache, masking, mxfp4, and heterogeneous modeling utilities, so it best represents the underlying feature work.", + "canonical_pr_reason": "#45332 is the broader, more complete PR: it includes the per-layer config work plus modeling/runtime integration, cache/masking changes, and related tests/docs. #45333 is a narrower config-only slice.", "best_issue_reason": null, - "best_pr_reason": "#45332 is the best PR to anchor the cluster because it subsumes the config work and extends into the core code paths needed for actual heterogeneous model support.", + "best_pr_reason": "#45332 is the best representative because it covers the full heterogeneous model support feature rather than only the configuration portion.", "soft_edge_verdicts": [ { "left": "pull_request:45332", "right": "pull_request:45333", "accept": false, - "reason": "They share the same feature area and some files, but #45333 is only per-layer configuration while #45332 also adds modeling/runtime support. That makes #45333 a subset, not the same concrete PR-level change." + "reason": "Related feature work, but not the same concrete change: #45333 is config-only, while #45332 includes additional modeling/runtime code paths. They look like a subset/superset pair, not duplicate PRs that should be merged into one." } ] }, @@ -23465,12 +23435,12 @@ "soft_pairs": [ "pull_request:45415|pull_request:45425" ], - "prepared_review_unit_hash": "bb1ce2ec1b2af41f1f415c3778ed90eac49c748e6290c7c41765f4065b02ff99", + "prepared_review_unit_hash": "2736b7b6836cee37ef71cc7e6f27a1447dcbc3c4414090777f5f1f50d446eb0e", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5513, + "serialized_chars": 5514, "estimated_input_tokens": 1379, "estimated_eval_tokens": 3014 }, @@ -23478,7 +23448,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5513, + "serialized_chars": 5514, "estimated_input_tokens": 1379, "estimated_eval_tokens": 3014 }, @@ -23487,18 +23457,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are both typing-focused PRs in transformers, but they address different scopes: #45415 is a broad type-checking sweep across many files, while #45425 adds/adjusts specific typing protocols and annotations around modeling_utils and related helpers. Shared typing infrastructure is not enough to make them the same change.", - "confidence": 0.86, + "summary": "These are both typing-focused pull requests, but they are not the same underlying change: #45415 is a broad type-checking sweep across many transformers modules, while #45425 adds shared typing protocols/aliases and related typing fixes. The overlap in a few shared files looks incidental rather than duplicate work.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "#45425 is the more focused, concrete typing change set; #45415 is a larger repo-wide type-checking cleanup and not the same underlying change.", + "canonical_pr_reason": "No single canonical PR: the shared files reflect overlapping typing infrastructure, but the diffs target different scopes and would not naturally collapse into one merged change.", "best_issue_reason": null, - "best_pr_reason": "#45425 is the better representative PR for this cluster because it makes a narrower, self-contained typing fix rather than a broad linting/type-check sweep.", + "best_pr_reason": "#45425 is the better representative of the cluster\u2019s narrower typing-infrastructure work; #45415 is a much broader repo-wide type-checking sweep.", "soft_edge_verdicts": [ { - "left": "pull_request:45415", - "right": "pull_request:45425", + "left": "pull_request:45415 (Adds type checking to `src/transformers/*py`)", + "right": "pull_request:45425 (chore(typing): added modeling_utils to ty)", "accept": false, - "reason": "Both are typing-related, but they fix different problems at different scopes. #45415 is a wide sweep to add type checking across many modules and silence type-check issues, while #45425 introduces specific typing protocols/aliases for modeling_utils and related integrations. They do not look mergeable as one PR." + "reason": "Reject: both touch typing-related code, but #45415 is a broad type-checking pass across many files, whereas #45425 adds specific typing protocol/alias support and targeted fixes. They are not the same concrete bug/change and would not plausibly be one PR." } ] }, @@ -23528,7 +23498,7 @@ "soft_pairs": [ "pull_request:45548|pull_request:45662" ], - "prepared_review_unit_hash": "50adef346135bb2fb9bd94d598c43001bf14ee390cb6ed197af01f39c891c4c2", + "prepared_review_unit_hash": "490f399ce59d210fee618ac65357427766137792b9a7e8af5c4b7a4010d6a181", "component_budget": { "node_count": 2, "item_count": 2, @@ -23550,18 +23520,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two PRs touch expert-parallel loading paths, but they fix different failure modes: one is about preserving EP-related config and launch-time loading for DeepSpeed ZeRO-3, while the other prevents EP experts from being overwritten during FSDP2/rank-0 broadcast. They are related by subsystem, not the same underlying change.", + "summary": "These PRs are related in theme (expert-parallel loading bugs) and both touch `modeling_utils.py`, but they fix different failure modes on different distributed backends: one targets DeepSpeed ZeRO-3 loading via `accelerate launch`, the other targets FSDP2/rank-0 broadcast overwriting expert weights. They are not duplicates.", "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45662 is the better canonical PR for this cluster because it targets a concrete EP loading corruption bug in the weight-loading path and spans the relevant post-shard / distributed tensor handling. It is the more substantive and behaviorally specific fix.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR #45662 is the best representative of the cluster because it addresses a clearer end-to-end correctness problem in the model-loading code path, whereas PR #45548 is a narrower config/launch integration fix.", + "best_pr_reason": "PR #45662 is the better representative of the cluster: it addresses a concrete expert-weight overwrite bug with a clearer root cause and broader EP/FSDP2 loading impact than the more backend-specific ZeRO-3 loading fix in #45548.", "soft_edge_verdicts": [ { - "left": "pull_request:45548", - "right": "pull_request:45662", + "left": "Fix EP + DeepSpeed ZeRO-3 loading via accelerate launch", + "right": "Fix EP + FSDP2: experts silently overwritten by rank-0 broadcast", "accept": false, - "reason": "Both involve expert parallelism, but they fix different bugs and code paths: #45548 preserves distributed config and adds an EP flag for accelerate/ZeRO-3 loading, while #45662 fixes experts being overwritten after shard wrapping in FSDP2. Not mergeable as one PR." + "reason": "Different underlying bugs and code paths: ZeRO-3/accelerate loading versus FSDP2 rank-0 broadcast overwriting experts. Shared EP/modeling_utils context is not enough to treat them as the same change." } ] }, @@ -23591,7 +23561,7 @@ "soft_pairs": [ "pull_request:45639|pull_request:45645" ], - "prepared_review_unit_hash": "c4eff0bb07013c6542e403b08445f3e8da8df8b54921f871c4c9ebb30fbde358", + "prepared_review_unit_hash": "0b9d444b00d1c81a54637da1c257740df4db7d4bcec0ac8fef7c66deba482ee0", "component_budget": { "node_count": 2, "item_count": 2, @@ -23613,18 +23583,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs target the same xdist-related captured_info logging problem in transformers testing utilities; PR 45645 is the broader, more complete version because it includes the core xdist-safe file handling plus the CI workflow change to aggregate multiple captured_info files.", - "confidence": 0.88, + "summary": "Both PRs target the same xdist-related captured_info logging problem and share the same code paths/files. PR 45645 is the broader, more complete fix because it includes the xdist-safe output change plus CI workflow handling for multiple log files.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45645 is the better canonical PR because it covers the underlying fix in testing_utils and also updates the CI workflow to read all xdist-split captured_info artifacts, making the end-to-end solution more complete.", + "canonical_pr_reason": "PR 45645 is the best canonical PR: it covers the underlying xdist collision fix in testing_utils and also updates CI to collect all captured_info*.txt files, making the end-to-end logging behavior complete.", "best_issue_reason": null, - "best_pr_reason": "PR 45645 is the strongest representative of the change set: it fixes the xdist-safe output path and preserves CI debug log collection across multiple worker files, whereas 45639 appears to cover only the code-side portion.", + "best_pr_reason": "PR 45645 is the strongest standalone representative because it addresses the same root bug and additionally preserves debug logs in workflow consumption, whereas 45639 is just the library-side half of the fix.", "soft_edge_verdicts": [ { "left": "pull_request:45639", "right": "pull_request:45645", "accept": true, - "reason": "Same underlying bug: captured_info logging breaks under xdist. 45645 looks like a superset of 45639 by including the same testing_utils fix plus the workflow-side aggregation of split log files, so they could plausibly be merged into one PR." + "reason": "Same underlying xdist/captured_info collision issue, overlapping testing_utils changes, and both are plausibly part of one merged fix; 45645 extends 45639 with CI log collection." } ] }, @@ -23654,12 +23624,12 @@ "soft_pairs": [ "pull_request:45720|pull_request:46259" ], - "prepared_review_unit_hash": "5ba90fed84929d18bddad0ac449ae14298bfeed26c270e5403b55b031364e63f", + "prepared_review_unit_hash": "c8344faf420d90a1998f681fbce3c2b37b166d58d3ba0d26f54797c4315fea6a", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 7234, + "serialized_chars": 7233, "estimated_input_tokens": 1809, "estimated_eval_tokens": 3874 }, @@ -23667,7 +23637,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 7234, + "serialized_chars": 7233, "estimated_input_tokens": 1809, "estimated_eval_tokens": 3874 }, @@ -23676,18 +23646,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two PRs are both linter-rule changes, but they implement different rules on different code paths (TRF003 vs TRF018). They should not be treated as duplicates.", - "confidence": 0.96, + "summary": "Two open PRs in transformers, but they target different linter rules and different code changes: one adds can_return_tuple to many model forwards, the other refactors weight initialization across many model classes. They are related only by broad modeling/linter cleanup, not the same underlying fix.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #46259 is the better cluster representative because it adds the more substantial modeling-rule implementation across many model files, while #45720 is a narrower partial TRF003 change.", + "canonical_pr_reason": "PR 45720 is the better anchor: it has a tighter, more specific code-path change (forward return handling via can_return_tuple) and is easier to compare against true duplicates than the broader initialization-rule sweep.", "best_issue_reason": null, - "best_pr_reason": "PR #46259 is the strongest standalone candidate: it has a broader, clearer scope and a more complete rule implementation, making it more suitable as the representative PR for this cluster.", + "best_pr_reason": "PR 45720 is the best representative of the cluster because it is the more narrowly scoped and concrete implementation of a single rule, whereas 46259 is a separate linter-driven refactor with different behavior.", "soft_edge_verdicts": [ { "left": "pull_request:45720", "right": "pull_request:46259", "accept": false, - "reason": "Reject: these are different lint rules with different intended behavior. #45720 implements part of TRF003, while #46259 adds TRF018; they do not appear to fix the same concrete bug or change." + "reason": "Different linter rules and different fixes: TRF003 adds can_return_tuple / return-dict handling in forward methods, while TRF018 changes model weight initialization logic. They do not patch the same concrete bug or code path and would not be merged as one PR." } ] }, @@ -23786,12 +23756,12 @@ "pull_request:45932|pull_request:46891", "pull_request:46100|pull_request:46891" ], - "prepared_review_unit_hash": "74cbde91c9343bcab1d2ba59e55a9ab7b4ca9a8704db2853d15f74f99cd10ea3", + "prepared_review_unit_hash": "9a1c9345cd02e5fc24201ea6b95f7435a9e998ab594f4893b6e5d6adfa6a60f5", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 7883, + "serialized_chars": 7882, "estimated_input_tokens": 1971, "estimated_eval_tokens": 4198 }, @@ -23799,7 +23769,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 7883, + "serialized_chars": 7882, "estimated_input_tokens": 1971, "estimated_eval_tokens": 4198 }, @@ -23808,30 +23778,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three unrelated open test/CI-oriented PRs that only overlap superficially in labels and workflow/testing areas. They do not appear to fix the same underlying problem or belong in one duplicate cluster.", - "confidence": 0.92, + "summary": "These PRs are all test/CI-oriented, but they target different concrete changes: grafana/CI secrets plus model tweaks, a kernel test migration, and a PEFT integration CI branch. They are not the same underlying bug or change, so this is not a true duplicate cluster.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45932 is the broadest and most central of the three, with the largest surface area and explicit CI/workflow changes, but it is not a true duplicate of the others.", + "canonical_pr_reason": "No clear canonical PR: the three PRs are unrelated in scope and code path, with only superficial overlap in testing/CI infrastructure.", "best_issue_reason": null, - "best_pr_reason": "#45932 is the best representative PR in this set because it is the most substantial and most centrally about CI/test plumbing, whereas #46100 is a test migration and #46891 is a separate PEFT integration CI experiment.", + "best_pr_reason": "PR 46891 is the narrowest and most self-contained of the three, but it still does not match the others as a duplicate; it is only the least broad representative if one must be chosen.", "soft_edge_verdicts": [ { "left": "pull_request:45932", "right": "pull_request:46100", "accept": false, - "reason": "Superficial overlap only: both are test-related, but #45932 changes model code and CI secrets, while #46100 adds kernel/model tests and a DDP YAML. Different concrete changes and no shared bug path." + "reason": "Both are test/CI flavored, but 45932 is a grafana/CI-secrets plus model-mapping change while 46100 is a kernel test migration; they do not fix the same code-path problem." }, { "left": "pull_request:45932", "right": "pull_request:46891", "accept": false, - "reason": "Both touch the CI caller workflow, but #45932 is a grafana/testing PR with model mapping changes, while #46891 is a PEFT integration test/CI branch update. Same subsystem, different change intent and code path." + "reason": "Shared workflow filename is not enough here: 45932 adds grafana-related CI secrets and model edits, while 46891 is a PEFT integration CI branch with peft.py/tests_fetcher changes." }, { "left": "pull_request:46100", "right": "pull_request:46891", "accept": false, - "reason": "Only shared theme is testing. #46100 is a broad test migration across vision models; #46891 is specifically about PEFT integration CI and related integration code. Not the same underlying change." + "reason": "46100 is a broad kernel test migration, whereas 46891 is a PEFT integration CI/test update; different functionality and no common concrete bug or merged change." } ] }, @@ -23861,7 +23831,7 @@ "soft_pairs": [ "pull_request:46126|pull_request:46269" ], - "prepared_review_unit_hash": "db55ce1e3880d9c4c7b2baae9c0df521c6a1beb5b018d0aa9f6e2ec1cf1f92fc", + "prepared_review_unit_hash": "1550df42fe245d17692552a8f0b6b2867eb4904d56d24422b679ed5f456d366e", "component_budget": { "node_count": 2, "item_count": 2, @@ -23883,18 +23853,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates: PR 46126 is a focused EP/Trainer integration change, while PR 46269 is a broad distributed-branch base that happens to touch overlapping distributed files but represents a much larger, different change set.", - "confidence": 0.92, + "summary": "Two open PRs touch overlapping distributed files, but they are not the same change: one is a focused EP/Trainer integration PR, the other is a broad distributed branch base with many unrelated edits.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46126 is the more canonical artifact for this cluster because it describes a specific, coherent feature/change with a clear scope; PR 46269 is a broad base branch with many unrelated distributed updates and CI/config churn.", + "canonical_pr_reason": "PR #46126 is the most focused and representative change in the cluster: it specifically implements EP + Trainer integration on top of DistributedConfig. PR #46269 is a broad branch-base PR spanning many distributed/FSDP areas and is not a clean canonical duplicate target.", "best_issue_reason": null, - "best_pr_reason": "PR 46126 is the best representative PR here: it targets one concrete code path and could be reviewed/merged on its own, unlike PR 46269 which is too sprawling to serve as the canonical duplicate target.", + "best_pr_reason": "PR #46126 is the best standalone PR to keep as the cluster representative because it has a single concrete purpose and tighter scope than #46269.", "soft_edge_verdicts": [ { "left": "pull_request:46126", "right": "pull_request:46269", "accept": false, - "reason": "Reject: the overlap is only at the level of shared distributed filenames. PR 46126 implements EP + Trainer integration, whereas PR 46269 is a wide distributed branch base touching CI, FSDP, modeling, and multiple integration paths; they are not the same underlying fix/change and would not plausibly be merged as one PR." + "reason": "They overlap on distributed filenames, but the PRs are not the same underlying change. #46126 is a targeted EP/Trainer integration, while #46269 is a large distributed branch base with many unrelated modifications; they would not plausibly merge into one PR." } ] }, @@ -23960,12 +23930,12 @@ "pull_request:46504|pull_request:46563", "pull_request:46504|pull_request:46716" ], - "prepared_review_unit_hash": "8f152e3bcec53dc722aa5106a0a0c4c6688499307cc2b0c643fdef4a16010dcb", + "prepared_review_unit_hash": "f5069f0f3b5dbbb67f1ce25e67993f4ea45c3c2858cbee4d33ae2acd06e3e61f", "component_budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 29, - "serialized_chars": 25931, + "serialized_chars": 25932, "estimated_input_tokens": 6483, "estimated_eval_tokens": 13222 }, @@ -23973,7 +23943,7 @@ "node_count": 10, "item_count": 10, "soft_pair_count": 29, - "serialized_chars": 25931, + "serialized_chars": 25932, "estimated_input_tokens": 6483, "estimated_eval_tokens": 13222 }, @@ -23982,198 +23952,186 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are mostly unrelated workflow-maintenance PRs that only look similar at the title level. Even the repeated pr-ci-security-gate.yml changes address different concerns, so there is no safe duplicate merge here.", - "confidence": 0.96, + "summary": "This cluster is mostly heterogeneous CI/workflow maintenance. I found no true duplicate PR pairs; the only tight subcluster is the pr-ci-security-gate.yml series, but those PRs are separate hardening tweaks rather than the same fix.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical duplicate PR stands out because the cluster is heterogeneous. If one representative is needed, PR 46554 is the most substantive and central security-gate workflow change, but it is not a duplicate of the others.", + "canonical_pr_reason": "No canonical PR: the cluster spans unrelated workflow chores, and the pr-ci-security-gate PRs are distinct edits to the same file rather than duplicates.", "best_issue_reason": null, - "best_pr_reason": "PR 46554 is the best single representative because it makes the core pr-ci-security-gate workflow logic change and has the strongest overlap with the other security-gate variants, though the group is not a true duplicate set.", + "best_pr_reason": "PR 46554 is the strongest representative of the only coherent subtheme (pr-ci-security-gate), because it contains the main functional workflow logic change.", "soft_edge_verdicts": [ { "left": "pull_request:46138", "right": "pull_request:46554", "accept": false, - "reason": "Different workflow files and different changes: self-comment CI stdin handling vs security-gate checks." + "reason": "Different workflows and concerns: stdin handling for self-comment CI vs security-gate logic changes." }, { "left": "pull_request:46138", "right": "pull_request:46558", "accept": false, - "reason": "Different workflow files; one updates comment parsing, the other changes PR CI gate permissions/approval flow." + "reason": "Different workflow files and different behavior; not the same underlying bug or change." }, { "left": "pull_request:46138", "right": "pull_request:46563", "accept": false, - "reason": "Unrelated workflows: self-comment CI command plumbing vs adding safety notes to the security gate reusable workflow." + "reason": "Self-comment CI input handling vs a comment-only security-gate update; unrelated patches." }, { "left": "pull_request:46138", "right": "pull_request:46716", "accept": false, - "reason": "Different workflow files and unrelated changes; not the same underlying bug or fix." + "reason": "Different workflow integrations: self-comment CI vs reusable-workflow secrets handling." }, { "left": "pull_request:46138", "right": "pull_request:46504", "accept": false, - "reason": "Different workflow files and different behavior; only a broad CI-workflow similarity." + "reason": "Same broad CI area, but one changes comment piping and the other changes commit-message parsing for image tagging." }, { "left": "pull_request:46432", "right": "pull_request:46503", "accept": false, - "reason": "SSH runner hardening vs benchmark trigger logic; same subsystem is too broad to merge." + "reason": "SSH-runner hardening vs benchmark workflow dispatch behavior; no same code path." }, { "left": "pull_request:46432", "right": "pull_request:46501", "accept": false, - "reason": "Unrelated workflow changes in different files: SSH runner hardening vs anti-slop permissions layout." + "reason": "SSH runner security changes vs anti-slop workflow permissions; unrelated changes." }, { "left": "pull_request:46432", "right": "pull_request:46502", "accept": false, - "reason": "Different workflows and problems; SSH runner secret handling is not the same as reviewer assignment checkout behavior." + "reason": "SSH runner hardening vs reviewer-assignment checkout behavior; not the same fix." }, { "left": "pull_request:46432", "right": "pull_request:46554", "accept": false, - "reason": "Different files and distinct fixes; SSH runner environment cleanup is unrelated to security-gate logic." + "reason": "SSH-runner changes are unrelated to pr-ci-security-gate workflow logic." }, { "left": "pull_request:46432", "right": "pull_request:46558", "accept": false, - "reason": "SSH runner hardening is unrelated to PR CI security-gate permissions/approval changes." + "reason": "Different workflows and different concerns; not a duplicate patch." }, { "left": "pull_request:46501", "right": "pull_request:46503", "accept": false, - "reason": "Different workflow files and different triggers/permissions; similar title only." + "reason": "Anti-slop permission scoping vs benchmark_v2 manual dispatch support; unrelated workflow updates." }, { "left": "pull_request:46501", "right": "pull_request:46502", "accept": false, - "reason": "Anti-slop permissions and reviewer checkout are separate workflow concerns, not one bug." + "reason": "Anti-slop permissions change vs base-ref checkout for assign-reviewers; different fixes." }, { "left": "pull_request:46501", "right": "pull_request:46504", "accept": false, - "reason": "Different workflow files and distinct behavior changes; not mergeable as one fix." + "reason": "Anti-slop workflow permissions vs build-ci-docker-images commit-message handling; not the same bug." }, { "left": "pull_request:46501", "right": "pull_request:46554", "accept": false, - "reason": "Different workflow files; anti-slop permissions vs security-gate code-path changes." + "reason": "Different workflows: anti-slop permissions vs security-gate checkout/check logic." }, { "left": "pull_request:46502", "right": "pull_request:46503", "accept": false, - "reason": "Reviewer checkout ref and benchmark trigger conditions are unrelated." + "reason": "Reviewer-assignment checkout change vs benchmark dispatch addition; unrelated." }, { "left": "pull_request:46502", "right": "pull_request:46504", "accept": false, - "reason": "Different workflow files and unrelated changes; only superficial CI similarity." + "reason": "Assign-reviewers base-ref checkout vs build-image tagging logic; different code paths." }, { "left": "pull_request:46502", "right": "pull_request:46554", "accept": false, - "reason": "Assign-reviewers checkout behavior is unrelated to security-gate logic." + "reason": "Assign-reviewers workflow tweak vs security-gate workflow hardening; not duplicates." }, { "left": "pull_request:46503", "right": "pull_request:46504", "accept": false, - "reason": "Benchmark dispatch trigger and build-ci docker commit parsing are distinct workflow fixes." + "reason": "Benchmark dispatch support vs build-ci-docker-images message parsing; different functionality." }, { "left": "pull_request:46503", "right": "pull_request:46554", "accept": false, - "reason": "Benchmark runner trigger logic is unrelated to PR CI security gate changes." + "reason": "Benchmark workflow trigger change vs security-gate logic; unrelated patches." }, { "left": "pull_request:46504", "right": "pull_request:46554", "accept": false, - "reason": "Build-ci tag selection from commit messages is unrelated to security-gate workflow hardening." + "reason": "Build-image tagging logic vs pr-ci-security-gate checks; different workflows and fixes." }, { "left": "pull_request:46504", "right": "pull_request:46558", "accept": false, - "reason": "Different workflow files and fixes; no shared underlying defect." + "reason": "Build-image commit parsing vs security-gate permission/approval changes; not the same change." }, { "left": "pull_request:46554", "right": "pull_request:46558", "accept": false, - "reason": "Same file, but different concrete changes: one adds checkout and file-change checks, the other adjusts permissions and approval gating." + "reason": "Same file, but one adds checkout/diff-based checks while the other adjusts permissions and approval gating." }, { "left": "pull_request:46554", "right": "pull_request:46563", "accept": false, - "reason": "Same workflow file, but one is security logic and the other is explanatory comments only." + "reason": "Same workflow file, but one changes gate logic and the other only adds explanatory comments." }, { "left": "pull_request:46554", "right": "pull_request:46716", "accept": false, - "reason": "Different security-gate edits: one adds checks, the other changes secrets inheritance; not the same change." + "reason": "Same workflow family, but one changes security-gate behavior and the other changes secrets handling for the reusable workflow call." }, { "left": "pull_request:46558", "right": "pull_request:46563", "accept": false, - "reason": "Permissions/approval logic vs comment-only update; not the same underlying fix." + "reason": "Both touch pr-ci-security-gate.yml, but permissions/if-condition edits are not the same as comment-only changes." }, { "left": "pull_request:46558", "right": "pull_request:46716", "accept": false, - "reason": "Different changes in the same workflow file: approval/permissions vs secrets handling." + "reason": "Permissions/approval gating vs secrets:{} change; separate hardening steps, not one duplicate fix." }, { "left": "pull_request:46563", "right": "pull_request:46716", "accept": false, - "reason": "Comment-only documentation change vs functional secrets scoping change." + "reason": "Comment-only metadata update vs secrets handling; clearly different changes." }, { "left": "pull_request:46504", "right": "pull_request:46563", "accept": false, - "reason": "Different workflow files and unrelated changes." + "reason": "Build-ci-docker-images commit parsing vs comment-only security-gate update; unrelated." }, { "left": "pull_request:46504", "right": "pull_request:46716", "accept": false, - "reason": "Different workflow files and unrelated fixes; not duplicate PRs." - }, - { - "left": "pull_request:46558", - "right": "pull_request:46563", - "accept": false, - "reason": "Different functional area in the same file; not the same concrete problem." - }, - { - "left": "pull_request:46558", - "right": "pull_request:46716", - "accept": false, - "reason": "Permissions/approval gating vs secrets scoping are distinct changes." + "reason": "Build-image tagging logic vs secrets handling in pr-ci-security-gate; not the same underlying bug." } ] }, @@ -24203,12 +24161,12 @@ "soft_pairs": [ "pull_request:46152|pull_request:46207" ], - "prepared_review_unit_hash": "8924788f6a3cf7debd8f6938790d69f7ffcd4ae233d2678bcc306843be5a83f4", + "prepared_review_unit_hash": "e2d3a7f39b99e164d928f13929aff8412dde369bf898520d7e62a4a82f49ef7a", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4275, + "serialized_chars": 4274, "estimated_input_tokens": 1069, "estimated_eval_tokens": 2394 }, @@ -24216,7 +24174,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4275, + "serialized_chars": 4274, "estimated_input_tokens": 1069, "estimated_eval_tokens": 2394 }, @@ -24225,18 +24183,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs modernize source-file inspection to avoid direct open() calls, but they address different code paths in different modules and are not duplicates.", - "confidence": 0.91, + "summary": "The two PRs are thematically related (removing direct file-open source reads), but they fix different code paths in different modules and should not be treated as duplicates.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #46152 is the broader representative of the shared theme: it introduces a reusable source-path resolution helper in import_utils to replace file-open based self-inspection. PR #46207 is a separate, narrower fix in modeling_utils.", + "canonical_pr_reason": "No single canonical PR for duplication: 46152 changes source access in import_utils, while 46207 changes source inspection in modeling_utils; they are separate fixes.", "best_issue_reason": null, - "best_pr_reason": "PR #46152 is the best cluster representative because it captures the more general refactor pattern and the core change in the shared code-reading approach, while PR #46207 is a specific follow-up in a different subsystem.", + "best_pr_reason": "46207 is the stronger standalone fix because it targets one concrete heuristic in _can_set_attn_implementation and includes tests, whereas 46152 is a different source-access refactor in another module.", "soft_edge_verdicts": [ { "left": "pull_request:46152", "right": "pull_request:46207", "accept": false, - "reason": "Same high-level goal, but different concrete functions and files with different implementation paths; they do not look mergeable into one PR or like the same underlying bug." + "reason": "Same broad theme, but not the same underlying bug/change: 46152 rewrites self-source resolution in import_utils, while 46207 updates source reading for attention-implementation detection in modeling_utils. They touch different code paths and would not plausibly be merged into one PR." } ] }, @@ -24329,12 +24287,12 @@ "soft_pairs": [ "pull_request:46281|pull_request:46890" ], - "prepared_review_unit_hash": "983f6eb82adc9cde687cb700fc7cc81aac57f8d404c3e3451bf20c605bd6dbeb", + "prepared_review_unit_hash": "d0ec78e721627db30d5465428f3f9dbdbc92cb9a26a92ec55846cf8349f48f80", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4271, + "serialized_chars": 4270, "estimated_input_tokens": 1068, "estimated_eval_tokens": 2392 }, @@ -24342,7 +24300,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4271, + "serialized_chars": 4270, "estimated_input_tokens": 1068, "estimated_eval_tokens": 2392 }, @@ -24351,18 +24309,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both items are security-focused PRs that harden path handling against traversal, but they target different untrusted inputs and different code paths: sharded checkpoint weight_map vs. transformers_weights config. Similar theme, not the same underlying fix.", - "confidence": 0.92, + "summary": "Both PRs harden path handling against traversal, but they target different untrusted inputs and different loader code paths: sharded checkpoint `weight_map` vs. `transformers_weights` config field. They are related in theme, not duplicates.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #46281 is the more concrete and self-contained fix: it addresses a specific loader path traversal in sharded-checkpoint resolution from the index weight_map, with a clear containment check and targeted test coverage.", + "canonical_pr_reason": "PR #46281 is the cleaner representative: it fixes a concrete containment bug in the sharded-checkpoint loader and shows an explicit path-validation change in one code path.", "best_issue_reason": null, - "best_pr_reason": "PR #46281 is the best representative of the cluster because it has the clearest, most explicit bug surface and a direct fix for a specific code path; #46890 is related security hardening but for a distinct input path.", + "best_pr_reason": "PR #46281 is the best single representative because it is a focused, self-contained fix with clear exploitability and minimal overlap with the other PR.", "soft_edge_verdicts": [ { "left": "pull_request:46281", "right": "pull_request:46890", "accept": false, - "reason": "Both are path-traversal mitigations, but they fix different sources of untrusted filenames in different loaders. They do not appear to be the same bug or a single change that could plausibly be merged into one PR." + "reason": "Same general security class (path traversal), but they modify different loaders and trust different fields. They are not the same underlying bug and would not be merged into one PR." } ] }, @@ -24392,12 +24350,12 @@ "soft_pairs": [ "pull_request:46310|pull_request:46546" ], - "prepared_review_unit_hash": "16bbd7e8b1dac8d6148a7dec9a38a54f938fda6e5d43dc199242550340da243e", + "prepared_review_unit_hash": "3de6542000db196c9d4cee095a36209f916e232e6265c9790af2fe84bbdbb99f", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4851, + "serialized_chars": 4850, "estimated_input_tokens": 1213, "estimated_eval_tokens": 2682 }, @@ -24405,7 +24363,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4851, + "serialized_chars": 4850, "estimated_input_tokens": 1213, "estimated_eval_tokens": 2682 }, @@ -24414,18 +24372,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs fix the same off-by-one in the right side of the sliding-window mask (`-dist <= right_window_size` instead of `<`); PR 46546 is the more complete change because it also updates Gemma4 and adds tests.", - "confidence": 0.93, + "summary": "Both PRs fix the same Moonshine streaming sliding-window boundary bug: the right window should include the edge index (`<=` instead of `<`). PR 46546 is the stronger representative because it includes the Moonshine fix plus a test, and also applies the same boundary correction in Gemma4.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46546 is the best canonical PR because it contains the full fix across all affected implementations and includes test coverage, while PR 46310 is a narrower partial fix.", + "canonical_pr_reason": "PR 46546 is the best canonical PR because it contains the same Moonshine boundary fix, adds test coverage, and is the more complete change set overall.", "best_issue_reason": null, - "best_pr_reason": "PR 46546 is the strongest representative of the change: broader coverage, same core code fix, and explicit tests make it the more complete/merge-ready patch.", + "best_pr_reason": "PR 46546 is the best overall PR for this cluster since it is the more complete, validated fix for the same underlying boundary bug.", "soft_edge_verdicts": [ { "left": "pull_request:46310", "right": "pull_request:46546", "accept": true, - "reason": "Same underlying boundary bug in the sliding-window right mask, with identical fix in the shared Moonshine streaming code path; 46546 is a superset rather than a different change." + "reason": "Same concrete bug: the Moonshine streaming right-window mask excludes the boundary element in both PRs, and both change the same code path to `<=`." } ] }, @@ -24464,12 +24422,12 @@ "pull_request:46425|pull_request:46850", "pull_request:46850|pull_request:46917" ], - "prepared_review_unit_hash": "3c005043df5839de116ea965e0c35267380c4722ace45042be3c834038934941", + "prepared_review_unit_hash": "e9c74e8fe14abdd1a0c1a4bf955fae223b99a938a9e957c9f41bb90baf24d381", "component_budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 6, - "serialized_chars": 16838, + "serialized_chars": 16837, "estimated_input_tokens": 4210, "estimated_eval_tokens": 8676 }, @@ -24477,7 +24435,7 @@ "node_count": 6, "item_count": 6, "soft_pair_count": 6, - "serialized_chars": 16838, + "serialized_chars": 16837, "estimated_input_tokens": 4210, "estimated_eval_tokens": 8676 }, @@ -24486,48 +24444,48 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are all distinct: three separate model additions, one Gemma sliding-window masking bugfix, one continuous-batching multimodal feature, and one deprecation cleanup. The soft links are broad-topic similarity only, not duplicate code changes.", - "confidence": 0.97, + "summary": "These PRs are only superficially similar; they span unrelated areas like model additions, continuous batching, masking logic, and deprecation cleanup. No pair looks like the same underlying change.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46425 is the best anchor for the cluster because it is the most central and broadly model-integration oriented PR, but it still addresses a different model family than the others.", + "canonical_pr_reason": "No true canonical duplicate PR: the set is heterogeneous, and even the closest items are different model additions or unrelated fixes.", "best_issue_reason": null, - "best_pr_reason": "PR 46425 is the best overall representative of the cluster\u2019s loose similarity pattern, since it is the hub with the most soft connections and the broadest feature-addition scope.", + "best_pr_reason": "PR 46425 is the broadest representative of the superficial model-support subgroup, but it is still a distinct RADIO model addition rather than a duplicate of the others.", "soft_edge_verdicts": [ { "left": "pull_request:46376", "right": "pull_request:46417", "accept": false, - "reason": "Different features for different model families: continuous batching multimodal support vs. HunYuan VL model support." + "reason": "Continuous batching multimodal support vs. adding the HunYuan VL model; different code paths and goals." }, { "left": "pull_request:46376", "right": "pull_request:46425", "accept": false, - "reason": "Unrelated PRs: continuous batching multimodal support is a runtime feature change; RADIO is a model implementation/docs addition." + "reason": "Continuous batching infrastructure vs. RADIO model support; not the same bug or change." }, { "left": "pull_request:46425", "right": "pull_request:46509", "accept": false, - "reason": "Both add model support, but for different architectures and code paths (RADIO vs. Nemotron Omni), so they are not the same change." + "reason": "RADIO model support vs. Nemotron Omni support; both are model additions but for different models and implementations." }, { "left": "pull_request:46425", "right": "pull_request:46917", "accept": false, - "reason": "RADIO model support and deprecated-args cleanup are unrelated changes touching different subsystems." + "reason": "New RADIO model support vs. removing deprecated training args; unrelated changes." }, { "left": "pull_request:46425", "right": "pull_request:46850", "accept": false, - "reason": "Model addition vs. Gemma masking bugfix; they do not fix the same concrete problem or change the same code path." + "reason": "Model registration/support work vs. a Gemma attention-masking bug fix; different underlying problems." }, { "left": "pull_request:46850", "right": "pull_request:46917", "accept": false, - "reason": "Gemma attention-mask boundary fix and deprecated training/image-processor args cleanup are unrelated." + "reason": "Gemma masking fix vs. deprecation cleanup; no shared concrete code-path problem." } ] }, @@ -24557,7 +24515,7 @@ "soft_pairs": [ "pull_request:46574|pull_request:46580" ], - "prepared_review_unit_hash": "916d065f7e80368b21f4ac9eb2733ff15fc216c2250c5735967cb41a42c7268c", + "prepared_review_unit_hash": "8aff577f9e37e22cb2e6d50ed21159af6bba346f2a581234c86e4bee9cfe361c", "component_budget": { "node_count": 2, "item_count": 2, @@ -24579,18 +24537,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two unrelated open PRs: one fixes CodeLlama decode stripping of real leading spaces, the other fixes DeepSeek-Coder/DeepSeek-LLM auto-tokenizer class selection. They should not be treated as duplicates.", - "confidence": 0.99, + "summary": "These two PRs are unrelated tokenizer fixes: one corrects CodeLlama decode behavior to preserve leading spaces, while the other fixes DeepSeek-Coder/DeepSeek-LLM checkpoints loading under the wrong tokenizer class. They do not look like the same underlying bug or change.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the PRs fix different bugs in different tokenizer pathways (CodeLlama decode behavior vs. auto-tokenizer checkpoint routing).", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "46580 is the broader tokenizer-routing fix, but it is still unrelated to 46574 and should not be merged with it as a duplicate.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:46574", "right": "pull_request:46580", "accept": false, - "reason": "Reject: 46574 changes CodeLlama decoder setup to preserve leading spaces on decode, while 46580 changes auto-tokenizer checkpoint-prefix logic for DeepSeek models. Different model families, files, and code paths." + "reason": "Different models and different code paths: 46574 changes CodeLlama decoder setup to stop stripping real leading spaces, while 46580 changes AutoTokenizer checkpoint-prefix routing for DeepSeek checkpoints that load the wrong tokenizer class. These are not mergeable as one fix." } ] }, @@ -24620,7 +24578,7 @@ "soft_pairs": [ "pull_request:46603|pull_request:46604" ], - "prepared_review_unit_hash": "566a339282b9105a15a9cb0ba86dc86ac6750ee494b66b7c69a371fff3c21513", + "prepared_review_unit_hash": "c79b4e399c793cc34f9ee76a73881d9ff19bc9ebb91cc25de0f592caab08ca91", "component_budget": { "node_count": 2, "item_count": 2, @@ -24642,18 +24600,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open Mistral PRs share files but address different changes: one refactors/moves the Mistral converter into an integrations package, while the other adds tekken tokenizer detection/conversion/save utilities. They are related maintenance around the same subsystem, but not the same concrete change.", - "confidence": 0.97, + "summary": "Both PRs are Mistral-tokenizer related and share file overlap, but they target different changes: one is a refactor/extraction of MistralConverter into a package, the other adds tekken tokenizer detection/conversion/save support. They are not the same underlying fix.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46604 is the broader functional change focused on tokenizer detection/conversion/save support; PR 46603 is mainly a code organization refactor.", + "canonical_pr_reason": "PR #46604 is the more representative canonical PR for this cluster because it delivers the broader user-facing Mistral tokenizer functionality, while #46603 is mainly a packaging/refactor change.", "best_issue_reason": null, - "best_pr_reason": "PR 46604 is the best representative of the cluster because it introduces the main user-facing tokenizer capability change, whereas 46603 is an internal move/refactor.", + "best_pr_reason": "PR #46604 is the best standalone PR here: it has a concrete functional goal, touches the relevant integration and tokenizer code paths, and includes corresponding tests.", "soft_edge_verdicts": [ { "left": "pull_request:46603", "right": "pull_request:46604", "accept": false, - "reason": "Different underlying changes: 46603 relocates MistralConverter into the integrations package, while 46604 implements tekken tokenizer detection/conversion/save utilities. Shared files/subsystem are not enough to merge them as duplicates." + "reason": "Reject: #46603 moves/extracts MistralConverter into the integrations package, while #46604 adds tekken tokenizer detection, conversion, and save utilities. Shared filenames reflect the same subsystem, not the same underlying change." } ] }, @@ -24685,12 +24643,12 @@ "pull_request:46741|pull_request:46773", "pull_request:46741|pull_request:46781" ], - "prepared_review_unit_hash": "f601a5a2ccc1087797963cafe9e386982baa67b5ba775adcb1c9c7da68eb7a55", + "prepared_review_unit_hash": "789465efed8a93bdfa78315690a9d6a1962a07730936f872abf9fe33a3f1c185", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 8456, + "serialized_chars": 8455, "estimated_input_tokens": 2114, "estimated_eval_tokens": 4484 }, @@ -24698,7 +24656,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 8456, + "serialized_chars": 8455, "estimated_input_tokens": 2114, "estimated_eval_tokens": 4484 }, @@ -24707,24 +24665,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three distinct PRs around linear-attention/cache behavior, but they fix different bugs: one for chunked-prefill/speculative decoding state reuse, one for padding-mask application at batch size 1, and one for QuantizedCache handling on hybrid models. None are true duplicates of each other.", + "summary": "All three pull requests target different bugs in adjacent model/cache codepaths, but they do not appear to be duplicates of one another. The two soft-similarity edges should be rejected.", "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #46741 is the best canonical PR because it is the broadest concrete fix in the shared model-cache area and the title/diff clearly center on a single bug family: incorrect Mamba2 chunked-prefill/speculative-decoding behavior across several affected models.", + "canonical_pr_reason": "PR 46741 is the broadest and most representative fix in the set: it addresses a concrete chunked-prefill/speculative-decoding cache-path bug across multiple Mamba-like models and has the richest model/test coverage.", "best_issue_reason": null, - "best_pr_reason": "PR #46741 is the strongest representative of the cluster since it addresses a specific forward/cache-path bug with clear model-level impact and the most substantial shared code changes.", + "best_pr_reason": "PR 46741 is the strongest standalone candidate overall because it fixes a specific runtime path with multi-model coverage and substantive implementation changes, whereas the others are narrower and unrelated bugs.", "soft_edge_verdicts": [ { "left": "pull_request:46741", "right": "pull_request:46773", "accept": false, - "reason": "Both touch similar linear-attention model files, but the bugs are different: #46741 changes cache/state reuse for chunked-prefill/speculative decoding, while #46773 fixes padding-mask application being skipped for batch size 1. Different code paths and failure modes." + "reason": "Different bugs: 46741 fixes chunked-prefill/speculative decoding cache handling, while 46773 fixes padding-mask application being skipped for batch-size-1 linear-attention models." }, { "left": "pull_request:46741", "right": "pull_request:46781", "accept": false, - "reason": "#46741 is a model forward/cache bug fix across several architectures, whereas #46781 fixes QuantizedCache construction for hybrid SSM+attention models in cache_utils. They are related to caches broadly but not the same underlying change." + "reason": "Different codepaths and failure modes: 46741 is about precomputed state handling in model forwards, while 46781 changes QuantizedCache construction for hybrid SSM+attention layer types." } ] }, @@ -24757,12 +24715,12 @@ "pull_request:46895|pull_request:46916", "pull_request:46898|pull_request:46916" ], - "prepared_review_unit_hash": "1b4af156e3127d63c0efe87c8cbbe60194522d25ba57d9e4c093528e2c16cb0d", + "prepared_review_unit_hash": "72136622b9be672a109a018ba8cf739b2cb3a2732ca0b7470e1e2f7209f9f8fe", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 7713, + "serialized_chars": 7715, "estimated_input_tokens": 1929, "estimated_eval_tokens": 4114 }, @@ -24770,7 +24728,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 7713, + "serialized_chars": 7715, "estimated_input_tokens": 1929, "estimated_eval_tokens": 4114 }, @@ -24779,30 +24737,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two of the PRs are model-specific variants of the same training-loss double-shift fix, while the third is a bundled PR that contains both fixes. The only clear non-duplicate pair is CohereASR vs Florence2.", - "confidence": 0.76, + "summary": "These PRs split into two model-specific training-loss double-shift fixes: CohereASR and Florence2. PR 46916 is the broadest representative because it bundles both fixes, while 46895 and 46898 are single-model variants. The cross-model pair 46895\u219446898 should not be treated as duplicates.", + "confidence": 0.82, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #46916 is the broadest representative: it contains the same concrete loss/shifting fix for CohereASR and Florence2, so it best captures the cluster\u2019s combined change.", + "canonical_pr_reason": "PR 46916 is the broadest, most representative change set: it covers both CohereASR and Florence2 double-shift fixes and shares the explicit Florence2 issue target.", "best_issue_reason": null, - "best_pr_reason": "PR #46916 is the best single representative because it subsumes the narrower CohereASR-only and Florence2-only fixes and points to both affected code paths.", + "best_pr_reason": "PR 46916 best represents the cluster because it unifies the recurring bug pattern across both affected models, whereas 46895 and 46898 are narrower single-model fixes.", "soft_edge_verdicts": [ { "left": "pull_request:46895", "right": "pull_request:46898", "accept": false, - "reason": "Different models and different issue targets; they share the symptom pattern but not the same concrete code path or change." + "reason": "Same high-level bug pattern, but they fix different model implementations and different issue targets. These are separate concrete code paths, not a duplicate PR pair." }, { "left": "pull_request:46895", "right": "pull_request:46916", "accept": true, - "reason": "46916 includes the same CohereASR training-loss double-shift fix in the same files; it is a superset PR, but this edge still matches the same underlying CohereASR bug/change." + "reason": "Both touch CohereASR's same training-loss double-shift path; 46916 just bundles that fix with an additional Florence2 change, so they plausibly overlap as the same underlying CohereASR bug." }, { "left": "pull_request:46898", "right": "pull_request:46916", "accept": true, - "reason": "46916 includes the same Florence2 training-loss double-shift fix in the same files; despite bundling CohereASR too, this is the same underlying Florence2 bug/change." + "reason": "Both address Florence2's same double-shift loss path, and 46916 simply combines that fix with the CohereASR variant, so the Florence2 portion is the same underlying bug." } ] }, @@ -24844,94 +24802,94 @@ "pull_request:46926|pull_request:46928", "pull_request:46927|pull_request:46928" ], - "prepared_review_unit_hash": "c668f32b49e267942925b11311db62ebd69e17449d695d494e85c96d8fa14584", + "prepared_review_unit_hash": "db2153021e1ee3a019fbf3eac0c3b269474a89d485b474f74a314deeadee128d", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 10, - "serialized_chars": 11925, - "estimated_input_tokens": 2982, - "estimated_eval_tokens": 6220 + "serialized_chars": 11922, + "estimated_input_tokens": 2981, + "estimated_eval_tokens": 6218 }, "budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 10, - "serialized_chars": 11925, - "estimated_input_tokens": 2982, - "estimated_eval_tokens": 6220 + "serialized_chars": 11922, + "estimated_input_tokens": 2981, + "estimated_eval_tokens": 6218 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are all model-specific fixes with different root causes: one real code-path bug in EdgeTam, two offload/load-test workarounds for different models, one dtype-cast fix in GLM46V, and one tokenizer registry update for BLIP-2. They are not duplicates of the same underlying change.", - "confidence": 0.95, + "summary": "All five PRs share a generic \u201cfix integration tests\u201d pattern, but they target different models and different failure modes. None of the soft-edge pairs look like the same underlying bug or change.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46907 is the strongest representative because it fixes an actual runtime bug in the model forward path, while the others are unrelated model-specific test or registration fixes.", + "canonical_pr_reason": "PR 46907 is the best representative only in a loose sense: it fixes a concrete model code-path bug in `edgetam`, whereas the others are mostly model-specific test/setup or registry tweaks. It is not a true duplicate of the others.", "best_issue_reason": null, - "best_pr_reason": "PR 46907 is the best single PR to represent the cluster since it contains a concrete code fix rather than a test-only workaround and is clearly distinct from the other PRs.", + "best_pr_reason": "PR 46907 is the strongest cluster representative because it changes actual model forwarding logic to drop an invalid kwarg, while the other PRs are isolated fixes for unrelated models or test harness setup.", "soft_edge_verdicts": [ { "left": "pull_request:46907", "right": "pull_request:46927", "accept": false, - "reason": "Different bugs in different models: EdgeTam drops processor-only kwargs, while GLM46V casts input_ids to long." + "reason": "Different models and different fixes: `edgetam` removes `original_sizes` before backbone call; `glm46v` casts `input_ids` to long. Same broad theme, not the same underlying bug." }, { "left": "pull_request:46907", "right": "pull_request:46926", "accept": false, - "reason": "EdgeTam forward-path fix vs MiniMax-M3-VL test offload_folder workaround; not the same underlying issue." + "reason": "`edgetam` code-path fix vs `minimax_m3_vl` test-only offload folder addition. Unrelated model failures and unrelated remedies." }, { "left": "pull_request:46907", "right": "pull_request:46908", "accept": false, - "reason": "Different models and fixes: EdgeTam code-path cleanup vs DeepSeek-V3.2 test offload_folder workaround." + "reason": "`edgetam` fixes an invalid kwarg in model forward; `deepseek_v32` adds `offload_folder` to test loading. Different subsystems and failure modes." }, { "left": "pull_request:46907", "right": "pull_request:46928", "accept": false, - "reason": "EdgeTam runtime fix vs BLIP-2 tokenizer auto-registration update; unrelated changes." + "reason": "`edgetam` model forward bug vs `blip_2` tokenizer registry update. No shared concrete code-path problem." }, { "left": "pull_request:46908", "right": "pull_request:46926", "accept": false, - "reason": "Both are load-related test changes, but they target different models and different load failures, so they are not the same bug." + "reason": "Both are load_error-related test fixes, but for different models and different offload/test setups. Not mergeable as one PR." }, { "left": "pull_request:46908", "right": "pull_request:46927", "accept": false, - "reason": "DeepSeek test offload workaround and GLM46V input dtype fix are unrelated." + "reason": "`deepseek_v32` test loading workaround vs `glm46v` forward-type fix. Different bug classes and code paths." }, { - "left": "pull_request:46926", - "right": "pull_request:46927", + "left": "pull_request:46908", + "right": "pull_request:46928", "accept": false, - "reason": "MiniMax offload_folder test workaround vs GLM46V forward input cast; different code paths and failures." + "reason": "`deepseek_v32` offload-folder test change is unrelated to the `blip_2` tokenizer mapping addition." }, { "left": "pull_request:46926", - "right": "pull_request:46928", + "right": "pull_request:46927", "accept": false, - "reason": "MiniMax load workaround and BLIP-2 tokenizer registry change are not the same underlying change." + "reason": "`minimax_m3_vl` adds a test offload folder; `glm46v` changes forward input typing. Distinct model-specific issues." }, { - "left": "pull_request:46927", + "left": "pull_request:46926", "right": "pull_request:46928", "accept": false, - "reason": "GLM46V input_ids casting fix vs BLIP-2 tokenizer registration update; unrelated model fixes." + "reason": "`minimax_m3_vl` load workaround vs `blip_2` tokenizer config change. Same generic area of tests/models, but not the same bug." }, { - "left": "pull_request:46908", + "left": "pull_request:46927", "right": "pull_request:46928", "accept": false, - "reason": "DeepSeek load_error workaround vs BLIP-2 tokenizer registry update; different models and different bugs." + "reason": "`glm46v` fixes input dtype handling; `blip_2` extends tokenizer auto-mapping. No common underlying defect." } ] }, diff --git a/analysis/current/manifest.json b/analysis/current/manifest.json index 5990750f80d9a49d0b8927a755e1df97a59f12d0..8b48e8d148de13aa7337bb84344f146e74974ed0 100644 --- a/analysis/current/manifest.json +++ b/analysis/current/manifest.json @@ -1,8 +1,8 @@ { - "analysis_id": "hybrid-model-20260627t060016z", + "analysis_id": "hybrid-model-20260627t120014z", "archived_artifacts": { - "hybrid": "snapshots/20260627T060016Z/analysis-runs/hybrid-model-20260627t060016z/analysis-report-hybrid.json", - "hybrid_reviews": "snapshots/20260627T060016Z/analysis-runs/hybrid-model-20260627t060016z/analysis-report-hybrid.llm-reviews.json" + "hybrid": "snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/analysis-report-hybrid.llm-reviews.json" }, "artifacts": { "hybrid": "analysis/current/analysis-report-hybrid.json", @@ -10,9 +10,9 @@ }, "channel": "canonical", "model": null, - "published_at": "2026-06-27T06:09:30Z", + "published_at": "2026-06-27T12:10:33Z", "repo": "huggingface/transformers", "schema_version": 1, - "snapshot_id": "20260627T060016Z", + "snapshot_id": "20260627T120014Z", "variant": "hybrid" } diff --git a/snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/analysis-report-hybrid.json b/snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/analysis-report-hybrid.json new file mode 100644 index 0000000000000000000000000000000000000000..0ae892a8f9c387844e35181400554bd04512d789 --- /dev/null +++ b/snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/analysis-report-hybrid.json @@ -0,0 +1,2816 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260627T120014Z", + "generated_at": "2026-06-27T12:10:31Z", + "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-46894-7", + "summary": "Cluster of 3 issues and 4 PRs centered on issue #46897.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 46897, + "canonical_pr_number": 46903, + "issue_numbers": [ + 46894, + 46897, + 46901 + ], + "pr_numbers": [ + 46895, + 46898, + 46903, + 46916 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 46895, + "right_pr_number": 46898, + "code_similarity": 0.117, + "size_similarity": 0.467, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.156, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46895, + "right_pr_number": 46903, + "code_similarity": 0.24, + "size_similarity": 0.844, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.476, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46895, + "right_pr_number": 46916, + "code_similarity": 0.335, + "size_similarity": 0.622, + "file_overlap": 0.4, + "area_overlap": 0.194, + "patch_similarity": 0.154, + "shared_filenames": [ + "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "src/transformers/models/cohere_asr/modular_cohere_asr.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "left_ranges": [ + [ + 615, + 627 + ], + [ + 639, + 654 + ] + ], + "right_ranges": [ + [ + 637, + 644 + ] + ] + }, + { + "filename": "src/transformers/models/cohere_asr/modular_cohere_asr.py", + "left_ranges": [ + [ + 478, + 490 + ], + [ + 502, + 517 + ] + ], + "right_ranges": [ + [ + 500, + 507 + ] + ] + } + ] + }, + { + "left_pr_number": 46898, + "right_pr_number": 46903, + "code_similarity": 0.125, + "size_similarity": 0.553, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.099, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46898, + "right_pr_number": 46916, + "code_similarity": 0.42, + "size_similarity": 0.75, + "file_overlap": 0.5, + "area_overlap": 0.194, + "patch_similarity": 0.35, + "shared_filenames": [ + "src/transformers/models/florence2/modeling_florence2.py", + "src/transformers/models/florence2/modular_florence2.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/florence2/modeling_florence2.py", + "left_ranges": [ + [ + 872, + 878 + ], + [ + 881, + 887 + ], + [ + 904, + 918 + ] + ], + "right_ranges": [ + [ + 902, + 909 + ] + ] + }, + { + "filename": "src/transformers/models/florence2/modular_florence2.py", + "left_ranges": [ + [ + 1580, + 1586 + ], + [ + 1589, + 1595 + ], + [ + 1612, + 1626 + ] + ], + "right_ranges": [ + [ + 1610, + 1617 + ] + ] + } + ] + }, + { + "left_pr_number": 46903, + "right_pr_number": 46916, + "code_similarity": 0.17, + "size_similarity": 0.737, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.154, + "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-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-46869-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #46869.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 46869, + "canonical_pr_number": 46886, + "issue_numbers": [ + 46869 + ], + "pr_numbers": [ + 46886, + 46922 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 46886, + "right_pr_number": 46922, + "code_similarity": 0.711, + "size_similarity": 0.857, + "file_overlap": 1.0, + "area_overlap": 0.5, + "patch_similarity": 0.429, + "shared_filenames": [ + "src/transformers/models/yolos/modeling_yolos.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/yolos/modeling_yolos.py", + "left_ranges": [ + [ + 446, + 452 + ] + ], + "right_ranges": [ + [ + 434, + 440 + ], + [ + 446, + 452 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-46159-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #46159.", + "status": "open", + "confidence": 0.55, + "canonical_issue_number": 46159, + "canonical_pr_number": 46174, + "issue_numbers": [ + 46159 + ], + "pr_numbers": [ + 46174, + 46780 + ], + "evidence_types": [ + "closing_reference" + ], + "pr_comparisons": [ + { + "left_pr_number": 46174, + "right_pr_number": 46780, + "code_similarity": 0.841, + "size_similarity": 0.87, + "file_overlap": 1.0, + "area_overlap": 0.767, + "patch_similarity": 0.657, + "shared_filenames": [ + "src/transformers/models/aimv2/modeling_aimv2.py", + "src/transformers/models/d_fine/modeling_d_fine.py", + "src/transformers/models/deimv2/modeling_deimv2.py", + "src/transformers/models/pp_doclayout_v2/modeling_pp_doclayout_v2.py", + "src/transformers/models/pp_doclayout_v3/modeling_pp_doclayout_v3.py", + "src/transformers/models/rt_detr/modeling_rt_detr.py", + "src/transformers/models/rt_detr_v2/modeling_rt_detr_v2.py", + "src/transformers/models/vit_mae/modeling_vit_mae.py", + "src/transformers/models/vit_mae/modular_vit_mae.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/aimv2/modeling_aimv2.py", + "left_ranges": [ + [ + 134, + 141 + ], + [ + 144, + 154 + ], + [ + 157, + 165 + ] + ], + "right_ranges": [ + [ + 134, + 140 + ], + [ + 143, + 153 + ], + [ + 156, + 162 + ] + ] + }, + { + "filename": "src/transformers/models/d_fine/modeling_d_fine.py", + "left_ranges": [ + [ + 631, + 638 + ], + [ + 641, + 651 + ], + [ + 654, + 662 + ] + ], + "right_ranges": [ + [ + 631, + 637 + ], + [ + 640, + 650 + ], + [ + 653, + 659 + ] + ] + }, + { + "filename": "src/transformers/models/deimv2/modeling_deimv2.py", + "left_ranges": [ + [ + 703, + 710 + ], + [ + 713, + 723 + ], + [ + 726, + 734 + ] + ], + "right_ranges": [ + [ + 703, + 709 + ], + [ + 712, + 722 + ], + [ + 725, + 731 + ] + ] + }, + { + "filename": "src/transformers/models/pp_doclayout_v2/modeling_pp_doclayout_v2.py", + "left_ranges": [ + [ + 1551, + 1558 + ], + [ + 1561, + 1571 + ], + [ + 1574, + 1582 + ] + ], + "right_ranges": [ + [ + 1551, + 1557 + ], + [ + 1560, + 1570 + ], + [ + 1573, + 1579 + ] + ] + }, + { + "filename": "src/transformers/models/pp_doclayout_v3/modeling_pp_doclayout_v3.py", + "left_ranges": [ + [ + 804, + 811 + ], + [ + 814, + 824 + ], + [ + 827, + 835 + ] + ], + "right_ranges": [ + [ + 804, + 810 + ], + [ + 813, + 823 + ], + [ + 826, + 832 + ] + ] + }, + { + "filename": "src/transformers/models/rt_detr/modeling_rt_detr.py", + "left_ranges": [ + [ + 847, + 854 + ], + [ + 857, + 867 + ], + [ + 870, + 878 + ] + ], + "right_ranges": [ + [ + 847, + 853 + ], + [ + 856, + 866 + ], + [ + 869, + 875 + ] + ] + }, + { + "filename": "src/transformers/models/rt_detr_v2/modeling_rt_detr_v2.py", + "left_ranges": [ + [ + 976, + 983 + ], + [ + 986, + 996 + ], + [ + 999, + 1007 + ] + ], + "right_ranges": [ + [ + 976, + 982 + ], + [ + 985, + 995 + ], + [ + 998, + 1004 + ] + ] + }, + { + "filename": "src/transformers/models/vit_mae/modeling_vit_mae.py", + "left_ranges": [ + [ + 152, + 159 + ], + [ + 162, + 172 + ], + [ + 175, + 183 + ] + ], + "right_ranges": [ + [ + 152, + 158 + ], + [ + 161, + 171 + ], + [ + 174, + 180 + ] + ] + }, + { + "filename": "src/transformers/models/vit_mae/modular_vit_mae.py", + "left_ranges": [ + [ + 62, + 69 + ], + [ + 72, + 82 + ], + [ + 85, + 93 + ] + ], + "right_ranges": [ + [ + 62, + 68 + ], + [ + 71, + 81 + ], + [ + 84, + 90 + ] + ] + } + ] + } + ] + }, + { + "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": 0.721, + "size_similarity": 0.973, + "file_overlap": 0.75, + "area_overlap": 0.563, + "patch_similarity": 0.696, + "shared_filenames": [ + "src/transformers/testing_utils.py", + "tests/utils/test_testing_utils.py", + "utils/notification_service.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/testing_utils.py", + "left_ranges": [ + [ + 3611, + 3618 + ], + [ + 3835, + 3869 + ] + ], + "right_ranges": [ + [ + 3525, + 3558 + ], + [ + 3782, + 3788 + ] + ] + }, + { + "filename": "tests/utils/test_testing_utils.py", + "left_ranges": [ + [ + 1, + 144 + ] + ], + "right_ranges": [ + [ + 1, + 114 + ] + ] + }, + { + "filename": "utils/notification_service.py", + "left_ranges": [ + [ + 935, + 967 + ] + ], + "right_ranges": [ + [ + 935, + 967 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-46305-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #46305.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 46305, + "canonical_pr_number": 46546, + "issue_numbers": [ + 46305 + ], + "pr_numbers": [ + 46310, + 46546 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 46310, + "right_pr_number": 46546, + "code_similarity": 0.566, + "size_similarity": 0.444, + "file_overlap": 0.5, + "area_overlap": 0.867, + "patch_similarity": 0.16, + "shared_filenames": [ + "src/transformers/models/moonshine_streaming/modeling_moonshine_streaming.py", + "src/transformers/models/moonshine_streaming/modular_moonshine_streaming.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/moonshine_streaming/modeling_moonshine_streaming.py", + "left_ranges": [ + [ + 356, + 362 + ] + ], + "right_ranges": [ + [ + 355, + 361 + ] + ] + }, + { + "filename": "src/transformers/models/moonshine_streaming/modular_moonshine_streaming.py", + "left_ranges": [ + [ + 269, + 275 + ] + ], + "right_ranges": [ + [ + 269, + 275 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-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-45415-2", + "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "status": "open", + "confidence": 0.5, + "canonical_issue_number": null, + "canonical_pr_number": 45425, + "issue_numbers": [], + "pr_numbers": [ + 45415, + 45425 + ], + "evidence_types": [ + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 45415, + "right_pr_number": 45425, + "code_similarity": 0.092, + "size_similarity": 0.171, + "file_overlap": 0.065, + "area_overlap": 0.053, + "patch_similarity": 0.13, + "shared_filenames": [ + "src/transformers/modeling_flash_attention_utils.py", + "src/transformers/modeling_utils.py", + "utils/check_types.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/modeling_flash_attention_utils.py", + "left_ranges": [ + [ + 12, + 18 + ], + [ + 148, + 156 + ] + ], + "right_ranges": [ + [ + 31, + 37 + ], + [ + 148, + 154 + ] + ] + }, + { + "filename": "src/transformers/modeling_utils.py", + "left_ranges": [ + [ + 29, + 35 + ], + [ + 175, + 181 + ], + [ + 187, + 193 + ], + [ + 295, + 301 + ], + [ + 449, + 458 + ], + [ + 840, + 851 + ], + [ + 1032, + 1038 + ], + [ + 1062, + 1068 + ], + [ + 1566, + 1572 + ], + [ + 1822, + 1828 + ], + [ + 1833, + 1841 + ], + [ + 2393, + 2399 + ], + [ + 2990, + 2996 + ], + [ + 3136, + 3143 + ], + [ + 3262, + 3268 + ], + [ + 3291, + 3299 + ], + [ + 3320, + 3326 + ], + [ + 4010, + 4016 + ], + [ + 4177, + 4183 + ], + [ + 4190, + 4196 + ], + [ + 4505, + 4511 + ], + [ + 4623, + 4629 + ], + [ + 4747, + 4753 + ], + [ + 4765, + 4771 + ] + ], + "right_ranges": [ + [ + 29, + 35 + ], + [ + 121, + 127 + ], + [ + 139, + 147 + ], + [ + 181, + 187 + ], + [ + 191, + 214 + ], + [ + 343, + 360 + ], + [ + 374, + 383 + ], + [ + 428, + 438 + ], + [ + 903, + 923 + ], + [ + 962, + 968 + ], + [ + 1010, + 1016 + ], + [ + 1079, + 1095 + ], + [ + 1112, + 1129 + ], + [ + 1626, + 1633 + ], + [ + 1853, + 1862 + ], + [ + 1873, + 1892 + ], + [ + 1895, + 1903 + ], + [ + 1924, + 1930 + ], + [ + 2448, + 2469 + ], + [ + 3059, + 3065 + ], + [ + 3205, + 3212 + ], + [ + 3328, + 3338 + ], + [ + 4095, + 4106 + ], + [ + 4293, + 4301 + ], + [ + 4635, + 4641 + ], + [ + 4757, + 4763 + ], + [ + 4836, + 4849 + ], + [ + 4889, + 4895 + ], + [ + 4907, + 4913 + ] + ] + }, + { + "filename": "utils/check_types.py", + "left_ranges": [ + [ + 4, + 10 + ], + [ + 15, + 28 + ], + [ + 32, + 54 + ] + ], + "right_ranges": [ + [ + 16, + 22 + ], + [ + 26, + 32 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-44018-2", + "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "status": "open", + "confidence": 0.5, + "canonical_issue_number": null, + "canonical_pr_number": 44068, + "issue_numbers": [], + "pr_numbers": [ + 44018, + 44068 + ], + "evidence_types": [ + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 44018, + "right_pr_number": 44068, + "code_similarity": 0.766, + "size_similarity": 0.939, + "file_overlap": 1.0, + "area_overlap": 0.425, + "patch_similarity": 0.866, + "shared_filenames": [ + "src/transformers/models/gpt_neo/modeling_gpt_neo.py", + "tests/models/gpt_neo/test_modeling_gpt_neo.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/gpt_neo/modeling_gpt_neo.py", + "left_ranges": [ + [ + 26, + 46 + ], + [ + 138, + 143 + ], + [ + 182, + 187 + ], + [ + 283, + 295 + ], + [ + 327, + 341 + ], + [ + 348, + 354 + ], + [ + 360, + 369 + ], + [ + 399, + 406 + ], + [ + 411, + 419 + ], + [ + 428, + 433 + ], + [ + 472, + 492 + ], + [ + 507, + 513 + ], + [ + 519, + 528 + ], + [ + 541, + 559 + ], + [ + 562, + 567 + ], + [ + 595, + 601 + ], + [ + 607, + 614 + ], + [ + 627, + 643 + ], + [ + 685, + 690 + ], + [ + 708, + 714 + ], + [ + 720, + 727 + ], + [ + 740, + 757 + ], + [ + 761, + 766 + ], + [ + 780, + 786 + ], + [ + 791, + 798 + ], + [ + 807, + 822 + ], + [ + 840, + 845 + ] + ], + "right_ranges": [ + [ + 26, + 31 + ], + [ + 34, + 44 + ], + [ + 136, + 141 + ], + [ + 180, + 185 + ], + [ + 281, + 293 + ], + [ + 325, + 339 + ], + [ + 346, + 352 + ], + [ + 358, + 367 + ], + [ + 397, + 404 + ], + [ + 409, + 417 + ], + [ + 426, + 434 + ], + [ + 463, + 483 + ], + [ + 498, + 504 + ], + [ + 510, + 519 + ], + [ + 532, + 550 + ], + [ + 553, + 564 + ], + [ + 586, + 592 + ], + [ + 598, + 605 + ], + [ + 618, + 634 + ], + [ + 676, + 688 + ], + [ + 699, + 705 + ], + [ + 711, + 718 + ], + [ + 731, + 748 + ], + [ + 752, + 762 + ], + [ + 771, + 777 + ], + [ + 782, + 789 + ], + [ + 798, + 813 + ], + [ + 831, + 836 + ] + ] + }, + { + "filename": "tests/models/gpt_neo/test_modeling_gpt_neo.py", + "left_ranges": [ + [ + 458, + 464 + ] + ], + "right_ranges": [ + [ + 458, + 464 + ] + ] + } + ] + } + ] + } + ], + "duplicate_issues": [ + { + "cluster_id": "cluster-46894-7", + "canonical_issue_number": 46897, + "duplicate_issue_numbers": [ + 46894, + 46901 + ], + "reason": "Issues in cluster-46894-7 are treated as duplicates because they share closing_reference, shared_issue_target, soft_similarity evidence." + } + ], + "duplicate_prs": [ + { + "cluster_id": "cluster-41211-3", + "canonical_pr_number": 44339, + "duplicate_pr_numbers": [ + 41356 + ], + "target_issue_number": 41211, + "reason": "PRs in cluster-41211-3 are treated as duplicates because they converge on issue #41211 with closing_reference evidence." + }, + { + "cluster_id": "cluster-43240-3", + "canonical_pr_number": 43251, + "duplicate_pr_numbers": [ + 43254 + ], + "target_issue_number": 43240, + "reason": "PRs in cluster-43240-3 are treated as duplicates because they converge on issue #43240 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-43698-3", + "canonical_pr_number": 43779, + "duplicate_pr_numbers": [ + 43816 + ], + "target_issue_number": 43698, + "reason": "PRs in cluster-43698-3 are treated as duplicates because they converge on issue #43698 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-43979-11", + "canonical_pr_number": 44007, + "duplicate_pr_numbers": [ + 43996, + 44013, + 44044, + 44066, + 44072, + 44085, + 44129, + 44154, + 44722 + ], + "target_issue_number": 43979, + "reason": "PRs in cluster-43979-11 are treated as duplicates because they converge on issue #43979 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-44018-2", + "canonical_pr_number": 44068, + "duplicate_pr_numbers": [ + 44018 + ], + "target_issue_number": null, + "reason": "PRs in cluster-44018-2 are treated as duplicates because they share soft_similarity evidence." + }, + { + "cluster_id": "cluster-45081-3", + "canonical_pr_number": 45317, + "duplicate_pr_numbers": [ + 45086 + ], + "target_issue_number": 45081, + "reason": "PRs in cluster-45081-3 are treated as duplicates because they converge on issue #45081 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-45415-2", + "canonical_pr_number": 45425, + "duplicate_pr_numbers": [ + 45415 + ], + "target_issue_number": null, + "reason": "PRs in cluster-45415-2 are treated as duplicates because they share soft_similarity evidence." + }, + { + "cluster_id": "cluster-45561-3", + "canonical_pr_number": 45645, + "duplicate_pr_numbers": [ + 45639 + ], + "target_issue_number": 45561, + "reason": "PRs in cluster-45561-3 are treated as duplicates because they converge on issue #45561 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-45647-3", + "canonical_pr_number": 45738, + "duplicate_pr_numbers": [ + 45737 + ], + "target_issue_number": 45647, + "reason": "PRs in cluster-45647-3 are treated as duplicates because they converge on issue #45647 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-46159-3", + "canonical_pr_number": 46174, + "duplicate_pr_numbers": [ + 46780 + ], + "target_issue_number": 46159, + "reason": "PRs in cluster-46159-3 are treated as duplicates because they converge on issue #46159 with closing_reference evidence." + }, + { + "cluster_id": "cluster-46305-3", + "canonical_pr_number": 46546, + "duplicate_pr_numbers": [ + 46310 + ], + "target_issue_number": 46305, + "reason": "PRs in cluster-46305-3 are treated as duplicates because they converge on issue #46305 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-46869-3", + "canonical_pr_number": 46886, + "duplicate_pr_numbers": [ + 46922 + ], + "target_issue_number": 46869, + "reason": "PRs in cluster-46869-3 are treated as duplicates because they converge on issue #46869 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-46894-7", + "canonical_pr_number": 46903, + "duplicate_pr_numbers": [ + 46895, + 46898, + 46916 + ], + "target_issue_number": 46897, + "reason": "PRs in cluster-46894-7 are treated as duplicates because they converge on issue #46897 with closing_reference, shared_issue_target, soft_similarity evidence." + } + ], + "best_issue": { + "cluster_id": "cluster-46894-7", + "issue_number": 46897, + "reason": "Issue #46897 is the strongest global issue candidate because it is open, belongs to a cluster with 7 artifacts, and carries 0 discussion comments plus 9 inbound references.", + "score": 179.0 + }, + "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/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/analysis-report-hybrid.llm-reviews.json b/snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/analysis-report-hybrid.llm-reviews.json new file mode 100644 index 0000000000000000000000000000000000000000..0f668dca6c0bf98d8430e74e77ec3a9dadbe66d5 --- /dev/null +++ b/snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/analysis-report-hybrid.llm-reviews.json @@ -0,0 +1,24904 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260627T120014Z", + "generated_at": "2026-06-27T12:10:31Z", + "model": "gpt-5.4-mini?service_tier=flex", + "reviews": [ + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 1, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:24643", + "issue:30277", + "issue:34634", + "issue:34928", + "issue:35545", + "issue:36010", + "issue:36331", + "issue:39290", + "issue:41720", + "issue:41762", + "issue:42915", + "issue:43039", + "issue:43425", + "issue:43541", + "issue:43596", + "issue:43638", + "issue:43716", + "issue:43827", + "issue:43828", + "issue:43975", + "issue:44292", + "issue:44322", + "issue:44457", + "issue:44509", + "issue:44512", + "issue:44560", + "issue:44805", + "issue:44841", + "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:46468", + "issue:46515", + "issue:46661", + "issue:46869" + ], + "soft_pairs": [ + "issue:44322|issue:45464", + "issue:45237|issue:45685", + "issue:43716|issue:45237", + "issue:43828|issue:45237", + "issue:45237|issue:46468", + "issue:45237|issue:45507", + "issue:30277|issue:35545", + "issue:35545|issue:42915", + "issue:35545|issue:41720", + "issue:45901|issue:46015", + "issue:44509|issue:44512", + "issue:35545|issue:36010", + "issue:44918|issue:46165", + "issue:46165|issue:46869", + "issue:44805|issue:45823", + "issue:43638|issue:44805", + "issue:43638|issue:46165", + "issue:43596|issue:44805", + "issue:44805|issue:46294", + "issue:44805|issue:45161", + "issue:46165|issue:46468", + "issue:44560|issue:46165", + "issue:34634|issue:35545", + "issue:30277|issue:43638", + "issue:43596|issue:43638", + "issue:43039|issue:45161", + "issue:43975|issue:44457", + "issue:44292|issue:44457", + "issue:44841|issue:46661", + "issue:43716|issue:46869", + "issue:24643|issue:30277", + "issue:34928|issue:44805", + "issue:45859|issue:45902", + "issue:45685|issue:46869", + "issue:43638|issue:45161", + "issue:34928|issue:36331", + "issue:43596|issue:45161", + "issue:43716|issue:45507", + "issue:44292|issue:45237", + "issue:45874|issue:46294", + "issue:39290|issue:45081", + "issue:45800|issue:46515", + "issue:45823|issue:45874", + "issue:43425|issue:44292", + "issue:34928|issue:41720", + "issue:43828|issue:45161", + "issue:43541|issue:43716", + "issue:43828|issue:44292", + "issue:43716|issue:46468", + "issue:44292|issue:44918", + "issue:45507|issue:45685", + "issue:43716|issue:44292", + "issue:44918|issue:45161", + "issue:43716|issue:45685", + "issue:43039|issue:44918", + "issue:41762|issue:43638", + "issue:43827|issue:44512", + "issue:43828|issue:45685", + "issue:43039|issue:43828", + "issue:30277|issue:36331", + "issue:34928|issue:36010", + "issue:45237|issue:46015", + "issue:43039|issue:44805" + ], + "prepared_review_unit_hash": "2247dcdb0d75e321b0f8af8999c1e7724d5da68b9d46e8102102482278cbdf56", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 63, + "serialized_chars": 46576, + "estimated_input_tokens": 11644, + "estimated_eval_tokens": 23544 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly unrelated bug reports. Only two soft-edge pairs look like true duplicates: the DeepSpeed ZeRO-3 Bert index-out-of-bounds reports and the torch 2.4.1 compatibility reports. The rest share model names or subsystems but not the same concrete failure.", + "confidence": 0.83, + "canonical_issue_reason": "issue:43596 is the best canonical issue: it states the ZeRO-3/Bert index-out-of-bounds failure clearly, and issue:43638 reads as the same bug in a non-pretrained Bert variant.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43596 is the strongest representative overall because it is concrete, well-scoped, and anchors the only strong duplicate pair in the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44322", + "right": "issue:45464", + "accept": false, + "reason": "Both are Qwen3.5-related, but the packet shows different failures: missing config attribute vs streaming chat/completions error." + }, + { + "left": "issue:45237", + "right": "issue:45685", + "accept": false, + "reason": "Different hardware/runtime failures: AMD GPU inference vs MPS histogram int support." + }, + { + "left": "issue:43716", + "right": "issue:45237", + "accept": false, + "reason": "Different bugs and paths: multimodal dtype mismatch vs AMD GPU execution failure." + }, + { + "left": "issue:43828", + "right": "issue:45237", + "accept": false, + "reason": "Autocast dtype mismatch is not the same bug as the AMD GPU failure." + }, + { + "left": "issue:45237", + "right": "issue:46468", + "accept": false, + "reason": "Different concrete failures: AMD GPU issue vs broken official example." + }, + { + "left": "issue:45237", + "right": "issue:45507", + "accept": false, + "reason": "No shared code-path problem; one is AMD GPU inference, the other an invalid-method error in a different model." + }, + { + "left": "issue:30277", + "right": "issue:35545", + "accept": false, + "reason": "Both are runtime issues, but DeepSpeed collective mismatch and ONNX export failure are different bugs." + }, + { + "left": "issue:35545", + "right": "issue:42915", + "accept": false, + "reason": "Different models and different failure modes; export error vs FineGrainedFP8 issue." + }, + { + "left": "issue:35545", + "right": "issue:41720", + "accept": false, + "reason": "ONNX export error and Qwen3 auto-device-map cuda assert are unrelated." + }, + { + "left": "issue:45901", + "right": "issue:46015", + "accept": false, + "reason": "Both are documentation/example problems, but they are about different docs and different breaks." + }, + { + "left": "issue:44509", + "right": "issue:44512", + "accept": false, + "reason": "Both are v5 doc cleanups, but they reference different removed features." + }, + { + "left": "issue:35545", + "right": "issue:36010", + "accept": false, + "reason": "Different regressions: ONNX export failure vs GenerationMixin import error." + }, + { + "left": "issue:44918", + "right": "issue:46165", + "accept": false, + "reason": "Different bugs: Qwen3.5 embedding unpacking vs resume-training with remote code." + }, + { + "left": "issue:46165", + "right": "issue:46869", + "accept": false, + "reason": "Different setups and symptoms: resume/trust-remote-code vs multi-GPU device-map failure." + }, + { + "left": "issue:44805", + "right": "issue:45823", + "accept": false, + "reason": "Mask/index shape mismatch is not the same as Gemma4 device mismatch in forward." + }, + { + "left": "issue:43638", + "right": "issue:44805", + "accept": false, + "reason": "Different failure mechanisms: ZeRO-3 Bert loading error vs mask/tensor shape mismatch." + }, + { + "left": "issue:43638", + "right": "issue:46165", + "accept": false, + "reason": "No overlap in code path or failure mode." + }, + { + "left": "issue:43596", + "right": "issue:44805", + "accept": false, + "reason": "Both are index-related, but the concrete bugs are different." + }, + { + "left": "issue:44805", + "right": "issue:46294", + "accept": false, + "reason": "Mask/index shape mismatch is unrelated to Qwen3.5 video token/template mismatch." + }, + { + "left": "issue:44805", + "right": "issue:45161", + "accept": false, + "reason": "Mask/index shape mismatch is unrelated to GPT-OSS tensor-parallel behavior." + }, + { + "left": "issue:46165", + "right": "issue:46468", + "accept": false, + "reason": "Different issues: remote-code resume training vs broken official example." + }, + { + "left": "issue:44560", + "right": "issue:46165", + "accept": false, + "reason": "Video chat-template StopIteration is unrelated to resume-training remote-code behavior." + }, + { + "left": "issue:34634", + "right": "issue:35545", + "accept": false, + "reason": "BarkProcessor voice_preset and ModernBERT ONNX export are unrelated." + }, + { + "left": "issue:30277", + "right": "issue:43638", + "accept": false, + "reason": "Both involve DeepSpeed, but the concrete failures differ: collective mismatch vs Bert ZeRO-3 index error." + }, + { + "left": "issue:43596", + "right": "issue:43638", + "accept": true, + "reason": "Same DeepSpeed ZeRO-3/Bert index-out-of-bounds bug; 43638 is a more specific non-pretrained Bert variant." + }, + { + "left": "issue:43039", + "right": "issue:45161", + "accept": false, + "reason": "Liger kernel cross-entropy invocation and GPT-OSS TP failure are different problems." + }, + { + "left": "issue:43975", + "right": "issue:44457", + "accept": false, + "reason": "Incorrect detokenization and LoRA merge/save/load inconsistency are different bugs." + }, + { + "left": "issue:44292", + "right": "issue:44457", + "accept": false, + "reason": "Different model/runtime failures with no shared concrete code path." + }, + { + "left": "issue:44841", + "right": "issue:46661", + "accept": false, + "reason": "Processor failure and remote-code config_class attribute error are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:46869", + "accept": false, + "reason": "Different multimodal/device-map failures; not the same underlying bug." + }, + { + "left": "issue:24643", + "right": "issue:30277", + "accept": false, + "reason": "Both mention DeepSpeed, but one is a weight-dimension training error and the other is a collective mismatch." + }, + { + "left": "issue:34928", + "right": "issue:44805", + "accept": false, + "reason": "Activation checkpointing/FSDP mismatch and mask-shape indexing are different code paths." + }, + { + "left": "issue:45859", + "right": "issue:45902", + "accept": false, + "reason": "Same class, but different concrete bugs: CPU offload incompatibility vs uninitialized memory after meta materialization." + }, + { + "left": "issue:45685", + "right": "issue:46869", + "accept": false, + "reason": "MPS histogram int support and YOLO multi-GPU auto device-map are unrelated." + }, + { + "left": "issue:43638", + "right": "issue:45161", + "accept": false, + "reason": "Bert ZeRO-3 loading failure is not the same as GPT-OSS tensor-parallel behavior." + }, + { + "left": "issue:34928", + "right": "issue:36331", + "accept": false, + "reason": "FSDP/checkpointing mismatch and CustomTrainer compute_loss signature breakage are different failures." + }, + { + "left": "issue:43596", + "right": "issue:45161", + "accept": false, + "reason": "DeepSpeed/Bert loading error and GPT-OSS tensor-parallel issue are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:45507", + "accept": false, + "reason": "Device/dtype mismatch in Mistral-3 is not the same as GraniteMoEHybrid invalid method calls." + }, + { + "left": "issue:44292", + "right": "issue:45237", + "accept": false, + "reason": "Different model/runtime failures; no shared underlying bug." + }, + { + "left": "issue:45874", + "right": "issue:46294", + "accept": false, + "reason": "Different Gemma4/Qwen3.5 bugs: reverse embedding lookup vs video token/template mismatch." + }, + { + "left": "issue:39290", + "right": "issue:45081", + "accept": false, + "reason": "Different models and failures: Gemma3 config attribute vs Mistral regex patch crash." + }, + { + "left": "issue:45800", + "right": "issue:46515", + "accept": true, + "reason": "Both report the same torch 2.4.1 compatibility regression; 46515 is a shorter duplicate report." + }, + { + "left": "issue:45823", + "right": "issue:45874", + "accept": false, + "reason": "Different Gemma4 issues: forward device mismatch vs expensive reverse embedding lookup." + }, + { + "left": "issue:43425", + "right": "issue:44292", + "accept": false, + "reason": "Torch version incompatibility is broader than the Qwen NVFP4 runtime error; not the same bug." + }, + { + "left": "issue:43827", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs after v5 removals, but they cover different removed features." + }, + { + "left": "issue:43828", + "right": "issue:45685", + "accept": false, + "reason": "Autocast dtype mismatch and MPS histogram int support are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:43828", + "accept": false, + "reason": "Liger kernel cross-entropy call and autocast dtype mismatch do not share a concrete code path." + }, + { + "left": "issue:30277", + "right": "issue:36331", + "accept": false, + "reason": "DeepSpeed collective mismatch and trainer compute_loss signature breakage are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36010", + "accept": false, + "reason": "FSDP/checkpointing mismatch and GenerationMixin import failure are different issues." + }, + { + "left": "issue:45237", + "right": "issue:46015", + "accept": false, + "reason": "AMD GPU inference failure and outdated RAG examples are not the same underlying change." + }, + { + "left": "issue:43039", + "right": "issue:44805", + "accept": false, + "reason": "Liger kernel cross-entropy call and mask-shape indexing error do not share a concrete code path." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "Mostly grounded, but at least one soft-edge reason overstates the evidence: issue:44322 vs issue:45464 are both Qwen3.5-related, so saying they are from \u201cdifferent model families\u201d is not supported by the packet. The duplicate conclusions on issue:43596/issue:43638 and issue:45800/issue:46515 look plausible, but the verdict set is not fully conservative." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 2, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:36010", + "issue:39290", + "issue:41093", + "issue:41720", + "issue:41762", + "issue:43039", + "issue:43296", + "issue:43404", + "issue:43425", + "issue:43531", + "issue:43541", + "issue:43572", + "issue:43596", + "issue:43638", + "issue:43792", + "issue:43828", + "issue:43854", + "issue:43866", + "issue:43872", + "issue:43901", + "issue:43975", + "issue:44292", + "issue:44512", + "issue:44560", + "issue:44589", + "issue:44841", + "issue:44863", + "issue:45020", + "issue:45070", + "issue:45084", + "issue:45237", + "issue:45362", + "issue:45399", + "issue:45464", + "issue:45507", + "issue:45685", + "issue:45800", + "issue:45812", + "issue:45823", + "issue:45901", + "issue:45920", + "issue:46015", + "issue:46294", + "issue:46304", + "issue:46402", + "issue:46661", + "issue:46772", + "issue:46869" + ], + "soft_pairs": [ + "issue:43866|issue:46402", + "issue:45800|issue:46304", + "issue:43296|issue:46304", + "issue:41762|issue:43596", + "issue:39290|issue:41720", + "issue:44589|issue:46304", + "issue:43901|issue:44512", + "issue:45237|issue:45901", + "issue:43975|issue:46015", + "issue:43039|issue:44560", + "issue:45823|issue:46294", + "issue:44841|issue:45823", + "issue:43866|issue:46772", + "issue:44863|issue:46015", + "issue:45020|issue:46304", + "issue:45812|issue:45920", + "issue:43866|issue:44863", + "issue:43872|issue:46304", + "issue:44292|issue:46015", + "issue:39290|issue:43531", + "issue:41093|issue:43638", + "issue:43792|issue:45901", + "issue:43854|issue:43866", + "issue:43425|issue:45070", + "issue:43975|issue:45901", + "issue:43792|issue:45070", + "issue:43596|issue:45800", + "issue:45362|issue:45464", + "issue:43541|issue:46869", + "issue:44292|issue:45901", + "issue:39290|issue:45823", + "issue:45464|issue:45685", + "issue:36010|issue:39290", + "issue:43541|issue:45685", + "issue:43541|issue:43828", + "issue:44841|issue:45084", + "issue:43404|issue:43866", + "issue:45507|issue:46869", + "issue:41093|issue:41762", + "issue:45399|issue:46661", + "issue:43572|issue:45070", + "issue:45800|issue:45823" + ], + "prepared_review_unit_hash": "1424eeb5af8adcc475c972493154a68449bf031bd53696fd06029bff5c3c4b67", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 42, + "serialized_chars": 40042, + "estimated_input_tokens": 10011, + "estimated_eval_tokens": 20278 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most pairs are distinct model-specific or subsystem-specific issues. Two pairs look like true duplicates: the v5 AutoTokenizer wrong-token-id regression across model families, and the ZeRO-3 load-time index error during model initialization.", + "confidence": 0.68, + "canonical_issue_reason": "issue:45920 is the best canonical issue because it describes the broad v5 AutoTokenizer regression affecting multiple model families, which subsumes the narrower Granite-only report.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45920 is the strongest global issue candidate: it is clearly framed, high-impact, and broad enough to represent the shared tokenizer regression pattern.