{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:NSRCHRTBARTEFK7QU2UYGOHWXH","short_pith_number":"pith:NSRCHRTB","schema_version":"1.0","canonical_sha256":"6ca223c661046642abf0a6a98338f6b9d16667031e59f8800c857bee11ad3b6b","source":{"kind":"arxiv","id":"2503.23013","version":1},"attestation_state":"computed","paper":{"title":"DAT: Dynamic Alpha Tuning for Hybrid Retrieval in Retrieval-Augmented Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Hsin-Ling Hsu, Jengnan Tzeng","submitted_at":"2025-03-29T08:35:01Z","abstract_excerpt":"Hybrid retrieval techniques in Retrieval-Augmented Generation (RAG) systems enhance information retrieval by combining dense and sparse (e.g., BM25-based) retrieval methods. However, existing approaches struggle with adaptability, as fixed weighting schemes fail to adjust to different queries. To address this, we propose DAT (Dynamic Alpha Tuning), a novel hybrid retrieval framework that dynamically balances dense retrieval and BM25 for each query. DAT leverages a large language model (LLM) to evaluate the effectiveness of the top-1 results from both retrieval methods, assigning an effectivene"},"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":"2503.23013","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2025-03-29T08:35:01Z","cross_cats_sorted":[],"title_canon_sha256":"c2c2387184049908fce5ef0f295b9aac067aa53c97658ce2b5c639fd1332ae7c","abstract_canon_sha256":"ab16f7d3b49b838bf87101fb1578a6f4c9140a4621c176c3ce638b829319a769"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:41:29.001473Z","signature_b64":"II7RUJrxSmF5ex+dwx+3JarItdp1QRfC/fAzC4pbM+BKB6C3EIx/iwqwvAy7Ysq3VeXXlQ4TCcFiN7+PLhsTAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6ca223c661046642abf0a6a98338f6b9d16667031e59f8800c857bee11ad3b6b","last_reissued_at":"2026-07-05T10:41:29.000996Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:41:29.000996Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DAT: Dynamic Alpha Tuning for Hybrid Retrieval in Retrieval-Augmented Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Hsin-Ling Hsu, Jengnan Tzeng","submitted_at":"2025-03-29T08:35:01Z","abstract_excerpt":"Hybrid retrieval techniques in Retrieval-Augmented Generation (RAG) systems enhance information retrieval by combining dense and sparse (e.g., BM25-based) retrieval methods. However, existing approaches struggle with adaptability, as fixed weighting schemes fail to adjust to different queries. To address this, we propose DAT (Dynamic Alpha Tuning), a novel hybrid retrieval framework that dynamically balances dense retrieval and BM25 for each query. DAT leverages a large language model (LLM) to evaluate the effectiveness of the top-1 results from both retrieval methods, assigning an effectivene"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.23013","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/2503.23013/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":"2503.23013","created_at":"2026-07-05T10:41:29.001055+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.23013v1","created_at":"2026-07-05T10:41:29.001055+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.23013","created_at":"2026-07-05T10:41:29.001055+00:00"},{"alias_kind":"pith_short_12","alias_value":"NSRCHRTBARTE","created_at":"2026-07-05T10:41:29.001055+00:00"},{"alias_kind":"pith_short_16","alias_value":"NSRCHRTBARTEFK7Q","created_at":"2026-07-05T10:41:29.001055+00:00"},{"alias_kind":"pith_short_8","alias_value":"NSRCHRTB","created_at":"2026-07-05T10:41:29.001055+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.13141","citing_title":"Rethinking RAG in Long Videos: What to Retrieve and How to Use It?","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01240","citing_title":"Efficient RAG with Intent-Aware Retrieval and Semantics-Preserving Chunking","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30711","citing_title":"SAGE: A Novelty Gate for Efficient Memory Evolution in Agentic LLMs","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03676","citing_title":"Are LLM-Based Retrievers Worth Their Cost? An Empirical Study of Efficiency, Robustness, and Reasoning Overhead","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03986","citing_title":"From Intent to Execution: Composing Agentic Workflows with Agent Recommendation","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00631","citing_title":"H-RAG at SemEval-2026 Task 8: Hierarchical Parent-Child Retrieval for Multi-Turn RAG Conversations","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NSRCHRTBARTEFK7QU2UYGOHWXH","json":"https://pith.science/pith/NSRCHRTBARTEFK7QU2UYGOHWXH.json","graph_json":"https://pith.science/api/pith-number/NSRCHRTBARTEFK7QU2UYGOHWXH/graph.json","events_json":"https://pith.science/api/pith-number/NSRCHRTBARTEFK7QU2UYGOHWXH/events.json","paper":"https://pith.science/paper/NSRCHRTB"},"agent_actions":{"view_html":"https://pith.science/pith/NSRCHRTBARTEFK7QU2UYGOHWXH","download_json":"https://pith.science/pith/NSRCHRTBARTEFK7QU2UYGOHWXH.json","view_paper":"https://pith.science/paper/NSRCHRTB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.23013&json=true","fetch_graph":"https://pith.science/api/pith-number/NSRCHRTBARTEFK7QU2UYGOHWXH/graph.json","fetch_events":"https://pith.science/api/pith-number/NSRCHRTBARTEFK7QU2UYGOHWXH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NSRCHRTBARTEFK7QU2UYGOHWXH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NSRCHRTBARTEFK7QU2UYGOHWXH/action/storage_attestation","attest_author":"https://pith.science/pith/NSRCHRTBARTEFK7QU2UYGOHWXH/action/author_attestation","sign_citation":"https://pith.science/pith/NSRCHRTBARTEFK7QU2UYGOHWXH/action/citation_signature","submit_replication":"https://pith.science/pith/NSRCHRTBARTEFK7QU2UYGOHWXH/action/replication_record"}},"created_at":"2026-07-05T10:41:29.001055+00:00","updated_at":"2026-07-05T10:41:29.001055+00:00"}