{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WGKZRAQY4ANYCTBVRJIMNPO3TU","short_pith_number":"pith:WGKZRAQY","schema_version":"1.0","canonical_sha256":"b195988218e01b814c358a50c6bddb9d049f6a6777a536352a20b1cfbee66545","source":{"kind":"arxiv","id":"2302.04304","version":3},"attestation_state":"computed","paper":{"title":"Q-Diffusion: Quantizing Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Daniel Kang, Huanrui Yang, Kurt Keutzer, Long Lian, Shanghang Zhang, Xiuyu Li, Yijiang Liu, Zhen Dong","submitted_at":"2023-02-08T19:38:59Z","abstract_excerpt":"Diffusion models have achieved great success in image synthesis through iterative noise estimation using deep neural networks. However, the slow inference, high memory consumption, and computation intensity of the noise estimation model hinder the efficient adoption of diffusion models. Although post-training quantization (PTQ) is considered a go-to compression method for other tasks, it does not work out-of-the-box on diffusion models. We propose a novel PTQ method specifically tailored towards the unique multi-timestep pipeline and model architecture of the diffusion models, which compresses"},"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":"2302.04304","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-02-08T19:38:59Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"dea2f913e6f557205591f39696874bec9e0764b07f97ea97c52eaa1b4dfacd27","abstract_canon_sha256":"e89befe90b4e202e475542ddf394bf02ca753209f1754912f253775fd9b26485"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:18:37.900393Z","signature_b64":"GNDnUsCBN/jlHbdXJai+HVNZde8bJ79D/LU/c8hM3/MPn0s4trl+H1ZWwHSQXsqbdQbGP5vgk+bousC+IY4NDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b195988218e01b814c358a50c6bddb9d049f6a6777a536352a20b1cfbee66545","last_reissued_at":"2026-07-05T06:18:37.899901Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:18:37.899901Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Q-Diffusion: Quantizing Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Daniel Kang, Huanrui Yang, Kurt Keutzer, Long Lian, Shanghang Zhang, Xiuyu Li, Yijiang Liu, Zhen Dong","submitted_at":"2023-02-08T19:38:59Z","abstract_excerpt":"Diffusion models have achieved great success in image synthesis through iterative noise estimation using deep neural networks. However, the slow inference, high memory consumption, and computation intensity of the noise estimation model hinder the efficient adoption of diffusion models. Although post-training quantization (PTQ) is considered a go-to compression method for other tasks, it does not work out-of-the-box on diffusion models. We propose a novel PTQ method specifically tailored towards the unique multi-timestep pipeline and model architecture of the diffusion models, which compresses"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.04304","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/2302.04304/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":"2302.04304","created_at":"2026-07-05T06:18:37.899958+00:00"},{"alias_kind":"arxiv_version","alias_value":"2302.04304v3","created_at":"2026-07-05T06:18:37.899958+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.04304","created_at":"2026-07-05T06:18:37.899958+00:00"},{"alias_kind":"pith_short_12","alias_value":"WGKZRAQY4ANY","created_at":"2026-07-05T06:18:37.899958+00:00"},{"alias_kind":"pith_short_16","alias_value":"WGKZRAQY4ANYCTBV","created_at":"2026-07-05T06:18:37.899958+00:00"},{"alias_kind":"pith_short_8","alias_value":"WGKZRAQY","created_at":"2026-07-05T06:18:37.899958+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.28421","citing_title":"JuZhou 1.0 Technical Report: The First Edge-Native Text-to-Image Foundation Model Trained Entirely on China-Developed AI Accelerators","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12668","citing_title":"OFA-Diffusion Compression: Compressing Diffusion Model in One-Shot Manner","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06916","citing_title":"FP4 Explore, BF16 Train: Diffusion Reinforcement Learning via Efficient Rollout Scaling","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18348","citing_title":"AdaCluster: Adaptive Query-Key Clustering for Sparse Attention in Video Generation","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WGKZRAQY4ANYCTBVRJIMNPO3TU","json":"https://pith.science/pith/WGKZRAQY4ANYCTBVRJIMNPO3TU.json","graph_json":"https://pith.science/api/pith-number/WGKZRAQY4ANYCTBVRJIMNPO3TU/graph.json","events_json":"https://pith.science/api/pith-number/WGKZRAQY4ANYCTBVRJIMNPO3TU/events.json","paper":"https://pith.science/paper/WGKZRAQY"},"agent_actions":{"view_html":"https://pith.science/pith/WGKZRAQY4ANYCTBVRJIMNPO3TU","download_json":"https://pith.science/pith/WGKZRAQY4ANYCTBVRJIMNPO3TU.json","view_paper":"https://pith.science/paper/WGKZRAQY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2302.04304&json=true","fetch_graph":"https://pith.science/api/pith-number/WGKZRAQY4ANYCTBVRJIMNPO3TU/graph.json","fetch_events":"https://pith.science/api/pith-number/WGKZRAQY4ANYCTBVRJIMNPO3TU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WGKZRAQY4ANYCTBVRJIMNPO3TU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WGKZRAQY4ANYCTBVRJIMNPO3TU/action/storage_attestation","attest_author":"https://pith.science/pith/WGKZRAQY4ANYCTBVRJIMNPO3TU/action/author_attestation","sign_citation":"https://pith.science/pith/WGKZRAQY4ANYCTBVRJIMNPO3TU/action/citation_signature","submit_replication":"https://pith.science/pith/WGKZRAQY4ANYCTBVRJIMNPO3TU/action/replication_record"}},"created_at":"2026-07-05T06:18:37.899958+00:00","updated_at":"2026-07-05T06:18:37.899958+00:00"}