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43866", + "right": "issue:46402", + "accept": false, + "reason": "Different models and different failures: corrupted checkpoint vs missing text-encoder weights." + }, + { + "left": "issue:45800", + "right": "issue:46304", + "accept": false, + "reason": "Unrelated compatibility regressions: torch/version mismatch vs USE_DISTRIBUTED=0 breakage." + }, + { + "left": "issue:43296", + "right": "issue:46304", + "accept": false, + "reason": "Different loading path and symptom: vLLM PaddleOCR-VL load failure vs distributed build breakage." + }, + { + "left": "issue:41762", + "right": "issue:43596", + "accept": true, + "reason": "Both report the same ZeRO-3 model-init/load IndexError ('index 0 out of bounds' / empty tensor) and could plausibly share one fix." + }, + { + "left": "issue:39290", + "right": "issue:41720", + "accept": false, + "reason": "Different bugs: Gemma3 config attribute missing with vLLM vs Qwen3 auto-device-map CUDA assert." + }, + { + "left": "issue:44589", + "right": "issue:46304", + "accept": false, + "reason": "Different low-level runtime issues: missing Float8 storage vs USE_DISTRIBUTED=0 support." + }, + { + "left": "issue:43901", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs issues, but they refer to different outdated commands and different doc sections." + }, + { + "left": "issue:45237", + "right": "issue:45901", + "accept": false, + "reason": "Different task/model failures: AMD GPU support vs table-question-answering crash." + }, + { + "left": "issue:43975", + "right": "issue:46015", + "accept": false, + "reason": "Different problems: detokenization regression vs outdated RAG examples." + }, + { + "left": "issue:43039", + "right": "issue:44560", + "accept": false, + "reason": "Different code paths: Liger kernel cross-entropy dispatch vs Qwen3-vl video StopIteration." + }, + { + "left": "issue:45823", + "right": "issue:46294", + "accept": false, + "reason": "Different failure modes: device-map PLE mismatch vs chat-template video token/feature mismatch." + }, + { + "left": "issue:44841", + "right": "issue:45823", + "accept": false, + "reason": "Different models and subsystems: processor loading vs device-map forward mismatch." + }, + { + "left": "issue:43866", + "right": "issue:46772", + "accept": false, + "reason": "Checkpoint corruption is not the same as NVFP4 compatibility for DiffusionGemma." + }, + { + "left": "issue:44863", + "right": "issue:46015", + "accept": false, + "reason": "NemotronH checkpoint loading is unrelated to RAG example docs drift." + }, + { + "left": "issue:45020", + "right": "issue:46304", + "accept": false, + "reason": "Remote-code model breakage is a different issue from the USE_DISTRIBUTED=0 torch support regression." + }, + { + "left": "issue:45812", + "right": "issue:45920", + "accept": true, + "reason": "Same underlying v5 AutoTokenizer regression: wrong token IDs across multiple model families, with Granite as one affected case." + }, + { + "left": "issue:43866", + "right": "issue:44863", + "accept": false, + "reason": "Different model-specific loading failures with no shared symptom or fix path." + }, + { + "left": "issue:43872", + "right": "issue:46304", + "accept": false, + "reason": "bitsandbytes parameter-construction incompatibility is unrelated to distributed build support." + }, + { + "left": "issue:44292", + "right": "issue:46015", + "accept": false, + "reason": "NVFP4 model runtime error is unrelated to documentation examples." + }, + { + "left": "issue:39290", + "right": "issue:43531", + "accept": false, + "reason": "Both mention sliding-window concepts, but they are different models and different concrete failures." + }, + { + "left": "issue:41093", + "right": "issue:43638", + "accept": false, + "reason": "Similar zero-sized index symptoms, but the mask-shape mismatch and ZeRO-3 Bert load error are not the same bug." + }, + { + "left": "issue:43792", + "right": "issue:45901", + "accept": false, + "reason": "Different domains: Whisper loading/runtime vs table-question-answering pipeline crash." + }, + { + "left": "issue:43854", + "right": "issue:43866", + "accept": false, + "reason": "Different model problems: unit-test load failure vs corrupted checkpoint." + }, + { + "left": "issue:43425", + "right": "issue:45070", + "accept": false, + "reason": "Torch version incompatibility is not the same as a PretrainedConfig pydantic-field regression." + }, + { + "left": "issue:43975", + "right": "issue:45901", + "accept": false, + "reason": "Detokenization regression and table QA crash are unrelated." + }, + { + "left": "issue:43792", + "right": "issue:45070", + "accept": false, + "reason": "Whisper runtime failure is unrelated to the pydantic config regression." + }, + { + "left": "issue:43596", + "right": "issue:45800", + "accept": false, + "reason": "Both mention load/runtime problems, but the ZeRO-3 index error is not a torch 2.4.1 compatibility issue." + }, + { + "left": "issue:45362", + "right": "issue:45464", + "accept": false, + "reason": "Same model family, but different interfaces and failure modes: transformers chat crash vs streaming chat/completions failure." + }, + { + "left": "issue:43541", + "right": "issue:46869", + "accept": false, + "reason": "Different model families and execution paths: grouped_mm tracing vs multi-GPU device-map failure." + }, + { + "left": "issue:44292", + "right": "issue:45901", + "accept": false, + "reason": "NVFP4 runtime failure and table QA crash are unrelated." + }, + { + "left": "issue:39290", + "right": "issue:45823", + "accept": false, + "reason": "Different models and different device/config issues; only broad version-related similarity." + }, + { + "left": "issue:45464", + "right": "issue:45685", + "accept": false, + "reason": "Chat streaming failure is unrelated to an MPS histogram kernel missing for Int." + }, + { + "left": "issue:36010", + "right": "issue:39290", + "accept": false, + "reason": "Import error for GenerationMixin is unrelated to the Gemma3TextConfig sliding-window attribute issue." + }, + { + "left": "issue:43541", + "right": "issue:45685", + "accept": false, + "reason": "Different execution failures: torch dynamo grouped_mm tracing vs MPS histogram kernel unsupported." + }, + { + "left": "issue:43541", + "right": "issue:43828", + "accept": false, + "reason": "Grouped_mm tracing error and autocast dtype mismatch are different bugs." + }, + { + "left": "issue:44841", + "right": "issue:45084", + "accept": false, + "reason": "Processor loading failure is unrelated to compile-time template-node errors." + }, + { + "left": "issue:43404", + "right": "issue:43866", + "accept": false, + "reason": "Weight-tying bug in Mistral3 is unrelated to a corrupted Ovis2 checkpoint." + }, + { + "left": "issue:45507", + "right": "issue:46869", + "accept": false, + "reason": "Different model families and different invalid-method/device-map failures." + }, + { + "left": "issue:41093", + "right": "issue:41762", + "accept": false, + "reason": "Both are IndexErrors, but one is a mask-length mismatch and the other is a ZeRO-3 zero-size index error." + }, + { + "left": "issue:45399", + "right": "issue:46661", + "accept": false, + "reason": "Fallback logic for flash-attn2 is unrelated to remote-code model_class config_class errors." + }, + { + "left": "issue:43572", + "right": "issue:45070", + "accept": false, + "reason": "A missing StableLmConfig field is not the same as a broader pydantic PretrainedConfig breakage." + }, + { + "left": "issue:45800", + "right": "issue:45823", + "accept": false, + "reason": "Different regressions: torch/distributed compatibility vs Gemma4 device mismatch." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: issue:45920 is clearly the broad v5 AutoTokenizer regression, and the duplicate judgments are conservative. The one accepted ZeRO-3 pair is reasonably plausible from the shared IndexError/load-time symptom, and the rest of the soft-edge rejections stay on the safe side without overstating equivalence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 3, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:39186", + "issue:40990", + "issue:41093", + "issue:42915", + "issue:43054", + "issue:43296", + "issue:43366", + "issue:43404", + "issue:43531", + "issue:43541", + "issue:43596", + "issue:43645", + "issue:43716", + "issue:43828", + "issue:43873", + "issue:43994", + "issue:44291", + "issue:44360", + "issue:44387", + "issue:44479", + "issue:44492", + "issue:44512", + "issue:44530", + "issue:44805", + "issue:44863", + "issue:44918", + "issue:44960", + "issue:45005", + "issue:45020", + "issue:45070", + "issue:45161", + "issue:45183", + "issue:45399", + "issue:45464", + "issue:45632", + "issue:45698", + "issue:45759", + "issue:45800", + "issue:45902", + "issue:46165", + "issue:46294", + "issue:46463", + "issue:46468", + "issue:46489", + "issue:46491", + "issue:46640", + "issue:46661", + "issue:46869" + ], + "soft_pairs": [ + "issue:43296|issue:43366", + "issue:44387|issue:45005", + "issue:44805|issue:44918", + "issue:43541|issue:45161", + "issue:43828|issue:45464", + "issue:43366|issue:45464", + "issue:44291|issue:45070", + "issue:43531|issue:45070", + "issue:41093|issue:43596", + "issue:45020|issue:45800", + "issue:43716|issue:46294", + "issue:43054|issue:43994", + "issue:44479|issue:46294", + "issue:46468|issue:46869", + "issue:45759|issue:45902", + "issue:43596|issue:46165", + "issue:46640|issue:46661", + "issue:43716|issue:44918", + "issue:39186|issue:44918", + "issue:40990|issue:44960", + "issue:43645|issue:45399", + "issue:45632|issue:45698", + "issue:43404|issue:45005", + "issue:42915|issue:45399", + "issue:43873|issue:45005", + "issue:44530|issue:45005", + "issue:44360|issue:44512", + "issue:44863|issue:45005", + "issue:44492|issue:44512", + "issue:45183|issue:46463", + "issue:46489|issue:46491" + ], + "prepared_review_unit_hash": "d636e2d5160dac2b6fe0c5e04345b931fd1d560c565665241c3785b4be4ac469", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36889, + "estimated_input_tokens": 9223, + "estimated_eval_tokens": 18702 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly unrelated issue reports. The strongest cluster anchor is the cross-model input-validation bug, and I\u2019m only accepting soft edges where the titles clearly point to the same underlying defect; the remote-code cache/module pair is too speculative and should be rejected.", + "confidence": 0.81, + "canonical_issue_reason": "Issue 45183 is the broad umbrella report for the shared input-validation bug across multiple models, which also covers the narrower T5 decoder-input case in 46463.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45183 is the clearest root-cause-oriented report in the set and best represents the shared validation logic defect.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43296", + "right": "issue:43366", + "accept": false, + "reason": "Different products and failure modes: PaddleOCR-VL loading in vLLM vs GGUF GPT-OSS architecture support." + }, + { + "left": "issue:44387", + "right": "issue:45005", + "accept": false, + "reason": "Int4 quantization OOM and tied-weights behavior are unrelated bugs." + }, + { + "left": "issue:44805", + "right": "issue:44918", + "accept": false, + "reason": "Both are shape-related errors, but they involve different model paths and different causes." + }, + { + "left": "issue:43541", + "right": "issue:45161", + "accept": false, + "reason": "Both mention MoE models, but grouped_mm tracing and tensor-parallel support are different code-path problems." + }, + { + "left": "issue:43828", + "right": "issue:45464", + "accept": false, + "reason": "Autocast dtype mismatch and streaming chat/completions failure are not the same bug." + }, + { + "left": "issue:43366", + "right": "issue:45464", + "accept": false, + "reason": "Unrelated issues: GPT-OSS GGUF architecture support vs Qwen3.5 streaming inference." + }, + { + "left": "issue:44291", + "right": "issue:45070", + "accept": false, + "reason": "init_empty_weights argument handling and pydantic PretrainedConfig breakage are different compatibility issues." + }, + { + "left": "issue:43531", + "right": "issue:45070", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior and a pydantic config regression are unrelated." + }, + { + "left": "issue:41093", + "right": "issue:43596", + "accept": false, + "reason": "Both are indexing errors, but they arise in different initialization/training flows and do not look like the same defect." + }, + { + "left": "issue:45020", + "right": "issue:45800", + "accept": false, + "reason": "Remote-code model breakage and a torch/transformers version incompatibility are separate compatibility reports." + }, + { + "left": "issue:43716", + "right": "issue:46294", + "accept": false, + "reason": "Both involve multimodal inputs, but image preprocessor dtype mismatch and Qwen3.5 video token accounting are different bugs." + }, + { + "left": "issue:43054", + "right": "issue:43994", + "accept": false, + "reason": "Same model family, but one is a relative text-embedding quality complaint and the other is a broader bad-output/loading symptom; not clearly the same underlying bug." + }, + { + "left": "issue:44479", + "right": "issue:46294", + "accept": true, + "reason": "Both describe the Qwen video-input regression where video features and video tokens are miscounted; 46294 looks like a specific manifestation of the broader regression in 44479." + }, + { + "left": "issue:46468", + "right": "issue:46869", + "accept": false, + "reason": "Nemotron example breakage and YOLO multi-GPU device-map auto failure are unrelated." + }, + { + "left": "issue:45759", + "right": "issue:45902", + "accept": false, + "reason": "Composite config unwrapping and uninitialized rotary embedding memory are different issues." + }, + { + "left": "issue:43596", + "right": "issue:46165", + "accept": false, + "reason": "deepspeed zero3 init failure and resume-training with trust_remote_code are not the same code-path problem." + }, + { + "left": "issue:46640", + "right": "issue:46661", + "accept": false, + "reason": "Flash Attention 2 issue and remote-code model_class/config_class attribute error are different failures." + }, + { + "left": "issue:45183", + "right": "issue:46463", + "accept": true, + "reason": "Both are about the same input-validation logic; 46463 is the T5-specific instance of the broader validation-message bug in 45183." + }, + { + "left": "issue:46489", + "right": "issue:46491", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different models and different behaviors." + }, + { + "left": "issue:45632", + "right": "issue:45698", + "accept": false, + "reason": "They are plausibly related to remote-code/custom-module loading, but the titles do not clearly show the same concrete bug or a single mergeable fix." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "Overall grounded, but one soft-edge acceptance is too speculative for the packet: issue:45632 vs issue:45698 is plausibly related to remote-code/custom-module loading, yet the titles do not clearly show the same bugfix or a single mergeable PR, so that verdict is not conservative enough. The other accepted pairs look more defensible." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 4, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:24643", + "issue:28530", + "issue:36331", + "issue:38175", + "issue:39692", + "issue:41553", + "issue:41762", + "issue:42915", + "issue:42947", + "issue:43054", + "issue:43493", + "issue:43643", + "issue:43644", + "issue:43782", + "issue:43824", + "issue:43828", + "issue:43856", + "issue:43872", + "issue:43881", + "issue:43883", + "issue:43950", + "issue:43975", + "issue:44292", + "issue:44387", + "issue:44403", + "issue:44451", + "issue:44488", + "issue:44534", + "issue:44589", + "issue:44910", + "issue:44918", + "issue:44928", + "issue:44960", + "issue:44991", + "issue:45020", + "issue:45399", + "issue:45507", + "issue:45520", + "issue:45663", + "issue:45685", + "issue:45701", + "issue:45823", + "issue:45859", + "issue:45902", + "issue:45920", + "issue:45987", + "issue:46489", + "issue:46858" + ], + "soft_pairs": [ + "issue:43975|issue:44292", + "issue:44918|issue:45507", + "issue:43950|issue:44534", + "issue:39692|issue:43054", + "issue:43644|issue:43950", + "issue:36331|issue:41762", + "issue:43872|issue:44589", + "issue:43828|issue:44928", + "issue:45987|issue:46858", + "issue:45920|issue:46489", + "issue:43054|issue:45701", + "issue:38175|issue:43054", + "issue:43950|issue:45902", + "issue:41553|issue:43883", + "issue:43054|issue:43493", + "issue:44910|issue:45859", + "issue:42947|issue:44387", + "issue:45663|issue:45902", + "issue:41553|issue:43881", + "issue:44403|issue:44451", + "issue:43975|issue:45520", + "issue:44403|issue:44488", + "issue:24643|issue:28530", + "issue:42947|issue:43856", + "issue:41762|issue:42915", + "issue:43643|issue:45020", + "issue:44991|issue:45399", + "issue:43824|issue:43872", + "issue:43782|issue:44960", + "issue:43881|issue:43883", + "issue:45685|issue:45823" + ], + "prepared_review_unit_hash": "4b356b265339cf228f0d8664f5a145396faa49c49540e3fb3cae909e39728d2b", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37030, + "estimated_input_tokens": 9258, + "estimated_eval_tokens": 18772 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most items are unrelated, but there is a clear duplicate cluster around the Transformers v5 non-persistent-buffer regression; `issue:43950` is the best representative of that bug family.", + "confidence": 0.77, + "canonical_issue_reason": "`issue:43950` most clearly states the core regression (`from_pretrained()` corrupting non-persistent buffers) and aligns with the closest duplicate reports (`issue:43644`, `issue:44534`).", + "canonical_pr_reason": null, + "best_issue_reason": "It is the most general and explicit report of the shared buffer-corruption bug, with the clearest path to a single fix.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43975", + "right": "issue:44292", + "accept": false, + "reason": "Different models and failure modes; both are tokenizer/loading bugs only at a broad level." + }, + { + "left": "issue:44918", + "right": "issue:45507", + "accept": false, + "reason": "Different concrete problems: TRL SFT embedding unpacking vs GraniteMoEHybrid method call." + }, + { + "left": "issue:43950", + "right": "issue:44534", + "accept": true, + "reason": "Same underlying regression: non-persistent buffers being filled/corrupted in Transformers v5." + }, + { + "left": "issue:39692", + "right": "issue:43054", + "accept": false, + "reason": "Docs/example errors vs model-quality discrepancy; not the same bug." + }, + { + "left": "issue:43644", + "right": "issue:43950", + "accept": true, + "reason": "Both report the same non-persistent-buffer junk/corruption regression." + }, + { + "left": "issue:36331", + "right": "issue:41762", + "accept": false, + "reason": "Unrelated trainer API mismatch vs ZeRO-3 model-loading IndexError." + }, + { + "left": "issue:43872", + "right": "issue:44589", + "accept": false, + "reason": "Different quantization/storage incompatibilities; no shared code-path evidence." + }, + { + "left": "issue:43828", + "right": "issue:44928", + "accept": false, + "reason": "ImportError for a class name vs Qwen3.5 attention/position-id crash." + }, + { + "left": "issue:45987", + "right": "issue:46858", + "accept": true, + "reason": "Same `cache.get_seq_length()` contract bug; the later report exposes the backend-specific failure." + }, + { + "left": "issue:45920", + "right": "issue:46489", + "accept": true, + "reason": "Same tokenizer regression in v5; `46489` looks like a model-specific gap in the broader fix." + }, + { + "left": "issue:43054", + "right": "issue:45701", + "accept": false, + "reason": "SigLIP2 embedding quality complaint vs unrelated tokenization change report." + }, + { + "left": "issue:38175", + "right": "issue:43054", + "accept": false, + "reason": "Both involve SigLIP2, but the symptoms are different and not clearly the same bug." + }, + { + "left": "issue:43950", + "right": "issue:45902", + "accept": false, + "reason": "`45902` is a model-specific `to_empty`/offload issue, not clearly the same bug as the general buffer-corruption regression." + }, + { + "left": "issue:41553", + "right": "issue:43883", + "accept": false, + "reason": "Different models and different loading errors." + }, + { + "left": "issue:43054", + "right": "issue:43493", + "accept": true, + "reason": "Both point to the same SigLIP2 implementation discrepancy behind degraded text embeddings." + }, + { + "left": "issue:44910", + "right": "issue:45859", + "accept": false, + "reason": "Flash-Attention packed-sequence crash vs CPU-offload rotary embedding incompatibility." + }, + { + "left": "issue:42947", + "right": "issue:44387", + "accept": false, + "reason": "Different memory issues with different mechanisms and settings." + }, + { + "left": "issue:45663", + "right": "issue:45902", + "accept": false, + "reason": "Different models and different defects; no shared concrete bug." + }, + { + "left": "issue:41553", + "right": "issue:43881", + "accept": false, + "reason": "Different model-loading failures." + }, + { + "left": "issue:44403", + "right": "issue:44451", + "accept": false, + "reason": "Unrelated: log noise vs a specific model load failure." + }, + { + "left": "issue:43975", + "right": "issue:45520", + "accept": false, + "reason": "Tokenizer detokenization bug vs flash-attn import issue." + }, + { + "left": "issue:44403", + "right": "issue:44488", + "accept": false, + "reason": "General loading noise vs a specific tokenizer/model load failure." + }, + { + "left": "issue:24643", + "right": "issue:28530", + "accept": false, + "reason": "Different training-time errors and unrelated root causes." + }, + { + "left": "issue:42947", + "right": "issue:43856", + "accept": false, + "reason": "Both mention memory, but the bugs are different (LoRA checkpointing vs MoE training footprint)." + }, + { + "left": "issue:41762", + "right": "issue:42915", + "accept": false, + "reason": "Different model families and different loading/configuration failures." + }, + { + "left": "issue:43643", + "right": "issue:45020", + "accept": false, + "reason": "Both involve remote code, but one is missing fields in AutoConfig while the other is a broader regression report." + }, + { + "left": "issue:44991", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer load failure vs flash-attn fallback selection problem." + }, + { + "left": "issue:43824", + "right": "issue:43872", + "accept": false, + "reason": "Import error for a missing class vs bitsandbytes parameter incompatibility." + }, + { + "left": "issue:43782", + "right": "issue:44960", + "accept": false, + "reason": "Different model-specific loading/runtime failures." + }, + { + "left": "issue:43881", + "right": "issue:43883", + "accept": false, + "reason": "Different models and unrelated errors." + }, + { + "left": "issue:45685", + "right": "issue:45823", + "accept": false, + "reason": "Different device/backend issues with different failure signatures." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the accepted soft pairs are conservative and supported by the issue titles/body context. No clear overstatement or non-mergeable PR pairing appears in the chosen cluster." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 5, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:24643", + "issue:30064", + "issue:36331", + "issue:41093", + "issue:41553", + "issue:42491", + "issue:42947", + "issue:43257", + "issue:43404", + "issue:43792", + "issue:43827", + "issue:43828", + "issue:43854", + "issue:43872", + "issue:43940", + "issue:43950", + "issue:44220", + "issue:44291", + "issue:44368", + "issue:44479", + "issue:44488", + "issue:44509", + "issue:44530", + "issue:44560", + "issue:44661", + "issue:44805", + "issue:44863", + "issue:44928", + "issue:44991", + "issue:45020", + "issue:45081", + "issue:45305", + "issue:45313", + "issue:45356", + "issue:45399", + "issue:45488", + "issue:45588", + "issue:45715", + "issue:45759", + "issue:45823", + "issue:45923", + "issue:46315", + "issue:46402", + "issue:46489", + "issue:46661", + "issue:46869", + "issue:46894", + "issue:46897" + ], + "soft_pairs": [ + "issue:45081|issue:45399", + "issue:45020|issue:45823", + "issue:44661|issue:45020", + "issue:42491|issue:44368", + "issue:43828|issue:45823", + "issue:45020|issue:45356", + "issue:45715|issue:46315", + "issue:45488|issue:46489", + "issue:43827|issue:44509", + "issue:43872|issue:46869", + "issue:46894|issue:46897", + "issue:43404|issue:43950", + "issue:45759|issue:46661", + "issue:43872|issue:44291", + "issue:45020|issue:46661", + "issue:43792|issue:44220", + "issue:43257|issue:45305", + "issue:41553|issue:44991", + "issue:36331|issue:41093", + "issue:42947|issue:45923", + "issue:43854|issue:44863", + "issue:43940|issue:45313", + "issue:41553|issue:44488", + "issue:44928|issue:45588", + "issue:41093|issue:44805", + "issue:44479|issue:44560", + "issue:24643|issue:30064", + "issue:44530|issue:44863", + "issue:44863|issue:46402" + ], + "prepared_review_unit_hash": "b053d3cd771ec9efffc27d7954a63581b25b52b933839d960c8bd53a78824221", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 37083, + "estimated_input_tokens": 9271, + "estimated_eval_tokens": 18798 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a heterogeneous set of unrelated bug reports. A few soft pairs are true duplicates or close enough to share one fix: the v5 docs pipeline-removal reports, the `_is_hf_initialized` load-path regression, the Qwen3/Qwen3.5 ZeRO-3 weight-loading failures, and the two training-loss double-shift reports. Most other pairs only share a broad subsystem or similar wording.", + "confidence": 0.84, + "canonical_issue_reason": "No single canonical issue covers the whole cluster; the strongest representative of the true duplicate-like reports is issue 46897, which is a concrete, high-signal instance of the repeated training-loss double-shift bug.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46897 is the cleanest standalone issue in the cluster: it is specific, clearly reproducible, and its sibling 46894 looks like the same underlying bug report with a different affected model.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45081", + "right": "issue:45399", + "accept": false, + "reason": "Different bugs: tokenizer regex patch crash vs flash-attn2 fallback gating." + }, + { + "left": "issue:45020", + "right": "issue:45823", + "accept": false, + "reason": "Both are remote/device-map regressions, but the failures and code paths are different." + }, + { + "left": "issue:44661", + "right": "issue:45020", + "accept": false, + "reason": "One is add-new-model-like tokenizer mapping logic; the other is a broad remote_code regression." + }, + { + "left": "issue:42491", + "right": "issue:44368", + "accept": false, + "reason": "Same model family, but one is a version-compatibility load failure and the other is a warning about tie_word_embeddings." + }, + { + "left": "issue:43828", + "right": "issue:45823", + "accept": false, + "reason": "Different symptoms: autocast dtype mismatch vs device_map forward device mismatch." + }, + { + "left": "issue:45020", + "right": "issue:45356", + "accept": false, + "reason": "Both mention tokenizer/model loading, but the root causes are distinct." + }, + { + "left": "issue:45715", + "right": "issue:46315", + "accept": false, + "reason": "Both are tokenizer performance issues, but they affect different methods and likely different hot paths." + }, + { + "left": "issue:45488", + "right": "issue:46489", + "accept": false, + "reason": "Same model family, but one is a tokenizer-pretokenizer override regression and the other is an output-correctness issue with a different described fix gap." + }, + { + "left": "issue:43827", + "right": "issue:44509", + "accept": true, + "reason": "Both report the same docs regression: removed pipeline tasks still referenced after v5." + }, + { + "left": "issue:43872", + "right": "issue:46869", + "accept": false, + "reason": "Unexpected kwarg vs multi-GPU YOLO device-map failure; not the same bug." + }, + { + "left": "issue:46894", + "right": "issue:46897", + "accept": true, + "reason": "Nearly identical training-loss double-shift reports; same underlying bug, just different affected model." + }, + { + "left": "issue:43404", + "right": "issue:43950", + "accept": false, + "reason": "lm_head tying bug vs non-persistent buffer corruption are unrelated model-loading issues." + }, + { + "left": "issue:45759", + "right": "issue:46661", + "accept": false, + "reason": "Different remote-code/config handling failures with no clear shared concrete code path." + }, + { + "left": "issue:43872", + "right": "issue:44291", + "accept": true, + "reason": "Both center on the unexpected `_is_hf_initialized` argument during model loading/initialization." + }, + { + "left": "issue:45020", + "right": "issue:46661", + "accept": false, + "reason": "Shared remote-code theme, but the reported failures are not specific enough to treat as the same bug." + }, + { + "left": "issue:43792", + "right": "issue:44220", + "accept": false, + "reason": "Different audio/model-loading failures; not the same code-path problem." + }, + { + "left": "issue:43257", + "right": "issue:45305", + "accept": false, + "reason": "Both involve DeepSpeed, but one is missing Qwen3 MOE weight conversion and the other is gradient averaging with loss kwargs." + }, + { + "left": "issue:41553", + "right": "issue:44991", + "accept": false, + "reason": "Tokenizer-loading failures, but for different models and different failure modes." + }, + { + "left": "issue:36331", + "right": "issue:41093", + "accept": false, + "reason": "Trainer signature mismatch vs tensor mask shape mismatch; unrelated." + }, + { + "left": "issue:42947", + "right": "issue:45923", + "accept": false, + "reason": "Both mention gradient checkpointing, but one is a LoRA/PEFT ineffectiveness issue and the other is a missing flag on a remote-code variant." + }, + { + "left": "issue:43854", + "right": "issue:44863", + "accept": false, + "reason": "Different model-specific loading failures with no shared concrete bug." + }, + { + "left": "issue:43940", + "right": "issue:45313", + "accept": true, + "reason": "Same underlying Qwen3-family ZeRO-3 weight-loading failure, just reported on different model variants." + }, + { + "left": "issue:41553", + "right": "issue:44488", + "accept": false, + "reason": "Bad error message vs model-loading failure for a specific checkpoint; not the same issue." + }, + { + "left": "issue:44928", + "right": "issue:45588", + "accept": false, + "reason": "Gradient explosion in RLHF vs flash-attention crash on sink-less models are unrelated." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": false, + "reason": "Same generic IndexError wording, but likely different masks/tensors and different root causes." + }, + { + "left": "issue:44479", + "right": "issue:44560", + "accept": false, + "reason": "Both are video-input regressions, but they affect different model families and symptoms." + }, + { + "left": "issue:24643", + "right": "issue:30064", + "accept": false, + "reason": "Training DeepSpeed weight-shape error vs segmentation-map processing failure; unrelated." + }, + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "PagedAttention linear_attention crash vs checkpoint-loading failure; different bugs." + }, + { + "left": "issue:44863", + "right": "issue:46402", + "accept": false, + "reason": "Different model-loading regressions for different model families." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the accepted soft edges are stated conservatively enough from the available evidence (mainly title-level similarity). The only caveat is that a couple of the duplicate-like calls rely on shared bug patterns rather than explicit cross-links, but that is still within a reasonable conservative read of 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": 23, + "review_unit_index": 6, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:36010", + "issue:38175", + "issue:39692", + "issue:42491", + "issue:42617", + "issue:42915", + "issue:43317", + "issue:43366", + "issue:43531", + "issue:43550", + "issue:43596", + "issue:43632", + "issue:43638", + "issue:43644", + "issue:43645", + "issue:43646", + "issue:43716", + "issue:43720", + "issue:43824", + "issue:43828", + "issue:43950", + "issue:43957", + "issue:43994", + "issue:44155", + "issue:44534", + "issue:44560", + "issue:44589", + "issue:44661", + "issue:44683", + "issue:44756", + "issue:44805", + "issue:44877", + "issue:44910", + "issue:44918", + "issue:45030", + "issue:45325", + "issue:45584", + "issue:45663", + "issue:45672", + "issue:45685", + "issue:45812", + "issue:45823", + "issue:45834", + "issue:45902", + "issue:46489", + "issue:46515", + "issue:46858", + "issue:46899" + ], + "soft_pairs": [ + "issue:43366|issue:43828", + "issue:43596|issue:46515", + "issue:45325|issue:45902", + "issue:43366|issue:43531", + "issue:43638|issue:46515", + "issue:45663|issue:45823", + "issue:44155|issue:45584", + "issue:43366|issue:44589", + "issue:43824|issue:43957", + "issue:43950|issue:45834", + "issue:42915|issue:43957", + "issue:44910|issue:46858", + "issue:43716|issue:44560", + "issue:43645|issue:43646", + "issue:44560|issue:45685", + "issue:43550|issue:44910", + "issue:44683|issue:46515", + "issue:42617|issue:43366", + "issue:44560|issue:44918", + "issue:36010|issue:42915", + "issue:45812|issue:46489", + "issue:45823|issue:46515", + "issue:44661|issue:45823", + "issue:44560|issue:44805", + "issue:38175|issue:43994", + "issue:45672|issue:46899", + "issue:44877|issue:45030", + "issue:43632|issue:45834", + "issue:43638|issue:44661", + "issue:43596|issue:44661", + "issue:42491|issue:43720", + "issue:44661|issue:44805", + "issue:43317|issue:44756", + "issue:38175|issue:39692", + "issue:43644|issue:44534" + ], + "prepared_review_unit_hash": "809b93a4ffc5c33e193f8488befbd30656ac3414fc7f1184af0634fc25adda80", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38346, + "estimated_input_tokens": 9587, + "estimated_eval_tokens": 19430 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: most pairs are only superficially similar, but a few look like true duplicates, especially the non-persistent-buffer regression and a couple of model-specific output/dtype regressions.", + "confidence": 0.71, + "canonical_issue_reason": "issue:43644 is the clearest canonical issue in the set: it states the root regression directly, is broad enough to subsume the near-identical follow-up report 44534, and is more precise than the more symptom-driven or model-specific issues.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43644 is the best consolidation target because it has the cleanest bug description and the strongest duplicate overlap, while most other issues in the cluster are unrelated.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43366", + "right": "issue:43828", + "accept": false, + "reason": "Different bugs: GGUF gpt-oss support vs a Phi-tiny-MoE dtype/autocast failure." + }, + { + "left": "issue:43596", + "right": "issue:46515", + "accept": false, + "reason": "Both mention failures, but one is a deepspeed/Bert index error and the other is a torch version incompatibility report." + }, + { + "left": "issue:45325", + "right": "issue:45902", + "accept": false, + "reason": "Different model families and different code paths: Qwen2.5-VL rope indexing vs Qwen3_5Moe rotary inv_freq materialization." + }, + { + "left": "issue:43366", + "right": "issue:43531", + "accept": false, + "reason": "Unrelated issues: GGUF support for gpt-oss vs Qwen3-MoE sliding_window behavior." + }, + { + "left": "issue:43638", + "right": "issue:46515", + "accept": false, + "reason": "A similar stack is involved, but the concrete bug is different: Bert/deepspeed zero3 index error vs general torch incompatibility." + }, + { + "left": "issue:45663", + "right": "issue:45823", + "accept": false, + "reason": "Both are Gemma4, but one is FSDP2 shared_kv_states rebuilding and the other is device_map auto PLE mismatch." + }, + { + "left": "issue:44155", + "right": "issue:45584", + "accept": false, + "reason": "Different subsystems and symptoms: AudioFlamingo3 batching leak vs Whisper empty-transcription generation failure." + }, + { + "left": "issue:43366", + "right": "issue:44589", + "accept": false, + "reason": "No shared bug: gpt-oss GGUF support vs a Float8 storage lookup error." + }, + { + "left": "issue:43824", + "right": "issue:43957", + "accept": false, + "reason": "ImportError for a specific class is not the same bug as meta-device loading breaking some models." + }, + { + "left": "issue:43950", + "right": "issue:45834", + "accept": false, + "reason": "The buffer-corruption regression is unrelated to Kosmos2.5 long-OCR indexing." + }, + { + "left": "issue:42915", + "right": "issue:43957", + "accept": false, + "reason": "Different failures: Qwen3Moe FP8 config vs generic meta-device model loading." + }, + { + "left": "issue:44910", + "right": "issue:46858", + "accept": false, + "reason": "Distinct attention/cache bugs: Qwen3.5 flash-attention position_ids vs static-cache seq_length tensor arithmetic." + }, + { + "left": "issue:43716", + "right": "issue:44560", + "accept": false, + "reason": "Different model paths and errors: Mistral-3 preprocessor dtype mismatch vs Qwen3-vl video StopIteration." + }, + { + "left": "issue:43645", + "right": "issue:43646", + "accept": true, + "reason": "These describe the same underlying regression: custom model initialization breaks in Transformers 5, with one report just being the Jupyter-specific case." + }, + { + "left": "issue:44560", + "right": "issue:45685", + "accept": false, + "reason": "Completely different issues: video embedding StopIteration vs MPS histogram on Int." + }, + { + "left": "issue:43550", + "right": "issue:44910", + "accept": false, + "reason": "Both are attention-related, but the concrete failures are different code paths and different root causes." + }, + { + "left": "issue:44683", + "right": "issue:46515", + "accept": false, + "reason": "Compiled flex_attention on torch >=2.9 is unrelated to a torch 2.4.1 incompatibility report." + }, + { + "left": "issue:42617", + "right": "issue:43366", + "accept": false, + "reason": "A runtime script failure is not the same as GGUF gpt-oss architecture support." + }, + { + "left": "issue:44560", + "right": "issue:44918", + "accept": false, + "reason": "Different bugs: Qwen3-vl video StopIteration vs TRL embedding unpacking failure." + }, + { + "left": "issue:36010", + "right": "issue:42915", + "accept": false, + "reason": "Import path regression for GenerationMixin is unrelated to Qwen3Moe FineGrainedFP8Config failure." + }, + { + "left": "issue:45812", + "right": "issue:46489", + "accept": false, + "reason": "Both are tokenizer-related, but they concern different model families and different underlying causes." + }, + { + "left": "issue:45823", + "right": "issue:46515", + "accept": false, + "reason": "Gemma4 device placement bug vs a generic torch version incompatibility." + }, + { + "left": "issue:44661", + "right": "issue:45823", + "accept": false, + "reason": "Tokenizer mapping registration failure is unrelated to Gemma4 PLE device mismatch." + }, + { + "left": "issue:44560", + "right": "issue:44805", + "accept": false, + "reason": "Different symptoms and code paths: video StopIteration vs mask/tensor shape mismatch." + }, + { + "left": "issue:38175", + "right": "issue:43994", + "accept": true, + "reason": "Same model and same class of bad-output regression for SigLIP2; both reports point to the same inference path failing." + }, + { + "left": "issue:45672", + "right": "issue:46899", + "accept": true, + "reason": "Both report the same Gemma4 multimodal 4-bit dtype-cast regression affecting audio/vision modules." + }, + { + "left": "issue:44877", + "right": "issue:45030", + "accept": false, + "reason": "Different model/config validation failures with no shared code path." + }, + { + "left": "issue:43632", + "right": "issue:45834", + "accept": false, + "reason": "_is_hf_initialized flag regression is unrelated to Kosmos2.5 OCR index errors." + }, + { + "left": "issue:43638", + "right": "issue:44661", + "accept": false, + "reason": "Bert/deepspeed zero3 initialization and tokenizer mapping registration are different bugs." + }, + { + "left": "issue:43596", + "right": "issue:44661", + "accept": false, + "reason": "No shared underlying bug: Bert zero3 index error vs tokenizer mapping issue." + }, + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "LoRA/model-compatibility regression is unrelated to packed-weight unloading in BitNet accelerate loading." + }, + { + "left": "issue:44661", + "right": "issue:44805", + "accept": false, + "reason": "Tokenizer mapping registration issue is unrelated to a tensor mask shape error." + }, + { + "left": "issue:43317", + "right": "issue:44756", + "accept": false, + "reason": "Both involve loading/memory, but one is dequantized offload on device_map=auto and the other is Strix Halo mmap OOM." + }, + { + "left": "issue:38175", + "right": "issue:39692", + "accept": false, + "reason": "Same model family, but one is bad inference outputs while the other is a docs-example/quantization failure; not the same concrete bug." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "Near-identical reports of the same non-persistent-buffer junk/corruption regression in Transformers 5.x." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The consolidation choice is grounded in the packet: issue 43644 and 44534 have nearly identical titles, and the other accepted pairs are likewise supported by the issue titles as same-bug follow-ups rather than merely similar symptoms. The rejected soft edges are conservative and mostly stay at the title level without overstating evidence. No mergeability concerns apply here since these are issue pairs, not PR pairs." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 7, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:39692", + "issue:41720", + "issue:42491", + "issue:42915", + "issue:43257", + "issue:43425", + "issue:43644", + "issue:43645", + "issue:43708", + "issue:43854", + "issue:43856", + "issue:43873", + "issue:43931", + "issue:43957", + "issue:43994", + "issue:44075", + "issue:44155", + "issue:44164", + "issue:44291", + "issue:44387", + "issue:44402", + "issue:44451", + "issue:44488", + "issue:44756", + "issue:44863", + "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:45507", + "issue:45588", + "issue:45812", + "issue:46050", + "issue:46459", + "issue:46468", + "issue:46512", + "issue:46846", + "issue:46897", + "issue:46901" + ], + "soft_pairs": [ + "issue:42491|issue:45276", + "issue:41720|issue:42915", + "issue:42915|issue:44756", + "issue:44451|issue:44488", + "issue:44960|issue:44977", + "issue:46459|issue:46512", + "issue:44863|issue:46846", + "issue:45313|issue:45357", + "issue:44488|issue:44991", + "issue:45507|issue:46468", + "issue:43644|issue:43645", + "issue:42491|issue:45292", + "issue:39692|issue:43994", + "issue:43257|issue:43854", + "issue:43856|issue:43873", + "issue:43856|issue:44387", + "issue:43708|issue:44075", + "issue:43854|issue:43931", + "issue:42491|issue:45335", + "issue:44960|issue:46050", + "issue:46897|issue:46901", + "issue:45488|issue:45812", + "issue:45276|issue:45335", + "issue:45362|issue:45406", + "issue:44164|issue:44291", + "issue:44910|issue:45588", + "issue:44402|issue:44488", + "issue:43425|issue:43957", + "issue:39692|issue:44960", + "issue:41720|issue:44155" + ], + "prepared_review_unit_hash": "8fb1f90a10c2bfc133ef8024730424e3ee1e93da1e83354d6593fd74fe2e6368", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36103, + "estimated_input_tokens": 9026, + "estimated_eval_tokens": 18308 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a heterogeneous mix of unrelated model-loading, tokenizer, quantization, and training bugs. Only one soft edge looks like a true duplicate: the two training-loss double-shift reports. The tokenizer/model-loading pairs are similar in theme but not strong enough to merge confidently.", + "confidence": 0.68, + "canonical_issue_reason": "Issue 43873 is the broadest active bug in the set and the best umbrella representative for the recurring runtime-memory/quantization theme, even though the cluster is not a tight duplicate group.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43873 is the best single triage anchor: it is open, clearly actionable, and closest to the common inference-memory/offloading failures that recur across several items.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42491", + "right": "issue:45276", + "accept": false, + "reason": "Different model families and different failures; only broad embedding/resize wording overlaps." + }, + { + "left": "issue:41720", + "right": "issue:42915", + "accept": false, + "reason": "Different Qwen3 failures: auto device mapping cuda assert vs FineGrainedFP8Config load failure." + }, + { + "left": "issue:42915", + "right": "issue:44756", + "accept": false, + "reason": "Unrelated bugs: Qwen3 MoE FP8 failure vs Strix Halo mmap/OOM behavior." + }, + { + "left": "issue:44451", + "right": "issue:44488", + "accept": false, + "reason": "Both are BERT-family load failures, but the titles point to different model-specific regressions without enough evidence of the same root cause." + }, + { + "left": "issue:44960", + "right": "issue:44977", + "accept": false, + "reason": "Different models and different flash-attention problems." + }, + { + "left": "issue:46459", + "right": "issue:46512", + "accept": false, + "reason": "Different code paths: dtype ignored on composite checkpoint load vs stale config.dtype after from_config." + }, + { + "left": "issue:44863", + "right": "issue:46846", + "accept": false, + "reason": "Different model families and different TP/load-planning failures." + }, + { + "left": "issue:45313", + "right": "issue:45357", + "accept": false, + "reason": "Both involve Qwen3.5, but one is ZeRO-3 weight loading and the other is save_pretrained visual-key regression." + }, + { + "left": "issue:44488", + "right": "issue:44991", + "accept": false, + "reason": "Same general tokenizer-loading theme, but different checkpoints and no clear proof of the same underlying bug." + }, + { + "left": "issue:45507", + "right": "issue:46468", + "accept": false, + "reason": "Different models and unrelated invalid-method vs broken-example failures." + }, + { + "left": "issue:43644", + "right": "issue:43645", + "accept": false, + "reason": "Same release, but distinct bugs: non-persistent buffer junk vs Jupyter custom-model initialization." + }, + { + "left": "issue:42491", + "right": "issue:45292", + "accept": false, + "reason": "Different issues; only a shared resize/token-embedding theme." + }, + { + "left": "issue:39692", + "right": "issue:43994", + "accept": false, + "reason": "Both mention SigLIP2, but one is a docs example with quantization/model-processor mismatch and the other is nonsensical inference output." + }, + { + "left": "issue:43257", + "right": "issue:43854", + "accept": false, + "reason": "Different models and different failure modes; only broad loading trouble overlaps." + }, + { + "left": "issue:43856", + "right": "issue:43873", + "accept": false, + "reason": "Both are memory-related, but one is Qwen3 MoE training efficiency and the other is quantization/offloading behavior." + }, + { + "left": "issue:43856", + "right": "issue:44387", + "accept": false, + "reason": "Related to memory pressure, but one is training inefficiency and the other is an int4 quantization regression." + }, + { + "left": "issue:43708", + "right": "issue:44075", + "accept": false, + "reason": "Completely different training scheduler vs optimizer-argument bug." + }, + { + "left": "issue:43854", + "right": "issue:43931", + "accept": false, + "reason": "Different models and different loading failures." + }, + { + "left": "issue:42491", + "right": "issue:45335", + "accept": false, + "reason": "Both mention embeddings, but the model subclasses and broken paths differ." + }, + { + "left": "issue:44960", + "right": "issue:46050", + "accept": false, + "reason": "Different serialization bugs: GLM5 model issue vs sinq save/load corruption." + }, + { + "left": "issue:46897", + "right": "issue:46901", + "accept": true, + "reason": "Both describe the same training-loss double-shift bug: loss is computed against shifted labels instead of the original labels." + }, + { + "left": "issue:45488", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but one is a LlamaTokenizer pre-tokenizer override and the other is wrong token IDs for Granite models." + }, + { + "left": "issue:45276", + "right": "issue:45335", + "accept": false, + "reason": "Same API surface, but likely model-specific resize_token_embeddings omissions rather than one shared bug." + }, + { + "left": "issue:45362", + "right": "issue:45406", + "accept": false, + "reason": "Unrelated crashes in transformers chat vs transformers serve, with different model/processor causes." + }, + { + "left": "issue:44164", + "right": "issue:44291", + "accept": false, + "reason": "Different save/load internals: extra_state handling vs init_empty_weights argument mismatch." + }, + { + "left": "issue:44910", + "right": "issue:45588", + "accept": false, + "reason": "Both are flash_attention.py issues, but the bugs are different: packed-sequence misinterpretation vs s_aux=None AttributeError." + }, + { + "left": "issue:44402", + "right": "issue:44488", + "accept": false, + "reason": "Related tokenizer/vocab symptoms, but one is a vocab-size mismatch and the other is a load failure." + }, + { + "left": "issue:43425", + "right": "issue:43957", + "accept": false, + "reason": "Torch version compatibility vs meta-device model-loading breakage are different problems." + }, + { + "left": "issue:39692", + "right": "issue:44960", + "accept": false, + "reason": "Unrelated SigLIP2 docs/runtime issue vs GLM5 issue." + }, + { + "left": "issue:41720", + "right": "issue:44155", + "accept": false, + "reason": "Different subsystems and failures: Qwen3 device mapping vs AudioFlamingo3 batched inference leakage." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft edge, issue:46897 vs issue:46901, is well supported by the titles and descriptions as the same training-loss double-shift pattern. The other pairwise rejections are cautious rather than overstated, and the umbrella/canonical choice of issue:43873 is defensible as an open, actionable anchor." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 8, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:36010", + "issue:39404", + "issue:42796", + "issue:42947", + "issue:43335", + "issue:43381", + "issue:43404", + "issue:43425", + "issue:43454", + "issue:43645", + "issue:43646", + "issue:43708", + "issue:43828", + "issue:43854", + "issue:43873", + "issue:43957", + "issue:44079", + "issue:44155", + "issue:44164", + "issue:44291", + "issue:44292", + "issue:44387", + "issue:44402", + "issue:44451", + "issue:44479", + "issue:44589", + "issue:44756", + "issue:44863", + "issue:44877", + "issue:44912", + "issue:44928", + "issue:45216", + "issue:45276", + "issue:45292", + "issue:45335", + "issue:45588", + "issue:45663", + "issue:45684", + "issue:45698", + "issue:45704", + "issue:45759", + "issue:45823", + "issue:45859", + "issue:46326", + "issue:46468", + "issue:46829", + "issue:46894", + "issue:46901" + ], + "soft_pairs": [ + "issue:45663|issue:45759", + "issue:44079|issue:46829", + "issue:43404|issue:43454", + "issue:44291|issue:44756", + "issue:42947|issue:45704", + "issue:43873|issue:44387", + "issue:44292|issue:44912", + "issue:36010|issue:39404", + "issue:45663|issue:45859", + "issue:44387|issue:44756", + "issue:42947|issue:43381", + "issue:44155|issue:45823", + "issue:43957|issue:44756", + "issue:43335|issue:45663", + "issue:45276|issue:45292", + "issue:44863|issue:46468", + "issue:44479|issue:44877", + "issue:43381|issue:44928", + "issue:44164|issue:44756", + "issue:45684|issue:45698", + "issue:45588|issue:45663", + "issue:43854|issue:45216", + "issue:42796|issue:46326", + "issue:45292|issue:45335", + "issue:46894|issue:46901", + "issue:43828|issue:43957", + "issue:44402|issue:44451", + "issue:43646|issue:43708", + "issue:43645|issue:44164", + "issue:44589|issue:44756", + "issue:43425|issue:43828" + ], + "prepared_review_unit_hash": "544cbb29b5bae9efff4d0eb1792a3139adff137110832aa7c9a80a435d63e111", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37130, + "estimated_input_tokens": 9283, + "estimated_eval_tokens": 18822 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly unrelated issues; only a small duplicate cluster around `resize_token_embeddings` and a likely shared training-loss double-shift bug stand out.", + "confidence": 0.81, + "canonical_issue_reason": "issue:45292 is the most generic and central `resize_token_embeddings` regression, and it has the strongest overlap with the closely related embedding-resize reports.", + "canonical_pr_reason": null, + "best_issue_reason": "Best representative of the cluster because it is the clearest, most general duplicate hub among the accepted issue pairs.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45663", + "right": "issue:45759", + "accept": false, + "reason": "Different model paths and failures: FSDP2 shared-kv state rebuild vs. composite Qwen config unwrapping." + }, + { + "left": "issue:44079", + "right": "issue:46829", + "accept": false, + "reason": "Both involve output structures, but they describe different bugs in different classes of outputs." + }, + { + "left": "issue:43404", + "right": "issue:43454", + "accept": false, + "reason": "Both mention untied `lm_head`, but they are separate model families and not clearly the same concrete code-path bug." + }, + { + "left": "issue:44291", + "right": "issue:44756", + "accept": false, + "reason": "Unrelated loading/initialization failure vs. mmap/OOM platform issue." + }, + { + "left": "issue:42947", + "right": "issue:45704", + "accept": false, + "reason": "Gradient checkpointing with LoRA is unrelated to the Apex RMSNorm memory leak." + }, + { + "left": "issue:43873", + "right": "issue:44387", + "accept": false, + "reason": "Both are quantization/memory related, but one is offloading behavior and the other is int4 OOM; not the same bug." + }, + { + "left": "issue:44292", + "right": "issue:44912", + "accept": false, + "reason": "Different quantization formats and outcomes: load failure vs. fallback to bf16." + }, + { + "left": "issue:36010", + "right": "issue:39404", + "accept": false, + "reason": "Import error for `GenerationMixin` is unrelated to Whisper pipeline `return_language` behavior." + }, + { + "left": "issue:45663", + "right": "issue:45859", + "accept": false, + "reason": "Different model-specific offload/forward problems with no clear shared code-path." + }, + { + "left": "issue:44387", + "right": "issue:44756", + "accept": false, + "reason": "Both mention OOM, but the causes are unrelated." + }, + { + "left": "issue:42947", + "right": "issue:43381", + "accept": false, + "reason": "Both involve gradient checkpointing, but one is LoRA ineffectiveness and the other is eval-mode incompatibility." + }, + { + "left": "issue:44155", + "right": "issue:45823", + "accept": false, + "reason": "Different model families and failures; no same underlying bug is evident." + }, + { + "left": "issue:43957", + "right": "issue:44756", + "accept": false, + "reason": "Meta-device loading issues are not the same as the Strix Halo mmap OOM bug." + }, + { + "left": "issue:43335", + "right": "issue:45663", + "accept": false, + "reason": "Separate bugs in different models: SwitchTransformers config construction vs. Gemma4 FSDP2 attention state." + }, + { + "left": "issue:45276", + "right": "issue:45292", + "accept": true, + "reason": "Same `resize_token_embeddings` regression; 45276 is a model-specific instance of the generic output_embeddings problem in 45292." + }, + { + "left": "issue:44863", + "right": "issue:46468", + "accept": false, + "reason": "Nemotron checkpoint loading and the broken official example are related by model family but not clearly the same bug." + }, + { + "left": "issue:44479", + "right": "issue:44877", + "accept": false, + "reason": "Different regressions in unrelated model/config loading paths." + }, + { + "left": "issue:43381", + "right": "issue:44928", + "accept": false, + "reason": "Gradient checkpointing in eval mode is unrelated to the Qwen3.5 RLHF NaN issue." + }, + { + "left": "issue:44164", + "right": "issue:44756", + "accept": false, + "reason": "Save/from_pretrained extra_state handling is unrelated to the Strix Halo mmap OOM bug." + }, + { + "left": "issue:45684", + "right": "issue:45698", + "accept": false, + "reason": "Both concern custom-model save/load, but one is permissions propagation and the other is wrong module loading." + }, + { + "left": "issue:45588", + "right": "issue:45663", + "accept": false, + "reason": "Flash-attention sinkless-model crash is unrelated to the Gemma4 FSDP2 KeyError." + }, + { + "left": "issue:43854", + "right": "issue:45216", + "accept": false, + "reason": "Different model-loading regressions and different failure modes." + }, + { + "left": "issue:42796", + "right": "issue:46326", + "accept": false, + "reason": "Tokenizer BPE-marker handling is not the same bug as `continue_final_message` on mistral_common tokenizers." + }, + { + "left": "issue:45292", + "right": "issue:45335", + "accept": true, + "reason": "Same class of bug: `resize_token_embeddings` fails to propagate resized embeddings to model-specific embedding references." + }, + { + "left": "issue:46894", + "right": "issue:46901", + "accept": true, + "reason": "Both describe the same training-loss double-shift bug: the loss is computed against labels shifted by one position instead of the original labels." + }, + { + "left": "issue:43828", + "right": "issue:43957", + "accept": false, + "reason": "Different dtype/meta-loading regressions with no evidence of the same underlying issue." + }, + { + "left": "issue:44402", + "right": "issue:44451", + "accept": false, + "reason": "Both are model-load problems, but one is vocab/token placement and the other is a separate loading failure." + }, + { + "left": "issue:43646", + "right": "issue:43708", + "accept": false, + "reason": "Custom model initialization breakage and Trainer `resume_from_checkpoint` max-step miscalculation are unrelated." + }, + { + "left": "issue:43645", + "right": "issue:44164", + "accept": false, + "reason": "Notebook custom-model initialization and extra_state save/load are different issues." + }, + { + "left": "issue:44589", + "right": "issue:44756", + "accept": false, + "reason": "Storage-object lookup failure is unrelated to the Strix Halo mmap/OOM issue." + }, + { + "left": "issue:43425", + "right": "issue:43828", + "accept": false, + "reason": "Torch version compatibility and autocast dtype mismatch are separate problems." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The analysis is grounded in the packet and stays conservative on the soft edges. The accepted duplicate pairs all look mergeable into single bugfix PRs: the resize_token_embeddings cluster (45276/45292/45335) and the training-loss double-shift pair (46894/46901). Rejections are generally cautious and not overstated." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 9, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:36032", + "issue:39404", + "issue:43299", + "issue:43425", + "issue:43452", + "issue:43596", + "issue:43611", + "issue:43644", + "issue:43645", + "issue:43716", + "issue:43856", + "issue:43901", + "issue:43906", + "issue:43931", + "issue:43994", + "issue:44079", + "issue:44164", + "issue:44265", + "issue:44479", + "issue:44488", + "issue:44509", + "issue:44530", + "issue:44589", + "issue:44671", + "issue:44683", + "issue:44756", + "issue:44805", + "issue:44898", + "issue:44912", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45479", + "issue:45701", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:46050", + "issue:46315", + "issue:46326", + "issue:46491", + "issue:46665", + "issue:46821", + "issue:46869", + "issue:46894", + "issue:46897", + "issue:46899" + ], + "soft_pairs": [ + "issue:46315|issue:46491", + "issue:45812|issue:46315", + "issue:39404|issue:45701", + "issue:44530|issue:45084", + "issue:43611|issue:43645", + "issue:39404|issue:46665", + "issue:44683|issue:45823", + "issue:43425|issue:46869", + "issue:43425|issue:43716", + "issue:43425|issue:45823", + "issue:44589|issue:45084", + "issue:44479|issue:44805", + "issue:43906|issue:44530", + "issue:43901|issue:44509", + "issue:44671|issue:44991", + "issue:43994|issue:44079", + "issue:44265|issue:44898", + "issue:36032|issue:43452", + "issue:39404|issue:44912", + "issue:46050|issue:46326", + "issue:44164|issue:45759", + "issue:43906|issue:46326", + "issue:44991|issue:45081", + "issue:43299|issue:43931", + "issue:45823|issue:46899", + "issue:45479|issue:46894", + "issue:43856|issue:44756", + "issue:43596|issue:44683", + "issue:45479|issue:46897", + "issue:44164|issue:46821", + "issue:44488|issue:44987", + "issue:44488|issue:45081", + "issue:43611|issue:43644" + ], + "prepared_review_unit_hash": "c01c773c21081d09e102b1f758de2084a71d89e1cea0150b5009665743e298c8", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37739, + "estimated_input_tokens": 9435, + "estimated_eval_tokens": 19126 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most items are separate regression reports across different models, tokenizer paths, or infra behaviors. The closest thematic area is v5 model-loading regressions, with issue 43611 as the best broad representative, but the soft pairs are still too heterogeneous to merge as duplicates.", + "confidence": 0.78, + "canonical_issue_reason": "Issue 43611 is the broadest and most central core regression report here: it targets Transformers 5.0.0 model-loading behavior and is closest in scope to several other loading/init complaints.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43611 is the best overall issue candidate because it is a clear, general loading regression rather than a model-specific symptom, so it best represents the cluster theme.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46315", + "right": "issue:46491", + "accept": false, + "reason": "One is a tokenizer performance regression in convert_tokens_to_ids; the other is a CodeLlama encode/decode space-roundtrip correctness bug. Different code paths and symptoms." + }, + { + "left": "issue:45812", + "right": "issue:46315", + "accept": false, + "reason": "Granite token IDs are wrong; the other is slower conversion for slow tokenizers with many added tokens. Same tokenizer area, but not the same bug." + }, + { + "left": "issue:39404", + "right": "issue:45701", + "accept": false, + "reason": "Whisper pipeline return_language regression is unrelated to a general tokenization-version-change report." + }, + { + "left": "issue:44530", + "right": "issue:45084", + "accept": false, + "reason": "PagedAttentionCache/linear_attention crash on Qwen3.5 is unrelated to a template-node compilation TypeError." + }, + { + "left": "issue:43611", + "right": "issue:43645", + "accept": false, + "reason": "Both involve v5 model init/loading regressions, but one is base_model_prefix loading and the other is defining custom models in notebooks; not the same concrete bug." + }, + { + "left": "issue:39404", + "right": "issue:46665", + "accept": false, + "reason": "Whisper pipeline return_language and from_pretrained inside Python classes are distinct behaviors with different failure modes." + }, + { + "left": "issue:44683", + "right": "issue:45823", + "accept": false, + "reason": "Torch >=2.9 flex_attention compilation failure is unrelated to Gemma4 device placement mismatch." + }, + { + "left": "issue:43425", + "right": "issue:46869", + "accept": false, + "reason": "A torch version incompatibility report is not the same as a YOLO multi-GPU device_map bug." + }, + { + "left": "issue:43425", + "right": "issue:43716", + "accept": false, + "reason": "Torch 2.10 incompatibility and Mistral-3 dtype mismatch are different problems." + }, + { + "left": "issue:43425", + "right": "issue:45823", + "accept": false, + "reason": "Library-version incompatibility is not the same as a device_map forward-pass mismatch." + }, + { + "left": "issue:44589", + "right": "issue:45084", + "accept": false, + "reason": "Float8 storage lookup failure and non-template-node compilation are unrelated runtime errors." + }, + { + "left": "issue:44479", + "right": "issue:44805", + "accept": false, + "reason": "Video-input regression for Qwen VL models and a generic mask/index shape mismatch are not the same underlying defect." + }, + { + "left": "issue:43906", + "right": "issue:44530", + "accept": false, + "reason": "An isolated reproduction of another issue is not the same as the PagedAttentionCache linear_attention crash." + }, + { + "left": "issue:43901", + "right": "issue:44509", + "accept": false, + "reason": "Both are docs-related, but they describe different pipeline task doc regressions." + }, + { + "left": "issue:44671", + "right": "issue:44991", + "accept": false, + "reason": "CamemBERT masked-LM prediction regression and tokenizer loading failure for a different model are distinct." + }, + { + "left": "issue:43994", + "right": "issue:44079", + "accept": false, + "reason": "SigLIP2 incorrect outputs and ModelOutput key assignment logic are different bugs in different layers." + }, + { + "left": "issue:44265", + "right": "issue:44898", + "accept": false, + "reason": "torch.export/exportability via torch_compilable_check is unrelated to Perceiver interpolation at non-default resolution." + }, + { + "left": "issue:36032", + "right": "issue:43452", + "accept": false, + "reason": "T5Tokenizer add_special_tokens conflict and gguf_file/from_pretrained breakage are different tokenizer/model-loading failures." + }, + { + "left": "issue:39404", + "right": "issue:44912", + "accept": false, + "reason": "Whisper return_language pipeline behavior is unrelated to MXFP4 quantization fallback on git-oss-20b." + }, + { + "left": "issue:46050", + "right": "issue:46326", + "accept": false, + "reason": "Quantized save/load breakage after sinq is not the same as the mistral_common continue_final_message bug." + }, + { + "left": "issue:44164", + "right": "issue:45759", + "accept": false, + "reason": "extra_state save/load handling and unwrapping text_config for composite Qwen configs are different code paths." + }, + { + "left": "issue:43906", + "right": "issue:46326", + "accept": false, + "reason": "A reproduction of another issue does not match the mistral_common continue_final_message regression." + }, + { + "left": "issue:44991", + "right": "issue:45081", + "accept": false, + "reason": "Both are tokenizer-related, but one is wrong loading behavior for a specific model and the other is a Mistral regex patch attribute error; not the same bug." + }, + { + "left": "issue:43299", + "right": "issue:43931", + "accept": false, + "reason": "Both involve Qwen3-VL loading, but one is a broad dev-version regression for MoE models and the other is a specific weight-shape mismatch; too different to merge." + }, + { + "left": "issue:45823", + "right": "issue:46899", + "accept": false, + "reason": "Gemma4 device_map mismatch during forward and unsafe dtype casting in multimodal embedders are separate multimodal bugs." + }, + { + "left": "issue:45479", + "right": "issue:46894", + "accept": false, + "reason": "Degenerate zero loss in sequence classification is unrelated to the ASR double-shift training-loss bug." + }, + { + "left": "issue:43856", + "right": "issue:44756", + "accept": false, + "reason": "Qwen3 MoE training memory inefficiency and disabling mmap on Strix Halo are unrelated performance/oom reports." + }, + { + "left": "issue:43596", + "right": "issue:44683", + "accept": false, + "reason": "Zero3/BertModel indexing failure and compiled flex_attention incompatibility are different failures." + }, + { + "left": "issue:45479", + "right": "issue:46897", + "accept": false, + "reason": "The sequence-classification zero-loss bug is not the same as the Florence2 training-loss double-shift regression." + }, + { + "left": "issue:44164", + "right": "issue:46821", + "accept": false, + "reason": "extra_state handling and TimesFM window_size AttributeError are unrelated issues." + }, + { + "left": "issue:44488", + "right": "issue:44987", + "accept": false, + "reason": "Both are loading failures, but for different models and likely different root causes." + }, + { + "left": "issue:44488", + "right": "issue:45081", + "accept": false, + "reason": "A sleng-bert load failure and the Mistral regex/backend_tokenizer error are distinct tokenizer-loading bugs." + }, + { + "left": "issue:43611", + "right": "issue:43644", + "accept": false, + "reason": "Both are v5 regressions in model handling, but one is about base_model_prefix loading and the other about non-persistent buffers; not the same concrete fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 10, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:32090", + "issue:34567", + "issue:36032", + "issue:39404", + "issue:40444", + "issue:40990", + "issue:42175", + "issue:42796", + "issue:42915", + "issue:43262", + "issue:43317", + "issue:43377", + "issue:43404", + "issue:43421", + "issue:43452", + "issue:43482", + "issue:43541", + "issue:43582", + "issue:43632", + "issue:43643", + "issue:43644", + "issue:43645", + "issue:43646", + "issue:43716", + "issue:43746", + "issue:43883", + "issue:44206", + "issue:44220", + "issue:44451", + "issue:44462", + "issue:44488", + "issue:44530", + "issue:44534", + "issue:44661", + "issue:44683", + "issue:44737", + "issue:44811", + "issue:44987", + "issue:44991", + "issue:45084", + "issue:45507", + "issue:45584", + "issue:45859", + "issue:46178", + "issue:46285", + "issue:46326", + "issue:46402", + "issue:46489" + ], + "soft_pairs": [ + "issue:44462|issue:46489", + "issue:42796|issue:43404", + "issue:46285|issue:46402", + "issue:36032|issue:42915", + "issue:43716|issue:45084", + "issue:32090|issue:39404", + "issue:44737|issue:45859", + "issue:44661|issue:44683", + "issue:43643|issue:44530", + "issue:44451|issue:44987", + "issue:40990|issue:43421", + "issue:42915|issue:44530", + "issue:43541|issue:45084", + "issue:43644|issue:43646", + "issue:43582|issue:43632", + "issue:42175|issue:46178", + "issue:34567|issue:40444", + "issue:43883|issue:44488", + "issue:40444|issue:46326", + "issue:44206|issue:44220", + "issue:44811|issue:45584", + "issue:36032|issue:43645", + "issue:43452|issue:43482", + "issue:44987|issue:44991", + "issue:43317|issue:43482", + "issue:45084|issue:45507", + "issue:43262|issue:43377", + "issue:43746|issue:46326", + "issue:43883|issue:44991", + "issue:43646|issue:44534" + ], + "prepared_review_unit_hash": "e7ad9176e7c6e9cc6a6295fda029b3ad1c4fdd435c78ae6fc9da670be8517064", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36691, + "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 heterogeneous: most items are unrelated Transformers regressions or model-specific loading/tokenizer bugs, with only superficial overlap around v5 and initialization/loading behavior. I would not treat these as duplicates; 43646 is only the broadest representative issue, not a true canonical duplicate target.", + "confidence": 0.31, + "canonical_issue_reason": "Issue 43646 is the broadest, most generic regression report in the set ('custom model initialization' under Transformers 5.0.0), so it is the least specific representative even though the cluster is not a true duplicate group.", + "canonical_pr_reason": null, + "best_issue_reason": "43646 is the best single representative because it describes a broad Transformers v5 regression affecting initialization, which is more general than the other model- or tokenizer-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44462", + "right": "issue:46489", + "accept": false, + "reason": "Both concern tokenizer loading, but one is about repo tokenizer.json being ignored and the other is a DeepSeek-Coder v1 tokenizer regression tied to a specific v5+ fix gap; not the same bug." + }, + { + "left": "issue:42796", + "right": "issue:43404", + "accept": false, + "reason": "Tokenizer BPE-marker handling and a tied lm_head weight bug in Mistral3 are different code paths and failure modes." + }, + { + "left": "issue:46285", + "right": "issue:46402", + "accept": false, + "reason": "Different models and different regressions: CLIP text model structure mismatch vs Sam3 text-encoder weights not loading." + }, + { + "left": "issue:36032", + "right": "issue:42915", + "accept": false, + "reason": "T5 tokenizer attribute conflict and Qwen3MoE FP8 config failure are unrelated tokenizer vs quantization issues." + }, + { + "left": "issue:43716", + "right": "issue:45084", + "accept": false, + "reason": "Image preprocessor dtype mismatch is not the same as a template compilation error." + }, + { + "left": "issue:32090", + "right": "issue:39404", + "accept": false, + "reason": "Trainer GPU broadcast TypeError and Whisper pipeline return_language breakage are unrelated subsystems and bugs." + }, + { + "left": "issue:44737", + "right": "issue:45859", + "accept": false, + "reason": "Both involve rotary embeddings / offload concerns, but one is CPU device placement and the other is a Qwen3.5MoE offload compatibility issue." + }, + { + "left": "issue:44661", + "right": "issue:44683", + "accept": false, + "reason": "Tokenizer mapping lookup failure and compiled flex_attention on torch>=2.9 are different failures." + }, + { + "left": "issue:43643", + "right": "issue:44530", + "accept": false, + "reason": "trust_remote_code config field loss and PagedAttentionCache linear_attention crash are separate issues." + }, + { + "left": "issue:44451", + "right": "issue:44987", + "accept": false, + "reason": "Both are loading failures for different models, but the causes are different and model-specific." + }, + { + "left": "issue:40990", + "right": "issue:43421", + "accept": false, + "reason": "High perplexity on gpt-oss and runtime post-processor updates for special tokens are not the same bug." + }, + { + "left": "issue:42915", + "right": "issue:44530", + "accept": false, + "reason": "Different Qwen model families and different problems: FP8 config vs cache/group type crash." + }, + { + "left": "issue:43541", + "right": "issue:45084", + "accept": false, + "reason": "Torch dynamo tracing on Mixtral float32 grouped_mm is unrelated to template-node compilation errors." + }, + { + "left": "issue:43644", + "right": "issue:43646", + "accept": false, + "reason": "Both are Transformers v5 regressions, but one is junk-filled non-persistent buffers and the other is custom model initialization failure; related theme, not the same underlying bug." + }, + { + "left": "issue:43582", + "right": "issue:43632", + "accept": false, + "reason": "AppleSilicon warmup TypeError and _is_hf_initialized flag breakage are distinct initialization/runtime issues." + }, + { + "left": "issue:42175", + "right": "issue:46178", + "accept": false, + "reason": "Missing tensorflow backend in '.[torch]' install and undefined tensorflow_text backend are both backend-mapping problems, but not the same missing dependency or code path." + }, + { + "left": "issue:34567", + "right": "issue:40444", + "accept": false, + "reason": "TrainerState token counting and Qwen2.5-VL iterable dataset multi-image failure are unrelated." + }, + { + "left": "issue:43883", + "right": "issue:44488", + "accept": false, + "reason": "Both mention model loading failures, but one is a missing tied-weights attribute and the other is a tokenizer-loading regression for a specific model." + }, + { + "left": "issue:40444", + "right": "issue:46326", + "accept": false, + "reason": "Multi-image IterableDataset finetuning failure and mistral_common continue_final_message behavior are different multimodal/tokenizer bugs." + }, + { + "left": "issue:44206", + "right": "issue:44220", + "accept": false, + "reason": "LasrFeatureExtractor unsupported center arg and _torch_extract_fbank_features are closely related audio preprocessing issues, but the reports point to different API breakages and likely different fixes." + }, + { + "left": "issue:44811", + "right": "issue:45584", + "accept": false, + "reason": "Both involve Whisper token handling, but batch_decode skip_special_tokens and empty-transcription generation after align_special_tokens are different behaviors." + }, + { + "left": "issue:36032", + "right": "issue:43645", + "accept": false, + "reason": "T5 tokenizer method conflict and Jupyter custom model initialization are not the same bug." + }, + { + "left": "issue:43452", + "right": "issue:43482", + "accept": false, + "reason": "gguf_file breaking from_pretrained and Qwen2.5-GGUF loading failure are both GGUF-related but not the same concrete failure mode." + }, + { + "left": "issue:44987", + "right": "issue:44991", + "accept": false, + "reason": "Both are loading regressions in transformers>=5, but they affect different models and likely different tokenizer/config paths." + }, + { + "left": "issue:43317", + "right": "issue:43482", + "accept": false, + "reason": "GPU+CPU offload dequantized load failure and Qwen2.5-GGUF v5 loading failure are separate loader/offload bugs." + }, + { + "left": "issue:45084", + "right": "issue:45507", + "accept": false, + "reason": "Template compilation error and GraniteMoEHybrid invalid method calls are unrelated." + }, + { + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Audio processor sampling-rate defaulting and MIMI padding-mask batch inconsistency are different processor/encoder bugs." + }, + { + "left": "issue:43746", + "right": "issue:46326", + "accept": false, + "reason": "PEFT adapter local checkpoint loading and mistral_common continue_final_message behavior do not share the same underlying issue." + }, + { + "left": "issue:43883", + "right": "issue:44991", + "accept": false, + "reason": "Missing all_tied_weights_keys and tokenizer loading regression for EMBEDDIA/est-roberta are different failures." + }, + { + "left": "issue:43646", + "right": "issue:44534", + "accept": false, + "reason": "These are very similar on the surface, but 44534 is the generic 'non-persistent buffers junk' regression, while 43646 is specifically custom model initialization; related family, not a duplicate." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 11, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:28530", + "issue:34567", + "issue:36032", + "issue:41720", + "issue:42915", + "issue:43089", + "issue:43296", + "issue:43322", + "issue:43377", + "issue:43404", + "issue:43425", + "issue:43531", + "issue:43540", + "issue:43632", + "issue:43645", + "issue:43716", + "issue:43742", + "issue:43792", + "issue:43819", + "issue:43994", + "issue:44008", + "issue:44062", + "issue:44220", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44451", + "issue:44464", + "issue:44466", + "issue:44479", + "issue:44534", + "issue:44589", + "issue:44610", + "issue:44683", + "issue:44849", + "issue:44877", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45216", + "issue:45325", + "issue:45357", + "issue:45987", + "issue:46315", + "issue:46821", + "issue:46869", + "issue:46899" + ], + "soft_pairs": [ + "issue:45084|issue:46869", + "issue:43792|issue:46869", + "issue:44403|issue:45081", + "issue:43645|issue:44534", + "issue:44849|issue:45325", + "issue:43377|issue:43819", + "issue:43404|issue:44466", + "issue:44464|issue:45987", + "issue:44987|issue:45081", + "issue:44008|issue:46899", + "issue:43994|issue:44610", + "issue:44451|issue:45081", + "issue:42915|issue:44291", + "issue:28530|issue:34567", + "issue:36032|issue:43742", + "issue:44062|issue:44987", + "issue:44265|issue:44610", + "issue:42915|issue:44589", + "issue:43089|issue:45987", + "issue:43322|issue:43540", + "issue:44479|issue:45325", + "issue:43632|issue:46821", + "issue:36032|issue:41720", + "issue:45216|issue:45357", + "issue:43531|issue:44877", + "issue:44991|issue:46315", + "issue:43425|issue:44403", + "issue:43296|issue:43531", + "issue:44220|issue:44683", + "issue:43716|issue:46899" + ], + "prepared_review_unit_hash": "b32089a9bd297c139f9979ca2ea916a80ee73c6009dfc57af8bdf95fa2e99e2e", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36729, + "estimated_input_tokens": 9183, + "estimated_eval_tokens": 18622 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a mix of unrelated bugs across loading, generation, multimodal processing, and trainer internals. Only one soft edge looks like a true duplicate/follow-up pair: the Qwen3.5 save_pretrained regression in 45216/45357.", + "confidence": 0.89, + "canonical_issue_reason": "Issue 45216 is the best canonical anchor: it reports the broader Qwen3.5 save_pretrained checkpoint regression, and 45357 reads like a later narrower follow-up about the same incorrect visual-encoder key serialization. The rest of the cluster is unrelated noise.", + "canonical_pr_reason": null, + "best_issue_reason": "45216 is the strongest representative issue because it is the earlier, broader report in the only true duplicate pair and gives the clearest problem statement.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45084", + "right": "issue:46869", + "accept": false, + "reason": "Different failures: template-node compilation vs YOLO multi-GPU device-mapping cuda assert." + }, + { + "left": "issue:43792", + "right": "issue:46869", + "accept": false, + "reason": "Unrelated model/runtime problems: Whisper inference failure vs YOLO auto device map failure." + }, + { + "left": "issue:44403", + "right": "issue:45081", + "accept": false, + "reason": "One is a logging/noise complaint; the other is a concrete tokenizer regex crash in Mistral loading." + }, + { + "left": "issue:43645", + "right": "issue:44534", + "accept": false, + "reason": "Both are v5 regressions, but they hit different code paths: notebook custom-model init vs non-persistent buffer initialization." + }, + { + "left": "issue:44849", + "right": "issue:45325", + "accept": false, + "reason": "Different Qwen model bugs: hidden-states output handling vs rope index scaling for still images/video positions." + }, + { + "left": "issue:43377", + "right": "issue:43819", + "accept": false, + "reason": "Both involve output mismatches, but they concern different models and mechanisms (padding-mask support vs missing STE)." + }, + { + "left": "issue:43404", + "right": "issue:44466", + "accept": false, + "reason": "Related theme of lm_head/tied weights, but one is a specific Mistral3 tying bug and the other is device-dependent serialization behavior." + }, + { + "left": "issue:44464", + "right": "issue:45987", + "accept": false, + "reason": "Chunked generation inconsistency is not the same bug as StaticCache.get_seq_length returning the wrong type." + }, + { + "left": "issue:44987", + "right": "issue:45081", + "accept": false, + "reason": "Both are load failures, but one is a tokenizer loading regression for a specific model and the other is a Mistral regex patch crash." + }, + { + "left": "issue:44008", + "right": "issue:46899", + "accept": false, + "reason": "Both are multimodal dtype-related, but they affect different model families and have different root causes." + }, + { + "left": "issue:43994", + "right": "issue:44610", + "accept": false, + "reason": "Different vision issues: nonsensical SigLIP2 outputs vs OmDet-Turbo processor/model input-size mismatch." + }, + { + "left": "issue:44451", + "right": "issue:45081", + "accept": false, + "reason": "Both are tokenizer/model loading failures, but the models and underlying causes differ." + }, + { + "left": "issue:42915", + "right": "issue:44291", + "accept": false, + "reason": "Different initialization failures: FineGrainedFP8Config vs unexpected _is_hf_initialized argument in init_empty_weights." + }, + { + "left": "issue:28530", + "right": "issue:34567", + "accept": false, + "reason": "Trainer early-stopping metric selection and num_input_tokens_seen updating are unrelated trainer issues." + }, + { + "left": "issue:36032", + "right": "issue:43742", + "accept": false, + "reason": "Different loading errors: T5Tokenizer add_special_tokens conflict vs MobileLLM key error." + }, + { + "left": "issue:44062", + "right": "issue:44987", + "accept": false, + "reason": "AddedToken special-argument duplication is a tokenizer constructor bug, not the same as a model-specific tokenizer load failure." + }, + { + "left": "issue:44265", + "right": "issue:44610", + "accept": false, + "reason": "torch.export/torch_compilable_check failure is unrelated to OmDet-Turbo processor image sizing." + }, + { + "left": "issue:42915", + "right": "issue:44589", + "accept": false, + "reason": "FineGrainedFP8Config failure is unrelated to missing Float8 storage type during serialization/loading." + }, + { + "left": "issue:43089", + "right": "issue:45987", + "accept": false, + "reason": "Generation performance overhead is not the same as a cache API contract/type bug." + }, + { + "left": "issue:43322", + "right": "issue:43540", + "accept": false, + "reason": "Different multimodal problems on different models: Llava Next segfault vs Qwen3OmniMoe video-input ValueError." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": false, + "reason": "Related Qwen VL/video area, but 44479 is a broad video regression while 45325 is a specific still-image temporal-position scaling bug." + }, + { + "left": "issue:43632", + "right": "issue:46821", + "accept": false, + "reason": "Transformers v5 init-flag breakage is unrelated to the TimesFM window_size AttributeError." + }, + { + "left": "issue:36032", + "right": "issue:41720", + "accept": false, + "reason": "Tokenizer-load conflict and Qwen3 auto device-map cuda assert are unrelated." + }, + { + "left": "issue:45216", + "right": "issue:45357", + "accept": true, + "reason": "Same underlying Qwen3.5 save_pretrained regression: both report incorrect checkpoint/visual-encoder key serialization, with 45357 reading like a narrower follow-up." + }, + { + "left": "issue:43531", + "right": "issue:44877", + "accept": false, + "reason": "Qwen3-MoE sliding_window handling and granite_speech strict-config loading are unrelated." + }, + { + "left": "issue:44991", + "right": "issue:46315", + "accept": false, + "reason": "Both involve slow tokenizers, but one is a load failure and the other is a performance regression in convert_tokens_to_ids." + }, + { + "left": "issue:43425", + "right": "issue:44403", + "accept": false, + "reason": "Torch version incompatibility is unrelated to the 'unnecessary noise' loading complaint." + }, + { + "left": "issue:43296", + "right": "issue:43531", + "accept": false, + "reason": "PaddleOCR-VL loading failure and Qwen3-MoE sliding_window behavior are unrelated." + }, + { + "left": "issue:44220", + "right": "issue:44683", + "accept": false, + "reason": "Audio fbank extraction and compiled flex_attention are different subsystems and failure modes." + }, + { + "left": "issue:43716", + "right": "issue:46899", + "accept": false, + "reason": "Both mention dtype issues in multimodal paths, but they are different models and different bugs." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet and the one accepted soft edge is supported by the issue titles: 45216 and 45357 both concern the Qwen3.5 save_pretrained checkpoint/visual-encoder key serialization regression. The other soft-edge rejections are phrased conservatively and do not overclaim duplicate status from similarity alone." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 12, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:33666", + "issue:35707", + "issue:40444", + "issue:40990", + "issue:41553", + "issue:41720", + "issue:42175", + "issue:42796", + "issue:43296", + "issue:43317", + "issue:43377", + "issue:43381", + "issue:43404", + "issue:43421", + "issue:43482", + "issue:43525", + "issue:43531", + "issue:43582", + "issue:43611", + "issue:43632", + "issue:43646", + "issue:43746", + "issue:43792", + "issue:43856", + "issue:44079", + "issue:44291", + "issue:44403", + "issue:44462", + "issue:44509", + "issue:44534", + "issue:44568", + "issue:44589", + "issue:44756", + "issue:44811", + "issue:44991", + "issue:45081", + "issue:45325", + "issue:45491", + "issue:45538", + "issue:45542", + "issue:45663", + "issue:45812", + "issue:45923", + "issue:46491", + "issue:46710", + "issue:46821", + "issue:46899", + "issue:46909" + ], + "soft_pairs": [ + "issue:43611|issue:44534", + "issue:40444|issue:45325", + "issue:45812|issue:46491", + "issue:44403|issue:44991", + "issue:43381|issue:45923", + "issue:43632|issue:44756", + "issue:42796|issue:43746", + "issue:43792|issue:44589", + "issue:43404|issue:46899", + "issue:33666|issue:40444", + "issue:40990|issue:44811", + "issue:44291|issue:44589", + "issue:42796|issue:46710", + "issue:43482|issue:43525", + "issue:35707|issue:40444", + "issue:44079|issue:45663", + "issue:41553|issue:43582", + "issue:43482|issue:45538", + "issue:43632|issue:44534", + "issue:43421|issue:44568", + "issue:43531|issue:44291", + "issue:42175|issue:45542", + "issue:41720|issue:45081", + "issue:44462|issue:45812", + "issue:43317|issue:43856", + "issue:44079|issue:44534", + "issue:44509|issue:46909", + "issue:43296|issue:46821", + "issue:43377|issue:45491", + "issue:43611|issue:43646" + ], + "prepared_review_unit_hash": "3dc730cc5a3f9c96fedc0bcdb8fdd7a026852b27bca4a5b251ccd83475a87e69", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36775, + "estimated_input_tokens": 9194, + "estimated_eval_tokens": 18644 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab-bag of unrelated bugs. The only loose common thread is Transformers v5 loading/init/tokenizer regressions, but the soft pairs still split across different concrete code paths and model families, so none look like true duplicates.", + "confidence": 0.47, + "canonical_issue_reason": "issue:43611 is the broadest, most central v5 regression here: it targets model loading behavior itself, which is the closest umbrella for several other init/load-related reports in the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43611 is the best single representative because it captures the broad model-loading regression theme without being as narrow as the model-specific or symptom-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43611", + "right": "issue:44534", + "accept": false, + "reason": "Both are v5 regressions, but one is model loading/base_model_prefix and the other is non-persistent buffer initialization; different code paths." + }, + { + "left": "issue:40444", + "right": "issue:45325", + "accept": false, + "reason": "Both are Qwen2.5-VL, but one fails finetuning with multiple images per prompt and the other mis-scales still-image rope positions." + }, + { + "left": "issue:45812", + "right": "issue:46491", + "accept": false, + "reason": "Tokenizer regressions, but different models and symptoms: Granite wrong token IDs vs CodeLlama leading-space round-trip." + }, + { + "left": "issue:44403", + "right": "issue:44991", + "accept": false, + "reason": "One is noisy load logging; the other is a tokenizer loading failure for EMBEDDIA/est-roberta." + }, + { + "left": "issue:43381", + "right": "issue:45923", + "accept": false, + "reason": "Both mention gradient checkpointing, but one forbids eval-mode use and the other is a missing support flag on a trust_remote_code variant." + }, + { + "left": "issue:43632", + "right": "issue:44756", + "accept": false, + "reason": "One is the `_is_hf_initialized` v5 init regression; the other is an mmap OOM workaround on Strix Halo." + }, + { + "left": "issue:42796", + "right": "issue:43746", + "accept": false, + "reason": "Tokenizer BPE marker handling vs PEFT local-checkpoint loading; unrelated concrete bugs." + }, + { + "left": "issue:43792", + "right": "issue:44589", + "accept": false, + "reason": "Whisper runtime failure vs missing Float8 storage type; unrelated." + }, + { + "left": "issue:43404", + "right": "issue:46899", + "accept": false, + "reason": "Mistral3 lm_head weight tying bug vs Gemma4 4-bit quantization dtype cast issue." + }, + { + "left": "issue:33666", + "right": "issue:40444", + "accept": false, + "reason": "Both involve Qwen VL training, but multi-GPU training and multi-image iterable-dataset failures are different bugs." + }, + { + "left": "issue:40990", + "right": "issue:44811", + "accept": false, + "reason": "High perplexity on gpt-oss vs Whisper batch_decode skip_special_tokens behavior." + }, + { + "left": "issue:44291", + "right": "issue:44589", + "accept": false, + "reason": "`init_empty_weights`/_is_hf_initialized arg issue vs Float8 storage lookup error." + }, + { + "left": "issue:42796", + "right": "issue:46710", + "accept": false, + "reason": "Tokenizer BPE-marker bug vs DeepSeek accuracy regression after upgrade." + }, + { + "left": "issue:43482", + "right": "issue:43525", + "accept": false, + "reason": "GGUF loading failure vs missing `pad_token_id` on Llama4Config." + }, + { + "left": "issue:35707", + "right": "issue:40444", + "accept": false, + "reason": "Progressive generation with inputs_embeds/past_key_values vs multi-image iterable-dataset finetuning." + }, + { + "left": "issue:44079", + "right": "issue:45663", + "accept": false, + "reason": "ModelOutput key assignment bug vs Gemma4 FSDP2 `shared_kv_states` KeyError." + }, + { + "left": "issue:41553", + "right": "issue:43582", + "accept": false, + "reason": "Bad AutoTokenizer error message for Voxtral vs Apple Silicon caching_allocator_warmup TypeError." + }, + { + "left": "issue:43482", + "right": "issue:45538", + "accept": false, + "reason": "GGUF/v5 loading failure vs CLIPTokenizer max_length bug." + }, + { + "left": "issue:43632", + "right": "issue:44534", + "accept": false, + "reason": "`_is_hf_initialized` flag regression vs non-persistent buffer junk fill." + }, + { + "left": "issue:43421", + "right": "issue:44568", + "accept": false, + "reason": "Runtime post-processor updates for special tokens vs mdeberta BOS/EOS not added." + }, + { + "left": "issue:43531", + "right": "issue:44291", + "accept": false, + "reason": "Qwen3-MoE sliding-window bug vs `init_empty_weights` constructor-arg regression." + }, + { + "left": "issue:42175", + "right": "issue:45542", + "accept": false, + "reason": "Both mention TensorFlow packaging, but one is missing tensorflow in `.