{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:STZFCDUH5TNXNPZUZ7NSBIF4CH","short_pith_number":"pith:STZFCDUH","schema_version":"1.0","canonical_sha256":"94f2510e87ecdb76bf34cfdb20a0bc11d2c836cabdfb1038a2ad3f383d25c4c7","source":{"kind":"arxiv","id":"2502.11228","version":2},"attestation_state":"computed","paper":{"title":"Vendi-RAG: Adaptively Trading-Off Diversity And Quality Significantly Improves Retrieval Augmented Generation With LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Adji Bousso Dieng, Mohammad Reza Rezaei","submitted_at":"2025-02-16T18:46:10Z","abstract_excerpt":"Retrieval-augmented generation (RAG) enhances large language models (LLMs) for domain-specific question-answering (QA) tasks by leveraging external knowledge sources. However, traditional RAG systems primarily focus on relevance-based retrieval and often struggle with redundancy, especially when reasoning requires connecting information from multiple sources. This paper introduces Vendi-RAG, a framework based on an iterative process that jointly optimizes retrieval diversity and answer quality. This joint optimization leads to significantly higher accuracy for multi-hop QA tasks. Vendi-RAG lev"},"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":"2502.11228","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-16T18:46:10Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d6dfe6628e31c92be950a7ba8eb72b6a4c8090c907d1eb893345357d623b8d72","abstract_canon_sha256":"8601584919687a08bcf20a5e10efd30d4c90a2fbc42f5ceb091c54db8fbb097d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:16.599611Z","signature_b64":"+0sa99dDfn5yEc6j+Lf31TrJ2jrKEVACPeS5K01key3+MEby2VlRb9AKq3iKjD3h7e4ensPDqyCx+LjGgtjgCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"94f2510e87ecdb76bf34cfdb20a0bc11d2c836cabdfb1038a2ad3f383d25c4c7","last_reissued_at":"2026-07-05T11:08:16.599090Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:16.599090Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Vendi-RAG: Adaptively Trading-Off Diversity And Quality Significantly Improves Retrieval Augmented Generation With LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Adji Bousso Dieng, Mohammad Reza Rezaei","submitted_at":"2025-02-16T18:46:10Z","abstract_excerpt":"Retrieval-augmented generation (RAG) enhances large language models (LLMs) for domain-specific question-answering (QA) tasks by leveraging external knowledge sources. However, traditional RAG systems primarily focus on relevance-based retrieval and often struggle with redundancy, especially when reasoning requires connecting information from multiple sources. This paper introduces Vendi-RAG, a framework based on an iterative process that jointly optimizes retrieval diversity and answer quality. This joint optimization leads to significantly higher accuracy for multi-hop QA tasks. Vendi-RAG lev"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.11228","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/2502.11228/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":"2502.11228","created_at":"2026-07-05T11:08:16.599157+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.11228v2","created_at":"2026-07-05T11:08:16.599157+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.11228","created_at":"2026-07-05T11:08:16.599157+00:00"},{"alias_kind":"pith_short_12","alias_value":"STZFCDUH5TNX","created_at":"2026-07-05T11:08:16.599157+00:00"},{"alias_kind":"pith_short_16","alias_value":"STZFCDUH5TNXNPZU","created_at":"2026-07-05T11:08:16.599157+00:00"},{"alias_kind":"pith_short_8","alias_value":"STZFCDUH","created_at":"2026-07-05T11:08:16.599157+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17910","citing_title":"Non-negative Elastic Net Decoding for Information Retrieval","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2507.21934","citing_title":"Culinary Crossroads: A RAG Framework for Enhancing Diversity in Cross-Cultural Recipe Adaptation","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08046","citing_title":"Guaranteeing Knowledge Integration with Joint Decoding for Retrieval-Augmented Generation","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/STZFCDUH5TNXNPZUZ7NSBIF4CH","json":"https://pith.science/pith/STZFCDUH5TNXNPZUZ7NSBIF4CH.json","graph_json":"https://pith.science/api/pith-number/STZFCDUH5TNXNPZUZ7NSBIF4CH/graph.json","events_json":"https://pith.science/api/pith-number/STZFCDUH5TNXNPZUZ7NSBIF4CH/events.json","paper":"https://pith.science/paper/STZFCDUH"},"agent_actions":{"view_html":"https://pith.science/pith/STZFCDUH5TNXNPZUZ7NSBIF4CH","download_json":"https://pith.science/pith/STZFCDUH5TNXNPZUZ7NSBIF4CH.json","view_paper":"https://pith.science/paper/STZFCDUH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.11228&json=true","fetch_graph":"https://pith.science/api/pith-number/STZFCDUH5TNXNPZUZ7NSBIF4CH/graph.json","fetch_events":"https://pith.science/api/pith-number/STZFCDUH5TNXNPZUZ7NSBIF4CH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/STZFCDUH5TNXNPZUZ7NSBIF4CH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/STZFCDUH5TNXNPZUZ7NSBIF4CH/action/storage_attestation","attest_author":"https://pith.science/pith/STZFCDUH5TNXNPZUZ7NSBIF4CH/action/author_attestation","sign_citation":"https://pith.science/pith/STZFCDUH5TNXNPZUZ7NSBIF4CH/action/citation_signature","submit_replication":"https://pith.science/pith/STZFCDUH5TNXNPZUZ7NSBIF4CH/action/replication_record"}},"created_at":"2026-07-05T11:08:16.599157+00:00","updated_at":"2026-07-05T11:08:16.599157+00:00"}