{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:DV6WN7FGM5XKP6LL52ALZNHNG6","short_pith_number":"pith:DV6WN7FG","schema_version":"1.0","canonical_sha256":"1d7d66fca6676ea7f96bee80bcb4ed37b9f97b22b868e7dd00549a1d29c287e1","source":{"kind":"arxiv","id":"2410.07589","version":1},"attestation_state":"computed","paper":{"title":"No Free Lunch: Retrieval-Augmented Generation Undermines Fairness in LLMs, Even for Vigilant Users","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.IR","authors_text":"Daiqing Qi, Dongliang Guo, Hongyi Wu, Mengxuan Hu, Ronghang Zhu, Sheng Li, Zihan Guan","submitted_at":"2024-10-10T03:51:58Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) is widely adopted for its effectiveness and cost-efficiency in mitigating hallucinations and enhancing the domain-specific generation capabilities of large language models (LLMs). However, is this effectiveness and cost-efficiency truly a free lunch? In this study, we comprehensively investigate the fairness costs associated with RAG by proposing a practical three-level threat model from the perspective of user awareness of fairness. Specifically, varying levels of user fairness awareness result in different degrees of fairness censorship on the external da"},"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":"2410.07589","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2024-10-10T03:51:58Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"f2aa25e1e779d4fec57ce1fb11f8c5e397d09bf88b5241da8c0d0938e5a8b37a","abstract_canon_sha256":"dcad1f0b7b27f8ae35ec29278a7e5ec2409767ce6c9a0a6e30e7a9e64f8a8977"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:18:40.045166Z","signature_b64":"9Ibk4MzkzrvIHkLEdNoMF2xXPaUoZDHax6gOw3/rJC+uX1Kz7b3sC3Rkzf119uww4N9z0CQFsGry7GKP1MjIAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1d7d66fca6676ea7f96bee80bcb4ed37b9f97b22b868e7dd00549a1d29c287e1","last_reissued_at":"2026-07-05T09:18:40.044559Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:18:40.044559Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"No Free Lunch: Retrieval-Augmented Generation Undermines Fairness in LLMs, Even for Vigilant Users","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.IR","authors_text":"Daiqing Qi, Dongliang Guo, Hongyi Wu, Mengxuan Hu, Ronghang Zhu, Sheng Li, Zihan Guan","submitted_at":"2024-10-10T03:51:58Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) is widely adopted for its effectiveness and cost-efficiency in mitigating hallucinations and enhancing the domain-specific generation capabilities of large language models (LLMs). However, is this effectiveness and cost-efficiency truly a free lunch? In this study, we comprehensively investigate the fairness costs associated with RAG by proposing a practical three-level threat model from the perspective of user awareness of fairness. Specifically, varying levels of user fairness awareness result in different degrees of fairness censorship on the external da"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.07589","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/2410.07589/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":"2410.07589","created_at":"2026-07-05T09:18:40.044613+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.07589v1","created_at":"2026-07-05T09:18:40.044613+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.07589","created_at":"2026-07-05T09:18:40.044613+00:00"},{"alias_kind":"pith_short_12","alias_value":"DV6WN7FGM5XK","created_at":"2026-07-05T09:18:40.044613+00:00"},{"alias_kind":"pith_short_16","alias_value":"DV6WN7FGM5XKP6LL","created_at":"2026-07-05T09:18:40.044613+00:00"},{"alias_kind":"pith_short_8","alias_value":"DV6WN7FG","created_at":"2026-07-05T09:18:40.044613+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.28343","citing_title":"The Crowded Embedding Space: A Mean-Field Mechanism for Emergent Marginalization in Retrieval-Augmented Agents","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15790","citing_title":"Fairness-Aware Retrieval Optimization for Retrieval-Augmented Generation","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DV6WN7FGM5XKP6LL52ALZNHNG6","json":"https://pith.science/pith/DV6WN7FGM5XKP6LL52ALZNHNG6.json","graph_json":"https://pith.science/api/pith-number/DV6WN7FGM5XKP6LL52ALZNHNG6/graph.json","events_json":"https://pith.science/api/pith-number/DV6WN7FGM5XKP6LL52ALZNHNG6/events.json","paper":"https://pith.science/paper/DV6WN7FG"},"agent_actions":{"view_html":"https://pith.science/pith/DV6WN7FGM5XKP6LL52ALZNHNG6","download_json":"https://pith.science/pith/DV6WN7FGM5XKP6LL52ALZNHNG6.json","view_paper":"https://pith.science/paper/DV6WN7FG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.07589&json=true","fetch_graph":"https://pith.science/api/pith-number/DV6WN7FGM5XKP6LL52ALZNHNG6/graph.json","fetch_events":"https://pith.science/api/pith-number/DV6WN7FGM5XKP6LL52ALZNHNG6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DV6WN7FGM5XKP6LL52ALZNHNG6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DV6WN7FGM5XKP6LL52ALZNHNG6/action/storage_attestation","attest_author":"https://pith.science/pith/DV6WN7FGM5XKP6LL52ALZNHNG6/action/author_attestation","sign_citation":"https://pith.science/pith/DV6WN7FGM5XKP6LL52ALZNHNG6/action/citation_signature","submit_replication":"https://pith.science/pith/DV6WN7FGM5XKP6LL52ALZNHNG6/action/replication_record"}},"created_at":"2026-07-05T09:18:40.044613+00:00","updated_at":"2026-07-05T09:18:40.044613+00:00"}