{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:6XVUN5UGJKRUCFVOJCM6GRFFJG","short_pith_number":"pith:6XVUN5UG","schema_version":"1.0","canonical_sha256":"f5eb46f6864aa34116ae4899e344a5499b7f75f221bacde673993d1ec7347a86","source":{"kind":"arxiv","id":"2411.10057","version":1},"attestation_state":"computed","paper":{"title":"KuaiFormer: Transformer-Based Retrieval at Kuaishou","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.IR","authors_text":"Chi Liu, Guorui Zhou, Jiangxia Cao, Kai Zheng, Kun Gai, Qiang Luo, Rui Huang","submitted_at":"2024-11-15T09:20:46Z","abstract_excerpt":"In large-scale content recommendation systems, retrieval serves as the initial stage in the pipeline, responsible for selecting thousands of candidate items from billions of options to pass on to ranking modules. Traditionally, the dominant retrieval method has been Embedding-Based Retrieval (EBR) using a Deep Neural Network (DNN) dual-tower structure. However, applying transformer in retrieval tasks has been the focus of recent research, though real-world industrial deployment still presents significant challenges. In this paper, we introduce KuaiFormer, a novel transformer-based retrieval fr"},"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":"2411.10057","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2024-11-15T09:20:46Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"6a42f2202bcfea893db4bd9529cc092532360d04b3f1befd95cb44128ae0c7d9","abstract_canon_sha256":"6c225d460f93fa5c3d0279374559c16f81f15c8b42595ddb78a43546ba722a3a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:35:58.011935Z","signature_b64":"l0yEY4a7l/Y4rvCsumxJ8CCkasnGuGfiIPF13b9QCFnSV2b0XJfHcMrpthmWfQ+sDI9istB0VHH16096AQYwCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f5eb46f6864aa34116ae4899e344a5499b7f75f221bacde673993d1ec7347a86","last_reissued_at":"2026-07-05T09:35:58.011448Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:35:58.011448Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"KuaiFormer: Transformer-Based Retrieval at Kuaishou","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.IR","authors_text":"Chi Liu, Guorui Zhou, Jiangxia Cao, Kai Zheng, Kun Gai, Qiang Luo, Rui Huang","submitted_at":"2024-11-15T09:20:46Z","abstract_excerpt":"In large-scale content recommendation systems, retrieval serves as the initial stage in the pipeline, responsible for selecting thousands of candidate items from billions of options to pass on to ranking modules. Traditionally, the dominant retrieval method has been Embedding-Based Retrieval (EBR) using a Deep Neural Network (DNN) dual-tower structure. However, applying transformer in retrieval tasks has been the focus of recent research, though real-world industrial deployment still presents significant challenges. In this paper, we introduce KuaiFormer, a novel transformer-based retrieval fr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.10057","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/2411.10057/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":"2411.10057","created_at":"2026-07-05T09:35:58.011507+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.10057v1","created_at":"2026-07-05T09:35:58.011507+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.10057","created_at":"2026-07-05T09:35:58.011507+00:00"},{"alias_kind":"pith_short_12","alias_value":"6XVUN5UGJKRU","created_at":"2026-07-05T09:35:58.011507+00:00"},{"alias_kind":"pith_short_16","alias_value":"6XVUN5UGJKRUCFVO","created_at":"2026-07-05T09:35:58.011507+00:00"},{"alias_kind":"pith_short_8","alias_value":"6XVUN5UG","created_at":"2026-07-05T09:35:58.011507+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.00448","citing_title":"Real-Time Hard Negative Sampling via LLM-based Clustering for Large-Scale Two-Tower Retrieval","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29970","citing_title":"From Extraction to Navigation: Progressive Retrieval with Indirectly Infinite Depth","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17779","citing_title":"Learning Variable-Length Tokenization for Generative Recommendation","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17994","citing_title":"Towards Sustainable Growth: A Multi-Value-Aware Retrieval Framework for E-Commerce Search","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2510.27157","citing_title":"A Survey on Generative Recommendation: Data, Model, and Tasks","ref_index":99,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25787","citing_title":"Harmonizing Generative Retrieval and Ranking in Chain-of-Recommendation","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6XVUN5UGJKRUCFVOJCM6GRFFJG","json":"https://pith.science/pith/6XVUN5UGJKRUCFVOJCM6GRFFJG.json","graph_json":"https://pith.science/api/pith-number/6XVUN5UGJKRUCFVOJCM6GRFFJG/graph.json","events_json":"https://pith.science/api/pith-number/6XVUN5UGJKRUCFVOJCM6GRFFJG/events.json","paper":"https://pith.science/paper/6XVUN5UG"},"agent_actions":{"view_html":"https://pith.science/pith/6XVUN5UGJKRUCFVOJCM6GRFFJG","download_json":"https://pith.science/pith/6XVUN5UGJKRUCFVOJCM6GRFFJG.json","view_paper":"https://pith.science/paper/6XVUN5UG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.10057&json=true","fetch_graph":"https://pith.science/api/pith-number/6XVUN5UGJKRUCFVOJCM6GRFFJG/graph.json","fetch_events":"https://pith.science/api/pith-number/6XVUN5UGJKRUCFVOJCM6GRFFJG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6XVUN5UGJKRUCFVOJCM6GRFFJG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6XVUN5UGJKRUCFVOJCM6GRFFJG/action/storage_attestation","attest_author":"https://pith.science/pith/6XVUN5UGJKRUCFVOJCM6GRFFJG/action/author_attestation","sign_citation":"https://pith.science/pith/6XVUN5UGJKRUCFVOJCM6GRFFJG/action/citation_signature","submit_replication":"https://pith.science/pith/6XVUN5UGJKRUCFVOJCM6GRFFJG/action/replication_record"}},"created_at":"2026-07-05T09:35:58.011507+00:00","updated_at":"2026-07-05T09:35:58.011507+00:00"}