{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OBHMWDBBNGTKGILI466LCN5KNI","short_pith_number":"pith:OBHMWDBB","schema_version":"1.0","canonical_sha256":"704ecb0c2169a6a32168e7bcb137aa6a017d2bee9bdf6f8f0b01729a89719225","source":{"kind":"arxiv","id":"2503.19092","version":2},"attestation_state":"computed","paper":{"title":"Rankers, Judges, and Assistants: Towards Understanding the Interplay of LLMs in Information Retrieval Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.IR","authors_text":"Donald Metzler, Krisztian Balog, Zhen Qin","submitted_at":"2025-03-24T19:24:40Z","abstract_excerpt":"Large language models (LLMs) are increasingly integral to information retrieval (IR), powering ranking, evaluation, and AI-assisted content creation. This widespread adoption necessitates a critical examination of potential biases arising from the interplay between these LLM-based components. This paper synthesizes existing research and presents novel experiment designs that explore how LLM-based rankers and assistants influence LLM-based judges. We provide the first empirical evidence of LLM judges exhibiting significant bias towards LLM-based rankers. Furthermore, we observe limitations in L"},"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":"2503.19092","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-03-24T19:24:40Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"e12d709d020500e476fa49d536fb2c71502e2a3372a1bfc9ee1c68a335afeb47","abstract_canon_sha256":"cb4d55b841150896fa8364613522b41a01d17f946d568050813f5f130cbedd85"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:34:49.177767Z","signature_b64":"Qne0ocziWUKQMCHBKvWCKFGDWRcgQygrew9wWJQIdBj/LclzUwi6MDT+nZEYcy2RHmyTiK/qoNoWVszt2/iGCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"704ecb0c2169a6a32168e7bcb137aa6a017d2bee9bdf6f8f0b01729a89719225","last_reissued_at":"2026-07-05T11:34:49.177262Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:34:49.177262Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Rankers, Judges, and Assistants: Towards Understanding the Interplay of LLMs in Information Retrieval Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.IR","authors_text":"Donald Metzler, Krisztian Balog, Zhen Qin","submitted_at":"2025-03-24T19:24:40Z","abstract_excerpt":"Large language models (LLMs) are increasingly integral to information retrieval (IR), powering ranking, evaluation, and AI-assisted content creation. This widespread adoption necessitates a critical examination of potential biases arising from the interplay between these LLM-based components. This paper synthesizes existing research and presents novel experiment designs that explore how LLM-based rankers and assistants influence LLM-based judges. We provide the first empirical evidence of LLM judges exhibiting significant bias towards LLM-based rankers. Furthermore, we observe limitations in L"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.19092","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/2503.19092/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":"2503.19092","created_at":"2026-07-05T11:34:49.177323+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.19092v2","created_at":"2026-07-05T11:34:49.177323+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.19092","created_at":"2026-07-05T11:34:49.177323+00:00"},{"alias_kind":"pith_short_12","alias_value":"OBHMWDBBNGTK","created_at":"2026-07-05T11:34:49.177323+00:00"},{"alias_kind":"pith_short_16","alias_value":"OBHMWDBBNGTKGILI","created_at":"2026-07-05T11:34:49.177323+00:00"},{"alias_kind":"pith_short_8","alias_value":"OBHMWDBB","created_at":"2026-07-05T11:34:49.177323+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.25032","citing_title":"Offline Evaluation Measures of Fairness in Recommender Systems","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16706","citing_title":"Evaluating Tool-Using Language Agents: Judge Reliability, Propagation Cascades, and Runtime Mitigation in AgentProp-Bench","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OBHMWDBBNGTKGILI466LCN5KNI","json":"https://pith.science/pith/OBHMWDBBNGTKGILI466LCN5KNI.json","graph_json":"https://pith.science/api/pith-number/OBHMWDBBNGTKGILI466LCN5KNI/graph.json","events_json":"https://pith.science/api/pith-number/OBHMWDBBNGTKGILI466LCN5KNI/events.json","paper":"https://pith.science/paper/OBHMWDBB"},"agent_actions":{"view_html":"https://pith.science/pith/OBHMWDBBNGTKGILI466LCN5KNI","download_json":"https://pith.science/pith/OBHMWDBBNGTKGILI466LCN5KNI.json","view_paper":"https://pith.science/paper/OBHMWDBB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.19092&json=true","fetch_graph":"https://pith.science/api/pith-number/OBHMWDBBNGTKGILI466LCN5KNI/graph.json","fetch_events":"https://pith.science/api/pith-number/OBHMWDBBNGTKGILI466LCN5KNI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OBHMWDBBNGTKGILI466LCN5KNI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OBHMWDBBNGTKGILI466LCN5KNI/action/storage_attestation","attest_author":"https://pith.science/pith/OBHMWDBBNGTKGILI466LCN5KNI/action/author_attestation","sign_citation":"https://pith.science/pith/OBHMWDBBNGTKGILI466LCN5KNI/action/citation_signature","submit_replication":"https://pith.science/pith/OBHMWDBBNGTKGILI466LCN5KNI/action/replication_record"}},"created_at":"2026-07-05T11:34:49.177323+00:00","updated_at":"2026-07-05T11:34:49.177323+00:00"}