{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GX5GXJXP7D7AF45W7ZHEN5JZ5T","short_pith_number":"pith:GX5GXJXP","schema_version":"1.0","canonical_sha256":"35fa6ba6eff8fe02f3b6fe4e46f539ecdb38066c6342aa47e77ca949ac128ddb","source":{"kind":"arxiv","id":"2503.06930","version":2},"attestation_state":"computed","paper":{"title":"Post-Training Quantization for Diffusion Transformer via Hierarchical Timestep Grouping","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chao Xu, Jing Han, Kai Han, Ning Ding, Yehui Tang, Yuchuan Tian","submitted_at":"2025-03-10T05:21:04Z","abstract_excerpt":"Diffusion Transformer (DiT) has now become the preferred choice for building image generation models due to its great generation capability. Unlike previous convolution-based UNet models, DiT is purely composed of a stack of transformer blocks, which renders DiT excellent in scalability like large language models. However, the growing model size and multi-step sampling paradigm bring about considerable pressure on deployment and inference. In this work, we propose a post-training quantization framework tailored for Diffusion Transforms to tackle these challenges. We firstly locate that the qua"},"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":"2503.06930","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-10T05:21:04Z","cross_cats_sorted":[],"title_canon_sha256":"f7c80fced2696698a4f6d4fde99d6aa0428d680a8b1074b0369a9a8dbc0a5fec","abstract_canon_sha256":"ff74c99ba00f2a091404ee735bc065ed373bc1f20ac18efca7007fe032bc1529"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:41:21.364171Z","signature_b64":"Le3VRdEguCGY6JcF4/Ss+u3UNpYG+pZjbvzfezBGU4tHVcMqaiwzaPLMc6AxJldYyG3QAXhboqriJAgSvGXsBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"35fa6ba6eff8fe02f3b6fe4e46f539ecdb38066c6342aa47e77ca949ac128ddb","last_reissued_at":"2026-07-05T10:41:21.363591Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:41:21.363591Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Post-Training Quantization for Diffusion Transformer via Hierarchical Timestep Grouping","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chao Xu, Jing Han, Kai Han, Ning Ding, Yehui Tang, Yuchuan Tian","submitted_at":"2025-03-10T05:21:04Z","abstract_excerpt":"Diffusion Transformer (DiT) has now become the preferred choice for building image generation models due to its great generation capability. Unlike previous convolution-based UNet models, DiT is purely composed of a stack of transformer blocks, which renders DiT excellent in scalability like large language models. However, the growing model size and multi-step sampling paradigm bring about considerable pressure on deployment and inference. In this work, we propose a post-training quantization framework tailored for Diffusion Transforms to tackle these challenges. We firstly locate that the qua"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.06930","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/2503.06930/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":"2503.06930","created_at":"2026-07-05T10:41:21.363653+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.06930v2","created_at":"2026-07-05T10:41:21.363653+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.06930","created_at":"2026-07-05T10:41:21.363653+00:00"},{"alias_kind":"pith_short_12","alias_value":"GX5GXJXP7D7A","created_at":"2026-07-05T10:41:21.363653+00:00"},{"alias_kind":"pith_short_16","alias_value":"GX5GXJXP7D7AF45W","created_at":"2026-07-05T10:41:21.363653+00:00"},{"alias_kind":"pith_short_8","alias_value":"GX5GXJXP","created_at":"2026-07-05T10:41:21.363653+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/GX5GXJXP7D7AF45W7ZHEN5JZ5T","json":"https://pith.science/pith/GX5GXJXP7D7AF45W7ZHEN5JZ5T.json","graph_json":"https://pith.science/api/pith-number/GX5GXJXP7D7AF45W7ZHEN5JZ5T/graph.json","events_json":"https://pith.science/api/pith-number/GX5GXJXP7D7AF45W7ZHEN5JZ5T/events.json","paper":"https://pith.science/paper/GX5GXJXP"},"agent_actions":{"view_html":"https://pith.science/pith/GX5GXJXP7D7AF45W7ZHEN5JZ5T","download_json":"https://pith.science/pith/GX5GXJXP7D7AF45W7ZHEN5JZ5T.json","view_paper":"https://pith.science/paper/GX5GXJXP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.06930&json=true","fetch_graph":"https://pith.science/api/pith-number/GX5GXJXP7D7AF45W7ZHEN5JZ5T/graph.json","fetch_events":"https://pith.science/api/pith-number/GX5GXJXP7D7AF45W7ZHEN5JZ5T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GX5GXJXP7D7AF45W7ZHEN5JZ5T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GX5GXJXP7D7AF45W7ZHEN5JZ5T/action/storage_attestation","attest_author":"https://pith.science/pith/GX5GXJXP7D7AF45W7ZHEN5JZ5T/action/author_attestation","sign_citation":"https://pith.science/pith/GX5GXJXP7D7AF45W7ZHEN5JZ5T/action/citation_signature","submit_replication":"https://pith.science/pith/GX5GXJXP7D7AF45W7ZHEN5JZ5T/action/replication_record"}},"created_at":"2026-07-05T10:41:21.363653+00:00","updated_at":"2026-07-05T10:41:21.363653+00:00"}