{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:AXXEKI4Z6XTDGZANIJXJEMUGLD","short_pith_number":"pith:AXXEKI4Z","canonical_record":{"source":{"id":"2405.02743","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-04T19:53:03Z","cross_cats_sorted":[],"title_canon_sha256":"aa9ee52c949ed04be34ba03cef15ca54d8fa3f1666765cc246b93f45c377f889","abstract_canon_sha256":"4476f03ff6f6eaddc036ff44f9d10e4dd0f807b5a4756d6aa5d2ea719684c4a4"},"schema_version":"1.0"},"canonical_sha256":"05ee452399f5e633640d426e92328658f5e529765e2911c1ffa7587ac29aede0","source":{"kind":"arxiv","id":"2405.02743","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.02743","created_at":"2026-07-05T08:15:53Z"},{"alias_kind":"arxiv_version","alias_value":"2405.02743v1","created_at":"2026-07-05T08:15:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.02743","created_at":"2026-07-05T08:15:53Z"},{"alias_kind":"pith_short_12","alias_value":"AXXEKI4Z6XTD","created_at":"2026-07-05T08:15:53Z"},{"alias_kind":"pith_short_16","alias_value":"AXXEKI4Z6XTDGZAN","created_at":"2026-07-05T08:15:53Z"},{"alias_kind":"pith_short_8","alias_value":"AXXEKI4Z","created_at":"2026-07-05T08:15:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:AXXEKI4Z6XTDGZANIJXJEMUGLD","target":"record","payload":{"canonical_record":{"source":{"id":"2405.02743","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-04T19:53:03Z","cross_cats_sorted":[],"title_canon_sha256":"aa9ee52c949ed04be34ba03cef15ca54d8fa3f1666765cc246b93f45c377f889","abstract_canon_sha256":"4476f03ff6f6eaddc036ff44f9d10e4dd0f807b5a4756d6aa5d2ea719684c4a4"},"schema_version":"1.0"},"canonical_sha256":"05ee452399f5e633640d426e92328658f5e529765e2911c1ffa7587ac29aede0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:15:53.215437Z","signature_b64":"XUpGhwaXaINChbOwSHXufTnMbX+sz6y3r9EeCsbMRp/K27w/zW9SOj03cPCx7HG38D4F+WTiniglrw9QCsjtBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"05ee452399f5e633640d426e92328658f5e529765e2911c1ffa7587ac29aede0","last_reissued_at":"2026-07-05T08:15:53.214927Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:15:53.214927Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.02743","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:15:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JRWE3GgMqXl3ve6Mm8UJOK99h1bonUEJvNuAQBnJ9TMTnAdZhObIJs1+IRIkV8T7tria6v+BmipgOOUQQCoODw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T10:57:27.391662Z"},"content_sha256":"3ead0b1aa7bfcccbd4fc29518e4c73f2ec046949391e9182cb83cdc8f3ddd143","schema_version":"1.0","event_id":"sha256:3ead0b1aa7bfcccbd4fc29518e4c73f2ec046949391e9182cb83cdc8f3ddd143"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:AXXEKI4Z6XTDGZANIJXJEMUGLD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Beyond Performance: Quantifying and Mitigating Label Bias in LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Roy Schwartz, Yuval Reif","submitted_at":"2024-05-04T19:53:03Z","abstract_excerpt":"Large language models (LLMs) have shown remarkable adaptability to diverse tasks, by leveraging context prompts containing instructions, or minimal input-output examples. However, recent work revealed they also exhibit label bias -- an undesirable preference toward predicting certain answers over others. Still, detecting and measuring this bias reliably and at scale has remained relatively unexplored. In this study, we evaluate different approaches to quantifying label bias in a model's predictions, conducting a comprehensive investigation across 279 classification tasks and ten LLMs. Our inve"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.02743","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2405.02743/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:15:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zS/9C72+zS8qCKxU2iPV92oEac5gUbHCIJG0KV/CVY5ERDrVS9CZNS6G9oO/Az/GS5dVA7X6maG0KoyR6/bUBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T10:57:27.394063Z"},"content_sha256":"04516394e647eecce1174090a9dea6440db5a08e74f4df0776a984ba1d1ccc32","schema_version":"1.0","event_id":"sha256:04516394e647eecce1174090a9dea6440db5a08e74f4df0776a984ba1d1ccc32"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AXXEKI4Z6XTDGZANIJXJEMUGLD/bundle.json","state_url":"https://