{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:EU3D6E2ML3IZEAAAFJNFUFP72I","short_pith_number":"pith:EU3D6E2M","schema_version":"1.0","canonical_sha256":"25363f134c5ed19200002a5a5a15ffd21a0d5554eed2c4f6399a7c64435725c5","source":{"kind":"arxiv","id":"2405.17873","version":2},"attestation_state":"computed","paper":{"title":"MixDQ: Memory-Efficient Few-Step Text-to-Image Diffusion Models with Metric-Decoupled Mixed Precision Quantization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Enshu Liu, Guohao Dai, Guyue Huang, Shengen Yan, Tianchen Zhao, Tongcheng Fang, Xuefei Ning, Yu Wang, Zinan Lin","submitted_at":"2024-05-28T06:50:58Z","abstract_excerpt":"Diffusion models have achieved significant visual generation quality. However, their significant computational and memory costs pose challenge for their application on resource-constrained mobile devices or even desktop GPUs. Recent few-step diffusion models reduces the inference time by reducing the denoising steps. However, their memory consumptions are still excessive. The Post Training Quantization (PTQ) replaces high bit-width FP representation with low-bit integer values (INT4/8) , which is an effective and efficient technique to reduce the memory cost. However, when applying to few-step"},"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":"2405.17873","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-28T06:50:58Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7b308940c129fde1ce299425217cf74b47d1cc999bad517581e49e9b5784cc66","abstract_canon_sha256":"0a89c7d8d53594cc7523d50b600fae753acb3fa31647b87403d284d2b1719153"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:25:05.798676Z","signature_b64":"b3kSTqVRO4Y787CDnYAo0g/OyM65Pqf+sBgNS7YN+e7EsfBXtmlUycQda3mMBtTzpK6miVvUSh3aHWh5qaXpBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"25363f134c5ed19200002a5a5a15ffd21a0d5554eed2c4f6399a7c64435725c5","last_reissued_at":"2026-07-05T08:25:05.798153Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:25:05.798153Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MixDQ: Memory-Efficient Few-Step Text-to-Image Diffusion Models with Metric-Decoupled Mixed Precision Quantization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Enshu Liu, Guohao Dai, Guyue Huang, Shengen Yan, Tianchen Zhao, Tongcheng Fang, Xuefei Ning, Yu Wang, Zinan Lin","submitted_at":"2024-05-28T06:50:58Z","abstract_excerpt":"Diffusion models have achieved significant visual generation quality. However, their significant computational and memory costs pose challenge for their application on resource-constrained mobile devices or even desktop GPUs. Recent few-step diffusion models reduces the inference time by reducing the denoising steps. However, their memory consumptions are still excessive. The Post Training Quantization (PTQ) replaces high bit-width FP representation with low-bit integer values (INT4/8) , which is an effective and efficient technique to reduce the memory cost. However, when applying to few-step"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.17873","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/2405.17873/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":"2405.17873","created_at":"2026-07-05T08:25:05.798210+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.17873v2","created_at":"2026-07-05T08:25:05.798210+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.17873","created_at":"2026-07-05T08:25:05.798210+00:00"},{"alias_kind":"pith_short_12","alias_value":"EU3D6E2ML3IZ","created_at":"2026-07-05T08:25:05.798210+00:00"},{"alias_kind":"pith_short_16","alias_value":"EU3D6E2ML3IZEAAA","created_at":"2026-07-05T08:25:05.798210+00:00"},{"alias_kind":"pith_short_8","alias_value":"EU3D6E2M","created_at":"2026-07-05T08:25:05.798210+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.04013","citing_title":"RUQuant: Towards Refining Uniform Quantization for Large Language Models","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18117","citing_title":"LoRaQ: Optimized Low Rank Approximation for 4-bit Quantization","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EU3D6E2ML3IZEAAAFJNFUFP72I","json":"https://pith.science/pith/EU3D6E2ML3IZEAAAFJNFUFP72I.json","graph_json":"https://pith.science/api/pith-number/EU3D6E2ML3IZEAAAFJNFUFP72I/graph.json","events_json":"https://pith.science/api/pith-number/EU3D6E2ML3IZEAAAFJNFUFP72I/events.json","paper":"https://pith.science/paper/EU3D6E2M"},"agent_actions":{"view_html":"https://pith.science/pith/EU3D6E2ML3IZEAAAFJNFUFP72I","download_json":"https://pith.science/pith/EU3D6E2ML3IZEAAAFJNFUFP72I.json","view_paper":"https://pith.science/paper/EU3D6E2M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.17873&json=true","fetch_graph":"https://pith.science/api/pith-number/EU3D6E2ML3IZEAAAFJNFUFP72I/graph.json","fetch_events":"https://pith.science/api/pith-number/EU3D6E2ML3IZEAAAFJNFUFP72I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EU3D6E2ML3IZEAAAFJNFUFP72I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EU3D6E2ML3IZEAAAFJNFUFP72I/action/storage_attestation","attest_author":"https://pith.science/pith/EU3D6E2ML3IZEAAAFJNFUFP72I/action/author_attestation","sign_citation":"https://pith.science/pith/EU3D6E2ML3IZEAAAFJNFUFP72I/action/citation_signature","submit_replication":"https://pith.science/pith/EU3D6E2ML3IZEAAAFJNFUFP72I/action/replication_record"}},"created_at":"2026-07-05T08:25:05.798210+00:00","updated_at":"2026-07-05T08:25:05.798210+00:00"}