{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7HHJ2SZTGXOBP2TM563FGDTOCR","short_pith_number":"pith:7HHJ2SZT","schema_version":"1.0","canonical_sha256":"f9ce9d4b3335dc17ea6cefb6530e6e145bcad9a19d4aa80aa1aeb8e854d3539c","source":{"kind":"arxiv","id":"2412.10208","version":3},"attestation_state":"computed","paper":{"title":"Efficient Generative Modeling with Residual Vector Quantization-Based Tokens","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jaehyeon Kim, Jaewoong Cho, Keon Lee, Taehong Moon","submitted_at":"2024-12-13T15:31:17Z","abstract_excerpt":"We introduce ResGen, an efficient Residual Vector Quantization (RVQ)-based generative model for high-fidelity generation with fast sampling. RVQ improves data fidelity by increasing the number of quantization steps, referred to as depth, but deeper quantization typically increases inference steps in generative models. To address this, ResGen directly predicts the vector embedding of collective tokens rather than individual ones, ensuring that inference steps remain independent of RVQ depth. Additionally, we formulate token masking and multi-token prediction within a probabilistic framework usi"},"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":"2412.10208","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-12-13T15:31:17Z","cross_cats_sorted":[],"title_canon_sha256":"b4cb41454c0ee8d7a076013849c82eb53f7305fe2df3b411c6d3486e4c782f99","abstract_canon_sha256":"b685b20f4c44162257b314f06e120feae5cdd8b5d3206a0e9e0c6da4017f9799"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:00.638275Z","signature_b64":"bCyN4/jnx38XcoTsVS0t55PsSnl0Oj3dRXUvbevrMuaR3+UvWlbQcbEYQT9cDuGV9+Hb0Q/2IBgFOmU6IHZ2DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f9ce9d4b3335dc17ea6cefb6530e6e145bcad9a19d4aa80aa1aeb8e854d3539c","last_reissued_at":"2026-07-05T11:14:00.637770Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:00.637770Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Generative Modeling with Residual Vector Quantization-Based Tokens","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jaehyeon Kim, Jaewoong Cho, Keon Lee, Taehong Moon","submitted_at":"2024-12-13T15:31:17Z","abstract_excerpt":"We introduce ResGen, an efficient Residual Vector Quantization (RVQ)-based generative model for high-fidelity generation with fast sampling. RVQ improves data fidelity by increasing the number of quantization steps, referred to as depth, but deeper quantization typically increases inference steps in generative models. To address this, ResGen directly predicts the vector embedding of collective tokens rather than individual ones, ensuring that inference steps remain independent of RVQ depth. Additionally, we formulate token masking and multi-token prediction within a probabilistic framework usi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.10208","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/2412.10208/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":"2412.10208","created_at":"2026-07-05T11:14:00.637837+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.10208v3","created_at":"2026-07-05T11:14:00.637837+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.10208","created_at":"2026-07-05T11:14:00.637837+00:00"},{"alias_kind":"pith_short_12","alias_value":"7HHJ2SZTGXOB","created_at":"2026-07-05T11:14:00.637837+00:00"},{"alias_kind":"pith_short_16","alias_value":"7HHJ2SZTGXOBP2TM","created_at":"2026-07-05T11:14:00.637837+00:00"},{"alias_kind":"pith_short_8","alias_value":"7HHJ2SZT","created_at":"2026-07-05T11:14:00.637837+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2507.08128","citing_title":"Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Models","ref_index":61,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7HHJ2SZTGXOBP2TM563FGDTOCR","json":"https://pith.science/pith/7HHJ2SZTGXOBP2TM563FGDTOCR.json","graph_json":"https://pith.science/api/pith-number/7HHJ2SZTGXOBP2TM563FGDTOCR/graph.json","events_json":"https://pith.science/api/pith-number/7HHJ2SZTGXOBP2TM563FGDTOCR/events.json","paper":"https://pith.science/paper/7HHJ2SZT"},"agent_actions":{"view_html":"https://pith.science/pith/7HHJ2SZTGXOBP2TM563FGDTOCR","download_json":"https://pith.science/pith/7HHJ2SZTGXOBP2TM563FGDTOCR.json","view_paper":"https://pith.science/paper/7HHJ2SZT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.10208&json=true","fetch_graph":"https://pith.science/api/pith-number/7HHJ2SZTGXOBP2TM563FGDTOCR/graph.json","fetch_events":"https://pith.science/api/pith-number/7HHJ2SZTGXOBP2TM563FGDTOCR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7HHJ2SZTGXOBP2TM563FGDTOCR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7HHJ2SZTGXOBP2TM563FGDTOCR/action/storage_attestation","attest_author":"https://pith.science/pith/7HHJ2SZTGXOBP2TM563FGDTOCR/action/author_attestation","sign_citation":"https://pith.science/pith/7HHJ2SZTGXOBP2TM563FGDTOCR/action/citation_signature","submit_replication":"https://pith.science/pith/7HHJ2SZTGXOBP2TM563FGDTOCR/action/replication_record"}},"created_at":"2026-07-05T11:14:00.637837+00:00","updated_at":"2026-07-05T11:14:00.637837+00:00"}