{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:VAVTSDQVHOCDF7JHNXASUOR4VR","short_pith_number":"pith:VAVTSDQV","schema_version":"1.0","canonical_sha256":"a82b390e153b8432fd276dc12a3a3cac6daaa7925bb616c7a9b0253d1e1ff2dd","source":{"kind":"arxiv","id":"2007.16122","version":2},"attestation_state":"computed","paper":{"title":"COLD: Towards the Next Generation of Pre-Ranking System","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.IR","authors_text":"Biye Jiang, Guorui Zhou, Kun Gai, Liqin Zhao, Xiaoqiang Zhu, Zhe Wang","submitted_at":"2020-07-31T15:06:43Z","abstract_excerpt":"Multi-stage cascade architecture exists widely in many industrial systems such as recommender systems and online advertising, which often consists of sequential modules including matching, pre-ranking, ranking, etc. For a long time, it is believed pre-ranking is just a simplified version of the ranking module, considering the larger size of the candidate set to be ranked. Thus, efforts are made mostly on simplifying ranking model to handle the explosion of computing power for online inference. In this paper, we rethink the challenge of the pre-ranking system from an algorithm-system co-design "},"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":"2007.16122","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2020-07-31T15:06:43Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"83f3a6d18f8a3790e0dc02e5a7a662fabca28e69fd76a796703fd044a89c2d3e","abstract_canon_sha256":"6238e4acda1c7ff023dbe598b1c72ee89baf63336b3c6e0cce8cf06a8633d218"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:27:26.266796Z","signature_b64":"UVVRbuPn57eBW/wpO315Pu91r054EDQAZlmtFQ2JCxNvA2pl6ICo71UEHKzTvTxjgOLw04YMYgrjr80lMs1qDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a82b390e153b8432fd276dc12a3a3cac6daaa7925bb616c7a9b0253d1e1ff2dd","last_reissued_at":"2026-07-05T01:27:26.266299Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:27:26.266299Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"COLD: Towards the Next Generation of Pre-Ranking System","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.IR","authors_text":"Biye Jiang, Guorui Zhou, Kun Gai, Liqin Zhao, Xiaoqiang Zhu, Zhe Wang","submitted_at":"2020-07-31T15:06:43Z","abstract_excerpt":"Multi-stage cascade architecture exists widely in many industrial systems such as recommender systems and online advertising, which often consists of sequential modules including matching, pre-ranking, ranking, etc. For a long time, it is believed pre-ranking is just a simplified version of the ranking module, considering the larger size of the candidate set to be ranked. Thus, efforts are made mostly on simplifying ranking model to handle the explosion of computing power for online inference. In this paper, we rethink the challenge of the pre-ranking system from an algorithm-system co-design "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.16122","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/2007.16122/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":"2007.16122","created_at":"2026-07-05T01:27:26.266365+00:00"},{"alias_kind":"arxiv_version","alias_value":"2007.16122v2","created_at":"2026-07-05T01:27:26.266365+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.16122","created_at":"2026-07-05T01:27:26.266365+00:00"},{"alias_kind":"pith_short_12","alias_value":"VAVTSDQVHOCD","created_at":"2026-07-05T01:27:26.266365+00:00"},{"alias_kind":"pith_short_16","alias_value":"VAVTSDQVHOCDF7JH","created_at":"2026-07-05T01:27:26.266365+00:00"},{"alias_kind":"pith_short_8","alias_value":"VAVTSDQV","created_at":"2026-07-05T01:27:26.266365+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2502.18965","citing_title":"OneRec: Unifying Retrieve and Rank with Generative Recommender and Iterative Preference Alignment","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19269","citing_title":"CS3: Efficient Online Capability Synergy for Two-Tower Recommendation","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VAVTSDQVHOCDF7JHNXASUOR4VR","json":"https://pith.science/pith/VAVTSDQVHOCDF7JHNXASUOR4VR.json","graph_json":"https://pith.science/api/pith-number/VAVTSDQVHOCDF7JHNXASUOR4VR/graph.json","events_json":"https://pith.science/api/pith-number/VAVTSDQVHOCDF7JHNXASUOR4VR/events.json","paper":"https://pith.science/paper/VAVTSDQV"},"agent_actions":{"view_html":"https://pith.science/pith/VAVTSDQVHOCDF7JHNXASUOR4VR","download_json":"https://pith.science/pith/VAVTSDQVHOCDF7JHNXASUOR4VR.json","view_paper":"https://pith.science/paper/VAVTSDQV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2007.16122&json=true","fetch_graph":"https://pith.science/api/pith-number/VAVTSDQVHOCDF7JHNXASUOR4VR/graph.json","fetch_events":"https://pith.science/api/pith-number/VAVTSDQVHOCDF7JHNXASUOR4VR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VAVTSDQVHOCDF7JHNXASUOR4VR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VAVTSDQVHOCDF7JHNXASUOR4VR/action/storage_attestation","attest_author":"https://pith.science/pith/VAVTSDQVHOCDF7JHNXASUOR4VR/action/author_attestation","sign_citation":"https://pith.science/pith/VAVTSDQVHOCDF7JHNXASUOR4VR/action/citation_signature","submit_replication":"https://pith.science/pith/VAVTSDQVHOCDF7JHNXASUOR4VR/action/replication_record"}},"created_at":"2026-07-05T01:27:26.266365+00:00","updated_at":"2026-07-05T01:27:26.266365+00:00"}