{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:P3NVBT3TR2YYFXNI66VCVWSC6A","short_pith_number":"pith:P3NVBT3T","schema_version":"1.0","canonical_sha256":"7edb50cf738eb182dda8f7aa2ada42f02af015c3e095f878d9b51b1ffea4232c","source":{"kind":"arxiv","id":"2205.01289","version":5},"attestation_state":"computed","paper":{"title":"On Ranking Consistency of Pre-ranking Stage","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Siyu Gu, Xiangrong Sheng","submitted_at":"2022-05-03T03:47:42Z","abstract_excerpt":"Industrial ranking systems, such as advertising systems, rank items by aggregating multiple objectives into one final objective to satisfy user demand and commercial intent. Cascade architecture, composed of retrieval, pre-ranking, and ranking stages, is usually adopted to reduce the computational cost. Each stage may employ various models for different objectives and calculate the final objective by aggregating these models' outputs. The multi-stage ranking strategy causes a new problem - the ranked lists of the ranking stage and previous stages may be inconsistent. For example, items that sh"},"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":"2205.01289","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2022-05-03T03:47:42Z","cross_cats_sorted":[],"title_canon_sha256":"f28e788938cd276c8760236765c8516aafbbad8e83b519d80c8e9ded61a8a7d5","abstract_canon_sha256":"b1d610c5862a81b4b98993722714189d81e8890606aca8db1459b741452a724e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:12:51.273621Z","signature_b64":"/GRcFNcleXdIk+HUaVa4r1kpaDCJ45re3ZzQFqI/seRhib4bDfkzJGMZZTS1rmUeEVVxqOkdDKIHqQrWgRVvDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7edb50cf738eb182dda8f7aa2ada42f02af015c3e095f878d9b51b1ffea4232c","last_reissued_at":"2026-07-05T05:12:51.273189Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:12:51.273189Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Ranking Consistency of Pre-ranking Stage","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Siyu Gu, Xiangrong Sheng","submitted_at":"2022-05-03T03:47:42Z","abstract_excerpt":"Industrial ranking systems, such as advertising systems, rank items by aggregating multiple objectives into one final objective to satisfy user demand and commercial intent. Cascade architecture, composed of retrieval, pre-ranking, and ranking stages, is usually adopted to reduce the computational cost. Each stage may employ various models for different objectives and calculate the final objective by aggregating these models' outputs. The multi-stage ranking strategy causes a new problem - the ranked lists of the ranking stage and previous stages may be inconsistent. For example, items that sh"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.01289","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/2205.01289/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":"2205.01289","created_at":"2026-07-05T05:12:51.273245+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.01289v5","created_at":"2026-07-05T05:12:51.273245+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.01289","created_at":"2026-07-05T05:12:51.273245+00:00"},{"alias_kind":"pith_short_12","alias_value":"P3NVBT3TR2YY","created_at":"2026-07-05T05:12:51.273245+00:00"},{"alias_kind":"pith_short_16","alias_value":"P3NVBT3TR2YYFXNI","created_at":"2026-07-05T05:12:51.273245+00:00"},{"alias_kind":"pith_short_8","alias_value":"P3NVBT3T","created_at":"2026-07-05T05:12:51.273245+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.19987","citing_title":"UniRank: Benchmarking Ranking Models for Unified Sequential Modeling and Feature Interaction","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P3NVBT3TR2YYFXNI66VCVWSC6A","json":"https://pith.science/pith/P3NVBT3TR2YYFXNI66VCVWSC6A.json","graph_json":"https://pith.science/api/pith-number/P3NVBT3TR2YYFXNI66VCVWSC6A/graph.json","events_json":"https://pith.science/api/pith-number/P3NVBT3TR2YYFXNI66VCVWSC6A/events.json","paper":"https://pith.science/paper/P3NVBT3T"},"agent_actions":{"view_html":"https://pith.science/pith/P3NVBT3TR2YYFXNI66VCVWSC6A","download_json":"https://pith.science/pith/P3NVBT3TR2YYFXNI66VCVWSC6A.json","view_paper":"https://pith.science/paper/P3NVBT3T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.01289&json=true","fetch_graph":"https://pith.science/api/pith-number/P3NVBT3TR2YYFXNI66VCVWSC6A/graph.json","fetch_events":"https://pith.science/api/pith-number/P3NVBT3TR2YYFXNI66VCVWSC6A/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P3NVBT3TR2YYFXNI66VCVWSC6A/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P3NVBT3TR2YYFXNI66VCVWSC6A/action/storage_attestation","attest_author":"https://pith.science/pith/P3NVBT3TR2YYFXNI66VCVWSC6A/action/author_attestation","sign_citation":"https://pith.science/pith/P3NVBT3TR2YYFXNI66VCVWSC6A/action/citation_signature","submit_replication":"https://pith.science/pith/P3NVBT3TR2YYFXNI66VCVWSC6A/action/replication_record"}},"created_at":"2026-07-05T05:12:51.273245+00:00","updated_at":"2026-07-05T05:12:51.273245+00:00"}