{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:7PL5CZRKHPGSLTDTNJN67BIM3T","short_pith_number":"pith:7PL5CZRK","schema_version":"1.0","canonical_sha256":"fbd7d1662a3bcd25cc736a5bef850cdcd687e14a92a4bcc4f1b8cdbf95d939e1","source":{"kind":"arxiv","id":"2605.17779","version":1},"attestation_state":"computed","paper":{"title":"Learning Variable-Length Tokenization for Generative Recommendation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bowen Wu, Minhao Wang, Wei Zhang","submitted_at":"2026-05-18T02:57:19Z","abstract_excerpt":"Generative recommendation reformulates recommendation as next-token prediction over discrete semantic identifiers (IDs). A fundamental yet unexplored design choice is that existing methods employ fixed-length tokenization for all items, implicitly assuming uniform encoding capacity regardless of item characteristics. Through systematic experiments across four datasets, we discover the Popularity-Length Paradox: popular items achieve optimal performance with short IDs, while tail items require substantially longer codes to capture discriminative semantics. This reveals a critical mismatch where"},"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":"2605.17779","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-18T02:57:19Z","cross_cats_sorted":[],"title_canon_sha256":"16c34cab9f06ca8e079036c5432a5b5b81dba476ff38cf85a6d98f633ba80970","abstract_canon_sha256":"bf4d17ec53d0a0c9a031e069ed029949542f14a081606965a269c055d4d70b0d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:57.814424Z","signature_b64":"PypLQ38ycfCNyrrBQ+iF97qHS6nwPFgmuDeNPJuWNtq/4pTx9FTS2ZkVjOlvhcZu4KJ1oAcaFlLFljI5cZWGCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fbd7d1662a3bcd25cc736a5bef850cdcd687e14a92a4bcc4f1b8cdbf95d939e1","last_reissued_at":"2026-05-20T00:04:57.813600Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:57.813600Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Variable-Length Tokenization for Generative Recommendation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bowen Wu, Minhao Wang, Wei Zhang","submitted_at":"2026-05-18T02:57:19Z","abstract_excerpt":"Generative recommendation reformulates recommendation as next-token prediction over discrete semantic identifiers (IDs). A fundamental yet unexplored design choice is that existing methods employ fixed-length tokenization for all items, implicitly assuming uniform encoding capacity regardless of item characteristics. Through systematic experiments across four datasets, we discover the Popularity-Length Paradox: popular items achieve optimal performance with short IDs, while tail items require substantially longer codes to capture discriminative semantics. This reveals a critical mismatch where"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.17779","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/2605.17779/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":"2605.17779","created_at":"2026-05-20T00:04:57.813725+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.17779v1","created_at":"2026-05-20T00:04:57.813725+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.17779","created_at":"2026-05-20T00:04:57.813725+00:00"},{"alias_kind":"pith_short_12","alias_value":"7PL5CZRKHPGS","created_at":"2026-05-20T00:04:57.813725+00:00"},{"alias_kind":"pith_short_16","alias_value":"7PL5CZRKHPGSLTDT","created_at":"2026-05-20T00:04:57.813725+00:00"},{"alias_kind":"pith_short_8","alias_value":"7PL5CZRK","created_at":"2026-05-20T00:04:57.813725+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/7PL5CZRKHPGSLTDTNJN67BIM3T","json":"https://pith.science/pith/7PL5CZRKHPGSLTDTNJN67BIM3T.json","graph_json":"https://pith.science/api/pith-number/7PL5CZRKHPGSLTDTNJN67BIM3T/graph.json","events_json":"https://pith.science/api/pith-number/7PL5CZRKHPGSLTDTNJN67BIM3T/events.json","paper":"https://pith.science/paper/7PL5CZRK"},"agent_actions":{"view_html":"https://pith.science/pith/7PL5CZRKHPGSLTDTNJN67BIM3T","download_json":"https://pith.science/pith/7PL5CZRKHPGSLTDTNJN67BIM3T.json","view_paper":"https://pith.science/paper/7PL5CZRK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.17779&json=true","fetch_graph":"https://pith.science/api/pith-number/7PL5CZRKHPGSLTDTNJN67BIM3T/graph.json","fetch_events":"https://pith.science/api/pith-number/7PL5CZRKHPGSLTDTNJN67BIM3T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7PL5CZRKHPGSLTDTNJN67BIM3T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7PL5CZRKHPGSLTDTNJN67BIM3T/action/storage_attestation","attest_author":"https://pith.science/pith/7PL5CZRKHPGSLTDTNJN67BIM3T/action/author_attestation","sign_citation":"https://pith.science/pith/7PL5CZRKHPGSLTDTNJN67BIM3T/action/citation_signature","submit_replication":"https://pith.science/pith/7PL5CZRKHPGSLTDTNJN67BIM3T/action/replication_record"}},"created_at":"2026-05-20T00:04:57.813725+00:00","updated_at":"2026-05-20T00:04:57.813725+00:00"}