[torch]` and the other is tensorboard-only backend detection; not the same failure." + }, + { + "left": "issue:41720", + "right": "issue:45081", + "accept": false, + "reason": "Qwen3 auto device mapping cudaErrorAssert vs Mistral regex patch crash while loading tokenizer." + }, + { + "left": "issue:44462", + "right": "issue:45812", + "accept": false, + "reason": "AutoTokenizer ignoring tokenizer.json vs Granite token-ID regression." + }, + { + "left": "issue:43317", + "right": "issue:43856", + "accept": false, + "reason": "Dequantized model offload loading failure vs Qwen3 MoE training memory inefficiency." + }, + { + "left": "issue:44079", + "right": "issue:44534", + "accept": false, + "reason": "ModelOutput key assignment vs non-persistent buffer junk fill." + }, + { + "left": "issue:44509", + "right": "issue:46909", + "accept": false, + "reason": "Docs about removed pipeline tasks vs a broken DIA documentation example." + }, + { + "left": "issue:43296", + "right": "issue:46821", + "accept": false, + "reason": "PaddleOCR-VL load failure in vLLM vs TimesFM `window_size` AttributeError." + }, + { + "left": "issue:43377", + "right": "issue:45491", + "accept": false, + "reason": "MIMI padding-mask batch discrepancy vs Gemma3 NaNs with sliding-window mixed lengths." + }, + { + "left": "issue:43611", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 loading/init regressions, but one is `base_model_prefix` loading and the other is custom model initialization; related theme, different bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 13, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:34634", + "issue:35707", + "issue:39692", + "issue:40444", + "issue:40990", + "issue:41720", + "issue:42175", + "issue:42796", + "issue:42915", + "issue:43262", + "issue:43296", + "issue:43377", + "issue:43454", + "issue:43482", + "issue:43493", + "issue:43550", + "issue:43606", + "issue:43644", + "issue:43701", + "issue:43746", + "issue:43818", + "issue:43819", + "issue:43828", + "issue:43873", + "issue:43881", + "issue:43994", + "issue:44079", + "issue:44186", + "issue:44265", + "issue:44291", + "issue:44466", + "issue:44479", + "issue:44488", + "issue:44610", + "issue:44829", + "issue:44877", + "issue:45072", + "issue:45325", + "issue:45357", + "issue:45381", + "issue:45479", + "issue:45584", + "issue:45701", + "issue:45823", + "issue:45834", + "issue:46402", + "issue:46821" + ], + "soft_pairs": [ + "issue:43818|issue:46402", + "issue:44829|issue:45479", + "issue:43644|issue:44079", + "issue:42175|issue:43296", + "issue:43818|issue:45357", + "issue:44265|issue:45072", + "issue:43493|issue:43994", + "issue:40444|issue:45381", + "issue:45325|issue:45381", + "issue:45325|issue:45834", + "issue:43454|issue:44466", + "issue:42915|issue:43482", + "issue:44291|issue:44877", + "issue:43701|issue:45325", + "issue:43873|issue:45701", + "issue:42796|issue:45381", + "issue:43606|issue:44265", + "issue:40990|issue:45584", + "issue:43262|issue:43746", + "issue:44479|issue:45823", + "issue:43881|issue:44488", + "issue:43828|issue:46821", + "issue:35707|issue:41720", + "issue:34634|issue:35707", + "issue:39692|issue:40444", + "issue:43454|issue:43550", + "issue:43606|issue:43819", + "issue:44186|issue:44265", + "issue:41720|issue:45823", + "issue:43377|issue:44610" + ], + "prepared_review_unit_hash": "bc0e84454553d1a498491c1ff9724f9e3df9d80d641825e75e8add6efa59b113", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36278, + "estimated_input_tokens": 9070, + "estimated_eval_tokens": 18396 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The items are heterogeneous and do not form a single duplicate cluster. All listed soft pairs look like false positives: they share broad model/subsystem themes but not the same concrete bug or change path.", + "confidence": 0.92, + "canonical_issue_reason": "No clear canonical issue: the set spans many unrelated bugs (model loading, tokenization, quantization, device mapping, generation, and export), so selecting one as the duplicate anchor would be misleading.", + "canonical_pr_reason": null, + "best_issue_reason": "No single issue is globally best as a representative duplicate target; the soft pairs are mostly broad-topic similarities rather than the same underlying defect.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43818", + "right": "issue:46402", + "accept": false, + "reason": "Different bugs in different models: one is a Video-LLaVA temporal-attention/video-tower issue, the other is Sam3Model text-encoder weight loading." + }, + { + "left": "issue:44829", + "right": "issue:45479", + "accept": false, + "reason": "Both mention bad training/loss behavior, but the causes differ: flash_attention_3 regression vs. single-label classification with num_labels=1." + }, + { + "left": "issue:43644", + "right": "issue:44079", + "accept": false, + "reason": "Unrelated code paths: non-persistent buffer junk initialization vs. ModelOutput key assignment when a key was previously None." + }, + { + "left": "issue:42175", + "right": "issue:43296", + "accept": false, + "reason": "Package/backend installation problem vs. a vLLM model-load failure; no shared concrete bug." + }, + { + "left": "issue:43818", + "right": "issue:45357", + "accept": false, + "reason": "One is a model architecture/loading defect, the other is save_pretrained writing incorrect visual encoder keys." + }, + { + "left": "issue:44265", + "right": "issue:45072", + "accept": false, + "reason": "torch.export/checking failure and bf16 dtype mismatch are different failure modes, even though both involve transformers internals." + }, + { + "left": "issue:43493", + "right": "issue:43994", + "accept": false, + "reason": "Same model family, but one is HF-vs-JAX implementation divergence while the other is bad AutoModel/pipeline output; not clearly the same bug." + }, + { + "left": "issue:40444", + "right": "issue:45381", + "accept": false, + "reason": "Different Qwen2.5-VL scenarios: finetuning on multi-image prompts vs. video-input position-id regression." + }, + { + "left": "issue:45325", + "right": "issue:45381", + "accept": false, + "reason": "Both involve Qwen2.5-VL position handling, but one is still-image temporal scaling and the other is video vision_position_ids; too different to treat as duplicates." + }, + { + "left": "issue:45325", + "right": "issue:45834", + "accept": false, + "reason": "One is a rope-index scaling bug, the other is a long-OCR index error; shared model family is not enough." + }, + { + "left": "issue:43454", + "right": "issue:44466", + "accept": false, + "reason": "Both mention lm_head/tied weights, but one is failure to tie weights causing bad generation and the other is device-dependent serialization inconsistency." + }, + { + "left": "issue:42915", + "right": "issue:43482", + "accept": false, + "reason": "FP8 configuration failure vs. GGUF loading failure in v5 are unrelated." + }, + { + "left": "issue:44291", + "right": "issue:44877", + "accept": false, + "reason": "init_empty_weights argument mismatch is a different problem from strict config rejection of granite_speech." + }, + { + "left": "issue:43701", + "right": "issue:45325", + "accept": false, + "reason": "Checkpoint key mismatch and Qwen2.5-VL position-id scaling are unrelated bugs." + }, + { + "left": "issue:43873", + "right": "issue:45701", + "accept": false, + "reason": "Quantization/offload behavior and tokenization version changes are distinct issues." + }, + { + "left": "issue:42796", + "right": "issue:45381", + "accept": false, + "reason": "Tokenizer BPE-marker handling and Qwen2.5-VL video position IDs are different defects." + }, + { + "left": "issue:43606", + "right": "issue:44265", + "accept": false, + "reason": "CPU-offload device mismatch is not the same as export failing on torch_compilable_check." + }, + { + "left": "issue:40990", + "right": "issue:45584", + "accept": false, + "reason": "High perplexity on WikiText-2 and Whisper empty-transcription failure are unrelated symptoms and code paths." + }, + { + "left": "issue:43262", + "right": "issue:43746", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting and GraniteSpeech PEFT checkpoint loading are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:45823", + "accept": false, + "reason": "Both are inference regressions, but for different models and different root causes: Qwen video regression vs. Gemma4 PLE device mismatch." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "Different models and different load failures; not a duplicate bug." + }, + { + "left": "issue:43828", + "right": "issue:46821", + "accept": false, + "reason": "Autocast dtype mismatch and TimesFM window_size AttributeError are unrelated." + }, + { + "left": "issue:35707", + "right": "issue:41720", + "accept": false, + "reason": "Progressive generation with inputs_embeds/past_key_values is a different code path from auto device mapping causing cudaErrorAssert." + }, + { + "left": "issue:34634", + "right": "issue:35707", + "accept": false, + "reason": "BarkProcessor voice_preset and progressive generation with embeddings are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:40444", + "accept": false, + "reason": "SigLIP2 doc-example problems and Qwen2.5-VL multi-image finetuning failures do not share the same concrete bug." + }, + { + "left": "issue:43454", + "right": "issue:43550", + "accept": false, + "reason": "Tied-weight serialization/generation bug vs. Bamba-9B-v2 torch.compile+SDPA failure; different root causes." + }, + { + "left": "issue:43606", + "right": "issue:43819", + "accept": false, + "reason": "CPU offload device mismatch and DAC forward/from_latents mismatch are unrelated." + }, + { + "left": "issue:44186", + "right": "issue:44265", + "accept": false, + "reason": "Tokenizer crashes on NER/padding are not the same as torch.export failures." + }, + { + "left": "issue:41720", + "right": "issue:45823", + "accept": false, + "reason": "Different models and different device-related failures; no shared concrete bug." + }, + { + "left": "issue:43377", + "right": "issue:44610", + "accept": false, + "reason": "MIMI padding-mask output mismatch and OmDet processor/model input-size mismatch are distinct 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": 23, + "review_unit_index": 14, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:26413", + "issue:28530", + "issue:36331", + "issue:40990", + "issue:41720", + "issue:42617", + "issue:42796", + "issue:42915", + "issue:42947", + "issue:43065", + "issue:43257", + "issue:43317", + "issue:43322", + "issue:43377", + "issue:43450", + "issue:43454", + "issue:43526", + "issue:43550", + "issue:43572", + "issue:43720", + "issue:43749", + "issue:43873", + "issue:43881", + "issue:43883", + "issue:44186", + "issue:44387", + "issue:44466", + "issue:44534", + "issue:44568", + "issue:44756", + "issue:44877", + "issue:44898", + "issue:44977", + "issue:44991", + "issue:45072", + "issue:45357", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45823", + "issue:45920", + "issue:45923", + "issue:46050", + "issue:46209", + "issue:46305", + "issue:46315", + "issue:46402", + "issue:46821" + ], + "soft_pairs": [ + "issue:46305|issue:46315", + "issue:43257|issue:45923", + "issue:44568|issue:45920", + "issue:42796|issue:46209", + "issue:43454|issue:45072", + "issue:45823|issue:46821", + "issue:43377|issue:44898", + "issue:43454|issue:43720", + "issue:40990|issue:42947", + "issue:42617|issue:42915", + "issue:43749|issue:44534", + "issue:43317|issue:43873", + "issue:43572|issue:45759", + "issue:43526|issue:43883", + "issue:43322|issue:46050", + "issue:41720|issue:43322", + "issue:43881|issue:44991", + "issue:36331|issue:42617", + "issue:44387|issue:45704", + "issue:41720|issue:44977", + "issue:36331|issue:42915", + "issue:44466|issue:45357", + "issue:43450|issue:45701", + "issue:44877|issue:44991", + "issue:43065|issue:46402", + "issue:40990|issue:45701", + "issue:44186|issue:45072", + "issue:26413|issue:28530", + "issue:41720|issue:42617", + "issue:43550|issue:43720", + "issue:43873|issue:44756" + ], + "prepared_review_unit_hash": "d4887696d7b7ea2d8a65df39955eb1a8da7b1ca3f072ecbc7715ba4c736e86ee", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36975, + "estimated_input_tokens": 9244, + "estimated_eval_tokens": 18744 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: several unrelated bug reports were surfaced by soft text similarity, but none of the candidate pairs look like the same underlying issue. The closest themes are tokenizer regressions and device_map/offloading/quantization problems, yet each pair differs in concrete failure mode or affected path.", + "confidence": 0.78, + "canonical_issue_reason": "issue:43873 is the broadest and most central bug in the set: an open, highly discussed offloading/quantization regression with multiple nearby similarity hits, making it the best representative issue for this cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 has the strongest cluster-level relevance and the widest problem scope among the issues shown, so it is the best single issue to anchor triage even though it is not a duplicate of the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46305", + "right": "issue:46315", + "accept": false, + "reason": "Both are performance bugs, but one is a MoonshineStreaming masking off-by-one and the other is slow `convert_tokens_to_ids`; different code paths and symptoms." + }, + { + "left": "issue:43257", + "right": "issue:45923", + "accept": false, + "reason": "Qwen3 MoE weight conversion under accelerate+deepspeed is unrelated to a missing `supports_gradient_checkpointing` flag on a trust_remote_code variant." + }, + { + "left": "issue:44568", + "right": "issue:45920", + "accept": false, + "reason": "`add_special_tokens` not adding BOS/EOS for mdeberta-v3 is a tokenizer-special-case bug, while the other is wrong token IDs across several model families after a v5 regression." + }, + { + "left": "issue:42796", + "right": "issue:46209", + "accept": false, + "reason": "Tokenizer BPE marker handling and FP8 inference on B200 are unrelated failures in different subsystems." + }, + { + "left": "issue:43454", + "right": "issue:45072", + "accept": false, + "reason": "Incorrect batched video processor shape and dtype mismatches in bfloat16 inference are different bugs with different affected components." + }, + { + "left": "issue:45823", + "right": "issue:46821", + "accept": false, + "reason": "Gemma4 PLE device mismatch during forward is not the same as TimesFM `window_size` raising `AttributeError`; different models and error classes." + }, + { + "left": "issue:43377", + "right": "issue:44898", + "accept": false, + "reason": "MIMI encoder batching/padding-mask behavior is unrelated to Perceiver image classification failing despite `interpolate_pos_encoding=True`." + }, + { + "left": "issue:43454", + "right": "issue:43720", + "accept": false, + "reason": "Video processor batching shape issues do not match BitNet packed-weights unpacking during accelerate loading." + }, + { + "left": "issue:40990", + "right": "issue:42947", + "accept": false, + "reason": "High perplexity on gpt-oss-20b is a model quality/regression report, while the other is a gradient checkpointing effectiveness issue with PEFT LoRA." + }, + { + "left": "issue:42617", + "right": "issue:42915", + "accept": false, + "reason": "Failure to run `3d_parallel.py` is unrelated to Qwen3MoE failing with FineGrainedFP8Config." + }, + { + "left": "issue:43749", + "right": "issue:44534", + "accept": false, + "reason": "Broken FSDP CPU RAM efficient loading and non-persistent buffers filled with junk are different regression classes." + }, + { + "left": "issue:43317", + "right": "issue:43873", + "accept": false, + "reason": "Both involve loading/offloading, but one is a dequantized model load failure with gpu+cpu offload and the other is a broader quantization offloading issue; not the same concrete bug." + }, + { + "left": "issue:43572", + "right": "issue:45759", + "accept": false, + "reason": "Missing `pad_token_idx` in StableLmConfig is a config regression, while the other is `from_config` not unwrapping `text_config` for composite Qwen multimodal configs." + }, + { + "left": "issue:43526", + "right": "issue:43883", + "accept": false, + "reason": "`reduce_labels` returning one label in BeitImageProcessorFast is unrelated to a missing `all_tied_weights_keys` attribute in MolmoForCausalLM." + }, + { + "left": "issue:43322", + "right": "issue:46050", + "accept": false, + "reason": "Segmentation fault loading Llava Next models and broken sinq quantized save/load behavior are not the same code-path problem." + }, + { + "left": "issue:41720", + "right": "issue:43322", + "accept": false, + "reason": "Qwen3 auto device mapping cudaErrorAssert and Llava Next segmentation fault are both load-time crashes, but they affect different models and failure mechanisms." + }, + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "glm-4v-9b loading failure and tokenizer loading failure for EMBEDDIA/est-roberta are unrelated model-specific regressions." + }, + { + "left": "issue:36331", + "right": "issue:42617", + "accept": false, + "reason": "Unexpected `num_items_in_batch` in `compute_loss` is a trainer API compatibility issue, not the same as failing to run `3d_parallel.py`." + }, + { + "left": "issue:44387", + "right": "issue:45704", + "accept": false, + "reason": "Int4 quantization OOM in Transformers 5.x and T5's apex FusedRMSNorm memory leak are different memory issues with different root causes." + }, + { + "left": "issue:41720", + "right": "issue:44977", + "accept": false, + "reason": "Qwen3 auto device mapping crashes on A800, whereas Qwen3.5 flash-attention generation is a different generation-path regression." + }, + { + "left": "issue:36331", + "right": "issue:42915", + "accept": false, + "reason": "Trainer `compute_loss` signature incompatibility is unrelated to Qwen3MoE FineGrainedFP8Config failure." + }, + { + "left": "issue:44466", + "right": "issue:45357", + "accept": false, + "reason": "Both mention Qwen3.5 and weights, but one is inconsistent lm_head serialization by device and the other is incorrect visual encoder keys on save; different bugs." + }, + { + "left": "issue:43450", + "right": "issue:45701", + "accept": false, + "reason": "Incorrect batched video processor shape and version-dependent tokenization changes are different subsystems." + }, + { + "left": "issue:44877", + "right": "issue:44991", + "accept": false, + "reason": "Strict config blocking granite_speech loading is not the same as tokenizer loading failure for EMBEDDIA/est-roberta." + }, + { + "left": "issue:43065", + "right": "issue:46402", + "accept": false, + "reason": "Dummy `nn.Conv2d` in Sam3PixelDecoder and Sam3Model text encoder weights not loading are separate Sam3 issues with different affected modules." + }, + { + "left": "issue:40990", + "right": "issue:45701", + "accept": false, + "reason": "Perplexity on gpt-oss-20b and tokenization changing across versions are not the same bug." + }, + { + "left": "issue:44186", + "right": "issue:45072", + "accept": false, + "reason": "LayoutLMv2Tokenizer crashes on NER/batched padding and bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel are unrelated." + }, + { + "left": "issue:26413", + "right": "issue:28530", + "accept": false, + "reason": "Checkpoint resume epoch iterator failure and early stopping metric lookup failure are distinct training-loop issues." + }, + { + "left": "issue:41720", + "right": "issue:42617", + "accept": false, + "reason": "Qwen3 auto device map cudaErrorAssert and inability to run `3d_parallel.py` are different runtime failures." + }, + { + "left": "issue:43550", + "right": "issue:43720", + "accept": false, + "reason": "Bamba-9B-v2 torch.compile SDPA failure and BitNet packed-weight unpacking during accelerate loading are different model/backend bugs." + }, + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "Quantization offloading problems and disabling mmap on Strix Halo to avoid OOM are related only at a high level; they are not the same concrete issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 15, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:28530", + "issue:30064", + "issue:33666", + "issue:36010", + "issue:36331", + "issue:41720", + "issue:42617", + "issue:42915", + "issue:43295", + "issue:43316", + "issue:43317", + "issue:43322", + "issue:43404", + "issue:43425", + "issue:43526", + "issue:43531", + "issue:43550", + "issue:43572", + "issue:43577", + "issue:43606", + "issue:43646", + "issue:43708", + "issue:43881", + "issue:44008", + "issue:44186", + "issue:44361", + "issue:44423", + "issue:44464", + "issue:44610", + "issue:44671", + "issue:44734", + "issue:44743", + "issue:44898", + "issue:44913", + "issue:45161", + "issue:45357", + "issue:45406", + "issue:45632", + "issue:45684", + "issue:45685", + "issue:45701", + "issue:45823", + "issue:46291", + "issue:46612", + "issue:46665", + "issue:46821", + "issue:46858", + "issue:46869" + ], + "soft_pairs": [ + "issue:46612|issue:46858", + "issue:43577|issue:43708", + "issue:43295|issue:45701", + "issue:45685|issue:46821", + "issue:43550|issue:43606", + "issue:44186|issue:44610", + "issue:43646|issue:46665", + "issue:43404|issue:45357", + "issue:44734|issue:46858", + "issue:44610|issue:44898", + "issue:42915|issue:43317", + "issue:36010|issue:42617", + "issue:44423|issue:45406", + "issue:44008|issue:45823", + "issue:46821|issue:46869", + "issue:45632|issue:45684", + "issue:44464|issue:46858", + "issue:41720|issue:45161", + "issue:28530|issue:33666", + "issue:44671|issue:44743", + "issue:43531|issue:43572", + "issue:42915|issue:46291", + "issue:36331|issue:41720", + "issue:44186|issue:44361", + "issue:43526|issue:43881", + "issue:43316|issue:44913", + "issue:43425|issue:46821", + "issue:42915|issue:43322", + "issue:28530|issue:30064" + ], + "prepared_review_unit_hash": "1a3bbb3d74fc120cd30bc5e20b91ea2123d977837ab0e2522d72af36e86ee658", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36434, + "estimated_input_tokens": 9109, + "estimated_eval_tokens": 18474 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: training, generation, serving, tokenizer, config, and model-specific regressions are mixed together, with no clear duplicate center and no PRs provided.", + "confidence": 0.96, + "canonical_issue_reason": "No single issue is a credible canonical duplicate target because the reports cover unrelated bugs and features across many subsystems.", + "canonical_pr_reason": "No PRs are present in this cluster, so there is no canonical PR.", + "best_issue_reason": "No issue is a strong global representative; the set is too diverse to collapse onto one underlying bug.", + "best_pr_reason": "No PRs are present, so there is no best PR candidate.", + "soft_edge_verdicts": [ + { + "left": "issue:46612", + "right": "issue:46858", + "accept": false, + "reason": "Both mention cache/generation, but one is beam-search reorder skipping and the other is static-cache seq-length arithmetic on non-CUDA backends; different bugs." + }, + { + "left": "issue:43577", + "right": "issue:43708", + "accept": false, + "reason": "One is dtype loading for BLIP2, the other is resume_from_checkpoint max_steps math; unrelated failures." + }, + { + "left": "issue:43295", + "right": "issue:45701", + "accept": false, + "reason": "One is a processor.tokenizer regression, the other is a broad tokenization-version change; not the same concrete bug." + }, + { + "left": "issue:45685", + "right": "issue:46821", + "accept": false, + "reason": "MPS histogram/int error and TimesFM window_size AttributeError are different model/API issues." + }, + { + "left": "issue:43550", + "right": "issue:43606", + "accept": false, + "reason": "Torch.compile+SDPA on Bamba and CPU offload device mismatch on bark-small are distinct code-path failures." + }, + { + "left": "issue:44186", + "right": "issue:44610", + "accept": false, + "reason": "LayoutLMv2 tokenizer crashes on NER padding/truncation, while OmDet-Turbo has an image-size mismatch; different subsystems." + }, + { + "left": "issue:43646", + "right": "issue:46665", + "accept": false, + "reason": "Both involve loading/custom model behavior, but one is Transformers 5.0 custom initialization breakage and the other is from_pretrained inside Python classes; too broad and not the same bug." + }, + { + "left": "issue:43404", + "right": "issue:45357", + "accept": false, + "reason": "One is untied lm_head weights in Mistral3, the other is incorrect visual encoder keys on save_pretrained for Qwen3.5." + }, + { + "left": "issue:44734", + "right": "issue:46858", + "accept": false, + "reason": "Both are cache-related, but one is serve KV-cache continuation indexing and the other is prefill seq-length tensor arithmetic; different failure points." + }, + { + "left": "issue:44610", + "right": "issue:44898", + "accept": false, + "reason": "Processor output size mismatch vs Perceiver non-default-resolution interpolation failure are not the same underlying issue." + }, + { + "left": "issue:42915", + "right": "issue:43317", + "accept": false, + "reason": "FineGrainedFP8Config failure for Qwen3Moe and dequantized model load with gpu+cpu offload are unrelated." + }, + { + "left": "issue:36010", + "right": "issue:42617", + "accept": false, + "reason": "GenerationMixin import failure and inability to run 3d_parallel.py are unrelated." + }, + { + "left": "issue:44423", + "right": "issue:45406", + "accept": false, + "reason": "Both are serving crashes, but one is multimodal continuous batching string-to(). error and the other is missing _tokenizer on Gemma4Processor; different bugs." + }, + { + "left": "issue:44008", + "right": "issue:45823", + "accept": false, + "reason": "Gemma3n variable-name collision in forward and Gemma4 PLE device mismatch are different model-specific regressions." + }, + { + "left": "issue:46821", + "right": "issue:46869", + "accept": false, + "reason": "TimesFM window_size AttributeError and YOLO multi-GPU device_map failure do not share the same code path." + }, + { + "left": "issue:45632", + "right": "issue:45684", + "accept": false, + "reason": "One is a trust_remote_code cache-path collision, the other is permissions propagation in save_pretrained; unrelated." + }, + { + "left": "issue:44464", + "right": "issue:46858", + "accept": false, + "reason": "Chunked generation inconsistency under compiled forward is a different bug from static-cache seq-length arithmetic." + }, + { + "left": "issue:41720", + "right": "issue:45161", + "accept": false, + "reason": "Auto device-mapped Qwen3 cudaErrorAssert and GPT-OSS MoE TP malfunction are different parallelism/device issues." + }, + { + "left": "issue:28530", + "right": "issue:33666", + "accept": false, + "reason": "Early-stopping metric lookup and Qwen2-VL multi-GPU training are unrelated." + }, + { + "left": "issue:44671", + "right": "issue:44743", + "accept": false, + "reason": "CamemBERT masked-LM prediction regression and Qwen3.5 recurrent-state reset with cache are different model behaviors." + }, + { + "left": "issue:43531", + "right": "issue:43572", + "accept": false, + "reason": "Qwen3-MoE sliding_window handling and StableLmConfig pad_token_idx omission are unrelated configuration/model issues." + }, + { + "left": "issue:42915", + "right": "issue:46291", + "accept": false, + "reason": "FP8 model failure and kernels import/version requirement error are separate dependency/runtime problems." + }, + { + "left": "issue:36331", + "right": "issue:41720", + "accept": false, + "reason": "Trainer compute_loss signature change and Qwen3 auto device mapping CUDA assert are unrelated." + }, + { + "left": "issue:44186", + "right": "issue:44361", + "accept": false, + "reason": "Different tokenizers, different models, and different crash conditions; not the same bug." + }, + { + "left": "issue:43526", + "right": "issue:43881", + "accept": false, + "reason": "Beit reduce_labels bug and glm-4v-9b loading failure are unrelated." + }, + { + "left": "issue:43316", + "right": "issue:44913", + "accept": false, + "reason": "Gemma3TextConfig API mismatch and GPTNeoXConfig rotary_pct reload regression are distinct config issues." + }, + { + "left": "issue:43425", + "right": "issue:46821", + "accept": false, + "reason": "Torch 2.10 compatibility and TimesFM window_size AttributeError are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:43322", + "accept": false, + "reason": "FineGrainedFP8Config failure and Llava Next segmentation fault are different bugs." + }, + { + "left": "issue:28530", + "right": "issue:30064", + "accept": false, + "reason": "Early stopping metric resolution and void segmentation-map processing 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": 23, + "review_unit_index": 16, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:29159", + "issue:30064", + "issue:33357", + "issue:34567", + "issue:34634", + "issue:35707", + "issue:36032", + "issue:41720", + "issue:42490", + "issue:43329", + "issue:43388", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43606", + "issue:43644", + "issue:43653", + "issue:43720", + "issue:43746", + "issue:43782", + "issue:43873", + "issue:44062", + "issue:44077", + "issue:44387", + "issue:44451", + "issue:44492", + "issue:44509", + "issue:44534", + "issue:44589", + "issue:44625", + "issue:44898", + "issue:44987", + "issue:44991", + "issue:45042", + "issue:45059", + "issue:45072", + "issue:45357", + "issue:45406", + "issue:45464", + "issue:45479", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45859", + "issue:46402", + "issue:46665", + "issue:46821", + "issue:46858" + ], + "soft_pairs": [ + "issue:33357|issue:35707", + "issue:29159|issue:36032", + "issue:43540|issue:43782", + "issue:43550|issue:43653", + "issue:44387|issue:45701", + "issue:30064|issue:34634", + "issue:43873|issue:45704", + "issue:43606|issue:43720", + "issue:44492|issue:44509", + "issue:44534|issue:46665", + "issue:43531|issue:45759", + "issue:41720|issue:43746", + "issue:42490|issue:43329", + "issue:43388|issue:45059", + "issue:45406|issue:45464", + "issue:43606|issue:43653", + "issue:34567|issue:35707", + "issue:43531|issue:46821", + "issue:34634|issue:41720", + "issue:43540|issue:46858", + "issue:45759|issue:45859", + "issue:44451|issue:44991", + "issue:45357|issue:46402", + "issue:43531|issue:44589", + "issue:44077|issue:44625", + "issue:43644|issue:46665", + "issue:44062|issue:44451", + "issue:44898|issue:45072", + "issue:44987|issue:45042", + "issue:44625|issue:45479" + ], + "prepared_review_unit_hash": "4679ea54ab8bb52a7c12b6f82d5ca966221e7fe72f56c1a2e617cfe9798664d1", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36838, + "estimated_input_tokens": 9210, + "estimated_eval_tokens": 18676 + }, + "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 tokenizer, multimodal, quantization/offload, serving, and training issues. I would reject every soft edge candidate.", + "confidence": 0.93, + "canonical_issue_reason": "No single issue is a good canonical representative because the cluster is a loose similarity bucket covering many unrelated bug classes and code paths.", + "canonical_pr_reason": null, + "best_issue_reason": "No issue is especially suitable as a global best representative for dedupe; the artifacts do not converge on one underlying bug or fix.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:33357", + "right": "issue:35707", + "accept": false, + "reason": "Different problems: MacOS CLIP bus error vs generation with inputs_embeds/past_key_values." + }, + { + "left": "issue:29159", + "right": "issue:36032", + "accept": false, + "reason": "Both tokenizer-related, but one is slow/fast inconsistency while the other is a T5 tokenizer load-time method conflict." + }, + { + "left": "issue:43540", + "right": "issue:43782", + "accept": false, + "reason": "Both involve Qwen multimodal loading/processing, but they hit different model paths and failure modes." + }, + { + "left": "issue:43550", + "right": "issue:43653", + "accept": false, + "reason": "Unrelated: torch.compile/SDPA failure in Bamba vs BigBirdTokenizer special-token decode bug." + }, + { + "left": "issue:44387", + "right": "issue:45701", + "accept": false, + "reason": "Int4 quantization memory growth is unrelated to tokenization changes across versions." + }, + { + "left": "issue:30064", + "right": "issue:34634", + "accept": false, + "reason": "Different processors and symptoms: void segmentation map handling vs Bark voice_preset behavior." + }, + { + "left": "issue:43873", + "right": "issue:45704", + "accept": false, + "reason": "Offloading/quantization mismatch is not the same as T5's apex RMSNorm memory leak." + }, + { + "left": "issue:43606", + "right": "issue:43720", + "accept": false, + "reason": "Both are load/offload regressions, but they affect different models and concrete code paths." + }, + { + "left": "issue:44492", + "right": "issue:44509", + "accept": false, + "reason": "A documentation typo is not the same as stale pipeline-task docs removed in v5." + }, + { + "left": "issue:44534", + "right": "issue:46665", + "accept": false, + "reason": "Non-persistent buffer initialization is unrelated to from_pretrained behavior inside Python classes." + }, + { + "left": "issue:43531", + "right": "issue:45759", + "accept": false, + "reason": "Different Qwen config bugs: sliding_window handling vs failing to unwrap text_config in from_config." + }, + { + "left": "issue:41720", + "right": "issue:43746", + "accept": false, + "reason": "Qwen3 auto device-mapping CUDA assert is unrelated to GraniteSpeech PEFT checkpoint loading." + }, + { + "left": "issue:42490", + "right": "issue:43329", + "accept": false, + "reason": "Local-path import warning and multimodal token counting are separate bugs." + }, + { + "left": "issue:43388", + "right": "issue:45059", + "accept": false, + "reason": "Metric gathering truncation of tuple labels is unrelated to SAM3 text/bbox behavior." + }, + { + "left": "issue:45406", + "right": "issue:45464", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer in serve is a different failure than Qwen streaming inference." + }, + { + "left": "issue:43606", + "right": "issue:43653", + "accept": false, + "reason": "Device-mismatch CPU offload bug and tokenizer special-token registration bug do not share a code path." + }, + { + "left": "issue:34567", + "right": "issue:35707", + "accept": false, + "reason": "Trainer token counters and progressive generation with cache are unrelated." + }, + { + "left": "issue:43531", + "right": "issue:46821", + "accept": false, + "reason": "Both mention windowing, but Qwen3-MoE sliding_window and TimesFM window_size are different APIs and bugs." + }, + { + "left": "issue:34634", + "right": "issue:41720", + "accept": false, + "reason": "BarkProcessor voice_preset and Qwen3 auto device mapping are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:46858", + "accept": false, + "reason": "Qwen3OmniMoe video-processing ValueError is not the same as static-cache slice arithmetic on non-CUDA backends." + }, + { + "left": "issue:45759", + "right": "issue:45859", + "accept": false, + "reason": "Both are Qwen3.5 config/offload issues, but one is config unwrapping while the other is embedding forward compatibility under CPU offload." + }, + { + "left": "issue:44451", + "right": "issue:44991", + "accept": false, + "reason": "Different model/tokenizer loading regressions for different checkpoints." + }, + { + "left": "issue:45357", + "right": "issue:46402", + "accept": false, + "reason": "Both concern multimodal save/load state, but they affect different models and different weight keys." + }, + { + "left": "issue:43531", + "right": "issue:44589", + "accept": false, + "reason": "Sliding-window handling in Qwen3-MoE is unrelated to missing Float8 storage errors during load." + }, + { + "left": "issue:44077", + "right": "issue:44625", + "accept": false, + "reason": "Optional post_init validation and num_labels propagation are different configuration bugs." + }, + { + "left": "issue:43644", + "right": "issue:46665", + "accept": false, + "reason": "Non-persistent buffer junk in v5 is not the same as from_pretrained behavior inside user-defined Python classes." + }, + { + "left": "issue:44062", + "right": "issue:44451", + "accept": false, + "reason": "AddedToken keyword duplication is a tokenizer construction bug, not a general model-loading failure." + }, + { + "left": "issue:44898", + "right": "issue:45072", + "accept": false, + "reason": "Perceiver image resolution handling and bfloat16 dtype mismatches are unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45042", + "accept": false, + "reason": "One is a model loading regression, the other is an image-processor dependency requirement bug." + }, + { + "left": "issue:44625", + "right": "issue:45479", + "accept": false, + "reason": "Config propagation to text_config is distinct from the degenerate zero-loss issue in sequence classification." + } + ] + }, + "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": 23, + "review_unit_index": 17, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:30064", + "issue:36010", + "issue:39692", + "issue:41720", + "issue:42886", + "issue:43295", + "issue:43381", + "issue:43421", + "issue:43493", + "issue:43550", + "issue:43653", + "issue:43720", + "issue:43746", + "issue:43782", + "issue:43819", + "issue:43844", + "issue:43873", + "issue:43927", + "issue:43994", + "issue:44186", + "issue:44361", + "issue:44368", + "issue:44514", + "issue:44545", + "issue:44610", + "issue:44671", + "issue:44843", + "issue:44918", + "issue:44936", + "issue:44995", + "issue:45081", + "issue:45356", + "issue:45357", + "issue:45381", + "issue:45704", + "issue:45759", + "issue:45834", + "issue:45907", + "issue:45923", + "issue:45987", + "issue:46032", + "issue:46050", + "issue:46463", + "issue:46655", + "issue:46693", + "issue:46829", + "issue:46832", + "issue:46858" + ], + "soft_pairs": [ + "issue:43381|issue:43844", + "issue:44368|issue:45357", + "issue:39692|issue:44610", + "issue:46655|issue:46693", + "issue:43421|issue:43927", + "issue:41720|issue:45704", + "issue:43746|issue:46050", + "issue:43381|issue:45704", + "issue:42886|issue:44843", + "issue:45704|issue:45923", + "issue:44918|issue:44936", + "issue:44514|issue:44545", + "issue:45759|issue:45923", + "issue:43746|issue:43873", + "issue:43421|issue:43653", + "issue:43994|issue:44671", + "issue:39692|issue:43493", + "issue:46032|issue:46463", + "issue:43782|issue:45381", + "issue:43421|issue:44186", + "issue:45081|issue:45356", + "issue:30064|issue:36010", + "issue:44995|issue:46858", + "issue:45356|issue:45907", + "issue:39692|issue:43873", + "issue:45987|issue:46829", + "issue:43720|issue:43819", + "issue:46829|issue:46832", + "issue:43653|issue:43720", + "issue:43295|issue:45834", + "issue:43550|issue:44361" + ], + "prepared_review_unit_hash": "2fff6a4147a75e5534bed2495d51020b71d658c7a05c01542e19cd7352d497d8", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37467, + "estimated_input_tokens": 9367, + "estimated_eval_tokens": 18990 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose semantic grouping of tokenizer/processor/model-loading bug reports, not a single duplicate set. The only clear same-issue duplicate is the Qwen2_5_VLProcessor batched apply_chat_template crash pair; the rest are related only at subsystem level and should stay separate.", + "confidence": 0.81, + "canonical_issue_reason": "Issue 44514 is the clearest exact duplicate representative: 44545 reports the same Qwen2_5_VLProcessor.apply_chat_template batched-input crash with padding=False.", + "canonical_pr_reason": null, + "best_issue_reason": "44514 is the best global representative because it is a concrete, exact duplicate pair with a specific reproducible crash; the other candidates are broader or only loosely related.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43381", + "right": "issue:43844", + "accept": false, + "reason": "Both touch training-time behavior, but one is gradient checkpointing in eval mode and the other is a ZeRO-3/DeepSpeed gradient anomaly; different bugs and code paths." + }, + { + "left": "issue:44368", + "right": "issue:45357", + "accept": false, + "reason": "Both mention Qwen model warnings/regressions, but one is a tie_word_embeddings warning and the other is incorrect visual-encoder save keys; not the same bug." + }, + { + "left": "issue:39692", + "right": "issue:44610", + "accept": false, + "reason": "Both are vision/multimodal issues, but one is a SigLIP2 doc/model mismatch plus quantization failure and the other is an OmDet-Turbo image-size mismatch." + }, + { + "left": "issue:46655", + "right": "issue:46693", + "accept": false, + "reason": "Video-related, but one is a padding calculation bug in the video processor and the other is a FA2 long-context performance regression." + }, + { + "left": "issue:43421", + "right": "issue:43927", + "accept": false, + "reason": "Both involve token handling, but one is runtime post-processor updates and the other is save/load losing custom token IDs; different failure modes." + }, + { + "left": "issue:41720", + "right": "issue:45704", + "accept": false, + "reason": "Unrelated root causes: Qwen3 auto device mapping cudaErrorAssert versus a T5 apex fused RMSNorm memory leak." + }, + { + "left": "issue:43746", + "right": "issue:46050", + "accept": false, + "reason": "Both are load/save regressions, but one is PEFT adapter loading from local checkpoints and the other is sinq quantized model persistence." + }, + { + "left": "issue:43381", + "right": "issue:45704", + "accept": false, + "reason": "Gradient checkpointing in eval mode and an apex RMSNorm leak are different bugs with no shared code path." + }, + { + "left": "issue:42886", + "right": "issue:44843", + "accept": false, + "reason": "Both break offline workflows, but one is generic cache loading with HF_HUB_OFFLINE and the other is a specific model_info() call in _patch_mistral_regex." + }, + { + "left": "issue:45704", + "right": "issue:45923", + "accept": false, + "reason": "T5 RMSNorm leak versus a missing supports_gradient_checkpointing flag on a trust_remote_code model; unrelated issues." + }, + { + "left": "issue:44918", + "right": "issue:44936", + "accept": false, + "reason": "Both involve training/evaluation workflows, but one is SFT embedding unpacking and the other is trainer.evaluate() failing after train(); different code paths." + }, + { + "left": "issue:44514", + "right": "issue:44545", + "accept": true, + "reason": "Same title and same reported Qwen2_5_VLProcessor.apply_chat_template crash on batched input when padding=False; clear duplicate." + }, + { + "left": "issue:45759", + "right": "issue:45923", + "accept": false, + "reason": "Composite Qwen config unwrapping and a missing gradient-checkpointing flag are separate model-initialization issues." + }, + { + "left": "issue:43746", + "right": "issue:43873", + "accept": false, + "reason": "Local checkpoint loading for PEFT adapters versus quantization offloading behavior; too different to merge." + }, + { + "left": "issue:43421", + "right": "issue:43653", + "accept": false, + "reason": "Runtime post-processor refresh and BigBird mask-token special-token registration are related only broadly through tokenizers." + }, + { + "left": "issue:43994", + "right": "issue:44671", + "accept": false, + "reason": "Different models and failures: SigLIP2 nonsensical outputs versus CamemBERT masked-LM prediction issues." + }, + { + "left": "issue:39692", + "right": "issue:43493", + "accept": false, + "reason": "Both are SigLIP2-related, but one is a doc/example mismatch and quantization issue while the other is HF/JAX implementation parity; not the same concrete bug." + }, + { + "left": "issue:46032", + "right": "issue:46463", + "accept": false, + "reason": "One is incorrect use_cache results for Mamba2Mixer, the other is just an error-message bug in T5 decoder validation." + }, + { + "left": "issue:43782", + "right": "issue:45381", + "accept": false, + "reason": "Different Qwen VL bugs: from_pretrained weight_only loading versus incorrect vision_position_ids in video input." + }, + { + "left": "issue:43421", + "right": "issue:44186", + "accept": false, + "reason": "Both tokenizer-related, but one is special-token post-processor updates and the other is LayoutLMv2 NER/padding crashes; not duplicates." + }, + { + "left": "issue:45081", + "right": "issue:45356", + "accept": false, + "reason": "Both involve _patch_mistral_regex, but one is an AttributeError loading Mistral tokenizers and the other is Kimi-K2.5 codec/warning regression; not the same failure." + }, + { + "left": "issue:30064", + "right": "issue:36010", + "accept": false, + "reason": "Completely unrelated: void segmentation map processing versus an ImportError for GenerationMixin." + }, + { + "left": "issue:44995", + "right": "issue:46858", + "accept": false, + "reason": "Both mention cache/sequence length, but one is a stale indexer cache on second forward pass and the other is a static-cache device-tensor arithmetic bug under torch.compile." + }, + { + "left": "issue:45356", + "right": "issue:45907", + "accept": false, + "reason": "Tokenizer regression versus a hub API revision-forwarding bug; unrelated." + }, + { + "left": "issue:39692", + "right": "issue:43873", + "accept": false, + "reason": "SigLIP2 doc/model mismatch versus offloading with quantization; no shared underlying bug." + }, + { + "left": "issue:45987", + "right": "issue:46829", + "accept": false, + "reason": "StaticCache.get_seq_length() returning a tensor and VideoPrism output-field misassignment are unrelated." + }, + { + "left": "issue:43720", + "right": "issue:43819", + "accept": false, + "reason": "BitNet packed-weight loading and DAC from_latents STE mismatch are different model-specific bugs." + }, + { + "left": "issue:46829", + "right": "issue:46832", + "accept": false, + "reason": "One is an output-field bug in VideoPrism classification, the other is an RT-DETR forward-shape mismatch; different code paths." + }, + { + "left": "issue:43653", + "right": "issue:43720", + "accept": false, + "reason": "BigBird special-token decode failure and BitNet accelerate-loading unpacking are unrelated." + }, + { + "left": "issue:43295", + "right": "issue:45834", + "accept": false, + "reason": "Processor.tokenizer regression in v4.57.5 versus a Kosmos2.5 long-OCR index error; different issues." + }, + { + "left": "issue:43550", + "right": "issue:44361", + "accept": false, + "reason": "A torch.compile/SDPA crash in Bamba is unrelated to an MLukeTokenizer AttributeError." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The analysis is grounded in the packet and stays conservative. The only clearly supported duplicate pair is issue:44514 / issue:44545, which share the same Qwen2_5_VLProcessor.apply_chat_template batched-input crash. The remaining pairwise rejections are appropriately cautious based on the issue titles alone, and the summary does not overclaim broader duplication beyond subsystem-level similarity." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 18, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:30064", + "issue:34634", + "issue:36010", + "issue:36331", + "issue:39692", + "issue:41720", + "issue:43335", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43596", + "issue:43606", + "issue:43611", + "issue:43618", + "issue:43643", + "issue:43644", + "issue:43653", + "issue:43697", + "issue:43749", + "issue:43756", + "issue:43819", + "issue:43927", + "issue:44008", + "issue:44062", + "issue:44186", + "issue:44220", + "issue:44442", + "issue:44448", + "issue:44488", + "issue:44661", + "issue:44671", + "issue:44821", + "issue:44843", + "issue:44877", + "issue:44898", + "issue:44991", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45183", + "issue:45381", + "issue:45479", + "issue:45584", + "issue:45834", + "issue:46294", + "issue:46665", + "issue:46693", + "issue:46858" + ], + "soft_pairs": [ + "issue:44062|issue:44488", + "issue:43550|issue:43927", + "issue:43596|issue:44220", + "issue:45381|issue:46693", + "issue:43540|issue:46294", + "issue:44008|issue:46858", + "issue:43531|issue:43643", + "issue:43611|issue:46665", + "issue:43653|issue:43819", + "issue:44220|issue:44661", + "issue:44186|issue:44898", + "issue:43618|issue:43697", + "issue:41720|issue:44991", + "issue:30064|issue:36331", + "issue:43643|issue:44877", + "issue:43550|issue:43819", + "issue:43335|issue:43756", + "issue:44843|issue:45081", + "issue:44821|issue:45084", + "issue:44186|issue:44442", + "issue:39692|issue:45834", + "issue:44448|issue:44671", + "issue:43540|issue:44008", + "issue:43644|issue:43749", + "issue:34634|issue:36010", + "issue:45183|issue:45479", + "issue:45584|issue:45834", + "issue:43606|issue:45072" + ], + "prepared_review_unit_hash": "6e034c7e02e035893ceed71979bff5673b37259e1a9d9144c08e7c84406b879c", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36296, + "estimated_input_tokens": 9074, + "estimated_eval_tokens": 18404 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose collection of mostly unrelated Transformers bug reports. A few items cluster around loading/config/tokenizer regressions, but the soft pairs still point to different concrete failure modes, models, or code paths, so none should be merged as duplicates.", + "confidence": 0.87, + "canonical_issue_reason": "issue:43611 is the most broadly representative loading regression in the set: it describes a Transformers 5.0.0 model-loading breakage around `base_model_prefix`, which aligns with the cluster\u2019s strongest repeated theme of load/config failures.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43611 is the best single issue to anchor this cluster because it is a clear, concrete regression with broader applicability than the more model-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44062", + "right": "issue:44488", + "accept": false, + "reason": "Both involve tokenizer loading/special-token behavior, but one is an `AddedToken(..., special=...)` argument error and the other is a model-specific tokenizer load failure; different bugs." + }, + { + "left": "issue:43550", + "right": "issue:43927", + "accept": false, + "reason": "One is a `torch.compile`/SDPA failure in Bamba, the other is a save/load config token-ID loss in Dia; unrelated code paths." + }, + { + "left": "issue:43596", + "right": "issue:44220", + "accept": false, + "reason": "`BertModel` zero3 init IndexError and `_torch_extract_fbank_features()` are different subsystems and different failure modes." + }, + { + "left": "issue:45381", + "right": "issue:46693", + "accept": false, + "reason": "Qwen2.5-VL video position IDs and FA2 long-context performance regression are unrelated issues." + }, + { + "left": "issue:43540", + "right": "issue:46294", + "accept": false, + "reason": "Both are video-related, but one is a Qwen3OmniMoe video-processing ValueError and the other is a chat-template token/feature count mismatch in Qwen3.5-9B." + }, + { + "left": "issue:44008", + "right": "issue:46858", + "accept": false, + "reason": "Gemma3n variable-name collision in forward is unrelated to `_prefill` cache sequence-length tensor arithmetic on non-CUDA backends." + }, + { + "left": "issue:43531", + "right": "issue:43643", + "accept": false, + "reason": "Sliding-window handling in Qwen3-MoE and missing fields from `AutoConfig.from_pretrained(..., trust_remote_code=True)` are different bugs." + }, + { + "left": "issue:43611", + "right": "issue:46665", + "accept": false, + "reason": "Both touch model loading, but one is a specific `base_model_prefix` regression and the other is inconsistent `from_pretrained` behavior inside Python classes." + }, + { + "left": "issue:43653", + "right": "issue:43819", + "accept": false, + "reason": "BigBirdTokenizer special-token registration and DAC `from_latents`/forward mismatch are unrelated." + }, + { + "left": "issue:44220", + "right": "issue:44661", + "accept": false, + "reason": "Audio feature extraction and `TOKENIZER_MAPPING_NAMES` model-addition logic are separate paths." + }, + { + "left": "issue:44186", + "right": "issue:44898", + "accept": false, + "reason": "LayoutLMv2 tokenizer padding/truncation crashes and Perceiver image-classification failures are different model families and bugs." + }, + { + "left": "issue:43618", + "right": "issue:43697", + "accept": false, + "reason": "Both are output regressions, but CLIP attentions missing and RTDetrV2 differing outputs are not the same underlying defect." + }, + { + "left": "issue:41720", + "right": "issue:44991", + "accept": false, + "reason": "Qwen3 auto device mapping CUDA assert and tokenizer loading for `EMBEDDIA/est-roberta` are unrelated." + }, + { + "left": "issue:30064", + "right": "issue:36331", + "accept": false, + "reason": "Void segmentation-map processing and `compute_loss(..., num_items_in_batch)` are different issues in different layers of the stack." + }, + { + "left": "issue:43643", + "right": "issue:44877", + "accept": false, + "reason": "Both concern config loading strictness, but one is missing fields from `trust_remote_code` and the other is a `granite_speech` config rejection; different root causes." + }, + { + "left": "issue:43550", + "right": "issue:43819", + "accept": false, + "reason": "`torch.compile`+SDPA in Bamba and DAC latent decoding mismatch are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:43756", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation and Smollm3 RoPE-layer dropping are different model-specific architecture bugs." + }, + { + "left": "issue:44843", + "right": "issue:45081", + "accept": false, + "reason": "Both involve `_patch_mistral_regex`, but one is an offline `model_info()` call and the other is an AttributeError on `backend_tokenizer`; separate failure modes." + }, + { + "left": "issue:44821", + "right": "issue:45084", + "accept": false, + "reason": "AutoImageProcessor URL loading and template compilation errors are unrelated." + }, + { + "left": "issue:44186", + "right": "issue:44442", + "accept": false, + "reason": "LayoutLMv2 tokenizer input handling and FastSpeech2Conformer tokenizer loading are different tokenizer bugs." + }, + { + "left": "issue:39692", + "right": "issue:45834", + "accept": false, + "reason": "SigLIP2 doc-example errors and Kosmos2.5 long OCR index errors are not the same underlying bug." + }, + { + "left": "issue:44448", + "right": "issue:44671", + "accept": false, + "reason": "Both are v5 output differences, but they affect different models and likely different code paths." + }, + { + "left": "issue:43540", + "right": "issue:44008", + "accept": false, + "reason": "Qwen3OmniMoe video-input processing and Gemma3n audio mask attribute collision are different modalities and bugs." + }, + { + "left": "issue:43644", + "right": "issue:43749", + "accept": false, + "reason": "Non-persistent-buffer junk values and FSDP CPU RAM efficient loading are both v5/loading-adjacent, but not the same defect." + }, + { + "left": "issue:34634", + "right": "issue:36010", + "accept": false, + "reason": "BarkProcessor voice preset and `GenerationMixin` import failure are unrelated." + }, + { + "left": "issue:45183", + "right": "issue:45479", + "accept": false, + "reason": "Input validation XOR messaging and degenerate zero loss in sequence classification are different model-behavior bugs." + }, + { + "left": "issue:45584", + "right": "issue:45834", + "accept": false, + "reason": "Whisper empty transcription after special-token alignment and Kosmos2.5 long OCR index errors are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:45072", + "accept": false, + "reason": "CPU-offload device mismatch in Bark and dtype mismatches in bfloat16 inference are distinct runtime 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": 23, + "review_unit_index": 19, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:28218", + "issue:29159", + "issue:42175", + "issue:43322", + "issue:43388", + "issue:43421", + "issue:43540", + "issue:43550", + "issue:43577", + "issue:43582", + "issue:43645", + "issue:43716", + "issue:43720", + "issue:43749", + "issue:43782", + "issue:43818", + "issue:43825", + "issue:43827", + "issue:43931", + "issue:43950", + "issue:43994", + "issue:44008", + "issue:44291", + "issue:44479", + "issue:44561", + "issue:44610", + "issue:44671", + "issue:44805", + "issue:44811", + "issue:44936", + "issue:44977", + "issue:44991", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45381", + "issue:45685", + "issue:45701", + "issue:45834", + "issue:45987", + "issue:46032", + "issue:46291", + "issue:46294", + "issue:46463", + "issue:46512", + "issue:46665", + "issue:46846" + ], + "soft_pairs": [ + "issue:43322|issue:45685", + "issue:43550|issue:45072", + "issue:43388|issue:46032", + "issue:44479|issue:45834", + "issue:43782|issue:44977", + "issue:43825|issue:43994", + "issue:43582|issue:46512", + "issue:43577|issue:43818", + "issue:44991|issue:46291", + "issue:43421|issue:44811", + "issue:43950|issue:45987", + "issue:44561|issue:45020", + "issue:43716|issue:44008", + "issue:43931|issue:46846", + "issue:46294|issue:46463", + "issue:44805|issue:44936", + "issue:28218|issue:29159", + "issue:44610|issue:45072", + "issue:43540|issue:45381", + "issue:43720|issue:45072", + "issue:43749|issue:44291", + "issue:43825|issue:43827", + "issue:43577|issue:43582", + "issue:43645|issue:46665", + "issue:44671|issue:45701", + "issue:42175|issue:45081" + ], + "prepared_review_unit_hash": "5751aff18ffb06408a9d2d78faa1ef1b9a877bee07f551d210e8069e4cc2ccfe", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35339, + "estimated_input_tokens": 8835, + "estimated_eval_tokens": 17926 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: tokenizer/pipeline docs issues, multimodal/video regressions, dtype/loading bugs, trainer/cache problems, and a few API-contract issues. None of the soft pairs look like the same underlying bug or change, so they should stay separate.", + "confidence": 0.86, + "canonical_issue_reason": "No canonical issue: the cluster is a mix of unrelated tokenizer, video, loading, dtype, and trainer bugs.", + "canonical_pr_reason": "No pull requests are present in this cluster.", + "best_issue_reason": "No clear global representative; the issue set is too heterogeneous for a meaningful duplicate anchor.", + "best_pr_reason": "No pull requests are present in this cluster.", + "soft_edge_verdicts": [ + { + "left": "issue:43322", + "right": "issue:45685", + "accept": false, + "reason": "Different failures: Llava Next loading crash vs an MPS histogram/int support error in moe code." + }, + { + "left": "issue:43550", + "right": "issue:45072", + "accept": false, + "reason": "Different models and root causes: torch.compile/SDPA failure vs dtype mismatch in bfloat16 inference." + }, + { + "left": "issue:43388", + "right": "issue:46032", + "accept": false, + "reason": "Trainer metric gathering drops labels; Mamba2Mixer is a cache correctness bug." + }, + { + "left": "issue:44479", + "right": "issue:45834", + "accept": false, + "reason": "Both are multimodal regressions, but they affect different models and different code paths." + }, + { + "left": "issue:43782", + "right": "issue:44977", + "accept": false, + "reason": "Loading with weight_only=True is not the same as Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:43825", + "right": "issue:43994", + "accept": false, + "reason": "One is an error-message regression; the other is a model/pipeline inference-quality problem." + }, + { + "left": "issue:43582", + "right": "issue:46512", + "accept": false, + "reason": "Apple Silicon allocator warmup TypeError is unrelated to stale config.dtype after from_config()." + }, + { + "left": "issue:43577", + "right": "issue:43818", + "accept": false, + "reason": "Blip2 dtype propagation and Video-LLaVA temporal-attention/weight-sharing are different bugs." + }, + { + "left": "issue:44991", + "right": "issue:46291", + "accept": false, + "reason": "Tokenizer-loading failure and pipelines import failure are different subsystems and distinct errors." + }, + { + "left": "issue:43421", + "right": "issue:44811", + "accept": false, + "reason": "Runtime post-processor updates are not the same bug as batch_decode ignoring skip_special_tokens." + }, + { + "left": "issue:43950", + "right": "issue:45987", + "accept": false, + "reason": "Non-persistent buffer corruption and StaticCache return-type mismatch are unrelated contract bugs." + }, + { + "left": "issue:44561", + "right": "issue:45020", + "accept": false, + "reason": "Both mention remote_code breakage, but the underlying removed APIs/problems are different." + }, + { + "left": "issue:43716", + "right": "issue:44008", + "accept": false, + "reason": "dtype mismatch in a processor/model pair is unrelated to a variable name collision in forward()." + }, + { + "left": "issue:43931", + "right": "issue:46846", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch and missing TP-plan entries are different loading vs tensor-parallel bugs." + }, + { + "left": "issue:46294", + "right": "issue:46463", + "accept": false, + "reason": "Video chat-template token/feature mismatch is unrelated to T5 decoder input validation messaging." + }, + { + "left": "issue:44805", + "right": "issue:44936", + "accept": false, + "reason": "A tensor-mask shape error is not the same issue as trainer.evaluate() failing after train()." + }, + { + "left": "issue:28218", + "right": "issue:29159", + "accept": false, + "reason": "Both are tokenizer regressions, but one is about added-token spacing while the other is a broader slow/fast inconsistency." + }, + { + "left": "issue:44610", + "right": "issue:45072", + "accept": false, + "reason": "Different models and symptoms: processor size mismatch vs dtype mismatches." + }, + { + "left": "issue:43540", + "right": "issue:45381", + "accept": false, + "reason": "Different Qwen video-input bugs with different reported failures and likely different code paths." + }, + { + "left": "issue:43720", + "right": "issue:45072", + "accept": false, + "reason": "Packed-weight unpacking during accelerate loading is unrelated to bfloat16 dtype mismatches." + }, + { + "left": "issue:43749", + "right": "issue:44291", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and init_empty_weights argument handling are distinct loading-path regressions." + }, + { + "left": "issue:43825", + "right": "issue:43827", + "accept": false, + "reason": "A bad v5 error message and stale documentation are related topics, but not the same underlying bug/change." + }, + { + "left": "issue:43577", + "right": "issue:43582", + "accept": false, + "reason": "Blip2 dtype propagation and Apple Silicon allocator warmup are unrelated." + }, + { + "left": "issue:43645", + "right": "issue:46665", + "accept": false, + "reason": "Jupyter custom-model initialization regression is different from from_pretrained behavior inside Python classes." + }, + { + "left": "issue:44671", + "right": "issue:45701", + "accept": false, + "reason": "CamemBERT prediction errors are not the same as a general tokenization-change report." + }, + { + "left": "issue:42175", + "right": "issue:45081", + "accept": false, + "reason": "Missing TensorFlow in an install extra is unrelated to the Mistral tokenizer regex crash." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 20, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:34634", + "issue:36010", + "issue:36331", + "issue:42175", + "issue:42490", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43452", + "issue:43526", + "issue:43531", + "issue:43550", + "issue:43582", + "issue:43638", + "issue:43645", + "issue:43653", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43782", + "issue:43819", + "issue:43906", + "issue:43927", + "issue:44060", + "issue:44368", + "issue:44554", + "issue:44589", + "issue:44610", + "issue:44857", + "issue:44912", + "issue:44936", + "issue:44991", + "issue:45003", + "issue:45020", + "issue:45042", + "issue:45081", + "issue:45084", + "issue:45325", + "issue:45372", + "issue:45406", + "issue:45538", + "issue:45600", + "issue:46209", + "issue:46285", + "issue:46291", + "issue:46326", + "issue:46697", + "issue:46728" + ], + "soft_pairs": [ + "issue:34634|issue:36331", + "issue:45003|issue:46285", + "issue:46697|issue:46728", + "issue:43653|issue:44912", + "issue:43782|issue:45325", + "issue:43262|issue:43819", + "issue:45081|issue:45406", + "issue:43582|issue:44857", + "issue:45372|issue:46326", + "issue:44589|issue:45042", + "issue:43550|issue:44912", + "issue:43742|issue:45020", + "issue:43329|issue:45600", + "issue:45084|issue:45538", + "issue:44060|issue:44368", + "issue:43452|issue:44991", + "issue:43526|issue:43697", + "issue:43257|issue:43329", + "issue:43526|issue:44554", + "issue:43906|issue:46209", + "issue:36010|issue:36331", + "issue:43819|issue:43927", + "issue:36010|issue:43645", + "issue:43531|issue:45042", + "issue:42175|issue:46291", + "issue:43638|issue:44936", + "issue:42490|issue:43257", + "issue:43720|issue:44610" + ], + "prepared_review_unit_hash": "ac181fda98c744b47840a70625f63e9ada62b896795779db17107b0c8daad113", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36528, + "estimated_input_tokens": 9132, + "estimated_eval_tokens": 18520 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is highly heterogeneous: the soft-similarity links mostly connect unrelated model-, tokenizer-, backend-, or training-specific regressions. I would not merge any of the suggested pairs; they do not look like the same underlying bug or change.", + "confidence": 0.81, + "canonical_issue_reason": "issue:45020 is the broadest and most central report in the set ('Recent transformers versions break models using `remote_code`'), so it is the best anchor if one must pick a representative issue. It is still not a duplicate of the others, but it best matches the general theme of upstream loading regressions.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the strongest cluster representative because it is a broadly scoped, user-facing regression with enough generality to sit near other loading/import failures, even though the rest of the items are not duplicates of it.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:34634", + "right": "issue:36331", + "accept": false, + "reason": "Different subsystems and failure modes: BarkProcessor voice_preset behavior vs CustomTrainer compute_loss signature mismatch." + }, + { + "left": "issue:45003", + "right": "issue:46285", + "accept": false, + "reason": "Both involve modeling internals, but one is unsafe sys.modules access and the other is a CLIP wrapper structure inconsistency; not the same bug." + }, + { + "left": "issue:46697", + "right": "issue:46728", + "accept": false, + "reason": "Both are image-processor sizing issues, but they affect different models and different calculations (padding vs patch-count logic)." + }, + { + "left": "issue:43653", + "right": "issue:44912", + "accept": false, + "reason": "Tokenizer special-token registration is unrelated to MXFP4 quantization fallback during loading." + }, + { + "left": "issue:43782", + "right": "issue:45325", + "accept": false, + "reason": "Different VL model bugs: one is a weight_only load error, the other is a RoPE / temporal position scaling issue." + }, + { + "left": "issue:43262", + "right": "issue:43819", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting is unrelated to DAC latent reconstruction / STE mismatch." + }, + { + "left": "issue:45081", + "right": "issue:45406", + "accept": false, + "reason": "Both involve Mistral/Gemma processor loading, but one is a regex patch crash and the other is a missing _tokenizer attribute; different code paths." + }, + { + "left": "issue:43582", + "right": "issue:44857", + "accept": false, + "reason": "A TypeError in caching_allocator_warmup is not the same issue as LwDetrImageLoss crashing under float16 AMP." + }, + { + "left": "issue:45372", + "right": "issue:46326", + "accept": false, + "reason": "Importing ReasoningEffort from mistral_common and continue_final_message behavior are distinct processor/tokenizer problems." + }, + { + "left": "issue:44589", + "right": "issue:45042", + "accept": false, + "reason": "Missing Float8 storage support and an erroneous torchvision dependency are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:44912", + "accept": false, + "reason": "torch.compile with SDPA on Bamba is a different failure mode from MXFP4 quantization fallback on git-oss-20b." + }, + { + "left": "issue:43742", + "right": "issue:45020", + "accept": false, + "reason": "A key error loading MobileLLM-125M is a specific model/config issue, whereas 45020 is a broader remote_code regression; not the same concrete bug." + }, + { + "left": "issue:43329", + "right": "issue:45600", + "accept": false, + "reason": "Undefined video-token helper variables in multimodal token counting is unrelated to stale auto_mappings references." + }, + { + "left": "issue:45084", + "right": "issue:45538", + "accept": false, + "reason": "A template compilation TypeError and CLIPTokenizer's model_max_length bug are unrelated." + }, + { + "left": "issue:44060", + "right": "issue:44368", + "accept": false, + "reason": "Both mention tied-weights warnings, but they involve different models, triggers, and warning behavior; not clearly the same underlying defect." + }, + { + "left": "issue:43452", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer/model loading failures, but one concerns gguf_file handling and the other a specific tokenizer incompatibility for EMBEDDIA/est-roberta." + }, + { + "left": "issue:43526", + "right": "issue:43697", + "accept": false, + "reason": "BeitImageProcessorFast label reduction and RTDetrV2 output divergence are different model-specific bugs." + }, + { + "left": "issue:43257", + "right": "issue:43329", + "accept": false, + "reason": "Qwen3 MoE weight conversion with accelerate/deepspeed is unrelated to multimodal token counting for video processors." + }, + { + "left": "issue:43526", + "right": "issue:44554", + "accept": false, + "reason": "These are unrelated: image-processor label reduction vs an MPS attention correctness issue." + }, + { + "left": "issue:43906", + "right": "issue:46209", + "accept": false, + "reason": "An isolated reproduction of an older issue is not the same as an FP8 inference failure on B200; different hardware and bug class." + }, + { + "left": "issue:36010", + "right": "issue:36331", + "accept": false, + "reason": "ImportError for GenerationMixin and a Trainer signature change are separate breakages." + }, + { + "left": "issue:43819", + "right": "issue:43927", + "accept": false, + "reason": "DAC latent mismatch and DiaConfig losing custom token IDs are different save/load vs generation bugs." + }, + { + "left": "issue:36010", + "right": "issue:43645", + "accept": false, + "reason": "One is a transformers import regression; the other is a Jupyter custom-model initialization breakage in Transformers 5.0.0." + }, + { + "left": "issue:43531", + "right": "issue:45042", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior is unrelated to PIL backend processors requiring torchvision." + }, + { + "left": "issue:42175", + "right": "issue:46291", + "accept": false, + "reason": "Package install/backend dependency omission is not the same as kernels v0.15 requiring version/revision metadata." + }, + { + "left": "issue:43638", + "right": "issue:44936", + "accept": false, + "reason": "A deepspeed zero3/non-pretrained Bert indexing error is unrelated to trainer.evaluate() failing after trainer.train()." + }, + { + "left": "issue:42490", + "right": "issue:43257", + "accept": false, + "reason": "Local-path import warning with trailing slash and Qwen3 MoE weight conversion are unrelated." + }, + { + "left": "issue:43720", + "right": "issue:44610", + "accept": false, + "reason": "Packed-weight unpacking during accelerate loading is unrelated to OmDet-Turbo processor/model input-size mismatch." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 21, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:28282", + "issue:29159", + "issue:29870", + "issue:34567", + "issue:38175", + "issue:39692", + "issue:40990", + "issue:42915", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43450", + "issue:43454", + "issue:43525", + "issue:43540", + "issue:43577", + "issue:43606", + "issue:43643", + "issue:43650", + "issue:43696", + "issue:43742", + "issue:43746", + "issue:43749", + "issue:43854", + "issue:43906", + "issue:43927", + "issue:43931", + "issue:44291", + "issue:44387", + "issue:44479", + "issue:44568", + "issue:44610", + "issue:44617", + "issue:44756", + "issue:44811", + "issue:44821", + "issue:44912", + "issue:44977", + "issue:45042", + "issue:45356", + "issue:45459", + "issue:45759", + "issue:45803", + "issue:45859", + "issue:46291", + "issue:46421", + "issue:46424", + "issue:46701" + ], + "soft_pairs": [ + "issue:39692|issue:44912", + "issue:43854|issue:44977", + "issue:43643|issue:45759", + "issue:43746|issue:44811", + "issue:38175|issue:44610", + "issue:43257|issue:43696", + "issue:43262|issue:44811", + "issue:44387|issue:44617", + "issue:43257|issue:43931", + "issue:45459|issue:45803", + "issue:43329|issue:43450", + "issue:46421|issue:46424", + "issue:43329|issue:44479", + "issue:42915|issue:43742", + "issue:43749|issue:44756", + "issue:34567|issue:40990", + "issue:44291|issue:45042", + "issue:43650|issue:43906", + "issue:42915|issue:44821", + "issue:29159|issue:34567", + "issue:43606|issue:45859", + "issue:43927|issue:44912", + "issue:43540|issue:46701", + "issue:28282|issue:29870", + "issue:43454|issue:43577", + "issue:43525|issue:46291", + "issue:44568|issue:45356" + ], + "prepared_review_unit_hash": "d27fc01cb6871579ab1f3cc4dd94254c3e2790bbc88cbcbe24ff997e5a6bf86b", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 36011, + "estimated_input_tokens": 9003, + "estimated_eval_tokens": 18262 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous; I don\u2019t see any true duplicate pairs among the soft-edge candidates. The only partial theme is multimodal/video regressions, but even there the reports point to different concrete failures.", + "confidence": 0.24, + "canonical_issue_reason": "issue:44479 is the broadest, highest-signal report in the set: a concrete video-input regression affecting multiple Qwen VL/MoE variants, making it the best anchor for the small multimodal/video subtheme.", + "canonical_pr_reason": null, + "best_issue_reason": "Same as canonical: it is the clearest, most representative bug report among otherwise unrelated issues, with a specific regression title and multiple affected model families.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:39692", + "right": "issue:44912", + "accept": false, + "reason": "Different failures: SigLIP2 doc example/model-processor mismatch vs GPT-OSS MXFP4 quantization fallback." + }, + { + "left": "issue:43854", + "right": "issue:44977", + "accept": false, + "reason": "Different model/code paths: GLM-4.7 loading in tests vs Qwen3.5 flash-attention generation." + }, + { + "left": "issue:43643", + "right": "issue:45759", + "accept": false, + "reason": "Both involve config/loading behavior, but one is missing fields from trust_remote_code and the other is text_config unwrapping for composite Qwen multimodal configs; not the same bug." + }, + { + "left": "issue:43746", + "right": "issue:44811", + "accept": false, + "reason": "Different components: GraniteSpeech PEFT local checkpoint loading vs Whisper batch_decode skip_special_tokens handling." + }, + { + "left": "issue:38175", + "right": "issue:44610", + "accept": false, + "reason": "Different symptoms and models: unexpected zero probabilities in SigLIP2 vs OmDet-Turbo processor/model input-size mismatch." + }, + { + "left": "issue:43257", + "right": "issue:43696", + "accept": false, + "reason": "Qwen3 MoE weight conversion/loading issue vs GPT-oss CUDA OOM; unrelated root causes." + }, + { + "left": "issue:43262", + "right": "issue:44811", + "accept": false, + "reason": "Audio chat-template sampling-rate default vs Whisper batch_decode token-decoding behavior; different bugs." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Both are OOM-related, but one is a Transformers 5.x int4 reserved-memory regression and the other is a Sam3Video model OOM; not the same code-path problem." + }, + { + "left": "issue:43257", + "right": "issue:43931", + "accept": false, + "reason": "Different Qwen model loading failures: MoE weight conversion under accelerate/deepspeed vs Qwen3-VL weight-shape mismatch." + }, + { + "left": "issue:45459", + "right": "issue:45803", + "accept": false, + "reason": "Both mention bad exception handling, but they are different functions and different failure modes (protobuf decode masking vs bare except in Fuyu tensor conversion)." + }, + { + "left": "issue:43329", + "right": "issue:43450", + "accept": false, + "reason": "Both involve video/multimodal processing, but one is undefined helpers in multimodal token counting and the other is batched video shape handling; insufficient evidence of the same bug." + }, + { + "left": "issue:46421", + "right": "issue:46424", + "accept": false, + "reason": "Both are cache-related, but one is lazy-init recompilation and the other is max_cache_length being ignored except for Quantized Cache." + }, + { + "left": "issue:43329", + "right": "issue:44479", + "accept": false, + "reason": "Related to video inputs, but one is undefined helper variables in _get_num_multimodal_tokens and the other is a broader v5.3.0 regression across several Qwen VL models." + }, + { + "left": "issue:42915", + "right": "issue:43742", + "accept": false, + "reason": "Different failures: FineGrainedFP8Config support in Qwen3MoE vs key error loading MobileLLM-125M." + }, + { + "left": "issue:43749", + "right": "issue:44756", + "accept": false, + "reason": "Different memory/loading issues: FSDP CPU RAM efficient loading vs mmap on Strix Halo causing OOM." + }, + { + "left": "issue:34567", + "right": "issue:40990", + "accept": false, + "reason": "Unrelated training-state counter bug vs high perplexity on GPT-oss-20b." + }, + { + "left": "issue:44291", + "right": "issue:45042", + "accept": false, + "reason": "Different unrelated regressions: init_empty_weights/_is_hf_initialized vs PIL image processors requiring torchvision." + }, + { + "left": "issue:43650", + "right": "issue:43906", + "accept": false, + "reason": "No clear overlap: one is an ambiguous 'ADD THE DATA' issue and the other is a reproduction of a different upstream bug." + }, + { + "left": "issue:42915", + "right": "issue:44821", + "accept": false, + "reason": "Different areas: Qwen3MoE FP8 config vs AutoImageProcessor loading from URL." + }, + { + "left": "issue:29159", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer slow/fast inconsistency is unrelated to TrainerState.num_input_tokens_seen not updating." + }, + { + "left": "issue:43606", + "right": "issue:45859", + "accept": false, + "reason": "Both mention CPU offload, but they affect different models/components and don\u2019t look like the same underlying fix." + }, + { + "left": "issue:43927", + "right": "issue:44912", + "accept": false, + "reason": "Different bugs: DiaConfig token IDs lost on save/load vs GPT-oss MXFP4 quantization fallback." + }, + { + "left": "issue:43540", + "right": "issue:46701", + "accept": false, + "reason": "Both are multimodal inference issues, but one is Qwen3OmniMoe video-input processing and the other is DiffusionGemma image-text-to-text pipeline behavior." + }, + { + "left": "issue:28282", + "right": "issue:29870", + "accept": false, + "reason": "Completely different problems: missing PyTorch ImportError vs Bart generate() unbound decoder_attention_mask." + }, + { + "left": "issue:43454", + "right": "issue:43577", + "accept": false, + "reason": "Different model bugs: AyaVision weight tying vs Blip2 dtype propagation." + }, + { + "left": "issue:43525", + "right": "issue:46291", + "accept": false, + "reason": "Unrelated: missing pad_token_id on Llama4Config vs kernels dependency/version import failure." + }, + { + "left": "issue:44568", + "right": "issue:45356", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different models and different tokenization/codecs behavior." + } + ] + }, + "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": 23, + "review_unit_index": 22, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:24643", + "issue:27957", + "issue:28282", + "issue:29127", + "issue:29159", + "issue:33357", + "issue:34567", + "issue:34689", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43404", + "issue:43425", + "issue:43452", + "issue:43540", + "issue:43577", + "issue:43582", + "issue:43697", + "issue:43742", + "issue:43749", + "issue:43824", + "issue:44062", + "issue:44291", + "issue:44479", + "issue:44488", + "issue:44560", + "issue:44568", + "issue:44792", + "issue:44849", + "issue:44857", + "issue:44936", + "issue:44964", + "issue:44987", + "issue:44991", + "issue:45127", + "issue:45137", + "issue:45357", + "issue:45405", + "issue:45488", + "issue:45538", + "issue:45701", + "issue:45907", + "issue:46291", + "issue:46512", + "issue:46693", + "issue:46821", + "issue:46832", + "issue:46869" + ], + "soft_pairs": [ + "issue:44987|issue:45538", + "issue:28282|issue:34689", + "issue:45538|issue:46869", + "issue:44857|issue:46832", + "issue:43697|issue:45357", + "issue:43425|issue:46291", + "issue:33357|issue:34567", + "issue:43404|issue:45127", + "issue:42915|issue:45538", + "issue:45137|issue:45405", + "issue:42175|issue:43824", + "issue:28282|issue:29127", + "issue:44062|issue:44991", + "issue:44479|issue:46693", + "issue:43742|issue:44291", + "issue:44936|issue:46821", + "issue:43577|issue:46512", + "issue:42915|issue:43749", + "issue:44849|issue:45701", + "issue:43452|issue:43582", + "issue:24643|issue:29159", + "issue:27957|issue:28282", + "issue:44987|issue:45907", + "issue:41720|issue:43540", + "issue:42175|issue:44488", + "issue:43540|issue:44560", + "issue:44568|issue:45488", + "issue:44792|issue:44964" + ], + "prepared_review_unit_hash": "f49754b0ed942496412476e5df6819e66bafa067b27eec344e56f4870677dd70", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36019, + "estimated_input_tokens": 9005, + "estimated_eval_tokens": 18266 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a broad mix of mostly unrelated Transformers issues; many are in adjacent areas like tokenization, model loading, vision/video regressions, and dtype handling, but the soft pairs do not look like true duplicates of the same concrete bug.", + "confidence": 0.88, + "canonical_issue_reason": "issue:45488 is the clearest, most central representative of the tokenizer-behavior regression theme in this cluster and has the strongest cross-model impact.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45488 is the best global issue candidate because it is specific enough to be actionable yet broad enough to represent the recurring tokenizer/version regression pattern seen here.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44987", + "right": "issue:45538", + "accept": false, + "reason": "Different bugs: one is a model load failure for a specific repo, the other is CLIPTokenizer max_length behavior." + }, + { + "left": "issue:28282", + "right": "issue:34689", + "accept": false, + "reason": "Unrelated failure modes: missing PyTorch import vs a Llama Vision model-loading regression." + }, + { + "left": "issue:45538", + "right": "issue:46869", + "accept": false, + "reason": "Tokenizer max_length bug vs YOLO auto device-map failure are unrelated." + }, + { + "left": "issue:44857", + "right": "issue:46832", + "accept": false, + "reason": "Both are object-detection related, but the concrete code paths differ: loss/AMP crash vs forward shape/feature-level mismatch." + }, + { + "left": "issue:43697", + "right": "issue:45357", + "accept": false, + "reason": "Different models and symptoms: RT-DETR output drift vs Qwen3.5 save_pretrained key naming." + }, + { + "left": "issue:43425", + "right": "issue:46291", + "accept": false, + "reason": "Separate dependency issues: Torch compatibility vs kernels revision/version requirement." + }, + { + "left": "issue:33357", + "right": "issue:34567", + "accept": false, + "reason": "Mac bus error on CLIP model load is unrelated to TrainerState token counting." + }, + { + "left": "issue:43404", + "right": "issue:45127", + "accept": false, + "reason": "Weight-tying bug in Mistral3 is not the same as LoRA merge collapse with extended vocab." + }, + { + "left": "issue:42915", + "right": "issue:45538", + "accept": false, + "reason": "FP8 config failure for Qwen3Moe is unrelated to CLIPTokenizer max_length." + }, + { + "left": "issue:45137", + "right": "issue:45405", + "accept": false, + "reason": "DeepSpeed deque error and unreleased PEFT version bump are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:43824", + "accept": false, + "reason": "Packaging/backend installation issue vs missing model class import are different problems." + }, + { + "left": "issue:28282", + "right": "issue:29127", + "accept": false, + "reason": "Missing PyTorch import and LayoutLMv3 box-validation messaging are not the same bug." + }, + { + "left": "issue:44062", + "right": "issue:44991", + "accept": false, + "reason": "AddedToken keyword collision is not evidenced as the same issue as est-roberta tokenizer loading." + }, + { + "left": "issue:44479", + "right": "issue:46693", + "accept": false, + "reason": "Video-input regression and FA2 long-context performance regression are distinct changes." + }, + { + "left": "issue:43742", + "right": "issue:44291", + "accept": false, + "reason": "MobileLLM key error and init_empty_weights argument mismatch are different loading failures." + }, + { + "left": "issue:44936", + "right": "issue:46821", + "accept": false, + "reason": "Trainer evaluate/train state issue is unrelated to TimesFM window_size attribute handling." + }, + { + "left": "issue:43577", + "right": "issue:46512", + "accept": false, + "reason": "Both involve dtype consistency, but one is submodule weights staying float32 while the other is stale config dtype; not the same concrete bug." + }, + { + "left": "issue:42915", + "right": "issue:43749", + "accept": false, + "reason": "Qwen3Moe FP8 failure and FSDP CPU RAM efficient loading are unrelated." + }, + { + "left": "issue:44849", + "right": "issue:45701", + "accept": false, + "reason": "Qwen3.5 hidden-state bug is not the same as version-dependent tokenization changes." + }, + { + "left": "issue:43452", + "right": "issue:43582", + "accept": false, + "reason": "gguf_file loading breakage and Apple Silicon allocator TypeError are unrelated." + }, + { + "left": "issue:24643", + "right": "issue:29159", + "accept": false, + "reason": "DeepSpeed weight-dimension runtime error and tokenizer slow/fast inconsistency are different subsystems." + }, + { + "left": "issue:27957", + "right": "issue:28282", + "accept": false, + "reason": "Flash Attention 2 with XLM-Roberta is unrelated to missing PyTorch import." + }, + { + "left": "issue:44987", + "right": "issue:45907", + "accept": false, + "reason": "Model loading failure for a specific repo and list_repo_files revision forwarding are different bugs." + }, + { + "left": "issue:41720", + "right": "issue:43540", + "accept": false, + "reason": "Auto device-map CUDA assert for Qwen3 and video-input processing error for Qwen3OmniMoe are not the same issue." + }, + { + "left": "issue:42175", + "right": "issue:44488", + "accept": false, + "reason": "Backend install packaging bug vs model-loading issue for cjvt/sleng-bert are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44560", + "accept": false, + "reason": "Both mention video, but they affect different Qwen models and failure points; too broad to merge." + }, + { + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "Different tokenizer regressions: special-token addition vs pre-tokenizer override causing DeepSeek breakage." + }, + { + "left": "issue:44792", + "right": "issue:44964", + "accept": false, + "reason": "Janus test failure and Phi-4 multimodal loading failure are distinct multimodal 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": 23, + "review_unit_index": 23, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:24643", + "issue:29870", + "issue:34567", + "issue:36331", + "issue:38175", + "issue:43257", + "issue:43452", + "issue:43531", + "issue:43540", + "issue:43577", + "issue:43582", + "issue:43697", + "issue:43716", + "issue:43742", + "issue:43756", + "issue:43906", + "issue:44062", + "issue:44291", + "issue:44479", + "issue:44488", + "issue:44568", + "issue:44589", + "issue:44625", + "issue:44737", + "issue:44821", + "issue:44877", + "issue:44898", + "issue:45042", + "issue:45072", + "issue:45081", + "issue:45325", + "issue:45357", + "issue:45381", + "issue:45405", + "issue:45600", + "issue:45672", + "issue:45674", + "issue:45823", + "issue:45920", + "issue:46399", + "issue:46463", + "issue:46468", + "issue:46491", + "issue:46693", + "issue:46701", + "issue:46821", + "issue:46829" + ], + "soft_pairs": [ + "issue:44479|issue:45381", + "issue:38175|issue:44898", + "issue:44291|issue:44821", + "issue:43540|issue:44479", + "issue:34567|issue:36331", + "issue:43257|issue:43756", + "issue:44062|issue:45081", + "issue:45042|issue:45081", + "issue:43582|issue:46821", + "issue:45672|issue:45674", + "issue:43742|issue:44589", + "issue:43756|issue:46399", + "issue:43756|issue:45325", + "issue:44568|issue:46491", + "issue:45405|issue:45600", + "issue:43716|issue:46701", + "issue:43577|issue:45072", + "issue:43531|issue:44821", + "issue:43906|issue:46468", + "issue:24643|issue:29870", + "issue:44737|issue:46463", + "issue:43906|issue:46693", + "issue:45823|issue:46463", + "issue:43452|issue:44488", + "issue:43697|issue:46829", + "issue:44625|issue:44877", + "issue:45357|issue:45920" + ], + "prepared_review_unit_hash": "1338f770209de7cc1ef8462118c4b0a36c595467f89931a000f7e54008e4b78c", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35602, + "estimated_input_tokens": 8901, + "estimated_eval_tokens": 18058 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Only one soft pair looks like the same underlying bug: the Qwen2.5-VL/Qwen3-VL video position-id regression. The rest are mostly topic-level similarity (tokenizers, dtype, multimodal, training, loading) but appear to be different concrete failures.", + "confidence": 0.79, + "canonical_issue_reason": "issue:44479 is the best issue anchor because it describes the broad video-input regression and is the more general formulation of the only likely duplicate pair; issue:45381 looks like a narrower reproduction of the same problem.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44479 is the strongest representative issue in the cluster: it is broader, more user-facing, and closest to the one plausible duplicate relationship in the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44479", + "right": "issue:45381", + "accept": true, + "reason": "Both report the same Qwen2.5-VL/Qwen3-VL video-input position-id regression; 45381 looks like a narrower reproduction of the broader 44479 bug." + }, + { + "left": "issue:38175", + "right": "issue:44898", + "accept": false, + "reason": "Both are vision-model correctness issues, but they involve different models and failure modes (SigLIP zero probabilities vs Perceiver resizing/interpolation)." + }, + { + "left": "issue:44291", + "right": "issue:44821", + "accept": false, + "reason": "Different loading failures: one is an init_empty_weights/_is_hf_initialized argument mismatch, the other is loading AutoImageProcessor from a URL." + }, + { + "left": "issue:43540", + "right": "issue:44479", + "accept": false, + "reason": "Both involve video inputs, but they target different model families and likely different code paths (Qwen3OmniMoe vs Qwen2.5-VL/Qwen3-VL)." + }, + { + "left": "issue:34567", + "right": "issue:36331", + "accept": false, + "reason": "Both touch Trainer behavior, but one is token accounting and the other is a compute_loss API signature break." + }, + { + "left": "issue:43257", + "right": "issue:43756", + "accept": false, + "reason": "Unrelated model-loading/weight-conversion issue versus a RoPE-layer-count bug in Smollm3." + }, + { + "left": "issue:44062", + "right": "issue:45081", + "accept": false, + "reason": "Both are tokenizer bugs, but the concrete failures differ: AddedToken keyword collision vs Mistral regex patch accessing backend_tokenizer." + }, + { + "left": "issue:45042", + "right": "issue:45081", + "accept": false, + "reason": "Completely different subsystems: PIL image processors requiring torchvision vs a Mistral tokenizer regex patch crash." + }, + { + "left": "issue:43582", + "right": "issue:46821", + "accept": false, + "reason": "Apple Silicon allocator warmup TypeError is unrelated to the TimesFM window_size AttributeError." + }, + { + "left": "issue:45672", + "right": "issue:45674", + "accept": false, + "reason": "Both are 4-bit inference problems, but one is a uint8 torch.finfo() error in Gemma4 audio modules and the other is a BitsAndBytesConfig skip-module regression." + }, + { + "left": "issue:43742", + "right": "issue:44589", + "accept": false, + "reason": "Different model-load failures: MobileLLM key error vs missing Float8 storage type." + }, + { + "left": "issue:43756", + "right": "issue:46399", + "accept": false, + "reason": "Both mention RoPE/rotary concepts, but one is a missing RoPE layers bug and the other is a kernel registration decorator crash." + }, + { + "left": "issue:43756", + "right": "issue:45325", + "accept": false, + "reason": "Different rotary-position bugs in different models: missing RoPE layers vs incorrect still-image temporal scaling in Qwen2.5-VL." + }, + { + "left": "issue:44568", + "right": "issue:46491", + "accept": false, + "reason": "Different tokenizer regressions: missing BOS/EOS insertion vs leading-space stripping on encode/decode." + }, + { + "left": "issue:45405", + "right": "issue:45600", + "accept": false, + "reason": "A dependency-version bump issue and a removed-config reference in auto_mappings are not the same bug." + }, + { + "left": "issue:43716", + "right": "issue:46701", + "accept": false, + "reason": "Both are multimodal, but one is a dtype mismatch between preprocessor and model; the other is a pipeline/output wiring failure." + }, + { + "left": "issue:43577", + "right": "issue:45072", + "accept": false, + "reason": "Both mention dtype mismatches, but they affect different architectures and likely different conversion/loading paths." + }, + { + "left": "issue:43531", + "right": "issue:44821", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to AutoImageProcessor URL loading." + }, + { + "left": "issue:43906", + "right": "issue:46468", + "accept": false, + "reason": "One is only an isolated reproduction of issue 38071; the other is a broken official Nemotron example. Not enough evidence they are the same underlying bug." + }, + { + "left": "issue:24643", + "right": "issue:29870", + "accept": false, + "reason": "Training-with-Deepspeed RuntimeError and BART generate() UnboundLocalError are unrelated failures." + }, + { + "left": "issue:44737", + "right": "issue:46463", + "accept": false, + "reason": "XLNet CPU placement bug and T5 decoder-input validation error are different code paths and symptoms." + }, + { + "left": "issue:43906", + "right": "issue:46693", + "accept": false, + "reason": "Isolated reproduction of an older issue is not the same as a FA2 long-context performance regression." + }, + { + "left": "issue:45823", + "right": "issue:46463", + "accept": false, + "reason": "Gemma4 device-map mismatch is unrelated to T5 decoder input validation." + }, + { + "left": "issue:43452", + "right": "issue:44488", + "accept": false, + "reason": "Both are model-loading complaints, but one is specifically about gguf_file support and the other is a separate sleng-bert loading failure." + }, + { + "left": "issue:43697", + "right": "issue:46829", + "accept": false, + "reason": "Different model families and outputs: RTDetrV2 forward differences vs VideoPrism output field misassignment." + }, + { + "left": "issue:44625", + "right": "issue:44877", + "accept": false, + "reason": "Different config issues: Qwen3.5 num_labels propagation vs strict config blocking granite_speech loading." + }, + { + "left": "issue:45357", + "right": "issue:45920", + "accept": false, + "reason": "One is a Qwen3.5 save_pretrained visual-encoder key regression; the other is a tokenizer ID regression across several models." