{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:6YEBUQ3ASTHHD5SSOQA4ON3B2A","short_pith_number":"pith:6YEBUQ3A","schema_version":"1.0","canonical_sha256":"f6081a436094ce71f6527401c73761d01c48590bed8c6c2d719178cd59da96b9","source":{"kind":"arxiv","id":"2406.00231","version":2},"attestation_state":"computed","paper":{"title":"LLM-RankFusion: Mitigating Intrinsic Inconsistency in LLM-based Ranking","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.IR","authors_text":"Huazheng Wang, Lizhong Chen, Ojas Tendolkar, Qingyun Wu, Raymond Baartmans, Yifan Zeng","submitted_at":"2024-05-31T23:29:42Z","abstract_excerpt":"Ranking passages by prompting a large language model (LLM) can achieve promising performance in modern information retrieval (IR) systems. A common approach to sort the ranking list is by prompting LLMs for a pairwise or setwise comparison which often relies on sorting algorithms. However, sorting-based methods require consistent comparisons to correctly sort the passages, which we show that LLMs often violate. We identify two kinds of intrinsic inconsistency in LLM-based pairwise comparisons: order inconsistency which leads to conflicting results when switching the passage order, and transiti"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2406.00231","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2024-05-31T23:29:42Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"fb8655767a7ce86adfb7f78011a6ea87137b7275f0e0e86d35debee3e35a8cdc","abstract_canon_sha256":"9cfb274318e339da01057c0d2b75387238b7de866cb804ab0e007121ba118e16"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:40:17.727367Z","signature_b64":"2HCfy8g/LPuR4WEFuSA0BA4YgpxIPOmN/d71jWiYjlrnBbEz2EFmT6fLanDz/Z/fQ+UC2tOZJYThiqBk7IMtBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f6081a436094ce71f6527401c73761d01c48590bed8c6c2d719178cd59da96b9","last_reissued_at":"2026-07-05T09:40:17.726826Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:40:17.726826Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LLM-RankFusion: Mitigating Intrinsic Inconsistency in LLM-based Ranking","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.IR","authors_text":"Huazheng Wang, Lizhong Chen, Ojas Tendolkar, Qingyun Wu, Raymond Baartmans, Yifan Zeng","submitted_at":"2024-05-31T23:29:42Z","abstract_excerpt":"Ranking passages by prompting a large language model (LLM) can achieve promising performance in modern information retrieval (IR) systems. A common approach to sort the ranking list is by prompting LLMs for a pairwise or setwise comparison which often relies on sorting algorithms. However, sorting-based methods require consistent comparisons to correctly sort the passages, which we show that LLMs often violate. We identify two kinds of intrinsic inconsistency in LLM-based pairwise comparisons: order inconsistency which leads to conflicting results when switching the passage order, and transiti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.00231","kind":"arxiv","version":2},"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/2406.00231/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2406.00231","created_at":"2026-07-05T09:40:17.726894+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.00231v2","created_at":"2026-07-05T09:40:17.726894+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.00231","created_at":"2026-07-05T09:40:17.726894+00:00"},{"alias_kind":"pith_short_12","alias_value":"6YEBUQ3ASTHH","created_at":"2026-07-05T09:40:17.726894+00:00"},{"alias_kind":"pith_short_16","alias_value":"6YEBUQ3ASTHHD5SS","created_at":"2026-07-05T09:40:17.726894+00:00"},{"alias_kind":"pith_short_8","alias_value":"6YEBUQ3A","created_at":"2026-07-05T09:40:17.726894+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01782","citing_title":"Whole-Pool Setwise Reranking with Long-Context Language Models","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28353","citing_title":"From Regulatory Approvals to Patents: Cross-Domain Linking for Cardiovascular Device Traceability","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10150","citing_title":"Learning from Emptiness: De-biasing Listwise Rerankers with Content-Agnostic Probability Calibration","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6YEBUQ3ASTHHD5SSOQA4ON3B2A","json":"https://pith.science/pith/6YEBUQ3ASTHHD5SSOQA4ON3B2A.json","graph_json":"https://pith.science/api/pith-number/6YEBUQ3ASTHHD5SSOQA4ON3B2A/graph.json","events_json":"https://pith.science/api/pith-number/6YEBUQ3ASTHHD5SSOQA4ON3B2A/events.json","paper":"https://pith.science/paper/6YEBUQ3A"},"agent_actions":{"view_html":"https://pith.science/pith/6YEBUQ3ASTHHD5SSOQA4ON3B2A","download_json":"https://pith.science/pith/6YEBUQ3ASTHHD5SSOQA4ON3B2A.json","view_paper":"https://pith.science/paper/6YEBUQ3A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.00231&json=true","fetch_graph":"https://pith.science/api/pith-number/6YEBUQ3ASTHHD5SSOQA4ON3B2A/graph.json","fetch_events":"https://pith.science/api/pith-number/6YEBUQ3ASTHHD5SSOQA4ON3B2A/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6YEBUQ3ASTHHD5SSOQA4ON3B2A/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6YEBUQ3ASTHHD5SSOQA4ON3B2A/action/storage_attestation","attest_author":"https://pith.science/pith/6YEBUQ3ASTHHD5SSOQA4ON3B2A/action/author_attestation","sign_citation":"https://pith.science/pith/6YEBUQ3ASTHHD5SSOQA4ON3B2A/action/citation_signature","submit_replication":"https://pith.science/pith/6YEBUQ3ASTHHD5SSOQA4ON3B2A/action/replication_record"}},"created_at":"2026-07-05T09:40:17.726894+00:00","updated_at":"2026-07-05T09:40:17.726894+00:00"}