{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:L5TT2IFKJOD46HCOMKJRSKEVO5","short_pith_number":"pith:L5TT2IFK","schema_version":"1.0","canonical_sha256":"5f673d20aa4b87cf1c4e629319289577555155720a8d79612acfadcc3cc1bb44","source":{"kind":"arxiv","id":"2504.04400","version":3},"attestation_state":"computed","paper":{"title":"Pre-training Generative Recommender with Multi-Identifier Item Tokenization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Bowen Zheng, Enze Liu, Ji-Rong Wen, Wayne Xin Zhao, Yue Wang, Zhongfu Chen, Zhongrui Ma","submitted_at":"2025-04-06T08:03:03Z","abstract_excerpt":"Generative recommendation autoregressively generates item identifiers to recommend potential items. Existing methods typically adopt a one-to-one mapping strategy, where each item is represented by a single identifier. However, this scheme poses issues, such as suboptimal semantic modeling for low-frequency items and limited diversity in token sequence data. To overcome these limitations, we propose MTGRec, which leverages Multi-identifier item Tokenization to augment token sequence data for Generative Recommender pre-training. Our approach involves two key innovations: multi-identifier item t"},"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":"2504.04400","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2025-04-06T08:03:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"199d117e56f3ef785c08b586b5b8bee362f7c1e95acff30ce7859385401135ca","abstract_canon_sha256":"67d86331a7d06b0a8197132d456c4ca1175fa23ba01cb0c5adfc42aa676f56b2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:09:10.930861Z","signature_b64":"Zyjp/def2ZUShzIfes0Yzz+LVrJmKfL25hRGNXwnu9PbnO7MCHZQEZDywuKZbsfIDUNg2unlB9ZmzSIGWm2SDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5f673d20aa4b87cf1c4e629319289577555155720a8d79612acfadcc3cc1bb44","last_reissued_at":"2026-07-05T11:09:10.930386Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:09:10.930386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pre-training Generative Recommender with Multi-Identifier Item Tokenization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Bowen Zheng, Enze Liu, Ji-Rong Wen, Wayne Xin Zhao, Yue Wang, Zhongfu Chen, Zhongrui Ma","submitted_at":"2025-04-06T08:03:03Z","abstract_excerpt":"Generative recommendation autoregressively generates item identifiers to recommend potential items. Existing methods typically adopt a one-to-one mapping strategy, where each item is represented by a single identifier. However, this scheme poses issues, such as suboptimal semantic modeling for low-frequency items and limited diversity in token sequence data. To overcome these limitations, we propose MTGRec, which leverages Multi-identifier item Tokenization to augment token sequence data for Generative Recommender pre-training. Our approach involves two key innovations: multi-identifier item t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.04400","kind":"arxiv","version":3},"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/2504.04400/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":"2504.04400","created_at":"2026-07-05T11:09:10.930441+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.04400v3","created_at":"2026-07-05T11:09:10.930441+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.04400","created_at":"2026-07-05T11:09:10.930441+00:00"},{"alias_kind":"pith_short_12","alias_value":"L5TT2IFKJOD4","created_at":"2026-07-05T11:09:10.930441+00:00"},{"alias_kind":"pith_short_16","alias_value":"L5TT2IFKJOD46HCO","created_at":"2026-07-05T11:09:10.930441+00:00"},{"alias_kind":"pith_short_8","alias_value":"L5TT2IFK","created_at":"2026-07-05T11:09:10.930441+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2601.19711","citing_title":"Differentiable Semantic ID for Generative Recommendation","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06331","citing_title":"Expressiveness Limits of Autoregressive Semantic ID Generation in Generative Recommendation","ref_index":62,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L5TT2IFKJOD46HCOMKJRSKEVO5","json":"https://pith.science/pith/L5TT2IFKJOD46HCOMKJRSKEVO5.json","graph_json":"https://pith.science/api/pith-number/L5TT2IFKJOD46HCOMKJRSKEVO5/graph.json","events_json":"https://pith.science/api/pith-number/L5TT2IFKJOD46HCOMKJRSKEVO5/events.json","paper":"https://pith.science/paper/L5TT2IFK"},"agent_actions":{"view_html":"https://pith.science/pith/L5TT2IFKJOD46HCOMKJRSKEVO5","download_json":"https://pith.science/pith/L5TT2IFKJOD46HCOMKJRSKEVO5.json","view_paper":"https://pith.science/paper/L5TT2IFK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.04400&json=true","fetch_graph":"https://pith.science/api/pith-number/L5TT2IFKJOD46HCOMKJRSKEVO5/graph.json","fetch_events":"https://pith.science/api/pith-number/L5TT2IFKJOD46HCOMKJRSKEVO5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L5TT2IFKJOD46HCOMKJRSKEVO5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L5TT2IFKJOD46HCOMKJRSKEVO5/action/storage_attestation","attest_author":"https://pith.science/pith/L5TT2IFKJOD46HCOMKJRSKEVO5/action/author_attestation","sign_citation":"https://pith.science/pith/L5TT2IFKJOD46HCOMKJRSKEVO5/action/citation_signature","submit_replication":"https://pith.science/pith/L5TT2IFKJOD46HCOMKJRSKEVO5/action/replication_record"}},"created_at":"2026-07-05T11:09:10.930441+00:00","updated_at":"2026-07-05T11:09:10.930441+00:00"}