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The accepted pair (issue:44479 / issue:45381) is well-supported by the titles and scope difference, and the rejected pairs are appropriately framed as topic similarity rather than the same bug. No overstatement stands out." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 24, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:28282", + "issue:32090", + "issue:38175", + "issue:41762", + "issue:43295", + "issue:43452", + "issue:43525", + "issue:43526", + "issue:43540", + "issue:43643", + "issue:43653", + "issue:43688", + "issue:43697", + "issue:43720", + "issue:43749", + "issue:43761", + "issue:43819", + "issue:43994", + "issue:44361", + "issue:44442", + "issue:44466", + "issue:44493", + "issue:44737", + "issue:44811", + "issue:44843", + "issue:44898", + "issue:44913", + "issue:44936", + "issue:45005", + "issue:45200", + "issue:45216", + "issue:45356", + "issue:45372", + "issue:45375", + "issue:45479", + "issue:45542", + "issue:45588", + "issue:45663", + "issue:46050", + "issue:46121", + "issue:46170", + "issue:46178", + "issue:46463", + "issue:46693", + "issue:46701", + "issue:46829", + "issue:46846", + "issue:46899" + ], + "soft_pairs": [ + "issue:45663|issue:46846", + "issue:43540|issue:46693", + "issue:44913|issue:46121", + "issue:44737|issue:45588", + "issue:28282|issue:32090", + "issue:46463|issue:46899", + "issue:43697|issue:45216", + "issue:43526|issue:45479", + "issue:43720|issue:44898", + "issue:43643|issue:45375", + "issue:43761|issue:46829", + "issue:45200|issue:46463", + "issue:44466|issue:45005", + "issue:44361|issue:44442", + "issue:44493|issue:46050", + "issue:43653|issue:44361", + "issue:41762|issue:44936", + "issue:43452|issue:44843", + "issue:43819|issue:44811", + "issue:43688|issue:43749", + "issue:38175|issue:43295", + "issue:43994|issue:44898", + "issue:43525|issue:46701", + "issue:45372|issue:46170", + "issue:45542|issue:46178", + "issue:43295|issue:45356" + ], + "prepared_review_unit_hash": "76a9d14727f863f01a171768c685cf57f9c0e3e4349e7595346fba0bb2df2293", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 26, + "serialized_chars": 36196, + "estimated_input_tokens": 9049, + "estimated_eval_tokens": 18354 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is highly heterogeneous: it mixes unrelated bugs in tokenizers, model configs, generation, quantization, distributed loading, and backend detection. None of the soft-edge pairs look like the same underlying bug or change, so all are rejected as duplicates.", + "confidence": 0.92, + "canonical_issue_reason": "Issue 45005 is the broadest and most representative entry here: it describes a v5 tied-weights regression with wider model-family impact, making it the best umbrella candidate despite the cluster\u2019s overall heterogeneity.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45005 is the clearest and most general issue in the set, with a concrete regression description and broad applicability across translation/tied-weight models.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45663", + "right": "issue:46846", + "accept": false, + "reason": "Different model families and failure modes: Gemma4 FSDP2 layer state bug vs Qwen3.5 TP plan OOM/shape error." + }, + { + "left": "issue:43540", + "right": "issue:46693", + "accept": false, + "reason": "Unrelated code paths: Qwen3OmniMoe video-input validation vs FA2 long-context performance regression." + }, + { + "left": "issue:44913", + "right": "issue:46121", + "accept": false, + "reason": "Both touch RoPE/rotation config handling, but one is reload persistence of a config field and the other is validation/type conversion for a JSON-loaded list." + }, + { + "left": "issue:44737", + "right": "issue:45588", + "accept": false, + "reason": "XLNet CPU device placement bug and flash_attention crash on missing s_aux are unrelated implementation defects." + }, + { + "left": "issue:28282", + "right": "issue:32090", + "accept": false, + "reason": "Import-time PyTorch missing error vs Trainer GPU broadcast TypeError; completely different surfaces and causes." + }, + { + "left": "issue:46463", + "right": "issue:46899", + "accept": false, + "reason": "T5 decoder-input validation error message vs Gemma4 multimodal dtype-cast quantization bug; not the same problem." + }, + { + "left": "issue:43697", + "right": "issue:45216", + "accept": false, + "reason": "RTDetrV2 output drift in v5 and Qwen3.5 save_pretrained regression are different models and code paths." + }, + { + "left": "issue:43526", + "right": "issue:45479", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug is unrelated to the single-label-classification zero-loss issue." + }, + { + "left": "issue:43720", + "right": "issue:44898", + "accept": false, + "reason": "BitNet packed-weight loading bug vs Perceiver non-default-resolution interpolation failure; different subsystems and symptoms." + }, + { + "left": "issue:43643", + "right": "issue:45375", + "accept": false, + "reason": "Generic trust_remote_code config reconstruction issue vs a specific Qwen3_5MoeVisionConfig field being dropped; related only at a very high level." + }, + { + "left": "issue:43761", + "right": "issue:46829", + "accept": false, + "reason": "Both concern output fields, but CLIPVision hidden_states regression and VideoPrism hidden_states misassignment are different model implementations." + }, + { + "left": "issue:45200", + "right": "issue:46463", + "accept": false, + "reason": "Gemma4 missing default mm_token_type_ids and T5 decoder-input error messaging are unrelated bugs." + }, + { + "left": "issue:44466", + "right": "issue:45005", + "accept": false, + "reason": "Both mention tied weights, but one is device-dependent lm_head serialization while the other is a broader translation-model regression without the same concrete failure mode." + }, + { + "left": "issue:44361", + "right": "issue:44442", + "accept": false, + "reason": "Two different tokenizer load/task failures for different tokenizers; same broad area, not the same bug." + }, + { + "left": "issue:44493", + "right": "issue:46050", + "accept": false, + "reason": "Position-id key warnings and sinq save/load breakage are distinct model-state issues." + }, + { + "left": "issue:43653", + "right": "issue:44361", + "accept": false, + "reason": "BigBird mask-token special-token registration and MLuke task AttributeError are unrelated tokenizer defects." + }, + { + "left": "issue:41762", + "right": "issue:44936", + "accept": false, + "reason": "Gemma3 ZeRO-3 load-time index error and trainer.evaluate() after train() failure have no shared concrete code path." + }, + { + "left": "issue:43452", + "right": "issue:44843", + "accept": false, + "reason": "GGUF from_pretrained handling and an offline model_info call in mistral regex patch are different loading bugs." + }, + { + "left": "issue:43819", + "right": "issue:44811", + "accept": false, + "reason": "DAC latent conversion mismatch and Whisper batch_decode skip_special_tokens handling are unrelated audio/model API issues." + }, + { + "left": "issue:43688", + "right": "issue:43749", + "accept": false, + "reason": "Aux-loss normalization in MoE models vs FSDP CPU RAM efficient loading failure are different problems." + }, + { + "left": "issue:38175", + "right": "issue:43295", + "accept": false, + "reason": "Siglip2 zero probabilities and processor.tokenizer regression are separate multimodal issues." + }, + { + "left": "issue:43994", + "right": "issue:44898", + "accept": false, + "reason": "Siglip2 nonsensical outputs and Perceiver interpolation failure are different model-specific regressions." + }, + { + "left": "issue:43525", + "right": "issue:46701", + "accept": false, + "reason": "Llama4Config missing pad_token_id and DiffusionGemma image-text pipeline breakage are unrelated." + }, + { + "left": "issue:45372", + "right": "issue:46170", + "accept": false, + "reason": "Gemma4 processor import dependency issue and Windows ROCm torch.distributed dependency issue are different environment/import failures." + }, + { + "left": "issue:45542", + "right": "issue:46178", + "accept": false, + "reason": "TensorFlow backend detection error and tensorflow_text backend mapping error are separate backend-registration problems." + }, + { + "left": "issue:43295", + "right": "issue:45356", + "accept": false, + "reason": "Processor.tokenizer regression and Kimi-K2.5 codec/fix_mistral_regex warning regression do not share the same concrete bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 25, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:29159", + "issue:32090", + "issue:33357", + "issue:33666", + "issue:36010", + "issue:41720", + "issue:42898", + "issue:43452", + "issue:43504", + "issue:43531", + "issue:43550", + "issue:43577", + "issue:43643", + "issue:43646", + "issue:43653", + "issue:43688", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43756", + "issue:43927", + "issue:44464", + "issue:44466", + "issue:44479", + "issue:44484", + "issue:44554", + "issue:44561", + "issue:44568", + "issue:44625", + "issue:44704", + "issue:44734", + "issue:44743", + "issue:44821", + "issue:44843", + "issue:44977", + "issue:44991", + "issue:45440", + "issue:45491", + "issue:45701", + "issue:45704", + "issue:46002", + "issue:46050", + "issue:46402", + "issue:46491", + "issue:46612", + "issue:46693", + "issue:46821" + ], + "soft_pairs": [ + "issue:43531|issue:43742", + "issue:42898|issue:46491", + "issue:29159|issue:33357", + "issue:43653|issue:44568", + "issue:43646|issue:43749", + "issue:43452|issue:43577", + "issue:43756|issue:46402", + "issue:44843|issue:44991", + "issue:43688|issue:45704", + "issue:43653|issue:43927", + "issue:44464|issue:44743", + "issue:43688|issue:45701", + "issue:44484|issue:44704", + "issue:45491|issue:46002", + "issue:44479|issue:44625", + "issue:44821|issue:46821", + "issue:43688|issue:45440", + "issue:33666|issue:36010", + "issue:32090|issue:33666", + "issue:44734|issue:46612", + "issue:44466|issue:46050", + "issue:43504|issue:43720", + "issue:44977|issue:46693", + "issue:43504|issue:43550", + "issue:44464|issue:44977", + "issue:41720|issue:44464", + "issue:43643|issue:44561", + "issue:43577|issue:44554" + ], + "prepared_review_unit_hash": "5a5fbcedb9929434bb86ede3e43c9d1fd73d03fe8dd9c41c9ebc890ce8b0547b", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35543, + "estimated_input_tokens": 8886, + "estimated_eval_tokens": 18028 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the soft pairs span different models, APIs, and failure modes, so none look like true duplicates. The issues are mostly separate regressions in tokenization, loading, generation, config propagation, and attention paths.", + "confidence": 0.66, + "canonical_issue_reason": "Issue 43646 is the broadest and most representative regression in the set: a clear Transformers 5.0 model-initialization break with a concrete code path, making it the best single issue to anchor nearby reports.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43646 is the strongest standalone representative because it is a general v5 regression with a clearly described failure mode and broad duplicate-triage utility.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43531", + "right": "issue:43742", + "accept": false, + "reason": "Different failures: Qwen3-MoE sliding_window behavior vs a MobileLLM pretrained-load key error." + }, + { + "left": "issue:42898", + "right": "issue:46491", + "accept": false, + "reason": "Both are tokenization regressions, but one is about v5 cleanup behavior and the other is a CodeLlama round-trip spacing bug." + }, + { + "left": "issue:29159", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer inconsistency vs a MacOS bus error on a CLIP model; different symptom and subsystem." + }, + { + "left": "issue:43653", + "right": "issue:44568", + "accept": false, + "reason": "Both involve special tokens, but they affect different tokenizers and different concrete behaviors." + }, + { + "left": "issue:43646", + "right": "issue:43749", + "accept": false, + "reason": "Model initialization break in v5 vs FSDP CPU RAM efficient loading; unrelated code paths." + }, + { + "left": "issue:43452", + "right": "issue:43577", + "accept": false, + "reason": "GGUF/AutoModel loading break and BLIP2 dtype propagation are distinct loading bugs." + }, + { + "left": "issue:43756", + "right": "issue:46402", + "accept": false, + "reason": "Smollm3 RoPE-layer mismatch and Sam3Model text-encoder weight loading are different model-specific regressions." + }, + { + "left": "issue:44843", + "right": "issue:44991", + "accept": false, + "reason": "Offline model_info call in a regex patch vs tokenizer loading failure for a specific repo; not the same bug." + }, + { + "left": "issue:43688", + "right": "issue:45704", + "accept": false, + "reason": "Aux-loss normalization and an apex RMSNorm memory leak are unrelated issues." + }, + { + "left": "issue:43653", + "right": "issue:43927", + "accept": false, + "reason": "BigBird special-token registration and DiaConfig custom token IDs being lost are different mechanisms." + }, + { + "left": "issue:44464", + "right": "issue:44743", + "accept": false, + "reason": "Both involve generation/caching, but one is compiled forward chunked generation and the other is recurrent-state reset in modular_qwen3_5." + }, + { + "left": "issue:43688", + "right": "issue:45701", + "accept": false, + "reason": "Auxiliary-loss normalization is unrelated to a tokenization version-change report." + }, + { + "left": "issue:44484", + "right": "issue:44704", + "accept": false, + "reason": "A save_pretrained shard-size question and missing kwargs in AutoProcessor loading are not the same bug." + }, + { + "left": "issue:45491", + "right": "issue:46002", + "accept": false, + "reason": "Both mention NaNs, but one is Gemma3 sliding-window attention on GPU and the other is classification-head NaNs after checkout/init." + }, + { + "left": "issue:44479", + "right": "issue:44625", + "accept": false, + "reason": "Video-input regression for Qwen variants vs num_labels propagation in Qwen3.5; different failures." + }, + { + "left": "issue:44821", + "right": "issue:46821", + "accept": false, + "reason": "AutoImageProcessor URL loading and TimesFM window_size AttributeError are unrelated." + }, + { + "left": "issue:43688", + "right": "issue:45440", + "accept": false, + "reason": "Aux-loss normalization is not the same as DeepSeekV3 native-vs-remote divergence." + }, + { + "left": "issue:33666", + "right": "issue:36010", + "accept": false, + "reason": "Qwen2-VL multi-GPU training and GenerationMixin import failure are different topics and code paths." + }, + { + "left": "issue:32090", + "right": "issue:33666", + "accept": false, + "reason": "Trainer GPU broadcast NoneType error is unrelated to Qwen2-VL multi-GPU training." + }, + { + "left": "issue:44734", + "right": "issue:46612", + "accept": false, + "reason": "Both touch cache/generation, but one is a server KV-cache indexing crash and the other is beam-search reorder being skipped for certain model classes." + }, + { + "left": "issue:44466", + "right": "issue:46050", + "accept": false, + "reason": "Serialization of lm_head.weight by device and quantized sinq save/load breakage are distinct persistence bugs." + }, + { + "left": "issue:43504", + "right": "issue:43720", + "accept": false, + "reason": "Legacy-field preset loading and packed-weight unpacking during accelerate loading are different load-time issues." + }, + { + "left": "issue:44977", + "right": "issue:46693", + "accept": false, + "reason": "Qwen3.5 flash-attention generation correctness vs FA2 long-context performance regression; not the same underlying bug." + }, + { + "left": "issue:43504", + "right": "issue:43550", + "accept": false, + "reason": "Beit legacy-field loading and Bamba torch.compile+SDPA failure are unrelated." + }, + { + "left": "issue:44464", + "right": "issue:44977", + "accept": false, + "reason": "Compiled-forward chunked generation inconsistency and Qwen3.5 flash-attention generation failure are different regressions." + }, + { + "left": "issue:41720", + "right": "issue:44464", + "accept": false, + "reason": "Auto device mapping cudaErrorAssert and compiled-forward chunked generation are separate problems." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "Both involve trust_remote_code, but one is missing config fields and the other is removal of is_torch_fx_available; different breakages." + }, + { + "left": "issue:43577", + "right": "issue:44554", + "accept": false, + "reason": "BLIP2 dtype propagation and an MPS attention correctness 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": 23, + "review_unit_index": 26, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:28218", + "issue:33357", + "issue:33666", + "issue:36331", + "issue:39692", + "issue:42796", + "issue:43262", + "issue:43278", + "issue:43295", + "issue:43452", + "issue:43493", + "issue:43504", + "issue:43525", + "issue:43572", + "issue:43606", + "issue:43611", + "issue:43646", + "issue:43688", + "issue:43701", + "issue:43720", + "issue:43782", + "issue:43825", + "issue:43844", + "issue:43927", + "issue:43994", + "issue:44361", + "issue:44451", + "issue:44466", + "issue:44514", + "issue:44554", + "issue:44568", + "issue:44610", + "issue:44625", + "issue:44849", + "issue:44964", + "issue:44977", + "issue:45290", + "issue:45405", + "issue:45584", + "issue:45823", + "issue:46050", + "issue:46192", + "issue:46209", + "issue:46365", + "issue:46489", + "issue:46693", + "issue:46710", + "issue:46846" + ], + "soft_pairs": [ + "issue:43611|issue:46846", + "issue:43720|issue:43927", + "issue:43504|issue:44610", + "issue:33666|issue:36331", + "issue:43606|issue:43927", + "issue:43782|issue:46693", + "issue:43278|issue:43688", + "issue:43262|issue:43295", + "issue:43720|issue:44361", + "issue:44514|issue:45290", + "issue:44964|issue:45405", + "issue:43493|issue:43825", + "issue:44554|issue:46365", + "issue:43525|issue:43572", + "issue:44977|issue:46209", + "issue:44625|issue:45584", + "issue:43452|issue:44451", + "issue:28218|issue:33357", + "issue:45823|issue:46050", + "issue:42796|issue:44977", + "issue:43994|issue:46192", + "issue:43493|issue:43844", + "issue:39692|issue:43295", + "issue:43278|issue:43844", + "issue:43646|issue:44466", + "issue:46209|issue:46710", + "issue:44849|issue:46209", + "issue:33357|issue:33666", + "issue:44568|issue:46489", + "issue:43278|issue:43701" + ], + "prepared_review_unit_hash": "d0cc6bfe47aeef879f8e3e97515ca7db5fdea046272500748d9cf3127a571260", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36792, + "estimated_input_tokens": 9198, + "estimated_eval_tokens": 18652 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is highly heterogeneous. Most pairs only share a broad subsystem label like tokenizer, processor, config, or model loading, but they describe different concrete failures and code paths. No soft pair looks safely mergeable into a single duplicate issue.", + "confidence": 0.94, + "canonical_issue_reason": "issue:44568 is the cleanest, most self-contained tokenizer regression in the set, and the only nearby soft overlap is another tokenizer/template-related bug rather than a model-loading or training issue.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44568 is the strongest representative of the limited tokenizer-regression theme in this cluster because it is specific, reproducible, and narrowly scoped.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43611", + "right": "issue:46846", + "accept": false, + "reason": "Both mention model internals, but one is a base_model_prefix loading issue and the other is a tensor-parallel plan gap; different bugs." + }, + { + "left": "issue:43720", + "right": "issue:43927", + "accept": false, + "reason": "Packed-weight loading failure and missing token-id propagation are unrelated code paths." + }, + { + "left": "issue:43504", + "right": "issue:44610", + "accept": false, + "reason": "Legacy field loading in BEiT and an OmDet processor input-size mismatch are different problems." + }, + { + "left": "issue:33666", + "right": "issue:36331", + "accept": false, + "reason": "Multi-GPU training support and a custom trainer signature error are not the same defect." + }, + { + "left": "issue:43606", + "right": "issue:43927", + "accept": false, + "reason": "CPU offload device mismatch is unrelated to DiaConfig token-id persistence." + }, + { + "left": "issue:43782", + "right": "issue:46693", + "accept": false, + "reason": "Qwen3VL weight-only loading and FA2 long-context performance regression are different failure modes." + }, + { + "left": "issue:43278", + "right": "issue:43688", + "accept": false, + "reason": "Embedding dtype drift and auxiliary-loss normalization are separate training-time issues." + }, + { + "left": "issue:43262", + "right": "issue:43295", + "accept": false, + "reason": "Both involve processors, but one is a sampling-rate default and the other is processor/tokenizer API breakage." + }, + { + "left": "issue:43720", + "right": "issue:44361", + "accept": false, + "reason": "Accelerate packed-weight loading and MLukeTokenizer task handling are unrelated." + }, + { + "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/no-content handling; different edge cases." + }, + { + "left": "issue:44964", + "right": "issue:45405", + "accept": false, + "reason": "Phi-4 multimodal loading and a PEFT version pin are not the same bug." + }, + { + "left": "issue:43493", + "right": "issue:43825", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and an incorrect pipeline error message are unrelated." + }, + { + "left": "issue:44554", + "right": "issue:46365", + "accept": false, + "reason": "An MPS attention correctness issue and a Trainer import conflict with datasets are different failures." + }, + { + "left": "issue:43525", + "right": "issue:43572", + "accept": false, + "reason": "Both are missing-pad-token config complaints, but they affect different model configs and attributes." + }, + { + "left": "issue:44977", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3.5 flash-attention generation and FP8 inference on B200 are distinct issues." + }, + { + "left": "issue:44625", + "right": "issue:45584", + "accept": false, + "reason": "num_labels propagation in Qwen3.5 and Whisper empty-transcription failures are unrelated." + }, + { + "left": "issue:43452", + "right": "issue:44451", + "accept": false, + "reason": "GGUF loading breakage and a specific ScandiBERT loading failure are different model-loading bugs." + }, + { + "left": "issue:28218", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer spacing after added tokens and a MacOS bus error in CLIP are unrelated." + }, + { + "left": "issue:45823", + "right": "issue:46050", + "accept": false, + "reason": "Gemma4 device-mismatch forward errors and sinq save/load corruption are different code paths." + }, + { + "left": "issue:42796", + "right": "issue:44977", + "accept": false, + "reason": "Ministral tokenizer BPE-marker handling and Qwen3.5 flash-attn generation are unrelated." + }, + { + "left": "issue:43994", + "right": "issue:46192", + "accept": false, + "reason": "SigLIP2 nonsensical outputs and silently ignored return_tensors are different API bugs." + }, + { + "left": "issue:43493", + "right": "issue:43844", + "accept": false, + "reason": "SigLIP2 implementation mismatch and a ZeRO-3 gradient regression do not describe the same bug." + }, + { + "left": "issue:39692", + "right": "issue:43295", + "accept": false, + "reason": "Documentation example errors and processor/tokenizer regression are not the same underlying issue." + }, + { + "left": "issue:43278", + "right": "issue:43844", + "accept": false, + "reason": "Embedding dtype drift and training-gradient explosion are different training regressions." + }, + { + "left": "issue:43646", + "right": "issue:44466", + "accept": false, + "reason": "Custom model initialization breakage and lm_head serialization inconsistency are separate model lifecycle bugs." + }, + { + "left": "issue:46209", + "right": "issue:46710", + "accept": false, + "reason": "FP8 inference failure and an accuracy regression after version upgrade are not the same defect." + }, + { + "left": "issue:44849", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3.5 hidden-state bug and FP8/B200 inference failure are unrelated." + }, + { + "left": "issue:33357", + "right": "issue:33666", + "accept": false, + "reason": "MacOS bus error in CLIP and Qwen2-VL multi-GPU training are different subsystems and failure modes." + }, + { + "left": "issue:44568", + "right": "issue:46489", + "accept": false, + "reason": "Both are tokenizer regressions in v5+, but they affect different models and different special-token/BPE paths; not the same concrete bug." + }, + { + "left": "issue:43278", + "right": "issue:43701", + "accept": false, + "reason": "Embedding dtype change and resume_from_checkpoint key mismatch 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": 23, + "review_unit_index": 27, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:29942", + "issue:30333", + "issue:43257", + "issue:43299", + "issue:43334", + "issue:43504", + "issue:43572", + "issue:43575", + "issue:43606", + "issue:43653", + "issue:43824", + "issue:43844", + "issue:43873", + "issue:43927", + "issue:43937", + "issue:44112", + "issue:44291", + "issue:44361", + "issue:44464", + "issue:44488", + "issue:44493", + "issue:44568", + "issue:44589", + "issue:44792", + "issue:44843", + "issue:44849", + "issue:44869", + "issue:44964", + "issue:45072", + "issue:45200", + "issue:45229", + "issue:45245", + "issue:45357", + "issue:45405", + "issue:45520", + "issue:45759", + "issue:45812", + "issue:46050", + "issue:46285", + "issue:46399", + "issue:46424", + "issue:46701", + "issue:46710", + "issue:46752", + "issue:46821", + "issue:46832", + "issue:46846", + "issue:46858" + ], + "soft_pairs": [ + "issue:44488|issue:44843", + "issue:44792|issue:45405", + "issue:43927|issue:44361", + "issue:44568|issue:45812", + "issue:43824|issue:44589", + "issue:43575|issue:45405", + "issue:44849|issue:45357", + "issue:43504|issue:43606", + "issue:43257|issue:43299", + "issue:46821|issue:46832", + "issue:46399|issue:46846", + "issue:45245|issue:46710", + "issue:44589|issue:45520", + "issue:43334|issue:43572", + "issue:44493|issue:46285", + "issue:43606|issue:44112", + "issue:43334|issue:45759", + "issue:45072|issue:46832", + "issue:43504|issue:43653", + "issue:43572|issue:44291", + "issue:44464|issue:46424", + "issue:43937|issue:46701", + "issue:46832|issue:46858", + "issue:43844|issue:44964", + "issue:45200|issue:45229", + "issue:43873|issue:46050", + "issue:29942|issue:30333", + "issue:44869|issue:46752" + ], + "prepared_review_unit_hash": "4c9fce7a591a0371a7d3a15a55b0add0cf7a627656bc2a3293368898be025f0d", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36304, + "estimated_input_tokens": 9076, + "estimated_eval_tokens": 18408 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose collection of unrelated issue reports with similar wording, not a true duplicate set. The soft edges mostly connect superficially similar regressions across different models, tokenizers, cache/offload paths, and CI failures; none look like the same underlying bug or change.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue stands out: the items span many unrelated bugs and regressions across different subsystems and models.", + "canonical_pr_reason": "No pull requests are present in this cluster.", + "best_issue_reason": "No issue is a clear global representative; several are broad regressions, but none meaningfully subsume the rest of the cluster.", + "best_pr_reason": "No pull requests are present in this cluster.", + "soft_edge_verdicts": [ + { + "left": "issue:44488", + "right": "issue:44843", + "accept": false, + "reason": "Both concern loading/tokenizer behavior, but one is a model load failure for cjvt/sleng-bert and the other is an offline-mode regression in _patch_mistral_regex; different root causes." + }, + { + "left": "issue:44792", + "right": "issue:45405", + "accept": false, + "reason": "Janus test failure and an unreleased PEFT version bump are unrelated issues." + }, + { + "left": "issue:43927", + "right": "issue:44361", + "accept": false, + "reason": "Different AttributeErrors in different code paths: Dia config save/load vs MLuke tokenizer task handling." + }, + { + "left": "issue:44568", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but one is missing BOS/EOS on mdeberta-v3 and the other is wrong token IDs for Granite models; not the same bug." + }, + { + "left": "issue:43824", + "right": "issue:44589", + "accept": false, + "reason": "ImportError for a missing Qwen2.5-VL class and a Float8 storage TypeError are unrelated." + }, + { + "left": "issue:43575", + "right": "issue:45405", + "accept": false, + "reason": "Qwen2 OOM with tensor parallelism and a PEFT release pin issue are unrelated." + }, + { + "left": "issue:44849", + "right": "issue:45357", + "accept": false, + "reason": "Both are Qwen3.5-related, but hidden-state output bug and incorrect save_pretrained visual keys are different failures." + }, + { + "left": "issue:43504", + "right": "issue:43606", + "accept": false, + "reason": "BEiT legacy-field loading and Bark CPU-offload device mismatch are different code paths and defects." + }, + { + "left": "issue:43257", + "right": "issue:43299", + "accept": false, + "reason": "Both mention Qwen3 MOE, but one is accelerate+deepspeed weight conversion and the other is transformers-dev model loading; not the same fix." + }, + { + "left": "issue:46821", + "right": "issue:46832", + "accept": false, + "reason": "TimesFM window_size AttributeError and RT-DETR forward TypeError are unrelated." + }, + { + "left": "issue:46399", + "right": "issue:46846", + "accept": false, + "reason": "Both involve Qwen3.5 TP/kernels, but missing decorator registration and missing TP plan entries are different concrete bugs." + }, + { + "left": "issue:45245", + "right": "issue:46710", + "accept": false, + "reason": "Category-cardinality limit error and an accuracy regression are not the same underlying issue." + }, + { + "left": "issue:44589", + "right": "issue:45520", + "accept": false, + "reason": "Float8 storage lookup failure and Python 3.13 flash_attn import keyerror are unrelated." + }, + { + "left": "issue:43334", + "right": "issue:43572", + "accept": false, + "reason": "Both are missing config fields after v5, but Qwen3VL pad_token_id and StableLm pad_token_idx are different regressions." + }, + { + "left": "issue:44493", + "right": "issue:46285", + "accept": false, + "reason": "Unexpected position-id keys and CLIP text-model structure changes are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:44112", + "accept": false, + "reason": "Both are CI/device-related, but Bark CPU offload mismatch and GraniteSpeech stale device override test are different failures." + }, + { + "left": "issue:43334", + "right": "issue:45759", + "accept": false, + "reason": "Qwen3VL pad_token_id loss and composite Qwen 3.5/3.6 config unwrapping are distinct loading issues." + }, + { + "left": "issue:45072", + "right": "issue:46832", + "accept": false, + "reason": "bf16 dtype mismatch in SwitchTransformers/TimmWrapperModel and RT-DETR feature-level mismatch are unrelated." + }, + { + "left": "issue:43504", + "right": "issue:43653", + "accept": false, + "reason": "BEiT legacy-field loading and BigBirdTokenizer special-token registration are unrelated." + }, + { + "left": "issue:43572", + "right": "issue:44291", + "accept": false, + "reason": "StableLm missing pad_token_idx and init_empty_weights unexpected argument are different regressions." + }, + { + "left": "issue:44464", + "right": "issue:46424", + "accept": false, + "reason": "Compiled chunked generation inconsistency and max_cache_length handling for quantized cache are unrelated." + }, + { + "left": "issue:43937", + "right": "issue:46701", + "accept": false, + "reason": "DiaConfig generation error and DiffusionGemma pipeline failure are separate bugs." + }, + { + "left": "issue:46832", + "right": "issue:46858", + "accept": false, + "reason": "RT-DETR num_feature_levels mismatch and cache.get_seq_length device-tensor arithmetic are unrelated." + }, + { + "left": "issue:43844", + "right": "issue:44964", + "accept": false, + "reason": "ZeRO-3 gradient-growth behavior and Phi-4 multimodal load failure do not share a concrete bug." + }, + { + "left": "issue:45200", + "right": "issue:45229", + "accept": false, + "reason": "Gemma 4 token-type defaulting and multi-GPU OOM are different problems." + }, + { + "left": "issue:43873", + "right": "issue:46050", + "accept": false, + "reason": "Both mention quantization/offloading, but one is an offloading bug and the other is broken save/load for sinq quantization; not mergeable as one fix." + }, + { + "left": "issue:29942", + "right": "issue:30333", + "accept": false, + "reason": "Flash Attention 2 test failures and MLFlow reporting on failed training jobs are unrelated." + }, + { + "left": "issue:44869", + "right": "issue:46752", + "accept": false, + "reason": "Whisper timestamp decoding crash and empty-list chat template IndexError are different code paths and 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": 23, + "review_unit_index": 28, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:15354", + "issue:28218", + "issue:29159", + "issue:29870", + "issue:29942", + "issue:32090", + "issue:34567", + "issue:37428", + "issue:38175", + "issue:41720", + "issue:42490", + "issue:43065", + "issue:43295", + "issue:43450", + "issue:43493", + "issue:43540", + "issue:43572", + "issue:43696", + "issue:43701", + "issue:43873", + "issue:43957", + "issue:43994", + "issue:44186", + "issue:44423", + "issue:44493", + "issue:44554", + "issue:44610", + "issue:44734", + "issue:44977", + "issue:45072", + "issue:45084", + "issue:45198", + "issue:45200", + "issue:45229", + "issue:45406", + "issue:45446", + "issue:45520", + "issue:45588", + "issue:45759", + "issue:45823", + "issue:46050", + "issue:46121", + "issue:46192", + "issue:46369", + "issue:46701", + "issue:46710", + "issue:46821" + ], + "soft_pairs": [ + "issue:45229|issue:45823", + "issue:43994|issue:44493", + "issue:45588|issue:46050", + "issue:42490|issue:43696", + "issue:29159|issue:32090", + "issue:44977|issue:46701", + "issue:45084|issue:45520", + "issue:28218|issue:34567", + "issue:43065|issue:46369", + "issue:44554|issue:45446", + "issue:44423|issue:44734", + "issue:15354|issue:38175", + "issue:43493|issue:43701", + "issue:41720|issue:46710", + "issue:45406|issue:46121", + "issue:44977|issue:46710", + "issue:45072|issue:45198", + "issue:43295|issue:43450", + "issue:37428|issue:43957", + "issue:45588|issue:45759", + "issue:43540|issue:45200", + "issue:43295|issue:43873", + "issue:44186|issue:46192", + "issue:43572|issue:46821", + "issue:43295|issue:44610", + "issue:45200|issue:45823", + "issue:45823|issue:46192", + "issue:29870|issue:29942" + ], + "prepared_review_unit_hash": "8b4d7554374534a84fdebf0a2c3e03e641558b37db5959384cb0665b8c8e69aa", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35567, + "estimated_input_tokens": 8892, + "estimated_eval_tokens": 18040 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The items are mostly unrelated issue reports across tokenizer, generation, serving, quantization, and multimodal code paths. I rejected every soft pair because they describe different concrete failures, not the same underlying bug.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:45200 is the strongest standalone issue: it is specific, reproducible, and has a clear expected fix (default mm_token_type_ids to zeros for text-only fine-tuning in Gemma 4).", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45229", + "right": "issue:45823", + "accept": false, + "reason": "Same model family, but one is CUDA OOM and the other is a device-map mismatch during forward; different failure modes." + }, + { + "left": "issue:43994", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 result-quality regression vs unexpected position-id key warnings; unrelated bugs." + }, + { + "left": "issue:45588", + "right": "issue:46050", + "accept": false, + "reason": "Flash-attention sinkless-model crash vs quantized save/load corruption; different code paths and symptoms." + }, + { + "left": "issue:42490", + "right": "issue:43696", + "accept": false, + "reason": "Trailing-slash import warning and GPU OOM are unrelated." + }, + { + "left": "issue:29159", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer slow/fast inconsistency vs Trainer broadcast NoneType error; no shared underlying defect." + }, + { + "left": "issue:44977", + "right": "issue:46701", + "accept": false, + "reason": "Qwen flash-attention generation bug vs DiffusionGemma pipeline failure; different models and subsystems." + }, + { + "left": "issue:45084", + "right": "issue:45520", + "accept": false, + "reason": "Template-node compilation error vs flash_attn import KeyError; unrelated failures." + }, + { + "left": "issue:28218", + "right": "issue:34567", + "accept": false, + "reason": "Added-token spacing bug vs TrainerState counter not updating; different concerns." + }, + { + "left": "issue:43065", + "right": "issue:46369", + "accept": false, + "reason": "Dummy Conv2d placeholder issue vs memory overflow in sam3-litetext; not the same bug." + }, + { + "left": "issue:44554", + "right": "issue:45446", + "accept": false, + "reason": "MPS attention correctness issue vs PyTorch version guard for AuxRequest import; different layers." + }, + { + "left": "issue:44423", + "right": "issue:44734", + "accept": false, + "reason": "Both are serve crashes, but one is multimodal .to() handling and the other is KV-cache continuation indexing; distinct code paths." + }, + { + "left": "issue:15354", + "right": "issue:38175", + "accept": false, + "reason": "torch.jit.script GeneratorExp export failure vs SigLIP2 zero-probability output issue; unrelated." + }, + { + "left": "issue:43493", + "right": "issue:43701", + "accept": false, + "reason": "SigLIP2 HF/JAX discrepancy vs checkpoint key mismatch; different bugs." + }, + { + "left": "issue:41720", + "right": "issue:46710", + "accept": false, + "reason": "CUDA assert under auto device mapping vs accuracy regression after upgrade; not the same defect." + }, + { + "left": "issue:45406", + "right": "issue:46121", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer vs rope-validation type error; different failure surfaces." + }, + { + "left": "issue:44977", + "right": "issue:46710", + "accept": false, + "reason": "Flash-attention generation bug vs model accuracy regression; unrelated." + }, + { + "left": "issue:45072", + "right": "issue:45198", + "accept": false, + "reason": "Bfloat16 dtype mismatch in CI vs Wav2Vec2 save/tokenization failure; different models and paths." + }, + { + "left": "issue:43295", + "right": "issue:43450", + "accept": false, + "reason": "Processor/tokenizer regression vs video batch shape bug; distinct issues." + }, + { + "left": "issue:37428", + "right": "issue:43957", + "accept": false, + "reason": "Both concern loading/imports, but one is a missing flash-attention helper import and the other is meta-device loading breakage; not the same bug." + }, + { + "left": "issue:45588", + "right": "issue:45759", + "accept": false, + "reason": "Flash-attention None/s_aux crash vs composite Qwen config unwrapping; unrelated." + }, + { + "left": "issue:43540", + "right": "issue:45200", + "accept": false, + "reason": "Video-input processing error in Qwen3OmniMoe vs Gemma4 token-type defaulting bug; different models and code paths." + }, + { + "left": "issue:43295", + "right": "issue:43873", + "accept": false, + "reason": "Processor tokenizer regression vs quantization offloading problem; no shared root cause." + }, + { + "left": "issue:44186", + "right": "issue:46192", + "accept": false, + "reason": "NER/batched-padding tokenizer crash vs return_tensors being ignored; different tokenizer behaviors." + }, + { + "left": "issue:43572", + "right": "issue:46821", + "accept": false, + "reason": "StableLmConfig pad_token_idx missing vs TimesFM window_size AttributeError; unrelated configuration bugs." + }, + { + "left": "issue:43295", + "right": "issue:44610", + "accept": false, + "reason": "Processor/tokenizer regression vs OmDet-Turbo image-size mismatch; different model-specific issues." + }, + { + "left": "issue:45200", + "right": "issue:45823", + "accept": false, + "reason": "Both are Gemma4 issues, but one is token-type defaults and the other is device-map placement; not the same bug." + }, + { + "left": "issue:45823", + "right": "issue:46192", + "accept": false, + "reason": "Device placement mismatch vs tokenizer return_tensors handling; unrelated." + }, + { + "left": "issue:29870", + "right": "issue:29942", + "accept": false, + "reason": "Bart decoder_attention_mask assignment bug vs Flash Attention 2 test failures; 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": 23, + "review_unit_index": 29, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:28218", + "issue:35141", + "issue:37428", + "issue:39692", + "issue:42490", + "issue:42915", + "issue:43232", + "issue:43262", + "issue:43278", + "issue:43295", + "issue:43334", + "issue:43450", + "issue:43452", + "issue:43493", + "issue:43525", + "issue:43550", + "issue:43572", + "issue:43575", + "issue:43653", + "issue:43824", + "issue:43844", + "issue:43856", + "issue:43950", + "issue:43975", + "issue:44112", + "issue:44442", + "issue:44466", + "issue:44521", + "issue:44554", + "issue:44737", + "issue:44792", + "issue:44849", + "issue:44987", + "issue:45127", + "issue:45198", + "issue:45200", + "issue:45290", + "issue:45356", + "issue:45381", + "issue:45520", + "issue:45563", + "issue:46178", + "issue:46192", + "issue:46209", + "issue:46285", + "issue:46512", + "issue:46894", + "issue:46899" + ], + "soft_pairs": [ + "issue:43550|issue:43844", + "issue:45520|issue:46178", + "issue:43450|issue:43575", + "issue:43572|issue:44987", + "issue:44554|issue:45356", + "issue:37428|issue:43824", + "issue:42490|issue:43493", + "issue:43232|issue:44849", + "issue:43295|issue:43493", + "issue:44466|issue:45127", + "issue:39692|issue:46192", + "issue:44521|issue:45290", + "issue:43653|issue:44112", + "issue:45381|issue:46209", + "issue:44737|issue:46894", + "issue:28218|issue:35141", + "issue:43278|issue:43856", + "issue:42915|issue:43824", + "issue:43950|issue:46285", + "issue:43824|issue:43975", + "issue:43452|issue:46512", + "issue:43334|issue:43525", + "issue:43262|issue:43450", + "issue:44442|issue:45198", + "issue:45200|issue:46899", + "issue:44792|issue:45563" + ], + "prepared_review_unit_hash": "0d5eeed72df9239567d8d922979b3c0510e8b6a5c3964ad73b83b204f755870c", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 26, + "serialized_chars": 35884, + "estimated_input_tokens": 8971, + "estimated_eval_tokens": 18198 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous: nearly all pairs are distinct bugs spanning tokenization, model config/load, multimodal processing, import errors, and training/runtime regressions. I found no duplicate PRs and no strong same-bug pair among the soft candidates.", + "confidence": 0.92, + "canonical_issue_reason": "issue:44521 is the strongest standalone representative: it is open, recently active, and has the most inbound references among the listed issues, while the rest of the set is largely unrelated.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44521 is the best global issue anchor in this set because it is the most active and broadly scoped current bug report, even though the cluster does not form a true duplicate group.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43550", + "right": "issue:43844", + "accept": false, + "reason": "Different failure modes: torch.compile/SDPA on Bamba vs gradient blow-up under ZeRO-3. No shared bug or code path indicated." + }, + { + "left": "issue:45520", + "right": "issue:46178", + "accept": false, + "reason": "Both are import/backend errors, but they involve different subsystems and packages (`flash_attn` vs `tensorflow_text`) with unrelated root causes." + }, + { + "left": "issue:43450", + "right": "issue:43575", + "accept": false, + "reason": "Unrelated issues: batched video processor shape vs tensor-parallel OOM on Qwen2. Different code paths and symptoms." + }, + { + "left": "issue:43572", + "right": "issue:44987", + "accept": false, + "reason": "Both concern model loading/configuration, but the specific bugs differ and there is no evidence of the same underlying regression." + }, + { + "left": "issue:44554", + "right": "issue:45356", + "accept": false, + "reason": "Completely different areas: MPS attention correctness vs tokenizer codec handling/warning regression." + }, + { + "left": "issue:37428", + "right": "issue:43824", + "accept": false, + "reason": "Both are ImportErrors, but for different symbols and model families; this looks like separate missing-export issues." + }, + { + "left": "issue:42490", + "right": "issue:43493", + "accept": false, + "reason": "Path-trailing-slash import warning is unrelated to the SigLIP2 implementation discrepancy." + }, + { + "left": "issue:43232", + "right": "issue:44849", + "accept": false, + "reason": "Different generation bugs: sync_gpus kwargs update vs output_hidden_states behavior in Qwen3.5." + }, + { + "left": "issue:43295", + "right": "issue:43493", + "accept": false, + "reason": "Custom processor/tokenizer regression and SigLIP2 model discrepancy are unrelated failures." + }, + { + "left": "issue:44466", + "right": "issue:45127", + "accept": false, + "reason": "Both mention tied weights/lm_head, but one is serialization/device-dependent state and the other is LoRA merge collapse. Not the same concrete bug." + }, + { + "left": "issue:39692", + "right": "issue:46192", + "accept": false, + "reason": "Documentation/model mismatch and quantization failure are unrelated to `return_tensors` being ignored when `text_kwargs` is passed." + }, + { + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Same API surface, but different bugs: all-zero assistant masks vs crash on tool-call assistant messages." + }, + { + "left": "issue:43653", + "right": "issue:44112", + "accept": false, + "reason": "BigBirdTokenizer special-token bug and a GraniteSpeech CI test flake are unrelated." + }, + { + "left": "issue:45381", + "right": "issue:46209", + "accept": false, + "reason": "Qwen2.5-VL vision position ids and FP8 inference on B200 are separate hardware/model issues." + }, + { + "left": "issue:44737", + "right": "issue:46894", + "accept": false, + "reason": "XLNet device placement bug and CohereASR double-shift training bug are unrelated." + }, + { + "left": "issue:28218", + "right": "issue:35141", + "accept": false, + "reason": "Tokenizer spacing after added tokens is not the same as embedding reinitialization after resize/post_init." + }, + { + "left": "issue:43278", + "right": "issue:43856", + "accept": false, + "reason": "Dtype drift between train/eval is unrelated to MoE memory inefficiency." + }, + { + "left": "issue:42915", + "right": "issue:43824", + "accept": false, + "reason": "FineGrainedFP8Config failure on Qwen3Moe and a missing Qwen2.5-VL export are unrelated." + }, + { + "left": "issue:43950", + "right": "issue:46285", + "accept": false, + "reason": "Non-persistent buffer corruption in `from_pretrained` is a different regression from CLIP text-model structure inconsistency." + }, + { + "left": "issue:43824", + "right": "issue:43975", + "accept": false, + "reason": "Missing export ImportError and DeepSeek detokenization regression do not share an underlying fix." + }, + { + "left": "issue:43452", + "right": "issue:46512", + "accept": false, + "reason": "gguf_file loading failures are unrelated to stale `config.dtype` after `from_config(dtype=...)`." + }, + { + "left": "issue:43334", + "right": "issue:43525", + "accept": false, + "reason": "Both mention missing `pad_token_id`, but they affect different config classes/model families; no evidence of a single shared bug." + }, + { + "left": "issue:43262", + "right": "issue:43450", + "accept": false, + "reason": "Audio chat-template sampling-rate default and video batch-shape handling are different preprocessing bugs." + }, + { + "left": "issue:44442", + "right": "issue:45198", + "accept": false, + "reason": "AutoTokenizer failing for FastSpeech2Conformer and Wav2Vec2 save/tokenization failure are separate audio-tokenizer issues." + }, + { + "left": "issue:45200", + "right": "issue:46899", + "accept": false, + "reason": "Text-only Gemma4 fine-tuning mm_token_type_ids issue and 4-bit quantized multimodal dtype-cast issue are different failures." + }, + { + "left": "issue:44792", + "right": "issue:45563", + "accept": false, + "reason": "Janus image-generation test failure and paged-generate warning behavior are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 30, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:28282", + "issue:29942", + "issue:32090", + "issue:33357", + "issue:37428", + "issue:38175", + "issue:41720", + "issue:43278", + "issue:43335", + "issue:43388", + "issue:43425", + "issue:43504", + "issue:43526", + "issue:43550", + "issue:43606", + "issue:43611", + "issue:43653", + "issue:43819", + "issue:43825", + "issue:43827", + "issue:43901", + "issue:43931", + "issue:44112", + "issue:44206", + "issue:44448", + "issue:44561", + "issue:44610", + "issue:44743", + "issue:44855", + "issue:44857", + "issue:44898", + "issue:45083", + "issue:45200", + "issue:45325", + "issue:45412", + "issue:45520", + "issue:45584", + "issue:45593", + "issue:45759", + "issue:46170", + "issue:46192", + "issue:46209", + "issue:46234", + "issue:46285", + "issue:46424", + "issue:46512", + "issue:46821", + "issue:46858" + ], + "soft_pairs": [ + "issue:43819|issue:44112", + "issue:44898|issue:45412", + "issue:43526|issue:46192", + "issue:45200|issue:45325", + "issue:43827|issue:43901", + "issue:41720|issue:46209", + "issue:43335|issue:43550", + "issue:44561|issue:46234", + "issue:43388|issue:43526", + "issue:43819|issue:45083", + "issue:44857|issue:45593", + "issue:44610|issue:44857", + "issue:46424|issue:46858", + "issue:45759|issue:46285", + "issue:46512|issue:46821", + "issue:45520|issue:46170", + "issue:43606|issue:45083", + "issue:38175|issue:45412", + "issue:44206|issue:45584", + "issue:43335|issue:43653", + "issue:37428|issue:43425", + "issue:43550|issue:44112", + "issue:29942|issue:32090", + "issue:28282|issue:33357", + "issue:43611|issue:43931", + "issue:44448|issue:44743", + "issue:43335|issue:43504", + "issue:44855|issue:45083", + "issue:43278|issue:43825" + ], + "prepared_review_unit_hash": "9fa4d0ccde49d7b9ae5156a3f7570d9fe4c0a0be971cedffa6969ea48b57f548", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36751, + "estimated_input_tokens": 9188, + "estimated_eval_tokens": 18632 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a collection of unrelated closed issues spanning model-loading errors, cache/device bugs, docs regressions, and model-specific failures. None of the soft pairs look like the same underlying bug or change, so no pair should be merged as duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "issue:43825 is the most broadly user-facing and clearly framed regression in the set, with a concrete behavior mismatch and a wider impact than the more model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43825 is the best representative issue here because it is specific, reproducible, and broadly applicable compared with the many narrow hardware/model/config-only cases.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43819", + "right": "issue:44112", + "accept": false, + "reason": "Different subsystems and failures: DAC latent conversion mismatch vs GraniteSpeech stale device override CI test." + }, + { + "left": "issue:44898", + "right": "issue:45412", + "accept": false, + "reason": "One is a Perceiver image-size/interpolation bug; the other is an RT-DETR memory-release issue." + }, + { + "left": "issue:43526", + "right": "issue:46192", + "accept": false, + "reason": "Different code paths: BEiT fast image label reduction vs generic processor argument handling for return_tensors." + }, + { + "left": "issue:45200", + "right": "issue:45325", + "accept": false, + "reason": "Both are multimodal-position related, but they affect different models and different logic (mm_token_type_ids vs rope index scaling)." + }, + { + "left": "issue:43827", + "right": "issue:43901", + "accept": false, + "reason": "Both are docs regressions after v5, but they concern different APIs and different incorrect statements." + }, + { + "left": "issue:41720", + "right": "issue:46209", + "accept": false, + "reason": "Different models and environments: Qwen3 auto device mapping on A800 vs FP8 inference on B200." + }, + { + "left": "issue:43335", + "right": "issue:43550", + "accept": false, + "reason": "Unrelated bugs: SwitchTransformers config construction vs Bamba torch.compile/SDPA failure." + }, + { + "left": "issue:44561", + "right": "issue:46234", + "accept": false, + "reason": "Both mention missing helpers, but they are different symbols with different affected models and import paths." + }, + { + "left": "issue:43388", + "right": "issue:43526", + "accept": false, + "reason": "Different label-processing bugs in different components; one is metric gathering, the other is image processor label reduction." + }, + { + "left": "issue:43819", + "right": "issue:45083", + "accept": false, + "reason": "Different helper/functions and model families; no shared underlying defect is evident." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "Different failures: AMP/CUDA crash in LwDetrImageLoss vs missing auxiliary losses in D-FINE." + }, + { + "left": "issue:44610", + "right": "issue:44857", + "accept": false, + "reason": "Processor input-size mismatch vs numerical crash in a different model loss path." + }, + { + "left": "issue:46424", + "right": "issue:46858", + "accept": false, + "reason": "Both involve cache internals, but one is about honoring max_cache_length and the other about device-tensor seq length arithmetic under torch.compile." + }, + { + "left": "issue:45759", + "right": "issue:46285", + "accept": false, + "reason": "Different model-loading structure issues: composite Qwen config unwrapping vs CLIP text-model attribute inconsistency." + }, + { + "left": "issue:46512", + "right": "issue:46821", + "accept": false, + "reason": "Unrelated issues: stale config dtype after from_config vs TimesFM window_size AttributeError." + }, + { + "left": "issue:45520", + "right": "issue:46170", + "accept": false, + "reason": "Different platform-specific import/runtime problems with different dependencies and root causes." + }, + { + "left": "issue:43606", + "right": "issue:45083", + "accept": false, + "reason": "A CPU offload device mismatch is unrelated to qwen3_omni_moe feature-length helper behavior." + }, + { + "left": "issue:38175", + "right": "issue:45412", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 is not the same bug as RT-DETR memory not being freed." + }, + { + "left": "issue:44206", + "right": "issue:45584", + "accept": false, + "reason": "Different audio/model pipeline bugs with different triggers and failure modes." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "Different components and symptoms: SwitchTransformers sparse-layer config vs BigBirdTokenizer special-token registration." + }, + { + "left": "issue:37428", + "right": "issue:43425", + "accept": false, + "reason": "A missing internal import is not the same as a broad torch-version incompatibility report." + }, + { + "left": "issue:43550", + "right": "issue:44112", + "accept": false, + "reason": "Model compile/SDPA failure vs a stale CI test; no shared code-path bug." + }, + { + "left": "issue:29942", + "right": "issue:32090", + "accept": false, + "reason": "FlashAttention test failures and trainer broadcast NoneType errors are unrelated." + }, + { + "left": "issue:28282", + "right": "issue:33357", + "accept": false, + "reason": "PyTorch-not-found import error vs MacOS bus error on a community CLIP model are different issues." + }, + { + "left": "issue:43611", + "right": "issue:43931", + "accept": false, + "reason": "Different model-loading failures: base_model_prefix behavior change vs a Qwen3-VL weight-shape mismatch." + }, + { + "left": "issue:44448", + "right": "issue:44743", + "accept": false, + "reason": "Different regressions: Pegasus output drift across versions vs Qwen3 recurrent-state reset with cache." + }, + { + "left": "issue:43335", + "right": "issue:43504", + "accept": false, + "reason": "SwitchTransformers config bug and Beit semantic-segmentation legacy-field load failure are unrelated." + }, + { + "left": "issue:44855", + "right": "issue:45083", + "accept": false, + "reason": "Python 3.13 torch.jit parsing/IndentationError is unrelated to qwen3_omni_moe output-length helper behavior." + }, + { + "left": "issue:43278", + "right": "issue:43825", + "accept": false, + "reason": "Different bugs in different areas: dtype drift in eval vs pipeline error-message regression." + } + ] + }, + "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": 23, + "review_unit_index": 31, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:26413", + "issue:28218", + "issue:29127", + "issue:31501", + "issue:32090", + "issue:33357", + "issue:34689", + "issue:35141", + "issue:37428", + "issue:39692", + "issue:42490", + "issue:43335", + "issue:43526", + "issue:43575", + "issue:43606", + "issue:43701", + "issue:43720", + "issue:43819", + "issue:43854", + "issue:44112", + "issue:44206", + "issue:44351", + "issue:44360", + "issue:44448", + "issue:44493", + "issue:44514", + "issue:44521", + "issue:44561", + "issue:44617", + "issue:44779", + "issue:44857", + "issue:45072", + "issue:45083", + "issue:45127", + "issue:45245", + "issue:45290", + "issue:45310", + "issue:45459", + "issue:45478", + "issue:45588", + "issue:45593", + "issue:45823", + "issue:45859", + "issue:46234", + "issue:46326", + "issue:46431", + "issue:46566", + "issue:46697" + ], + "soft_pairs": [ + "issue:29127|issue:32090", + "issue:43720|issue:44112", + "issue:43335|issue:45083", + "issue:37428|issue:44351", + "issue:44514|issue:44521", + "issue:43335|issue:44112", + "issue:43335|issue:45593", + "issue:45290|issue:46326", + "issue:45127|issue:45245", + "issue:45588|issue:45859", + "issue:34689|issue:35141", + "issue:42490|issue:43854", + "issue:46431|issue:46566", + "issue:45310|issue:45478", + "issue:44448|issue:44779", + "issue:43819|issue:44493", + "issue:39692|issue:43526", + "issue:43606|issue:44857", + "issue:26413|issue:31501", + "issue:28218|issue:32090", + "issue:45459|issue:46234", + "issue:39692|issue:44493", + "issue:44617|issue:46697", + "issue:43701|issue:45083", + "issue:44206|issue:45823", + "issue:32090|issue:33357", + "issue:42490|issue:43575", + "issue:44857|issue:45072", + "issue:44360|issue:44561" + ], + "prepared_review_unit_hash": "f3f7e8029975bef796e344a61cb2987f0ea96fc55b41800f249ffaa894ea6058", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36464, + "estimated_input_tokens": 9116, + "estimated_eval_tokens": 18488 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly unrelated issues from different subsystems; the only clear duplicate is the two identical Qwen3.5 MoE from_pretrained reports. Everything else is just broad thematic overlap, not the same underlying bug.", + "confidence": 0.96, + "canonical_issue_reason": "issue:45310 is the older of the exact duplicate pair (45310/45478) and cleanly captures the Qwen3.5 MoE from_pretrained regression.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the strongest canonical issue because it is an exact duplicate match with 45478 and is a concrete, narrowly scoped bug report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:29127", + "right": "issue:32090", + "accept": false, + "reason": "Different failures in different code paths: LayoutLMv3 error messaging vs Trainer NoneType GPU broadcast." + }, + { + "left": "issue:43720", + "right": "issue:44112", + "accept": false, + "reason": "Both involve loading/device handling, but they are model-specific CI failures with different root causes." + }, + { + "left": "issue:43335", + "right": "issue:45083", + "accept": false, + "reason": "Both touch sparse/Transformer internals, but one is SwitchTransformers config creation and the other is a qwen3_omni_moe helper bug." + }, + { + "left": "issue:37428", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols and different APIs." + }, + { + "left": "issue:44514", + "right": "issue:44521", + "accept": false, + "reason": "Same general apply_chat_template area, but one crashes on batched padding=False input and the other produces all-zero assistant masks for multimodal inputs." + }, + { + "left": "issue:43335", + "right": "issue:44112", + "accept": false, + "reason": "Unrelated: SwitchTransformers config bug vs a GraniteSpeech stale device override CI failure." + }, + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "Different models and symptoms; sparse-layer creation is not the same as missing auxiliary losses." + }, + { + "left": "issue:45290", + "right": "issue:46326", + "accept": false, + "reason": "Both concern chat templates, but they fail in different tokenizer flows and on different inputs." + }, + { + "left": "issue:45127", + "right": "issue:45245", + "accept": false, + "reason": "Completely different bugs: LoRA merging/output collapse vs a category-cardinality runtime limit." + }, + { + "left": "issue:45588", + "right": "issue:45859", + "accept": false, + "reason": "Different code paths: flash_attention sinkless-model handling vs Qwen3_5Moe rotary embedding CPU-offload incompatibility." + }, + { + "left": "issue:34689", + "right": "issue:35141", + "accept": false, + "reason": "Both involve model loading/weights, but the actual bugs are unrelated." + }, + { + "left": "issue:42490", + "right": "issue:43854", + "accept": false, + "reason": "Local-path import warning vs a specific model-loading failure are not the same bug." + }, + { + "left": "issue:46431", + "right": "issue:46566", + "accept": false, + "reason": "Processor chat-template absence and DiffusionGemma meta-device loading are unrelated." + }, + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Exact same title and same reported Qwen3.5 MoE from_pretrained failure on transformers>=5.4.0." + }, + { + "left": "issue:44448", + "right": "issue:44779", + "accept": false, + "reason": "Both are version regressions, but they affect different models/tokenizers and likely different code paths." + }, + { + "left": "issue:43819", + "right": "issue:44493", + "accept": false, + "reason": "DAC from_latents mismatch is unrelated to unexpected position-id keys in model loading." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "One is a docs/example issue for SigLIP2, the other is a BeitImageProcessorFast functional bug." + }, + { + "left": "issue:43606", + "right": "issue:44857", + "accept": false, + "reason": "Different failures: CPU offload device mismatch vs AMP/CUDA crash in a loss module." + }, + { + "left": "issue:26413", + "right": "issue:31501", + "accept": false, + "reason": "Checkpoint resume iterator failure is not the same as a FineWeb training startup failure." + }, + { + "left": "issue:28218", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer spacing after added tokens is unrelated to Trainer GPU broadcast errors." + }, + { + "left": "issue:45459", + "right": "issue:46234", + "accept": false, + "reason": "Tokenizer protobuf error handling and a missing import_utils helper are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:44493", + "accept": false, + "reason": "Different issues: SigLIP2 docs/example errors vs unexpected position-id key behavior." + }, + { + "left": "issue:44617", + "right": "issue:46697", + "accept": false, + "reason": "CUDA OOM for Sam3Video and Swin2SR image over-padding are unrelated." + }, + { + "left": "issue:43701", + "right": "issue:45083", + "accept": false, + "reason": "resume_from_checkpoint key mismatch is not the same as qwen3_omni_moe output-length helper behavior." + }, + { + "left": "issue:44206", + "right": "issue:45823", + "accept": false, + "reason": "Different model/runtime bugs: unsupported center arg in LasrFeatureExtractor vs Gemma4 PLE device mismatch." + }, + { + "left": "issue:32090", + "right": "issue:33357", + "accept": false, + "reason": "NoneType GPU broadcast and MacOS bus error on CLIP loading do not share the same root cause." + }, + { + "left": "issue:42490", + "right": "issue:43575", + "accept": false, + "reason": "A local-path import warning is unrelated to tensor-parallel OOM on Qwen2-57B." + }, + { + "left": "issue:44857", + "right": "issue:45072", + "accept": false, + "reason": "Both mention numeric precision, but they are different modules and failure modes." + }, + { + "left": "issue:44360", + "right": "issue:44561", + "accept": false, + "reason": "DSA indexer activation question and v5 trust_remote_code import breakage are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative: it identifies the only clearly duplicate pair (issue:45310 / issue:45478) and does not overclaim the softer overlaps. The accepted duplicate verdict is well supported by the identical title and closely matching regression description. 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": 23, + "review_unit_index": 32, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:10000", + "issue:15354", + "issue:25251", + "issue:27957", + "issue:29870", + "issue:29942", + "issue:32090", + "issue:33290", + "issue:34689", + "issue:35141", + "issue:37428", + "issue:38175", + "issue:40444", + "issue:42175", + "issue:42490", + "issue:43278", + "issue:43329", + "issue:43381", + "issue:43526", + "issue:43550", + "issue:43575", + "issue:43873", + "issue:43927", + "issue:43976", + "issue:44448", + "issue:44466", + "issue:44493", + "issue:44561", + "issue:44568", + "issue:44779", + "issue:44855", + "issue:44857", + "issue:44871", + "issue:44977", + "issue:45083", + "issue:45229", + "issue:45290", + "issue:45381", + "issue:45440", + "issue:45491", + "issue:45588", + "issue:45698", + "issue:45920", + "issue:46192", + "issue:46399", + "issue:46421", + "issue:46655", + "issue:46858" + ], + "soft_pairs": [ + "issue:43550|issue:45083", + "issue:27957|issue:34689", + "issue:29942|issue:35141", + "issue:38175|issue:44493", + "issue:29870|issue:32090", + "issue:43278|issue:43329", + "issue:44448|issue:44977", + "issue:45920|issue:46192", + "issue:45290|issue:45381", + "issue:10000|issue:15354", + "issue:44977|issue:45229", + "issue:43278|issue:43381", + "issue:43927|issue:45698", + "issue:44855|issue:46399", + "issue:29870|issue:33290", + "issue:45290|issue:46655", + "issue:40444|issue:43526", + "issue:44857|issue:45588", + "issue:37428|issue:42175", + "issue:32090|issue:33290", + "issue:46421|issue:46858", + "issue:44561|issue:44779", + "issue:43873|issue:45229", + "issue:44448|issue:44466", + "issue:43976|issue:45440", + "issue:25251|issue:32090", + "issue:42490|issue:43873", + "issue:44568|issue:45491", + "issue:43278|issue:43575", + "issue:44871|issue:44977" + ], + "prepared_review_unit_hash": "1c85fcbe3630875a89b5eab988b6444c73d2f7efec45c00daa7097011a137328", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 37075, + "estimated_input_tokens": 9269, + "estimated_eval_tokens": 18794 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is highly heterogeneous: most items are distinct bug reports spanning different models, subsystems, and failure modes. None of the soft-edge pairs look like true duplicates of the same underlying defect. If one issue has to represent the cluster, the broad v5 tokenizer regression (multiple model families affected) is the strongest anchor.", + "confidence": 0.74, + "canonical_issue_reason": "issue:45920 is the broadest, highest-impact report here: a v5 AutoTokenizer regression affecting multiple model families, with a concrete and widely reproducible symptom.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45920 is the best overall representative because it spans several models, clearly describes a regression, and is more general than the narrower single-model bugs in the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43550", + "right": "issue:45083", + "accept": false, + "reason": "Different models and different failures: torch.compile+SDPA on Bamba vs an undefined helper-variable bug in qwen3_omni_moe." + }, + { + "left": "issue:27957", + "right": "issue:34689", + "accept": false, + "reason": "Flash Attention 2 support for XLM-Roberta and a Llama 3.2 Vision loading regression are unrelated problems." + }, + { + "left": "issue:29942", + "right": "issue:35141", + "accept": false, + "reason": "Failing Flash Attention 2 tests and token-embedding resize/post_init reinitialization are separate code paths." + }, + { + "left": "issue:38175", + "right": "issue:44493", + "accept": false, + "reason": "Unexpected zero probabilities in siglip2 and unexpected position-id keys in many models are different symptoms and likely different bugs." + }, + { + "left": "issue:29870", + "right": "issue:32090", + "accept": false, + "reason": "Bart generate() decoder mask assignment and Trainer _gpu_broadcast_one NoneType support are unrelated runtime errors." + }, + { + "left": "issue:43278", + "right": "issue:43329", + "accept": false, + "reason": "Embedding dtype drift in evaluate and a multimodal token-count helper using undefined names are distinct issues." + }, + { + "left": "issue:44448", + "right": "issue:44977", + "accept": false, + "reason": "Pegasus v4/v5 output differences and Qwen3.5 flash-attention generation failures do not describe the same defect." + }, + { + "left": "issue:45920", + "right": "issue:46192", + "accept": false, + "reason": "Tokenizer ID regressions across many models are not the same as silently ignoring return_tensors when text_kwargs is passed." + }, + { + "left": "issue:45290", + "right": "issue:45381", + "accept": false, + "reason": "Chat-template crashes on tool-call assistant messages and Qwen2.5-VL video vision_position_ids issues are separate features." + }, + { + "left": "issue:10000", + "right": "issue:15354", + "accept": false, + "reason": "German DistilBertModel issue and GeneratorExp/torch.jit.script export failure are unrelated." + }, + { + "left": "issue:44977", + "right": "issue:45229", + "accept": false, + "reason": "Flash-attention generation behavior and multi-GPU CUDA OOM on Gemma4 are different failure modes." + }, + { + "left": "issue:43278", + "right": "issue:43381", + "accept": false, + "reason": "BF16-to-FP32 embedding dtype change in eval is not the same as gradient checkpointing being disallowed in eval mode." + }, + { + "left": "issue:43927", + "right": "issue:45698", + "accept": false, + "reason": "Saving/loading custom token IDs and loading the wrong custom module after save_pretrained are both serialization-related, but they are different bugs." + }, + { + "left": "issue:44855", + "right": "issue:46399", + "accept": false, + "reason": "Python 3.13 torch.jit parsing/IndentationError and a missing decorator in Qwen3.5 kernel registration are distinct import/runtime issues." + }, + { + "left": "issue:29870", + "right": "issue:33290", + "accept": false, + "reason": "A generation-time UnboundLocalError and deepspeed+adafactor OOM are not the same underlying problem." + }, + { + "left": "issue:45290", + "right": "issue:46655", + "accept": false, + "reason": "Chat template tool-call crashes and SmolVLM video padding miscalculation affect different parts of the stack." + }, + { + "left": "issue:40444", + "right": "issue:43526", + "accept": false, + "reason": "IterableDataset multi-image finetuning failure and BeitImageProcessorFast label reduction bug are unrelated." + }, + { + "left": "issue:44857", + "right": "issue:45588", + "accept": false, + "reason": "LwDetrImageLoss AMP/CUDA crash and flash_attention.py s_aux=None AttributeError are separate code paths." + }, + { + "left": "issue:37428", + "right": "issue:42175", + "accept": false, + "reason": "Flash-attention utility import breakage and missing TensorFlow in a torch-only install are unrelated packaging/import issues." + }, + { + "left": "issue:32090", + "right": "issue:33290", + "accept": false, + "reason": "Trainer broadcast NoneType support and deepspeed optimizer OOM are different training-time failures." + }, + { + "left": "issue:46421", + "right": "issue:46858", + "accept": false, + "reason": "Lazy cache init causing recompilation and device-tensor seq-length arithmetic in static cache are both cache-related but not the same bug." + }, + { + "left": "issue:44561", + "right": "issue:44779", + "accept": false, + "reason": "Removal of is_torch_fx_available breaking trust_remote_code and DeepSeek tokenizer regressions are separate regressions." + }, + { + "left": "issue:43873", + "right": "issue:45229", + "accept": false, + "reason": "Quantization offloading problems and Gemma4 multi-GPU OOM are both memory-related but not the same defect." + }, + { + "left": "issue:44448", + "right": "issue:44466", + "accept": false, + "reason": "Pegasus output divergence across versions and lm_head.weight serialization differing by device are different issues." + }, + { + "left": "issue:43976", + "right": "issue:45440", + "accept": false, + "reason": "Python-version compatibility in Transformers 5.1.0 and DeepSeekV3MoE parity with remote code are unrelated." + }, + { + "left": "issue:25251", + "right": "issue:32090", + "accept": false, + "reason": "Pipeline top_k nesting behavior and Trainer broadcast NoneType errors are unrelated." + }, + { + "left": "issue:42490", + "right": "issue:43873", + "accept": false, + "reason": "Trailing-slash local model warning and quantization offloading malfunction are unrelated." + }, + { + "left": "issue:44568", + "right": "issue:45491", + "accept": false, + "reason": "mDeBERTa add_special_tokens behavior and Gemma3 NaN embeddings with mixed-length batches are different tokenizer/model bugs." + }, + { + "left": "issue:43278", + "right": "issue:43575", + "accept": false, + "reason": "Eval dtype change and Qwen2-57B tensor-parallel OOM are not the same issue." + }, + { + "left": "issue:44871", + "right": "issue:44977", + "accept": false, + "reason": "Gemma eos_token_id config mismatch and Qwen3.5 flash-attention generation problems are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 33, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:29127", + "issue:29870", + "issue:29942", + "issue:30064", + "issue:33357", + "issue:34567", + "issue:34689", + "issue:39692", + "issue:41720", + "issue:42490", + "issue:42915", + "issue:42994", + "issue:43065", + "issue:43066", + "issue:43381", + "issue:43412", + "issue:43575", + "issue:43611", + "issue:43653", + "issue:43701", + "issue:43756", + "issue:43994", + "issue:44186", + "issue:44206", + "issue:44265", + "issue:44361", + "issue:44442", + "issue:44462", + "issue:44479", + "issue:44743", + "issue:44977", + "issue:45198", + "issue:45229", + "issue:45245", + "issue:45372", + "issue:45406", + "issue:45412", + "issue:45440", + "issue:45459", + "issue:45491", + "issue:45563", + "issue:45588", + "issue:45593", + "issue:45712", + "issue:45803", + "issue:45823", + "issue:46121", + "issue:46693" + ], + "soft_pairs": [ + "issue:43381|issue:43701", + "issue:45588|issue:45823", + "issue:41720|issue:45229", + "issue:33357|issue:39692", + "issue:29127|issue:34567", + "issue:44977|issue:45563", + "issue:45198|issue:45803", + "issue:44361|issue:45198", + "issue:44186|issue:45198", + "issue:43653|issue:44743", + "issue:43611|issue:46121", + "issue:29870|issue:34689", + "issue:45372|issue:45588", + "issue:42915|issue:42994", + "issue:43756|issue:45440", + "issue:43653|issue:44442", + "issue:29942|issue:30064", + "issue:44206|issue:44479", + "issue:45372|issue:45406", + "issue:43412|issue:45588", + "issue:45459|issue:45712", + "issue:45491|issue:46693", + "issue:44265|issue:45198", + "issue:29127|issue:33357", + "issue:43065|issue:43994", + "issue:43065|issue:45412", + "issue:45245|issue:45563", + "issue:43575|issue:43701", + "issue:43066|issue:44462", + "issue:45440|issue:45593", + "issue:45372|issue:45459", + "issue:42490|issue:43381" + ], + "prepared_review_unit_hash": "4a729ac2ae51120b355e85f814ade252dffb233b4778854d10a371eb95a6a4cf", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37284, + "estimated_input_tokens": 9321, + "estimated_eval_tokens": 18898 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the soft pairs span unrelated tokenizer, generation, vision, training, export, and memory issues, so I rejected all duplicate candidates. I\u2019d use Gemma4 multi-GPU CUDA OOM as the best standalone issue anchor because it is a concrete runtime failure with substantial discussion.", + "confidence": 0.9, + "canonical_issue_reason": "issue:45229 is the clearest standalone bug in the set: it has the strongest discussion signal and a concrete, reproducible multi-GPU CUDA OOM failure mode.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45229 is the most triage-worthy representative issue: specific, user-facing, and likely to be useful as a cluster anchor.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43381", + "right": "issue:43701", + "accept": false, + "reason": "Different bugs: eval-mode gradient checkpointing restriction vs resume-from-checkpoint key mismatch." + }, + { + "left": "issue:45588", + "right": "issue:45823", + "accept": false, + "reason": "Different code paths: flash-attention crash on sink-less models vs Gemma4 device-map forward mismatch." + }, + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "Both are GPU inference failures, but one is a Qwen3 device-map assert and the other is a Gemma4 OOM; not the same fix." + }, + { + "left": "issue:33357", + "right": "issue:39692", + "accept": false, + "reason": "Unrelated symptoms: MacOS CLIP bus error vs SigLIP2 documentation/example errors." + }, + { + "left": "issue:29127", + "right": "issue:34567", + "accept": false, + "reason": "LayoutLMv3 box/error-message issue and TrainerState token-count update issue are unrelated." + }, + { + "left": "issue:44977", + "right": "issue:45563", + "accept": false, + "reason": "Flash-attention generation failure and stale warning for num_return_sequences are different bugs." + }, + { + "left": "issue:45198", + "right": "issue:45803", + "accept": false, + "reason": "Wav2Vec2 save/tokenization failure is not the same as a bare-except error-swallowing bug in Fuyu." + }, + { + "left": "issue:44361", + "right": "issue:45198", + "accept": false, + "reason": "Different tokenizer/model failures: MLuke AttributeError vs Wav2Vec2 save/tokenization breakage." + }, + { + "left": "issue:44186", + "right": "issue:45198", + "accept": false, + "reason": "LayoutLMv2 NER/padding crash and Wav2Vec2 save/tokenization failure are separate issues." + }, + { + "left": "issue:43653", + "right": "issue:44743", + "accept": false, + "reason": "Tokenizer special-token registration bug and Qwen recurrent-state reset bug are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:46121", + "accept": false, + "reason": "Model-loading regression from base_model_prefix is unrelated to rope-validation type handling." + }, + { + "left": "issue:29870", + "right": "issue:34689", + "accept": false, + "reason": "BART generate decoder-mask bug and Llama 3.2 Vision model-loading regression are different failures." + }, + { + "left": "issue:45372", + "right": "issue:45588", + "accept": false, + "reason": "Gemma4 processor import failure is a different bug than the flash_attention s_aux=None crash." + }, + { + "left": "issue:42915", + "right": "issue:42994", + "accept": false, + "reason": "FP8 config loading failure and quantized model saving failure are distinct operations." + }, + { + "left": "issue:43756", + "right": "issue:45440", + "accept": false, + "reason": "Smollm3 RoPE-layer mismatch and DeepseekV3MoE divergence are different model-implementation bugs." + }, + { + "left": "issue:43653", + "right": "issue:44442", + "accept": false, + "reason": "BigBird special-token decode bug and AutoTokenizer loading failure are unrelated." + }, + { + "left": "issue:29942", + "right": "issue:30064", + "accept": false, + "reason": "Flash Attention 2 test failures and void segmentation-map processing failure are not the same bug." + }, + { + "left": "issue:44206", + "right": "issue:44479", + "accept": false, + "reason": "Unsupported center-arg regression and video-input regression affect different code paths and models." + }, + { + "left": "issue:45372", + "right": "issue:45406", + "accept": false, + "reason": "Gemma4 processor import breakage and missing _tokenizer attribute are related only broadly; they are different failure points." + }, + { + "left": "issue:43412", + "right": "issue:45588", + "accept": false, + "reason": "Executorch export guard/erfinv problems are unrelated to the flash_attention sink-less crash." + }, + { + "left": "issue:45459", + "right": "issue:45712", + "accept": false, + "reason": "Protobuf decode-error handling and leftover dummy-class cleanup are completely different issues." + }, + { + "left": "issue:45491", + "right": "issue:46693", + "accept": false, + "reason": "Both mention FA2/long-context, but one is a correctness NaN bug in Gemma3 batching and the other is a performance regression." + }, + { + "left": "issue:44265", + "right": "issue:45198", + "accept": false, + "reason": "torch.export failure with torch_compilable_check is unrelated to Wav2Vec2 save/tokenization failure." + }, + { + "left": "issue:29127", + "right": "issue:33357", + "accept": false, + "reason": "LayoutLMv3 box-clarity issue and CLIP MacOS bus error are unrelated." + }, + { + "left": "issue:43065", + "right": "issue:43994", + "accept": false, + "reason": "Dummy Conv2d cleanup in Sam3PixelDecoder is unrelated to SigLIP2 model/pipeline mismatch." + }, + { + "left": "issue:43065", + "right": "issue:45412", + "accept": false, + "reason": "Dummy-class cleanup and RT-DETR memory leak are different bugs." + }, + { + "left": "issue:45245", + "right": "issue:45563", + "accept": false, + "reason": "Category-count runtime error and stale warning for num_return_sequences are unrelated." + }, + { + "left": "issue:43575", + "right": "issue:43701", + "accept": false, + "reason": "Tensor-parallel OOM and checkpoint key mismatch are different failure modes." + }, + { + "left": "issue:43066", + "right": "issue:44462", + "accept": false, + "reason": "Wrong tokenizer decoder type in v5 and ignoring tokenizer.json are separate tokenizer-loading bugs." + }, + { + "left": "issue:45440", + "right": "issue:45593", + "accept": false, + "reason": "DeepSeek implementation divergence and D-FINE auxiliary-loss behavior are unrelated model bugs." + }, + { + "left": "issue:45372", + "right": "issue:45459", + "accept": false, + "reason": "Gemma4 processor import error and protobuf-related tokenizer error handling are different issues." + }, + { + "left": "issue:42490", + "right": "issue:43381", + "accept": false, + "reason": "Local-path trailing-slash import warning and gradient-checkpointing eval-mode restriction 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": 23, + "review_unit_index": 34, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:27957", + "issue:29127", + "issue:33290", + "issue:34689", + "issue:35141", + "issue:42490", + "issue:42915", + "issue:43262", + "issue:43334", + "issue:43381", + "issue:43421", + "issue:43526", + "issue:43531", + "issue:43575", + "issue:43646", + "issue:43761", + "issue:43824", + "issue:43873", + "issue:43931", + "issue:43994", + "issue:44077", + "issue:44186", + "issue:44206", + "issue:44279", + "issue:44387", + "issue:44442", + "issue:44464", + "issue:44514", + "issue:44743", + "issue:44811", + "issue:44898", + "issue:44977", + "issue:45072", + "issue:45245", + "issue:45250", + "issue:45381", + "issue:45412", + "issue:45491", + "issue:45584", + "issue:45588", + "issue:45600", + "issue:45834", + "issue:45859", + "issue:46121", + "issue:46369", + "issue:46399", + "issue:46640", + "issue:46693" + ], + "soft_pairs": [ + "issue:29127|issue:35141", + "issue:43421|issue:45491", + "issue:43994|issue:45412", + "issue:45412|issue:46369", + "issue:43575|issue:44387", + "issue:44442|issue:45072", + "issue:44464|issue:45584", + "issue:42915|issue:43931", + "issue:43526|issue:45834", + "issue:43526|issue:46693", + "issue:43824|issue:44279", + "issue:27957|issue:29127", + "issue:27957|issue:33290", + "issue:45250|issue:46640", + "issue:43994|issue:46369", + "issue:33290|issue:34689", + "issue:43526|issue:44186", + "issue:33290|issue:35141", + "issue:43334|issue:43531", + "issue:46121|issue:46399", + "issue:45588|issue:45600", + "issue:43381|issue:43873", + "issue:44514|issue:45381", + "issue:43761|issue:45859", + "issue:43646|issue:44077", + "issue:44977|issue:45245", + "issue:44743|issue:44811", + "issue:44442|issue:44898", + "issue:43262|issue:44206", + "issue:42490|issue:44743" + ], + "prepared_review_unit_hash": "b4bd286a654f6a491b565464c9315dca0c2914eca89713147f94e508d54dae50", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36569, + "estimated_input_tokens": 9143, + "estimated_eval_tokens": 18542 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous issue reports with only superficial overlap around Flash Attention, Qwen, memory/OOM, and processor bugs; no soft-edge pair looks like the same concrete underlying bug or change.", + "confidence": 0.96, + "canonical_issue_reason": "Issue 46693 is the best representative: it is a specific, recent, clearly described FA2 regression with current activity, while the rest of the cluster is mostly unrelated single-bug reports.", + "canonical_pr_reason": null, + "best_issue_reason": "46693 is the most actionable issue in the set because it states a concrete regression, a concrete workload (long-context generation), and has active references; it is the strongest anchor for the cluster even though the cluster itself is not duplicate-heavy.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:29127", + "right": "issue:35141", + "accept": false, + "reason": "Different subsystems and failures: LayoutLMv3 error messaging vs token-embedding reinitialization during post_init." + }, + { + "left": "issue:43421", + "right": "issue:45491", + "accept": false, + "reason": "Tokenizer post-processor update behavior is unrelated to Gemma3 NaN embeddings on mixed-length batching." + }, + { + "left": "issue:43994", + "right": "issue:45412", + "accept": false, + "reason": "Different models and bug classes: nonsensical predictions vs model memory not being released." + }, + { + "left": "issue:45412", + "right": "issue:46369", + "accept": false, + "reason": "Both are memory-related, but one is GC/release behavior in RT-DETR and the other is an OOM in sam3-litetext; no shared code-path is evident." + }, + { + "left": "issue:43575", + "right": "issue:44387", + "accept": false, + "reason": "Both mention OOM, but one is tensor-parallel model loading and the other is int4 quantization memory growth; too different to merge." + }, + { + "left": "issue:44442", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer loading failure vs dtype mismatch in inference; different components and failure modes." + }, + { + "left": "issue:44464", + "right": "issue:45584", + "accept": false, + "reason": "Chunked generation with compiled forward is unrelated to Whisper empty-transcription handling after align_special_tokens." + }, + { + "left": "issue:42915", + "right": "issue:43931", + "accept": false, + "reason": "Qwen3Moe FP8 config failure is unrelated to Qwen3-VL weight-shape mismatch during loading." + }, + { + "left": "issue:43526", + "right": "issue:45834", + "accept": false, + "reason": "BEiT reduce_labels bug and Kosmos2.5 long-OCR index error are different processor issues." + }, + { + "left": "issue:43526", + "right": "issue:46693", + "accept": false, + "reason": "Different area and bug type: image processor label reduction vs FA2 long-context performance regression." + }, + { + "left": "issue:43824", + "right": "issue:44279", + "accept": false, + "reason": "Specific missing import for a Qwen2.5-VL class is not the same as a generic dependency issue." + }, + { + "left": "issue:27957", + "right": "issue:29127", + "accept": false, + "reason": "XLMRoberta Flash Attention 2 support and LayoutLMv3 error clarity are unrelated." + }, + { + "left": "issue:27957", + "right": "issue:33290", + "accept": false, + "reason": "Flash Attention 2 support and AdaFactor/DeepSpeed OOM are different bugs." + }, + { + "left": "issue:45250", + "right": "issue:46640", + "accept": false, + "reason": "Both are generic FA2 titles, but neither identifies the same concrete bug or code path; insufficient evidence to merge." + }, + { + "left": "issue:43994", + "right": "issue:46369", + "accept": false, + "reason": "Different model families and symptoms: bad outputs vs memory overflow." + }, + { + "left": "issue:33290", + "right": "issue:34689", + "accept": false, + "reason": "OOM in DeepSpeed with AdaFactor is unrelated to a model-loading regression for Llama 3.2 Vision." + }, + { + "left": "issue:43526", + "right": "issue:44186", + "accept": false, + "reason": "ImageProcessor reduce_labels bug and LayoutLMv2Tokenizer padding/NER crash are different components." + }, + { + "left": "issue:33290", + "right": "issue:35141", + "accept": false, + "reason": "Optimizer-related OOM and embedding reinitialization are unrelated." + }, + { + "left": "issue:43334", + "right": "issue:43531", + "accept": false, + "reason": "Missing pad_token_id on Qwen3-VL text config is unrelated to sliding_window behavior in Qwen3-MoE." + }, + { + "left": "issue:46121", + "right": "issue:46399", + "accept": false, + "reason": "Both involve RoPE-adjacent code, but one is parameter conversion validation and the other is kernel registration/decorator failure; not the same bug." + }, + { + "left": "issue:45588", + "right": "issue:45600", + "accept": false, + "reason": "Flash-attention sink handling and removed auto_mappings configs are separate CI/runtime issues." + }, + { + "left": "issue:43381", + "right": "issue:43873", + "accept": false, + "reason": "Gradient checkpointing in eval mode and quantization offloading are unrelated behaviors." + }, + { + "left": "issue:44514", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen2.5-VL issues, but one is batched chat-template padding and the other is video vision_position_ids; different bugs." + }, + { + "left": "issue:43761", + "right": "issue:45859", + "accept": false, + "reason": "CLIP hidden_states regression and Qwen3.5MoE CPU-offload incompatibility are unrelated." + }, + { + "left": "issue:43646", + "right": "issue:44077", + "accept": false, + "reason": "Custom model initialization breakage is not the same as patchtsmixer's post_init contract issue." + }, + { + "left": "issue:44977", + "right": "issue:45245", + "accept": false, + "reason": "Flash-attention generation issues and a categories-limit RuntimeError are unrelated." + }, + { + "left": "issue:44743", + "right": "issue:44811", + "accept": false, + "reason": "Qwen3.5 recurrent-state reset and Whisper skip_special_tokens handling are different code paths." + }, + { + "left": "issue:44442", + "right": "issue:44898", + "accept": false, + "reason": "FastSpeech2ConformerTokenizer loading and Perceiver interpolation failure are unrelated." + }, + { + "left": "issue:43262", + "right": "issue:44206", + "accept": false, + "reason": "Audio chat-template sample-rate defaulting and feature-extractor center-arg crashes are different audio bugs." + }, + { + "left": "issue:42490", + "right": "issue:44743", + "accept": false, + "reason": "Trailing-slash import warning and Qwen3.5 recurrent-state reset 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": 23, + "review_unit_index": 35, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:27957", + "issue:29870", + "issue:32090", + "issue:42175", + "issue:42617", + "issue:42907", + "issue:43232", + "issue:43316", + "issue:43388", + "issue:43493", + "issue:43525", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43643", + "issue:43701", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43761", + "issue:43827", + "issue:43873", + "issue:43931", + "issue:44062", + "issue:44077", + "issue:44079", + "issue:44188", + "issue:44190", + "issue:44442", + "issue:44479", + "issue:44496", + "issue:44855", + "issue:44871", + "issue:44913", + "issue:44987", + "issue:45003", + "issue:45137", + "issue:45216", + "issue:45325", + "issue:45440", + "issue:45698", + "issue:45759", + "issue:45803", + "issue:45907", + "issue:46178", + "issue:46234", + "issue:46506", + "issue:46697" + ], + "soft_pairs": [ + "issue:45003|issue:45440", + "issue:43720|issue:44442", + "issue:42175|issue:46234", + "issue:27957|issue:29870", + "issue:43316|issue:43525", + "issue:43493|issue:43827", + "issue:44062|issue:45907", + "issue:44077|issue:44479", + "issue:45325|issue:46697", + "issue:43873|issue:44871", + "issue:44855|issue:45137", + "issue:43643|issue:44913", + "issue:43749|issue:43931", + "issue:43316|issue:45759", + "issue:43761|issue:44079", + "issue:44190|issue:44496", + "issue:44987|issue:46178", + "issue:44442|issue:45803", + "issue:43931|issue:45216", + "issue:43761|issue:44188", + "issue:43232|issue:43388", + "issue:43701|issue:46506", + "issue:42907|issue:45698", + "issue:42617|issue:43575", + "issue:27957|issue:32090", + "issue:43575|issue:43742", + "issue:43550|issue:43701", + "issue:43606|issue:43701" + ], + "prepared_review_unit_hash": "b47d8598c56dbc6b1b0f2183eeca55a20715f9709a93923d5c2cc40d5827fb2d", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36215, + "estimated_input_tokens": 9054, + "estimated_eval_tokens": 18364 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is not a real duplicate set; it spans unrelated issues across model loading, configs, generation, distributed training, docs, and processors. All soft edges should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": "Issue 43701 is the closest representative only because it has multiple weak loading/runtime links, but the overall cluster is too heterogeneous to treat as one duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43701 is the best available cluster representative due to repeated soft-linking, but it is still not a strong duplicate anchor.