pith.science/pith/AXXEKI4Z6XTDGZANIJXJEMUGLD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AXXEKI4Z6XTDGZANIJXJEMUGLD/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-22T10:57:27Z","links":{"resolver":"https://pith.science/pith/AXXEKI4Z6XTDGZANIJXJEMUGLD","bundle":"https://pith.science/pith/AXXEKI4Z6XTDGZANIJXJEMUGLD/bundle.json","state":"https://pith.science/pith/AXXEKI4Z6XTDGZANIJXJEMUGLD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AXXEKI4Z6XTDGZANIJXJEMUGLD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:AXXEKI4Z6XTDGZANIJXJEMUGLD","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":"4476f03ff6f6eaddc036ff44f9d10e4dd0f807b5a4756d6aa5d2ea719684c4a4","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-04T19:53:03Z","title_canon_sha256":"aa9ee52c949ed04be34ba03cef15ca54d8fa3f1666765cc246b93f45c377f889"},"schema_version":"1.0","source":{"id":"2405.02743","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.02743","created_at":"2026-07-05T08:15:53Z"},{"alias_kind":"arxiv_version","alias_value":"2405.02743v1","created_at":"2026-07-05T08:15:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.02743","created_at":"2026-07-05T08:15:53Z"},{"alias_kind":"pith_short_12","alias_value":"AXXEKI4Z6XTD","created_at":"2026-07-05T08:15:53Z"},{"alias_kind":"pith_short_16","alias_value":"AXXEKI4Z6XTDGZAN","created_at":"2026-07-05T08:15:53Z"},{"alias_kind":"pith_short_8","alias_value":"AXXEKI4Z","created_at":"2026-07-05T08:15:53Z"}],"graph_snapshots":[{"event_id":"sha256:04516394e647eecce1174090a9dea6440db5a08e74f4df0776a984ba1d1ccc32","target":"graph","created_at":"2026-07-05T08:15:53Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2405.02743/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have shown remarkable adaptability to diverse tasks, by leveraging context prompts containing instructions, or minimal input-output examples. However, recent work revealed they also exhibit label bias -- an undesirable preference toward predicting certain answers over others. Still, detecting and measuring this bias reliably and at scale has remained relatively unexplored. In this study, we evaluate different approaches to quantifying label bias in a model's predictions, conducting a comprehensive investigation across 279 classification tasks and ten LLMs. Our inve","authors_text":"Roy Schwartz, Yuval Reif","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-04T19:53:03Z","title":"Beyond Performance: Quantifying and Mitigating Label Bias in LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.02743","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3ead0b1aa7bfcccbd4fc29518e4c73f2ec046949391e9182cb83cdc8f3ddd143","target":"record","created_at":"2026-07-05T08:15:53Z","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":"4476f03ff6f6eaddc036ff44f9d10e4dd0f807b5a4756d6aa5d2ea719684c4a4","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-04T19:53:03Z","title_canon_sha256":"aa9ee52c949ed04be34ba03cef15ca54d8fa3f1666765cc246b93f45c377f889"},"schema_version":"1.0","source":{"id":"2405.02743","kind":"arxiv","version":1}},"canonical_sha256":"05ee452399f5e633640d426e92328658f5e529765e2911c1ffa7587ac29aede0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"05ee452399f5e633640d426e92328658f5e529765e2911c1ffa7587ac29aede0","first_computed_at":"2026-07-05T08:15:53.214927Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:15:53.214927Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XUpGhwaXaINChbOwSHXufTnMbX+sz6y3r9EeCsbMRp/K27w/zW9SOj03cPCx7HG38D4F+WTiniglrw9QCsjtBg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:15:53.215437Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.02743","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3ead0b1aa7bfcccbd4fc29518e4c73f2ec046949391e9182cb83cdc8f3ddd143","sha256:04516394e647eecce1174090a9dea6440db5a08e74f4df0776a984ba1d1ccc32"],"state_sha256":"5b75512601f68e2c3751812e869be7e0691ac68dfbcfaf6ab58ba8cce37fe654"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RLwoAvMEF21TTcpO+dBsp0HtMBm5DlUCxb2SY5IV/nCU3dLshzXDIl50YCzOgwl+2hafe9Gy+VgfH0gqZZs/BQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-22T10:57:27.398113Z","bundle_sha256":"32e871ecc1076df6a2c93ecfcd1b43208a5b226024e6cb5cf6e6f7d3a01cbde1"}}