{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:GZF2V5K4Y7GXYRYRBX4BQA5V3D","short_pith_number":"pith:GZF2V5K4","schema_version":"1.0","canonical_sha256":"364baaf55cc7cd7c47110df81803b5d8cad9dfc6745c4d24e63b0e182174523d","source":{"kind":"arxiv","id":"2305.10657","version":4},"attestation_state":"computed","paper":{"title":"PTQD: Accurate Post-Training Quantization for Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bohan Zhuang, Hong Zhou, Jing Liu, Luping Liu, Weijia Wu, Yefei He","submitted_at":"2023-05-18T02:28:42Z","abstract_excerpt":"Diffusion models have recently dominated image synthesis tasks. However, the iterative denoising process is expensive in computations at inference time, making diffusion models less practical for low-latency and scalable real-world applications. Post-training quantization (PTQ) of diffusion models can significantly reduce the model size and accelerate the sampling process without re-training. Nonetheless, applying existing PTQ methods directly to low-bit diffusion models can significantly impair the quality of generated samples. Specifically, for each denoising step, quantization noise leads t"},"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":"2305.10657","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-18T02:28:42Z","cross_cats_sorted":[],"title_canon_sha256":"b214bc7fc51134879e0fde6ad1e6df6cdb700ec103ad309a7e76c9afa10d371e","abstract_canon_sha256":"feb9bdac4e696af5f3aa0804b2a78794db08c5126e4f63a72b67b89ac2a3b2b1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:07:53.469694Z","signature_b64":"fEsxMmlFTWRMDa6A6dHfZWbuMldMSU7a9Oj4xnxuMSx+W2oK8PGberU4sGoQtgY/QxLXh6wuVzxoCAbn+sFyDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"364baaf55cc7cd7c47110df81803b5d8cad9dfc6745c4d24e63b0e182174523d","last_reissued_at":"2026-07-05T07:07:53.469129Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:07:53.469129Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PTQD: Accurate Post-Training Quantization for Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bohan Zhuang, Hong Zhou, Jing Liu, Luping Liu, Weijia Wu, Yefei He","submitted_at":"2023-05-18T02:28:42Z","abstract_excerpt":"Diffusion models have recently dominated image synthesis tasks. However, the iterative denoising process is expensive in computations at inference time, making diffusion models less practical for low-latency and scalable real-world applications. Post-training quantization (PTQ) of diffusion models can significantly reduce the model size and accelerate the sampling process without re-training. Nonetheless, applying existing PTQ methods directly to low-bit diffusion models can significantly impair the quality of generated samples. Specifically, for each denoising step, quantization noise leads t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.10657","kind":"arxiv","version":4},"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/2305.10657/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":"2305.10657","created_at":"2026-07-05T07:07:53.469185+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.10657v4","created_at":"2026-07-05T07:07:53.469185+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.10657","created_at":"2026-07-05T07:07:53.469185+00:00"},{"alias_kind":"pith_short_12","alias_value":"GZF2V5K4Y7GX","created_at":"2026-07-05T07:07:53.469185+00:00"},{"alias_kind":"pith_short_16","alias_value":"GZF2V5K4Y7GXYRYR","created_at":"2026-07-05T07:07:53.469185+00:00"},{"alias_kind":"pith_short_8","alias_value":"GZF2V5K4","created_at":"2026-07-05T07:07:53.469185+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12280","citing_title":"Holding the FP8 Quality Ceiling at 8-Bit Weights and Activations: INT8 and GGUF Post-Training Quantization of Ideogram 4.0 for Consumer GPUs","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23381","citing_title":"VDE: Training-Free Accelerating Rectified Flow Model via Velocity Decomposition and Estimation","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21072","citing_title":"Q-ARVD: Quantizing Autoregressive Video Diffusion Models","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20179","citing_title":"TIDE: Efficient and Lossless MoE Diffusion LLM Inference with I/O-aware Expert Offload","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12668","citing_title":"OFA-Diffusion Compression: Compressing Diffusion Model in One-Shot Manner","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06916","citing_title":"FP4 Explore, BF16 Train: Diffusion Reinforcement Learning via Efficient Rollout Scaling","ref_index":59,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GZF2V5K4Y7GXYRYRBX4BQA5V3D","json":"https://pith.science/pith/GZF2V5K4Y7GXYRYRBX4BQA5V3D.json","graph_json":"https://pith.science/api/pith-number/GZF2V5K4Y7GXYRYRBX4BQA5V3D/graph.json","events_json":"https://pith.science/api/pith-number/GZF2V5K4Y7GXYRYRBX4BQA5V3D/events.json","paper":"https://pith.science/paper/GZF2V5K4"},"agent_actions":{"view_html":"https://pith.science/pith/GZF2V5K4Y7GXYRYRBX4BQA5V3D","download_json":"https://pith.science/pith/GZF2V5K4Y7GXYRYRBX4BQA5V3D.json","view_paper":"https://pith.science/paper/GZF2V5K4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.10657&json=true","fetch_graph":"https://pith.science/api/pith-number/GZF2V5K4Y7GXYRYRBX4BQA5V3D/graph.json","fetch_events":"https://pith.science/api/pith-number/GZF2V5K4Y7GXYRYRBX4BQA5V3D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GZF2V5K4Y7GXYRYRBX4BQA5V3D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GZF2V5K4Y7GXYRYRBX4BQA5V3D/action/storage_attestation","attest_author":"https://pith.science/pith/GZF2V5K4Y7GXYRYRBX4BQA5V3D/action/author_attestation","sign_citation":"https://pith.science/pith/GZF2V5K4Y7GXYRYRBX4BQA5V3D/action/citation_signature","submit_replication":"https://pith.science/pith/GZF2V5K4Y7GXYRYRBX4BQA5V3D/action/replication_record"}},"created_at":"2026-07-05T07:07:53.469185+00:00","updated_at":"2026-07-05T07:07:53.469185+00:00"}