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45003", + "right": "issue:45440", + "accept": false, + "reason": "Different bugs: unsafe sys.modules access vs a DeepSeekV3 model divergence. No shared code-path problem." + }, + { + "left": "issue:43720", + "right": "issue:44442", + "accept": false, + "reason": "Unrelated failures: BitNet loading/packed weights vs tokenizer loading for FastSpeech2Conformer." + }, + { + "left": "issue:42175", + "right": "issue:46234", + "accept": false, + "reason": "Different backend/package issues: pip extras omitting tensorflow vs a missing import_utils availability helper." + }, + { + "left": "issue:27957", + "right": "issue:29870", + "accept": false, + "reason": "Distinct model/runtime problems: XLM-RoBERTa Flash Attention 2 vs BART generate() decoder_attention_mask bug." + }, + { + "left": "issue:43316", + "right": "issue:43525", + "accept": false, + "reason": "Both are config-related, but one is a Gemma3TextConfig API discrepancy and the other is a missing Llama4Config attribute; not the same bug." + }, + { + "left": "issue:43493", + "right": "issue:43827", + "accept": false, + "reason": "Completely different domains: SigLIP2 implementation mismatch vs docs still referencing removed pipeline() APIs." + }, + { + "left": "issue:44062", + "right": "issue:45907", + "accept": false, + "reason": "Tokenizer AddedToken construction error vs Hub repo file listing not forwarding revision; unrelated paths." + }, + { + "left": "issue:44077", + "right": "issue:44479", + "accept": false, + "reason": "PatchTSMixer post_init validation issue vs Qwen video input regression; different models and failure modes." + }, + { + "left": "issue:45325", + "right": "issue:46697", + "accept": false, + "reason": "Different preprocessing bugs: Qwen2.5-VL rope indexing vs Swin2SR over-padding." + }, + { + "left": "issue:43873", + "right": "issue:44871", + "accept": false, + "reason": "Quantization/offloading behavior vs Gemma-3 eos_token_id config mismatch; not the same underlying defect." + }, + { + "left": "issue:44855", + "right": "issue:45137", + "accept": false, + "reason": "Python 3.13 torch.jit parsing error vs DeepSpeed ZeRO3 deque underflow; unrelated failures." + }, + { + "left": "issue:43643", + "right": "issue:44913", + "accept": false, + "reason": "Both touch config persistence, but one is trust_remote_code missing fields and the other is GPTNeoX rotary_pct reload behavior; not the same bug." + }, + { + "left": "issue:43749", + "right": "issue:43931", + "accept": false, + "reason": "Different loading problems: FSDP CPU RAM efficient loading vs Qwen3-VL weight shape mismatch." + }, + { + "left": "issue:43316", + "right": "issue:45759", + "accept": false, + "reason": "Both involve config objects, but one is an API discrepancy in Gemma3TextConfig and the other is AutoModelForCausalLM.from_config not unwrapping text_config." + }, + { + "left": "issue:43761", + "right": "issue:44079", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and ModelOutput key assignment are related at a high level but not clearly the same concrete bug." + }, + { + "left": "issue:44190", + "right": "issue:44496", + "accept": false, + "reason": "Local dataset loading failure in a training script vs unrecognized model/config loading for Olmo-Hybrid-Instruct." + }, + { + "left": "issue:44987", + "right": "issue:46178", + "accept": false, + "reason": "Different backend issues: transformers>=5.1.0 loading regression vs tensorflow_text backend mapping error." + }, + { + "left": "issue:44442", + "right": "issue:45803", + "accept": false, + "reason": "Tokenizer loading failure vs FuyuBatchFeature bare except swallowing errors; unrelated code paths." + }, + { + "left": "issue:43931", + "right": "issue:45216", + "accept": false, + "reason": "Different model-loading regressions: Qwen3-VL shape mismatch vs Qwen3.5 save_pretrained checkpoint corruption." + }, + { + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "CLIPVisionModel output regression vs attention-kernel divergence under torch.compile; not the same issue." + }, + { + "left": "issue:43232", + "right": "issue:43388", + "accept": false, + "reason": "Generation kwargs sync issue vs gather_for_metrics dropping tuple label elements; unrelated." + }, + { + "left": "issue:43701", + "right": "issue:46506", + "accept": false, + "reason": "resume_from_checkpoint key mismatch is unrelated to CPU/CUDA divergence in PPDocLayoutV3." + }, + { + "left": "issue:42907", + "right": "issue:45698", + "accept": false, + "reason": "Saving dequantized models vs loading the wrong custom module after save_pretrained; both involve save/load but different bugs." + }, + { + "left": "issue:42617", + "right": "issue:43575", + "accept": false, + "reason": "3d_parallel.py runtime failure vs tensor-parallel OOM when loading a specific Qwen2 model; not the same defect." + }, + { + "left": "issue:27957", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention 2 model issue vs Trainer gpu_broadcast_one NoneType error; unrelated." + }, + { + "left": "issue:43575", + "right": "issue:43742", + "accept": false, + "reason": "Tensor-parallel OOM for Qwen2-57B vs key error when loading MobileLLM-125M; distinct loading failures." + }, + { + "left": "issue:43550", + "right": "issue:43701", + "accept": false, + "reason": "torch.compile/SDPA failure in Bamba vs resume_from_checkpoint key mismatch; different subsystems and symptoms." + }, + { + "left": "issue:43606", + "right": "issue:43701", + "accept": false, + "reason": "CPU offload device mismatch for bark-small vs checkpoint key mismatch; 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": 23, + "review_unit_index": 36, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:16998", + "issue:22355", + "issue:25251", + "issue:28282", + "issue:42175", + "issue:43295", + "issue:43316", + "issue:43352", + "issue:43404", + "issue:43441", + "issue:43452", + "issue:43504", + "issue:43540", + "issue:43572", + "issue:43646", + "issue:43650", + "issue:43716", + "issue:43761", + "issue:43825", + "issue:44077", + "issue:44162", + "issue:44242", + "issue:44464", + "issue:44479", + "issue:44496", + "issue:44514", + "issue:44534", + "issue:44610", + "issue:44792", + "issue:44871", + "issue:44913", + "issue:45542", + "issue:45584", + "issue:45632", + "issue:45698", + "issue:45803", + "issue:46050", + "issue:46170", + "issue:46178", + "issue:46202", + "issue:46234", + "issue:46291", + "issue:46365", + "issue:46369", + "issue:46421", + "issue:46424", + "issue:46640", + "issue:46894" + ], + "soft_pairs": [ + "issue:25251|issue:28282", + "issue:46234|issue:46291", + "issue:43295|issue:43825", + "issue:16998|issue:22355", + "issue:43295|issue:44871", + "issue:44464|issue:46421", + "issue:44871|issue:45584", + "issue:43316|issue:43716", + "issue:42175|issue:43316", + "issue:44077|issue:44534", + "issue:44871|issue:46050", + "issue:43646|issue:44913", + "issue:45803|issue:46365", + "issue:44162|issue:44792", + "issue:43452|issue:44496", + "issue:43404|issue:43441", + "issue:44242|issue:46894", + "issue:45632|issue:46424", + "issue:45803|issue:46170", + "issue:44610|issue:46369", + "issue:43650|issue:46202", + "issue:43761|issue:44479", + "issue:43504|issue:45698", + "issue:45542|issue:46234", + "issue:43540|issue:44514", + "issue:43572|issue:46178", + "issue:43352|issue:46640" + ], + "prepared_review_unit_hash": "518b60358e2439ab4c8890a005e64615beb9759b4977b7aaaffbcc84650b160e", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35663, + "estimated_input_tokens": 8916, + "estimated_eval_tokens": 18088 + }, + "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, loosely unified by v5 regressions and model/loading/config edge cases. None of the soft pairs look like the same underlying bug or change, so I would not merge any of them as duplicates.", + "confidence": 0.91, + "canonical_issue_reason": "issue:43646 is the broadest umbrella for the cluster\u2019s recurring theme of Transformers v5 breaking custom model initialization/loading behavior.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45698 is the most actionable representative: it is open, actively discussed, and specifically about a custom-model loading regression after save_pretrained/from_pretrained.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Different subsystems and failures: pipeline output shape vs missing PyTorch backend import." + }, + { + "left": "issue:46234", + "right": "issue:46291", + "accept": false, + "reason": "One is a missing utility symbol, the other is a kernels dependency/version requirement on pipelines." + }, + { + "left": "issue:43295", + "right": "issue:43825", + "accept": false, + "reason": "Different regressions: processor/tokenizer image handling vs pipeline translation error messaging." + }, + { + "left": "issue:16998", + "right": "issue:22355", + "accept": false, + "reason": "Unrelated: model_max_length behavior for DeBERTa vs ONNX module import failure." + }, + { + "left": "issue:43295", + "right": "issue:44871", + "accept": false, + "reason": "Both are v5-era regressions, but they affect different models and code paths (multimodal processor vs Gemma EOS config)." + }, + { + "left": "issue:44464", + "right": "issue:46421", + "accept": false, + "reason": "Both mention chunking/caching, but one is inconsistent generation output and the other is spurious recompilation." + }, + { + "left": "issue:44871", + "right": "issue:45584", + "accept": false, + "reason": "Different models and bugs: Gemma EOS config mismatch vs Whisper empty-transcription generation failure." + }, + { + "left": "issue:43316", + "right": "issue:43716", + "accept": false, + "reason": "API discrepancy in config classes is not the same as image preprocessor/model dtype mismatch." + }, + { + "left": "issue:42175", + "right": "issue:43316", + "accept": false, + "reason": "Backend packaging/install issue vs config API inconsistency; not the same bug." + }, + { + "left": "issue:44077", + "right": "issue:44534", + "accept": false, + "reason": "PatchTSMixer post_init validation and non-persistent buffer initialization are separate changes." + }, + { + "left": "issue:44871", + "right": "issue:46050", + "accept": false, + "reason": "Gemma EOS token mismatch is unrelated to sinq quantization save/load breakage." + }, + { + "left": "issue:43646", + "right": "issue:44913", + "accept": false, + "reason": "Broad custom initialization regression vs GPTNeoX config field not round-tripping on reload." + }, + { + "left": "issue:45803", + "right": "issue:46365", + "accept": false, + "reason": "Different import/runtime problems: Fuyu tensor conversion exception handling vs Trainer import breaking with datasets installed." + }, + { + "left": "issue:44162", + "right": "issue:44792", + "accept": false, + "reason": "ESM2 model breakage and a janus image-generation test failure are not the same underlying issue." + }, + { + "left": "issue:43452", + "right": "issue:44496", + "accept": false, + "reason": "Both concern model loading, but one is gguf_file handling and the other is missing model_type in config.json." + }, + { + "left": "issue:43404", + "right": "issue:43441", + "accept": false, + "reason": "Weight tying in Mistral3 and FlashAttention support errors for Ministral-3 are distinct bugs." + }, + { + "left": "issue:44242", + "right": "issue:46894", + "accept": false, + "reason": "Both are training-loss issues, but they affect different models and different loss computation bugs." + }, + { + "left": "issue:45632", + "right": "issue:46424", + "accept": false, + "reason": "Cache path collision for trust_remote_code is unrelated to max_cache_length handling in quantized cache." + }, + { + "left": "issue:45803", + "right": "issue:46170", + "accept": false, + "reason": "Bare except swallowing errors is unrelated to Windows/ROCm import failures around torch.distributed." + }, + { + "left": "issue:44610", + "right": "issue:46369", + "accept": false, + "reason": "Processor input-size mismatch and sam3-litetext memory overflow are different failure modes." + }, + { + "left": "issue:43650", + "right": "issue:46202", + "accept": false, + "reason": "Placeholder/data issue vs Gemma pipeline round-trip tokenization bug; no shared underlying defect." + }, + { + "left": "issue:43761", + "right": "issue:44479", + "accept": false, + "reason": "Hidden-states regression in CLIPVisionModel is not the same as Qwen VL video-input regression." + }, + { + "left": "issue:43504", + "right": "issue:45698", + "accept": false, + "reason": "Both involve loading saved models, but one is a legacy field in BEiT presets and the other is wrong custom-module resolution after save/load." + }, + { + "left": "issue:45542", + "right": "issue:46234", + "accept": false, + "reason": "TensorFlow backend packaging issue vs missing import_utils symbol; unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Different Qwen multimodal bugs: video processing in Qwen3OmniMoe vs batched chat-template padding crash in Qwen2.5-VL." + }, + { + "left": "issue:43572", + "right": "issue:46178", + "accept": false, + "reason": "StableLm config field regression and tensorflow_text backend registration failure are unrelated." + }, + { + "left": "issue:43352", + "right": "issue:46640", + "accept": false, + "reason": "FlashAttention support errors for different models are not the same code-path problem." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 37, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:25251", + "issue:27957", + "issue:29870", + "issue:35141", + "issue:41950", + "issue:42490", + "issue:43065", + "issue:43388", + "issue:43475", + "issue:43525", + "issue:43531", + "issue:43644", + "issue:43673", + "issue:43688", + "issue:43827", + "issue:43931", + "issue:43976", + "issue:43994", + "issue:44060", + "issue:44190", + "issue:44242", + "issue:44534", + "issue:44617", + "issue:44734", + "issue:44743", + "issue:44792", + "issue:44855", + "issue:44869", + "issue:44913", + "issue:44933", + "issue:44964", + "issue:44977", + "issue:45127", + "issue:45200", + "issue:45245", + "issue:45362", + "issue:45406", + "issue:45584", + "issue:45588", + "issue:45600", + "issue:45632", + "issue:46082", + "issue:46421", + "issue:46617", + "issue:46697", + "issue:46832", + "issue:46858", + "issue:46897" + ], + "soft_pairs": [ + "issue:43065|issue:44617", + "issue:44792|issue:45600", + "issue:43388|issue:43688", + "issue:44792|issue:44977", + "issue:45127|issue:45200", + "issue:43673|issue:44743", + "issue:41950|issue:43475", + "issue:44190|issue:44964", + "issue:43525|issue:43531", + "issue:44534|issue:44913", + "issue:35141|issue:42490", + "issue:44734|issue:45406", + "issue:46082|issue:46697", + "issue:45632|issue:46617", + "issue:44855|issue:45588", + "issue:45245|issue:46858", + "issue:44869|issue:46832", + "issue:43827|issue:43994", + "issue:43673|issue:46421", + "issue:43931|issue:44060", + "issue:44242|issue:46897", + "issue:29870|issue:35141", + "issue:44617|issue:45584", + "issue:43644|issue:44913", + "issue:44933|issue:45362", + "issue:25251|issue:27957", + "issue:43976|issue:44855" + ], + "prepared_review_unit_hash": "75acb37c90654aa7fb4d2a4efbb4521e1b88608b9e66da2cf5e726ac7501f832", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "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": "No soft-edge pair looks like the same underlying bug/change; they mostly share only broad subsystem keywords. The clearest recurring theme in the set is the v5 non-persistent-buffer regression, best represented by the broader buffer-junk report.", + "confidence": 0.88, + "canonical_issue_reason": "issue:44534 is the most canonical issue in this set: it states the shared regression most broadly and cleanly, and is the best representative of the repeated non-persistent-buffer problem.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44534 is the strongest global representative because it captures a concrete, repository-wide v5 regression without overfitting to a narrower version string.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43065", + "right": "issue:44617", + "accept": false, + "reason": "Same model family (Sam3) but different failures: a dummy Conv2d implementation versus CUDA OOM in video inference." + }, + { + "left": "issue:44792", + "right": "issue:45600", + "accept": false, + "reason": "Different code paths: janus image-generation test failure versus CI breakage from removed Sam3LiteText config mappings." + }, + { + "left": "issue:43388", + "right": "issue:43688", + "accept": false, + "reason": "One is a dataloader/gather bug dropping labels; the other is an MoE auxiliary-loss normalization issue." + }, + { + "left": "issue:44792", + "right": "issue:44977", + "accept": false, + "reason": "Unrelated model-specific failures: janus test generation issue versus Qwen3.5 flash-attention generation behavior." + }, + { + "left": "issue:45127", + "right": "issue:45200", + "accept": false, + "reason": "LoRA merge / tied-embedding collapse is a different bug from Gemma 4 mm_token_type_ids defaulting in text-only fine-tuning." + }, + { + "left": "issue:43673", + "right": "issue:44743", + "accept": false, + "reason": "Both involve cache/state handling, but one is missing GenerationMixin cache in chunked prefill while the other resets recurrent states for Qwen3.5 modular code." + }, + { + "left": "issue:41950", + "right": "issue:43475", + "accept": false, + "reason": "Video-classification pipeline processor lookup is unrelated to a SAM 3 VisionEncoderOutput attribute error." + }, + { + "left": "issue:44190", + "right": "issue:44964", + "accept": false, + "reason": "Local dataset loading in a training script is different from a multimodal model load failure in latest transformers." + }, + { + "left": "issue:43525", + "right": "issue:43531", + "accept": false, + "reason": "LlamaConfig pad_token_id handling is unrelated to the Qwen3-MoE sliding_window bug." + }, + { + "left": "issue:44534", + "right": "issue:44913", + "accept": false, + "reason": "A v5 buffer-initialization regression is unrelated to GPTNeoXConfig rotary_pct not persisting on reload." + }, + { + "left": "issue:35141", + "right": "issue:42490", + "accept": false, + "reason": "Resized embeddings being reinitialized is a model-weight bug, not the local-path trailing-slash import warning." + }, + { + "left": "issue:44734", + "right": "issue:45406", + "accept": false, + "reason": "KV-cache continuation indexing failure in serve is unrelated to Gemma4Processor missing _tokenizer." + }, + { + "left": "issue:46082", + "right": "issue:46697", + "accept": false, + "reason": "LlamaConfig head_dim validation is a config constraint issue, not an image processor padding bug." + }, + { + "left": "issue:45632", + "right": "issue:46617", + "accept": false, + "reason": "Both touch trust_remote_code caching, but one is a leaf-directory cache collision and the other is a cache load FileNotFoundError; not the same bug." + }, + { + "left": "issue:44855", + "right": "issue:45588", + "accept": false, + "reason": "Python 3.13 torch.jit parsing/IndentationError is unrelated to flash-attention crashing on s_aux=None." + }, + { + "left": "issue:45245", + "right": "issue:46858", + "accept": false, + "reason": "Category cardinality overflow is unrelated to cache.get_seq_length() returning a device tensor in static-cache prefill." + }, + { + "left": "issue:44869", + "right": "issue:46832", + "accept": false, + "reason": "Whisper timestamp decoding crash and RT-DETR feature-level mismatch are different model-specific bugs." + }, + { + "left": "issue:43827", + "right": "issue:43994", + "accept": false, + "reason": "Stale docs referencing pipeline() are not the same as a model producing nonsense with AutoModel and pipeline." + }, + { + "left": "issue:43673", + "right": "issue:46421", + "accept": false, + "reason": "Missing cache in chunked prefill and cache lazy initialization causing recompilation are related only at a high level, not the same code-path bug." + }, + { + "left": "issue:43931", + "right": "issue:44060", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch is a load-time checkpoint issue, not the tied-weights warning in Qwen3-Next." + }, + { + "left": "issue:44242", + "right": "issue:46897", + "accept": false, + "reason": "Load-balancing loss omission and Florence2 double-shift training loss are different MoE/training bugs." + }, + { + "left": "issue:29870", + "right": "issue:35141", + "accept": false, + "reason": "Bart decoder_attention_mask initialization error is unrelated to post_init reinitializing output embeddings after resize." + }, + { + "left": "issue:44617", + "right": "issue:45584", + "accept": false, + "reason": "Sam3Video OOM and Whisper empty-transcription generation failure are unrelated failures." + }, + { + "left": "issue:43644", + "right": "issue:44913", + "accept": false, + "reason": "A v5 non-persistent-buffer regression is unrelated to GPTNeoXConfig rotary_pct persistence." + }, + { + "left": "issue:44933", + "right": "issue:45362", + "accept": false, + "reason": "An import from image_utils is a library API break, not the Qwen3.5 chat crash." + }, + { + "left": "issue:25251", + "right": "issue:27957", + "accept": false, + "reason": "top_k pipeline output nesting and XLMRoberta Flash Attention 2 are unrelated issues." + }, + { + "left": "issue:43976", + "right": "issue:44855", + "accept": false, + "reason": "Python-version packaging incompatibility is not the same as the DebertaV2Model torch.jit parsing error." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 38, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:25251", + "issue:29870", + "issue:39692", + "issue:41720", + "issue:43065", + "issue:43295", + "issue:43335", + "issue:43526", + "issue:43540", + "issue:43550", + "issue:43673", + "issue:43854", + "issue:44077", + "issue:44112", + "issue:44190", + "issue:44462", + "issue:44464", + "issue:44479", + "issue:44521", + "issue:44610", + "issue:44617", + "issue:44734", + "issue:44737", + "issue:44779", + "issue:44877", + "issue:44938", + "issue:44964", + "issue:45127", + "issue:45245", + "issue:45276", + "issue:45290", + "issue:45357", + "issue:45381", + "issue:45542", + "issue:45563", + "issue:45584", + "issue:45600", + "issue:45632", + "issue:45803", + "issue:45907", + "issue:46234", + "issue:46369", + "issue:46489", + "issue:46655", + "issue:46697", + "issue:46752", + "issue:46858" + ], + "soft_pairs": [ + "issue:45563|issue:45584", + "issue:43550|issue:44190", + "issue:45381|issue:46369", + "issue:44077|issue:44877", + "issue:41720|issue:45245", + "issue:44479|issue:44734", + "issue:43335|issue:44190", + "issue:43065|issue:44610", + "issue:44464|issue:45563", + "issue:39692|issue:44190", + "issue:45127|issue:45357", + "issue:45381|issue:46655", + "issue:44112|issue:45600", + "issue:44779|issue:46489", + "issue:43295|issue:45276", + "issue:44617|issue:46369", + "issue:43526|issue:46697", + "issue:45290|issue:46752", + "issue:44462|issue:45907", + "issue:45276|issue:45584", + "issue:43540|issue:44734", + "issue:44737|issue:45803", + "issue:44938|issue:45542", + "issue:25251|issue:29870", + "issue:44521|issue:45381", + "issue:43673|issue:46858", + "issue:43854|issue:44964", + "issue:45632|issue:46234" + ], + "prepared_review_unit_hash": "27bb6137e0ac92ccc387cd96c140244638c98ae80c7e82ea6f30d74306de031c", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35696, + "estimated_input_tokens": 8924, + "estimated_eval_tokens": 18104 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a grab-bag of unrelated Transformers bug reports spanning different models, generation paths, processors, configs, and environment issues. None of the soft pairs look like the same underlying bug or a mergeable duplicate PR.", + "confidence": 0.83, + "canonical_issue_reason": "No clear canonical issue: the items are mostly unrelated bugs/regressions across different subsystems, and none of the soft pairs describe the same underlying defect.", + "canonical_pr_reason": null, + "best_issue_reason": "If a representative issue must be chosen, issue:44521 is the most active/open and broadly user-facing, but it is not a true cluster canonical.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45563", + "right": "issue:45584", + "accept": false, + "reason": "Both are generation-related, but one is a stale warning and the other is an empty-transcription crash; different failure modes and code paths." + }, + { + "left": "issue:43550", + "right": "issue:44190", + "accept": false, + "reason": "A torch.compile SDPA failure and a local-dataset loading error in an example script are unrelated." + }, + { + "left": "issue:45381", + "right": "issue:46369", + "accept": false, + "reason": "Qwen video position-id corruption and Sam3LiteText memory overflow are different model-specific problems." + }, + { + "left": "issue:44077", + "right": "issue:44877", + "accept": false, + "reason": "One is about optional post_init behavior, the other about strict config loading; same area, not the same bug." + }, + { + "left": "issue:41720", + "right": "issue:45245", + "accept": false, + "reason": "cudaErrorAssert in Qwen3 auto device mapping is unrelated to the 2^24 category-limit runtime error." + }, + { + "left": "issue:44479", + "right": "issue:44734", + "accept": false, + "reason": "A video-input regression in Qwen VL models and a serve KV-cache indexing crash are distinct bugs." + }, + { + "left": "issue:43335", + "right": "issue:44190", + "accept": false, + "reason": "Sparse-layer construction in SwitchTransformers is unrelated to the image-classification dataset loading issue." + }, + { + "left": "issue:43065", + "right": "issue:44610", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder and a 640px-vs-224px processor mismatch are different code paths." + }, + { + "left": "issue:44464", + "right": "issue:45563", + "accept": false, + "reason": "Chunked generation inconsistencies with compiled forward are not the same as a stale warning for num_return_sequences." + }, + { + "left": "issue:39692", + "right": "issue:44190", + "accept": false, + "reason": "SigLIP2 documentation/example errors and local dataset loading failures are unrelated." + }, + { + "left": "issue:45127", + "right": "issue:45357", + "accept": false, + "reason": "LoRA/vocab-collapse after merge and incorrect visual-encoder keys in save_pretrained are separate model-save/merge bugs." + }, + { + "left": "issue:45381", + "right": "issue:46655", + "accept": false, + "reason": "Both touch video processing, but one is wrong vision_position_ids and the other is incorrect padding calculation." + }, + { + "left": "issue:44112", + "right": "issue:45600", + "accept": false, + "reason": "A stale CI test and broken auto_mappings references are unrelated maintenance issues." + }, + { + "left": "issue:44779", + "right": "issue:46489", + "accept": false, + "reason": "Both concern DeepSeek tokenizers, but they describe different regressions and cannot be safely merged as one bug." + }, + { + "left": "issue:43295", + "right": "issue:45276", + "accept": false, + "reason": "A processor/tokenizer regression and a Gemma4 embedding-resize bug are not the same defect." + }, + { + "left": "issue:44617", + "right": "issue:46369", + "accept": false, + "reason": "Sam3Video OOM and Sam3-LiteText memory overflow are different variants with different symptoms." + }, + { + "left": "issue:43526", + "right": "issue:46697", + "accept": false, + "reason": "Label reduction in BeitImageProcessorFast and Swin2SR over-padding are unrelated image-processing bugs." + }, + { + "left": "issue:45290", + "right": "issue:46752", + "accept": false, + "reason": "Both are chat-template crashes, but one is tool-call-specific and the other is an empty-list IndexError; too different to merge." + }, + { + "left": "issue:44462", + "right": "issue:45907", + "accept": false, + "reason": "AutoTokenizer ignoring tokenizer.json and list_repo_files forwarding revision are unrelated Hub behaviors." + }, + { + "left": "issue:45276", + "right": "issue:45584", + "accept": false, + "reason": "Gemma4 embedding resizing and Whisper empty-transcription handling are unrelated bugs." + }, + { + "left": "issue:43540", + "right": "issue:44734", + "accept": false, + "reason": "Qwen3OmniMoe video-input validation and serve KV-cache continuation indexing are different code paths." + }, + { + "left": "issue:44737", + "right": "issue:45803", + "accept": false, + "reason": "XLNet CPU tensor allocation and Fuyu bare-except error handling are unrelated implementation issues." + }, + { + "left": "issue:44938", + "right": "issue:45542", + "accept": false, + "reason": "Python 3.14 import/load failure and undefined tf backend due to tensorboard-only install are different environment issues." + }, + { + "left": "issue:25251", + "right": "issue:29870", + "accept": false, + "reason": "Pipeline top_k nesting behavior and Bart generate decoder_attention_mask unbound local error are unrelated generation bugs." + }, + { + "left": "issue:44521", + "right": "issue:45381", + "accept": false, + "reason": "All-zero assistant masks and wrong video vision_position_ids are both multimodal, but they are distinct bugs." + }, + { + "left": "issue:43673", + "right": "issue:46858", + "accept": false, + "reason": "Missing GenerationMixin cache in chunked_prefill and device-tensor seq_length arithmetic are separate cache/compile issues." + }, + { + "left": "issue:43854", + "right": "issue:44964", + "accept": false, + "reason": "Model-load failures for GLM-4.7-Flash and Phi-4 multimodal are different reports with different failure contexts." + }, + { + "left": "issue:45632", + "right": "issue:46234", + "accept": false, + "reason": "A trust_remote_code cache-path collision and a missing import_utils symbol are unrelated bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 39, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:27957", + "issue:32090", + "issue:35141", + "issue:36246", + "issue:38617", + "issue:41628", + "issue:41950", + "issue:42175", + "issue:42796", + "issue:42907", + "issue:43295", + "issue:43381", + "issue:43404", + "issue:43441", + "issue:43479", + "issue:43526", + "issue:43531", + "issue:43643", + "issue:43650", + "issue:43673", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44112", + "issue:44220", + "issue:44246", + "issue:44279", + "issue:44295", + "issue:44336", + "issue:44496", + "issue:44521", + "issue:44683", + "issue:44792", + "issue:44964", + "issue:44987", + "issue:45003", + "issue:45042", + "issue:45083", + "issue:45216", + "issue:45335", + "issue:45375", + "issue:45446", + "issue:45593", + "issue:45600", + "issue:46082", + "issue:46121", + "issue:46424", + "issue:46506" + ], + "soft_pairs": [ + "issue:44246|issue:45003", + "issue:44964|issue:45600", + "issue:43643|issue:44496", + "issue:44112|issue:44964", + "issue:42175|issue:44279", + "issue:27957|issue:35141", + "issue:46082|issue:46121", + "issue:36246|issue:41628", + "issue:42796|issue:44521", + "issue:36246|issue:38617", + "issue:43404|issue:45216", + "issue:32090|issue:35141", + "issue:44295|issue:46506", + "issue:41950|issue:44016", + "issue:43531|issue:45600", + "issue:43650|issue:44220", + "issue:43673|issue:46424", + "issue:43976|issue:44987", + "issue:43295|issue:45335", + "issue:44016|issue:44220", + "issue:45083|issue:46506", + "issue:42907|issue:43441", + "issue:44683|issue:45446", + "issue:43976|issue:44279", + "issue:43526|issue:43901", + "issue:43381|issue:45593", + "issue:43479|issue:45375", + "issue:44496|issue:45042", + "issue:43901|issue:44496", + "issue:43673|issue:44336", + "issue:44792|issue:46424" + ], + "prepared_review_unit_hash": "c19925f4c81758a65165a9f2496a60e7bd45907e65666c4d7a1ad97f4682a25c", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36934, + "estimated_input_tokens": 9234, + "estimated_eval_tokens": 18724 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a set of unrelated issues with only superficial lexical overlap (imports, configs, cache, multimodal, etc.). None of the soft pairs look like the same underlying bug or change, so no merges are justified. The strongest standalone issue here is the Phi-4 multimodal model-loading regression because it is concrete and reproducible.", + "confidence": 0.78, + "canonical_issue_reason": "issue:44964 is the most concrete, user-facing regression in the set: it names a specific model, a clear failure mode, and a current-transformers incompatibility, making it a reasonable representative issue for this cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44964 is the best issue candidate because it has a specific model, a specific broken workflow, and a clear reproduction target; it is more actionable than the broader import/config reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44246", + "right": "issue:45003", + "accept": false, + "reason": "Both mention import-related behavior, but one is about slow imports while the other is about unsafe sys.modules access; these are different bugs and code paths." + }, + { + "left": "issue:44964", + "right": "issue:45600", + "accept": false, + "reason": "Phi-4 multimodal loading failure and removed Sam3LiteText config references are unrelated problems in different parts of the library." + }, + { + "left": "issue:43643", + "right": "issue:44496", + "accept": false, + "reason": "One is a trust_remote_code config field loss issue; the other is an unrecognized model/model_type error. Similar theme, different root cause." + }, + { + "left": "issue:44112", + "right": "issue:44964", + "accept": false, + "reason": "A stale CI test for GraniteSpeech and a Phi-4 model load regression are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:44279", + "accept": false, + "reason": "Missing TensorFlow in the torch extra and a generic dependency issue are not the same concrete failure." + }, + { + "left": "issue:27957", + "right": "issue:35141", + "accept": false, + "reason": "FlashAttention support for XLMRoberta and token embedding reinitialization after resize are separate model behaviors." + }, + { + "left": "issue:46082", + "right": "issue:46121", + "accept": false, + "reason": "LlamaConfig head_dim validation and rope-parameter list handling are different configuration bugs." + }, + { + "left": "issue:36246", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but for different symbols and modules; no evidence they share the same underlying fix." + }, + { + "left": "issue:42796", + "right": "issue:44521", + "accept": false, + "reason": "Tokenizer BPE-marker handling and assistant-mask generation for multimodal chat templates are distinct issues." + }, + { + "left": "issue:36246", + "right": "issue:38617", + "accept": false, + "reason": "Different missing imports in different modules; they are not the same bug." + }, + { + "left": "issue:43404", + "right": "issue:45216", + "accept": false, + "reason": "lm_head weight tying in Mistral3 and Qwen3.5 checkpoint saving are separate model-specific regressions." + }, + { + "left": "issue:32090", + "right": "issue:35141", + "accept": false, + "reason": "Trainer GPU broadcast NoneType failure and post_init reinitializing output embeddings are unrelated." + }, + { + "left": "issue:44295", + "right": "issue:46506", + "accept": false, + "reason": "A position_ids buffer read error and CPU/CUDA output divergence are different failure modes." + }, + { + "left": "issue:41950", + "right": "issue:44016", + "accept": false, + "reason": "A pipeline using image processors and a notebook syntax error are not the same problem." + }, + { + "left": "issue:43531", + "right": "issue:45600", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and removed config references in auto_mappings.py are unrelated." + }, + { + "left": "issue:43650", + "right": "issue:44220", + "accept": false, + "reason": "The placeholder 'ADD THE DATA' issue does not correspond to the fbank feature extraction bug." + }, + { + "left": "issue:43673", + "right": "issue:46424", + "accept": false, + "reason": "A missing GenerationMixin cache in chunked_prefill and cache max-length handling for quantized cache are related only at a high level, not as the same concrete bug." + }, + { + "left": "issue:43976", + "right": "issue:44987", + "accept": false, + "reason": "Python version compatibility for Transformers 5.1.0 and failing to load a specific model are different issues." + }, + { + "left": "issue:43295", + "right": "issue:45335", + "accept": false, + "reason": "Processor/tokenizer regression and t5gemma decoder embedding resize behavior are separate code-path bugs." + }, + { + "left": "issue:44016", + "right": "issue:44220", + "accept": false, + "reason": "Notebook syntax error and feature-extraction helper behavior are unrelated." + }, + { + "left": "issue:45083", + "right": "issue:46506", + "accept": false, + "reason": "A helper-function behavior bug in qwen3_omni_moe and cross-device result differences are not the same issue." + }, + { + "left": "issue:42907", + "right": "issue:43441", + "accept": false, + "reason": "Saving dequantized Ministrals/Devstrals and FlashAttention failures in Ministral-3 are different problems, even if they touch the same model family." + }, + { + "left": "issue:44683", + "right": "issue:45446", + "accept": false, + "reason": "Compiled flex_attention failing on torch >= 2.9 and an incorrect version check for AuxRequest import are separate flex_attention issues." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Package Python-version compatibility and a generic dependency issue are not the same bug." + }, + { + "left": "issue:43526", + "right": "issue:43901", + "accept": false, + "reason": "BeitImageProcessorFast reduce_labels behavior and TextClassificationPipeline docs mismatch are unrelated." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Gradient checkpointing in eval mode and D-FINE auxiliary losses are different model/training concerns." + }, + { + "left": "issue:43479", + "right": "issue:45375", + "accept": false, + "reason": "Different multimodal config classes, different missing/defaulted fields, and different model families." + }, + { + "left": "issue:44496", + "right": "issue:45042", + "accept": false, + "reason": "Unrecognized model/model_type loading and PIL image processors incorrectly requiring torchvision are unrelated bugs." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "Pipeline documentation wording and model loading failure are not the same underlying issue." + }, + { + "left": "issue:43673", + "right": "issue:44336", + "accept": false, + "reason": "Cache handling and ANSI escape output from loading_report are completely different concerns." + }, + { + "left": "issue:44792", + "right": "issue:46424", + "accept": false, + "reason": "A failing Janus image-generation test and cache max-length semantics 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": 23, + "review_unit_index": 40, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:39692", + "issue:41950", + "issue:42796", + "issue:43295", + "issue:43299", + "issue:43441", + "issue:43577", + "issue:43650", + "issue:43653", + "issue:43824", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44062", + "issue:44075", + "issue:44242", + "issue:44279", + "issue:44327", + "issue:44351", + "issue:44368", + "issue:44466", + "issue:44496", + "issue:44521", + "issue:44779", + "issue:44933", + "issue:45020", + "issue:45200", + "issue:45216", + "issue:45276", + "issue:45278", + "issue:45335", + "issue:45357", + "issue:45405", + "issue:45584", + "issue:45588", + "issue:45593", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:46082", + "issue:46123", + "issue:46369", + "issue:46399", + "issue:46512", + "issue:46515", + "issue:46693", + "issue:46752" + ], + "soft_pairs": [ + "issue:44016|issue:44062", + "issue:44496|issue:45759", + "issue:43441|issue:45588", + "issue:43824|issue:44933", + "issue:46123|issue:46369", + "issue:44327|issue:46752", + "issue:44521|issue:46693", + "issue:44279|issue:46515", + "issue:45335|issue:45584", + "issue:43299|issue:45216", + "issue:44279|issue:45405", + "issue:46082|issue:46399", + "issue:43295|issue:44779", + "issue:44075|issue:44368", + "issue:43976|issue:45405", + "issue:39692|issue:43901", + "issue:43577|issue:45216", + "issue:44351|issue:45278", + "issue:46082|issue:46512", + "issue:42796|issue:43650", + "issue:44779|issue:45812", + "issue:41950|issue:45020", + "issue:43299|issue:45357", + "issue:44466|issue:45276", + "issue:43299|issue:43854", + "issue:43653|issue:45593", + "issue:41950|issue:45823", + "issue:44242|issue:45200" + ], + "prepared_review_unit_hash": "2b8f867207fee7b1603edabf832676bbec941c7058c41ad306046147829a8518", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36214, + "estimated_input_tokens": 9054, + "estimated_eval_tokens": 18364 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: it mixes tokenizer, model-loading, pipeline, config, and documentation regressions, but none of the soft-pair candidates look like the same underlying bug/change. The closest thing to an umbrella is a broad upgrade/import-regression report, not a true duplicate set.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 45278 is the broadest umbrella-style report here ('many import errors after update from 4.57.0 to 5.5.0'), so it is the closest anchor among otherwise unrelated bugs, though it is not a true duplicate of the others.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45278 is the best representative of the cluster only because it is the most general post-upgrade breakage report; the rest are narrower, concrete failures in different code paths.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44016", + "right": "issue:44062", + "accept": false, + "reason": "One is a notebook syntax error; the other is an AddedToken keyword conflict. Different failure modes and code paths." + }, + { + "left": "issue:44496", + "right": "issue:45759", + "accept": false, + "reason": "Both concern model loading, but one is unknown-model/config typing and the other is from_config failing to unwrap text_config for composite Qwen multimodal configs." + }, + { + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "Both mention FlashAttention, but one is a Ministral model-specific crash and the other is an AttributeError in flash_attention.py for sink-less models." + }, + { + "left": "issue:43824", + "right": "issue:44933", + "accept": false, + "reason": "Both are import errors, but they target different missing symbols in different modules." + }, + { + "left": "issue:46123", + "right": "issue:46369", + "accept": false, + "reason": "A batching/pipeline bug vs a model memory overflow; not the same underlying issue." + }, + { + "left": "issue:44327", + "right": "issue:46752", + "accept": false, + "reason": "Both crash on edge cases, but one is QA span top-k bounds and the other is apply_chat_template on an empty list." + }, + { + "left": "issue:44521", + "right": "issue:46693", + "accept": false, + "reason": "Assistant-mask generation for multimodal chat templates is unrelated to the FA2 long-context performance regression." + }, + { + "left": "issue:44279", + "right": "issue:46515", + "accept": false, + "reason": "A generic dependency issue vs a torch 2.4.1 incompatibility; different root causes." + }, + { + "left": "issue:45335", + "right": "issue:45584", + "accept": false, + "reason": "One is resize_token_embeddings not propagating in t5gemma; the other is Whisper generation failing after align_special_tokens." + }, + { + "left": "issue:43299", + "right": "issue:45216", + "accept": false, + "reason": "Qwen3VL loading failure vs Qwen3.5 save_pretrained serialization regression; related model family, but different concrete bugs." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "Dependency/install trouble vs a bumped MIN_PEFT_VERSION pointing to an unreleased package version." + }, + { + "left": "issue:46082", + "right": "issue:46399", + "accept": false, + "reason": "Config validation around head_dim is unrelated to a missing decorator for apply_rotary_pos_emb kernel registration." + }, + { + "left": "issue:43295", + "right": "issue:44779", + "accept": false, + "reason": "A regression in custom model code using processor.tokenizer/images vs Deepseek tokenizer output corruption; different APIs and symptoms." + }, + { + "left": "issue:44075", + "right": "issue:44368", + "accept": false, + "reason": "SGD args being ignored and a tie_word_embeddings warning are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:45405", + "accept": false, + "reason": "Python version support mismatch vs an unreleased PEFT minimum-version bump; different dependency problems." + }, + { + "left": "issue:39692", + "right": "issue:43901", + "accept": false, + "reason": "Different docs issues: SigLIP2 example errors vs TextClassificationPipeline docs/behavior mismatch." + }, + { + "left": "issue:43577", + "right": "issue:45216", + "accept": false, + "reason": "BLIP2 dtype propagation bug vs Qwen3.5 checkpoint serialization regression; not the same bug." + }, + { + "left": "issue:44351", + "right": "issue:45278", + "accept": false, + "reason": "Both are import failures after updates, but one is HybridCache specifically and the other is a broad set of import errors; not the same concrete defect." + }, + { + "left": "issue:46082", + "right": "issue:46512", + "accept": false, + "reason": "Both involve config handling, but one rejects a head_dim value while the other leaves model.config.dtype stale after from_config." + }, + { + "left": "issue:42796", + "right": "issue:43650", + "accept": false, + "reason": "Tokenizer BPE marker handling vs an empty/low-signal 'ADD THE DATA' issue; no shared bug." + }, + { + "left": "issue:44779", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and different token-ID/encoding logic." + }, + { + "left": "issue:41950", + "right": "issue:45020", + "accept": false, + "reason": "A video-classification processor lookup bug is not the same as a broad remote_code model-loading regression." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Qwen3VL loading vs Qwen3.5 save_pretrained still saving wrong visual-encoder keys; distinct regression points." + }, + { + "left": "issue:44466", + "right": "issue:45276", + "accept": false, + "reason": "lm_head.weight serialization differs by device, whereas this is about resize_token_embeddings not updating embeddings." + }, + { + "left": "issue:43299", + "right": "issue:43854", + "accept": false, + "reason": "Different models and failure sites: Qwen3VL loading vs GLM-4.7-Flash unit-test loading." + }, + { + "left": "issue:43653", + "right": "issue:45593", + "accept": false, + "reason": "BigBirdTokenizer special-token registration is unrelated to D-FINE losing auxiliary losses when denoising is off." + }, + { + "left": "issue:41950", + "right": "issue:45823", + "accept": false, + "reason": "Pipeline processor mismatch vs Gemma4 PLE device placement mismatch; different subsystems and symptoms." + }, + { + "left": "issue:44242", + "right": "issue:45200", + "accept": false, + "reason": "Load-balancing loss omission in MoE routing vs Gemma4 requiring mm_token_type_ids to default to zeros." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "Dependency issue vs unreleased PEFT minimum-version bump; 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": 23, + "review_unit_index": 41, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:33290", + "issue:42175", + "issue:42796", + "issue:42994", + "issue:43317", + "issue:43441", + "issue:43479", + "issue:43525", + "issue:43650", + "issue:43673", + "issue:43723", + "issue:43761", + "issue:43824", + "issue:43881", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44162", + "issue:44246", + "issue:44261", + "issue:44263", + "issue:44291", + "issue:44297", + "issue:44336", + "issue:44361", + "issue:44462", + "issue:44623", + "issue:44671", + "issue:44683", + "issue:44749", + "issue:44779", + "issue:45081", + "issue:45103", + "issue:45372", + "issue:45397", + "issue:45542", + "issue:45593", + "issue:45600", + "issue:45656", + "issue:45712", + "issue:45800", + "issue:45834", + "issue:45859", + "issue:45907", + "issue:46209", + "issue:46291", + "issue:46399", + "issue:46851" + ], + "soft_pairs": [ + "issue:43650|issue:46209", + "issue:44297|issue:44462", + "issue:43479|issue:43525", + "issue:44683|issue:45542", + "issue:44016|issue:44291", + "issue:44361|issue:45593", + "issue:45081|issue:45907", + "issue:33290|issue:45656", + "issue:43441|issue:45834", + "issue:42994|issue:43317", + "issue:44623|issue:44749", + "issue:44261|issue:44263", + "issue:45103|issue:45372", + "issue:44336|issue:46851", + "issue:45600|issue:45712", + "issue:45859|issue:46851", + "issue:42175|issue:43976", + "issue:44623|issue:45397", + "issue:44671|issue:44779", + "issue:43723|issue:44162", + "issue:43901|issue:44246", + "issue:43673|issue:43761", + "issue:43824|issue:43976", + "issue:45542|issue:45800", + "issue:42796|issue:43441", + "issue:43525|issue:43881", + "issue:46291|issue:46399" + ], + "prepared_review_unit_hash": "bb35b3bf788fe5a0c52de8dfee5a882c3e429fd426850c5b2c712c5b697c5506", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35703, + "estimated_input_tokens": 8926, + "estimated_eval_tokens": 18108 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a set of unrelated bugs/regressions that only look superficially similar by shared subsystem or wording. I would not merge any of the soft pairs as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": "No true duplicate cluster is present; if one issue must serve as an anchor, #44246 is the broadest/highest-engagement report and the least specific of the set.", + "canonical_pr_reason": null, + "best_issue_reason": "#44246 is the best triage anchor because it is a general, package-level performance issue with strong discussion activity, but it should not absorb the other reports as duplicates.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43650", + "right": "issue:46209", + "accept": false, + "reason": "'ADD THE DATA' is unrelated to FP8 inference on B200; no shared bug or code path." + }, + { + "left": "issue:44297", + "right": "issue:44462", + "accept": false, + "reason": "Tokenizer save mismatch is different from AutoTokenizer ignoring repository tokenizer.json; both involve tokenizers but not the same failure." + }, + { + "left": "issue:43479", + "right": "issue:43525", + "accept": false, + "reason": "Different models and bugs: Phi4Multimodal default sub-config initialization vs Llama4Config missing pad_token_id." + }, + { + "left": "issue:44683", + "right": "issue:45542", + "accept": false, + "reason": "Compiled flex_attention on torch>=2.9 is unrelated to missing TensorFlow backend installation." + }, + { + "left": "issue:44016", + "right": "issue:44291", + "accept": false, + "reason": "Notebook syntax error and init_empty_weights TypeError are unrelated issues." + }, + { + "left": "issue:44361", + "right": "issue:45593", + "accept": false, + "reason": "MLLukeTokenizer task AttributeError is distinct from D-FINE auxiliary loss behavior." + }, + { + "left": "issue:45081", + "right": "issue:45907", + "accept": false, + "reason": "Mistral regex patch crash is unrelated to list_repo_files revision forwarding." + }, + { + "left": "issue:33290", + "right": "issue:45656", + "accept": false, + "reason": "AdamFactor OOM in deepspeed and optimizer step called twice are different DeepSpeed optimizer problems." + }, + { + "left": "issue:43441", + "right": "issue:45834", + "accept": false, + "reason": "FlashAttention failure for Ministral-3 is unrelated to Kosmos2.5 long OCR index errors." + }, + { + "left": "issue:42994", + "right": "issue:43317", + "accept": false, + "reason": "Quantized model save failure and dequantized model load with cpu offload are different quantization lifecycle bugs." + }, + { + "left": "issue:44623", + "right": "issue:44749", + "accept": false, + "reason": "Missing processor save files and slow filtering after upgrade are unrelated." + }, + { + "left": "issue:44261", + "right": "issue:44263", + "accept": false, + "reason": "MLA q_a_layernorm epsilon precision issue is not the same as torch.split return values in GlmMoeDsaIndexer." + }, + { + "left": "issue:45103", + "right": "issue:45372", + "accept": false, + "reason": "auto_docstring future-annotations crash is unrelated to mistral_common ReasoningEffort import failure." + }, + { + "left": "issue:44336", + "right": "issue:46851", + "accept": false, + "reason": "ANSI codes in loading_report and Qwen3.5 variable-length position_ids are separate bugs." + }, + { + "left": "issue:45600", + "right": "issue:45712", + "accept": false, + "reason": "auto_mappings references removed configs, while dummy_pt_objects leaks dummy classes; shared repo area only, not same bug." + }, + { + "left": "issue:45859", + "right": "issue:46851", + "accept": false, + "reason": "CPU offload rotary embedding incompatibility is different from variable-length position_ids handling." + }, + { + "left": "issue:42175", + "right": "issue:43976", + "accept": false, + "reason": "Backend dependency omission with pip extras is unrelated to Python version support regression." + }, + { + "left": "issue:44623", + "right": "issue:45397", + "accept": false, + "reason": "Processor save missing files is not the same as gemma-4 zero3 from_pretrained loading behavior." + }, + { + "left": "issue:44671", + "right": "issue:44779", + "accept": false, + "reason": "CamemBERT masked LM predictions and Deepseek tokenizer incorrect results are different model/tokenizer regressions." + }, + { + "left": "issue:43723", + "right": "issue:44162", + "accept": false, + "reason": "Tokenizer loading failure in v5 and ESM2 breakage are separate model-family issues." + }, + { + "left": "issue:43901", + "right": "issue:44246", + "accept": false, + "reason": "Docs mentioning return_all_scores is a documentation mismatch, not the same as import time performance variance." + }, + { + "left": "issue:43673", + "right": "issue:43761", + "accept": false, + "reason": "Chunked prefill cache missing and CLIPVision hidden_states regression are different code paths." + }, + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "Qwen2.5-VL import failure is unrelated to Python version compatibility." + }, + { + "left": "issue:45542", + "right": "issue:45800", + "accept": false, + "reason": "TensorBoard/TensorFlow backend packaging issue is unrelated to torch/transformers version incompatibility." + }, + { + "left": "issue:42796", + "right": "issue:43441", + "accept": false, + "reason": "Ministral tokenizer BPE marker handling and FlashAttention failure are separate tokenizer vs attention bugs." + }, + { + "left": "issue:43525", + "right": "issue:43881", + "accept": false, + "reason": "Llama4Config missing pad_token_id and glm-4v-9b loading failure are distinct model config/loading issues." + }, + { + "left": "issue:46291", + "right": "issue:46399", + "accept": false, + "reason": "kerns v0.15 revision/version requirement is unrelated to Qwen3.5 rotary embedding registration crash." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 42, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:41628", + "issue:42491", + "issue:42886", + "issue:43299", + "issue:43525", + "issue:43526", + "issue:43531", + "issue:43540", + "issue:43643", + "issue:43784", + "issue:43976", + "issue:44016", + "issue:44188", + "issue:44242", + "issue:44246", + "issue:44295", + "issue:44297", + "issue:44336", + "issue:44373", + "issue:44451", + "issue:44485", + "issue:44496", + "issue:44589", + "issue:44871", + "issue:45003", + "issue:45072", + "issue:45083", + "issue:45103", + "issue:45325", + "issue:45362", + "issue:45397", + "issue:45542", + "issue:45561", + "issue:45563", + "issue:45588", + "issue:45593", + "issue:45823", + "issue:46132", + "issue:46399", + "issue:46614", + "issue:46655", + "issue:46665", + "issue:46726", + "issue:46752", + "issue:46821", + "issue:46829", + "issue:46851" + ], + "soft_pairs": [ + "issue:44188|issue:44336", + "issue:44016|issue:45823", + "issue:44589|issue:45362", + "issue:44336|issue:44373", + "issue:43525|issue:43643", + "issue:46655|issue:46726", + "issue:45325|issue:46851", + "issue:41628|issue:43976", + "issue:44297|issue:45397", + "issue:43784|issue:45542", + "issue:44246|issue:44496", + "issue:45083|issue:46851", + "issue:46399|issue:46851", + "issue:45561|issue:45563", + "issue:45103|issue:46399", + "issue:43526|issue:45003", + "issue:43540|issue:46752", + "issue:45072|issue:46829", + "issue:46614|issue:46665", + "issue:44295|issue:45397", + "issue:43531|issue:45362", + "issue:42491|issue:43299", + "issue:44246|issue:44871", + "issue:44242|issue:45593", + "issue:45588|issue:46851", + "issue:44336|issue:44485", + "issue:45003|issue:46821", + "issue:42886|issue:46132", + "issue:44451|issue:46752" + ], + "prepared_review_unit_hash": "5862de650ef5cbf4d9b5e6be7d67ab24b916a18f90bd8306599f68dc35198438", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35843, + "estimated_input_tokens": 8961, + "estimated_eval_tokens": 18178 + }, + "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 spanning Qwen/Gemma/other model regressions, tokenizer/processor edge cases, import/config problems, and a few docs/testing bugs. The apparent similarity is mostly lexical (shared model names or API terms), not evidence of true duplicates.", + "confidence": 0.67, + "canonical_issue_reason": "Issue 46851 is the most concrete and actionable report among the closest subtheme here (Qwen3.5 / position_ids / variable-length inference). No stronger true-duplicate anchor is evident from the titles alone.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46851 is the best representative issue to anchor this cluster because it is narrowly scoped, user-facing, and clearly describes a reproducible runtime failure.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "Different subsystems and failure modes: torch.compile attention-kernel divergence vs ANSI codes in loading_report." + }, + { + "left": "issue:44016", + "right": "issue:45823", + "accept": false, + "reason": "Notebook syntax error is unrelated to a Gemma4 forward device-mismatch bug." + }, + { + "left": "issue:44589", + "right": "issue:45362", + "accept": false, + "reason": "Float8 storage lookup failure and a Qwen3.5 chat crash are distinct issues." + }, + { + "left": "issue:44336", + "right": "issue:44373", + "accept": false, + "reason": "ANSI output in a loading report is not the same as a wrong docstring for position_ids." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": false, + "reason": "Both involve config fields, but one is Llama4Config pad_token_id and the other is trust_remote_code returning incomplete AutoConfig fields; different code paths." + }, + { + "left": "issue:46655", + "right": "issue:46726", + "accept": false, + "reason": "Both are vision-related, but one is video padding math in SmolVLM and the other is missing vision-backend handling in AutoProcessor." + }, + { + "left": "issue:45325", + "right": "issue:46851", + "accept": false, + "reason": "Same broad family name, but different bugs: temporal scaling in Qwen2.5-VL vs variable-length position_ids in Qwen3.5." + }, + { + "left": "issue:41628", + "right": "issue:43976", + "accept": false, + "reason": "AutoImageProcessor import error and Python version compatibility failure are unrelated." + }, + { + "left": "issue:44297", + "right": "issue:45397", + "accept": false, + "reason": "Tokenizer config save/load mismatch is not the same as a Gemma4 zero3 from_pretrained failure." + }, + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "Sentence-transformers import NameError and missing TensorFlow backend due to tensorboard-only install are different import/runtime problems." + }, + { + "left": "issue:44246", + "right": "issue:44496", + "accept": false, + "reason": "Import latency and unrecognized model config are not the same bug." + }, + { + "left": "issue:45083", + "right": "issue:46851", + "accept": false, + "reason": "Both mention Qwen position-related behavior, but one is a helper length calculation and the other is variable-length input handling; not the same underlying defect." + }, + { + "left": "issue:46399", + "right": "issue:46851", + "accept": false, + "reason": "Kernel registration decorator crash and variable-length position_ids are unrelated code paths." + }, + { + "left": "issue:45561", + "right": "issue:45563", + "accept": false, + "reason": "A pytest-xdist race in testing utils is unrelated to a stale generate() warning." + }, + { + "left": "issue:45103", + "right": "issue:46399", + "accept": false, + "reason": "Auto-docstring annotation handling and attention-kernel registration are separate issues." + }, + { + "left": "issue:43526", + "right": "issue:45003", + "accept": false, + "reason": "BeitImageProcessorFast label reduction and unsafe sys.modules access are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:46752", + "accept": false, + "reason": "Qwen3OmniMoe video-input processing and empty chat-template input handling are different bugs." + }, + { + "left": "issue:45072", + "right": "issue:46829", + "accept": false, + "reason": "dtype mismatch in CI inference is unrelated to VideoPrism returning the wrong hidden_states tensor." + }, + { + "left": "issue:46614", + "right": "issue:46665", + "accept": false, + "reason": "Stale phonemizer language state and inconsistent from_pretrained behavior inside Python classes do not describe the same defect." + }, + { + "left": "issue:44295", + "right": "issue:45397", + "accept": false, + "reason": "Reading a registered position_ids buffer incorrectly is not the same as a Gemma4 zero3 loading issue." + }, + { + "left": "issue:43531", + "right": "issue:45362", + "accept": false, + "reason": "Sliding_window behavior in Qwen3-MoE is distinct from a Qwen3.5 chat crash." + }, + { + "left": "issue:42491", + "right": "issue:43299", + "accept": false, + "reason": "Both are version-related Qwen/MoE regressions, but they concern different models and failure modes, so they should not be merged." + }, + { + "left": "issue:44246", + "right": "issue:44871", + "accept": false, + "reason": "Import slowness and Gemma-3 eos_token_id inconsistency are unrelated." + }, + { + "left": "issue:44242", + "right": "issue:45593", + "accept": false, + "reason": "Both mention missing auxiliary losses under flags, but they are different models and loss implementations; too broad to treat as duplicates." + }, + { + "left": "issue:45588", + "right": "issue:46851", + "accept": false, + "reason": "flash_attention s_aux handling and variable-length position_ids are unrelated bugs." + }, + { + "left": "issue:44336", + "right": "issue:44485", + "accept": false, + "reason": "ANSI codes in a loading report and GLM-5 RoPE implementation are unrelated." + }, + { + "left": "issue:45003", + "right": "issue:46821", + "accept": false, + "reason": "Unsafe sys.modules access and a TimesFM window_size AttributeError are different code paths." + }, + { + "left": "issue:42886", + "right": "issue:46132", + "accept": false, + "reason": "Offline tokenizer cache loading and AttentionInterface.register behavior are unrelated." + }, + { + "left": "issue:44451", + "right": "issue:46752", + "accept": false, + "reason": "Model loading failure for ScandiBERT is unrelated to apply_chat_template on an empty list." + } + ] + }, + "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": 23, + "review_unit_index": 43, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:29870", + "issue:30333", + "issue:36246", + "issue:41950", + "issue:42757", + "issue:43257", + "issue:43262", + "issue:43295", + "issue:43475", + "issue:43526", + "issue:43540", + "issue:43550", + "issue:43596", + "issue:43901", + "issue:44038", + "issue:44162", + "issue:44220", + "issue:44242", + "issue:44246", + "issue:44295", + "issue:44351", + "issue:44466", + "issue:44479", + "issue:44496", + "issue:44623", + "issue:44749", + "issue:44779", + "issue:44857", + "issue:44877", + "issue:44987", + "issue:45003", + "issue:45137", + "issue:45200", + "issue:45292", + "issue:45335", + "issue:45375", + "issue:45459", + "issue:45561", + "issue:45584", + "issue:45600", + "issue:45759", + "issue:45859", + "issue:46369", + "issue:46399", + "issue:46506", + "issue:46655", + "issue:46726", + "issue:46897" + ], + "soft_pairs": [ + "issue:43550|issue:46506", + "issue:44779|issue:45292", + "issue:36246|issue:42757", + "issue:36246|issue:44351", + "issue:41950|issue:44220", + "issue:43901|issue:45003", + "issue:43257|issue:45137", + "issue:45003|issue:45561", + "issue:44496|issue:45003", + "issue:43540|issue:46655", + "issue:44162|issue:45200", + "issue:45375|issue:45759", + "issue:44479|issue:46655", + "issue:43262|issue:44246", + "issue:44295|issue:44749", + "issue:43475|issue:46369", + "issue:45859|issue:46399", + "issue:45137|issue:45600", + "issue:36246|issue:43596", + "issue:44877|issue:45375", + "issue:43295|issue:45292", + "issue:45292|issue:45584", + "issue:44623|issue:46726", + "issue:44466|issue:45335", + "issue:44242|issue:44857", + "issue:43526|issue:46897", + "issue:44038|issue:46506", + "issue:44987|issue:45459", + "issue:29870|issue:30333" + ], + "prepared_review_unit_hash": "639e7ddbde4d1a8b7d7da9db914fd7200fe30648d22e0c1f23051f6eaa3fe6a1", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36555, + "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 heterogeneous: it mixes unrelated bugs across model loading, processors, configs, generation, deepspeed, docs, and CI. None of the soft pairs appear to be the same underlying issue, so all soft edges should be rejected.", + "confidence": 0.96, + "canonical_issue_reason": "No clear canonical issue for deduplication: the items do not form one duplicate family and span many unrelated subsystems/models.", + "canonical_pr_reason": null, + "best_issue_reason": "If a representative issue is needed, issue:44479 is the broadest concrete regression in the set, but it is not a true cluster canonical.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43550", + "right": "issue:46506", + "accept": false, + "reason": "Different failures in different models/subsystems: Bamba SDPA torch.compile bug vs PPDocLayoutV3 CPU/CUDA mismatch." + }, + { + "left": "issue:44779", + "right": "issue:45292", + "accept": false, + "reason": "Tokenizer corruption in DeepSeek is unrelated to resize_token_embeddings not updating output embeddings." + }, + { + "left": "issue:36246", + "right": "issue:42757", + "accept": false, + "reason": "Different missing imports from different packages: Qwen2.5 VL image processor vs huggingface_hub is_offline_mode." + }, + { + "left": "issue:36246", + "right": "issue:44351", + "accept": false, + "reason": "Different missing symbols and code paths: Qwen2.5 VL image processor vs HybridCache import." + }, + { + "left": "issue:41950", + "right": "issue:44220", + "accept": false, + "reason": "Video pipeline image-processor lookup bug is unrelated to audio fbank extraction issues." + }, + { + "left": "issue:43901", + "right": "issue:45003", + "accept": false, + "reason": "Docs mismatch for return_all_scores is unrelated to unsafe sys.modules access in modeling_utils." + }, + { + "left": "issue:43257", + "right": "issue:45137", + "accept": false, + "reason": "Both involve DeepSpeed, but one is MOE weight conversion and the other is an empty deque crash; different bugs." + }, + { + "left": "issue:45003", + "right": "issue:45561", + "accept": false, + "reason": "Unsafe sys.modules access is unrelated to pytest-xdist captured_info.txt race conditions." + }, + { + "left": "issue:44496", + "right": "issue:45003", + "accept": false, + "reason": "Unrecognized model config loading is a different problem from sys.modules access." + }, + { + "left": "issue:43540", + "right": "issue:46655", + "accept": false, + "reason": "Both are video-related, but Qwen3OmniMoe ValueError and SmolVLM padding bug have different model paths and root causes." + }, + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "ESM2 breakage is unrelated to Gemma 4 mm_token_type_ids default behavior." + }, + { + "left": "issue:45375", + "right": "issue:45759", + "accept": false, + "reason": "Both mention Qwen multimodal configs, but one drops a missing field under strict validation and the other fails to unwrap text_config; not the same bug." + }, + { + "left": "issue:44479", + "right": "issue:46655", + "accept": false, + "reason": "Different models and code paths: Qwen VL video regression vs SmolVLM video processor padding." + }, + { + "left": "issue:43262", + "right": "issue:44246", + "accept": false, + "reason": "Audio sampling-rate defaults are unrelated to import latency." + }, + { + "left": "issue:44295", + "right": "issue:44749", + "accept": false, + "reason": "Position_ids buffer access error is unrelated to a transformers version slowdown." + }, + { + "left": "issue:43475", + "right": "issue:46369", + "accept": false, + "reason": "SAM3 video missing attribute and SAM3LiteText memory overflow are distinct issues." + }, + { + "left": "issue:45859", + "right": "issue:46399", + "accept": false, + "reason": "Both are Qwen3.5-related, but CPU offload incompatibility and missing registration decorator are different concrete failures." + }, + { + "left": "issue:45137", + "right": "issue:45600", + "accept": false, + "reason": "ZeRO3 deque crash is unrelated to auto_mappings referencing removed Sam3LiteText configs." + }, + { + "left": "issue:36246", + "right": "issue:43596", + "accept": false, + "reason": "Missing Qwen2.5 VL import and DeepSpeed/BertModel index error are unrelated." + }, + { + "left": "issue:44877", + "right": "issue:45375", + "accept": false, + "reason": "Strict config rejection for granite_speech is unrelated to Qwen3_5MoeVisionConfig field dropping." + }, + { + "left": "issue:43295", + "right": "issue:45292", + "accept": false, + "reason": "Processor/tokenizer regression is unrelated to resize_token_embeddings not updating output embeddings." + }, + { + "left": "issue:45292", + "right": "issue:45584", + "accept": false, + "reason": "Embedding resize bug is unrelated to Whisper empty transcription after align_special_tokens." + }, + { + "left": "issue:44623", + "right": "issue:46726", + "accept": false, + "reason": "Missing files on processor.save_pretrained is unrelated to AutoProcessor complaining about a missing vision backend." + }, + { + "left": "issue:44466", + "right": "issue:45335", + "accept": false, + "reason": "Serialization of lm_head.weight is unrelated to t5gemma decoder.embed_tokens not updating on resize." + }, + { + "left": "issue:44242", + "right": "issue:44857", + "accept": false, + "reason": "MoE load-balancing loss omission is unrelated to LwDetrImageLoss AMP/CUDA crashes." + }, + { + "left": "issue:43526", + "right": "issue:46897", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug is unrelated to Florence2 training-loss double-shift." + }, + { + "left": "issue:44038", + "right": "issue:46506", + "accept": false, + "reason": "Qwen3-VL-Moe regression and PPDocLayoutV3 CPU/CUDA mismatch are different bugs." + }, + { + "left": "issue:44987", + "right": "issue:45459", + "accept": false, + "reason": "Loading physical-intelligence/fast and hidden tokenizer errors from protobuf absence are unrelated." + }, + { + "left": "issue:29870", + "right": "issue:30333", + "accept": false, + "reason": "Bart generate UnboundLocalError is unrelated to MLflow not being reported after failed training jobs." + } + ] + }, + "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": 23, + "review_unit_index": 44, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:42199", + "issue:42886", + "issue:43065", + "issue:43352", + "issue:43408", + "issue:43475", + "issue:43526", + "issue:43540", + "issue:43653", + "issue:43723", + "issue:43906", + "issue:43931", + "issue:44060", + "issue:44077", + "issue:44162", + "issue:44188", + "issue:44297", + "issue:44368", + "issue:44479", + "issue:44483", + "issue:44623", + "issue:44877", + "issue:44977", + "issue:44995", + "issue:45103", + "issue:45200", + "issue:45229", + "issue:45250", + "issue:45397", + "issue:45412", + "issue:45563", + "issue:45588", + "issue:45656", + "issue:45685", + "issue:45987", + "issue:46082", + "issue:46123", + "issue:46170", + "issue:46202", + "issue:46421", + "issue:46431", + "issue:46710", + "issue:46829", + "issue:46832", + "issue:46851", + "issue:46858", + "issue:46894" + ], + "soft_pairs": [ + "issue:43352|issue:45250", + "issue:42199|issue:46851", + "issue:43526|issue:46894", + "issue:44060|issue:44077", + "issue:44368|issue:44977", + "issue:43653|issue:44162", + "issue:45412|issue:46123", + "issue:44479|issue:46170", + "issue:42886|issue:45229", + "issue:45397|issue:45656", + "issue:43065|issue:43408", + "issue:43475|issue:43540", + "issue:46170|issue:46832", + "issue:44877|issue:46170", + "issue:44368|issue:45563", + "issue:43723|issue:46710", + "issue:44995|issue:45200", + "issue:43931|issue:44368", + "issue:44188|issue:46202", + "issue:44483|issue:46202", + "issue:46829|issue:46858", + "issue:45987|issue:46082", + "issue:43906|issue:46202", + "issue:44995|issue:45588", + "issue:44188|issue:46421", + "issue:45685|issue:46431", + "issue:45103|issue:46858", + "issue:43065|issue:46123", + "issue:44297|issue:44623" + ], + "prepared_review_unit_hash": "7e84ec6cc689ec9025e5e3ef6744f949f816a0ada7b1834d5fb0edc8b61693ed", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35966, + "estimated_input_tokens": 8992, + "estimated_eval_tokens": 18240 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: tokenizer/loading, multimodal model config, cache/compile, and generation bugs. Only the two StaticCache.get_seq_length reports look like the same underlying defect; the rest are distinct issues that should stay separate.", + "confidence": 0.91, + "canonical_issue_reason": "issue:46858 is the best canonical because it states the root cause and concrete failure mode most clearly: get_seq_length() returning a device tensor breaks slice arithmetic under static cache + torch.compile on non-CUDA backends. issue:45987 is the narrower contract complaint that this subsumes.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:46858 is the strongest representative of the only true duplicate pair in this set: it is specific, actionable, and covers the broader downstream impact of the same get_seq_length bug.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43352", + "right": "issue:45250", + "accept": false, + "reason": "Both mention Flash Attention 2.0, but one is a model-support error for NemotronH and the other is a generic topic stub; not the same bug or change." + }, + { + "left": "issue:42199", + "right": "issue:46851", + "accept": false, + "reason": "Different models and failures: DETR cardinality/background-class handling vs Qwen3.5 variable-length position_ids." + }, + { + "left": "issue:43526", + "right": "issue:46894", + "accept": false, + "reason": "Unrelated bugs in different processors/models: BeitImageProcessorFast label reduction vs CohereASR training-loss double shift." + }, + { + "left": "issue:44060", + "right": "issue:44077", + "accept": false, + "reason": "Different subsystems and symptoms: Qwen3-Next tied-weights warning vs patchtsmixer post_init policy." + }, + { + "left": "issue:44368", + "right": "issue:44977", + "accept": false, + "reason": "Same family name (Qwen3.5) but distinct problems: tied-word-embeddings warning vs flash-attention generation failure." + }, + { + "left": "issue:43653", + "right": "issue:44162", + "accept": false, + "reason": "Tokenizer special-token registration bug vs ESM2 regression; no shared code path." + }, + { + "left": "issue:45412", + "right": "issue:46123", + "accept": false, + "reason": "RT-DETR memory cleanup vs MaskGenerationPipeline partial-batch handling; different behavior and code paths." + }, + { + "left": "issue:44479", + "right": "issue:46170", + "accept": false, + "reason": "Video-input regression in Qwen VL models and a Windows ROCm/torch.distributed import problem are unrelated." + }, + { + "left": "issue:42886", + "right": "issue:45229", + "accept": false, + "reason": "Offline tokenizer cache loading and Gemma4 multi-GPU CUDA OOM are unrelated issues." + }, + { + "left": "issue:45397", + "right": "issue:45656", + "accept": false, + "reason": "Gemma-4 ZeRO-3 loading vs DeepSpeed optimizer-step duplication are different training/runtime bugs." + }, + { + "left": "issue:43065", + "right": "issue:43408", + "accept": false, + "reason": "Both are SAM3-related, but one is a dummy Conv2d in the pixel decoder and the other is a model-type mismatch warning." + }, + { + "left": "issue:43475", + "right": "issue:43540", + "accept": false, + "reason": "SAM3 video encoder missing attribute vs Qwen3OmniMoe video-input ValueError; distinct failures." + }, + { + "left": "issue:46170", + "right": "issue:46832", + "accept": false, + "reason": "Windows ROCm import/type dependency issue vs RT-DETR forward shape/type error; unrelated." + }, + { + "left": "issue:44877", + "right": "issue:46170", + "accept": false, + "reason": "Strict granite_speech config loading and Windows ROCm dependency failures are not the same bug." + }, + { + "left": "issue:44368", + "right": "issue:45563", + "accept": false, + "reason": "Both are warnings, but one is a tied-embeddings config warning and the other is a stale num_return_sequences warning in paged generate." + }, + { + "left": "issue:43723", + "right": "issue:46710", + "accept": false, + "reason": "Tokenizer loading in v5 and an accuracy regression after upgrade are different classes of issue." + }, + { + "left": "issue:44995", + "right": "issue:45200", + "accept": false, + "reason": "GlmMoeDsa second-forward cache bug vs Gemma 4 mm_token_type_ids defaulting; unrelated." + }, + { + "left": "issue:43931", + "right": "issue:44368", + "accept": false, + "reason": "Qwen3-VL weight-shape loading mismatch vs Qwen3.5 tied-embeddings warning; different code paths." + }, + { + "left": "issue:44188", + "right": "issue:46202", + "accept": false, + "reason": "torch.compile branching in attention kernels and Gemma token roundtrip on \u2581 are unrelated." + }, + { + "left": "issue:44483", + "right": "issue:46202", + "accept": false, + "reason": "Chat/completions API rejection vs tokenizer roundtrip failure; no shared underlying defect." + }, + { + "left": "issue:46829", + "right": "issue:46858", + "accept": false, + "reason": "VideoPrism output field mix-up is a model-output bug, not the StaticCache get_seq_length contract problem." + }, + { + "left": "issue:45987", + "right": "issue:46082", + "accept": false, + "reason": "StaticCache sequence-length return type and LlamaConfig head_dim validation are different issues." + }, + { + "left": "issue:43906", + "right": "issue:46202", + "accept": false, + "reason": "Isolated reproduction of another issue vs Gemma roundtrip failure; not the same bug." + }, + { + "left": "issue:44995", + "right": "issue:45588", + "accept": false, + "reason": "Both are runtime bugs, but one is stale indexer cache and the other is flash_attention s_aux=None handling." + }, + { + "left": "issue:44188", + "right": "issue:46421", + "accept": false, + "reason": "Both mention compile/cache behavior, but one is kernel divergence under torch.compile and the other is lazy cache init causing recompilation." + }, + { + "left": "issue:45685", + "right": "issue:46431", + "accept": false, + "reason": "MPS histogram on Int vs missing chat template in a processor are unrelated." + }, + { + "left": "issue:45103", + "right": "issue:46858", + "accept": false, + "reason": "Docstring annotation handling and StaticCache get_seq_length return type are separate bugs." + }, + { + "left": "issue:43065", + "right": "issue:46123", + "accept": false, + "reason": "SAM3 pixel-decoder dummy layer and MaskGenerationPipeline partial-batch handling are unrelated." + }, + { + "left": "issue:44297", + "right": "issue:44623", + "accept": false, + "reason": "Both involve save_pretrained, but tokenizer_class metadata mismatch and processor missing files are different serialization 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": 23, + "review_unit_index": 45, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:28282", + "issue:30990", + "issue:39692", + "issue:41628", + "issue:42371", + "issue:42994", + "issue:43232", + "issue:43441", + "issue:43475", + "issue:43482", + "issue:43525", + "issue:43643", + "issue:43646", + "issue:43673", + "issue:43701", + "issue:43723", + "issue:43986", + "issue:43994", + "issue:44038", + "issue:44162", + "issue:44188", + "issue:44361", + "issue:44373", + "issue:44623", + "issue:44749", + "issue:44869", + "issue:44871", + "issue:44933", + "issue:44964", + "issue:45092", + "issue:45137", + "issue:45230", + "issue:45310", + "issue:45356", + "issue:45375", + "issue:45397", + "issue:45406", + "issue:45478", + "issue:45656", + "issue:46132", + "issue:46202", + "issue:46326", + "issue:46463", + "issue:46726", + "issue:46752", + "issue:46821", + "issue:46851", + "issue:46877" + ], + "soft_pairs": [ + "issue:44361|issue:44749", + "issue:43994|issue:46726", + "issue:44749|issue:45656", + "issue:43701|issue:44749", + "issue:44871|issue:46132", + "issue:44623|issue:45656", + "issue:43646|issue:44162", + "issue:43441|issue:46752", + "issue:42994|issue:43482", + "issue:45375|issue:45478", + "issue:43475|issue:43643", + "issue:43232|issue:43673", + "issue:44188|issue:46851", + "issue:28282|issue:30990", + "issue:43475|issue:46821", + "issue:41628|issue:44933", + "issue:45310|issue:45375", + "issue:45230|issue:45397", + "issue:43986|issue:46726", + "issue:44964|issue:45137", + "issue:44869|issue:45356", + "issue:44038|issue:44623", + "issue:39692|issue:46726", + "issue:46463|issue:46877", + "issue:44373|issue:46851", + "issue:42994|issue:43525", + "issue:42371|issue:45092", + "issue:43723|issue:46326", + "issue:45406|issue:46202" + ], + "prepared_review_unit_hash": "75c1f2ebdb055c603ba94d11ddb851d40df412e844b261505250fc6903d40ff8", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36056, + "estimated_input_tokens": 9014, + "estimated_eval_tokens": 18284 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is a grab bag of unrelated Transformers issues; none of the soft-edge pairs look like the same concrete bug or fix. Most similarities are broad subsystem/model-family overlap rather than duplicate reports.", + "confidence": 0.91, + "canonical_issue_reason": "issue:44162 is the strongest canonical issue candidate because it describes a severe, clearly stated regression with broad user impact; the other issues are narrower, more model-specific, or more ambiguous.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44162 is the best overall issue for triage quality and impact: it is concrete, high-severity, and well-scoped compared with the rest.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44361", + "right": "issue:44749", + "accept": false, + "reason": "Different bugs: tokenizer/task AttributeError vs a performance regression after upgrading Transformers." + }, + { + "left": "issue:43994", + "right": "issue:46726", + "accept": false, + "reason": "Both involve vision/AutoProcessor paths, but one is wrong SigLIP2 outputs and the other is a missing-backend error message." + }, + { + "left": "issue:44749", + "right": "issue:45656", + "accept": false, + "reason": "Unrelated problems: data filtering slowdown versus DeepSpeed optimizer stepping twice." + }, + { + "left": "issue:43701", + "right": "issue:44749", + "accept": false, + "reason": "Checkpoint key mismatch is a separate training/resume issue, not the same as the performance regression." + }, + { + "left": "issue:44871", + "right": "issue:46132", + "accept": false, + "reason": "Different code paths: Gemma EOS token config inconsistency vs AttentionInterface registration behavior." + }, + { + "left": "issue:44623", + "right": "issue:45656", + "accept": false, + "reason": "Saving processor files is unrelated to the DeepSpeed optimizer-step bug." + }, + { + "left": "issue:43646", + "right": "issue:44162", + "accept": false, + "reason": "AutoConfig trust_remote_code field loss is a different bug from the ESM2 regression." + }, + { + "left": "issue:43441", + "right": "issue:46752", + "accept": false, + "reason": "FlashAttention/Ministral failure and empty-chat-template IndexError are unrelated." + }, + { + "left": "issue:42994", + "right": "issue:43482", + "accept": false, + "reason": "Both are quantization/model-loading themed, but one is saving a quantized model and the other is loading a GGUF model." + }, + { + "left": "issue:45375", + "right": "issue:45478", + "accept": false, + "reason": "Same model family, but one is a missing config field dropped by strict validation and the other is a from_pretrained error; not clearly the same bug." + }, + { + "left": "issue:43475", + "right": "issue:43643", + "accept": false, + "reason": "SAM video encoder attribute failure is unrelated to AutoConfig remote-code field handling." + }, + { + "left": "issue:43232", + "right": "issue:43673", + "accept": false, + "reason": "Both are generation-related, but they describe different failures in different parts of generation state handling." + }, + { + "left": "issue:44188", + "right": "issue:46851", + "accept": false, + "reason": "Attention kernel divergence under torch.compile is not the same as variable-length position_ids handling in Qwen3.5." + }, + { + "left": "issue:28282", + "right": "issue:30990", + "accept": false, + "reason": "Missing PyTorch import error is unrelated to Sentence Transformers hanging during load." + }, + { + "left": "issue:43475", + "right": "issue:46821", + "accept": false, + "reason": "Different models and failures: SAM3 vision encoder attribute error vs TimesFM window_size attribute error." + }, + { + "left": "issue:41628", + "right": "issue:44933", + "accept": false, + "reason": "Both mention import/export issues, but the specific missing symbols and failure modes are not clearly the same." + }, + { + "left": "issue:45310", + "right": "issue:45375", + "accept": false, + "reason": "Related Qwen3.5 topics, but one is a from_pretrained error and the other is a strict-validation field-drop bug." + }, + { + "left": "issue:45230", + "right": "issue:45397", + "accept": false, + "reason": "Generic bug report is not the same concrete issue as Gemma-4 ZeRO3 from_pretrained failure." + }, + { + "left": "issue:43986", + "right": "issue:46726", + "accept": false, + "reason": "Both involve missing vision dependencies, but one is a video-model AutoProcessor crash and the other is an image-processor recognition error." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Multimodal model loading failure is unrelated to DeepSpeed ZeRO3 deque underflow." + }, + { + "left": "issue:44869", + "right": "issue:45356", + "accept": false, + "reason": "Whisper decode crash from replacement character is a different tokenizer bug than Kimi-K2.5 codec handling regression." + }, + { + "left": "issue:44038", + "right": "issue:44623", + "accept": false, + "reason": "Qwen3-VL-Moe loading error and processor.save_pretrained missing files are distinct failures." + }, + { + "left": "issue:39692", + "right": "issue:46726", + "accept": false, + "reason": "Documentation example errors are not the same as the AutoProcessor missing-backend crash." + }, + { + "left": "issue:46463", + "right": "issue:46877", + "accept": false, + "reason": "Incorrect T5 validation error message is unrelated to auto_docstring printing diagnostics at import time." + }, + { + "left": "issue:44373", + "right": "issue:46851", + "accept": false, + "reason": "A wrong docstring is not the same as a functional bug in position_ids handling." + }, + { + "left": "issue:42994", + "right": "issue:43525", + "accept": false, + "reason": "Quantized model saving failure and missing pad_token_id on Llama4Config are separate issues." + }, + { + "left": "issue:42371", + "right": "issue:45092", + "accept": false, + "reason": "TF32 settings guidance is unrelated to InternVL2 meta-initialization incompatibility." + }, + { + "left": "issue:43723", + "right": "issue:46326", + "accept": false, + "reason": "AutoTokenizer loading regression and continue_final_message behavior for mistral_common tokenizers are different bugs." + }, + { + "left": "issue:45406", + "right": "issue:46202", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer and Gemma pipeline roundtrip failure 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": 23, + "review_unit_index": 46, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:30333", + "issue:30990", + "issue:33290", + "issue:35141", + "issue:36010", + "issue:41628", + "issue:42175", + "issue:42673", + "issue:42831", + "issue:42898", + "issue:42994", + "issue:43232", + "issue:43278", + "issue:43479", + "issue:43504", + "issue:43704", + "issue:43784", + "issue:43824", + "issue:43873", + "issue:43976", + "issue:44038", + "issue:44222", + "issue:44351", + "issue:44496", + "issue:44568", + "issue:44623", + "issue:44792", + "issue:44857", + "issue:44933", + "issue:44945", + "issue:45081", + "issue:45103", + "issue:45137", + "issue:45310", + "issue:45405", + "issue:45561", + "issue:45593", + "issue:45600", + "issue:45712", + "issue:45753", + "issue:46077", + "issue:46123", + "issue:46202", + "issue:46399", + "issue:46693", + "issue:46701", + "issue:46829", + "issue:46877" + ], + "soft_pairs": [ + "issue:43873|issue:44945", + "issue:44945|issue:45753", + "issue:43784|issue:46399", + "issue:41628|issue:44351", + "issue:44857|issue:46877", + "issue:45593|issue:46877", + "issue:42673|issue:43704", + "issue:44038|issue:45310", + "issue:44222|issue:45310", + "issue:36010|issue:41628", + "issue:46202|issue:46701", + "issue:42994|issue:43824", + "issue:43278|issue:45137", + "issue:45081|issue:45103", + "issue:42831|issue:44222", + "issue:43232|issue:44792", + "issue:42175|issue:44933", + "issue:43504|issue:45561", + "issue:45561|issue:45600", + "issue:43976|issue:44933", + "issue:44496|issue:45561", + "issue:30333|issue:33290", + "issue:45405|issue:46877", + "issue:30990|issue:35141", + "issue:42898|issue:44568", + "issue:42175|issue:45712", + "issue:43479|issue:46829", + "issue:46077|issue:46123", + "issue:44623|issue:45310", + "issue:42673|issue:46693" + ], + "prepared_review_unit_hash": "fb74c9f6b450fa7ae2564345acba5b29eb6b1558545bab7f0a3393c14df2dd97", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36536, + "estimated_input_tokens": 9134, + "estimated_eval_tokens": 18524 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous and does not form a single coherent duplicate set. The only clear duplicate-looking pair is the repeated Qwen3ForCausalLM VRAM leak report (42673/43704).", + "confidence": 0.74, + "canonical_issue_reason": "Issue 42673 is the earlier, fuller report of the exact duplicate pair with 43704 (same Qwen3ForCausalLM VRAM leak title and symptom), so it is the best canonical issue candidate.", + "canonical_pr_reason": null, + "best_issue_reason": "42673 is the clearest standalone representative in this noisy cluster because it matches 43704 exactly and has more discussion/context than the later duplicate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43873", + "right": "issue:44945", + "accept": false, + "reason": "Both are generation/runtime complaints, but one is quantization/offloading and the other is pipeline-parallel output corruption; different bugs." + }, + { + "left": "issue:44945", + "right": "issue:45753", + "accept": false, + "reason": "Different failure modes: pipeline parallelism output issue vs Qwen3_5 image-specific infinite loop." + }, + { + "left": "issue:43784", + "right": "issue:46399", + "accept": false, + "reason": "Import NameError in sentence-transformers vs Qwen3.5 kernel registration ValueError; unrelated code paths." + }, + { + "left": "issue:41628", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different symbols/classes and likely different breakages." + }, + { + "left": "issue:44857", + "right": "issue:46877", + "accept": false, + "reason": "CUDA AMP loss crash vs auto_docstring import-time logging; unrelated." + }, + { + "left": "issue:45593", + "right": "issue:46877", + "accept": false, + "reason": "Auxiliary-loss behavior in D-FINE vs docstring diagnostics printing; unrelated." + }, + { + "left": "issue:42673", + "right": "issue:43704", + "accept": true, + "reason": "Same title and same concrete symptom: Qwen3ForCausalLM VRAM leak when used in multiple dataloader threads; later report is a duplicate." + }, + { + "left": "issue:44038", + "right": "issue:45310", + "accept": false, + "reason": "Both mention Qwen3/VL-Moe loading, but the model/version context and concrete failure differ." + }, + { + "left": "issue:44222", + "right": "issue:45310", + "accept": false, + "reason": "FP8 save_pretrained failure vs Qwen3.5 from_pretrained load error; different issues." + }, + { + "left": "issue:36010", + "right": "issue:41628", + "accept": false, + "reason": "Import errors for different missing names (GenerationMixin vs AutoImageProcessor)." + }, + { + "left": "issue:46202", + "right": "issue:46701", + "accept": false, + "reason": "Both are Gemma-related, but one is token roundtrip on \u2581 and the other is image-text pipeline failure; different bugs." + }, + { + "left": "issue:42994", + "right": "issue:43824", + "accept": false, + "reason": "Quantized save_pretrained failure vs missing Qwen2_5_VLForConditionalGeneration import." + }, + { + "left": "issue:43278", + "right": "issue:45137", + "accept": false, + "reason": "Embedding dtype drift vs ZeRO3 deque IndexError; no overlap." + }, + { + "left": "issue:45081", + "right": "issue:45103", + "accept": false, + "reason": "Tokenizer regex crash vs docstring annotation AttributeError; unrelated." + }, + { + "left": "issue:42831", + "right": "issue:44222", + "accept": false, + "reason": "Both touch FP8, but one is an accuracy regression and the other is a MoE save_pretrained bug." + }, + { + "left": "issue:43232", + "right": "issue:44792", + "accept": false, + "reason": "Generation kwargs bug after sync_gpus vs janus image generation test failure; different code paths." + }, + { + "left": "issue:42175", + "right": "issue:44933", + "accept": false, + "reason": "Backend extras packaging issue vs missing image_utils import; unrelated." + }, + { + "left": "issue:43504", + "right": "issue:45561", + "accept": false, + "reason": "Pretrained model loading legacy-field bug vs xdist race in testing utils." + }, + { + "left": "issue:45561", + "right": "issue:45600", + "accept": false, + "reason": "Captured_info.txt race vs auto_mappings stale config references; different subsystems." + }, + { + "left": "issue:43976", + "right": "issue:44933", + "accept": false, + "reason": "Python version support regression vs missing image_utils import; unrelated." + }, + { + "left": "issue:44496", + "right": "issue:45561", + "accept": false, + "reason": "Unrecognized model/config loading vs xdist race in patched testing utils." + }, + { + "left": "issue:30333", + "right": "issue:33290", + "accept": false, + "reason": "MLflow reporting failure vs DeepSpeed/Adafactor OOM; different problems." + }, + { + "left": "issue:45405", + "right": "issue:46877", + "accept": false, + "reason": "MIN_PEFT_VERSION packaging issue vs docstring stdout/stderr logging." + }, + { + "left": "issue:30990", + "right": "issue:35141", + "accept": false, + "reason": "Loading hang in sentence-transformers vs post_init embedding reinitialization; different behaviors." + }, + { + "left": "issue:42898", + "right": "issue:44568", + "accept": false, + "reason": "Tokenizer cleanup behavior change vs add_special_tokens/BOS-EOS regression; different tokenizer bugs." + }, + { + "left": "issue:42175", + "right": "issue:45712", + "accept": false, + "reason": "TensorFlow backend packaging issue vs leftover dummy classes/check_repo failure." + }, + { + "left": "issue:43479", + "right": "issue:46829", + "accept": false, + "reason": "Default multimodal config initialization bug vs hidden_states output bug in video classification." + }, + { + "left": "issue:46077", + "right": "issue:46123", + "accept": false, + "reason": "Attention mask construction vs pipeline batching state flag; unrelated." + }, + { + "left": "issue:44623", + "right": "issue:45310", + "accept": false, + "reason": "Processor save_pretrained missing files vs Qwen3.5 from_pretrained error; unrelated." + }, + { + "left": "issue:42673", + "right": "issue:46693", + "accept": false, + "reason": "VRAM leak in dataloader threads vs FA2 long-context performance regression; different symptoms and code paths." