{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:MH3MLOFOS6MSMHE74D4KQRNPN4","short_pith_number":"pith:MH3MLOFO","schema_version":"1.0","canonical_sha256":"61f6c5b8ae9799261c9fe0f8a845af6f0dc8d94ffe3292a5b838202970ebf332","source":{"kind":"arxiv","id":"2306.17563","version":2},"attestation_state":"computed","paper":{"title":"Large Language Models are Effective Text Rankers with Pairwise Ranking Prompting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.IR","authors_text":"Donald Metzler, Honglei Zhuang, Jialu Liu, Jiaming Shen, Junru Wu, Kai Hui, Le Yan, Michael Bendersky, Rolf Jagerman, Tianqi Liu, Xuanhui Wang, Zhen Qin","submitted_at":"2023-06-30T11:32:25Z","abstract_excerpt":"Ranking documents using Large Language Models (LLMs) by directly feeding the query and candidate documents into the prompt is an interesting and practical problem. However, researchers have found it difficult to outperform fine-tuned baseline rankers on benchmark datasets. We analyze pointwise and listwise ranking prompts used by existing methods and argue that off-the-shelf LLMs do not fully understand these challenging ranking formulations. In this paper, we propose to significantly reduce the burden on LLMs by using a new technique called Pairwise Ranking Prompting (PRP). Our results are th"},"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":"2306.17563","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2023-06-30T11:32:25Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"0ab79d2c46ee0caab868e8b66ef37be647e55d8a6ecc44864658fbde748ea1d6","abstract_canon_sha256":"75deb7bbde7e914b482d208925ac683217f95f4df1cf719216c145d4d455a1dd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:01:31.165232Z","signature_b64":"AYyr++wSEkmJoRZhBbR6P0eYhyDdfFu22cmQBWAAPn/n9Ra9KGfJc89R1zG3elI41mfBwVOqsCT0X+jScR1jCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"61f6c5b8ae9799261c9fe0f8a845af6f0dc8d94ffe3292a5b838202970ebf332","last_reissued_at":"2026-07-05T08:01:31.164783Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:01:31.164783Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Large Language Models are Effective Text Rankers with Pairwise Ranking Prompting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.IR","authors_text":"Donald Metzler, Honglei Zhuang, Jialu Liu, Jiaming Shen, Junru Wu, Kai Hui, Le Yan, Michael Bendersky, Rolf Jagerman, Tianqi Liu, Xuanhui Wang, Zhen Qin","submitted_at":"2023-06-30T11:32:25Z","abstract_excerpt":"Ranking documents using Large Language Models (LLMs) by directly feeding the query and candidate documents into the prompt is an interesting and practical problem. However, researchers have found it difficult to outperform fine-tuned baseline rankers on benchmark datasets. We analyze pointwise and listwise ranking prompts used by existing methods and argue that off-the-shelf LLMs do not fully understand these challenging ranking formulations. In this paper, we propose to significantly reduce the burden on LLMs by using a new technique called Pairwise Ranking Prompting (PRP). Our results are th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.17563","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/2306.17563/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":"2306.17563","created_at":"2026-07-05T08:01:31.164839+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.17563v2","created_at":"2026-07-05T08:01:31.164839+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.17563","created_at":"2026-07-05T08:01:31.164839+00:00"},{"alias_kind":"pith_short_12","alias_value":"MH3MLOFOS6MS","created_at":"2026-07-05T08:01:31.164839+00:00"},{"alias_kind":"pith_short_16","alias_value":"MH3MLOFOS6MSMHE7","created_at":"2026-07-05T08:01:31.164839+00:00"},{"alias_kind":"pith_short_8","alias_value":"MH3MLOFO","created_at":"2026-07-05T08:01:31.164839+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05968","citing_title":"InfluMatch: Frontier-Quality KOL Search at 4B-Model Cost","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2605.27429","citing_title":"Ocean4Rec: Offline LLM-Derived OCEAN Profiles for Request-Time VOD Reranking","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27810","citing_title":"LRanker: LLM Ranker for Massive Candidates","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2502.00709","citing_title":"RankFlow: A Multi-Role Collaborative Reranking Workflow Utilizing Large Language Models","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2509.00303","citing_title":"Access Paths for Efficient Ordering with Large Language Models","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2310.02170","citing_title":"A Dynamic LLM-Powered Agent Network for Task-Oriented Agent Collaboration","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15299","citing_title":"Fortress: A Case Study in Stabilizing Search Recommendations via Temporal Data Augmentation and Feature Pruning","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2506.01481","citing_title":"TSGuard: Automated User-Centric Incident Diagnosis for AI Workloads in the Cloud","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2506.14092","citing_title":"Fragile Preferences: A Deep Dive Into Order Effects in Large Language Models","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2508.07050","citing_title":"ReasonRank: Empowering Passage Ranking with Strong Reasoning Ability","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2312.02724","citing_title":"RankZephyr: Effective and Robust Zero-Shot Listwise Reranking is a Breeze!","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26483","citing_title":"Efficient Listwise Reranking with Compressed Document Representations","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2412.05579","citing_title":"LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods","ref_index":186,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22180","citing_title":"ResRank: Unifying Retrieval and Listwise Reranking via End-to-End Joint Training with Residual Passage Compression","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15621","citing_title":"Rethinking the Necessity of Adaptive Retrieval-Augmented Generation through the Lens of Adaptive Listwise Ranking","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MH3MLOFOS6MSMHE74D4KQRNPN4","json":"https://pith.science/pith/MH3MLOFOS6MSMHE74D4KQRNPN4.json","graph_json":"https://pith.science/api/pith-number/MH3MLOFOS6MSMHE74D4KQRNPN4/graph.json","events_json":"https://pith.science/api/pith-number/MH3MLOFOS6MSMHE74D4KQRNPN4/events.json","paper":"https://pith.science/paper/MH3MLOFO"},"agent_actions":{"view_html":"https://pith.science/pith/MH3MLOFOS6MSMHE74D4KQRNPN4","download_json":"https://pith.science/pith/MH3MLOFOS6MSMHE74D4KQRNPN4.json","view_paper":"https://pith.science/paper/MH3MLOFO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.17563&json=true","fetch_graph":"https://pith.science/api/pith-number/MH3MLOFOS6MSMHE74D4KQRNPN4/graph.json","fetch_events":"https://pith.science/api/pith-number/MH3MLOFOS6MSMHE74D4KQRNPN4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MH3MLOFOS6MSMHE74D4KQRNPN4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MH3MLOFOS6MSMHE74D4KQRNPN4/action/storage_attestation","attest_author":"https://pith.science/pith/MH3MLOFOS6MSMHE74D4KQRNPN4/action/author_attestation","sign_citation":"https://pith.science/pith/MH3MLOFOS6MSMHE74D4KQRNPN4/action/citation_signature","submit_replication":"https://pith.science/pith/MH3MLOFOS6MSMHE74D4KQRNPN4/action/replication_record"}},"created_at":"2026-07-05T08:01:31.164839+00:00","updated_at":"2026-07-05T08:01:31.164839+00:00"}