{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:NM7Z2UQQAHQMMACJOJ77HQ34HB","short_pith_number":"pith:NM7Z2UQQ","schema_version":"1.0","canonical_sha256":"6b3f9d521001e0c60049727ff3c37c384904d0abcf8753f3528513546667d470","source":{"kind":"arxiv","id":"2407.02031","version":2},"attestation_state":"computed","paper":{"title":"SwiftDiffusion: Efficient Diffusion Model Serving with Add-on Modules","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.DC","authors_text":"Dakai An, Guodong Yang, Hanfeng Lu, Jiawei Chen, Kan Liu, Lingyun Yang, Lin Qu, Liping Zhang, Suyi Li, Tao Lan, Wei Wang, Weiyi Lu, Xiaoxiao Jiang, Yinghao Yu, Zhipeng Di","submitted_at":"2024-07-02T07:59:08Z","abstract_excerpt":"Text-to-image (T2I) generation using diffusion models has become a blockbuster service in today's AI cloud. A production T2I service typically involves a serving workflow where a base diffusion model is augmented with various \"add-on\" modules, notably ControlNet and LoRA, to enhance image generation control. Compared to serving the base model alone, these add-on modules introduce significant loading and computational overhead, resulting in increased latency. In this paper, we present SwiftDiffusion, a system that efficiently serves a T2I workflow through a holistic approach. SwiftDiffusion dec"},"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":"2407.02031","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-07-02T07:59:08Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"6b8b3292c384101027ac29851a75418063f360e8eea7e22f27ee0cafe977c24c","abstract_canon_sha256":"4861fc532e4527176e0f01270fce82c40e9952d7bad37053a46d9dd90ae00b51"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:45:11.517013Z","signature_b64":"KfYz6e5qTSP7IVID997CJ3+UPhSfF4MDkdFO6bddZNl6vWgzmMTuQTHY/4O2QhWPWpMYeqOp5mGtz78HOzPZBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6b3f9d521001e0c60049727ff3c37c384904d0abcf8753f3528513546667d470","last_reissued_at":"2026-07-05T09:45:11.516521Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:45:11.516521Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SwiftDiffusion: Efficient Diffusion Model Serving with Add-on Modules","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.DC","authors_text":"Dakai An, Guodong Yang, Hanfeng Lu, Jiawei Chen, Kan Liu, Lingyun Yang, Lin Qu, Liping Zhang, Suyi Li, Tao Lan, Wei Wang, Weiyi Lu, Xiaoxiao Jiang, Yinghao Yu, Zhipeng Di","submitted_at":"2024-07-02T07:59:08Z","abstract_excerpt":"Text-to-image (T2I) generation using diffusion models has become a blockbuster service in today's AI cloud. A production T2I service typically involves a serving workflow where a base diffusion model is augmented with various \"add-on\" modules, notably ControlNet and LoRA, to enhance image generation control. Compared to serving the base model alone, these add-on modules introduce significant loading and computational overhead, resulting in increased latency. In this paper, we present SwiftDiffusion, a system that efficiently serves a T2I workflow through a holistic approach. SwiftDiffusion dec"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.02031","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/2407.02031/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":"2407.02031","created_at":"2026-07-05T09:45:11.516575+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.02031v2","created_at":"2026-07-05T09:45:11.516575+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.02031","created_at":"2026-07-05T09:45:11.516575+00:00"},{"alias_kind":"pith_short_12","alias_value":"NM7Z2UQQAHQM","created_at":"2026-07-05T09:45:11.516575+00:00"},{"alias_kind":"pith_short_16","alias_value":"NM7Z2UQQAHQMMACJ","created_at":"2026-07-05T09:45:11.516575+00:00"},{"alias_kind":"pith_short_8","alias_value":"NM7Z2UQQ","created_at":"2026-07-05T09:45:11.516575+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.25550","citing_title":"DisagFusion: Asynchronous Pipeline Parallelism and Elastic Scheduling for Disaggregated Diffusion Serving","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NM7Z2UQQAHQMMACJOJ77HQ34HB","json":"https://pith.science/pith/NM7Z2UQQAHQMMACJOJ77HQ34HB.json","graph_json":"https://pith.science/api/pith-number/NM7Z2UQQAHQMMACJOJ77HQ34HB/graph.json","events_json":"https://pith.science/api/pith-number/NM7Z2UQQAHQMMACJOJ77HQ34HB/events.json","paper":"https://pith.science/paper/NM7Z2UQQ"},"agent_actions":{"view_html":"https://pith.science/pith/NM7Z2UQQAHQMMACJOJ77HQ34HB","download_json":"https://pith.science/pith/NM7Z2UQQAHQMMACJOJ77HQ34HB.json","view_paper":"https://pith.science/paper/NM7Z2UQQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.02031&json=true","fetch_graph":"https://pith.science/api/pith-number/NM7Z2UQQAHQMMACJOJ77HQ34HB/graph.json","fetch_events":"https://pith.science/api/pith-number/NM7Z2UQQAHQMMACJOJ77HQ34HB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NM7Z2UQQAHQMMACJOJ77HQ34HB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NM7Z2UQQAHQMMACJOJ77HQ34HB/action/storage_attestation","attest_author":"https://pith.science/pith/NM7Z2UQQAHQMMACJOJ77HQ34HB/action/author_attestation","sign_citation":"https://pith.science/pith/NM7Z2UQQAHQMMACJOJ77HQ34HB/action/citation_signature","submit_replication":"https://pith.science/pith/NM7Z2UQQAHQMMACJOJ77HQ34HB/action/replication_record"}},"created_at":"2026-07-05T09:45:11.516575+00:00","updated_at":"2026-07-05T09:45:11.516575+00:00"}