{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:3HVBDNT45GM6VTQHA672SMBWAO","short_pith_number":"pith:3HVBDNT4","schema_version":"1.0","canonical_sha256":"d9ea11b67ce999eace0707bfa930360383ba027acf1851629a9c199a97f620f8","source":{"kind":"arxiv","id":"2502.18460","version":2},"attestation_state":"computed","paper":{"title":"DRAMA: Diverse Augmentation from Large Language Models to Smaller Dense Retrievers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Barlas Oguz, Jimmy Lin, Wen-tau Yih, Xilun Chen, Xi Victoria Lin, Xueguang Ma","submitted_at":"2025-02-25T18:59:07Z","abstract_excerpt":"Large language models (LLMs) have demonstrated strong effectiveness and robustness while fine-tuned as dense retrievers. However, their large parameter size brings significant inference time computational challenges, including high encoding costs for large-scale corpora and increased query latency, limiting their practical deployment. While smaller retrievers offer better efficiency, they often fail to generalize effectively with limited supervised fine-tuning data. In this work, we introduce DRAMA, a training framework that leverages LLMs to train smaller generalizable dense retrievers. In pa"},"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":"2502.18460","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-25T18:59:07Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"2f1eaaa95e2a6653d7ef5b9e258fbc60ae385bf8944fd0b2884735aef4e6e49f","abstract_canon_sha256":"864f7a2a196774d2726d7736184f7f9127e760726fcb0ec785d3cd2fa2bd2557"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:54.455565Z","signature_b64":"htHKJEwpSjwti0i0FBrIhMOJN9NqdCbwecy2+ZnI1+s6Owf+gKBqhANIOwAMQYxV1n+DIQHH2XpIUiv22AvgCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d9ea11b67ce999eace0707bfa930360383ba027acf1851629a9c199a97f620f8","last_reissued_at":"2026-07-05T11:14:54.455118Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:54.455118Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DRAMA: Diverse Augmentation from Large Language Models to Smaller Dense Retrievers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Barlas Oguz, Jimmy Lin, Wen-tau Yih, Xilun Chen, Xi Victoria Lin, Xueguang Ma","submitted_at":"2025-02-25T18:59:07Z","abstract_excerpt":"Large language models (LLMs) have demonstrated strong effectiveness and robustness while fine-tuned as dense retrievers. However, their large parameter size brings significant inference time computational challenges, including high encoding costs for large-scale corpora and increased query latency, limiting their practical deployment. While smaller retrievers offer better efficiency, they often fail to generalize effectively with limited supervised fine-tuning data. In this work, we introduce DRAMA, a training framework that leverages LLMs to train smaller generalizable dense retrievers. In pa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.18460","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/2502.18460/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":"2502.18460","created_at":"2026-07-05T11:14:54.455171+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.18460v2","created_at":"2026-07-05T11:14:54.455171+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.18460","created_at":"2026-07-05T11:14:54.455171+00:00"},{"alias_kind":"pith_short_12","alias_value":"3HVBDNT45GM6","created_at":"2026-07-05T11:14:54.455171+00:00"},{"alias_kind":"pith_short_16","alias_value":"3HVBDNT45GM6VTQH","created_at":"2026-07-05T11:14:54.455171+00:00"},{"alias_kind":"pith_short_8","alias_value":"3HVBDNT4","created_at":"2026-07-05T11:14:54.455171+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.11141","citing_title":"Reducing Hallucination in Enterprise AI Workflows via Hybrid Utility Minimum Bayes Risk (HUMBR)","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3HVBDNT45GM6VTQHA672SMBWAO","json":"https://pith.science/pith/3HVBDNT45GM6VTQHA672SMBWAO.json","graph_json":"https://pith.science/api/pith-number/3HVBDNT45GM6VTQHA672SMBWAO/graph.json","events_json":"https://pith.science/api/pith-number/3HVBDNT45GM6VTQHA672SMBWAO/events.json","paper":"https://pith.science/paper/3HVBDNT4"},"agent_actions":{"view_html":"https://pith.science/pith/3HVBDNT45GM6VTQHA672SMBWAO","download_json":"https://pith.science/pith/3HVBDNT45GM6VTQHA672SMBWAO.json","view_paper":"https://pith.science/paper/3HVBDNT4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.18460&json=true","fetch_graph":"https://pith.science/api/pith-number/3HVBDNT45GM6VTQHA672SMBWAO/graph.json","fetch_events":"https://pith.science/api/pith-number/3HVBDNT45GM6VTQHA672SMBWAO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3HVBDNT45GM6VTQHA672SMBWAO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3HVBDNT45GM6VTQHA672SMBWAO/action/storage_attestation","attest_author":"https://pith.science/pith/3HVBDNT45GM6VTQHA672SMBWAO/action/author_attestation","sign_citation":"https://pith.science/pith/3HVBDNT45GM6VTQHA672SMBWAO/action/citation_signature","submit_replication":"https://pith.science/pith/3HVBDNT45GM6VTQHA672SMBWAO/action/replication_record"}},"created_at":"2026-07-05T11:14:54.455171+00:00","updated_at":"2026-07-05T11:14:54.455171+00:00"}