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted duplicate pair, issue 42673/43704, is well supported by identical titles and matching VRAM-leak symptom, with 42673 having more discussion/context. The remaining soft-edge rejections avoid overclaiming and are phrased cautiously." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 47, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:28282", + "issue:30333", + "issue:34689", + "issue:35141", + "issue:39692", + "issue:42222", + "issue:42796", + "issue:42831", + "issue:42898", + "issue:43116", + "issue:43295", + "issue:43479", + "issue:43723", + "issue:43784", + "issue:43824", + "issue:43874", + "issue:44016", + "issue:44060", + "issue:44075", + "issue:44079", + "issue:44190", + "issue:44222", + "issue:44295", + "issue:44297", + "issue:44360", + "issue:44568", + "issue:44623", + "issue:44625", + "issue:44734", + "issue:44779", + "issue:44869", + "issue:44933", + "issue:45092", + "issue:45103", + "issue:45125", + "issue:45200", + "issue:45356", + "issue:45405", + "issue:45656", + "issue:45712", + "issue:45812", + "issue:46231", + "issue:46762", + "issue:46877", + "issue:46909" + ], + "soft_pairs": [ + "issue:43479|issue:44869", + "issue:43723|issue:44568", + "issue:42898|issue:45812", + "issue:44016|issue:46231", + "issue:44625|issue:46877", + "issue:42898|issue:44779", + "issue:39692|issue:46909", + "issue:43479|issue:44079", + "issue:43874|issue:45092", + "issue:43784|issue:43824", + "issue:27957|issue:30333", + "issue:44734|issue:45103", + "issue:42222|issue:45200", + "issue:30333|issue:34689", + "issue:42898|issue:43295", + "issue:42898|issue:45356", + "issue:44295|issue:45656", + "issue:43116|issue:44190", + "issue:30333|issue:35141", + "issue:45405|issue:45712", + "issue:44222|issue:44297", + "issue:22355|issue:28282", + "issue:44060|issue:45125", + "issue:44075|issue:46231", + "issue:44933|issue:45712", + "issue:44360|issue:46762", + "issue:42831|issue:44623", + "issue:44297|issue:45656", + "issue:42796|issue:43723", + "issue:44222|issue:44623" + ], + "prepared_review_unit_hash": "b02bf6580a1bb1e7f010f72f973b253b5700f20fec028230956dc62303bfb43b", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36261, + "estimated_input_tokens": 9066, + "estimated_eval_tokens": 18388 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous; the soft matches are topic-adjacent but not true duplicates. The closest anchor is a tokenizer regression report, but I would not merge any of the soft pairs.", + "confidence": 0.58, + "canonical_issue_reason": "Issue 42898 is the most broadly representative regression report in the set: it concerns a versioned tokenizer behavior change and is the clearest anchor among otherwise unrelated issues.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 42898 has the highest general triage value because it describes a concrete user-facing v5 regression with likely broad impact, even though the cluster does not form a true duplicate set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43479", + "right": "issue:44869", + "accept": false, + "reason": "Different bugs: multimodal config default initialization vs Whisper word-timestamp decode crash." + }, + { + "left": "issue:43723", + "right": "issue:44568", + "accept": false, + "reason": "Both involve tokenizers, but one is a tokenizer loading/import issue and the other is a specific add_special_tokens regression on mdeberta-v3." + }, + { + "left": "issue:42898", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions in v5, but they affect different code paths: cleanup behavior vs Granite token IDs." + }, + { + "left": "issue:44016", + "right": "issue:46231", + "accept": false, + "reason": "Notebook syntax error and doc parameter mismatch are unrelated documentation issues." + }, + { + "left": "issue:44625", + "right": "issue:46877", + "accept": false, + "reason": "Qwen config propagation bug and auto_docstring stdout logging are unrelated." + }, + { + "left": "issue:42898", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but the affected models and failure mechanisms differ." + }, + { + "left": "issue:39692", + "right": "issue:46909", + "accept": false, + "reason": "Both are broken documentation examples, but they are different examples with different errors." + }, + { + "left": "issue:43479", + "right": "issue:44079", + "accept": false, + "reason": "Multimodal config initialization and ModelOutput key assignment are different code paths." + }, + { + "left": "issue:43874", + "right": "issue:45092", + "accept": false, + "reason": "Vision processor API missing method vs meta-initialization incompatibility for old remote-code checkpoints." + }, + { + "left": "issue:43784", + "right": "issue:43824", + "accept": false, + "reason": "Distinct import failures with different missing symbols and causes." + }, + { + "left": "issue:27957", + "right": "issue:30333", + "accept": false, + "reason": "Flash Attention model support and MLflow reporting on failed jobs are unrelated." + }, + { + "left": "issue:44734", + "right": "issue:45103", + "accept": false, + "reason": "Serving KV-cache indexing crash and auto_docstring annotations crash are unrelated." + }, + { + "left": "issue:42222", + "right": "issue:45200", + "accept": false, + "reason": "Both are multimodal-model issues, but they concern different models and different failures." + }, + { + "left": "issue:30333", + "right": "issue:34689", + "accept": false, + "reason": "MLflow failed-job reporting and model loading regression are unrelated." + }, + { + "left": "issue:42898", + "right": "issue:43295", + "accept": false, + "reason": "Both are transformer regressions, but one is tokenizer cleanup behavior and the other is processor/tokenizer image-handling compatibility." + }, + { + "left": "issue:42898", + "right": "issue:45356", + "accept": false, + "reason": "Different tokenizer regressions with different symptoms and code paths." + }, + { + "left": "issue:44295", + "right": "issue:45656", + "accept": false, + "reason": "Position_ids buffer access bug and deepspeed optimizer double-step are unrelated." + }, + { + "left": "issue:43116", + "right": "issue:44190", + "accept": false, + "reason": "Different example-script failures: empty multi-label outputs vs local dataset loading." + }, + { + "left": "issue:30333", + "right": "issue:35141", + "accept": false, + "reason": "MLflow reporting on training failure and embedding reinitialization on resize are unrelated." + }, + { + "left": "issue:45405", + "right": "issue:45712", + "accept": false, + "reason": "A dependency/version bump issue and leftover dummy-class cleanup are unrelated." + }, + { + "left": "issue:44222", + "right": "issue:44297", + "accept": false, + "reason": "Different save_pretrained bugs: FP8 MoE export vs tokenizer_class mismatch." + }, + { + "left": "issue:22355", + "right": "issue:28282", + "accept": false, + "reason": "Both are import/dependency errors, but they stem from different missing components and are not the same bug." + }, + { + "left": "issue:44060", + "right": "issue:45125", + "accept": false, + "reason": "Both are Qwen3-family issues, but one is a tied-weights warning and the other is missing tensor-parallel planning metadata." + }, + { + "left": "issue:44075", + "right": "issue:46231", + "accept": false, + "reason": "Optimizer argument usage bug and documentation parameter-name mismatch are unrelated." + }, + { + "left": "issue:44933", + "right": "issue:45712", + "accept": false, + "reason": "Import exposure in image_utils and leftover dummy classes failing checks are different hygiene issues." + }, + { + "left": "issue:44360", + "right": "issue:46762", + "accept": false, + "reason": "DSA indexer ReLU discussion and Minimax numerical differences are unrelated math/model issues." + }, + { + "left": "issue:42831", + "right": "issue:44623", + "accept": false, + "reason": "FP8 accuracy issue and processor.save_pretrained missing files are unrelated." + }, + { + "left": "issue:44297", + "right": "issue:45656", + "accept": false, + "reason": "Tokenizer save metadata mismatch and optimizer double-step are unrelated." + }, + { + "left": "issue:42796", + "right": "issue:43723", + "accept": false, + "reason": "Both concern tokenizers, but Ministral BPE-marker handling and tokenizer loading/import failure are different bugs." + }, + { + "left": "issue:44222", + "right": "issue:44623", + "accept": false, + "reason": "Different save_pretrained problems affecting different objects and failure modes." + } + ] + }, + "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": 23, + "review_unit_index": 48, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:30990", + "issue:38175", + "issue:41628", + "issue:42757", + "issue:42831", + "issue:42907", + "issue:43122", + "issue:43232", + "issue:43475", + "issue:43479", + "issue:43644", + "issue:43723", + "issue:44190", + "issue:44222", + "issue:44261", + "issue:44263", + "issue:44279", + "issue:44295", + "issue:44297", + "issue:44360", + "issue:44373", + "issue:44448", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44623", + "issue:44977", + "issue:45071", + "issue:45092", + "issue:45245", + "issue:45310", + "issue:45357", + "issue:45381", + "issue:45406", + "issue:45834", + "issue:46050", + "issue:46082", + "issue:46132", + "issue:46209", + "issue:46305", + "issue:46328", + "issue:46369", + "issue:46424", + "issue:46431", + "issue:46655", + "issue:46710", + "issue:46909" + ], + "soft_pairs": [ + "issue:44448|issue:46132", + "issue:45245|issue:46132", + "issue:43479|issue:43644", + "issue:41628|issue:42757", + "issue:44297|issue:45310", + "issue:44373|issue:44485", + "issue:46132|issue:46710", + "issue:45071|issue:46082", + "issue:44222|issue:44484", + "issue:43723|issue:44977", + "issue:43122|issue:46209", + "issue:44263|issue:45092", + "issue:44190|issue:46082", + "issue:43232|issue:46424", + "issue:43232|issue:45834", + "issue:44295|issue:44623", + "issue:43475|issue:46305", + "issue:46209|issue:46431", + "issue:46305|issue:46655", + "issue:44263|issue:44360", + "issue:44263|issue:44373", + "issue:30990|issue:38175", + "issue:42831|issue:45310", + "issue:45092|issue:45357", + "issue:46369|issue:46909", + "issue:42907|issue:46050", + "issue:44279|issue:46328", + "issue:45381|issue:46305", + "issue:44261|issue:45092", + "issue:44483|issue:45406" + ], + "prepared_review_unit_hash": "167b51dcb1d9227b801fb6a7f1dc8dc62d2b15834e879a4fead0216822bb7ea0", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 35675, + "estimated_input_tokens": 8919, + "estimated_eval_tokens": 18094 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose set of Transformer-related issues with some superficial textual similarity, but none of the suggested pairs look like true duplicates. No PRs are present.", + "confidence": 0.66, + "canonical_issue_reason": "issue:46132 is the most central and broad framework-level regression candidate in the set, appearing in multiple soft-similarity links and covering a shared API behavior change.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:46132 is the best representative issue because it is the broadest, most reusable bug report among the candidates; the others are narrower model- or feature-specific failures.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44448", + "right": "issue:46132", + "accept": false, + "reason": "Different bugs: model output drift vs AttentionInterface.register behavior." + }, + { + "left": "issue:45245", + "right": "issue:46132", + "accept": false, + "reason": "Unrelated: category cardinality limit vs attention registration semantics." + }, + { + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Different model/config init bug vs non-persistent buffer corruption." + }, + { + "left": "issue:41628", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but for different missing symbols/dependencies." + }, + { + "left": "issue:44297", + "right": "issue:45310", + "accept": false, + "reason": "Tokenizer save metadata mismatch vs Qwen3.5 from_pretrained failure." + }, + { + "left": "issue:44373", + "right": "issue:44485", + "accept": false, + "reason": "Docstring issue vs RoPE implementation bug." + }, + { + "left": "issue:46132", + "right": "issue:46710", + "accept": false, + "reason": "API behavior regression vs model accuracy regression; not the same code path." + }, + { + "left": "issue:45071", + "right": "issue:46082", + "accept": false, + "reason": "Config type-checking regression vs head_dim validation bug." + }, + { + "left": "issue:44222", + "right": "issue:44484", + "accept": false, + "reason": "FP8 save_pretrained bug vs unrelated shard-size default question." + }, + { + "left": "issue:43723", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer loading issue vs flash-attention generation bug." + }, + { + "left": "issue:43122", + "right": "issue:46209", + "accept": false, + "reason": "Tokenization change across versions vs FP8 B200 inference failure." + }, + { + "left": "issue:44263", + "right": "issue:45092", + "accept": false, + "reason": "Indexer tensor-split bug vs remote-code/meta-init incompatibility." + }, + { + "left": "issue:44190", + "right": "issue:46082", + "accept": false, + "reason": "Local dataset loading script issue vs LlamaConfig validation." + }, + { + "left": "issue:43232", + "right": "issue:46424", + "accept": false, + "reason": "Generation kwargs sync bug vs cache-length config bug." + }, + { + "left": "issue:43232", + "right": "issue:45834", + "accept": false, + "reason": "Generation state update bug vs Kosmos OCR index error." + }, + { + "left": "issue:44295", + "right": "issue:44623", + "accept": false, + "reason": "Buffer access error vs processor.save_pretrained file omission." + }, + { + "left": "issue:43475", + "right": "issue:46305", + "accept": false, + "reason": "SAM3 missing attribute vs Moonshine off-by-one masking." + }, + { + "left": "issue:46209", + "right": "issue:46431", + "accept": false, + "reason": "FP8 inference on B200 vs missing chat template error." + }, + { + "left": "issue:46305", + "right": "issue:46655", + "accept": false, + "reason": "Different video/window bugs with unrelated calculations." + }, + { + "left": "issue:44263", + "right": "issue:44360", + "accept": false, + "reason": "torch.split return bug vs missing ReLU in another indexer." + }, + { + "left": "issue:44263", + "right": "issue:44373", + "accept": false, + "reason": "Runtime indexing bug vs docstring typo." + }, + { + "left": "issue:30990", + "right": "issue:38175", + "accept": false, + "reason": "Unrelated symptoms: loading hang vs zero-probability outputs." + }, + { + "left": "issue:42831", + "right": "issue:45310", + "accept": false, + "reason": "FP8 accuracy regression vs Qwen3.5 from_pretrained failure." + }, + { + "left": "issue:45092", + "right": "issue:45357", + "accept": false, + "reason": "Checkpoint meta-init incompatibility vs incorrect save_pretrained keys." + }, + { + "left": "issue:46369", + "right": "issue:46909", + "accept": false, + "reason": "Memory overflow in a model vs broken documentation example." + }, + { + "left": "issue:42907", + "right": "issue:46050", + "accept": false, + "reason": "Different serialization issues: dequantized save failure vs quantized round-trip breakage." + }, + { + "left": "issue:44279", + "right": "issue:46328", + "accept": false, + "reason": "Generic dependency issue vs pad performance regression." + }, + { + "left": "issue:45381", + "right": "issue:46305", + "accept": false, + "reason": "Video position_ids bug vs streaming mask off-by-one." + }, + { + "left": "issue:44261", + "right": "issue:45092", + "accept": false, + "reason": "Layernorm epsilon bug vs checkpoint/meta-init incompatibility." + }, + { + "left": "issue:44483", + "right": "issue:45406", + "accept": false, + "reason": "Chat-completions request handling vs Gemma4Processor missing tokenizer." + } + ] + }, + "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": 23, + "review_unit_index": 49, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:28218", + "issue:30990", + "issue:33357", + "issue:39401", + "issue:41628", + "issue:42371", + "issue:42757", + "issue:42796", + "issue:42831", + "issue:42886", + "issue:42907", + "issue:42994", + "issue:43066", + "issue:43122", + "issue:43232", + "issue:43408", + "issue:43540", + "issue:43576", + "issue:43644", + "issue:43646", + "issue:43784", + "issue:43824", + "issue:43937", + "issue:43976", + "issue:44038", + "issue:44188", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44373", + "issue:44488", + "issue:44568", + "issue:44829", + "issue:44945", + "issue:44964", + "issue:45278", + "issue:45381", + "issue:45593", + "issue:45753", + "issue:46002", + "issue:46172", + "issue:46305", + "issue:46369", + "issue:46402", + "issue:46762", + "issue:46909" + ], + "soft_pairs": [ + "issue:22355|issue:27957", + "issue:43644|issue:46305", + "issue:44829|issue:46002", + "issue:42831|issue:44038", + "issue:41628|issue:43824", + "issue:43408|issue:46402", + "issue:42831|issue:45753", + "issue:22355|issue:28218", + "issue:39401|issue:44568", + "issue:42886|issue:43066", + "issue:42886|issue:42994", + "issue:42757|issue:45278", + "issue:27957|issue:30990", + "issue:44188|issue:44373", + "issue:43408|issue:46369", + "issue:43232|issue:45593", + "issue:42796|issue:43122", + "issue:43646|issue:46002", + "issue:45381|issue:46909", + "issue:43540|issue:46909", + "issue:30990|issue:33357", + "issue:42907|issue:44945", + "issue:44295|issue:44297", + "issue:43576|issue:43784", + "issue:42371|issue:44263", + "issue:39401|issue:44488", + "issue:42907|issue:44964", + "issue:43976|issue:44945", + "issue:46172|issue:46762", + "issue:42907|issue:43937" + ], + "prepared_review_unit_hash": "fe26b91339a52622e2db01298c59f19706406c48024580ef4c683520048083c2", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36050, + "estimated_input_tokens": 9013, + "estimated_eval_tokens": 18282 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is very heterogeneous: most soft pairs are clearly different bugs, with only a few broad import/tokenizer regressions showing superficial similarity. No pair is strong enough to merge as the same duplicate issue under a strict triage standard.", + "confidence": 0.93, + "canonical_issue_reason": "issue:45278 is the broadest umbrella-style report in the set (multiple import errors after upgrading to 5.5.0), so it is the best candidate for a canonical issue among these mostly unrelated reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the most general and centrally framed issue here, making it the least-specific representative of the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:22355", + "right": "issue:27957", + "accept": false, + "reason": "Different failures: missing transformers.onnx module vs XLMRoberta/Flash Attention 2 behavior." + }, + { + "left": "issue:43644", + "right": "issue:46305", + "accept": false, + "reason": "Unrelated bugs: non-persistent buffer corruption vs MoonshineStreaming window off-by-one." + }, + { + "left": "issue:44829", + "right": "issue:46002", + "accept": false, + "reason": "Different code paths: flash_attention_3 training collapse vs NaNs in classification head init/checkout." + }, + { + "left": "issue:42831", + "right": "issue:44038", + "accept": false, + "reason": "Different problems: FineGrainedFP8 accuracy regression vs Qwen3-VL-Moe model bug." + }, + { + "left": "issue:41628", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but for different missing symbols and likely different fixes." + }, + { + "left": "issue:43408", + "right": "issue:46402", + "accept": false, + "reason": "Different SAM3 issues: wrong model-type warning vs text encoder weights not loading." + }, + { + "left": "issue:42831", + "right": "issue:45753", + "accept": false, + "reason": "Different model families and symptoms: FP8 accuracy vs Qwen3_5 image loop." + }, + { + "left": "issue:22355", + "right": "issue:28218", + "accept": false, + "reason": "Missing module import vs tokenizer spacing after added token are unrelated." + }, + { + "left": "issue:39401", + "right": "issue:44568", + "accept": false, + "reason": "Different tokenizer bugs: wrong offset mapping vs missing BOS/EOS on add_special_tokens." + }, + { + "left": "issue:42886", + "right": "issue:43066", + "accept": false, + "reason": "Offline cache loading failure is not the same as wrong decoder type selection in v5." + }, + { + "left": "issue:42886", + "right": "issue:42994", + "accept": false, + "reason": "Different areas: offline tokenizer cache loading vs quantized model saving failure." + }, + { + "left": "issue:42757", + "right": "issue:45278", + "accept": false, + "reason": "42257 is one specific missing import; 45278 is a broad import-regression report, not the same concrete bug." + }, + { + "left": "issue:27957", + "right": "issue:30990", + "accept": false, + "reason": "Unrelated: Flash Attention 2 issue vs Sentence-Transformers loading hang." + }, + { + "left": "issue:44188", + "right": "issue:44373", + "accept": false, + "reason": "Attention kernel divergence is unrelated to a docstring typo." + }, + { + "left": "issue:43408", + "right": "issue:46369", + "accept": false, + "reason": "Both involve SAM3, but one is a type warning and the other is a memory overflow on a different path." + }, + { + "left": "issue:43232", + "right": "issue:45593", + "accept": false, + "reason": "Generation kwargs handling is unrelated to D-FINE auxiliary-loss behavior." + }, + { + "left": "issue:42796", + "right": "issue:43122", + "accept": false, + "reason": "Similar theme, but different bugs: BPE marker handling vs version-to-version tokenization changes." + }, + { + "left": "issue:43646", + "right": "issue:46002", + "accept": false, + "reason": "Custom model initialization breakage is not the same as NaNs in classification heads." + }, + { + "left": "issue:45381", + "right": "issue:46909", + "accept": false, + "reason": "Different model-specific issues: Qwen2.5-VL video position IDs vs a DIA documentation example." + }, + { + "left": "issue:43540", + "right": "issue:46909", + "accept": false, + "reason": "Video-input ValueError in Qwen3OmniMoe is unrelated to a broken docs example." + }, + { + "left": "issue:30990", + "right": "issue:33357", + "accept": false, + "reason": "Loading hang and MacOS bus error are distinct failures." + }, + { + "left": "issue:42907", + "right": "issue:44945", + "accept": false, + "reason": "Quantized saving failure is unrelated to incorrect pipeline-parallel LLM output." + }, + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "Buffer registration/read error is unrelated to tokenizer_config class mismatch." + }, + { + "left": "issue:43576", + "right": "issue:43784", + "accept": false, + "reason": "Broken env CLI and sentence-transformers import NameError are different regressions." + }, + { + "left": "issue:42371", + "right": "issue:44263", + "accept": false, + "reason": "TF32 settings guidance is unrelated to torch.split return behavior." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "Wrong offset_mapping and failure to load a specific model are different tokenizer/model-loading bugs." + }, + { + "left": "issue:42907", + "right": "issue:44964", + "accept": false, + "reason": "Saving dequantized Ministrals/Devstrals is unrelated to loading Phi-4 multimodal instruct." + }, + { + "left": "issue:43976", + "right": "issue:44945", + "accept": false, + "reason": "Python version incompatibility is unrelated to pipeline-parallel output corruption." + }, + { + "left": "issue:46172", + "right": "issue:46762", + "accept": false, + "reason": "Different numerical issues in different models/code paths." + }, + { + "left": "issue:42907", + "right": "issue:43937", + "accept": false, + "reason": "Quantized saving failure is unrelated to GLM-5 GenerationConfig validation." + } + ] + }, + "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": 23, + "review_unit_index": 50, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:22355", + "issue:29127", + "issue:29159", + "issue:36296", + "issue:39401", + "issue:41628", + "issue:42175", + "issue:42757", + "issue:42907", + "issue:43116", + "issue:43122", + "issue:43278", + "issue:43408", + "issue:43452", + "issue:43475", + "issue:43493", + "issue:43756", + "issue:43784", + "issue:43824", + "issue:44038", + "issue:44060", + "issue:44117", + "issue:44188", + "issue:44222", + "issue:44295", + "issue:44297", + "issue:44351", + "issue:44360", + "issue:44485", + "issue:44521", + "issue:44625", + "issue:44704", + "issue:44779", + "issue:44829", + "issue:44938", + "issue:44945", + "issue:45200", + "issue:45216", + "issue:45278", + "issue:45290", + "issue:45478", + "issue:45600", + "issue:45753", + "issue:46231", + "issue:46326", + "issue:46328", + "issue:46614", + "issue:46877" + ], + "soft_pairs": [ + "issue:43756|issue:44060", + "issue:41628|issue:45278", + "issue:44117|issue:44704", + "issue:43784|issue:45600", + "issue:39401|issue:44521", + "issue:44704|issue:45290", + "issue:44625|issue:44829", + "issue:44360|issue:44485", + "issue:43408|issue:43475", + "issue:43122|issue:44521", + "issue:44038|issue:44297", + "issue:43122|issue:46326", + "issue:42907|issue:43452", + "issue:44038|issue:44295", + "issue:45216|issue:45478", + "issue:22355|issue:29159", + "issue:39401|issue:46614", + "issue:44038|issue:44222", + "issue:43824|issue:44351", + "issue:43116|issue:45200", + "issue:43122|issue:44779", + "issue:42757|issue:44351", + "issue:36296|issue:43116", + "issue:43278|issue:44945", + "issue:42175|issue:44938", + "issue:22355|issue:29127", + "issue:43493|issue:45753", + "issue:46231|issue:46877", + "issue:44188|issue:46328" + ], + "prepared_review_unit_hash": "76f41f6ad899fc48ee9098e477a1886a1a3cbe9c876c147d176c50da82f31162", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36242, + "estimated_input_tokens": 9061, + "estimated_eval_tokens": 18378 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most pairs are related only by broad subsystem or symptom similarity and should stay separate. The strongest duplicate-like pair is the Qwen3.5 save/load regression (`save_pretrained` producing a bad checkpoint and `from_pretrained` failing on it).", + "confidence": 0.71, + "canonical_issue_reason": "issue:45216 is the clearest root-cause description for the Qwen3.5 serialization regression, and it plausibly covers the downstream `from_pretrained` failure in issue:45478.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45216 is the best representative issue because it states the regression and version boundary more concretely than the related load-time symptom issue.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43756", + "right": "issue:44060", + "accept": false, + "reason": "Both are model-specific architecture bugs, but they concern different models and different failure modes (RoPE layer behavior vs tied-weights warning)." + }, + { + "left": "issue:41628", + "right": "issue:45278", + "accept": false, + "reason": "Both are import-related, but one is a single missing export and the other is a broad upgrade-wide import breakage; not the same bug." + }, + { + "left": "issue:44117", + "right": "issue:44704", + "accept": false, + "reason": "Tokenizer mapping fallback and `AutoProcessor` kwarg forwarding are unrelated code paths with different observed failures." + }, + { + "left": "issue:43784", + "right": "issue:45600", + "accept": false, + "reason": "Both mention import-time breakage, but they are caused by different missing symbols/config references in different integrations." + }, + { + "left": "issue:39401", + "right": "issue:44521", + "accept": false, + "reason": "Both involve tokenization outputs, but one is offset mapping for Qwen3 and the other is assistant mask handling for multimodal chat templates." + }, + { + "left": "issue:44704", + "right": "issue:45290", + "accept": false, + "reason": "Different APIs and failure modes: cached-file kwargs propagation versus chat template crashes on tool-call messages." + }, + { + "left": "issue:44625", + "right": "issue:44829", + "accept": false, + "reason": "One is config propagation for Qwen3.5 labels; the other is a training-quality regression tied to flash attention 3." + }, + { + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "Unrelated model/math implementation issues: DSA indexer activation vs GLM-5 RoPE behavior." + }, + { + "left": "issue:43408", + "right": "issue:43475", + "accept": false, + "reason": "Both are SAM3-related, but one is a model-type mismatch warning and the other is a missing attribute runtime error." + }, + { + "left": "issue:43122", + "right": "issue:44521", + "accept": false, + "reason": "Both are tokenizer/template regressions, but they affect different features and outputs." + }, + { + "left": "issue:44038", + "right": "issue:44297", + "accept": false, + "reason": "One is a Qwen3-VL-Moe compatibility issue; the other is tokenizer save metadata mismatch. Different concrete bugs." + }, + { + "left": "issue:43122", + "right": "issue:46326", + "accept": false, + "reason": "Both are tokenizer-related regressions, but they concern different APIs and symptoms." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Different save/load failures in different model paths; not the same underlying bug." + }, + { + "left": "issue:44038", + "right": "issue:44295", + "accept": false, + "reason": "General Qwen3-VL-Moe issue vs a position_ids buffer read error; no concrete overlap." + }, + { + "left": "issue:45216", + "right": "issue:45478", + "accept": true, + "reason": "Same Qwen3.5 MoE save/load regression around the 5.4.0 change: a bad checkpoint saved by `save_pretrained` can plausibly surface as `from_pretrained` failure." + }, + { + "left": "issue:22355", + "right": "issue:29159", + "accept": false, + "reason": "A missing `transformers.onnx` module import is unrelated to slow/fast tokenizer inconsistency." + }, + { + "left": "issue:39401", + "right": "issue:46614", + "accept": false, + "reason": "Different tokenizer subsystems and different bugs: offset mapping vs stale phonemizer language state." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "Both mention MoE, but one is a model/version bug and the other is FP8 `save_pretrained` behavior; too broad to merge." + }, + { + "left": "issue:43824", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but they are for different exported classes and different dependency/version contexts." + }, + { + "left": "issue:43116", + "right": "issue:45200", + "accept": false, + "reason": "Different application paths: example classification script output vs Gemma 4 multimodal training defaults." + }, + { + "left": "issue:43122", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but they are model-specific and describe different incorrect behaviors." + }, + { + "left": "issue:42757", + "right": "issue:44351", + "accept": false, + "reason": "Different missing imports from different packages; same symptom class only." + }, + { + "left": "issue:36296", + "right": "issue:43116", + "accept": false, + "reason": "Tensor-parallel training and classification example behavior are unrelated." + }, + { + "left": "issue:43278", + "right": "issue:44945", + "accept": false, + "reason": "Different distributed/eval bugs with no shared concrete code path." + }, + { + "left": "issue:42175", + "right": "issue:44938", + "accept": false, + "reason": "Backend-install dependency issue versus Python 3.14 compatibility failure; not the same problem." + }, + { + "left": "issue:22355", + "right": "issue:29127", + "accept": false, + "reason": "Different issue types and different subsystems: missing ONNX module vs LayoutLMv3 error messaging." + }, + { + "left": "issue:43493", + "right": "issue:45753", + "accept": false, + "reason": "Different model families and different failures: SigLIP2 implementation discrepancy vs Qwen3_5 infinite loop on a specific image." + }, + { + "left": "issue:46231", + "right": "issue:46877", + "accept": false, + "reason": "Documentation parameter-name mismatch is unrelated to import-time docstring lint output." + }, + { + "left": "issue:44188", + "right": "issue:46328", + "accept": false, + "reason": "Attention-kernel divergence under `torch.compile` is unrelated to feature-extractor padding inefficiency." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The only accepted soft pair, issue:45216 vs issue:45478, is plausibly the same Qwen3.5 save/load regression, and the other verdicts are appropriately cautious by distinguishing different model families, APIs, 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": 23, + "review_unit_index": 51, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:28348", + "issue:33453", + "issue:38617", + "issue:39401", + "issue:42222", + "issue:42371", + "issue:42673", + "issue:42898", + "issue:43066", + "issue:43116", + "issue:43441", + "issue:43493", + "issue:43611", + "issue:43618", + "issue:43825", + "issue:43827", + "issue:43874", + "issue:43937", + "issue:43976", + "issue:43992", + "issue:44038", + "issue:44075", + "issue:44263", + "issue:44355", + "issue:44485", + "issue:44610", + "issue:44625", + "issue:44704", + "issue:44779", + "issue:44936", + "issue:44995", + "issue:45072", + "issue:45083", + "issue:45092", + "issue:45278", + "issue:45357", + "issue:45446", + "issue:45478", + "issue:46002", + "issue:46231", + "issue:46614", + "issue:46640", + "issue:46697", + "issue:46710", + "issue:46752", + "issue:46762", + "issue:46829", + "issue:46909" + ], + "soft_pairs": [ + "issue:43937|issue:44485", + "issue:38617|issue:45278", + "issue:42898|issue:46614", + "issue:43066|issue:44779", + "issue:43611|issue:46002", + "issue:44038|issue:45478", + "issue:43976|issue:45446", + "issue:44355|issue:46231", + "issue:43116|issue:43493", + "issue:42673|issue:45072", + "issue:45357|issue:45478", + "issue:44625|issue:46002", + "issue:42371|issue:43874", + "issue:45083|issue:45446", + "issue:44610|issue:46909", + "issue:43618|issue:46829", + "issue:43992|issue:44704", + "issue:44075|issue:44936", + "issue:43116|issue:43825", + "issue:43066|issue:46710", + "issue:28348|issue:46640", + "issue:46614|issue:46752", + "issue:44263|issue:46762", + "issue:44995|issue:46697", + "issue:45092|issue:46762", + "issue:42222|issue:43441", + "issue:43827|issue:45446", + "issue:33453|issue:39401" + ], + "prepared_review_unit_hash": "88c7488e776d16dcda7c5725398641a0508a392162150b45373a894f9f8acd10", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 35638, + "estimated_input_tokens": 8910, + "estimated_eval_tokens": 18076 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose theme match, not a true duplicate set. The soft pairs generally share broad subsystem language (tokenizers, Qwen, v5 regressions, flash attention) but describe different code paths or different symptoms, so I rejected them all.", + "confidence": 0.89, + "canonical_issue_reason": "issue:45278 is the broadest umbrella-style report in the set: it describes multiple import failures after a version upgrade, making it the most plausible canonical target if any of these are related.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the most suitable global issue to anchor review because it is the most general and could subsume specific import-regression reports; the rest are mostly narrower or unrelated symptoms.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43937", + "right": "issue:44485", + "accept": false, + "reason": "Both are GLM-5 related, but one is a GenerationConfig validation failure and the other is a RoPE implementation discussion; different bugs/code paths." + }, + { + "left": "issue:38617", + "right": "issue:45278", + "accept": false, + "reason": "Both mention import problems, but one is a specific missing symbol import and the other is a broad multi-import regression report; not the same concrete failure." + }, + { + "left": "issue:42898", + "right": "issue:46614", + "accept": false, + "reason": "Tokenizer cleanup behavior and a stale phonemizer backend are unrelated regressions." + }, + { + "left": "issue:43066", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but one is about decoder type in v5 and the other about DeepSeek output correctness; too different to merge." + }, + { + "left": "issue:43611", + "right": "issue:46002", + "accept": false, + "reason": "Base model prefix loading breakage and NaNs in classification heads are different failures with different symptoms and code paths." + }, + { + "left": "issue:44038", + "right": "issue:45478", + "accept": false, + "reason": "Both involve Qwen/MoE and Transformers v5, but they concern different model variants and different loading errors." + }, + { + "left": "issue:43976", + "right": "issue:45446", + "accept": false, + "reason": "A Python version compatibility issue is unrelated to a PyTorch version gate in flex_attention." + }, + { + "left": "issue:44355", + "right": "issue:46231", + "accept": false, + "reason": "Compiled Python file errors and a documentation parameter-name mismatch are not the same bug." + }, + { + "left": "issue:43116", + "right": "issue:43493", + "accept": false, + "reason": "Example-script multi-label classification output failure is unrelated to a SigLIP2 implementation discrepancy." + }, + { + "left": "issue:42673", + "right": "issue:45072", + "accept": false, + "reason": "VRAM leakage in Qwen3ForCausalLM and dtype mismatches in SwitchTransformers/TimmWrapperModel are different runtime issues." + }, + { + "left": "issue:45357", + "right": "issue:45478", + "accept": false, + "reason": "Both are Qwen3.5-related, but one is save_pretrained key serialization and the other is from_pretrained loading failure; not the same bug." + }, + { + "left": "issue:44625", + "right": "issue:46002", + "accept": false, + "reason": "num_labels propagation in Qwen3.5 is unrelated to NaNs in classification heads." + }, + { + "left": "issue:42371", + "right": "issue:43874", + "accept": false, + "reason": "TF32 API guidance and a missing image-patch helper in GLM46VImageProcessorFast are unrelated." + }, + { + "left": "issue:45083", + "right": "issue:45446", + "accept": false, + "reason": "A feature-extraction length helper bug and a flex_attention import/version check are unrelated." + }, + { + "left": "issue:44610", + "right": "issue:46909", + "accept": false, + "reason": "OmDet-Turbo preprocessing dimension mismatch and a broken DIA documentation example are not the same issue." + }, + { + "left": "issue:43618", + "right": "issue:46829", + "accept": false, + "reason": "Both involve output objects, but they affect different models and different misassigned fields." + }, + { + "left": "issue:43992", + "right": "issue:44704", + "accept": false, + "reason": "A missing weight in UMT5Encoder loading is unrelated to AutoProcessor not forwarding kwargs to cached_file." + }, + { + "left": "issue:44075", + "right": "issue:44936", + "accept": false, + "reason": "SGD arguments being ignored and trainer.evaluate() failing after train() are different trainer/optimizer bugs." + }, + { + "left": "issue:43116", + "right": "issue:43825", + "accept": false, + "reason": "An example classification script bug is unrelated to a pipeline() error-message regression." + }, + { + "left": "issue:43066", + "right": "issue:46710", + "accept": false, + "reason": "Tokenizer decoder-type regression and a DeepSeek accuracy regression may both be version-related, but they are not the same underlying defect." + }, + { + "left": "issue:28348", + "right": "issue:46640", + "accept": false, + "reason": "Flash Attention 2 support for GPT2LMHeadModel is a feature request; the other is a separate flash-attention issue." + }, + { + "left": "issue:46614", + "right": "issue:46752", + "accept": false, + "reason": "A stale phonemizer language/backend bug is unrelated to an empty-list apply_chat_template crash." + }, + { + "left": "issue:44263", + "right": "issue:46762", + "accept": false, + "reason": "torch.split return-value handling in one GLM MoE component and numerical differences in Minimax MSA attention are different bugs." + }, + { + "left": "issue:44995", + "right": "issue:46697", + "accept": false, + "reason": "A stale indexer cache crash and over-padding in an image processor are unrelated." + }, + { + "left": "issue:45092", + "right": "issue:46762", + "accept": false, + "reason": "Meta-initialization incompatibility for old remote-code checkpoints is unrelated to Minimax attention numerical drift." + }, + { + "left": "issue:42222", + "right": "issue:43441", + "accept": false, + "reason": "VitPose model breakage and a FlashAttention failure in Ministral-3 are different models and different failure modes." + }, + { + "left": "issue:43827", + "right": "issue:45446", + "accept": false, + "reason": "Documentation still mentioning pipeline() and a PyTorch import/version check in flex_attention are unrelated." + }, + { + "left": "issue:33453", + "right": "issue:39401", + "accept": false, + "reason": "A general tokenizer loading regression and a Qwen3 offset_mapping bug are not the same concrete bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 52, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:38617", + "issue:39401", + "issue:41628", + "issue:42222", + "issue:42617", + "issue:42673", + "issue:42796", + "issue:43064", + "issue:43122", + "issue:43408", + "issue:43450", + "issue:43673", + "issue:43742", + "issue:43937", + "issue:43986", + "issue:44117", + "issue:44291", + "issue:44380", + "issue:44410", + "issue:44466", + "issue:44484", + "issue:44779", + "issue:44829", + "issue:44877", + "issue:44977", + "issue:45071", + "issue:45081", + "issue:45216", + "issue:45310", + "issue:45341", + "issue:45356", + "issue:45362", + "issue:45405", + "issue:45446", + "issue:45561", + "issue:45563", + "issue:45588", + "issue:46002", + "issue:46077", + "issue:46178", + "issue:46231", + "issue:46298", + "issue:46328", + "issue:46512", + "issue:46614", + "issue:46655", + "issue:46752" + ], + "soft_pairs": [ + "issue:43064|issue:43673", + "issue:42673|issue:42796", + "issue:43937|issue:45563", + "issue:44380|issue:46077", + "issue:45356|issue:46614", + "issue:43673|issue:46328", + "issue:42673|issue:46655", + "issue:42222|issue:46752", + "issue:44291|issue:45071", + "issue:45405|issue:45446", + "issue:44117|issue:45588", + "issue:44466|issue:46002", + "issue:39401|issue:44779", + "issue:44410|issue:45310", + "issue:43937|issue:44977", + "issue:43450|issue:43986", + "issue:45341|issue:46328", + "issue:45341|issue:45561", + "issue:44829|issue:45216", + "issue:45071|issue:46512", + "issue:44117|issue:45081", + "issue:43408|issue:43742", + "issue:38617|issue:41628", + "issue:43122|issue:46298", + "issue:42617|issue:45341", + "issue:45362|issue:46328", + "issue:42222|issue:46178", + "issue:44484|issue:45310", + "issue:44877|issue:46231" + ], + "prepared_review_unit_hash": "2fdae60072f1b997bc60f611e52a57431edbf2319fd3169ae08e73822ce047bf", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35866, + "estimated_input_tokens": 8967, + "estimated_eval_tokens": 18190 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous; the soft pairs mostly share only broad subsystem similarity (tokenizers, config/loading, video processors, testing utils) rather than the same underlying bug. I would not merge any of the proposed soft edges.", + "confidence": 0.92, + "canonical_issue_reason": "Issue 43122 is the most general and representative version-regression report in the set (tokenization behavior changed from 4.57.3 to 5.0). That said, the cluster is too mixed to treat as one true duplicate group.", + "canonical_pr_reason": null, + "best_issue_reason": "43122 is the best single issue to anchor this cluster because it is broad, clearly regression-oriented, and closest to the recurring tokenization/version-change theme seen in several nearby issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43064", + "right": "issue:43673", + "accept": false, + "reason": "Training optimizer-state corruption and generation cache missing are unrelated code paths and bugs." + }, + { + "left": "issue:42673", + "right": "issue:42796", + "accept": false, + "reason": "VRAM leak in Qwen3ForCausalLM is unrelated to a tokenizer BPE-marker bug." + }, + { + "left": "issue:43937", + "right": "issue:45563", + "accept": false, + "reason": "Invalid GenerationConfig and a stale paging warning are different failure modes." + }, + { + "left": "issue:44380", + "right": "issue:46077", + "accept": false, + "reason": "Both touch attention behavior, but they are model-specific and not the same concrete bug." + }, + { + "left": "issue:45356", + "right": "issue:46614", + "accept": false, + "reason": "Tokenizer codec-handling regression and stale phonemizer backend state are different subsystems and code paths." + }, + { + "left": "issue:43673", + "right": "issue:46328", + "accept": false, + "reason": "Generation cache handling and SequenceFeatureExtractor.pad performance are unrelated." + }, + { + "left": "issue:42673", + "right": "issue:46655", + "accept": false, + "reason": "VRAM leak in model threading and video padding miscalculation are not the same bug." + }, + { + "left": "issue:42222", + "right": "issue:46752", + "accept": false, + "reason": "Vitpose model breakage and empty-list chat templating crash are unrelated." + }, + { + "left": "issue:44291", + "right": "issue:45071", + "accept": false, + "reason": "Both involve v5 API changes, but unexpected init args and PretrainedConfig type checking are different defects." + }, + { + "left": "issue:45405", + "right": "issue:45446", + "accept": false, + "reason": "A dependency minimum bump and a PyTorch version guard bug are unrelated." + }, + { + "left": "issue:44117", + "right": "issue:45588", + "accept": false, + "reason": "Tokenizer mapping None handling and flash_attention sink-less crash are unrelated." + }, + { + "left": "issue:44466", + "right": "issue:46002", + "accept": false, + "reason": "State dict serialization of tied weights and NaNs in classification heads are different problems." + }, + { + "left": "issue:39401", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different models and appear to be separate bugs." + }, + { + "left": "issue:44410", + "right": "issue:45310", + "accept": false, + "reason": "Missing layer projections and a from_pretrained loading error are not the same concrete issue." + }, + { + "left": "issue:43937", + "right": "issue:44977", + "accept": false, + "reason": "GenerationConfig validation and flash-attention generation failure are different bugs." + }, + { + "left": "issue:43450", + "right": "issue:43986", + "accept": false, + "reason": "Incorrect batched video processor shapes and missing-torchvision crash are separate issues." + }, + { + "left": "issue:45341", + "right": "issue:46328", + "accept": false, + "reason": "A testing_utils bug and a SequenceFeatureExtractor.pad inefficiency are unrelated." + }, + { + "left": "issue:45341", + "right": "issue:45561", + "accept": false, + "reason": "Generic testing_utils bug and xdist race on captured_info.txt are not the same underlying failure." + }, + { + "left": "issue:44829", + "right": "issue:45216", + "accept": false, + "reason": "Flash-attention training degeneration and Qwen3.5 checkpoint save regression are different code paths." + }, + { + "left": "issue:45071", + "right": "issue:46512", + "accept": false, + "reason": "Type-checking breakage and stale config.dtype after from_config are related only loosely, not the same bug." + }, + { + "left": "issue:44117", + "right": "issue:45081", + "accept": false, + "reason": "Missing tokenizer mapping entries and backend_tokenizer AttributeError in regex patching are distinct failures." + }, + { + "left": "issue:43408", + "right": "issue:43742", + "accept": false, + "reason": "Model-type warning and MobileLLM load key error are unrelated." + }, + { + "left": "issue:38617", + "right": "issue:41628", + "accept": false, + "reason": "These are different import errors involving different symbols and modules." + }, + { + "left": "issue:43122", + "right": "issue:46298", + "accept": false, + "reason": "Tokenization regression and noisy logging are unrelated." + }, + { + "left": "issue:42617", + "right": "issue:45341", + "accept": false, + "reason": "A script execution failure and a testing_utils bug are unrelated." + }, + { + "left": "issue:45362", + "right": "issue:46328", + "accept": false, + "reason": "Qwen3.5 chat crash and a SequenceFeatureExtractor.pad inefficiency are unrelated." + }, + { + "left": "issue:42222", + "right": "issue:46178", + "accept": false, + "reason": "Vitpose model breakage and unsupported backend mapping error are unrelated." + }, + { + "left": "issue:44484", + "right": "issue:45310", + "accept": false, + "reason": "save_pretrained shard-size question and Qwen3.5 loading error are unrelated." + }, + { + "left": "issue:44877", + "right": "issue:46231", + "accept": false, + "reason": "Strict config loading and a documentation parameter-name mismatch are not the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 53, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:27957", + "issue:38617", + "issue:39401", + "issue:41628", + "issue:42222", + "issue:42371", + "issue:43012", + "issue:43066", + "issue:43097", + "issue:43122", + "issue:43208", + "issue:43352", + "issue:43441", + "issue:43452", + "issue:43577", + "issue:43673", + "issue:43874", + "issue:43931", + "issue:43937", + "issue:43986", + "issue:44038", + "issue:44075", + "issue:44220", + "issue:44261", + "issue:44263", + "issue:44360", + "issue:44410", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44493", + "issue:44496", + "issue:44829", + "issue:44933", + "issue:45003", + "issue:45071", + "issue:45278", + "issue:45341", + "issue:45464", + "issue:45478", + "issue:45542", + "issue:45656", + "issue:46172", + "issue:46209", + "issue:46617", + "issue:46762", + "issue:46894" + ], + "soft_pairs": [ + "issue:42371|issue:44261", + "issue:43937|issue:44496", + "issue:39401|issue:46209", + "issue:44410|issue:45478", + "issue:42371|issue:46617", + "issue:44261|issue:46762", + "issue:43208|issue:46894", + "issue:44933|issue:45278", + "issue:38617|issue:44220", + "issue:44075|issue:45656", + "issue:43066|issue:43122", + "issue:44263|issue:44485", + "issue:43097|issue:44493", + "issue:43931|issue:44410", + "issue:43673|issue:45341", + "issue:41628|issue:45071", + "issue:42222|issue:45542", + "issue:45071|issue:45542", + "issue:44483|issue:45464", + "issue:44484|issue:45478", + "issue:43874|issue:44261", + "issue:44263|issue:46172", + "issue:44360|issue:46172", + "issue:27957|issue:43352", + "issue:43012|issue:43577", + "issue:43931|issue:44038", + "issue:43441|issue:44829", + "issue:43097|issue:43452", + "issue:45003|issue:45341", + "issue:43986|issue:45341", + "issue:44261|issue:44485" + ], + "prepared_review_unit_hash": "f960eb6e45a71d1ecc15691f3494b70846ab10a123c34bd2bed8fa10538254f2", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 35864, + "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 mostly a set of unrelated Transformers regressions and model-specific bugs. The only plausible duplicate pair is issue:44483 and issue:45464, which both point to a Qwen3.5 chat/completions streaming failure, but the packet only supports a likely duplicate relationship rather than a confirmed one.", + "confidence": 0.79, + "canonical_issue_reason": "issue:45464 is the best canonical candidate for the likely Qwen3.5 chat/completions streaming bug because it is the more detailed report. The duplicate relationship to issue:44483 looks plausible from the symptom match, but the packet does not provide explicit confirmation.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45464 is the strongest issue to anchor this cluster because it contains the most complete problem description and appears to describe the same underlying streaming chat/completions failure as issue:44483. That said, the evidence is only soft similarity, not hard confirmation.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42371", + "right": "issue:44261", + "accept": false, + "reason": "Both mention precision-related behavior, but one is about TF32 API control and the other about an MLA layernorm config/epsilon mismatch. Different bugs." + }, + { + "left": "issue:43937", + "right": "issue:44496", + "accept": false, + "reason": "GenerationConfig validation and unrecognized model/config loading are separate failure modes, not the same root cause." + }, + { + "left": "issue:39401", + "right": "issue:46209", + "accept": false, + "reason": "Tokenizer offset mapping and FP8 inference on B200 are unrelated symptoms and code paths." + }, + { + "left": "issue:44410", + "right": "issue:45478", + "accept": false, + "reason": "Both involve Qwen family models, but one is missing projections in qwen3next while the other is a Qwen3.5-MoE from_pretrained error. Different defects." + }, + { + "left": "issue:42371", + "right": "issue:46617", + "accept": false, + "reason": "TF32 API behavior and trust_remote_code cache loading are unrelated." + }, + { + "left": "issue:44261", + "right": "issue:46762", + "accept": false, + "reason": "Numerical precision concerns are too broad here; these are different model implementations and different bugs." + }, + { + "left": "issue:43208", + "right": "issue:46894", + "accept": false, + "reason": "Both discuss training issues, but they affect different models and different failure mechanisms." + }, + { + "left": "issue:44933", + "right": "issue:45278", + "accept": false, + "reason": "The latter is a broad report of many import regressions after upgrade; the former is one specific import failure. Not enough evidence they are the same underlying bug." + }, + { + "left": "issue:38617", + "right": "issue:44220", + "accept": false, + "reason": "ImportError for layer_type_validation is unrelated to the _torch_extract_fbank_features issue." + }, + { + "left": "issue:44075", + "right": "issue:45656", + "accept": false, + "reason": "Both are optimizer-related, but one says SGD args are ignored while the other says the optimizer step is called twice under deepspeed." + }, + { + "left": "issue:43066", + "right": "issue:43122", + "accept": false, + "reason": "Both are tokenizer/version-regression reports, but one is about decoder type and the other about output tokenization differences. Too loose to merge." + }, + { + "left": "issue:44263", + "right": "issue:44485", + "accept": false, + "reason": "A torch.split return-value issue and a RoPE implementation discussion in GLM-5 are different concrete bugs." + }, + { + "left": "issue:43097", + "right": "issue:44493", + "accept": false, + "reason": "Removed tie_embeddings_and_encoder_decoder and unexpected position-id keys do not point to the same root cause." + }, + { + "left": "issue:43931", + "right": "issue:44410", + "accept": false, + "reason": "Different Qwen3-VL loading/model-shape problems; model family overlap is not enough." + }, + { + "left": "issue:43673", + "right": "issue:45341", + "accept": false, + "reason": "Chunked prefill cache behavior and a testing_utils bug are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:45071", + "accept": false, + "reason": "AutoImageProcessor import failure and PreTrainedConfig type-checking breakage are separate regressions." + }, + { + "left": "issue:42222", + "right": "issue:45542", + "accept": false, + "reason": "Vitpose model breakage and TensorFlow/tensorboard backend detection are different issues." + }, + { + "left": "issue:45071", + "right": "issue:45542", + "accept": false, + "reason": "Type checking regression is unrelated to undefined tf backend errors." + }, + { + "left": "issue:44483", + "right": "issue:45464", + "accept": true, + "reason": "These appear to describe the same Qwen3.5 chat/completions streaming failure; 45464 is the more detailed report. The match is based on symptom similarity, not explicit confirmation in the packet." + }, + { + "left": "issue:44484", + "right": "issue:45478", + "accept": false, + "reason": "max_shard_size behavior and a Qwen3.5-MoE from_pretrained error are unrelated." + }, + { + "left": "issue:43874", + "right": "issue:44261", + "accept": false, + "reason": "Missing image patch-count method and MLA q_a_layernorm epsilon/config behavior are different model bugs." + }, + { + "left": "issue:44263", + "right": "issue:46172", + "accept": false, + "reason": "torch.split output handling in GLM-MoE-DSA is not the same as the deepseek-v4 dtype/transpose issue." + }, + { + "left": "issue:44360", + "right": "issue:46172", + "accept": false, + "reason": "A missing ReLU in the DSA indexer and a DeepSeek-V4 tensor conversion/transpose issue are distinct." + }, + { + "left": "issue:27957", + "right": "issue:43352", + "accept": false, + "reason": "Flash Attention support for XLM-Roberta and NemotronHForCausalLM are different model support gaps." + }, + { + "left": "issue:43012", + "right": "issue:43577", + "accept": false, + "reason": "PyTorch warning around bfloat16 compilation and BLIP2 dtype propagation are related only at a very high level." + }, + { + "left": "issue:43931", + "right": "issue:44038", + "accept": false, + "reason": "Both are Qwen3-VL-related, but one is a weight-shape mismatch and the other is a Qwen3-VL-Moe loading problem. Different concrete failures." + }, + { + "left": "issue:43441", + "right": "issue:44829", + "accept": false, + "reason": "FlashAttention failures for Ministral-3 and degenerate training with flash_attention_3 are not the same code-path problem." + }, + { + "left": "issue:43097", + "right": "issue:43452", + "accept": false, + "reason": "Removed tie_embeddings/encoder-decoder behavior and gguf_file handling for from_pretrained are unrelated." + }, + { + "left": "issue:45003", + "right": "issue:45341", + "accept": false, + "reason": "Unsafe sys.modules access in modeling_utils and a testing_utils bug do not match." + }, + { + "left": "issue:43986", + "right": "issue:45341", + "accept": false, + "reason": "AutoProcessor crash without torchvision and a testing_utils bug are unrelated." + }, + { + "left": "issue:44261", + "right": "issue:44485", + "accept": false, + "reason": "Precision/epsilon concerns are too generic; these are different GLM-related bugs with different symptoms." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The cluster summary is mostly cautious, but it overstates the evidence for issue:44483 \u2194 issue:45464. The packet only shows soft similarity and no explicit link or other hard confirmation, so calling it the \"only confirmed duplicate pair\" and saying 45464 is the clearer duplicate is stronger than the evidence supports. Please downgrade that language to a tentative/likely duplicate and avoid asserting confirmation or reproducibility from the packet alone." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 54, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:38617", + "issue:42757", + "issue:42890", + "issue:43012", + "issue:43064", + "issue:43097", + "issue:43208", + "issue:43502", + "issue:43531", + "issue:43673", + "issue:43873", + "issue:43874", + "issue:43986", + "issue:43992", + "issue:44038", + "issue:44062", + "issue:44117", + "issue:44263", + "issue:44315", + "issue:44351", + "issue:44380", + "issue:44410", + "issue:44484", + "issue:44704", + "issue:44936", + "issue:44987", + "issue:45103", + "issue:45341", + "issue:45372", + "issue:45375", + "issue:45440", + "issue:45588", + "issue:45753", + "issue:46050", + "issue:46123", + "issue:46172", + "issue:46298", + "issue:46365", + "issue:46369", + "issue:46421", + "issue:46655", + "issue:46664", + "issue:46701", + "issue:46710", + "issue:46752", + "issue:46821", + "issue:46897" + ], + "soft_pairs": [ + "issue:45753|issue:46369", + "issue:44117|issue:44987", + "issue:43874|issue:44263", + "issue:45440|issue:46172", + "issue:43531|issue:45341", + "issue:43673|issue:46123", + "issue:43097|issue:45588", + "issue:44410|issue:45375", + "issue:44038|issue:44410", + "issue:38617|issue:42757", + "issue:43208|issue:46897", + "issue:43064|issue:44315", + "issue:43986|issue:45372", + "issue:43097|issue:46050", + "issue:43986|issue:44484", + "issue:42890|issue:46298", + "issue:46655|issue:46664", + "issue:46172|issue:46710", + "issue:38617|issue:44351", + "issue:46298|issue:46752", + "issue:44062|issue:44117", + "issue:43097|issue:43873", + "issue:43012|issue:46821", + "issue:44117|issue:46701", + "issue:43502|issue:43992", + "issue:44936|issue:46365", + "issue:44704|issue:45103", + "issue:44380|issue:46421" + ], + "prepared_review_unit_hash": "3513f0a0be88f3861cc68588d9e167366059f4f4c5ee1f794c484e0acaa06a97", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35338, + "estimated_input_tokens": 8835, + "estimated_eval_tokens": 17926 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The soft pairs are mostly superficial overlaps (same model family, same API surface, or both are import/load errors) rather than the same underlying bug. I would not merge any of the candidate pairs. Among the listed issues, issue 43097 is the strongest canonical anchor because it is a clearly scoped v5 regression with broad impact and inbound references.", + "confidence": 0.84, + "canonical_issue_reason": "Issue 43097 is the best anchor: it describes a concrete, broad v5 regression with a clear user-visible API break and multiple inbound references.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43097 is the strongest standalone issue in this set because it is specific, actionable, and likely to attract duplicates around the same breaking change.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45753", + "right": "issue:46369", + "accept": false, + "reason": "Same general area of model inference, but one is an infinite loop in Qwen3_5 on a specific image and the other is a SAM memory overflow; different models and failure modes." + }, + { + "left": "issue:44117", + "right": "issue:44987", + "accept": false, + "reason": "Both involve from_pretrained/loading, but one is a tokenizer mapping None bug and the other is a specific model load failure; no clear shared underlying defect." + }, + { + "left": "issue:43874", + "right": "issue:44263", + "accept": false, + "reason": "Both are GLM-related, but one is a missing image-patch helper in the image processor and the other is a torch.split indexing issue; different code paths." + }, + { + "left": "issue:45440", + "right": "issue:46172", + "accept": false, + "reason": "Different DeepSeek model families and different bugs: remote-vs-native implementation divergence versus a dtype/transpose issue in deepseek-v4." + }, + { + "left": "issue:43531", + "right": "issue:45341", + "accept": false, + "reason": "Unrelated: a Qwen3-MoE sliding_window bug versus a testing_utils bug." + }, + { + "left": "issue:43673", + "right": "issue:46123", + "accept": false, + "reason": "Both are runtime behavior issues, but one is GenerationMixin cache/chunked_prefill and the other is MaskGenerationPipeline batch bookkeeping; not the same bug." + }, + { + "left": "issue:43097", + "right": "issue:45588", + "accept": false, + "reason": "Different subsystems and symptoms: a removed config option versus a flash-attention crash on s_aux=None." + }, + { + "left": "issue:44410", + "right": "issue:45375", + "accept": false, + "reason": "Both are Qwen3-family issues, but one is missing attention projections in a layer and the other is a missing config field silently dropped by strict deserialization." + }, + { + "left": "issue:44038", + "right": "issue:44410", + "accept": false, + "reason": "Both concern Qwen3 variants, but one is a version-specific loading bug and the other is a missing projections bug in a different architecture." + }, + { + "left": "issue:38617", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but they involve different modules and different missing symbols; too little evidence of a single underlying change." + }, + { + "left": "issue:43208", + "right": "issue:46897", + "accept": false, + "reason": "Both mention training bugs, but they are different models and different failure patterns; no shared code-path bug is evident." + }, + { + "left": "issue:43064", + "right": "issue:44315", + "accept": false, + "reason": "Both touch training setup, but one is a wrong optimizer-state bug with FSDP2/PEFT/cpu_ram_efficient_loading and the other is Liger Kernel not being applied with model_init." + }, + { + "left": "issue:43986", + "right": "issue:45372", + "accept": false, + "reason": "Both are processor-loading failures, but caused by different missing dependencies/imports and different processor stacks." + }, + { + "left": "issue:43097", + "right": "issue:46050", + "accept": false, + "reason": "A removed API/config option and a quantization save/load corruption bug are unrelated." + }, + { + "left": "issue:43986", + "right": "issue:44484", + "accept": false, + "reason": "Processor loading without torchvision and max_shard_size defaults in save_pretrained are unrelated." + }, + { + "left": "issue:42890", + "right": "issue:46298", + "accept": false, + "reason": "A flaky test-seeding issue and a logging-message issue are not the same underlying bug." + }, + { + "left": "issue:46655", + "right": "issue:46664", + "accept": false, + "reason": "Same SmolVLM video processor area, but one is padding calculation and the other is num_frames semantics; distinct defects." + }, + { + "left": "issue:46172", + "right": "issue:46710", + "accept": false, + "reason": "Different model families and different symptoms: a low-level tensor operation bug versus an accuracy regression after upgrade." + }, + { + "left": "issue:38617", + "right": "issue:44351", + "accept": false, + "reason": "Both are import-name errors from transformers, but they involve different symbols and different exported APIs." + }, + { + "left": "issue:46298", + "right": "issue:46752", + "accept": false, + "reason": "A noisy logging message and an empty-list IndexError in apply_chat_template are unrelated." + }, + { + "left": "issue:44062", + "right": "issue:44117", + "accept": false, + "reason": "One is a tokenizers AddedToken constructor bug; the other is a tokenizer mapping/loading assumption bug." + }, + { + "left": "issue:43097", + "right": "issue:43873", + "accept": false, + "reason": "Removed config/API behavior and quantization offloading behavior are different issues." + }, + { + "left": "issue:43012", + "right": "issue:46821", + "accept": false, + "reason": "A PyTorch precision warning during bfloat16 compilation and a TimesFM window_size AttributeError are unrelated." + }, + { + "left": "issue:44117", + "right": "issue:46701", + "accept": false, + "reason": "Different areas and failure modes: tokenizer mapping/loading versus an image-text-to-text pipeline failure." + }, + { + "left": "issue:43502", + "right": "issue:43992", + "accept": false, + "reason": "One is unintended API requests despite local_files_only=True; the other is a missing embed_tokens weight during UMT5Encoder loading." + }, + { + "left": "issue:44936", + "right": "issue:46365", + "accept": false, + "reason": "A train-then-evaluate lifecycle bug and an import conflict preventing Trainer import are different problems." + }, + { + "left": "issue:44704", + "right": "issue:45103", + "accept": false, + "reason": "AutoProcessor kwargs forwarding and auto_docstring annotation handling are unrelated code paths." + }, + { + "left": "issue:44380", + "right": "issue:46421", + "accept": false, + "reason": "GPT2 attention scaling with SDPA/FlashAttention and chunked-prefill recompilation from cache lazy init are different bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 55, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:36683", + "issue:38617", + "issue:42175", + "issue:42490", + "issue:42757", + "issue:42907", + "issue:43012", + "issue:43064", + "issue:43066", + "issue:43097", + "issue:43208", + "issue:43352", + "issue:43502", + "issue:43526", + "issue:43550", + "issue:43582", + "issue:43618", + "issue:43756", + "issue:43824", + "issue:43825", + "issue:43906", + "issue:43976", + "issue:43986", + "issue:43992", + "issue:44038", + "issue:44077", + "issue:44117", + "issue:44162", + "issue:44222", + "issue:44230", + "issue:44368", + "issue:44380", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44514", + "issue:44655", + "issue:44938", + "issue:45478", + "issue:45479", + "issue:45584", + "issue:45712", + "issue:45753", + "issue:46132", + "issue:46172", + "issue:46178", + "issue:46298", + "issue:46365" + ], + "soft_pairs": [ + "issue:44938|issue:46178", + "issue:43986|issue:44117", + "issue:42907|issue:44655", + "issue:42907|issue:45753", + "issue:43825|issue:44655", + "issue:43012|issue:43550", + "issue:45712|issue:46298", + "issue:43208|issue:43550", + "issue:43012|issue:44368", + "issue:43992|issue:44484", + "issue:43352|issue:43976", + "issue:43352|issue:44380", + "issue:43066|issue:43906", + "issue:44162|issue:46172", + "issue:43097|issue:46298", + "issue:44380|issue:46132", + "issue:45584|issue:46365", + "issue:44077|issue:46365", + "issue:44222|issue:45478", + "issue:36683|issue:43582", + "issue:44380|issue:45479", + "issue:42175|issue:46365", + "issue:42757|issue:43824", + "issue:43064|issue:44380", + "issue:42490|issue:43012", + "issue:43208|issue:43526", + "issue:44483|issue:44514", + "issue:43618|issue:44380", + "issue:38617|issue:43502", + "issue:44038|issue:44230", + "issue:43756|issue:44485" + ], + "prepared_review_unit_hash": "c8f4d1cc0cf4d399fbd95fe99ab3cc0c7aa92bd38022277317f07fd3e32f7413", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36743, + "estimated_input_tokens": 9186, + "estimated_eval_tokens": 18628 + }, + "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 surface keywords (imports, save/load, warnings, attention, backend) but not the same underlying bug. I would not merge any of the listed soft edges as duplicates.", + "confidence": 0.89, + "canonical_issue_reason": "issue:44380 is the closest thing to a representative issue because it is a concrete backend/config bug with the most plausible related edges, but the cluster is too mixed for it to serve as a true duplicate root.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44380 is the best available issue representative, but only for a narrow attention-backend subtheme; it is not a strong global canonical for this cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44938", + "right": "issue:46178", + "accept": false, + "reason": "Different failures: Python 3.14 import/load compatibility vs an invalid BACKENDS_MAPPING backend name." + }, + { + "left": "issue:43986", + "right": "issue:44117", + "accept": false, + "reason": "Both are import-related, but one is a missing torchvision dependency path and the other is a tokenizer mapping assumption bug." + }, + { + "left": "issue:42907", + "right": "issue:44655", + "accept": false, + "reason": "Both mention save/pretrained behavior, but one is a dequantized-model save failure and the other is Pipeline object serialization." + }, + { + "left": "issue:42907", + "right": "issue:45753", + "accept": false, + "reason": "Completely different problems: model saving for dequantized Ministrals/Devstrals vs an inference infinite-loop bug on Qwen3_5." + }, + { + "left": "issue:43825", + "right": "issue:44655", + "accept": false, + "reason": "One is an error-message/task-support issue in pipeline(), the other is failure to save Pipeline objects; different code paths." + }, + { + "left": "issue:43012", + "right": "issue:43550", + "accept": false, + "reason": "A compiler warning about bfloat16 precision is not the same as a torch.compile SDPA failure on Bamba." + }, + { + "left": "issue:45712", + "right": "issue:46298", + "accept": false, + "reason": "Repo cleanup/dummy-class leak vs noisy logging; unrelated maintenance vs logging behavior." + }, + { + "left": "issue:43208", + "right": "issue:43550", + "accept": false, + "reason": "xLSTM training bugs and Bamba torch.compile SDPA failure are different model-specific problems." + }, + { + "left": "issue:43012", + "right": "issue:44368", + "accept": false, + "reason": "Both are warning-style issues, but one is PyTorch precision logging and the other is a tie_word_embeddings config warning." + }, + { + "left": "issue:43992", + "right": "issue:44484", + "accept": false, + "reason": "Missing embed_tokens weight during load is unrelated to a question about max_shard_size in save_pretrained()." + }, + { + "left": "issue:43352", + "right": "issue:43976", + "accept": false, + "reason": "Flash Attention support on NemotronH vs Python-version compatibility for Transformers 5.1.0 are unrelated." + }, + { + "left": "issue:43352", + "right": "issue:44380", + "accept": false, + "reason": "Both involve attention, but one is unsupported Flash Attention on a specific model and the other is GPT2 attention scaling being ignored; different bugs." + }, + { + "left": "issue:43066", + "right": "issue:43906", + "accept": false, + "reason": "Tokenizer decoder type regression and an isolated reproduction of another issue are not enough to establish the same concrete bug." + }, + { + "left": "issue:44162", + "right": "issue:46172", + "accept": false, + "reason": "ESM2 model breakage and a DeepSeek tensor operation bug are unrelated model-specific failures." + }, + { + "left": "issue:43097", + "right": "issue:46298", + "accept": false, + "reason": "Removal of tie_embeddings_and_encoder_decoder and a noisy logging message are unrelated." + }, + { + "left": "issue:44380", + "right": "issue:46132", + "accept": false, + "reason": "Both touch attention behavior, but one is GPT2 attention scaling being ignored and the other is AttentionInterface.register changing callback behavior; not the same code-path bug." + }, + { + "left": "issue:45584", + "right": "issue:46365", + "accept": false, + "reason": "Whisper empty-transcription generation failure and inability to import Trainer due to datasets are unrelated." + }, + { + "left": "issue:44077", + "right": "issue:46365", + "accept": false, + "reason": "A patchtsmixer post_init validation issue has nothing to do with a Trainer import conflict caused by datasets." + }, + { + "left": "issue:44222", + "right": "issue:45478", + "accept": false, + "reason": "Both are MoE/FP8 flavored, but one is save_pretrained for FP8 and the other is from_pretrained failure on Qwen3.5 MoE; different operations and symptoms." + }, + { + "left": "issue:36683", + "right": "issue:43582", + "accept": false, + "reason": "Gemma3Config missing vocab_size and an Apple Silicon caching_allocator TypeError are unrelated." + }, + { + "left": "issue:44380", + "right": "issue:45479", + "accept": false, + "reason": "Attention scaling being ignored is unrelated to the degenerate zero-loss classification bug." + }, + { + "left": "issue:42175", + "right": "issue:46365", + "accept": false, + "reason": "TensorFlow backend installation mix-up and a Trainer import problem involving datasets are unrelated." + }, + { + "left": "issue:42757", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols from different modules and different root causes." + }, + { + "left": "issue:43064", + "right": "issue:44380", + "accept": false, + "reason": "FSDP2 optimizer-state corruption and GPT2 attention scaling behavior are unrelated code paths." + }, + { + "left": "issue:42490", + "right": "issue:43012", + "accept": false, + "reason": "Local-path trailing-slash warning and bfloat16 precision warning are unrelated." + }, + { + "left": "issue:43208", + "right": "issue:43526", + "accept": false, + "reason": "xLSTM training blockers and BeitImageProcessorFast reduce_labels behavior are unrelated." + }, + { + "left": "issue:44483", + "right": "issue:44514", + "accept": false, + "reason": "An OpenAI-compatible chat endpoint rejection and a batched chat-template crash are different API paths." + }, + { + "left": "issue:43618", + "right": "issue:44380", + "accept": false, + "reason": "CLIPOutput attentions assignment and GPT2 attention scaling are both attention-related, but they concern different components and failures." + }, + { + "left": "issue:38617", + "right": "issue:43502", + "accept": false, + "reason": "Missing layer_type_validation import and local_files_only still making API requests are unrelated import vs network behavior bugs." + }, + { + "left": "issue:44038", + "right": "issue:44230", + "accept": false, + "reason": "Both are Qwen3-family/MoE-related, but one is a model compatibility bug and the other is fp8 inference support; not the same concrete issue." + }, + { + "left": "issue:43756", + "right": "issue:44485", + "accept": false, + "reason": "Smollm3 RoPE-layer mismatch and GLM-5 RoPE implementation discussion are different model-specific RoPE 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": 23, + "review_unit_index": 56, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:29870", + "issue:33453", + "issue:36010", + "issue:36296", + "issue:36683", + "issue:42757", + "issue:42890", + "issue:42994", + "issue:43010", + "issue:43064", + "issue:43208", + "issue:43299", + "issue:43352", + "issue:43475", + "issue:43525", + "issue:43611", + "issue:43618", + "issue:43761", + "issue:43867", + "issue:43901", + "issue:43937", + "issue:43976", + "issue:43992", + "issue:44075", + "issue:44077", + "issue:44355", + "issue:44483", + "issue:44492", + "issue:44625", + "issue:44655", + "issue:44855", + "issue:44908", + "issue:44938", + "issue:45071", + "issue:45216", + "issue:45276", + "issue:45278", + "issue:45290", + "issue:45310", + "issue:45375", + "issue:45561", + "issue:46077", + "issue:46082", + "issue:46202", + "issue:46365", + "issue:46829", + "issue:46851", + "issue:46877" + ], + "soft_pairs": [ + "issue:42994|issue:44655", + "issue:43618|issue:43761", + "issue:43299|issue:43867", + "issue:36010|issue:42757", + "issue:44625|issue:46365", + "issue:45276|issue:46077", + "issue:44355|issue:46877", + "issue:43352|issue:43937", + "issue:36683|issue:43475", + "issue:44483|issue:45290", + "issue:43611|issue:44655", + "issue:43901|issue:44655", + "issue:43867|issue:45216", + "issue:43976|issue:44938", + "issue:36683|issue:44077", + "issue:44625|issue:46077", + "issue:46077|issue:46082", + "issue:43618|issue:45216", + "issue:43208|issue:46829", + "issue:36683|issue:43525", + "issue:36683|issue:44625", + "issue:29870|issue:36296", + "issue:43992|issue:45310", + "issue:43064|issue:44492", + "issue:43010|issue:45278", + "issue:44075|issue:46365", + "issue:44908|issue:46851", + "issue:29870|issue:33453", + "issue:44938|issue:45071", + "issue:44355|issue:44855", + "issue:43867|issue:46202", + "issue:36683|issue:43064", + "issue:43867|issue:45375", + "issue:42890|issue:45561" + ], + "prepared_review_unit_hash": "ad5b501bfcf3964dc6a8c1a2b24a2b1259313d102225097f7ca59a1b3ab50123", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37630, + "estimated_input_tokens": 9408, + "estimated_eval_tokens": 19072 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is broad and heterogeneous: import/load regressions, config bugs, training issues, docs/test fixes, and model-specific failures. None of the soft pairs look like the same underlying bug or change, so no duplicates were accepted.", + "confidence": 0.95, + "canonical_issue_reason": "No canonical issue: the items span unrelated bugs across different models, APIs, and failure modes, and none of the soft pairs are true duplicates.", + "canonical_pr_reason": null, + "best_issue_reason": "No single best issue: the cluster is too heterogeneous to nominate one representative duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42994", + "right": "issue:44655", + "accept": false, + "reason": "Both mention saving, but one is quantized model saving while the other is Pipeline.save_pretrained; different object types and failure modes." + }, + { + "left": "issue:43618", + "right": "issue:43761", + "accept": false, + "reason": "Both are CLIP regressions, but one is missing attentions in the output object and the other is hidden_states being None; different bugs." + }, + { + "left": "issue:43299", + "right": "issue:43867", + "accept": false, + "reason": "Both concern loading, but one is a Qwen3VL model-loading regression and the other is a state_dict sorting load error; not the same root cause." + }, + { + "left": "issue:36010", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols from different packages and unrelated breakages." + }, + { + "left": "issue:44625", + "right": "issue:46365", + "accept": false, + "reason": "One is about num_labels propagation in Qwen3.5, the other is importing Trainer being blocked by datasets; unrelated issues." + }, + { + "left": "issue:45276", + "right": "issue:46077", + "accept": false, + "reason": "Both touch Gemma4, but one is embedding resize propagation and the other is attention-mask construction; different code paths." + }, + { + "left": "issue:44355", + "right": "issue:46877", + "accept": false, + "reason": "Both are import-time problems, but one is compiled Python file handling and the other is auto_docstring printing diagnostics; distinct bugs." + }, + { + "left": "issue:43352", + "right": "issue:43937", + "accept": false, + "reason": "Different models and different failures: unsupported Flash Attention 2 vs invalid GenerationConfig." + }, + { + "left": "issue:36683", + "right": "issue:43475", + "accept": false, + "reason": "Both are AttributeErrors, but they hit different model configs/outputs and different missing attributes." + }, + { + "left": "issue:44483", + "right": "issue:45290", + "accept": false, + "reason": "One is a chat/completions request rejection and the other is apply_chat_template crashing on tool-call messages; different layers and causes." + }, + { + "left": "issue:43611", + "right": "issue:44655", + "accept": false, + "reason": "Loading with base_model_prefix and saving Pipeline objects are unrelated failures." + }, + { + "left": "issue:43901", + "right": "issue:44655", + "accept": false, + "reason": "Docs mentioning return_all_scores is a documentation mismatch, while save_pretrained on Pipeline objects is a runtime bug." + }, + { + "left": "issue:43867", + "right": "issue:45216", + "accept": false, + "reason": "Both involve model persistence, but one is load failure with sorted state_dict and the other is an incorrect saved Qwen3.5 checkpoint." + }, + { + "left": "issue:43976", + "right": "issue:44938", + "accept": false, + "reason": "Both are Python-version compatibility reports, but they affect different versions and likely different incompatibilities." + }, + { + "left": "issue:36683", + "right": "issue:44077", + "accept": false, + "reason": "Missing vocab_size on Gemma3Config and ignored SGD args are completely unrelated." + }, + { + "left": "issue:44625", + "right": "issue:46077", + "accept": false, + "reason": "One is Qwen3.5 num_labels propagation, the other is Gemma4 bidirectional attention masking; unrelated." + }, + { + "left": "issue:46077", + "right": "issue:46082", + "accept": false, + "reason": "Different configs and different bugs: Gemma4 attention-mask logic vs LlamaConfig head_dim validation." + }, + { + "left": "issue:43618", + "right": "issue:45216", + "accept": false, + "reason": "CLIP attentions regression and Qwen3.5 save_pretrained regression are unrelated." + }, + { + "left": "issue:43208", + "right": "issue:46829", + "accept": false, + "reason": "xLSTM training bugs and VideoPrism output field mix-up are unrelated model-specific issues." + }, + { + "left": "issue:36683", + "right": "issue:43525", + "accept": false, + "reason": "Both are missing config attributes, but they are different models and different attributes, so not the same bug." + }, + { + "left": "issue:36683", + "right": "issue:44625", + "accept": false, + "reason": "Different missing/propagated config fields in different model families; not a duplicate." + }, + { + "left": "issue:29870", + "right": "issue:36296", + "accept": false, + "reason": "Bart generate() decoder_attention_mask error and tensor-parallel training bug are unrelated." + }, + { + "left": "issue:43992", + "right": "issue:45310", + "accept": false, + "reason": "Both are from_pretrained-style failures, but they affect different models and different missing/loading mechanics." + }, + { + "left": "issue:43064", + "right": "issue:44492", + "accept": false, + "reason": "Trainer/FSDP2 optimizer-state corruption and a typo in cache strategies are unrelated." + }, + { + "left": "issue:43010", + "right": "issue:45278", + "accept": false, + "reason": "Cache update no_grad behavior and broad import errors after an update are different problem classes." + }, + { + "left": "issue:44075", + "right": "issue:46365", + "accept": false, + "reason": "Optimizer arguments not being used and Trainer import failing due to datasets are unrelated." + }, + { + "left": "issue:44908", + "right": "issue:46851", + "accept": false, + "reason": "Scheduler kwargs being ignored and Qwen3.5 variable-length position_ids are unrelated bugs." + }, + { + "left": "issue:29870", + "right": "issue:33453", + "accept": false, + "reason": "Bart generate() decoder_attention_mask error and tokenizer loading regression do not share the same failure mode." + }, + { + "left": "issue:44938", + "right": "issue:45071", + "accept": false, + "reason": "Both are version regressions, but one is Python 3.14 load failure and the other is PretrainedConfig type checking; not the same bug." + }, + { + "left": "issue:44355", + "right": "issue:44855", + "accept": false, + "reason": "Compiled-file import errors and a Python 3.13 IndentationError from torch.jit.script parsing are distinct causes." + }, + { + "left": "issue:43867", + "right": "issue:46202", + "accept": false, + "reason": "Model load failure with sorted state_dict and Gemma pipeline roundtrip failure are unrelated." + }, + { + "left": "issue:36683", + "right": "issue:43064", + "accept": false, + "reason": "Gemma3Config missing vocab_size and FSDP2 optimizer state corruption are unrelated." + }, + { + "left": "issue:43867", + "right": "issue:45375", + "accept": false, + "reason": "A load-model error and a missing deepstack_visual_indexes config field are different bugs." + }, + { + "left": "issue:42890", + "right": "issue:45561", + "accept": false, + "reason": "One is missing set_seed in integration tests, the other is a pytest-xdist race on captured_info.txt; different test infra 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": 23, + "review_unit_index": 57, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:28282", + "issue:30333", + "issue:33453", + "issue:36296", + "issue:41669", + "issue:42548", + "issue:42890", + "issue:42898", + "issue:43010", + "issue:43065", + "issue:43232", + "issue:43502", + "issue:43575", + "issue:43576", + "issue:43618", + "issue:43723", + "issue:43824", + "issue:43867", + "issue:43874", + "issue:43976", + "issue:43992", + "issue:44062", + "issue:44077", + "issue:44112", + "issue:44222", + "issue:44246", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44387", + "issue:44485", + "issue:44556", + "issue:44734", + "issue:44749", + "issue:44961", + "issue:44964", + "issue:45125", + "issue:45357", + "issue:45375", + "issue:45440", + "issue:45478", + "issue:45600", + "issue:45850", + "issue:46132", + "issue:46285", + "issue:46617", + "issue:46664", + "issue:46726" + ], + "soft_pairs": [ + "issue:43010|issue:45850", + "issue:41669|issue:44246", + "issue:36296|issue:43575", + "issue:43010|issue:43976", + "issue:43618|issue:44077", + "issue:43992|issue:45478", + "issue:43867|issue:45357", + "issue:28282|issue:33453", + "issue:43618|issue:46285", + "issue:44360|issue:45440", + "issue:42890|issue:44964", + "issue:42890|issue:44112", + "issue:42890|issue:45600", + "issue:42898|issue:46664", + "issue:43502|issue:44062", + "issue:44734|issue:46617", + "issue:43723|issue:44556", + "issue:44222|issue:45125", + "issue:44387|issue:44556", + "issue:43824|issue:46617", + "issue:43502|issue:44485", + "issue:43874|issue:46617", + "issue:43576|issue:46132", + "issue:43010|issue:43065", + "issue:45125|issue:45478", + "issue:30333|issue:33453", + "issue:44355|issue:44749", + "issue:42548|issue:44351", + "issue:45125|issue:45375", + "issue:46664|issue:46726", + "issue:43502|issue:44222", + "issue:43010|issue:43232", + "issue:43502|issue:44961" + ], + "prepared_review_unit_hash": "7a589fb6ec8a07858e484a10a956a55ad5a3f5421568dc14cb3f2e8d01f92348", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37185, + "estimated_input_tokens": 9297, + "estimated_eval_tokens": 18850 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous and title-similar rather than true duplicates. Nearly all soft edges are false positives across unrelated bugs; the only plausible duplicate pair is the Qwen3.5 MoE / from_pretrained tensor-parallel regression.", + "confidence": 0.63, + "canonical_issue_reason": "issue:45125 is the most specific root-cause report in the set: it identifies the concrete missing _tp_plan problem in the Qwen3.5 MoE path, which cleanly anchors the only near-duplicate-looking pair here.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45125 is the best representative issue because it describes an actionable, narrowly scoped code-path failure rather than a broad symptom or unrelated regression.