{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:JPMX7XLE7CNNCOX2UACBNIA3WA","short_pith_number":"pith:JPMX7XLE","schema_version":"1.0","canonical_sha256":"4bd97fdd64f89ad13afaa00416a01bb038f3cb66da04842292d2227c65bf0661","source":{"kind":"arxiv","id":"2209.00625","version":1},"attestation_state":"computed","paper":{"title":"SwiftPruner: Reinforced Evolutionary Pruning for Efficient Ad Relevance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.IR","authors_text":"Li Lyna Zhang, Mao Yang, Min Wu, Ruofei Zhang, Ting Cao, Wei Shen, Youkow Homma, Yujing Wang","submitted_at":"2022-08-30T03:05:56Z","abstract_excerpt":"Ad relevance modeling plays a critical role in online advertising systems including Microsoft Bing. To leverage powerful transformers like BERT in this low-latency setting, many existing approaches perform ad-side computations offline. While efficient, these approaches are unable to serve cold start ads, resulting in poor relevance predictions for such ads. This work aims to design a new, low-latency BERT via structured pruning to empower real-time online inference for cold start ads relevance on a CPU platform. Our challenge is that previous methods typically prune all layers of the transform"},"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":"2209.00625","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2022-08-30T03:05:56Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"01121887c1d78c4f350cbe647dde762e922f60186e3c9a5bae8af090fc74c08a","abstract_canon_sha256":"b7b1f48a71e402e384f439a330720647a6b68ad7fcfc65ee2805c97cde712cba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:53:52.123469Z","signature_b64":"iAJc8ndTclUOTuJj1Mj2PdzL5CsxEhO8MGyDvHVDolbxnOLHDPIi7TVxi+9s2K+iegac2Ek27/ui7VTWV+MBCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4bd97fdd64f89ad13afaa00416a01bb038f3cb66da04842292d2227c65bf0661","last_reissued_at":"2026-07-05T04:53:52.122957Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:53:52.122957Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SwiftPruner: Reinforced Evolutionary Pruning for Efficient Ad Relevance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.IR","authors_text":"Li Lyna Zhang, Mao Yang, Min Wu, Ruofei Zhang, Ting Cao, Wei Shen, Youkow Homma, Yujing Wang","submitted_at":"2022-08-30T03:05:56Z","abstract_excerpt":"Ad relevance modeling plays a critical role in online advertising systems including Microsoft Bing. To leverage powerful transformers like BERT in this low-latency setting, many existing approaches perform ad-side computations offline. While efficient, these approaches are unable to serve cold start ads, resulting in poor relevance predictions for such ads. This work aims to design a new, low-latency BERT via structured pruning to empower real-time online inference for cold start ads relevance on a CPU platform. Our challenge is that previous methods typically prune all layers of the transform"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.00625","kind":"arxiv","version":1},"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/2209.00625/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":"2209.00625","created_at":"2026-07-05T04:53:52.123014+00:00"},{"alias_kind":"arxiv_version","alias_value":"2209.00625v1","created_at":"2026-07-05T04:53:52.123014+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.00625","created_at":"2026-07-05T04:53:52.123014+00:00"},{"alias_kind":"pith_short_12","alias_value":"JPMX7XLE7CNN","created_at":"2026-07-05T04:53:52.123014+00:00"},{"alias_kind":"pith_short_16","alias_value":"JPMX7XLE7CNNCOX2","created_at":"2026-07-05T04:53:52.123014+00:00"},{"alias_kind":"pith_short_8","alias_value":"JPMX7XLE","created_at":"2026-07-05T04:53:52.123014+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/JPMX7XLE7CNNCOX2UACBNIA3WA","json":"https://pith.science/pith/JPMX7XLE7CNNCOX2UACBNIA3WA.json","graph_json":"https://pith.science/api/pith-number/JPMX7XLE7CNNCOX2UACBNIA3WA/graph.json","events_json":"https://pith.science/api/pith-number/JPMX7XLE7CNNCOX2UACBNIA3WA/events.json","paper":"https://pith.science/paper/JPMX7XLE"},"agent_actions":{"view_html":"https://pith.science/pith/JPMX7XLE7CNNCOX2UACBNIA3WA","download_json":"https://pith.science/pith/JPMX7XLE7CNNCOX2UACBNIA3WA.json","view_paper":"https://pith.science/paper/JPMX7XLE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2209.00625&json=true","fetch_graph":"https://pith.science/api/pith-number/JPMX7XLE7CNNCOX2UACBNIA3WA/graph.json","fetch_events":"https://pith.science/api/pith-number/JPMX7XLE7CNNCOX2UACBNIA3WA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JPMX7XLE7CNNCOX2UACBNIA3WA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JPMX7XLE7CNNCOX2UACBNIA3WA/action/storage_attestation","attest_author":"https://pith.science/pith/JPMX7XLE7CNNCOX2UACBNIA3WA/action/author_attestation","sign_citation":"https://pith.science/pith/JPMX7XLE7CNNCOX2UACBNIA3WA/action/citation_signature","submit_replication":"https://pith.science/pith/JPMX7XLE7CNNCOX2UACBNIA3WA/action/replication_record"}},"created_at":"2026-07-05T04:53:52.123014+00:00","updated_at":"2026-07-05T04:53:52.123014+00:00"}