{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QV2MY4TKVAVOYR4QSXMM65EIMA","short_pith_number":"pith:QV2MY4TK","schema_version":"1.0","canonical_sha256":"8574cc726aa82aec479095d8cf7488600ad3282e9827c6be32eb02aa1491962a","source":{"kind":"arxiv","id":"2404.14774","version":2},"attestation_state":"computed","paper":{"title":"CoST: Contrastive Quantization based Semantic Tokenization for Generative Recommendation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Jieming Zhu, mengqun Jin, Qijiong Liu, Xiu Li, Zexuan Qiu, Zhenhua Dong","submitted_at":"2024-04-23T06:29:48Z","abstract_excerpt":"Embedding-based retrieval serves as a dominant approach to candidate item matching for industrial recommender systems. With the success of generative AI, generative retrieval has recently emerged as a new retrieval paradigm for recommendation, which casts item retrieval as a generation problem. Its model consists of two stages: semantic tokenization and autoregressive generation. The first stage involves item tokenization that constructs discrete semantic tokens to index items, while the second stage autoregressively generates semantic tokens of candidate items. Therefore, semantic tokenizatio"},"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":"2404.14774","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2024-04-23T06:29:48Z","cross_cats_sorted":[],"title_canon_sha256":"e9494dda76e8b48ae03c16ee81b41a2a6c357d1a5530da03734abd409aba4445","abstract_canon_sha256":"b41139219b3ae35de8c8911546277f4e69e538f3a18dcdd9a4618dcb46386229"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:04:09.451903Z","signature_b64":"gldloimmNkuju72S/LZ3fMxDGzykBcjGWsrdJHgB2fBq6HVky4m/2v5InOtcrWJz6U/hs6cMaE304/eJ1eRZAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8574cc726aa82aec479095d8cf7488600ad3282e9827c6be32eb02aa1491962a","last_reissued_at":"2026-07-05T09:04:09.451405Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:04:09.451405Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CoST: Contrastive Quantization based Semantic Tokenization for Generative Recommendation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Jieming Zhu, mengqun Jin, Qijiong Liu, Xiu Li, Zexuan Qiu, Zhenhua Dong","submitted_at":"2024-04-23T06:29:48Z","abstract_excerpt":"Embedding-based retrieval serves as a dominant approach to candidate item matching for industrial recommender systems. With the success of generative AI, generative retrieval has recently emerged as a new retrieval paradigm for recommendation, which casts item retrieval as a generation problem. Its model consists of two stages: semantic tokenization and autoregressive generation. The first stage involves item tokenization that constructs discrete semantic tokens to index items, while the second stage autoregressively generates semantic tokens of candidate items. Therefore, semantic tokenizatio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.14774","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/2404.14774/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":"2404.14774","created_at":"2026-07-05T09:04:09.451465+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.14774v2","created_at":"2026-07-05T09:04:09.451465+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.14774","created_at":"2026-07-05T09:04:09.451465+00:00"},{"alias_kind":"pith_short_12","alias_value":"QV2MY4TKVAVO","created_at":"2026-07-05T09:04:09.451465+00:00"},{"alias_kind":"pith_short_16","alias_value":"QV2MY4TKVAVOYR4Q","created_at":"2026-07-05T09:04:09.451465+00:00"},{"alias_kind":"pith_short_8","alias_value":"QV2MY4TK","created_at":"2026-07-05T09:04:09.451465+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.10468","citing_title":"Learning Decomposed Contextual Token Representations from Pretrained and Collaborative Signals for Generative Recommendation","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QV2MY4TKVAVOYR4QSXMM65EIMA","json":"https://pith.science/pith/QV2MY4TKVAVOYR4QSXMM65EIMA.json","graph_json":"https://pith.science/api/pith-number/QV2MY4TKVAVOYR4QSXMM65EIMA/graph.json","events_json":"https://pith.science/api/pith-number/QV2MY4TKVAVOYR4QSXMM65EIMA/events.json","paper":"https://pith.science/paper/QV2MY4TK"},"agent_actions":{"view_html":"https://pith.science/pith/QV2MY4TKVAVOYR4QSXMM65EIMA","download_json":"https://pith.science/pith/QV2MY4TKVAVOYR4QSXMM65EIMA.json","view_paper":"https://pith.science/paper/QV2MY4TK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.14774&json=true","fetch_graph":"https://pith.science/api/pith-number/QV2MY4TKVAVOYR4QSXMM65EIMA/graph.json","fetch_events":"https://pith.science/api/pith-number/QV2MY4TKVAVOYR4QSXMM65EIMA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QV2MY4TKVAVOYR4QSXMM65EIMA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QV2MY4TKVAVOYR4QSXMM65EIMA/action/storage_attestation","attest_author":"https://pith.science/pith/QV2MY4TKVAVOYR4QSXMM65EIMA/action/author_attestation","sign_citation":"https://pith.science/pith/QV2MY4TKVAVOYR4QSXMM65EIMA/action/citation_signature","submit_replication":"https://pith.science/pith/QV2MY4TKVAVOYR4QSXMM65EIMA/action/replication_record"}},"created_at":"2026-07-05T09:04:09.451465+00:00","updated_at":"2026-07-05T09:04:09.451465+00:00"}