{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KXJISZX2D57ARTSULXQ3UW6SQ2","short_pith_number":"pith:KXJISZX2","schema_version":"1.0","canonical_sha256":"55d28966fa1f7e08ce545de1ba5bd286abd146b79cc7402557b72791fae99005","source":{"kind":"arxiv","id":"2504.07998","version":1},"attestation_state":"computed","paper":{"title":"CDM-QTA: Quantized Training Acceleration for Efficient LoRA Fine-Tuning of Diffusion Model","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI","cs.AR","cs.CV"],"primary_cat":"cs.GR","authors_text":"Jinming Lu, Minghao She, Wendong Mao, Zhongfeng Wang","submitted_at":"2025-04-08T22:40:29Z","abstract_excerpt":"Fine-tuning large diffusion models for custom applications demands substantial power and time, which poses significant challenges for efficient implementation on mobile devices. In this paper, we develop a novel training accelerator specifically for Low-Rank Adaptation (LoRA) of diffusion models, aiming to streamline the process and reduce computational complexity. By leveraging a fully quantized training scheme for LoRA fine-tuning, we achieve substantial reductions in memory usage and power consumption while maintaining high model fidelity. The proposed accelerator features flexible dataflow"},"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":"2504.07998","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.GR","submitted_at":"2025-04-08T22:40:29Z","cross_cats_sorted":["cs.AI","cs.AR","cs.CV"],"title_canon_sha256":"095fae1663b7a5e45697db039dd7250bf1132276476a4a0b3b0fe46cc6df0618","abstract_canon_sha256":"2fc733b9b8db159ef0b7ae915db01fe5273c455ec0035e40e2ad08eb80359167"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:47:40.769573Z","signature_b64":"6TSCoyWD+OsGywbXP8sEarRaWumE5jbRFKzQIFg0a1P4Hng+2feep8NxEX3RcgG+5VAjDcmyIMi1QzvWw6a3Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"55d28966fa1f7e08ce545de1ba5bd286abd146b79cc7402557b72791fae99005","last_reissued_at":"2026-07-05T10:47:40.769024Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:47:40.769024Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CDM-QTA: Quantized Training Acceleration for Efficient LoRA Fine-Tuning of Diffusion Model","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI","cs.AR","cs.CV"],"primary_cat":"cs.GR","authors_text":"Jinming Lu, Minghao She, Wendong Mao, Zhongfeng Wang","submitted_at":"2025-04-08T22:40:29Z","abstract_excerpt":"Fine-tuning large diffusion models for custom applications demands substantial power and time, which poses significant challenges for efficient implementation on mobile devices. In this paper, we develop a novel training accelerator specifically for Low-Rank Adaptation (LoRA) of diffusion models, aiming to streamline the process and reduce computational complexity. By leveraging a fully quantized training scheme for LoRA fine-tuning, we achieve substantial reductions in memory usage and power consumption while maintaining high model fidelity. The proposed accelerator features flexible dataflow"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.07998","kind":"arxiv","version":1},"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/2504.07998/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":"2504.07998","created_at":"2026-07-05T10:47:40.769090+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.07998v1","created_at":"2026-07-05T10:47:40.769090+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.07998","created_at":"2026-07-05T10:47:40.769090+00:00"},{"alias_kind":"pith_short_12","alias_value":"KXJISZX2D57A","created_at":"2026-07-05T10:47:40.769090+00:00"},{"alias_kind":"pith_short_16","alias_value":"KXJISZX2D57ARTSU","created_at":"2026-07-05T10:47:40.769090+00:00"},{"alias_kind":"pith_short_8","alias_value":"KXJISZX2","created_at":"2026-07-05T10:47:40.769090+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/KXJISZX2D57ARTSULXQ3UW6SQ2","json":"https://pith.science/pith/KXJISZX2D57ARTSULXQ3UW6SQ2.json","graph_json":"https://pith.science/api/pith-number/KXJISZX2D57ARTSULXQ3UW6SQ2/graph.json","events_json":"https://pith.science/api/pith-number/KXJISZX2D57ARTSULXQ3UW6SQ2/events.json","paper":"https://pith.science/paper/KXJISZX2"},"agent_actions":{"view_html":"https://pith.science/pith/KXJISZX2D57ARTSULXQ3UW6SQ2","download_json":"https://pith.science/pith/KXJISZX2D57ARTSULXQ3UW6SQ2.json","view_paper":"https://pith.science/paper/KXJISZX2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.07998&json=true","fetch_graph":"https://pith.science/api/pith-number/KXJISZX2D57ARTSULXQ3UW6SQ2/graph.json","fetch_events":"https://pith.science/api/pith-number/KXJISZX2D57ARTSULXQ3UW6SQ2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KXJISZX2D57ARTSULXQ3UW6SQ2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KXJISZX2D57ARTSULXQ3UW6SQ2/action/storage_attestation","attest_author":"https://pith.science/pith/KXJISZX2D57ARTSULXQ3UW6SQ2/action/author_attestation","sign_citation":"https://pith.science/pith/KXJISZX2D57ARTSULXQ3UW6SQ2/action/citation_signature","submit_replication":"https://pith.science/pith/KXJISZX2D57ARTSULXQ3UW6SQ2/action/replication_record"}},"created_at":"2026-07-05T10:47:40.769090+00:00","updated_at":"2026-07-05T10:47:40.769090+00:00"}