{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:UFNMQTM3CZRSICCJNKAEVPYBPU","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"cfc3a751c58c0db75e5635e0c86944b4554226db4975fafa682d72901497c7d7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-25T07:18:30Z","title_canon_sha256":"33773dbbee0db37c531b7fba373163b94d36dcd257911d480bd5ae8c2699e8e5"},"schema_version":"1.0","source":{"id":"2604.23178","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.23178","created_at":"2026-06-25T01:18:38Z"},{"alias_kind":"arxiv_version","alias_value":"2604.23178v2","created_at":"2026-06-25T01:18:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.23178","created_at":"2026-06-25T01:18:38Z"},{"alias_kind":"pith_short_12","alias_value":"UFNMQTM3CZRS","created_at":"2026-06-25T01:18:38Z"},{"alias_kind":"pith_short_16","alias_value":"UFNMQTM3CZRSICCJ","created_at":"2026-06-25T01:18:38Z"},{"alias_kind":"pith_short_8","alias_value":"UFNMQTM3","created_at":"2026-06-25T01:18:38Z"}],"graph_snapshots":[{"event_id":"sha256:f3a9c64291407e42afe0a4824cda372fe7e5b9d678f0a990c3d929abe77ae9da","target":"graph","created_at":"2026-06-25T01:18:38Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Style bias is the dominant bias (0.76-0.92 across all models), far exceeding position bias (<= 0.04), yet has received minimal research attention."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The chosen benchmarks (MT-Bench, LLMBar, and the custom set) and bias measurement protocols accurately isolate the targeted biases without introducing confounding artifacts from the test data itself or from how responses were generated."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Style bias dominates LLM-as-a-Judge systems far more than position bias, with debiasing strategies providing model-dependent gains and public tools released for replication."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Style bias dominates LLM judges at 0.76-0.92 strength across models, dwarfing position bias, while combined debiasing improves agreement for some judges by over 11 points."}],"snapshot_sha256":"2bd7de3189d8e0eec5c84f313280246433f95c57d0da21e811a399e27bbd3a16"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-21T09:37:57.140419Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T23:24:23.663274Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2604.23178/integrity.json","findings":[],"snapshot_sha256":"15c156ba6924d6fd6919298bff656bcddecd689c43cba5e7ba01d474861a1b0f","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM-as-a-Judge has become the dominant paradigm for evaluating language model outputs, yet LLM judges exhibit systematic biases that compromise evaluation reliability. We present a comprehensive empirical study comparing nine debiasing strategies across five judge models from four provider families (Google, Anthropic, OpenAI, Meta), three benchmarks (MT-Bench n=400, LLMBar n=200, custom n=375), and four bias types. Our headline practical finding is that a mid-tier model with the right debiasing can outperform frontier judges at a fraction of the cost: Gemini 2.5 Flash with the Combined Budget ","authors_text":"Sadman Kabir Soumik","cross_cats":[],"headline":"Style bias dominates LLM judges at 0.76-0.92 strength across models, dwarfing position bias, while combined debiasing improves agreement for some judges by over 11 points.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-25T07:18:30Z","title":"Judging the Judges: A Systematic Evaluation of Bias Mitigation Strategies in LLM-as-a-Judge Pipelines"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.23178","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-08T08:11:19.583183Z","id":"6e8af9b5-9826-4494-a886-382390b93bb8","model_set":{"reader":"grok-4.3"},"one_line_summary":"Style bias dominates LLM-as-a-Judge systems far more than position bias, with debiasing strategies providing model-dependent gains and public tools released for replication.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Style bias dominates LLM judges at 0.76-0.92 strength across models, dwarfing position bias, while combined debiasing improves agreement for some judges by over 11 points.","strongest_claim":"Style bias is the dominant bias (0.76-0.92 across all models), far exceeding position bias (<= 0.04), yet has received minimal research attention.","weakest_assumption":"The chosen benchmarks (MT-Bench, LLMBar, and the custom set) and bias measurement protocols accurately isolate the targeted biases without introducing confounding artifacts from the test data itself or from how responses were generated."}},"verdict_id":"6e8af9b5-9826-4494-a886-382390b93bb8"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:025d41d37d4429ae4ebbec7eefd27d5e3c6d12e1c2ba3db368074b8d5866bda1","target":"record","created_at":"2026-06-25T01:18:38Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"cfc3a751c58c0db75e5635e0c86944b4554226db4975fafa682d72901497c7d7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-25T07:18:30Z","title_canon_sha256":"33773dbbee0db37c531b7fba373163b94d36dcd257911d480bd5ae8c2699e8e5"},"schema_version":"1.0","source":{"id":"2604.23178","kind":"arxiv","version":2}},"canonical_sha256":"a15ac84d9b16632408496a804abf017d37fff889dd71edf862751ed24cb6be94","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a15ac84d9b16632408496a804abf017d37fff889dd71edf862751ed24cb6be94","first_computed_at":"2026-06-25T01:18:38.181048Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-25T01:18:38.181048Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1rntRvxvvmEvq0RZerUN3zasDNFKvtaireEOYYvxMIlZN9hCXTHxKbdsXw+OXEebA6ErsFPOAmOjaiBgT+BzDA==","signature_status":"signed_v1","signed_at":"2026-06-25T01:18:38.181648Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.23178","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:025d41d37d4429ae4ebbec7eefd27d5e3c6d12e1c2ba3db368074b8d5866bda1","sha256:f3a9c64291407e42afe0a4824cda372fe7e5b9d678f0a990c3d929abe77ae9da"],"state_sha256":"31afe89d69f41e3473c16f311f5c2f3390a49a460fd263bd918dd05dd3d35523"}