{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:FS5KSGUF75ZFVA7GOJALEV63DM","short_pith_number":"pith:FS5KSGUF","schema_version":"1.0","canonical_sha256":"2cbaa91a85ff725a83e67240b257db1b1f18cae5c7e42ff9676fe36b1767b576","source":{"kind":"arxiv","id":"2212.03185","version":2},"attestation_state":"computed","paper":{"title":"Rethinking the Objectives of Vector-Quantized Tokenizers for Image Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Mike Zheng Shou, Xiaohu Qie, Xintao Wang, Ying Shan, Yixiao Ge, Yuchao Gu","submitted_at":"2022-12-06T17:58:38Z","abstract_excerpt":"Vector-Quantized (VQ-based) generative models usually consist of two basic components, i.e., VQ tokenizers and generative transformers. Prior research focuses on improving the reconstruction fidelity of VQ tokenizers but rarely examines how the improvement in reconstruction affects the generation ability of generative transformers. In this paper, we surprisingly find that improving the reconstruction fidelity of VQ tokenizers does not necessarily improve the generation. Instead, learning to compress semantic features within VQ tokenizers significantly improves generative transformers' ability "},"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":"2212.03185","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-12-06T17:58:38Z","cross_cats_sorted":[],"title_canon_sha256":"e0b2d48ed4f71a3e00b764ee1f160552c0c965c126f35c6e4c83156560f5d863","abstract_canon_sha256":"f30025c26ffe18b7a8c90b822f7bbfcd2c7ae5fd2e9c6881b9447a12b518c590"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:49:31.884480Z","signature_b64":"/ByibUJEiNlTnZiqXO8yzxZl4ry5Aqzcb2W9fLQYggb3IDz3oZ+e5bpSZ5z05mQak+oMV54V/RuRh06an8QSAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2cbaa91a85ff725a83e67240b257db1b1f18cae5c7e42ff9676fe36b1767b576","last_reissued_at":"2026-07-05T05:49:31.884025Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:49:31.884025Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Rethinking the Objectives of Vector-Quantized Tokenizers for Image Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Mike Zheng Shou, Xiaohu Qie, Xintao Wang, Ying Shan, Yixiao Ge, Yuchao Gu","submitted_at":"2022-12-06T17:58:38Z","abstract_excerpt":"Vector-Quantized (VQ-based) generative models usually consist of two basic components, i.e., VQ tokenizers and generative transformers. Prior research focuses on improving the reconstruction fidelity of VQ tokenizers but rarely examines how the improvement in reconstruction affects the generation ability of generative transformers. In this paper, we surprisingly find that improving the reconstruction fidelity of VQ tokenizers does not necessarily improve the generation. Instead, learning to compress semantic features within VQ tokenizers significantly improves generative transformers' ability "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.03185","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/2212.03185/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":"2212.03185","created_at":"2026-07-05T05:49:31.884092+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.03185v2","created_at":"2026-07-05T05:49:31.884092+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.03185","created_at":"2026-07-05T05:49:31.884092+00:00"},{"alias_kind":"pith_short_12","alias_value":"FS5KSGUF75ZF","created_at":"2026-07-05T05:49:31.884092+00:00"},{"alias_kind":"pith_short_16","alias_value":"FS5KSGUF75ZFVA7G","created_at":"2026-07-05T05:49:31.884092+00:00"},{"alias_kind":"pith_short_8","alias_value":"FS5KSGUF","created_at":"2026-07-05T05:49:31.884092+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/FS5KSGUF75ZFVA7GOJALEV63DM","json":"https://pith.science/pith/FS5KSGUF75ZFVA7GOJALEV63DM.json","graph_json":"https://pith.science/api/pith-number/FS5KSGUF75ZFVA7GOJALEV63DM/graph.json","events_json":"https://pith.science/api/pith-number/FS5KSGUF75ZFVA7GOJALEV63DM/events.json","paper":"https://pith.science/paper/FS5KSGUF"},"agent_actions":{"view_html":"https://pith.science/pith/FS5KSGUF75ZFVA7GOJALEV63DM","download_json":"https://pith.science/pith/FS5KSGUF75ZFVA7GOJALEV63DM.json","view_paper":"https://pith.science/paper/FS5KSGUF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.03185&json=true","fetch_graph":"https://pith.science/api/pith-number/FS5KSGUF75ZFVA7GOJALEV63DM/graph.json","fetch_events":"https://pith.science/api/pith-number/FS5KSGUF75ZFVA7GOJALEV63DM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FS5KSGUF75ZFVA7GOJALEV63DM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FS5KSGUF75ZFVA7GOJALEV63DM/action/storage_attestation","attest_author":"https://pith.science/pith/FS5KSGUF75ZFVA7GOJALEV63DM/action/author_attestation","sign_citation":"https://pith.science/pith/FS5KSGUF75ZFVA7GOJALEV63DM/action/citation_signature","submit_replication":"https://pith.science/pith/FS5KSGUF75ZFVA7GOJALEV63DM/action/replication_record"}},"created_at":"2026-07-05T05:49:31.884092+00:00","updated_at":"2026-07-05T05:49:31.884092+00:00"}