{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:JTFGD2KGOWB53LHJHKI3OZWVSB","short_pith_number":"pith:JTFGD2KG","schema_version":"1.0","canonical_sha256":"4cca61e9467583ddace93a91b766d590680a8c92b853f1b141ebfa843b0405ef","source":{"kind":"arxiv","id":"2110.03611","version":5},"attestation_state":"computed","paper":{"title":"Adversarial Retriever-Ranker for dense text retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hang Zhang, Jiancheng Lv, Nan Duan, Weizhu Chen, Yelong Shen, Yeyun Gong","submitted_at":"2021-10-07T16:41:15Z","abstract_excerpt":"Current dense text retrieval models face two typical challenges. First, they adopt a siamese dual-encoder architecture to encode queries and documents independently for fast indexing and searching, while neglecting the finer-grained term-wise interactions. This results in a sub-optimal recall performance. Second, their model training highly relies on a negative sampling technique to build up the negative documents in their contrastive losses. To address these challenges, we present Adversarial Retriever-Ranker (AR2), which consists of a dual-encoder retriever plus a cross-encoder ranker. The t"},"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":"2110.03611","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-10-07T16:41:15Z","cross_cats_sorted":[],"title_canon_sha256":"d2805568b2e7e751df3244c68f1c8510c69064151b9a1bd7ffa12cc869747414","abstract_canon_sha256":"3fb8f9055f3413772d0254c7b63c2a699d3ec594be8d16b027183532b735ec28"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:11:35.181843Z","signature_b64":"QeyybMRrPKTXqksKHwJfhczuea0GA/kYCD4lkoQQW0QmQHuXxxRkVUywj0BMFY5udrMBIsFrS9Ippp6ZRGzzCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4cca61e9467583ddace93a91b766d590680a8c92b853f1b141ebfa843b0405ef","last_reissued_at":"2026-07-05T05:11:35.181363Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:11:35.181363Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adversarial Retriever-Ranker for dense text retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hang Zhang, Jiancheng Lv, Nan Duan, Weizhu Chen, Yelong Shen, Yeyun Gong","submitted_at":"2021-10-07T16:41:15Z","abstract_excerpt":"Current dense text retrieval models face two typical challenges. First, they adopt a siamese dual-encoder architecture to encode queries and documents independently for fast indexing and searching, while neglecting the finer-grained term-wise interactions. This results in a sub-optimal recall performance. Second, their model training highly relies on a negative sampling technique to build up the negative documents in their contrastive losses. To address these challenges, we present Adversarial Retriever-Ranker (AR2), which consists of a dual-encoder retriever plus a cross-encoder ranker. The t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.03611","kind":"arxiv","version":5},"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/2110.03611/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":"2110.03611","created_at":"2026-07-05T05:11:35.181426+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.03611v5","created_at":"2026-07-05T05:11:35.181426+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.03611","created_at":"2026-07-05T05:11:35.181426+00:00"},{"alias_kind":"pith_short_12","alias_value":"JTFGD2KGOWB5","created_at":"2026-07-05T05:11:35.181426+00:00"},{"alias_kind":"pith_short_16","alias_value":"JTFGD2KGOWB53LHJ","created_at":"2026-07-05T05:11:35.181426+00:00"},{"alias_kind":"pith_short_8","alias_value":"JTFGD2KG","created_at":"2026-07-05T05:11:35.181426+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.27449","citing_title":"Checking Fact with Better Retrieval: Dynamic Contrastive Learning for Evidence Retrieval","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2402.03216","citing_title":"M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04734","citing_title":"Beyond Hard Negatives: The Importance of Score Distribution in Knowledge Distillation for Dense Retrieval","ref_index":39,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JTFGD2KGOWB53LHJHKI3OZWVSB","json":"https://pith.science/pith/JTFGD2KGOWB53LHJHKI3OZWVSB.json","graph_json":"https://pith.science/api/pith-number/JTFGD2KGOWB53LHJHKI3OZWVSB/graph.json","events_json":"https://pith.science/api/pith-number/JTFGD2KGOWB53LHJHKI3OZWVSB/events.json","paper":"https://pith.science/paper/JTFGD2KG"},"agent_actions":{"view_html":"https://pith.science/pith/JTFGD2KGOWB53LHJHKI3OZWVSB","download_json":"https://pith.science/pith/JTFGD2KGOWB53LHJHKI3OZWVSB.json","view_paper":"https://pith.science/paper/JTFGD2KG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.03611&json=true","fetch_graph":"https://pith.science/api/pith-number/JTFGD2KGOWB53LHJHKI3OZWVSB/graph.json","fetch_events":"https://pith.science/api/pith-number/JTFGD2KGOWB53LHJHKI3OZWVSB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JTFGD2KGOWB53LHJHKI3OZWVSB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JTFGD2KGOWB53LHJHKI3OZWVSB/action/storage_attestation","attest_author":"https://pith.science/pith/JTFGD2KGOWB53LHJHKI3OZWVSB/action/author_attestation","sign_citation":"https://pith.science/pith/JTFGD2KGOWB53LHJHKI3OZWVSB/action/citation_signature","submit_replication":"https://pith.science/pith/JTFGD2KGOWB53LHJHKI3OZWVSB/action/replication_record"}},"created_at":"2026-07-05T05:11:35.181426+00:00","updated_at":"2026-07-05T05:11:35.181426+00:00"}