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43010", + "right": "issue:45850", + "accept": false, + "reason": "Broad transformer regressions and both mention v5-era behavior, but one is about cache update no_grad decoration and the other is a generic complaint about many broken models." + }, + { + "left": "issue:41669", + "right": "issue:44246", + "accept": false, + "reason": "Both touch import performance, but 41669 is a specific import-* cleanup regression while 44246 is a general intermittent slow-import report with no shared code path shown." + }, + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Tensor-parallel is the only overlap; one is a training bug and the other is an OOM on loading a very large model, which are different failure modes and fixes." + }, + { + "left": "issue:43010", + "right": "issue:43976", + "accept": false, + "reason": "These are unrelated: cache/layer no_grad decoration versus a Python-version compatibility issue." + }, + { + "left": "issue:43618", + "right": "issue:44077", + "accept": false, + "reason": "Different models and different bugs: CLIP attentions assignment versus patchtsmixer post_init policy." + }, + { + "left": "issue:43992", + "right": "issue:45478", + "accept": false, + "reason": "Both involve loading, but the reported model families and failure causes differ; missing embed_tokens.weight is not the same as a Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:43867", + "right": "issue:45357", + "accept": false, + "reason": "One is a sorted state_dict load problem, the other is incorrect save_pretrained key mapping for a different model family." + }, + { + "left": "issue:28282", + "right": "issue:33453", + "accept": false, + "reason": "ImportError for missing PyTorch is unrelated to a tokenizer-loading regression." + }, + { + "left": "issue:43618", + "right": "issue:46285", + "accept": false, + "reason": "These are different CLIP-related issues with distinct symptoms and code paths." + }, + { + "left": "issue:44360", + "right": "issue:45440", + "accept": false, + "reason": "Completely different areas: a DSA indexer discussion versus a DeepSeekV3 implementation divergence." + }, + { + "left": "issue:42890", + "right": "issue:44964", + "accept": false, + "reason": "Test seeding flakiness in SamHQ is unrelated to loading Phi-4 multimodal." + }, + { + "left": "issue:42890", + "right": "issue:44112", + "accept": false, + "reason": "Both are CI/test failures, but they affect different tests and root causes; no shared bug is evident." + }, + { + "left": "issue:42890", + "right": "issue:45600", + "accept": false, + "reason": "SamHQ test seeding and auto_mappings stale config references are unrelated maintenance issues." + }, + { + "left": "issue:42898", + "right": "issue:46664", + "accept": false, + "reason": "Tokenization cleanup behavior and SmolVLM frame-count semantics are separate behavior changes in different components." + }, + { + "left": "issue:43502", + "right": "issue:44062", + "accept": false, + "reason": "local_files_only network leakage is unrelated to the AddedToken special-argument TypeError." + }, + { + "left": "issue:44734", + "right": "issue:46617", + "accept": false, + "reason": "One is a KV-cache continuation indexing crash in serving; the other is a cache FileNotFoundError for trust_remote_code models." + }, + { + "left": "issue:43723", + "right": "issue:44556", + "accept": false, + "reason": "Both are loading regressions, but one is tokenizer loading and the other is checkpoint reload across versions; the concrete failure paths differ." + }, + { + "left": "issue:44222", + "right": "issue:45125", + "accept": false, + "reason": "FP8 save_pretrained for MoE and missing _tp_plan for tensor parallelism are different bugs in different APIs." + }, + { + "left": "issue:44387", + "right": "issue:44556", + "accept": false, + "reason": "Int4 memory growth/OOM is unrelated to checkpoint reload compatibility." + }, + { + "left": "issue:43824", + "right": "issue:46617", + "accept": false, + "reason": "Missing exported Qwen2_5_VL class import is unrelated to a trust_remote_code cache FileNotFoundError." + }, + { + "left": "issue:43502", + "right": "issue:44485", + "accept": false, + "reason": "local_files_only request leakage and GLM-5 RoPE implementation are not the same issue." + }, + { + "left": "issue:43874", + "right": "issue:46617", + "accept": false, + "reason": "A missing image-processor method in GLM46V is unrelated to remote-code cache loading." + }, + { + "left": "issue:43576", + "right": "issue:46132", + "accept": false, + "reason": "Broken env CLI output and AttentionInterface.register behavior are distinct subsystems and fixes." + }, + { + "left": "issue:43010", + "right": "issue:43065", + "accept": false, + "reason": "Cache/layer update decoration and a dummy Conv2d in Sam3PixelDecoder are unrelated model implementation issues." + }, + { + "left": "issue:45125", + "right": "issue:45478", + "accept": true, + "reason": "These look like the same Qwen3.5 MoE tensor-parallel/from_pretrained regression: one states the concrete missing _tp_plan root cause, the other reports the resulting from_pretrained failure." + }, + { + "left": "issue:30333", + "right": "issue:33453", + "accept": false, + "reason": "MLflow job reporting and tokenizer loading regression have no shared code path." + }, + { + "left": "issue:44355", + "right": "issue:44749", + "accept": false, + "reason": "Compiled Python file errors and tokenizer-processing slowdowns after upgrade are different problems." + }, + { + "left": "issue:42548", + "right": "issue:44351", + "accept": false, + "reason": "cannot import PreTrainedModel and cannot import HybridCache are different missing-symbol/import-export regressions." + }, + { + "left": "issue:45125", + "right": "issue:45375", + "accept": false, + "reason": "Both are Qwen3.5-related, but one is missing _tp_plan and the other is a vision-config field silently dropped; different bugs." + }, + { + "left": "issue:46664", + "right": "issue:46726", + "accept": false, + "reason": "Video num_frames semantics and missing vision-backend handling in AutoProcessor are unrelated multimodal issues." + }, + { + "left": "issue:43502", + "right": "issue:44222", + "accept": false, + "reason": "Offline loading/network behavior and FP8 MoE save_pretrained are unrelated." + }, + { + "left": "issue:43010", + "right": "issue:43232", + "accept": false, + "reason": "Both mention generation/cache, but one is a no_grad decoration issue and the other is a sync_gpus update-model-kwargs bug." + }, + { + "left": "issue:43502", + "right": "issue:44961", + "accept": false, + "reason": "local_files_only API leakage and a one-word issue title are not the same bug." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is grounded in the packet and stays conservative overall. The only accepted soft edge, issue:45125 \u2194 issue:45478, is plausibly the same Qwen3.5 MoE tensor-parallel/from_pretrained bug, while the other pairwise rejections are supported by distinct titles and failure modes. The summary does not overclaim beyond the evidence shown." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 58, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:29870", + "issue:30333", + "issue:30990", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:36296", + "issue:38617", + "issue:41628", + "issue:42548", + "issue:42757", + "issue:43299", + "issue:43475", + "issue:43576", + "issue:43673", + "issue:43824", + "issue:43827", + "issue:43901", + "issue:44038", + "issue:44222", + "issue:44230", + "issue:44273", + "issue:44315", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44393", + "issue:44485", + "issue:44492", + "issue:44556", + "issue:44623", + "issue:44908", + "issue:44938", + "issue:44961", + "issue:45003", + "issue:45071", + "issue:45125", + "issue:45230", + "issue:45362", + "issue:45468", + "issue:45542", + "issue:45753", + "issue:45850", + "issue:46143", + "issue:46144", + "issue:46209", + "issue:46617", + "issue:46664" + ], + "soft_pairs": [ + "issue:30990|issue:31795", + "issue:43827|issue:44961", + "issue:30333|issue:36296", + "issue:43299|issue:44556", + "issue:29870|issue:31795", + "issue:44355|issue:45071", + "issue:33453|issue:34689", + "issue:42757|issue:43576", + "issue:43576|issue:45071", + "issue:44315|issue:44360", + "issue:31795|issue:38617", + "issue:43576|issue:43824", + "issue:44961|issue:45468", + "issue:44485|issue:44961", + "issue:44351|issue:46617", + "issue:44908|issue:44961", + "issue:43576|issue:44938", + "issue:36296|issue:43901", + "issue:46143|issue:46144", + "issue:43475|issue:46617", + "issue:44222|issue:44230", + "issue:41628|issue:42548", + "issue:44315|issue:46209", + "issue:44485|issue:46664", + "issue:43673|issue:44492", + "issue:45362|issue:45850", + "issue:44273|issue:45542", + "issue:44623|issue:45230", + "issue:44360|issue:46664", + "issue:44038|issue:45125", + "issue:44273|issue:45003", + "issue:44393|issue:45071", + "issue:44393|issue:45753", + "issue:43576|issue:44556" + ], + "prepared_review_unit_hash": "75dfac5f79e41c56fb7f11070583e1a4a7ddda6670847d21728766e341c7d596", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 36918, + "estimated_input_tokens": 9230, + "estimated_eval_tokens": 18716 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "All soft edges are false positives: the cluster mixes unrelated bugs, docs issues, import errors, and model-specific regressions. No pair looks like the same underlying change or fixable in one merged PR.", + "confidence": 0.95, + "canonical_issue_reason": "No single canonical issue fits this cluster; the items are too heterogeneous and the soft similarities are only lexical.", + "canonical_pr_reason": null, + "best_issue_reason": "If forced to pick an anchor, issue:43576 is the most central/referenced concrete v5 regression, but it still does not represent the other issues well.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:30990", + "right": "issue:31795", + "accept": false, + "reason": "Sentence-Transformers loading hang vs documentation confusion about input_ids/past_key_values; different bug types." + }, + { + "left": "issue:43827", + "right": "issue:44961", + "accept": false, + "reason": "Docs about v5 pipeline removals vs an unrelated one-word issue; no shared failure mode." + }, + { + "left": "issue:30333", + "right": "issue:36296", + "accept": false, + "reason": "MLflow job-status reporting vs tensor-parallel training bug; unrelated subsystems and symptoms." + }, + { + "left": "issue:43299", + "right": "issue:44556", + "accept": false, + "reason": "Different model-loading regressions with different version contexts and model families; not the same bug." + }, + { + "left": "issue:29870", + "right": "issue:31795", + "accept": false, + "reason": "Generation-time UnboundLocalError vs forward-signature documentation confusion; not the same underlying issue." + }, + { + "left": "issue:44355", + "right": "issue:45071", + "accept": false, + "reason": "Compiled-Python import/runtime errors are not the same as PretrainedConfig type-checking breakage." + }, + { + "left": "issue:33453", + "right": "issue:34689", + "accept": false, + "reason": "Both are loading regressions, but one is tokenizer loading and the other is a specific Llama 3.2 vision model-load breakage." + }, + { + "left": "issue:42757", + "right": "issue:43576", + "accept": false, + "reason": "ImportError from huggingface_hub vs broken transformers env command; different components and failure modes." + }, + { + "left": "issue:43576", + "right": "issue:45071", + "accept": false, + "reason": "Broken CLI command vs type-checking regression in PretrainedConfig; not the same code path." + }, + { + "left": "issue:44315", + "right": "issue:44360", + "accept": false, + "reason": "Liger Kernel not applied in model_init vs DSA indexer lacking ReLU; unrelated training vs debug-tooling behavior." + }, + { + "left": "issue:31795", + "right": "issue:38617", + "accept": false, + "reason": "Doc clarification for model.forward vs ImportError for layer_type_validation; no common bug." + }, + { + "left": "issue:43576", + "right": "issue:43824", + "accept": false, + "reason": "CLI/environment command failure vs missing Qwen2.5-VL export; different areas and fixes." + }, + { + "left": "issue:44961", + "right": "issue:45468", + "accept": false, + "reason": "A placeholder title vs Gemma-4 audio positional encoding bug; clearly unrelated." + }, + { + "left": "issue:44485", + "right": "issue:44961", + "accept": false, + "reason": "GLM-5 RoPE discussion vs placeholder issue; no substantive overlap." + }, + { + "left": "issue:44351", + "right": "issue:46617", + "accept": false, + "reason": "Missing top-level import vs file-not-found when loading cache; both import-related but not the same failure." + }, + { + "left": "issue:44908", + "right": "issue:44961", + "accept": false, + "reason": "Scheduler kwargs ignored vs placeholder issue; no duplicate signal." + }, + { + "left": "issue:43576", + "right": "issue:44938", + "accept": false, + "reason": "Broken env command in v5 vs Python 3.14 loading failure; different root causes." + }, + { + "left": "issue:36296", + "right": "issue:43901", + "accept": false, + "reason": "Tensor parallel training bug vs docs mismatch for TextClassificationPipeline; unrelated." + }, + { + "left": "issue:46143", + "right": "issue:46144", + "accept": false, + "reason": "Missing kwargs passthrough in RoFormer methods vs attention-interface implementation issue; related family, different concrete bug." + }, + { + "left": "issue:43475", + "right": "issue:46617", + "accept": false, + "reason": "SAM 3 Video attribute error vs trust_remote_code cache FileNotFoundError; unrelated models and code paths." + }, + { + "left": "issue:44222", + "right": "issue:44230", + "accept": false, + "reason": "FP8 save_pretrained problem vs FP8 inference support request; one is saving, the other is inference capability." + }, + { + "left": "issue:41628", + "right": "issue:42548", + "accept": false, + "reason": "Missing AutoImageProcessor export vs missing PreTrainedModel export; similar import symptom but different symbols and likely fixes." + }, + { + "left": "issue:44315", + "right": "issue:46209", + "accept": false, + "reason": "model_init/Liger Kernel application bug vs FP8 inference on B200 hardware; not the same code path." + }, + { + "left": "issue:44485", + "right": "issue:46664", + "accept": false, + "reason": "GLM-5 RoPE implementation discussion vs SmolVLM num_frames semantics; different models and bugs." + }, + { + "left": "issue:43673", + "right": "issue:44492", + "accept": false, + "reason": "GenerationMixin cache missing during chunked_prefill vs cache-strategy typo; one is functional regression, the other is wording." + }, + { + "left": "issue:45362", + "right": "issue:45850", + "accept": false, + "reason": "Qwen3.5 chat crash vs broad complaint that v5 breaks many models; different granularity and symptoms." + }, + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Lazy loading malfunction vs TensorBoard-without-TensorFlow backend error; unrelated initialization issues." + }, + { + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "Missing processor.save_pretrained files vs generic bug report; no shared underlying defect." + }, + { + "left": "issue:44360", + "right": "issue:46664", + "accept": false, + "reason": "DSA indexer ReLU discussion vs SmolVLM video-frame semantics; unrelated subsystems." + }, + { + "left": "issue:44038", + "right": "issue:45125", + "accept": false, + "reason": "Qwen3-VL / model loading report vs missing _tp_plan for tensor parallelism; different bugs." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Lazy loading issue vs unsafe sys.modules access in modeling_utils; distinct root causes." + }, + { + "left": "issue:44393", + "right": "issue:45071", + "accept": false, + "reason": "Qwen3-VL bounding-box output error vs PretrainedConfig type-checking breakage; different models and failure modes." + }, + { + "left": "issue:44393", + "right": "issue:45753", + "accept": false, + "reason": "Bounding-box hallucination/error vs infinite loop on a specific image; both Qwen3.5-VL but different concrete behaviors." + }, + { + "left": "issue:43576", + "right": "issue:44556", + "accept": false, + "reason": "Broken env command vs checkpoint reload regression after upgrade; unrelated symptoms and 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": 23, + "review_unit_index": 59, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:31795", + "issue:33453", + "issue:34689", + "issue:41669", + "issue:42371", + "issue:42890", + "issue:43352", + "issue:43408", + "issue:43519", + "issue:43525", + "issue:43572", + "issue:43576", + "issue:43673", + "issue:43704", + "issue:43976", + "issue:44077", + "issue:44230", + "issue:44246", + "issue:44273", + "issue:44295", + "issue:44297", + "issue:44315", + "issue:44351", + "issue:44360", + "issue:44393", + "issue:44485", + "issue:44492", + "issue:44556", + "issue:44655", + "issue:44704", + "issue:44829", + "issue:44861", + "issue:44908", + "issue:45071", + "issue:45092", + "issue:45230", + "issue:45278", + "issue:45310", + "issue:45341", + "issue:45468", + "issue:45478", + "issue:45850", + "issue:46095", + "issue:46218", + "issue:46328", + "issue:46655", + "issue:46894" + ], + "soft_pairs": [ + "issue:41669|issue:43673", + "issue:43525|issue:45850", + "issue:44556|issue:45092", + "issue:43572|issue:45850", + "issue:44492|issue:45850", + "issue:45278|issue:45850", + "issue:44230|issue:46095", + "issue:42371|issue:44492", + "issue:41669|issue:46218", + "issue:43408|issue:44315", + "issue:42890|issue:44908", + "issue:44246|issue:44273", + "issue:43576|issue:44908", + "issue:43352|issue:44393", + "issue:43976|issue:44393", + "issue:44908|issue:45071", + "issue:44295|issue:45230", + "issue:44297|issue:45230", + "issue:44704|issue:44908", + "issue:44492|issue:46328", + "issue:41669|issue:44351", + "issue:41669|issue:45341", + "issue:43976|issue:45850", + "issue:44393|issue:45478", + "issue:43976|issue:46218", + "issue:44077|issue:44861", + "issue:31795|issue:33453", + "issue:43576|issue:44273", + "issue:44315|issue:44829", + "issue:44360|issue:45468", + "issue:43519|issue:46655", + "issue:44273|issue:44908", + "issue:44655|issue:44861", + "issue:45230|issue:45310", + "issue:43576|issue:46218", + "issue:43519|issue:44485", + "issue:45468|issue:46894", + "issue:43519|issue:44908", + "issue:41669|issue:43704", + "issue:31795|issue:34689" + ], + "prepared_review_unit_hash": "4b2d0c81a3d6ddfc37ca664e41f054f4550e86b7c12ed4c3796c7665aed45b9a", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 40, + "serialized_chars": 38543, + "estimated_input_tokens": 9636, + "estimated_eval_tokens": 19528 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose semantic grab-bag of transformer regressions and model-specific bugs, with only superficial overlap around v5 upgrades, loading/import issues, and lazy-loading/performance complaints. I do not see any true duplicate pairs among the soft edges.", + "confidence": 0.9, + "canonical_issue_reason": "Issue 45850 is the closest umbrella-style report here because it broadly summarizes the v5 breakage theme, but the cluster is not a tight duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45850 is the best representative issue for the cluster since it is the most general and central to the shared 'v5 breaks things' pattern, even though most members are not actual duplicates.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41669", + "right": "issue:43673", + "accept": false, + "reason": "Different problems: import-star / import-time slowdown versus missing GenerationMixin cache during chunked prefill." + }, + { + "left": "issue:43525", + "right": "issue:45850", + "accept": false, + "reason": "One is a specific Llama4Config attribute error; the other is a broad complaint about many models breaking in v5." + }, + { + "left": "issue:44556", + "right": "issue:45092", + "accept": false, + "reason": "Both involve upgrade incompatibilities, but they describe different failure modes and model families." + }, + { + "left": "issue:43572", + "right": "issue:45850", + "accept": false, + "reason": "StableLmConfig pad_token_idx regression is a specific config bug, not the same as the generic v5 breakage report." + }, + { + "left": "issue:44492", + "right": "issue:45850", + "accept": false, + "reason": "A cache-strategy typo is unrelated to a broad model-loading regression." + }, + { + "left": "issue:45278", + "right": "issue:45850", + "accept": false, + "reason": "Both are post-v5 regressions, but one is import errors and the other is a vague many-model breakage complaint." + }, + { + "left": "issue:44230", + "right": "issue:46095", + "accept": false, + "reason": "fp8 support for Qwen3-VL/Qwen3.5 differs completely from a DeepSeek weight-conversion question." + }, + { + "left": "issue:42371", + "right": "issue:44492", + "accept": false, + "reason": "TF32 API guidance and cache-strategy typo are unrelated topics." + }, + { + "left": "issue:41669", + "right": "issue:46218", + "accept": false, + "reason": "Import-star cleanup/performance and a missing Qwen3_5 import error are not the same bug." + }, + { + "left": "issue:43408", + "right": "issue:44315", + "accept": false, + "reason": "SAM model-type mismatch is unrelated to Liger Kernel not being applied under model_init." + }, + { + "left": "issue:42890", + "right": "issue:44908", + "accept": false, + "reason": "A flaky integration test due to missing seeds is unrelated to an inverse_sqrt scheduler kwargs bug." + }, + { + "left": "issue:44246", + "right": "issue:44273", + "accept": false, + "reason": "Both mention import/lazy-loading behavior, but one is slow import timing and the other is a functional lazy-loading failure; too weak to merge as duplicates." + }, + { + "left": "issue:43576", + "right": "issue:44908", + "accept": false, + "reason": "A broken transformers env command is unrelated to scheduler kwargs handling." + }, + { + "left": "issue:43352", + "right": "issue:44393", + "accept": false, + "reason": "Nemotron Flash Attention 2 support and Qwen3-VL bounding-box errors are different model/code paths." + }, + { + "left": "issue:43976", + "right": "issue:44393", + "accept": false, + "reason": "Python-version compatibility and Qwen3-VL output errors are unrelated." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Missing scheduler kwargs and PretrainedConfig type-checking are different regressions." + }, + { + "left": "issue:44295", + "right": "issue:45230", + "accept": false, + "reason": "A position_ids buffer read error is too generic to equate with an empty 'bug report' placeholder." + }, + { + "left": "issue:44297", + "right": "issue:45230", + "accept": false, + "reason": "Tokenizer save_pretrained class mismatch is unrelated to the placeholder bug report." + }, + { + "left": "issue:44704", + "right": "issue:44908", + "accept": false, + "reason": "AutoProcessor kwargs forwarding is unrelated to inverse_sqrt scheduler kwargs." + }, + { + "left": "issue:44492", + "right": "issue:46328", + "accept": false, + "reason": "Cache-strategy typo and SequenceFeatureExtractor.pad list conversion are different issues." + }, + { + "left": "issue:41669", + "right": "issue:44351", + "accept": false, + "reason": "Import-star/lazy loading cleanup is not the same as a missing HybridCache export/import error." + }, + { + "left": "issue:41669", + "right": "issue:45341", + "accept": false, + "reason": "Import performance cleanup and a testing_utils bug are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:45850", + "accept": false, + "reason": "Python support incompatibility and a broad v5 breakage complaint are not the same underlying bug." + }, + { + "left": "issue:44393", + "right": "issue:45478", + "accept": false, + "reason": "Both are Qwen-related, but one is a 2D bbox hallucination/error and the other is a Qwen3.5 MoE from_pretrained failure." + }, + { + "left": "issue:43976", + "right": "issue:46218", + "accept": false, + "reason": "Python-version compatibility is unrelated to a specific missing Qwen3_5 import." + }, + { + "left": "issue:44077", + "right": "issue:44861", + "accept": false, + "reason": "Optional post_init enforcement and a tied-weights AttributeError are different code paths." + }, + { + "left": "issue:31795", + "right": "issue:33453", + "accept": false, + "reason": "Documentation confusion around forward arguments is unrelated to tokenizer loading regression." + }, + { + "left": "issue:43576", + "right": "issue:44273", + "accept": false, + "reason": "Broken env command and lazy-loading behavior are not the same bug." + }, + { + "left": "issue:44315", + "right": "issue:44829", + "accept": false, + "reason": "Liger Kernel application under model_init is unrelated to flash_attention_3 training degeneration." + }, + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "A DSA/indexer discussion and a Gemma-4 audio positional encoding bug are unrelated." + }, + { + "left": "issue:43519", + "right": "issue:46655", + "accept": false, + "reason": "Incorrect timestamp calculation and incorrect padding calculation affect different processors and modalities." + }, + { + "left": "issue:44273", + "right": "issue:44908", + "accept": false, + "reason": "Lazy-loading problems and scheduler kwargs handling are unrelated." + }, + { + "left": "issue:44655", + "right": "issue:44861", + "accept": false, + "reason": "Pipeline save_pretrained support and tied-weight key handling are distinct issues." + }, + { + "left": "issue:45230", + "right": "issue:45310", + "accept": false, + "reason": "A placeholder bug report is not a duplicate of the Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:43576", + "right": "issue:46218", + "accept": false, + "reason": "Broken env CLI and missing Qwen3_5 import are unrelated." + }, + { + "left": "issue:43519", + "right": "issue:44485", + "accept": false, + "reason": "Qwen3VL timestamp calculation and GLM-5 RoPE implementation are different model-specific bugs." + }, + { + "left": "issue:45468", + "right": "issue:46894", + "accept": false, + "reason": "Gemma-4 audio positional encoding and CohereASR training-loss double shift are different bugs." + }, + { + "left": "issue:43519", + "right": "issue:44908", + "accept": false, + "reason": "Video timestamp calculation and scheduler kwargs handling are unrelated." + }, + { + "left": "issue:41669", + "right": "issue:43704", + "accept": false, + "reason": "Import cleanup/performance and a VRAM leak in multi-threaded dataloaders are different issues." + }, + { + "left": "issue:31795", + "right": "issue:34689", + "accept": false, + "reason": "Forward-doc confusion and a Llama 3.2 Vision model-loading breakage are not duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 23, + "review_unit_index": 60, + "review_unit_count": 60, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-500", + "nodes": [ + "issue:28348", + "issue:29870", + "issue:30333", + "issue:31515", + "issue:31795", + "issue:36296", + "issue:37934", + "issue:38617", + "issue:42371", + "issue:42913", + "issue:42936", + "issue:43066", + "issue:43240", + "issue:43298", + "issue:43352", + "issue:43502", + "issue:43519", + "issue:43525", + "issue:43572", + "issue:43618", + "issue:43668", + "issue:43704", + "issue:43717", + "issue:43813", + "issue:44031", + "issue:44230", + "issue:44261", + "issue:44263", + "issue:44351", + "issue:44485", + "issue:44492", + "issue:44623", + "issue:44861", + "issue:44908", + "issue:44938", + "issue:44998", + "issue:45250", + "issue:45290", + "issue:45310", + "issue:45468", + "issue:45850", + "issue:46018", + "issue:46095", + "issue:46144", + "issue:46172", + "issue:46218" + ], + "soft_pairs": [ + "issue:46018|issue:46172", + "issue:43704|issue:44485", + "issue:44485|issue:45468", + "issue:43502|issue:43519", + "issue:42371|issue:43704", + "issue:45850|issue:46218", + "issue:43519|issue:44623", + "issue:44263|issue:45468", + "issue:44261|issue:45468", + "issue:43298|issue:46095", + "issue:43519|issue:43704", + "issue:44351|issue:46218", + "issue:43352|issue:46218", + "issue:42913|issue:43066", + "issue:28348|issue:45250", + "issue:44908|issue:44998", + "issue:44492|issue:44998", + "issue:43618|issue:44998", + "issue:44998|issue:45850", + "issue:44938|issue:44998", + "issue:44861|issue:45290", + "issue:31515|issue:38617", + "issue:43572|issue:44861", + "issue:43525|issue:44861", + "issue:44230|issue:44623", + "issue:28348|issue:43352", + "issue:29870|issue:31515", + "issue:30333|issue:31515", + "issue:31515|issue:36296", + "issue:31515|issue:31795", + "issue:44230|issue:45310", + "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": "b05f307253f5c209236b43df5d122ff62485dae1e9221a2e16dcd213192165be", + "component_budget": { + "node_count": 500, + "item_count": 500, + "soft_pair_count": 1839, + "serialized_chars": 953912, + "estimated_input_tokens": 238478, + "estimated_eval_tokens": 477212 + }, + "budget": { + "node_count": 46, + "item_count": 46, + "soft_pair_count": 44, + "serialized_chars": 38880, + "estimated_input_tokens": 9720, + "estimated_eval_tokens": 19696 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a loose cluster of mostly unrelated transformer/model regression issues, with many false-positive similarity hits. No soft pair looks like the same concrete bug/change, and there are no PRs in the set.", + "confidence": 0.42, + "canonical_issue_reason": "issue:45850 is the closest thing to an umbrella report for broad v5 breakage/regression complaints, so it is the best single representative even though it is very vague.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45850 best fits as a global representative because it spans the broadest failure theme seen here (many models breaking after 5.0), but it is still too generic to be an ideal canonical duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46018", + "right": "issue:46172", + "accept": false, + "reason": "Same model family, but different bugs: shared-expert gating vs a DeepSeek-V4 weight conversion/transpose issue." + }, + { + "left": "issue:43704", + "right": "issue:44485", + "accept": false, + "reason": "Completely different models and failures: Qwen3 VRAM leak vs GLM-5 RoPE implementation." + }, + { + "left": "issue:44485", + "right": "issue:45468", + "accept": false, + "reason": "Different architectures and symptoms: GLM-5 RoPE vs Gemma-4 audio positional encoding." + }, + { + "left": "issue:43502", + "right": "issue:43519", + "accept": false, + "reason": "One is unintended network access with local_files_only, the other is a timestamp calculation bug in a processor." + }, + { + "left": "issue:42371", + "right": "issue:43704", + "accept": false, + "reason": "TF32 configuration guidance is unrelated to a Qwen3 VRAM leak in dataloader threads." + }, + { + "left": "issue:45850", + "right": "issue:46218", + "accept": false, + "reason": "A broad v5 breakage complaint is not the same concrete issue as a Qwen3_5 import failure." + }, + { + "left": "issue:43519", + "right": "issue:44623", + "accept": false, + "reason": "Different code paths: Qwen3VL timestamp math vs missing files in processor.save_pretrained()." + }, + { + "left": "issue:44263", + "right": "issue:45468", + "accept": false, + "reason": "torch.split return handling is unrelated to Gemma-4 audio positional encoding." + }, + { + "left": "issue:44261", + "right": "issue:45468", + "accept": false, + "reason": "RMS norm epsilon precision issue is unrelated to Gemma-4 audio positional encoding." + }, + { + "left": "issue:43298", + "right": "issue:46095", + "accept": false, + "reason": "Audio continuous batching support and DeepSeek-V4 weight conversion are different feature requests." + }, + { + "left": "issue:43519", + "right": "issue:43704", + "accept": false, + "reason": "Timestamp calculation bug and VRAM leak are different failures in different models." + }, + { + "left": "issue:44351", + "right": "issue:46218", + "accept": false, + "reason": "Both are import errors, but for different symbols and likely different root causes." + }, + { + "left": "issue:43352", + "right": "issue:46218", + "accept": false, + "reason": "Flash Attention support for NemotronH is unrelated to a Qwen3_5 module import failure." + }, + { + "left": "issue:42913", + "right": "issue:43066", + "accept": false, + "reason": "Both involve tokenizer/v5 behavior, but one is a broad behavior regression and the other is a specific decoder-type bug." + }, + { + "left": "issue:28348", + "right": "issue:45250", + "accept": false, + "reason": "GPT2-specific Flash Attention 2.0 support request is not the same as a generic Flash Attention 2.0 issue." + }, + { + "left": "issue:44908", + "right": "issue:44998", + "accept": false, + "reason": "Scheduler kwargs handling has no relation to the unrelated 'Unemployment' issue." + }, + { + "left": "issue:44492", + "right": "issue:44998", + "accept": false, + "reason": "Cache-strategy typo and the unrelated title clearly do not describe the same bug." + }, + { + "left": "issue:43618", + "right": "issue:44998", + "accept": false, + "reason": "CLIPOutput attentions regression is unrelated to the unrelated title." + }, + { + "left": "issue:44938", + "right": "issue:44998", + "accept": false, + "reason": "Python 3.14 import/load failure is unrelated to the unrelated title." + }, + { + "left": "issue:44861", + "right": "issue:45290", + "accept": false, + "reason": "Different crashes in different subsystems: tied-weights helper vs chat template tool-call handling." + }, + { + "left": "issue:31515", + "right": "issue:38617", + "accept": false, + "reason": "Slow from_pretrained checkpoint loading is unrelated to a missing import symbol." + }, + { + "left": "issue:43572", + "right": "issue:44861", + "accept": false, + "reason": "Missing pad_token_idx in config is unrelated to _get_tied_weight_keys crashing on list input." + }, + { + "left": "issue:43525", + "right": "issue:44861", + "accept": false, + "reason": "Llama4Config pad_token_id missing and tied-weights handling are different issues." + }, + { + "left": "issue:44230", + "right": "issue:44623", + "accept": false, + "reason": "FP8 inference support request is unrelated to missing files in processor.save_pretrained()." + }, + { + "left": "issue:28348", + "right": "issue:43352", + "accept": false, + "reason": "Both mention Flash Attention 2.0, but for different models and different support gaps." + }, + { + "left": "issue:29870", + "right": "issue:31515", + "accept": false, + "reason": "BART generate decoder_attention_mask bug is unrelated to slow checkpoint loading." + }, + { + "left": "issue:30333", + "right": "issue:31515", + "accept": false, + "reason": "MLflow reporting on training failure is unrelated to from_pretrained load performance." + }, + { + "left": "issue:31515", + "right": "issue:36296", + "accept": false, + "reason": "Slow checkpoint loading and tensor-parallel training bugs are different problems." + }, + { + "left": "issue:31515", + "right": "issue:31795", + "accept": false, + "reason": "Load performance is unrelated to documentation confusion about forward inputs." + }, + { + "left": "issue:44230", + "right": "issue:45310", + "accept": false, + "reason": "FP8 inference support and Qwen3.5 from_pretrained import errors are different issues." + }, + { + "left": "issue:43298", + "right": "issue:44230", + "accept": false, + "reason": "Audio continuous batching support and FP8 inference support are different feature requests." + }, + { + "left": "issue:43717", + "right": "issue:46144", + "accept": false, + "reason": "Mamba init_weights behavior is unrelated to RoFormer not using the attention interface." + }, + { + "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 problem in ModernBERTConfig is unrelated to tokenizer regex warnings." + }, + { + "left": "issue:42936", + "right": "issue:43240", + "accept": false, + "reason": "Mask2former ignore_value and kwargs not passed to loss are different code paths." + }, + { + "left": "issue:43240", + "right": "issue:46144", + "accept": false, + "reason": "Loss kwargs propagation is unrelated to RoFormer attention interface usage." + }, + { + "left": "issue:42936", + "right": "issue:44031", + "accept": false, + "reason": "Mask2former ignore_value and tokenizer regex warnings are unrelated." + }, + { + "left": "issue:42936", + "right": "issue:46144", + "accept": false, + "reason": "Mask2former ignore_value and RoFormer attention interface are unrelated bugs." + }, + { + "left": "issue:42936", + "right": "issue:43813", + "accept": false, + "reason": "Ignore_value handling is unrelated to a typo in quantization_operation." + }, + { + "left": "issue:43240", + "right": "issue:43668", + "accept": false, + "reason": "Passing kwargs to loss and a config type-hint mismatch are different issues." + }, + { + "left": "issue:43668", + "right": "issue:43813", + "accept": false, + "reason": "ModernBERT norm_eps typing and a quantization typo are unrelated." + }, + { + "left": "issue:37934", + "right": "issue:43668", + "accept": false, + "reason": "Questioning whether Llama4TextL2Norm is RMS norm is unrelated to a ModernBERT type-hint issue." + }, + { + "left": "issue:42936", + "right": "issue:43668", + "accept": false, + "reason": "Mask2former ignore_value handling is unrelated to a ModernBERT norm_eps typing bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 23, + "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": "5ba1e4116af6d658a42b901e6fc66417465227b9805dc083af462984337500f8", + "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": "False-positive cluster: the three issues are unrelated feature requests in different areas (tokenizer truncation behavior, learning-rate scheduling, and Wav2Vec2 word boundary offsets).", + "confidence": 0.98, + "canonical_issue_reason": "No canonical issue: there is no shared underlying bug or enhancement across these tickets.", + "canonical_pr_reason": null, + "best_issue_reason": "None of the issues is a good cluster representative because they address different code paths and user needs; if forced, 10519 is the most concrete standalone request, but it is not a duplicate of the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:10519", + "right": "issue:11307", + "accept": false, + "reason": "Different features in different subsystems: tokenizer truncation vs Wav2Vec2 word offset extraction; no shared bug or fix path." + }, + { + "left": "issue:10519", + "right": "issue:11005", + "accept": false, + "reason": "Unrelated requests: truncation strategy for tokenizers vs ReduceLROnPlateau-like scheduler behavior." + }, + { + "left": "issue:11005", + "right": "issue:11307", + "accept": false, + "reason": "Distinct domains and problems: learning-rate scheduling vs speech model word timing offsets." + } + ] + }, + "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": 23, + "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": "1eaa8d29dc06c082ed75dc9d4e81e5090c54881f1547c2376391c13db944c9c2", + "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": "The three issues are only loosely related through model/tokenizer loading, but they describe different problems: Tapas tokenization mismatch, a private-model auth/error-message issue, and a bert-base-uncased loading failure. I would not merge any of the soft pairs as duplicates.", + "confidence": 0.88, + "canonical_issue_reason": "issue #16618 is the best representative of the cluster because it is the most general concrete model-loading failure and has the strongest apparent overlap with the other two, even though the underlying problems are still distinct.", + "canonical_pr_reason": null, + "best_issue_reason": "#16618 is the clearest central issue to keep as the cluster anchor: it is the most user-facing loading failure and the closest semantic match to the other titles, but not a true duplicate of either.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:13244", + "right": "issue:16618", + "accept": false, + "reason": "Reject: one is about Tapas tokenization diverging from TensorFlow code, the other is a failure to load bert-base-uncased. They involve different models and a different class of bug." + }, + { + "left": "issue:15980", + "right": "issue:16618", + "accept": false, + "reason": "Reject: #15980 is specifically about a bad error message when downloading a private model without login, while #16618 reports inability to load bert-base-uncased. Same broad area, but not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 4, + "component_count": 23, + "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": "fe1eddfd054fe5f129dade87e2db858dac6f1fb1bcc9aaf1648d9445199ea9ad", + "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 distinct community contribution feature requests: one for Flash Attention 2 support and one for GGUF support. They target different capabilities and architectures, so they should not be merged as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": "Issue #33260 is the better representative of the cluster only in a loose triage sense because it is newer and has more inbound references, but it is not a duplicate of #26350.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #33260 is the strongest standalone issue because it has higher visible engagement and is phrased as a clear support request, but it addresses a different feature than #26350.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:26350", + "right": "issue:33260", + "accept": false, + "reason": "Different underlying feature requests: Flash Attention 2 support vs GGUF support. They concern separate code paths and architectures, not the same bug or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 5, + "component_count": 23, + "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-27088-9", + "nodes": [ + "issue:27088", + "issue:28936", + "issue:38435", + "issue:43834", + "issue:44365", + "issue:44370", + "issue:46222", + "issue:46226", + "issue:46931" + ], + "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:43834|issue:46931", + "issue:44365|issue:44370", + "issue:44365|issue:46222", + "issue:44365|issue:46226", + "issue:44365|issue:46931", + "issue:44370|issue:46222", + "issue:44370|issue:46226", + "issue:44370|issue:46931", + "issue:46222|issue:46226", + "issue:46222|issue:46931", + "issue:46226|issue:46931" + ], + "prepared_review_unit_hash": "d0cf7f5ebf59053edb1fee650d26714b78066996893f8c602fa8503caef81443", + "component_budget": { + "node_count": 9, + "item_count": 9, + "soft_pair_count": 29, + "serialized_chars": 15983, + "estimated_input_tokens": 3996, + "estimated_eval_tokens": 8248 + }, + "budget": { + "node_count": 9, + "item_count": 9, + "soft_pair_count": 29, + "serialized_chars": 15983, + "estimated_input_tokens": 3996, + "estimated_eval_tokens": 8248 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The cluster is heterogeneous overall: the Turkish, Spanish, and Romanian translation requests look like separate language-specific issues, while there are two clearer duplicate pockets inside it \u2014 the generic [i18n-] template trio and the identical Thai-titled pair. I would only collapse those pockets internally, not the full cluster.", + "confidence": 0.74, + "canonical_issue_reason": "issue:43834 is the best canonical within the generic placeholder/template pocket: it is the clearest representative and the earliest of the near-identical [i18n-] issues.", + "canonical_pr_reason": null, + "best_issue_reason": "As a standalone issue, issue:43834 is the strongest anchor for one duplicate pocket because it matches the near-identical template-style issues most closely; the language-specific translation requests should remain separate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:27088", + "right": "issue:44370", + "accept": false, + "reason": "Different language-specific translation request; same broad theme, but not the same issue." + }, + { + "left": "issue:27088", + "right": "issue:44365", + "accept": false, + "reason": "Different language-specific request; not a duplicate." + }, + { + "left": "issue:27088", + "right": "issue:43834", + "accept": false, + "reason": "Turkish translation request vs generic template-style issue; different concrete request." + }, + { + "left": "issue:27088", + "right": "issue:38435", + "accept": false, + "reason": "Separate language-specific translation requests." + }, + { + "left": "issue:27088", + "right": "issue:28936", + "accept": false, + "reason": "Turkish and Spanish translation requests are distinct." + }, + { + "left": "issue:28936", + "right": "issue:43834", + "accept": false, + "reason": "Spanish translation request is different from the generic placeholder/template issue." + }, + { + "left": "issue:28936", + "right": "issue:44365", + "accept": false, + "reason": "Different language-specific request; not the same underlying issue." + }, + { + "left": "issue:28936", + "right": "issue:38435", + "accept": false, + "reason": "Spanish vs Romanian translation requests are separate." + }, + { + "left": "issue:28936", + "right": "issue:44370", + "accept": false, + "reason": "Different language-specific request; not a duplicate." + }, + { + "left": "issue:38435", + "right": "issue:44370", + "accept": false, + "reason": "Romanian translation request is distinct from the template-style duplicate pocket." + }, + { + "left": "issue:38435", + "right": "issue:44365", + "accept": false, + "reason": "Different target language and separate request." + }, + { + "left": "issue:38435", + "right": "issue:43834", + "accept": false, + "reason": "Romanian translation issue vs generic template-style issue; not the same request." + }, + { + "left": "issue:38435", + "right": "issue:46222", + "accept": false, + "reason": "Different topic and different apparent request; only superficial similarity." + }, + { + "left": "issue:38435", + "right": "issue:46226", + "accept": false, + "reason": "Different topic and different apparent request; not safely mergeable." + }, + { + "left": "issue:43834", + "right": "issue:44370", + "accept": true, + "reason": "Near-identical template-style i18n issue; same underlying request." + }, + { + "left": "issue:43834", + "right": "issue:44365", + "accept": true, + "reason": "Near-identical template-style i18n issue; same underlying request." + }, + { + "left": "issue:43834", + "right": "issue:46222", + "accept": false, + "reason": "Different title and apparent request; not the same duplicate pocket." + }, + { + "left": "issue:43834", + "right": "issue:46226", + "accept": false, + "reason": "Different title and apparent request; not a duplicate." + }, + { + "left": "issue:43834", + "right": "issue:46931", + "accept": false, + "reason": "Different Thai title and no clear shared underlying request." + }, + { + "left": "issue:44365", + "right": "issue:44370", + "accept": true, + "reason": "Near-identical template-style i18n issue; same underlying request." + }, + { + "left": "issue:44365", + "right": "issue:46222", + "accept": false, + "reason": "Different title/topic; similarity alone is not enough." + }, + { + "left": "issue:44365", + "right": "issue:46226", + "accept": false, + "reason": "Different title/topic; not the same issue." + }, + { + "left": "issue:44365", + "right": "issue:46931", + "accept": false, + "reason": "Different apparent request and title." + }, + { + "left": "issue:44370", + "right": "issue:46222", + "accept": false, + "reason": "Different title/topic; not the same duplicate pocket." + }, + { + "left": "issue:44370", + "right": "issue:46226", + "accept": false, + "reason": "Different title/topic; not a duplicate." + }, + { + "left": "issue:44370", + "right": "issue:46931", + "accept": false, + "reason": "Different apparent request." + }, + { + "left": "issue:46222", + "right": "issue:46226", + "accept": true, + "reason": "Identical Thai-titled issue with effectively identical content." + }, + { + "left": "issue:46222", + "right": "issue:46931", + "accept": false, + "reason": "Similar-looking body text is not enough; titles and apparent requests differ." + }, + { + "left": "issue:46226", + "right": "issue:46931", + "accept": false, + "reason": "Different issue title and no clear same underlying request." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The broad split between the language-specific translation issues and the placeholder-like issues is grounded, but the writeup is a bit too assertive in places. In particular, the packet supports at least two duplicate-looking subgroups: the [i18n-] trio and the identical Thai-titled pair 46222/46226. So saying 43834 is the anchor for the \"only clearly duplicate subset\" is not accurate, and \"exact duplicate\"/\"spam\" language goes beyond what the packet alone proves. A more conservative summary would note multiple near-duplicate pockets without claiming exactness." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 6, + "component_count": 23, + "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-36718-19", + "nodes": [ + "issue:36718", + "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", + "issue:46645", + "issue:46653" + ], + "soft_pairs": [ + "issue:36718|issue:41115", + "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:42738|issue:46653", + "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:43837|issue:44121", + "issue:43909|issue:44205", + "issue:43909|issue:46645", + "issue:44121|issue:45322", + "issue:43837|issue:45027", + "issue:37015|issue:45027", + "issue:43837|issue:45177", + "issue:45322|issue:46133", + "issue:42503|issue:46653", + "issue:45027|issue:46653" + ], + "prepared_review_unit_hash": "2ff544196ca0abe0f75a64aaf8f1bc340de03b75454ebdae721f34574ca7fd6d", + "component_budget": { + "node_count": 19, + "item_count": 19, + "soft_pair_count": 40, + "serialized_chars": 24674, + "estimated_input_tokens": 6169, + "estimated_eval_tokens": 12594 + }, + "budget": { + "node_count": 19, + "item_count": 19, + "soft_pair_count": 40, + "serialized_chars": 24674, + "estimated_input_tokens": 6169, + "estimated_eval_tokens": 12594 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This is a loose topic cluster of unrelated model-addition/support requests across text, vision, audio, and architectural changes. The soft links are mostly title-similarity false positives; none of the paired issues look like true duplicates.", + "confidence": 0.88, + "canonical_issue_reason": "Issue 43671 is the closest thing to a cluster center because it is a concrete, well-discussed support request with several adjacent audio/model soft matches. That said, the set is heterogeneous and there is no true duplicate canonical issue.", + "canonical_pr_reason": null, + "best_issue_reason": "43671 is the best representative of the overall pattern of 'add support for a specific model' requests. It has active discussion and is concrete enough to stand in for the cluster, even though the cluster itself is not a duplicate set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36718", + "right": "issue:41115", + "accept": false, + "reason": "RoMa keypoint matcher and MiniCPM3 architecture are different vision/model requests with no shared concrete code path." + }, + { + "left": "issue:36817", + "right": "issue:41211", + "accept": false, + "reason": "EuroBert config and DEIMv2 are unrelated model additions." + }, + { + "left": "issue:36817", + "right": "issue:43671", + "accept": false, + "reason": "EuroBert config is a text-model config request; Qwen3-TTS is an audio support request." + }, + { + "left": "issue:36817", + "right": "issue:36879", + "accept": false, + "reason": "EuroBert config and RF-DETR target different model families and changes." + }, + { + "left": "issue:36879", + "right": "issue:41211", + "accept": false, + "reason": "RF-DETR and DEIMv2 are distinct detector/model requests." + }, + { + "left": "issue:36879", + "right": "issue:43909", + "accept": false, + "reason": "RF-DETR is a vision detector request; LFM2.5 Audio 1.5B is an audio model request." + }, + { + "left": "issue:36879", + "right": "issue:44121", + "accept": false, + "reason": "RF-DETR model support and weight-sparse transformer support are unrelated changes." + }, + { + "left": "issue:36879", + "right": "issue:43837", + "accept": false, + "reason": "RF-DETR and Qwen3-ASR are different domains and code paths." + }, + { + "left": "issue:36879", + "right": "issue:43671", + "accept": false, + "reason": "RF-DETR vision support and Qwen3-TTS audio support are unrelated requests." + }, + { + "left": "issue:37015", + "right": "issue:43837", + "accept": false, + "reason": "NeoBERT and Qwen3-ASR are unrelated model requests." + }, + { + "left": "issue:37015", + "right": "issue:44121", + "accept": false, + "reason": "NeoBERT addition and weight-sparse transformer support are different changes." + }, + { + "left": "issue:37015", + "right": "issue:43909", + "accept": false, + "reason": "NeoBERT is a text model request; LFM2.5 Audio is an audio model request." + }, + { + "left": "issue:36879", + "right": "issue:37015", + "accept": false, + "reason": "RF-DETR and NeoBERT are unrelated model additions." + }, + { + "left": "issue:37015", + "right": "issue:42503", + "accept": false, + "reason": "NeoBERT and ModernVBERT are different BERT-like model requests, not the same change." + }, + { + "left": "issue:41115", + "right": "issue:43837", + "accept": false, + "reason": "MiniCPM3 architecture and Qwen3-ASR support are different model domains." + }, + { + "left": "issue:41115", + "right": "issue:44121", + "accept": false, + "reason": "MiniCPM3 support and weight-sparse transformer support are distinct requests." + }, + { + "left": "issue:41115", + "right": "issue:42503", + "accept": false, + "reason": "MiniCPM3 and ModernVBERT are different architecture/model requests." + }, + { + "left": "issue:41115", + "right": "issue:45027", + "accept": false, + "reason": "MiniCPM3 and Voxtral TTS target different model families and modalities." + }, + { + "left": "issue:41211", + "right": "issue:43671", + "accept": false, + "reason": "DEIMv2 and Qwen3-TTS are unrelated model-support requests." + }, + { + "left": "issue:42503", + "right": "issue:43837", + "accept": false, + "reason": "ModernVBERT models and Qwen3-ASR are unrelated." + }, + { + "left": "issue:42503", + "right": "issue:43671", + "accept": false, + "reason": "ModernVBERT and Qwen3-TTS are different model families/domains." + }, + { + "left": "issue:42503", + "right": "issue:42738", + "accept": false, + "reason": "Both mention BERT-like models, but one is a specific model set request and the other is a broad RoPE architecture request; not the same concrete change." + }, + { + "left": "issue:42738", + "right": "issue:45027", + "accept": false, + "reason": "BERT-like models with RoPE and Voxtral TTS are unrelated requests." + }, + { + "left": "issue:42738", + "right": "issue:46653", + "accept": false, + "reason": "BERT-like model architecture request and the 'Mirko Privitera' issue are unrelated." + }, + { + "left": "issue:43671", + "right": "issue:43837", + "accept": false, + "reason": "Qwen3-TTS and Qwen3-ASR are different concrete support requests, not one mergeable fix." + }, + { + "left": "issue:43671", + "right": "issue:45027", + "accept": false, + "reason": "Both are TTS-related, but they are different models and likely different implementations." + }, + { + "left": "issue:43671", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-TTS and DeepSeek-OCR2 are different modalities and requests." + }, + { + "left": "issue:43671", + "right": "issue:44121", + "accept": false, + "reason": "Qwen3-TTS support and weight-sparse transformer support are unrelated." + }, + { + "left": "issue:43837", + "right": "issue:43909", + "accept": false, + "reason": "Qwen3-ASR and LFM2.5 Audio are both speech-related, but they are distinct model requests." + }, + { + "left": "issue:43837", + "right": "issue:45322", + "accept": false, + "reason": "Qwen3-ASR and EUPE are different model families with no shared change." + }, + { + "left": "issue:43837", + "right": "issue:44121", + "accept": false, + "reason": "Qwen3-ASR and weight-sparse transformer are unrelated changes." + }, + { + "left": "issue:43909", + "right": "issue:44205", + "accept": false, + "reason": "LFM2.5 Audio and SAM3-LiteText are different model-support requests." + }, + { + "left": "issue:43909", + "right": "issue:46645", + "accept": false, + "reason": "Audio model support and Granite SWA+sink option are different code-path changes." + }, + { + "left": "issue:44121", + "right": "issue:45322", + "accept": false, + "reason": "Weight-sparse transformer support and EUPE are unrelated model requests." + }, + { + "left": "issue:43837", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-ASR and Voxtral TTS are both speech-related, but they are not the same model or fix." + }, + { + "left": "issue:37015", + "right": "issue:45027", + "accept": false, + "reason": "NeoBERT and Voxtral TTS are unrelated model requests." + }, + { + "left": "issue:43837", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-ASR and DeepSeek-OCR2 target different modalities." + }, + { + "left": "issue:45322", + "right": "issue:46133", + "accept": false, + "reason": "EUPE and TIPSv2 are different vision-language models and different additions." + }, + { + "left": "issue:42503", + "right": "issue:46653", + "accept": false, + "reason": "ModernVBERT models and a non-technical person-name issue are unrelated." + }, + { + "left": "issue:45027", + "right": "issue:46653", + "accept": false, + "reason": "Voxtral TTS support and a non-model person-name 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": 7, + "component_count": 23, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-38740-2", + "nodes": [ + "issue:38740", + "issue:46443" + ], + "soft_pairs": [ + "issue:38740|issue:46443" + ], + "prepared_review_unit_hash": "b6982293cca2028da48ee209d8117de862593459a8daa3c7e71b41a90a566fef", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1506, + "estimated_input_tokens": 377, + "estimated_eval_tokens": 1010 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1506, + "estimated_input_tokens": 377, + "estimated_eval_tokens": 1010 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are not duplicates: one is a docs request to add `pruna` as an optimization framework, the other is a Vision RoPE refactor. They target different areas and different kinds of work.", + "confidence": 0.99, + "canonical_issue_reason": "Issue 46443 is the better representative only in the sense that it is the more concrete engineering task; issue 38740 is a separate docs-oriented request. There is no shared underlying bug/change.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46443 is the stronger standalone issue for clustering purposes because it describes a specific code refactor, while 38740 is an unrelated documentation enhancement.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:38740", + "right": "issue:46443", + "accept": false, + "reason": "Different topics and work types: docs addition for `pruna` vs Vision RoPE refactor. No shared code path or underlying change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 8, + "component_count": 23, + "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": "61f45c00cabeb3d0f26d98fc7ddfa1b6e16346699f6cad2ee807dd76d8cc1afe", + "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 unrelated: one is about block decoding behavior, the other is a feature request for DeepSeek v3.2 support. They do not describe the same underlying bug or change.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41084", + "right": "issue:43037", + "accept": false, + "reason": "Different topics and likely different workstreams; no shared concrete bug, code path, or feature target." + } + ] + }, + "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": 23, + "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-41306-2", + "nodes": [ + "issue:41306", + "issue:41826" + ], + "soft_pairs": [ + "issue:41306|issue:41826" + ], + "prepared_review_unit_hash": "cb986bf3e46a9b7f18d5effcc5fda71527082b42a0b92553b515896f1c592004", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1585, + "estimated_input_tokens": 397, + "estimated_eval_tokens": 1050 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1585, + "estimated_input_tokens": 397, + "estimated_eval_tokens": 1050 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are about different features: one requests a `from_pretrained()` flag for `TiledFusedLogitsLoss`/FLCE, while the other asks to integrate `TiledMLP` for lower memory usage. They share the word \u201cTiled\u201d but not the same bug or change request.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Issue 41306 is the more concrete, narrowly scoped request and is the better representative of this cluster if one must be chosen, but it is not a duplicate of 41826.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41306", + "right": "issue:41826", + "accept": false, + "reason": "Different underlying changes: FLCE `from_pretrained()` flag support versus adding TiledMLP integration for memory reduction. No shared code path or same fix target 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": 10, + "component_count": 23, + "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": "b7801ea712b835a4d707614157fea621cee7378057f44167475e88160db76ca3", + "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 are in the same Trainer/customization area but 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": 0.9, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "No clear canonical issue here; if forced, 43598 is the more concrete code-path request, but it does not subsume 43600.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43598", + "right": "issue:43600", + "accept": false, + "reason": "They share a broad Trainer/custom trainer theme, but the underlying bugs are different: customizability of `remove_unused_column` vs FLOPS counting for custom trainers. They would not plausibly be one issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 11, + "component_count": 23, + "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": "591f232131158e896b60c92bb058044a2dcabcdaa8e5342c8319f433a0947d7e", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1612, + "estimated_input_tokens": 403, + "estimated_eval_tokens": 1062 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1612, + "estimated_input_tokens": 403, + "estimated_eval_tokens": 1062 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two are related only at a broad theme level (Qwen model support), but they target different models and different AutoModel heads, so they are not duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "Issue #43684 is the broader registration problem and is the better anchor for this cluster, but it still does not subsume #44405 because the latter is a separate Qwen3.5 sequence-classification support request.", + "canonical_pr_reason": null, + "best_issue_reason": "#43684 is the closest thing to a canonical item because it addresses model registration at the AutoModel/AutoModelForConditionalGeneration level, which is a more general code-path than the specific head support asked for in #44405.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43684", + "right": "issue:44405", + "accept": false, + "reason": "Different underlying changes: #43684 asks for Qwen3-Omni registration in AutoModel and AutoModelForConditionalGeneration, while #44405 asks for AutoModelForSequenceClassification support for Qwen3.5/Qwen3_5Config. Same ecosystem, but not the same bug or mergeable fix." + } + ] + }, + "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": 23, + "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": "1d7cb5e0889869cd1d78a2057de4aceaebe321eca79e572d030f992542eae1a0", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4262, + "estimated_input_tokens": 1066, + "estimated_eval_tokens": 2388 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4262, + "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 is mostly low-information issue spam/duplicates with two title groups: one unrelated \"Fett\" issue and three near-identical \"huggingface\" issues. There isn\u2019t enough concrete content to safely merge any pair as the same underlying bug.", + "confidence": 0.42, + "canonical_issue_reason": "If this cluster were deduped, issue 43845 is the best representative because it is the earliest of the repeated \"huggingface\" reports and appears to be the central duplicate target among the similar titles.", + "canonical_pr_reason": null, + "best_issue_reason": "43845 is the best overall issue to keep as the representative because it is the earliest repeated \"huggingface\" entry and has the strongest similarity to the other repeated title issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43835", + "right": "issue:43845", + "accept": false, + "reason": "Titles are unrelated and there is no body/detail evidence tying \"Fett\" to the \"huggingface\" issues." + }, + { + "left": "issue:43835", + "right": "issue:43846", + "accept": false, + "reason": "Titles are unrelated and there is no body/detail evidence tying \"Fett\" to the \"huggingface\" issues." + }, + { + "left": "issue:43835", + "right": "issue:43859", + "accept": false, + "reason": "Titles are unrelated and there is no body/detail evidence tying \"Fett\" to the \"huggingface\" issues." + }, + { + "left": "issue:43845", + "right": "issue:43846", + "accept": false, + "reason": "The titles match closely, but the bodies are not available and there is not enough concrete evidence to confirm the same underlying bug." + }, + { + "left": "issue:43845", + "right": "issue:43859", + "accept": false, + "reason": "The titles match closely, but the bodies are not available and there is not enough concrete evidence to confirm the same underlying bug." + }, + { + "left": "issue:43846", + "right": "issue:43859", + "accept": false, + "reason": "The titles match closely, but the bodies are not available and there is not enough concrete evidence to confirm 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": 13, + "component_count": 23, + "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": "8b2cfcc8eb592d2c3d219eaf806910440f3ad3993982b0e26ebc3ca6513be7a7", + "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 looks like low-signal noise rather than one real duplicated bug/change set: three generic spam issues, one documentation-template issue, and one unrelated promotional title. No pair has concrete evidence of the same underlying problem.", + "confidence": 0.91, + "canonical_issue_reason": "issue:44371 is the cleanest representative of the cluster because it is the earliest and most generic \"\" entry, with no extra wording tying it to a different topic.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44371 is the best global issue representative because it is the most neutral placeholder/noise item and does not introduce a separate topic like documentation or promotion.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44371", + "right": "issue:44572", + "accept": false, + "reason": "Both are spam-like placeholders, but there is no concrete bug report or shared code-path/change to justify treating them as the same underlying issue." + }, + { + "left": "issue:44371", + "right": "issue:44573", + "accept": false, + "reason": "Same generic spam title is not enough; these could be separate spam submissions and there is no substantive issue content to merge." + }, + { + "left": "issue:44371", + "right": "issue:44418", + "accept": false, + "reason": "\"\" and a documentation enhancement suggestion are clearly different topics with no evidence of a shared bug or fix." + }, + { + "left": "issue:44371", + "right": "issue:44937", + "accept": false, + "reason": "A spam placeholder and a promotional/title-like issue do not indicate the same underlying problem." + }, + { + "left": "issue:44418", + "right": "issue:44572", + "accept": false, + "reason": "Documentation enhancement versus spam placeholder: unrelated and not the same underlying change request." + }, + { + "left": "issue:44418", + "right": "issue:44573", + "accept": false, + "reason": "Documentation enhancement versus spam placeholder: no shared bug, feature, or target area." + }, + { + "left": "issue:44572", + "right": "issue:44573", + "accept": false, + "reason": "Although the titles match, there is no actual issue detail showing they describe the same bug or change; they are just duplicate spam-like submissions." + }, + { + "left": "issue:44572", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder and unrelated promotional title do not describe a common underlying issue." + }, + { + "left": "issue:44573", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder and unrelated promotional title are not the same concrete problem." + } + ] + }, + "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": 23, + "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": "8befd5f326187f06668bb0b83440dab5fd92458b168429a4030783631d9f03f9", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1646, + "estimated_input_tokens": 412, + "estimated_eval_tokens": 1080 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1646, + "estimated_input_tokens": 412, + "estimated_eval_tokens": 1080 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two closed issues have identical titles and very similar metadata, so #45145 is the best representative and likely duplicate target. However, the packet does not expose the bodies, so the soft match is not strong enough to confirm they are definitively the same issue.", + "confidence": 0.84, + "canonical_issue_reason": "#45145 is earlier and has the same title as #45151, making it the safest canonical duplicate target among the two.", + "canonical_pr_reason": null, + "best_issue_reason": "#45145 is the best issue representative because it is the earliest report and the visible metadata strongly suggests the same topic as #45151.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45145", + "right": "issue:45151", + "accept": false, + "reason": "The evidence only shows identical titles, similar body lengths, and a soft-similarity score. Without the issue bodies or clearer scope/goal evidence, this should be treated as a likely duplicate rather than confirmed as the same underlying issue." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The canonical/representative choice is grounded by chronology and identical titles, but the soft-edge acceptance overstates the evidence. The packet only shows identical titles, close body lengths, and a soft-similarity signal; it does not expose the issue bodies or confirm that the scope/goal is the same. So this should be phrased as a likely duplicate or potential duplicate, not accepted as definitively the same issue." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 15, + "component_count": 23, + "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": 16, + "component_count": 23, + "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": "d645e636d0c8d984fc2a68d0d9e90d3df7c4d510d989162153a46e2ae228ce6d", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two empty, contentless issues that look like separate spam/placeholder posts, not duplicates of a real bug report or change request.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a good canonical candidate: both have no body, no references, and no actionable problem description.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45762", + "right": "issue:45763", + "accept": false, + "reason": "The only similarity is the surname pattern in the titles; there is no shared bug, repro, subsystem, or issue target indicating the same underlying problem." + } + ] + }, + "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": 23, + "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": "68c550abe73c04bea659dc3dbac54bac9d8aec6808af5c7b777d79d992f61a18", + "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 closed issues are just title-only personal-name/date entries with empty bodies and no shared context, so they do not appear to describe the same underlying bug or change.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "No strong canonical issue exists here; if forced to pick one anchor, issue 45764 is only marginally earlier, but the pair does not look duplicate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45764", + "right": "issue:45765", + "accept": false, + "reason": "The titles are unrelated personal names with dates, both bodies are empty, and there is no shared target or code-context suggesting the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 18, + "component_count": 23, + "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": "2acacd0df53ef27bc1f27237e705a628803a680b9150c2e34bbb6a509101da4b", + "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 unrelated: one requests a new token-classification/value-model class for Qwen3_5, while the other asks for FP8 training support. No duplicate relationship is present.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a duplicate of the other; if one had to be prioritized as a standalone triage item, 46295 is the broader infrastructure request, while 45810 is a specific model-class feature request.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45810", + "right": "issue:46295", + "accept": false, + "reason": "Different underlying requests and code paths: model-class addition for token classification vs FP8 training support. They do not describe the same bug or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 19, + "component_count": 23, + "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": "27dae74efab0f3b559fccc88d47e8c46b714c83f79be748c0bb702378e6b048e", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5716, + "estimated_input_tokens": 1429, + "estimated_eval_tokens": 3114 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5716, + "estimated_input_tokens": 1429, + "estimated_eval_tokens": 3114 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two exact duplicate issue pairs stand out: 45870\u219445871 and 45872\u219445873. Issue 45869 is a separate dataset listing and should not be merged with the example/service requests.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue for the whole set; there are two distinct duplicate groups, while 45869 is unrelated.", + "canonical_pr_reason": null, + "best_issue_reason": "45872 is the strongest representative because it is specific and actionable, and it has an exact duplicate in 45873.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45869", + "right": "issue:45870", + "accept": false, + "reason": "Different requests: a financial dataset bundle vs an example about time series prediction." + }, + { + "left": "issue:45869", + "right": "issue:45871", + "accept": false, + "reason": "Different requests: a financial dataset bundle vs an example about time series prediction." + }, + { + "left": "issue:45870", + "right": "issue:45871", + "accept": true, + "reason": "Exact same title and clearly the same example request." + }, + { + "left": "issue:45870", + "right": "issue:45872", + "accept": false, + "reason": "Both are finance-related, but one is an example request and the other is a time-series plus scraping-service request." + }, + { + "left": "issue:45870", + "right": "issue:45873", + "accept": false, + "reason": "Both are finance-related, but one is an example request and the other is a time-series plus scraping-service request." + }, + { + "left": "issue:45871", + "right": "issue:45872", + "accept": false, + "reason": "Both are finance-related, but one is an example request and the other is a time-series plus scraping-service request." + }, + { + "left": "issue:45871", + "right": "issue:45873", + "accept": false, + "reason": "Both are finance-related, but one is an example request and the other is a time-series plus scraping-service request." + }, + { + "left": "issue:45872", + "right": "issue:45873", + "accept": true, + "reason": "Exact same title and clearly the same feature request." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet: 45870\u219445871 and 45872\u219445873 are supported as duplicate pairs by identical titles/body-length matches and perfect similarity, and 45869 is clearly a separate dataset-style issue. The soft-edge verdicts are also conservative, with no overreach beyond the evidence provided." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 20, + "component_count": 23, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46439-2", + "nodes": [ + "issue:46439", + "issue:46441" + ], + "soft_pairs": [ + "issue:46439|issue:46441" + ], + "prepared_review_unit_hash": "90fa1c913ccc2ee65b31dbb6f489ea7cd497fb916cef1df20b0d003fc2f5d7b8", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1637, + "estimated_input_tokens": 410, + "estimated_eval_tokens": 1076 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1637, + "estimated_input_tokens": 410, + "estimated_eval_tokens": 1076 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both issues involve `cache_implementation=\"static\"` and `linear_attention`, but they describe different failure modes: one is a shape corruption / `RuntimeError` during early initialization, the other is a `KeyError: 'linear_attention'` in Qwen3.5. I would not merge them as duplicates without stronger evidence of the same code path.", + "confidence": 0.78, + "canonical_issue_reason": "Issue 46439 is the better canonical issue because it describes the broader, more generic failure mode around static-cache early initialization affecting linear-attention layers, which can subsume narrower model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46439 is the best representative of the cluster because it states the underlying bug at the subsystem level rather than a single-model symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46439", + "right": "issue:46441", + "accept": false, + "reason": "Related area, but not clearly the same bug: one report is a shape mismatch/RuntimeError from early initialization, the other is a KeyError for `linear_attention` in Qwen3.5." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 21, + "component_count": 23, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46551-2", + "nodes": [ + "issue:46551", + "issue:46552" + ], + "soft_pairs": [ + "issue:46551|issue:46552" + ], + "prepared_review_unit_hash": "c5346631926601bd1aa98b04cf68fb38b7d5c538f9e4f2bc7c5731d2d1531f31", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1707, + "estimated_input_tokens": 427, + "estimated_eval_tokens": 1110 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1707, + "estimated_input_tokens": 427, + "estimated_eval_tokens": 1110 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are effectively the same report: identical title and body about Wav2Vec2CTCTokenizer v5 incorrectly treating the word delimiter as a special token. The open issue is the better canonical representative.", + "confidence": 0.99, + "canonical_issue_reason": "Issue #46552 is the better canonical issue because it is open and already has inbound references, while #46551 is closed and appears to be the earlier duplicate copy.", + "canonical_pr_reason": null, + "best_issue_reason": "#46552 is the best issue to keep because it is active, open, and has the stronger likelihood of being the one that should collect discussion and fixes.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46551", + "right": "issue:46552", + "accept": true, + "reason": "Same title, same body, and same underlying tokenizer bug; this is a clear duplicate pair." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet: the two issues have identical titles and body lengths, and the open issue has inbound references, so treating #46552 as the canonical representative is reasonable. The soft-edge duplicate verdict is conservative and supported by the evidence available." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 22, + "component_count": 23, + "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-46657-2", + "nodes": [ + "issue:46657", + "issue:46692" + ], + "soft_pairs": [ + "issue:46657|issue:46692" + ], + "prepared_review_unit_hash": "e3f96b716482df0cb0cd54cfb8100cad1a436cefef971ef37219c7e3563944b7", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1569, + "estimated_input_tokens": 393, + "estimated_eval_tokens": 1042 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1569, + "estimated_input_tokens": 393, + "estimated_eval_tokens": 1042 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are both about training object detection models, but they describe different failure modes: one is a tensor shape mismatch, the other is an out-of-memory problem occurring after several epochs in RT-DETR v2. They do not look like the same underlying bug.", + "confidence": 0.93, + "canonical_issue_reason": "Issue #46657 is the clearer standalone bug report: it names a specific runtime error (tensor size mismatch) and is more directly actionable than the broader OOM report.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #46657 is the best representative issue for the cluster because it is more concrete and specific, whereas #46692 appears to be a separate training stability/resource problem.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46657", + "right": "issue:46692", + "accept": false, + "reason": "The reports involve different concrete problems and failure signatures: tensor dimension mismatch vs memory exhaustion after training for some epochs. Shared object-detection context is too broad to treat as one bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 23, + "component_count": 23, + "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-46822-2", + "nodes": [ + "issue:46822", + "issue:46823" + ], + "soft_pairs": [ + "issue:46822|issue:46823" + ], + "prepared_review_unit_hash": "24e09325fbea200ff8ee3f472011ce1b45f117d97a8de040b934e330ab82bc77", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1671, + "estimated_input_tokens": 418, + "estimated_eval_tokens": 1092 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1671, + "estimated_input_tokens": 418, + "estimated_eval_tokens": 1092 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues both mention model-loading and memory, but they describe different failure modes in different code paths: ZeRO-3 composite module partitioning vs 4-bit device_map/max_memory dispatch. They should not be clustered as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": "No clear canonical issue for a duplicate cluster here; the reports target different bugs and should remain separate.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46823 is the more self-contained bug report with a clear repro and concrete expected/actual behavior, but it is not a duplicate of 46822.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46822", + "right": "issue:46823", + "accept": false, + "reason": "Different underlying bugs: ZeRO-3 zero.Init failing to partition a composite submodule and causing OOM versus 4-bit device_map='auto' with max_memory incorrectly dispatching modules to CPU and raising ValueError. They do not look mergeable into one 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": 1, + "component_count": 38, + "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": "dc968de77a880f1bc6352046630b4b135c7ea09a4a1e1c76cb94cbe9d0121330", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7223, + "estimated_input_tokens": 1806, + "estimated_eval_tokens": 3868 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7223, + "estimated_input_tokens": 1806, + "estimated_eval_tokens": 3868 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both pull requests are about VibeVoice, but they are not the same change: PR 40546 adds the base VibeVoice model integration, while PR 42781 adds a separate realtime variant plus acoustic tokenizer work and generation changes. They overlap on shared architecture files, but the concrete scope is different enough that they should not be deduplicated.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 40546 is the better canonical PR for the cluster because it is the foundational VibeVoice implementation and broader in scope, while PR 42781 looks like a follow-on feature branch for realtime support.", + "best_issue_reason": null, + "best_pr_reason": "PR 40546 best represents the underlying VibeVoice integration work; PR 42781 is a narrower extension for realtime functionality and tokenizer support, not the same mergeable change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40546", + "right": "pull_request:42781", + "accept": false, + "reason": "Shared VibeVoice files indicate related work, but 40546 is the base implementation and 42781 is a distinct realtime/tokenizer extension with additional generation and auto-mapping changes. They are related, not duplicate fixes or the same concrete 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": 2, + "component_count": 38, + "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": "75b7127fca093b69f4410becb91f66026c512e0bebf9de38974ddb0da194ebff", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4675, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4675, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are unrelated: one adds TimesFM covariate forecasting, the other adds DINOv3 image classification support. They do not fix the same bug or implement the same change, so they should not be clustered as duplicates.", + "confidence": 0.99, + "canonical_issue_reason": null, + "canonical_pr_reason": "No clear canonical PR: the two pull requests target different models and different features, so neither serves as a duplicate representative for the cluster.", + "best_issue_reason": null, + "best_pr_reason": "Neither PR is a good global canonical choice because they address separate code paths and independent feature additions.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40755", + "right": "pull_request:41224", + "accept": false, + "reason": "Different subsystems and different changes: TimesFM forecasting-with-covariates versus DINOv3 image classification support. They do not appear to be the same underlying bug or mergeable change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 3, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-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": 4, + "component_count": 38, + "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-41855-15", + "nodes": [ + "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:46707", + "pull_request:46911" + ], + "soft_pairs": [ + "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:46911", + "pull_request:43424|pull_request:46707", + "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:46911", + "pull_request:44827|pull_request:46707", + "pull_request:44827|pull_request:45170", + "pull_request:44827|pull_request:46267", + "pull_request:45170|pull_request:46707", + "pull_request:44827|pull_request:45213", + "pull_request:43823|pull_request:45213", + "pull_request:44827|pull_request:45655", + "pull_request:46048|pull_request:46707", + "pull_request:44827|pull_request:46048", + "pull_request:43823|pull_request:46707", + "pull_request:43823|pull_request:46911", + "pull_request:43251|pull_request:43254" + ], + "prepared_review_unit_hash": "d4d2dab884c1b13a28137e4aa864ac7c7f8b94fe9941e075c3021a118cfbd1a6", + "component_budget": { + "node_count": 15, + "item_count": 15, + "soft_pair_count": 26, + "serialized_chars": 42250, + "estimated_input_tokens": 10563, + "estimated_eval_tokens": 21382 + }, + "budget": { + "node_count": 15, + "item_count": 15, + "soft_pair_count": 26, + "serialized_chars": 42250, + "estimated_input_tokens": 10563, + "estimated_eval_tokens": 21382 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly noise from loose topical similarity. The only clear duplicate is the pair of `fixed_cross_entropy` PRs; the rest touch unrelated models, docs, infra, or tests and should stay separate.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "`pull_request:43251` is the strongest canonical anchor for the one true duplicate pair: it targets the same `cross_entropy` kwargs bug, has the explicit issue link, and is the more developed change.", + "best_issue_reason": null, + "best_pr_reason": "`pull_request:43251` is the best representative of the real duplicate pair because it directly fixes the `fixed_cross_entropy` kwargs gap and is better supported than `43254`; it is not a good canonical for the whole cluster since the other PRs are unrelated.", + "soft_edge_verdicts": [ + { + "left": "pull_request:41855", + "right": "pull_request:43823", + "accept": false, + "reason": "Different problems: Mistral tokenizer API gaps vs adding a new MobileLLM model." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43424", + "accept": false, + "reason": "Unrelated: tokenizer methods vs executorch dynamic-shape export testing." + }, + { + "left": "pull_request:42668", + "right": "pull_request:43823", + "accept": false, + "reason": "Processor auto-loading robustness is unrelated to adding MobileLLM." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43823", + "accept": false, + "reason": "Cross-entropy kwargs fix and MobileLLM model addition are unrelated." + }, + { + "left": "pull_request:43251", + "right": "pull_request:44827", + "accept": false, + "reason": "Same file path is incidental; the bugs are different (`cross_entropy` kwargs vs Mistral4 test fixes)." + }, + { + "left": "pull_request:43424", + "right": "pull_request:46911", + "accept": false, + "reason": "Executorch export tests do not match Olmo3 RoPE-layer behavior changes." + }, + { + "left": "pull_request:43424", + "right": "pull_request:46707", + "accept": false, + "reason": "Dynamic-shape export testing is unrelated to native FSDP2 migration." + }, + { + "left": "pull_request:43424", + "right": "pull_request:44827", + "accept": false, + "reason": "Executorch export coverage and Mistral4 fixes are different changes." + }, + { + "left": "pull_request:43424", + "right": "pull_request:43823", + "accept": false, + "reason": "Executorch export tests and MobileLLM model addition are unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45655", + "accept": false, + "reason": "Adding MobileLLM is unrelated to config resolution order fixes." + }, + { + "left": "pull_request:43823", + "right": "pull_request:43995", + "accept": false, + "reason": "New MobileLLM model code and Falcon output-interface refactoring are different changes." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45213", + "accept": false, + "reason": "Falcon model refactor is unrelated to the model-creation skill PR." + }, + { + "left": "pull_request:43995", + "right": "pull_request:44827", + "accept": false, + "reason": "Falcon output-capture refactor and Mistral4 test fixes are not the same bug." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46911", + "accept": false, + "reason": "Mistral4 test fixes and Olmo3 per-layer RoPE are unrelated model changes." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46707", + "accept": false, + "reason": "Mistral4 test/debug fixes do not match native FSDP2 migration." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45170", + "accept": false, + "reason": "Mistral4 tests and `layrnorm` rename/conversion mapping are different issues." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46267", + "accept": false, + "reason": "Mistral4 test fixes are unrelated to chat parsing/serve updates." + }, + { + "left": "pull_request:45170", + "right": "pull_request:46707", + "accept": false, + "reason": "Typo/weight-renaming cleanup is unrelated to FSDP2 migration." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45213", + "accept": false, + "reason": "Model test fixes are unrelated to the model-creation skill/meta PR." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45213", + "accept": false, + "reason": "A new model implementation and the model-creation skill are different artifacts." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45655", + "accept": false, + "reason": "Mistral4 tests and config resolution ordering fixes are unrelated." + }, + { + "left": "pull_request:46048", + "right": "pull_request:46707", + "accept": false, + "reason": "TDT loss kernel integration is unrelated to FSDP2 module migration." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46048", + "accept": false, + "reason": "Mistral4 test fixes and TDT loss kernel integration are different changes." + }, + { + "left": "pull_request:43823", + "right": "pull_request:46707", + "accept": false, + "reason": "Adding MobileLLM and FSDP2 migration are unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:46911", + "accept": false, + "reason": "MobileLLM model addition is unrelated to Olmo3 RoPE logic changes." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43254", + "accept": true, + "reason": "Both PRs fix the same `fixed_cross_entropy` kwargs support gap in `loss_utils.py`; they are near-duplicate implementations of the same concrete bug." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The analysis is grounded in the packet and stays conservative on the soft edges. The only accepted duplicate pair, `pull_request:43251` / `pull_request:43254`, is clearly mergeable as the same `fixed_cross_entropy` kwargs bug in the same file and with the same issue target. The rest of the rejected pairs are reasonably treated as topical similarities rather than evidence of the same bug." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 5, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "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": "70767d93f058c1ba4a1fa1f60e6a98f31e469f68b468ac812084043cd49ae74c", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5934, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5934, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs both do GitHub Actions dependency bumps, but they target different actions and different workflow sets. They are related maintenance, not the same concrete change, so they should not be deduplicated.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #42975 is the broader workflow-runtime upgrade: it updates checkout/cache/upload-artifact across many workflows for Node 24 compatibility, making it the more representative maintenance PR in this cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR #42975 is the best overall representative because it addresses the larger cross-workflow GitHub Actions compatibility update, while #42976 is a narrower docker-action version bump.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42975", + "right": "pull_request:42976", + "accept": false, + "reason": "Both are dependency refreshes in CI workflows, but they update different actions (core workflow actions vs docker/build-push-action) and different concrete paths. They could be part of a maintenance sweep, but they are not the same underlying bug/change and should not be merged as duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 6, + "component_count": 38, + "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-42978-6", + "nodes": [ + "pull_request:42978", + "pull_request:43448", + "pull_request:43451", + "pull_request:45490", + "pull_request:46562", + "pull_request:46658" + ], + "soft_pairs": [ + "pull_request:42978|pull_request:43448", + "pull_request:42978|pull_request:43451", + "pull_request:42978|pull_request:45490", + "pull_request:43448|pull_request:43451", + "pull_request:43448|pull_request:45490", + "pull_request:43451|pull_request:45490", + "pull_request:45490|pull_request:46562", + "pull_request:46562|pull_request:46658" + ], + "prepared_review_unit_hash": "47118af5c3c40cb0617d92c5bea787d4b0bfb8d81120f89a6a5257991fd06033", + "component_budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 8, + "serialized_chars": 21198, + "estimated_input_tokens": 5300, + "estimated_eval_tokens": 10856 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 8, + "serialized_chars": 21198, + "estimated_input_tokens": 5300, + "estimated_eval_tokens": 10856 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly a set of unrelated model-addition PRs. The only potentially close pair is Step3p7 Flash vs a WIP Step3 7 Flash conversion/script PR, but their scopes differ enough that they should not be treated as duplicates.