{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DEDR6O566EDCSOTUVN3CCVFJGG","short_pith_number":"pith:DEDR6O56","schema_version":"1.0","canonical_sha256":"19071f3bbef106293a74ab762154a931868e3b78f5fd3ebbc21307fa0365a1bf","source":{"kind":"arxiv","id":"2603.17205","version":3},"attestation_state":"computed","paper":{"title":"OPERA: Online Data Pruning for Efficient Retrieval Model Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.IR","authors_text":"Bernie Wang, Cuixiong Hu, George Karypis, Haoyang Fang, Hengyi Wang, Katrin Kirchhoff, Shuai Zhang, Yifei Ma","submitted_at":"2026-03-17T23:11:45Z","abstract_excerpt":"Domain-specific finetuning is essential for dense retrievers, yet not all data pairs contribute equally to the learning process. We introduce OPERA, a data pruning framework that exploits this heterogeneity to improve both the effectiveness and efficiency of retrieval model adaptation. We first investigate static pruning (SP), which retains only high-similarity query-document pairs, revealing an intrinsic quality-coverage tradeoff: ranking (NDCG) improves while retrieval (Recall) can degrade due to reduced query diversity. To resolve this tradeoff, we propose a two-stage dynamic pruning (DP) s"},"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":"2603.17205","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-03-17T23:11:45Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"e661b6fb1da2bc829e743ef4e5d597de8bf6d1abf928c37ff88d879dc7c63a7b","abstract_canon_sha256":"a9ccc887523edc0ef9cdb99bc6ae2d94b991076c101b91cb598b4668ae00cf22"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-03T01:14:50.565586Z","signature_b64":"tOmNrqarCwW07PmisiEA2s8PlyTKBNbynD/IsQ1sFcxI8Mx5JYGQWJaOcqRA2lX+NsnpommUQ/ZXEUfClbVFCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"19071f3bbef106293a74ab762154a931868e3b78f5fd3ebbc21307fa0365a1bf","last_reissued_at":"2026-08-03T01:14:50.563985Z","signature_status":"signed_v1","first_computed_at":"2026-08-03T01:14:50.563985Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"OPERA: Online Data Pruning for Efficient Retrieval Model Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.IR","authors_text":"Bernie Wang, Cuixiong Hu, George Karypis, Haoyang Fang, Hengyi Wang, Katrin Kirchhoff, Shuai Zhang, Yifei Ma","submitted_at":"2026-03-17T23:11:45Z","abstract_excerpt":"Domain-specific finetuning is essential for dense retrievers, yet not all data pairs contribute equally to the learning process. We introduce OPERA, a data pruning framework that exploits this heterogeneity to improve both the effectiveness and efficiency of retrieval model adaptation. We first investigate static pruning (SP), which retains only high-similarity query-document pairs, revealing an intrinsic quality-coverage tradeoff: ranking (NDCG) improves while retrieval (Recall) can degrade due to reduced query diversity. To resolve this tradeoff, we propose a two-stage dynamic pruning (DP) s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.17205","kind":"arxiv","version":3},"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/2603.17205/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":"2603.17205","created_at":"2026-08-03T01:14:50.564895+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.17205v3","created_at":"2026-08-03T01:14:50.564895+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.17205","created_at":"2026-08-03T01:14:50.564895+00:00"},{"alias_kind":"pith_short_12","alias_value":"DEDR6O566EDC","created_at":"2026-08-03T01:14:50.564895+00:00"},{"alias_kind":"pith_short_16","alias_value":"DEDR6O566EDCSOTU","created_at":"2026-08-03T01:14:50.564895+00:00"},{"alias_kind":"pith_short_8","alias_value":"DEDR6O56","created_at":"2026-08-03T01:14:50.564895+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DEDR6O566EDCSOTUVN3CCVFJGG","json":"https://pith.science/pith/DEDR6O566EDCSOTUVN3CCVFJGG.json","graph_json":"https://pith.science/api/pith-number/DEDR6O566EDCSOTUVN3CCVFJGG/graph.json","events_json":"https://pith.science/api/pith-number/DEDR6O566EDCSOTUVN3CCVFJGG/events.json","paper":"https://pith.science/paper/DEDR6O56"},"agent_actions":{"view_html":"https://pith.science/pith/DEDR6O566EDCSOTUVN3CCVFJGG","download_json":"https://pith.science/pith/DEDR6O566EDCSOTUVN3CCVFJGG.json","view_paper":"https://pith.science/paper/DEDR6O56","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.17205&json=true","fetch_graph":"https://pith.science/api/pith-number/DEDR6O566EDCSOTUVN3CCVFJGG/graph.json","fetch_events":"https://pith.science/api/pith-number/DEDR6O566EDCSOTUVN3CCVFJGG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DEDR6O566EDCSOTUVN3CCVFJGG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DEDR6O566EDCSOTUVN3CCVFJGG/action/storage_attestation","attest_author":"https://pith.science/pith/DEDR6O566EDCSOTUVN3CCVFJGG/action/author_attestation","sign_citation":"https://pith.science/pith/DEDR6O566EDCSOTUVN3CCVFJGG/action/citation_signature","submit_replication":"https://pith.science/pith/DEDR6O566EDCSOTUVN3CCVFJGG/action/replication_record"}},"created_at":"2026-08-03T01:14:50.564895+00:00","updated_at":"2026-08-03T01:14:50.564895+00:00"}