{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:EPSUAGAAHXOI7UEHLOFZDBQXQM","short_pith_number":"pith:EPSUAGAA","schema_version":"1.0","canonical_sha256":"23e54018003ddc8fd0875b8b91861783181488110bb9de7a8cc745f0d6d8bb59","source":{"kind":"arxiv","id":"2212.09598","version":3},"attestation_state":"computed","paper":{"title":"Query-as-context Pre-training for Dense Passage Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Guangyuan Ma, Songlin Hu, Wanhui Qian, Xing Wu, Zijia Lin","submitted_at":"2022-12-19T16:34:19Z","abstract_excerpt":"Recently, methods have been developed to improve the performance of dense passage retrieval by using context-supervised pre-training. These methods simply consider two passages from the same document to be relevant, without taking into account the possibility of weakly correlated pairs. Thus, this paper proposes query-as-context pre-training, a simple yet effective pre-training technique to alleviate the issue. Query-as-context pre-training assumes that the query derived from a passage is more likely to be relevant to that passage and forms a passage-query pair. These passage-query pairs are 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":"2212.09598","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2022-12-19T16:34:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1cd4b5199a4c646b723e5aa46014ab59bc5f12af4a5659c342f134d0a97a30ba","abstract_canon_sha256":"e54e84e00bb181161ce0ec817839205ef7d47d21b6e3c4060e629db1fb48922f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:00:58.174106Z","signature_b64":"HC1fDfWVpA5RiJL6p7GbHanm8+IqSSlmktJDqbFC6EcQ8QPPzHkULWS2wJxWplIgNJlv3+R7Em6HhoW0mxjiAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"23e54018003ddc8fd0875b8b91861783181488110bb9de7a8cc745f0d6d8bb59","last_reissued_at":"2026-07-05T07:00:58.173617Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:00:58.173617Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Query-as-context Pre-training for Dense Passage Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Guangyuan Ma, Songlin Hu, Wanhui Qian, Xing Wu, Zijia Lin","submitted_at":"2022-12-19T16:34:19Z","abstract_excerpt":"Recently, methods have been developed to improve the performance of dense passage retrieval by using context-supervised pre-training. These methods simply consider two passages from the same document to be relevant, without taking into account the possibility of weakly correlated pairs. Thus, this paper proposes query-as-context pre-training, a simple yet effective pre-training technique to alleviate the issue. Query-as-context pre-training assumes that the query derived from a passage is more likely to be relevant to that passage and forms a passage-query pair. These passage-query pairs are t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.09598","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/2212.09598/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":"2212.09598","created_at":"2026-07-05T07:00:58.173671+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.09598v3","created_at":"2026-07-05T07:00:58.173671+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.09598","created_at":"2026-07-05T07:00:58.173671+00:00"},{"alias_kind":"pith_short_12","alias_value":"EPSUAGAAHXOI","created_at":"2026-07-05T07:00:58.173671+00:00"},{"alias_kind":"pith_short_16","alias_value":"EPSUAGAAHXOI7UEH","created_at":"2026-07-05T07:00:58.173671+00:00"},{"alias_kind":"pith_short_8","alias_value":"EPSUAGAA","created_at":"2026-07-05T07:00:58.173671+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.12766","citing_title":"NExtLong: Toward Effective Long-Context Training without Long Documents","ref_index":97,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EPSUAGAAHXOI7UEHLOFZDBQXQM","json":"https://pith.science/pith/EPSUAGAAHXOI7UEHLOFZDBQXQM.json","graph_json":"https://pith.science/api/pith-number/EPSUAGAAHXOI7UEHLOFZDBQXQM/graph.json","events_json":"https://pith.science/api/pith-number/EPSUAGAAHXOI7UEHLOFZDBQXQM/events.json","paper":"https://pith.science/paper/EPSUAGAA"},"agent_actions":{"view_html":"https://pith.science/pith/EPSUAGAAHXOI7UEHLOFZDBQXQM","download_json":"https://pith.science/pith/EPSUAGAAHXOI7UEHLOFZDBQXQM.json","view_paper":"https://pith.science/paper/EPSUAGAA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.09598&json=true","fetch_graph":"https://pith.science/api/pith-number/EPSUAGAAHXOI7UEHLOFZDBQXQM/graph.json","fetch_events":"https://pith.science/api/pith-number/EPSUAGAAHXOI7UEHLOFZDBQXQM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EPSUAGAAHXOI7UEHLOFZDBQXQM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EPSUAGAAHXOI7UEHLOFZDBQXQM/action/storage_attestation","attest_author":"https://pith.science/pith/EPSUAGAAHXOI7UEHLOFZDBQXQM/action/author_attestation","sign_citation":"https://pith.science/pith/EPSUAGAAHXOI7UEHLOFZDBQXQM/action/citation_signature","submit_replication":"https://pith.science/pith/EPSUAGAAHXOI7UEHLOFZDBQXQM/action/replication_record"}},"created_at":"2026-07-05T07:00:58.173671+00:00","updated_at":"2026-07-05T07:00:58.173671+00:00"}