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "No true duplicate PR stands out in the cluster; 46562 is the best representative only because it is the core Step3p7 model implementation PR, but it is not a duplicate of the others.", + "best_issue_reason": null, + "best_pr_reason": "46562 is the most substantial and self-contained model implementation PR in the cluster. It has the strongest technical completeness and the clearest model-addition scope.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42978", + "right": "pull_request:43448", + "accept": false, + "reason": "Different model families: ViT NEPA vs Molmo. Shared docs/auto-registry edits are incidental, not the same underlying change." + }, + { + "left": "pull_request:42978", + "right": "pull_request:43451", + "accept": false, + "reason": "Different models: ViT NEPA and Molmo2. No shared bug or code-path problem." + }, + { + "left": "pull_request:42978", + "right": "pull_request:45490", + "accept": false, + "reason": "Completely different additions: ViT NEPA vs CTSM." + }, + { + "left": "pull_request:43448", + "right": "pull_request:43451", + "accept": false, + "reason": "Molmo and Molmo2 are related names, but these are distinct model integrations with different files and behavior." + }, + { + "left": "pull_request:43448", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo vs CTSM are separate model additions, not the same bug or feature." + }, + { + "left": "pull_request:43451", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo2 and CTSM are unrelated model PRs despite both touching auto-mapping and docs." + }, + { + "left": "pull_request:45490", + "right": "pull_request:46562", + "accept": false, + "reason": "CTSM and Step3p7 Flash are unrelated model implementations; overlap in registry/docs does not make them duplicates." + }, + { + "left": "pull_request:46562", + "right": "pull_request:46658", + "accept": false, + "reason": "Both reference Step 3.7 Flash, but 46562 is the actual Transformers model implementation while 46658 is a WIP script/mini-model conversion artifact set. They are related support work, not the same mergeable code-path change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 7, + "component_count": 38, + "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-43291-7", + "nodes": [ + "pull_request:43291", + "pull_request:43665", + "pull_request:44815", + "pull_request:46380", + "pull_request:46539", + "pull_request:46886", + "pull_request:46922" + ], + "soft_pairs": [ + "pull_request:43291|pull_request:43665", + "pull_request:43291|pull_request:44815", + "pull_request:43291|pull_request:46380", + "pull_request:43291|pull_request:46922", + "pull_request:43665|pull_request:46539", + "pull_request:43665|pull_request:44815", + "pull_request:43665|pull_request:46380", + "pull_request:43665|pull_request:46922", + "pull_request:44815|pull_request:46539", + "pull_request:44815|pull_request:46380", + "pull_request:44815|pull_request:46922", + "pull_request:46380|pull_request:46539", + "pull_request:46380|pull_request:46922", + "pull_request:46539|pull_request:46922", + "pull_request:46886|pull_request:46922" + ], + "prepared_review_unit_hash": "24a8026c812ab78609bab2024b12888fce515a40edc2f22f588195e11478e502", + "component_budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 15, + "serialized_chars": 19595, + "estimated_input_tokens": 4899, + "estimated_eval_tokens": 10054 + }, + "budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 15, + "serialized_chars": 19595, + "estimated_input_tokens": 4899, + "estimated_eval_tokens": 10054 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Most items are unrelated PRs that only share vague titles or the word \u201cfix.\u201d The only strong duplicate signal is the pair of YOLOS PRs, which both target the same device_map=\"auto\" mismatch in the same model file.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46886 is the clearest anchor: it has the specific YOLOS device_map=\"auto\" mismatch title, an explicit issue target, and a focused one-file fix.", + "best_issue_reason": null, + "best_pr_reason": "PR 46886 is the best representative of the cluster because it most clearly describes the concrete bug and the intended fix scope; the other PRs are either unrelated or much less specific.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43291", + "right": "pull_request:43665", + "accept": false, + "reason": "Both are generic test-fix PRs, but they touch different models and different test paths; not the same underlying bug." + }, + { + "left": "pull_request:43291", + "right": "pull_request:44815", + "accept": false, + "reason": "Whisper tokenizer/test changes and FP8 dequant changes are unrelated code paths." + }, + { + "left": "pull_request:43291", + "right": "pull_request:46380", + "accept": false, + "reason": "Whisper test cleanup is unrelated to generation-config watermarking handling and model-specific generation tests." + }, + { + "left": "pull_request:43291", + "right": "pull_request:46922", + "accept": false, + "reason": "Whisper test/tokenizer work and YOLOS device mismatch fixes are different bugs in different subsystems." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46539", + "accept": false, + "reason": "These are broad test edits in different areas (CLIP/ViT vs Phimoe integration); no shared concrete bug." + }, + { + "left": "pull_request:43665", + "right": "pull_request:44815", + "accept": false, + "reason": "CLIP/ViT test edits do not match the FP8 dequant/model-loading change." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46380", + "accept": false, + "reason": "Model test tweaks are unrelated to watermarking_config validation and generation behavior." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46922", + "accept": false, + "reason": "Generic model test fixes are unrelated to the YOLOS device placement bug." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46539", + "accept": false, + "reason": "FP8 dequant/core loading changes are unrelated to the K8S/Phimoe test environment fix." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46380", + "accept": false, + "reason": "Different concrete changes: FP8 kernel/padding logic versus generation watermarking validation." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46922", + "accept": false, + "reason": "FP8 dequant fixes and YOLOS device mismatch fixes are distinct bugs." + }, + { + "left": "pull_request:46380", + "right": "pull_request:46539", + "accept": false, + "reason": "Generation watermarking validation and Phimoe integration-test environment adjustments do not fix the same issue." + }, + { + "left": "pull_request:46380", + "right": "pull_request:46922", + "accept": false, + "reason": "Unrelated: generation config validation versus YOLOS device placement." + }, + { + "left": "pull_request:46539", + "right": "pull_request:46922", + "accept": false, + "reason": "CI/Phimoe test infrastructure changes are unrelated to the YOLOS model bug." + }, + { + "left": "pull_request:46886", + "right": "pull_request:46922", + "accept": true, + "reason": "Both PRs target the same YOLOS device_map=\"auto\" device-mismatch problem in the same file; the fixes are alternative pieces of the same underlying bug and could plausibly be merged together." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only clearly mergeable same-bug pair is the YOLOS duo, which is supported by the shared issue target and shared file. The other rejects are reasonable given the distinct subsystems and fixes shown in the packet." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 8, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43488-3", + "nodes": [ + "pull_request:43488", + "pull_request:43532", + "pull_request:46747" + ], + "soft_pairs": [ + "pull_request:43488|pull_request:46747", + "pull_request:43488|pull_request:43532", + "pull_request:43532|pull_request:46747" + ], + "prepared_review_unit_hash": "12879f44029a5a64348ccf622283775db7b9840b82256359fe129c84cc3e4ec3", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 6350, + "estimated_input_tokens": 1588, + "estimated_eval_tokens": 3432 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 6350, + "estimated_input_tokens": 1588, + "estimated_eval_tokens": 3432 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "All three pull requests are separate \"don't merge\"/test-workflow probes with different code paths and purposes; they are not true duplicates of one another.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46747 is the most focused and self-contained change, touching only a single workflow file, so it's the least misleading representative of the cluster.", + "best_issue_reason": null, + "best_pr_reason": "There is no strong single canonical duplicate here, but 46747 is the cleanest anchor because it is the narrowest change; 43488 and 43532 are broader and less representative.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43488", + "right": "pull_request:46747", + "accept": false, + "reason": "Different change sets: 43488 is a setup/version/dependency test PR, while 46747 only flips a workflow trigger. They do not fix the same bug or code path." + }, + { + "left": "pull_request:43488", + "right": "pull_request:43532", + "accept": false, + "reason": "43488 changes packaging/versioning and one test, whereas 43532 edits several workflows plus test utilities. The overlap is superficial, not the same underlying change." + }, + { + "left": "pull_request:43532", + "right": "pull_request:46747", + "accept": false, + "reason": "43532 is a broad CI/workflow test PR with multiple files; 46747 is a narrow security-gate workflow tweak. Different intent and implementation." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 9, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "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": "dd3b83a3d2c95699f853bae34a3c72da6d2eafbaaf1122401ae37d8cfb7a3af3", + "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 two separate feature PRs: one adds a Promptable Concept Segmentation pipeline, the other adds a Promptable Visual Segmentation pipeline. They share documentation and pipeline scaffolding, but the underlying tasks, code paths, and test/model coverage differ, so they are not duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR: the two PRs implement different segmentation pipelines rather than the same change.", + "best_issue_reason": null, + "best_pr_reason": "PR #43613 is the broader and more representative feature addition, with wider model/test coverage and more files touched.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43612", + "right": "pull_request:43613", + "accept": false, + "reason": "Reject: same broad segmentation area, but different concrete pipelines and model integrations (concept vs visual segmentation). Shared docs/scaffolding alone is not enough to treat them as one 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": 10, + "component_count": 38, + "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": "28bfb5bd1b1e001ea18974f98f42d1c6844a89f7bf2bf249e1ad49c712247198", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4293, + "estimated_input_tokens": 1074, + "estimated_eval_tokens": 2404 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4293, + "estimated_input_tokens": 1074, + "estimated_eval_tokens": 2404 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open trainer.py PRs, but they address different trainer extension points: gradient-accumulation loss scaling vs. signature-column filtering. They are not duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR: 43651 changes loss scaling behavior in training_step, while 43663 adds a separate helper for signature-column filtering in trainer input preparation.", + "best_issue_reason": null, + "best_pr_reason": "43651 is the more concrete bug-fix-style PR and is the better anchor for the cluster, but it does not subsume the unrelated column-filtering refactor in 43663.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43651", + "right": "pull_request:43663", + "accept": false, + "reason": "Different underlying trainer changes: 43651 adjusts gradient-accumulation loss scaling, while 43663 refactors column filtering for model signatures. They do not fix the same code-path problem and would not plausibly merge into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 11, + "component_count": 38, + "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": "e4bde49293e53bf41234504cd19cf64502a41ec271ee8af61738ef55fbc1c51f", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 3664, + "estimated_input_tokens": 916, + "estimated_eval_tokens": 2088 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 3664, + "estimated_input_tokens": 916, + "estimated_eval_tokens": 2088 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs target runtime annotation/import problems in `src/transformers/cli/serve.py`, but they use different fixes and do not clearly resolve the exact same bug. They are related, yet the evidence is not strong enough to treat them as one mergeable change.", + "confidence": 0.73, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #43656 is the broader representative of the cluster because it applies a file-wide annotation/runtime-safety fix in `serve.py`, whereas PR #43842 is a narrower `TypeAdapter` typing workaround.", + "best_issue_reason": null, + "best_pr_reason": "PR #43656 best represents the cluster\u2019s main theme: fixing annotation-related runtime failures in the CLI. It is the more general change and is less dependent on a specific import workaround.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43656", + "right": "pull_request:43842", + "accept": false, + "reason": "Related but not the same underlying fix: one uses `from __future__ import annotations`, the other uses `TYPE_CHECKING` plus a stringized `TypeAdapter` annotation. That is similar in area, but not enough to treat them as one mergeable bugfix." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The packet only supports a conservative similarity judgment: both PRs touch `src/transformers/cli/serve.py` and are about annotation/runtime import issues, but they use different fix strategies (`from __future__ import annotations` vs. `TYPE_CHECKING` import plus stringized annotation). That is enough to say they are related, not enough to confidently say they are the same underlying `TypeAdapter` NameError or that they are a single mergeable bugfix. The soft-pair acceptance is too strong, and the summary slightly overstates the evidence." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 12, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-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": 13, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 2, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43996-28", + "nodes": [ + "pull_request:43996", + "pull_request:43997", + "pull_request:44007", + "pull_request:44010", + "pull_request:44013", + "pull_request:44017", + "pull_request:44018", + "pull_request:44019", + "pull_request:44024", + "pull_request:44044", + "pull_request:44056", + "pull_request:44059", + "pull_request:44066", + "pull_request:44068", + "pull_request:44071", + "pull_request:44072", + "pull_request:44073", + "pull_request:44074", + "pull_request:44076", + "pull_request:44085", + "pull_request:44086", + "pull_request:44098", + "pull_request:44101", + "pull_request:44116", + "pull_request:44129", + "pull_request:44154", + "pull_request:44161", + "pull_request:44722" + ], + "soft_pairs": [ + "pull_request:43996|pull_request:44085", + "pull_request:43996|pull_request:44044", + "pull_request:44066|pull_request:44085", + "pull_request:44007|pull_request:44072", + "pull_request:44072|pull_request:44722", + "pull_request:44066|pull_request:44072", + "pull_request:44013|pull_request:44044", + "pull_request:44066|pull_request:44086", + "pull_request:44018|pull_request:44068", + "pull_request:44066|pull_request:44071", + "pull_request:44066|pull_request:44068", + "pull_request:44018|pull_request:44066", + "pull_request:44019|pull_request:44722", + "pull_request:44086|pull_request:44722", + "pull_request:44019|pull_request:44085", + "pull_request:44018|pull_request:44019", + "pull_request:44019|pull_request:44071", + "pull_request:44068|pull_request:44722", + "pull_request:44068|pull_request:44116", + "pull_request:44018|pull_request:44116", + "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:44085|pull_request:44086", + "pull_request:44059|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:44161", + "pull_request:44017|pull_request:44722", + "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:44073|pull_request:44161", + "pull_request:44059|pull_request:44071", + "pull_request:44056|pull_request:44101", + "pull_request:44018|pull_request:44059", + "pull_request:44059|pull_request:44068", + "pull_request:44101|pull_request:44161", + "pull_request:43997|pull_request:44017", + "pull_request:43997|pull_request:44056", + "pull_request:44071|pull_request:44076", + "pull_request:44074|pull_request:44101", + "pull_request: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:44010|pull_request:44071", + "pull_request:44010|pull_request:44073", + "pull_request:44018|pull_request:44129", + "pull_request:43997|pull_request:44074", + "pull_request:44010|pull_request:44018", + "pull_request:43997|pull_request:44129", + "pull_request:44071|pull_request:44129", + "pull_request:43996|pull_request:44161", + "pull_request:43996|pull_request:44017", + "pull_request:44010|pull_request:44044", + "pull_request:44044|pull_request:44074", + "pull_request:44066|pull_request:44722", + "pull_request:44085|pull_request:44722", + "pull_request:44085|pull_request:44129", + "pull_request:43996|pull_request:44007", + "pull_request:43996|pull_request:44129", + "pull_request:44044|pull_request:44129", + "pull_request:44044|pull_request:44154", + "pull_request:44024|pull_request:44076" + ], + "prepared_review_unit_hash": "dea8d8ade4c6d900e3bc8fcaeea99658361b0d7c20b7fd4ebf7101997c44b3a4", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 77, + "serialized_chars": 84500, + "estimated_input_tokens": 21125, + "estimated_eval_tokens": 42506 + }, + "budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 72, + "serialized_chars": 76935, + "estimated_input_tokens": 19234, + "estimated_eval_tokens": 38724 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a broad set of model-specific output-tracing refactors that are not duplicates of each other. The only clear duplicate families are within GPT-J/CodeGen and GPT-Neo; the rest touch different models and code paths despite similar wording and shared issue references.", + "confidence": 0.88, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44722 is the strongest representative of the real duplicate family: it targets the same output-tracing refactor as earlier GPT-J variants, explicitly references #43979, and is the broadest/latest concrete implementation in that subset.", + "best_issue_reason": null, + "best_pr_reason": "44722 is the best single PR to keep because it covers the genuine GPT-J/CodeGen duplicate thread, is explicit about the tracking issue, and looks like the most complete version of that change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43996", + "right": "pull_request:44085", + "accept": false, + "reason": "Different model families and different code paths; same issue target alone is not enough." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44044", + "accept": false, + "reason": "Unrelated models (CVT/FNet vs DeBERTa); only the generic output-tracing theme matches." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44085", + "accept": true, + "reason": "Same GPT-J output-tracing refactor family; 44085 looks like a narrower variant of the same code-path change." + }, + { + "left": "pull_request:44007", + "right": "pull_request:44072", + "accept": false, + "reason": "Both are standardized-output refactors, but for different backbones and different implementations." + }, + { + "left": "pull_request:44072", + "right": "pull_request:44722", + "accept": false, + "reason": "EfficientNet/FocalNet-style output tracing is unrelated to the GPT-J/CodeGen change in 44722." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44072", + "accept": false, + "reason": "Different model families and different concrete forward paths." + }, + { + "left": "pull_request:44013", + "right": "pull_request:44044", + "accept": false, + "reason": "MobileNetV2 and DeBERTa are unrelated code paths." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J/CodeGen vs MGP-STR; same theme but not the same underlying change." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44068", + "accept": true, + "reason": "Same GPT-Neo output-tracing refactor on the same file family; these look like duplicate iterations of one change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-J/CodeGen and MPT are different model implementations." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-J/CodeGen vs GPT-Neo; similar refactor pattern but distinct code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44066", + "accept": false, + "reason": "Different models and no shared concrete bug/change." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44722", + "accept": false, + "reason": "ResNet output tracing is unrelated to GPT-J/CodeGen output tracing." + }, + { + "left": "pull_request:44086", + "right": "pull_request:44722", + "accept": false, + "reason": "MGP-STR and GPT-J/CodeGen are different implementations." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44085", + "accept": false, + "reason": "ResNet and GPT-J are different code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44019", + "accept": false, + "reason": "GPT-Neo and ResNet are unrelated models." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44071", + "accept": false, + "reason": "ResNet vs MPT; only the generic tracing refactor theme overlaps." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-Neo and GPT-J/CodeGen are different model families." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert are unrelated code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo vs Flaubert; same tracing motif, different implementation." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44722", + "accept": false, + "reason": "MPT and GPT-J/CodeGen are separate model changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo and GPT-J are different code paths." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44086", + "accept": false, + "reason": "Different models and no single concrete bug shared." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-2 and MGP-STR are unrelated implementations." + }, + { + "left": "pull_request:44098", + "right": "pull_request:44101", + "accept": false, + "reason": "ViLT and Flaubert/XLM are different model families." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-2 vs GPT-J/CodeGen; similar refactor language but distinct code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo and MPT are different backbones." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44161", + "accept": false, + "reason": "SegFormer and LongT5 are unrelated model implementations." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44722", + "accept": false, + "reason": "SegFormer output tracing is not the same change as GPT-J/CodeGen tracing." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44073", + "accept": false, + "reason": "SegFormer and VisualBert are different code paths." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44101", + "accept": false, + "reason": "VisualBert vs Flaubert/XLM; only general refactor similarity." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44161", + "accept": false, + "reason": "VisualBert and LongT5 are unrelated." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-2 and MPT are different model families." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44101", + "accept": false, + "reason": "MPNet and Flaubert/XLM are unrelated code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44059", + "accept": false, + "reason": "GPT-Neo and GPT-2 are distinct model implementations." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-2 vs GPT-Neo; same broad goal, different concrete changes." + }, + { + "left": "pull_request:44101", + "right": "pull_request:44161", + "accept": false, + "reason": "Flaubert/XLM and LongT5 are different model families." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44017", + "accept": false, + "reason": "RegNet and SegFormer are unrelated models." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44056", + "accept": false, + "reason": "RegNet and MPNet are unrelated code paths." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44076", + "accept": false, + "reason": "MPT and ImageGPT are different implementations." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44101", + "accept": false, + "reason": "TextNet and Flaubert/XLM are unrelated changes." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44018", + "accept": false, + "reason": "RegNet and GPT-Neo are different model families." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44161", + "accept": false, + "reason": "MPNet and LongT5 are unrelated." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44161", + "accept": false, + "reason": "RegNet and LongT5 are not the same bug or change." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44161", + "accept": false, + "reason": "TextNet and LongT5 are different code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44044", + "accept": false, + "reason": "GPT-Neo and DeBERTa are unrelated implementations." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44071", + "accept": false, + "reason": "SqueezeBert and MPT are different models." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44073", + "accept": false, + "reason": "SqueezeBert and VisualBert are unrelated." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-Neo and SpeechT5 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:44044", + "accept": false, + "reason": "SqueezeBert and DeBERTa are different model families." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44074", + "accept": false, + "reason": "DeBERTa and TextNet are unrelated implementations." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J/CodeGen output-tracing refactor family; 44722 is the broader later version." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44722", + "accept": true, + "reason": "Both are GPT-J output-tracing refactors on the same file family, and 44722 subsumes the narrower version." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-J output tracing and SpeechT5 output tracing are unrelated." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44007", + "accept": false, + "reason": "CVT/FNet and RegNet/ResNet are different model code paths." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44129", + "accept": false, + "reason": "CVT/FNet and SpeechT5 are unrelated changes." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44129", + "accept": false, + "reason": "DeBERTa and SpeechT5 are different model implementations." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44154", + "accept": false, + "reason": "DeBERTa and Vits are unrelated code paths." + }, + { + "left": "pull_request:44024", + "right": "pull_request:44076", + "accept": false, + "reason": "FocalNet and ImageGPT are different models; same refactor pattern only." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative: it only groups the clearly overlapping GPT-J/CodeGen and GPT-Neo variants, while treating the other output-tracing PRs as model-specific changes. The accepted soft-edge pairs appear mergeable as single bugfix iterations within the same family, and the justification for keeping 44722 as the representative GPT-J/CodeGen PR is supported by the diff and issue links." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 13, + "component_count": 38, + "review_unit_index": 2, + "review_unit_count": 2, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43996-28", + "nodes": [ + "pull_request:43997", + "pull_request:44010", + "pull_request:44056", + "pull_request:44073", + "pull_request:44074", + "pull_request:44129" + ], + "soft_pairs": [ + "pull_request:44073|pull_request:44074", + "pull_request:44056|pull_request:44073", + "pull_request:44056|pull_request:44074", + "pull_request:43997|pull_request:44010", + "pull_request:44010|pull_request:44129" + ], + "prepared_review_unit_hash": "eb263695585382e64c4e445527e1e59ec6b1cc810296a1099196a2d021ba4c24", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 77, + "serialized_chars": 84500, + "estimated_input_tokens": 21125, + "estimated_eval_tokens": 42506 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 5, + "serialized_chars": 13065, + "estimated_input_tokens": 3267, + "estimated_eval_tokens": 6790 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These PRs are all in the same broad theme\u2014migrating model code to standardized output tracing/capture\u2014but they touch different model implementations and code paths, so they should not be deduplicated into one change.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "PR 44129 is the strongest standalone candidate because it has an explicit issue target and a focused SpeechT5-specific refactor, making its scope more concrete than the more generic model migration PRs.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44073", + "right": "pull_request:44074", + "accept": false, + "reason": "Same refactor pattern, but different models (VisualBert vs TextNet) and different files/tests; they are related migrations, not the same underlying code change." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44073", + "accept": false, + "reason": "Both standardize output tracing, but they modify different model internals and are not the same concrete bug fix or mergeable into one PR." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44074", + "accept": false, + "reason": "Shared output-capture refactor theme only; MPNet and TextNet are separate implementations with distinct forward paths." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44010", + "accept": false, + "reason": "RegNet and SqueezeBert are unrelated model code paths; similar decorator migration does not make them duplicates." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44129", + "accept": false, + "reason": "Both touch output tracing, but SpeechT5 includes a model-specific forward signature change and explicit issue linkage, so it is not the same change as SqueezeBert." + } + ] + }, + "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": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43998-13", + "nodes": [ + "pull_request:43998", + "pull_request:43999", + "pull_request:44000", + "pull_request:44001", + "pull_request:44002", + "pull_request:44003", + "pull_request:44004", + "pull_request:44025", + "pull_request:44026", + "pull_request:44027", + "pull_request:44028", + "pull_request:44029", + "pull_request:44030" + ], + "soft_pairs": [ + "pull_request:43998|pull_request:44002", + "pull_request:43998|pull_request:44000", + "pull_request:43998|pull_request:44001", + "pull_request:43998|pull_request:44003", + "pull_request:43998|pull_request:44029", + "pull_request:43999|pull_request:44001", + "pull_request:43999|pull_request:44002", + "pull_request:43999|pull_request:44025", + "pull_request:43999|pull_request:44026", + "pull_request:43999|pull_request:44027", + "pull_request:44000|pull_request:44001", + "pull_request:44000|pull_request:44002", + "pull_request:44000|pull_request:44003", + "pull_request:44000|pull_request:44025", + "pull_request:44001|pull_request:44002", + "pull_request:44001|pull_request:44003", + "pull_request:44001|pull_request:44025", + "pull_request:44002|pull_request:44029", + "pull_request:44002|pull_request:44003", + "pull_request:44003|pull_request:44028", + "pull_request:44003|pull_request:44004", + "pull_request:43998|pull_request:44004", + "pull_request:44001|pull_request:44004", + "pull_request:44002|pull_request:44004", + "pull_request:44000|pull_request:44004", + "pull_request:44025|pull_request:44026", + "pull_request:44025|pull_request:44027", + "pull_request:44025|pull_request:44028", + "pull_request:44025|pull_request:44029", + "pull_request:44025|pull_request:44030", + "pull_request:44026|pull_request:44029", + "pull_request:44026|pull_request:44027", + "pull_request:44026|pull_request:44028", + "pull_request:44026|pull_request:44030", + "pull_request:44027|pull_request:44028", + "pull_request:44027|pull_request:44029", + "pull_request:44027|pull_request:44030", + "pull_request:44028|pull_request:44029", + "pull_request:44028|pull_request:44030", + "pull_request:44029|pull_request:44030" + ], + "prepared_review_unit_hash": "eb233522de54da195a69271af531a66139b0f1c8d0ccfd8cca0a539f859e611d", + "component_budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40171, + "estimated_input_tokens": 10043, + "estimated_eval_tokens": 20342 + }, + "budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40171, + "estimated_input_tokens": 10043, + "estimated_eval_tokens": 20342 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is a set of open PRs that all refactor output tracing / `can_return_tuple` behavior, but each targets a different model or code path. They look thematically related, not duplicate changes.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44003 is the strongest representative of the theme: it makes a larger tracing refactor across related Mamba codepaths and adds the output-capturing plumbing, so it best captures the cluster\u2019s core change pattern.", + "best_issue_reason": null, + "best_pr_reason": "PR 44003 is the most substantial and representative example of the output-tracing refactor work, with the broadest implementation scope among the candidates.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43998", + "right": "pull_request:44002", + "accept": false, + "reason": "Different models and forward paths (`timm_backbone` vs `upernet`); same refactor style, but not the same bug/change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44000", + "accept": false, + "reason": "Separate model families (`timm_backbone` vs `vision_text_dual_encoder`) with distinct tracing behavior." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44001", + "accept": false, + "reason": "Both are output-tracing refactors, but they touch unrelated model codepaths (`timm_backbone` vs `univnet`)." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44003", + "accept": false, + "reason": "`timm_backbone` and `mamba` are unrelated implementations; this is a shared refactor pattern, not one duplicate fix." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44029", + "accept": false, + "reason": "Different architectures (`timm_backbone` vs `rwkv`) and different changes to output handling." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44001", + "accept": false, + "reason": "`mobilenet_v1` and `univnet` are unrelated models; both refactor output tracing but not the same change." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44002", + "accept": false, + "reason": "Different models and code paths (`mobilenet_v1` vs `upernet`); no shared concrete bug." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44025", + "accept": false, + "reason": "These are separate depth/vision model refactors, not one mergeable PR." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44026", + "accept": false, + "reason": "`mobilenet_v1` and `vision_encoder_decoder` affect different forward flows and output contracts." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44027", + "accept": false, + "reason": "Distinct model families (`mobilenet_v1` vs `speech_encoder_decoder`) despite similar tracing cleanup." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44001", + "accept": false, + "reason": "`vision_text_dual_encoder` and `univnet` are unrelated; the overlap is only the refactor pattern." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model implementations and return signatures; not the same underlying issue." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44003", + "accept": false, + "reason": "`vision_text_dual_encoder` and `mamba` change different code paths and output plumbing." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44025", + "accept": false, + "reason": "Separate models with different output APIs; similar tracing cleanup is too broad to treat as duplicate." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44002", + "accept": false, + "reason": "`univnet` and `upernet` are unrelated model families, so these are not the same fix." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44003", + "accept": false, + "reason": "Different architectures and implementation details; no concrete shared bug." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44025", + "accept": false, + "reason": "These target different model stacks and output contracts, not one duplicate change." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44029", + "accept": false, + "reason": "`upernet` and `rwkv` are unrelated; same general refactor theme only." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44003", + "accept": false, + "reason": "Different model families and distinct tracing implementations; not mergeable as one PR." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44028", + "accept": false, + "reason": "`mamba`/`falcon_mamba` output tracing is unrelated to `superpoint`; different code paths and APIs." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44004", + "accept": false, + "reason": "`mamba` and `codegen` have different forward semantics and output shapes; not the same underlying bug." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44004", + "accept": false, + "reason": "`timm_backbone` and `codegen` are unrelated codepaths; shared refactor wording is not enough." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "Different model families and different output-tracing changes." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44004", + "accept": false, + "reason": "`upernet` and `codegen` do not share a concrete fix target." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44004", + "accept": false, + "reason": "These are separate model implementations with unrelated output handling." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44026", + "accept": false, + "reason": "Both are tracing refactors, but `depth_anything` and `vision_encoder_decoder` are different models and APIs." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44027", + "accept": false, + "reason": "`depth_anything` and `speech_encoder_decoder` are unrelated code paths." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44028", + "accept": false, + "reason": "Different model families and output structures; not a duplicate fix." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44029", + "accept": false, + "reason": "`depth_anything` and `rwkv` touch different architectures and tracing mechanisms." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44030", + "accept": false, + "reason": "`depth_anything` and `dpr` are distinct model stacks; same broad theme only." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44029", + "accept": false, + "reason": "`vision_encoder_decoder` and `rwkv` modify unrelated forward/output code." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44027", + "accept": false, + "reason": "`vision_encoder_decoder` and `speech_encoder_decoder` are separate wrappers with different output handling." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44028", + "accept": false, + "reason": "Different model families (`vision_encoder_decoder` vs `superpoint`) and different return semantics." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44030", + "accept": false, + "reason": "These are unrelated model implementations with only superficial similarity in refactoring style." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44028", + "accept": false, + "reason": "`speech_encoder_decoder` and `superpoint` are different subsystems and not the same concrete change." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44029", + "accept": false, + "reason": "Different architectures and output plumbing; no duplicate fix here." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44030", + "accept": false, + "reason": "`speech_encoder_decoder` and `dpr` are distinct models with distinct changes." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44029", + "accept": false, + "reason": "`superpoint` and `rwkv` are unrelated code paths, despite both touching output handling." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44030", + "accept": false, + "reason": "Different models and different forward contracts; not the same underlying bug/change." + }, + { + "left": "pull_request:44029", + "right": "pull_request:44030", + "accept": false, + "reason": "`rwkv` and `dpr` are separate model families with distinct refactors." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 15, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-44662-2", + "nodes": [ + "pull_request:44662", + "pull_request:45401" + ], + "soft_pairs": [ + "pull_request:44662|pull_request:45401" + ], + "prepared_review_unit_hash": "bf69bb1bb9f0dfc13896bf2e2cc1d882c939b713c2e77de41eb3a002eddf73ee", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6551, + "estimated_input_tokens": 1638, + "estimated_eval_tokens": 3532 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6551, + "estimated_input_tokens": 1638, + "estimated_eval_tokens": 3532 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two unrelated model-implementation PRs: one adds PenguinVL, the other adds Voxtral TTS. They share some standard transformers scaffolding, but they do not fix the same code path or represent one underlying change.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: the two pull requests introduce different models and have different goals, so they are not duplicates of one another.", + "best_issue_reason": null, + "best_pr_reason": "PR 45401 is the more self-contained candidate only because it has an explicit issue target and a narrower model-specific scope, but it is not a better duplicate match to 44662.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44662", + "right": "pull_request:45401", + "accept": false, + "reason": "Reject: PenguinVL and Voxtral TTS are distinct model additions with different files, configs, and tests; shared auto-registration/docs infrastructure is generic and not evidence of the same underlying bug or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 16, + "component_count": 38, + "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-44965-2", + "nodes": [ + "pull_request:44965", + "pull_request:46845" + ], + "soft_pairs": [ + "pull_request:44965|pull_request:46845" + ], + "prepared_review_unit_hash": "e8bb4d101a0300222c0f7c554169c0414b1c76f5861dc2915f1be254a214cbfc", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4219, + "estimated_input_tokens": 1055, + "estimated_eval_tokens": 2366 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4219, + "estimated_input_tokens": 1055, + "estimated_eval_tokens": 2366 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two PRs are not duplicates: one is about changing a PR documentation workflow / converting PRs to draft, while the other is about switching CI to a smaller runner group and includes an unrelated code tweak in `modeling_utils.py`. They do not fix the same concrete problem and would not plausibly be merged into one PR.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46845 is the better representative because its title and diff are more specific and it contains a coherent CI-related change, whereas PR 44965 is titled generically and appears to be an unrelated workflow experiment.", + "best_issue_reason": null, + "best_pr_reason": "PR 46845 is the stronger cluster representative due to clearer intent and more substantive, identifiable changes across CI config and supporting files.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44965", + "right": "pull_request:46845", + "accept": false, + "reason": "Different underlying changes: 44965 alters the PR documentation/draft workflow, while 46845 changes CI routing to a smaller group and touches `modeling_utils.py`. They are not the same bug or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 17, + "component_count": 38, + "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": "617920c8a8addd1247bc2c82f1400e7b258c06d8bbec74716dada39108defe65", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4407, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4407, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related kernel-fusion efforts but not the same change: PR #44979 introduces a generic module fusion API, while PR #45363 implements n-to-1 kernel fusion via `KernelConfig` and integration plumbing. They share theme and terminology, but the code paths and deliverables are different enough that they should not be deduplicated as one PR.", + "confidence": 0.88, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45363 is the better canonical PR for this cluster because it is the more concrete, user-facing kernel-fusion implementation (`KernelConfig`, integration hooks, example), whereas #44979 is a lower-level generic module-fusion API foundation.", + "best_issue_reason": null, + "best_pr_reason": "PR #45363 is the best representative of the cluster\u2019s apparent intent: enabling kernel fusion behavior through `KernelConfig` and related runtime integration. PR #44979 is adjacent infrastructure, but not the same concrete feature change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44979", + "right": "pull_request:45363", + "accept": false, + "reason": "Related area, but not the same underlying change: #44979 adds a generic module fusion API, while #45363 wires up n-to-1 kernel fusion through `KernelConfig` and model integration. They could coexist in a larger roadmap, but they are not duplicates and 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": 18, + "component_count": 38, + "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": "328e30ad130ec62e1168e432302d04cbf9e104d8da910eb75234d419646c0094", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4414, + "estimated_input_tokens": 1104, + "estimated_eval_tokens": 2464 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4414, + "estimated_input_tokens": 1104, + "estimated_eval_tokens": 2464 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs fix the same Mistral regex pre-tokenizer AttributeError in `tokenization_utils_tokenizers.py`; `45317` is the more complete version because it adds tests.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "`45317` is the stronger canonical PR: it makes the same code-path fix as `45086` and also includes regression tests, so it is the more complete representation of the change.", + "best_issue_reason": null, + "best_pr_reason": "`45317` best captures the cluster\u2019s final fix because it addresses the same AttributeError and includes test coverage, making it the most suitable PR to keep.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45086", + "right": "pull_request:45317", + "accept": true, + "reason": "Same underlying bug fix: both change the same `_patch_mistral_regex` pre-tokenizer logic to avoid the AttributeError. `45317` only adds tests, so this is a true duplicate/superset pair." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 19, + "component_count": 38, + "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-45115-3", + "nodes": [ + "pull_request:45115", + "pull_request:45569", + "pull_request:46565" + ], + "soft_pairs": [ + "pull_request:45115|pull_request:45569", + "pull_request:45569|pull_request:46565" + ], + "prepared_review_unit_hash": "63271d45ee6da55a94ecee8cf34ff238036080f459bcd6b5e80b64a2f08631a4", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 9767, + "estimated_input_tokens": 2442, + "estimated_eval_tokens": 5140 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 9767, + "estimated_input_tokens": 2442, + "estimated_eval_tokens": 5140 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are three distinct PRs around Nemotron-related model support. PR 45569 is the broadest and most complete Nemotron H architecture split/refactor, while 45115 is a narrower conversion/inheritance refactor and 46565 is an unrelated Nemotron 3.5 ASR feature PR. I do not see a true duplicate pair here.", + "confidence": 0.78, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45569 is the best canonical PR because it is the most substantial and complete Nemotron H change set, covering the architecture split, auto-mapping updates, docs, and tests; the other Nemotron H PR is narrower and looks like an incremental precursor rather than the same final change.", + "best_issue_reason": null, + "best_pr_reason": "PR 45569 is the strongest representative of the cluster: it has the broadest implementation scope, the most discussion activity, and the clearest end-to-end model integration work.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45115", + "right": "pull_request:45569", + "accept": false, + "reason": "They overlap on Nemotron H internals, but 45115 is a focused refactor/conversion-mapping and inheritance change, while 45569 is a broader split into NemotronH dense/sparse packages with docs and tests. They are related, but not the same concrete change and would not cleanly merge as one PR." + }, + { + "left": "pull_request:45569", + "right": "pull_request:46565", + "accept": false, + "reason": "These target different models and problems: Nemotron H architecture refactoring versus Nemotron 3.5 ASR support. Shared auto/docs files are too broad to indicate a duplicate." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 20, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "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": "9f64ac059a920aebb1e7f189e49c169cefbada8e7bdea7c3b20a432cf6c370e3", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5953, + "estimated_input_tokens": 1489, + "estimated_eval_tokens": 3234 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5953, + "estimated_input_tokens": 1489, + "estimated_eval_tokens": 3234 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are closely related feature PRs, but not duplicates: #45333 appears to be the configuration-only subset, while #45332 adds the broader heterogeneous modeling/runtime support on top of that. I would keep them separate and use #45332 as the cluster representative.", + "confidence": 0.83, + "canonical_issue_reason": null, + "canonical_pr_reason": "#45332 is the broader, more complete PR: it includes the per-layer config work plus modeling/runtime integration, cache/masking changes, and related tests/docs. #45333 is a narrower config-only slice.", + "best_issue_reason": null, + "best_pr_reason": "#45332 is the best representative because it covers the full heterogeneous model support feature rather than only the configuration portion.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45332", + "right": "pull_request:45333", + "accept": false, + "reason": "Related feature work, but not the same concrete change: #45333 is config-only, while #45332 includes additional modeling/runtime code paths. They look like a subset/superset pair, not duplicate PRs that should be merged into one." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 21, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "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": "2736b7b6836cee37ef71cc7e6f27a1447dcbc3c4414090777f5f1f50d446eb0e", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5514, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5514, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are both typing-focused pull requests, but they are not the same underlying change: #45415 is a broad type-checking sweep across many transformers modules, while #45425 adds shared typing protocols/aliases and related typing fixes. The overlap in a few shared files looks incidental rather than duplicate work.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR: the shared files reflect overlapping typing infrastructure, but the diffs target different scopes and would not naturally collapse into one merged change.", + "best_issue_reason": null, + "best_pr_reason": "#45425 is the better representative of the cluster\u2019s narrower typing-infrastructure work; #45415 is a much broader repo-wide type-checking sweep.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45415 (Adds type checking to `src/transformers/*py`)", + "right": "pull_request:45425 (chore(typing): added modeling_utils to ty)", + "accept": false, + "reason": "Reject: both touch typing-related code, but #45415 is a broad type-checking pass across many files, whereas #45425 adds specific typing protocol/alias support and targeted fixes. They are not the same concrete bug/change and would not plausibly be one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 22, + "component_count": 38, + "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": "490f399ce59d210fee618ac65357427766137792b9a7e8af5c4b7a4010d6a181", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4508, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4508, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are related in theme (expert-parallel loading bugs) and both touch `modeling_utils.py`, but they fix different failure modes on different distributed backends: one targets DeepSpeed ZeRO-3 loading via `accelerate launch`, the other targets FSDP2/rank-0 broadcast overwriting expert weights. They are not duplicates.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "PR #45662 is the better representative of the cluster: it addresses a concrete expert-weight overwrite bug with a clearer root cause and broader EP/FSDP2 loading impact than the more backend-specific ZeRO-3 loading fix in #45548.", + "soft_edge_verdicts": [ + { + "left": "Fix EP + DeepSpeed ZeRO-3 loading via accelerate launch", + "right": "Fix EP + FSDP2: experts silently overwritten by rank-0 broadcast", + "accept": false, + "reason": "Different underlying bugs and code paths: ZeRO-3/accelerate loading versus FSDP2 rank-0 broadcast overwriting experts. Shared EP/modeling_utils context is not enough to treat them as the same change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 23, + "component_count": 38, + "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": "0b9d444b00d1c81a54637da1c257740df4db7d4bcec0ac8fef7c66deba482ee0", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4673, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4673, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs target the same xdist-related captured_info logging problem and share the same code paths/files. PR 45645 is the broader, more complete fix because it includes the xdist-safe output change plus CI workflow handling for multiple log files.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45645 is the best canonical PR: it covers the underlying xdist collision fix in testing_utils and also updates CI to collect all captured_info*.txt files, making the end-to-end logging behavior complete.", + "best_issue_reason": null, + "best_pr_reason": "PR 45645 is the strongest standalone representative because it addresses the same root bug and additionally preserves debug logs in workflow consumption, whereas 45639 is just the library-side half of the fix.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45639", + "right": "pull_request:45645", + "accept": true, + "reason": "Same underlying xdist/captured_info collision issue, overlapping testing_utils changes, and both are plausibly part of one merged fix; 45645 extends 45639 with CI log collection." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 24, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45720-2", + "nodes": [ + "pull_request:45720", + "pull_request:46259" + ], + "soft_pairs": [ + "pull_request:45720|pull_request:46259" + ], + "prepared_review_unit_hash": "c8344faf420d90a1998f681fbce3c2b37b166d58d3ba0d26f54797c4315fea6a", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7233, + "estimated_input_tokens": 1809, + "estimated_eval_tokens": 3874 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7233, + "estimated_input_tokens": 1809, + "estimated_eval_tokens": 3874 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs in transformers, but they target different linter rules and different code changes: one adds can_return_tuple to many model forwards, the other refactors weight initialization across many model classes. They are related only by broad modeling/linter cleanup, not the same underlying fix.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45720 is the better anchor: it has a tighter, more specific code-path change (forward return handling via can_return_tuple) and is easier to compare against true duplicates than the broader initialization-rule sweep.", + "best_issue_reason": null, + "best_pr_reason": "PR 45720 is the best representative of the cluster because it is the more narrowly scoped and concrete implementation of a single rule, whereas 46259 is a separate linter-driven refactor with different behavior.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45720", + "right": "pull_request:46259", + "accept": false, + "reason": "Different linter rules and different fixes: TRF003 adds can_return_tuple / return-dict handling in forward methods, while TRF018 changes model weight initialization logic. They do not patch the same concrete bug or code path 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": 25, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45737-2", + "nodes": [ + "pull_request:45737", + "pull_request:45738" + ], + "soft_pairs": [ + "pull_request:45737|pull_request:45738" + ], + "prepared_review_unit_hash": "cc2ade73a49ec6a8883a7a24c433988645588bcd049e8aad2bce56b1b24daa1c", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4131, + "estimated_input_tokens": 1033, + "estimated_eval_tokens": 2322 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4131, + "estimated_input_tokens": 1033, + "estimated_eval_tokens": 2322 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs address the same MusicgenMelody cache initialization bug in `forward()`: replacing `EncoderDecoderCache(DynamicCache, DynamicCache)` with a single `DynamicCache`. PR #45737 is the more complete duplicate because it bundles the fix with a regression test, while #45738 is the fix-only variant.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45737 is the best canonical PR because it contains the same code fix as #45738 plus a regression test, making it the most complete representation of the bug fix.", + "best_issue_reason": null, + "best_pr_reason": "PR #45737 is the strongest overall candidate: it fixes the concrete cache bug and adds a targeted regression test, whereas #45738 only changes the implementation.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45737", + "right": "pull_request:45738", + "accept": true, + "reason": "Same underlying code-path bug in `musicgen_melody` cache setup; the diffs match on the core fix and the test-only addition in #45737 could plausibly live in the same PR as #45738." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The assessment is grounded in the packet: both PRs make the same `musicgen_melody` cache initialization change, and #45737 only adds a regression test on top of that fix. The soft-edge accept is conservative enough because the diffs are mergeable into one PR for the same bugfix." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 26, + "component_count": 38, + "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-45932-3", + "nodes": [ + "pull_request:45932", + "pull_request:46100", + "pull_request:46891" + ], + "soft_pairs": [ + "pull_request:45932|pull_request:46100", + "pull_request:45932|pull_request:46891", + "pull_request:46100|pull_request:46891" + ], + "prepared_review_unit_hash": "9a1c9345cd02e5fc24201ea6b95f7435a9e998ab594f4893b6e5d6adfa6a60f5", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7882, + "estimated_input_tokens": 1971, + "estimated_eval_tokens": 4198 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7882, + "estimated_input_tokens": 1971, + "estimated_eval_tokens": 4198 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are all test/CI-oriented, but they target different concrete changes: grafana/CI secrets plus model tweaks, a kernel test migration, and a PEFT integration CI branch. They are not the same underlying bug or change, so this is not a true duplicate cluster.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "No clear canonical PR: the three PRs are unrelated in scope and code path, with only superficial overlap in testing/CI infrastructure.", + "best_issue_reason": null, + "best_pr_reason": "PR 46891 is the narrowest and most self-contained of the three, but it still does not match the others as a duplicate; it is only the least broad representative if one must be chosen.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45932", + "right": "pull_request:46100", + "accept": false, + "reason": "Both are test/CI flavored, but 45932 is a grafana/CI-secrets plus model-mapping change while 46100 is a kernel test migration; they do not fix the same code-path problem." + }, + { + "left": "pull_request:45932", + "right": "pull_request:46891", + "accept": false, + "reason": "Shared workflow filename is not enough here: 45932 adds grafana-related CI secrets and model edits, while 46891 is a PEFT integration CI branch with peft.py/tests_fetcher changes." + }, + { + "left": "pull_request:46100", + "right": "pull_request:46891", + "accept": false, + "reason": "46100 is a broad kernel test migration, whereas 46891 is a PEFT integration CI/test update; different functionality and no common concrete bug or merged 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": 38, + "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": "1550df42fe245d17692552a8f0b6b2867eb4904d56d24422b679ed5f456d366e", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5525, + "estimated_input_tokens": 1382, + "estimated_eval_tokens": 3020 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5525, + "estimated_input_tokens": 1382, + "estimated_eval_tokens": 3020 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs touch overlapping distributed files, but they are not the same change: one is a focused EP/Trainer integration PR, the other is a broad distributed branch base with many unrelated edits.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #46126 is the most focused and representative change in the cluster: it specifically implements EP + Trainer integration on top of DistributedConfig. PR #46269 is a broad branch-base PR spanning many distributed/FSDP areas and is not a clean canonical duplicate target.", + "best_issue_reason": null, + "best_pr_reason": "PR #46126 is the best standalone PR to keep as the cluster representative because it has a single concrete purpose and tighter scope than #46269.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46126", + "right": "pull_request:46269", + "accept": false, + "reason": "They overlap on distributed filenames, but the PRs are not the same underlying change. #46126 is a targeted EP/Trainer integration, while #46269 is a large distributed branch base with many unrelated modifications; they would not plausibly merge into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 28, + "component_count": 38, + "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-46138-10", + "nodes": [ + "pull_request:46138", + "pull_request:46432", + "pull_request:46501", + "pull_request:46502", + "pull_request:46503", + "pull_request:46504", + "pull_request:46554", + "pull_request:46558", + "pull_request:46563", + "pull_request:46716" + ], + "soft_pairs": [ + "pull_request:46138|pull_request:46554", + "pull_request:46138|pull_request:46558", + "pull_request:46138|pull_request:46563", + "pull_request:46138|pull_request:46716", + "pull_request:46138|pull_request:46504", + "pull_request:46432|pull_request:46503", + "pull_request:46432|pull_request:46501", + "pull_request:46432|pull_request:46502", + "pull_request:46432|pull_request:46554", + "pull_request:46432|pull_request:46558", + "pull_request:46501|pull_request:46503", + "pull_request:46501|pull_request:46502", + "pull_request:46501|pull_request:46504", + "pull_request:46501|pull_request:46554", + "pull_request:46502|pull_request:46503", + "pull_request:46502|pull_request:46504", + "pull_request:46502|pull_request:46554", + "pull_request:46503|pull_request:46504", + "pull_request:46503|pull_request:46554", + "pull_request:46504|pull_request:46554", + "pull_request:46504|pull_request:46558", + "pull_request:46554|pull_request:46558", + "pull_request:46554|pull_request:46563", + "pull_request:46554|pull_request:46716", + "pull_request:46558|pull_request:46563", + "pull_request:46558|pull_request:46716", + "pull_request:46563|pull_request:46716", + "pull_request:46504|pull_request:46563", + "pull_request:46504|pull_request:46716" + ], + "prepared_review_unit_hash": "f5069f0f3b5dbbb67f1ce25e67993f4ea45c3c2858cbee4d33ae2acd06e3e61f", + "component_budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 29, + "serialized_chars": 25932, + "estimated_input_tokens": 6483, + "estimated_eval_tokens": 13222 + }, + "budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 29, + "serialized_chars": 25932, + "estimated_input_tokens": 6483, + "estimated_eval_tokens": 13222 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous CI/workflow maintenance. I found no true duplicate PR pairs; the only tight subcluster is the pr-ci-security-gate.yml series, but those PRs are separate hardening tweaks rather than the same fix.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: the cluster spans unrelated workflow chores, and the pr-ci-security-gate PRs are distinct edits to the same file rather than duplicates.", + "best_issue_reason": null, + "best_pr_reason": "PR 46554 is the strongest representative of the only coherent subtheme (pr-ci-security-gate), because it contains the main functional workflow logic change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46138", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflows and concerns: stdin handling for self-comment CI vs security-gate logic changes." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46558", + "accept": false, + "reason": "Different workflow files and different behavior; not the same underlying bug or change." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46563", + "accept": false, + "reason": "Self-comment CI input handling vs a comment-only security-gate update; unrelated patches." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46716", + "accept": false, + "reason": "Different workflow integrations: self-comment CI vs reusable-workflow secrets handling." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46504", + "accept": false, + "reason": "Same broad CI area, but one changes comment piping and the other changes commit-message parsing for image tagging." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46503", + "accept": false, + "reason": "SSH-runner hardening vs benchmark workflow dispatch behavior; no same code path." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46501", + "accept": false, + "reason": "SSH runner security changes vs anti-slop workflow permissions; unrelated changes." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46502", + "accept": false, + "reason": "SSH runner hardening vs reviewer-assignment checkout behavior; not the same fix." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46554", + "accept": false, + "reason": "SSH-runner changes are unrelated to pr-ci-security-gate workflow logic." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46558", + "accept": false, + "reason": "Different workflows and different concerns; not a duplicate patch." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46503", + "accept": false, + "reason": "Anti-slop permission scoping vs benchmark_v2 manual dispatch support; unrelated workflow updates." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46502", + "accept": false, + "reason": "Anti-slop permissions change vs base-ref checkout for assign-reviewers; different fixes." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46504", + "accept": false, + "reason": "Anti-slop workflow permissions vs build-ci-docker-images commit-message handling; not the same bug." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflows: anti-slop permissions vs security-gate checkout/check logic." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46503", + "accept": false, + "reason": "Reviewer-assignment checkout change vs benchmark dispatch addition; unrelated." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46504", + "accept": false, + "reason": "Assign-reviewers base-ref checkout vs build-image tagging logic; different code paths." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46554", + "accept": false, + "reason": "Assign-reviewers workflow tweak vs security-gate workflow hardening; not duplicates." + }, + { + "left": "pull_request:46503", + "right": "pull_request:46504", + "accept": false, + "reason": "Benchmark dispatch support vs build-ci-docker-images message parsing; different functionality." + }, + { + "left": "pull_request:46503", + "right": "pull_request:46554", + "accept": false, + "reason": "Benchmark workflow trigger change vs security-gate logic; unrelated patches." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46554", + "accept": false, + "reason": "Build-image tagging logic vs pr-ci-security-gate checks; different workflows and fixes." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46558", + "accept": false, + "reason": "Build-image commit parsing vs security-gate permission/approval changes; not the same change." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46558", + "accept": false, + "reason": "Same file, but one adds checkout/diff-based checks while the other adjusts permissions and approval gating." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46563", + "accept": false, + "reason": "Same workflow file, but one changes gate logic and the other only adds explanatory comments." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46716", + "accept": false, + "reason": "Same workflow family, but one changes security-gate behavior and the other changes secrets handling for the reusable workflow call." + }, + { + "left": "pull_request:46558", + "right": "pull_request:46563", + "accept": false, + "reason": "Both touch pr-ci-security-gate.yml, but permissions/if-condition edits are not the same as comment-only changes." + }, + { + "left": "pull_request:46558", + "right": "pull_request:46716", + "accept": false, + "reason": "Permissions/approval gating vs secrets:{} change; separate hardening steps, not one duplicate fix." + }, + { + "left": "pull_request:46563", + "right": "pull_request:46716", + "accept": false, + "reason": "Comment-only metadata update vs secrets handling; clearly different changes." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46563", + "accept": false, + "reason": "Build-ci-docker-images commit parsing vs comment-only security-gate update; unrelated." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46716", + "accept": false, + "reason": "Build-image tagging logic vs secrets handling in pr-ci-security-gate; 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": "pull_request", + "component_index": 29, + "component_count": 38, + "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": "e2d3a7f39b99e164d928f13929aff8412dde369bf898520d7e62a4a82f49ef7a", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4274, + "estimated_input_tokens": 1069, + "estimated_eval_tokens": 2394 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4274, + "estimated_input_tokens": 1069, + "estimated_eval_tokens": 2394 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two PRs are thematically related (removing direct file-open source reads), but they fix different code paths in different modules and should not be treated as duplicates.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR for duplication: 46152 changes source access in import_utils, while 46207 changes source inspection in modeling_utils; they are separate fixes.", + "best_issue_reason": null, + "best_pr_reason": "46207 is the stronger standalone fix because it targets one concrete heuristic in _can_set_attn_implementation and includes tests, whereas 46152 is a different source-access refactor in another module.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46152", + "right": "pull_request:46207", + "accept": false, + "reason": "Same broad theme, but not the same underlying bug/change: 46152 rewrites self-source resolution in import_utils, while 46207 updates source reading for attention-implementation detection in modeling_utils. They touch different code paths and would not plausibly be merged into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 30, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46174-2", + "nodes": [ + "pull_request:46174", + "pull_request:46780" + ], + "soft_pairs": [ + "pull_request:46174|pull_request:46780" + ], + "prepared_review_unit_hash": "a48ce4ca162f555682f79411d7229ac1ccc08d7822dfab654bb881639fccddbc", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6030, + "estimated_input_tokens": 1508, + "estimated_eval_tokens": 3272 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6030, + "estimated_input_tokens": 1508, + "estimated_eval_tokens": 3272 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two PRs target the same reported MPS crash in the sinusoidal position embedding helper, but they implement different fixes and are not safe to treat as duplicates.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46174 is the broader, more complete fix for the float64-on-MPS crash: it updates the helper across all affected model files and keeps the internal math on CPU float64 before casting back.", + "best_issue_reason": null, + "best_pr_reason": "PR 46174 is the best representative of the cluster because it is the original, more comprehensive fix and has more discussion/review context than the narrower follow-up.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46174", + "right": "pull_request:46780", + "accept": false, + "reason": "They share the same issue target and helper, but the patches are different solutions: 46174 avoids float64 on MPS by doing the arithmetic on CPU first, while 46780 switches the helper to use dtype for internal arithmetic. Those are alternative implementations, not one mergeable duplicate PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 31, + "component_count": 38, + "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-46281-2", + "nodes": [ + "pull_request:46281", + "pull_request:46890" + ], + "soft_pairs": [ + "pull_request:46281|pull_request:46890" + ], + "prepared_review_unit_hash": "d0ec78e721627db30d5465428f3f9dbdbc92cb9a26a92ec55846cf8349f48f80", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4270, + "estimated_input_tokens": 1068, + "estimated_eval_tokens": 2392 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4270, + "estimated_input_tokens": 1068, + "estimated_eval_tokens": 2392 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs harden path handling against traversal, but they target different untrusted inputs and different loader code paths: sharded checkpoint `weight_map` vs. `transformers_weights` config field. They are related in theme, not duplicates.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #46281 is the cleaner representative: it fixes a concrete containment bug in the sharded-checkpoint loader and shows an explicit path-validation change in one code path.", + "best_issue_reason": null, + "best_pr_reason": "PR #46281 is the best single representative because it is a focused, self-contained fix with clear exploitability and minimal overlap with the other PR.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46281", + "right": "pull_request:46890", + "accept": false, + "reason": "Same general security class (path traversal), but they modify different loaders and trust different fields. They are not the same underlying bug and would not 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": 32, + "component_count": 38, + "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-46310-2", + "nodes": [ + "pull_request:46310", + "pull_request:46546" + ], + "soft_pairs": [ + "pull_request:46310|pull_request:46546" + ], + "prepared_review_unit_hash": "3de6542000db196c9d4cee095a36209f916e232e6265c9790af2fe84bbdbb99f", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4850, + "estimated_input_tokens": 1213, + "estimated_eval_tokens": 2682 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4850, + "estimated_input_tokens": 1213, + "estimated_eval_tokens": 2682 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs fix the same Moonshine streaming sliding-window boundary bug: the right window should include the edge index (`<=` instead of `<`). PR 46546 is the stronger representative because it includes the Moonshine fix plus a test, and also applies the same boundary correction in Gemma4.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46546 is the best canonical PR because it contains the same Moonshine boundary fix, adds test coverage, and is the more complete change set overall.", + "best_issue_reason": null, + "best_pr_reason": "PR 46546 is the best overall PR for this cluster since it is the more complete, validated fix for the same underlying boundary bug.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46310", + "right": "pull_request:46546", + "accept": true, + "reason": "Same concrete bug: the Moonshine streaming right-window mask excludes the boundary element in both PRs, and both change the same code path to `<=`." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 33, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46376-6", + "nodes": [ + "pull_request:46376", + "pull_request:46417", + "pull_request:46425", + "pull_request:46509", + "pull_request:46850", + "pull_request:46917" + ], + "soft_pairs": [ + "pull_request:46376|pull_request:46417", + "pull_request:46376|pull_request:46425", + "pull_request:46425|pull_request:46509", + "pull_request:46425|pull_request:46917", + "pull_request:46425|pull_request:46850", + "pull_request:46850|pull_request:46917" + ], + "prepared_review_unit_hash": "e9c74e8fe14abdd1a0c1a4bf955fae223b99a938a9e957c9f41bb90baf24d381", + "component_budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 6, + "serialized_chars": 16837, + "estimated_input_tokens": 4210, + "estimated_eval_tokens": 8676 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 6, + "serialized_chars": 16837, + "estimated_input_tokens": 4210, + "estimated_eval_tokens": 8676 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are only superficially similar; they span unrelated areas like model additions, continuous batching, masking logic, and deprecation cleanup. No pair looks like the same underlying change.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "No true canonical duplicate PR: the set is heterogeneous, and even the closest items are different model additions or unrelated fixes.", + "best_issue_reason": null, + "best_pr_reason": "PR 46425 is the broadest representative of the superficial model-support subgroup, but it is still a distinct RADIO model addition rather than a duplicate of the others.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46376", + "right": "pull_request:46417", + "accept": false, + "reason": "Continuous batching multimodal support vs. adding the HunYuan VL model; different code paths and goals." + }, + { + "left": "pull_request:46376", + "right": "pull_request:46425", + "accept": false, + "reason": "Continuous batching infrastructure vs. RADIO model support; not the same bug or change." + }, + { + "left": "pull_request:46425", + "right": "pull_request:46509", + "accept": false, + "reason": "RADIO model support vs. Nemotron Omni support; both are model additions but for different models and implementations." + }, + { + "left": "pull_request:46425", + "right": "pull_request:46917", + "accept": false, + "reason": "New RADIO model support vs. removing deprecated training args; unrelated changes." + }, + { + "left": "pull_request:46425", + "right": "pull_request:46850", + "accept": false, + "reason": "Model registration/support work vs. a Gemma attention-masking bug fix; different underlying problems." + }, + { + "left": "pull_request:46850", + "right": "pull_request:46917", + "accept": false, + "reason": "Gemma masking fix vs. deprecation cleanup; no shared concrete code-path problem." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 34, + "component_count": 38, + "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-46574-2", + "nodes": [ + "pull_request:46574", + "pull_request:46580" + ], + "soft_pairs": [ + "pull_request:46574|pull_request:46580" + ], + "prepared_review_unit_hash": "8aff577f9e37e22cb2e6d50ed21159af6bba346f2a581234c86e4bee9cfe361c", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4410, + "estimated_input_tokens": 1103, + "estimated_eval_tokens": 2462 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4410, + "estimated_input_tokens": 1103, + "estimated_eval_tokens": 2462 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two PRs are unrelated tokenizer fixes: one corrects CodeLlama decode behavior to preserve leading spaces, while the other fixes DeepSeek-Coder/DeepSeek-LLM checkpoints loading under the wrong tokenizer class. They do not look like the same underlying bug or change.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:46574", + "right": "pull_request:46580", + "accept": false, + "reason": "Different models and different code paths: 46574 changes CodeLlama decoder setup to stop stripping real leading spaces, while 46580 changes AutoTokenizer checkpoint-prefix routing for DeepSeek checkpoints that load the wrong tokenizer class. These are not mergeable as one 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": 35, + "component_count": 38, + "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-46603-2", + "nodes": [ + "pull_request:46603", + "pull_request:46604" + ], + "soft_pairs": [ + "pull_request:46603|pull_request:46604" + ], + "prepared_review_unit_hash": "c79b4e399c793cc34f9ee76a73881d9ff19bc9ebb91cc25de0f592caab08ca91", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5496, + "estimated_input_tokens": 1374, + "estimated_eval_tokens": 3004 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5496, + "estimated_input_tokens": 1374, + "estimated_eval_tokens": 3004 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs are Mistral-tokenizer related and share file overlap, but they target different changes: one is a refactor/extraction of MistralConverter into a package, the other adds tekken tokenizer detection/conversion/save support. They are not the same underlying fix.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #46604 is the more representative canonical PR for this cluster because it delivers the broader user-facing Mistral tokenizer functionality, while #46603 is mainly a packaging/refactor change.", + "best_issue_reason": null, + "best_pr_reason": "PR #46604 is the best standalone PR here: it has a concrete functional goal, touches the relevant integration and tokenizer code paths, and includes corresponding tests.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46603", + "right": "pull_request:46604", + "accept": false, + "reason": "Reject: #46603 moves/extracts MistralConverter into the integrations package, while #46604 adds tekken tokenizer detection, conversion, and save utilities. Shared filenames reflect the same subsystem, 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": 36, + "component_count": 38, + "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-46741-3", + "nodes": [ + "pull_request:46741", + "pull_request:46773", + "pull_request:46781" + ], + "soft_pairs": [ + "pull_request:46741|pull_request:46773", + "pull_request:46741|pull_request:46781" + ], + "prepared_review_unit_hash": "789465efed8a93bdfa78315690a9d6a1962a07730936f872abf9fe33a3f1c185", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 8455, + "estimated_input_tokens": 2114, + "estimated_eval_tokens": 4484 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 8455, + "estimated_input_tokens": 2114, + "estimated_eval_tokens": 4484 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "All three pull requests target different bugs in adjacent model/cache codepaths, but they do not appear to be duplicates of one another. The two soft-similarity edges should be rejected.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46741 is the broadest and most representative fix in the set: it addresses a concrete chunked-prefill/speculative-decoding cache-path bug across multiple Mamba-like models and has the richest model/test coverage.", + "best_issue_reason": null, + "best_pr_reason": "PR 46741 is the strongest standalone candidate overall because it fixes a specific runtime path with multi-model coverage and substantive implementation changes, whereas the others are narrower and unrelated bugs.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46741", + "right": "pull_request:46773", + "accept": false, + "reason": "Different bugs: 46741 fixes chunked-prefill/speculative decoding cache handling, while 46773 fixes padding-mask application being skipped for batch-size-1 linear-attention models." + }, + { + "left": "pull_request:46741", + "right": "pull_request:46781", + "accept": false, + "reason": "Different codepaths and failure modes: 46741 is about precomputed state handling in model forwards, while 46781 changes QuantizedCache construction for hybrid SSM+attention layer types." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 37, + "component_count": 38, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46895-3", + "nodes": [ + "pull_request:46895", + "pull_request:46898", + "pull_request:46916" + ], + "soft_pairs": [ + "pull_request:46895|pull_request:46898", + "pull_request:46895|pull_request:46916", + "pull_request:46898|pull_request:46916" + ], + "prepared_review_unit_hash": "72136622b9be672a109a018ba8cf739b2cb3a2732ca0b7470e1e2f7209f9f8fe", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7715, + "estimated_input_tokens": 1929, + "estimated_eval_tokens": 4114 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7715, + "estimated_input_tokens": 1929, + "estimated_eval_tokens": 4114 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs split into two model-specific training-loss double-shift fixes: CohereASR and Florence2. PR 46916 is the broadest representative because it bundles both fixes, while 46895 and 46898 are single-model variants. The cross-model pair 46895\u219446898 should not be treated as duplicates.", + "confidence": 0.82, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46916 is the broadest, most representative change set: it covers both CohereASR and Florence2 double-shift fixes and shares the explicit Florence2 issue target.", + "best_issue_reason": null, + "best_pr_reason": "PR 46916 best represents the cluster because it unifies the recurring bug pattern across both affected models, whereas 46895 and 46898 are narrower single-model fixes.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46895", + "right": "pull_request:46898", + "accept": false, + "reason": "Same high-level bug pattern, but they fix different model implementations and different issue targets. These are separate concrete code paths, not a duplicate PR pair." + }, + { + "left": "pull_request:46895", + "right": "pull_request:46916", + "accept": true, + "reason": "Both touch CohereASR's same training-loss double-shift path; 46916 just bundles that fix with an additional Florence2 change, so they plausibly overlap as the same underlying CohereASR bug." + }, + { + "left": "pull_request:46898", + "right": "pull_request:46916", + "accept": true, + "reason": "Both address Florence2's same double-shift loss path, and 46916 simply combines that fix with the CohereASR variant, so the Florence2 portion is the same underlying bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 38, + "component_count": 38, + "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-46907-5", + "nodes": [ + "pull_request:46907", + "pull_request:46908", + "pull_request:46926", + "pull_request:46927", + "pull_request:46928" + ], + "soft_pairs": [ + "pull_request:46907|pull_request:46927", + "pull_request:46907|pull_request:46926", + "pull_request:46907|pull_request:46908", + "pull_request:46907|pull_request:46928", + "pull_request:46908|pull_request:46926", + "pull_request:46908|pull_request:46928", + "pull_request:46908|pull_request:46927", + "pull_request:46926|pull_request:46927", + "pull_request:46926|pull_request:46928", + "pull_request:46927|pull_request:46928" + ], + "prepared_review_unit_hash": "db2153021e1ee3a019fbf3eac0c3b269474a89d485b474f74a314deeadee128d", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 10, + "serialized_chars": 11922, + "estimated_input_tokens": 2981, + "estimated_eval_tokens": 6218 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 10, + "serialized_chars": 11922, + "estimated_input_tokens": 2981, + "estimated_eval_tokens": 6218 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "All five PRs share a generic \u201cfix integration tests\u201d pattern, but they target different models and different failure modes. None of the soft-edge pairs look like the same underlying bug or change.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46907 is the best representative only in a loose sense: it fixes a concrete model code-path bug in `edgetam`, whereas the others are mostly model-specific test/setup or registry tweaks. It is not a true duplicate of the others.", + "best_issue_reason": null, + "best_pr_reason": "PR 46907 is the strongest cluster representative because it changes actual model forwarding logic to drop an invalid kwarg, while the other PRs are isolated fixes for unrelated models or test harness setup.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46907", + "right": "pull_request:46927", + "accept": false, + "reason": "Different models and different fixes: `edgetam` removes `original_sizes` before backbone call; `glm46v` casts `input_ids` to long. Same broad theme, not the same underlying bug." + }, + { + "left": "pull_request:46907", + "right": "pull_request:46926", + "accept": false, + "reason": "`edgetam` code-path fix vs `minimax_m3_vl` test-only offload folder addition. Unrelated model failures and unrelated remedies." + }, + { + "left": "pull_request:46907", + "right": "pull_request:46908", + "accept": false, + "reason": "`edgetam` fixes an invalid kwarg in model forward; `deepseek_v32` adds `offload_folder` to test loading. Different subsystems and failure modes." + }, + { + "left": "pull_request:46907", + "right": "pull_request:46928", + "accept": false, + "reason": "`edgetam` model forward bug vs `blip_2` tokenizer registry update. No shared concrete code-path problem." + }, + { + "left": "pull_request:46908", + "right": "pull_request:46926", + "accept": false, + "reason": "Both are load_error-related test fixes, but for different models and different offload/test setups. Not mergeable as one PR." + }, + { + "left": "pull_request:46908", + "right": "pull_request:46927", + "accept": false, + "reason": "`deepseek_v32` test loading workaround vs `glm46v` forward-type fix. Different bug classes and code paths." + }, + { + "left": "pull_request:46908", + "right": "pull_request:46928", + "accept": false, + "reason": "`deepseek_v32` offload-folder test change is unrelated to the `blip_2` tokenizer mapping addition." + }, + { + "left": "pull_request:46926", + "right": "pull_request:46927", + "accept": false, + "reason": "`minimax_m3_vl` adds a test offload folder; `glm46v` changes forward input typing. Distinct model-specific issues." + }, + { + "left": "pull_request:46926", + "right": "pull_request:46928", + "accept": false, + "reason": "`minimax_m3_vl` load workaround vs `blip_2` tokenizer config change. Same generic area of tests/models, but not the same bug." + }, + { + "left": "pull_request:46927", + "right": "pull_request:46928", + "accept": false, + "reason": "`glm46v` fixes input dtype handling; `blip_2` extends tokenizer auto-mapping. No common underlying defect." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + } + ] +} diff --git a/snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/manifest.json b/snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..f423f0fd95bd703c75674ebc1663b07c17289dfe --- /dev/null +++ b/snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/manifest.json @@ -0,0 +1,14 @@ +{ + "analysis_id": "hybrid-model-20260627t120014z", + "artifacts": { + "hybrid": "snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "model": null, + "published_at": "2026-06-27T12:10:33Z", + "repo": "huggingface/transformers", + "schema_version": 1, + "snapshot_id": "20260627T120014Z", + "variant": "hybrid" +} diff --git a/snapshots/20260627T120014Z/manifest.json b/snapshots/20260627T120014Z/manifest.json index ee08c5d4f293ddfa1c647966dce161ecf9c0d23f..865f48bf6299018c885bf036e416f5277ab889e1 100644 --- a/snapshots/20260627T120014Z/manifest.json +++ b/snapshots/20260627T120014Z/manifest.json @@ -32,6 +32,24 @@ "timeline_events": 312 }, "extracted_at": "2026-06-27T12:00:14Z", + "published_analysis": { + "canonical_analysis_id": "hybrid-model-20260627t120014z", + "runs": { + "hybrid-model-20260627t120014z": { + "analysis_id": "hybrid-model-20260627t120014z", + "artifacts": { + "hybrid": "snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "manifest_path": "snapshots/20260627T120014Z/analysis-runs/hybrid-model-20260627t120014z/manifest.json", + "model": null, + "published_at": "2026-06-27T12:10:33Z", + "variant": "hybrid" + } + }, + "schema_version": 1 + }, "repo": "huggingface/transformers", "snapshot_id": "20260627T120014Z", "watermark": {