{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:RZSRSVZ6C5GI6GVJ4TKSP2UH2V","short_pith_number":"pith:RZSRSVZ6","schema_version":"1.0","canonical_sha256":"8e6519573e174c8f1aa9e4d527ea87d55e53d95ccf02a75f15b7a55546aacaba","source":{"kind":"arxiv","id":"2205.16007","version":2},"attestation_state":"computed","paper":{"title":"Improved Vector Quantized Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dong Chen, Fang Wen, Jianmin Bao, Shuyang Gu, Zhicong Tang","submitted_at":"2022-05-31T17:59:53Z","abstract_excerpt":"Vector quantized diffusion (VQ-Diffusion) is a powerful generative model for text-to-image synthesis, but sometimes can still generate low-quality samples or weakly correlated images with text input. We find these issues are mainly due to the flawed sampling strategy. In this paper, we propose two important techniques to further improve the sample quality of VQ-Diffusion. 1) We explore classifier-free guidance sampling for discrete denoising diffusion model and propose a more general and effective implementation of classifier-free guidance. 2) We present a high-quality inference strategy to al"},"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":"2205.16007","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-05-31T17:59:53Z","cross_cats_sorted":[],"title_canon_sha256":"9d9582fbe2c178e73aeec2b0007de5f08e783e9629eba5ed3b5d0f6cfa1a85ff","abstract_canon_sha256":"eaa7db4130993c2d18e0e1f82466b930fe52ea47b8ec0bf14b472e963d2c53f6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:39:50.847329Z","signature_b64":"sDjhVKxdXEbGixLrbR/O4aS5iUb8UjipjrGC/p/elDOmYCliX6PC4XsSSXOamuPRHKLPoJXLPrWjfbKOn/D1DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8e6519573e174c8f1aa9e4d527ea87d55e53d95ccf02a75f15b7a55546aacaba","last_reissued_at":"2026-07-05T05:39:50.846901Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:39:50.846901Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improved Vector Quantized Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dong Chen, Fang Wen, Jianmin Bao, Shuyang Gu, Zhicong Tang","submitted_at":"2022-05-31T17:59:53Z","abstract_excerpt":"Vector quantized diffusion (VQ-Diffusion) is a powerful generative model for text-to-image synthesis, but sometimes can still generate low-quality samples or weakly correlated images with text input. We find these issues are mainly due to the flawed sampling strategy. In this paper, we propose two important techniques to further improve the sample quality of VQ-Diffusion. 1) We explore classifier-free guidance sampling for discrete denoising diffusion model and propose a more general and effective implementation of classifier-free guidance. 2) We present a high-quality inference strategy to al"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.16007","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/2205.16007/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":"2205.16007","created_at":"2026-07-05T05:39:50.846956+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.16007v2","created_at":"2026-07-05T05:39:50.846956+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.16007","created_at":"2026-07-05T05:39:50.846956+00:00"},{"alias_kind":"pith_short_12","alias_value":"RZSRSVZ6C5GI","created_at":"2026-07-05T05:39:50.846956+00:00"},{"alias_kind":"pith_short_16","alias_value":"RZSRSVZ6C5GI6GVJ","created_at":"2026-07-05T05:39:50.846956+00:00"},{"alias_kind":"pith_short_8","alias_value":"RZSRSVZ6","created_at":"2026-07-05T05:39:50.846956+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.03578","citing_title":"Diffusing in the Right Space: A Systematic Study of Latent Diffusability","ref_index":102,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22847","citing_title":"Dream-Cubed: Controllable Generative Modeling in Minecraft by Training on Billions of Cubes","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04590","citing_title":"From Diffusion to Rectified Flow: Rethinking Text-Based Segmentation","ref_index":44,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RZSRSVZ6C5GI6GVJ4TKSP2UH2V","json":"https://pith.science/pith/RZSRSVZ6C5GI6GVJ4TKSP2UH2V.json","graph_json":"https://pith.science/api/pith-number/RZSRSVZ6C5GI6GVJ4TKSP2UH2V/graph.json","events_json":"https://pith.science/api/pith-number/RZSRSVZ6C5GI6GVJ4TKSP2UH2V/events.json","paper":"https://pith.science/paper/RZSRSVZ6"},"agent_actions":{"view_html":"https://pith.science/pith/RZSRSVZ6C5GI6GVJ4TKSP2UH2V","download_json":"https://pith.science/pith/RZSRSVZ6C5GI6GVJ4TKSP2UH2V.json","view_paper":"https://pith.science/paper/RZSRSVZ6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.16007&json=true","fetch_graph":"https://pith.science/api/pith-number/RZSRSVZ6C5GI6GVJ4TKSP2UH2V/graph.json","fetch_events":"https://pith.science/api/pith-number/RZSRSVZ6C5GI6GVJ4TKSP2UH2V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RZSRSVZ6C5GI6GVJ4TKSP2UH2V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RZSRSVZ6C5GI6GVJ4TKSP2UH2V/action/storage_attestation","attest_author":"https://pith.science/pith/RZSRSVZ6C5GI6GVJ4TKSP2UH2V/action/author_attestation","sign_citation":"https://pith.science/pith/RZSRSVZ6C5GI6GVJ4TKSP2UH2V/action/citation_signature","submit_replication":"https://pith.science/pith/RZSRSVZ6C5GI6GVJ4TKSP2UH2V/action/replication_record"}},"created_at":"2026-07-05T05:39:50.846956+00:00","updated_at":"2026-07-05T05:39:50.846956+00:00"}