{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GYRVQFU47JGGWG65QIZFNRGXSQ","short_pith_number":"pith:GYRVQFU4","schema_version":"1.0","canonical_sha256":"362358169cfa4c6b1bdd823256c4d794177485bd73931275cf91f1bd2b462e8f","source":{"kind":"arxiv","id":"2405.20380","version":1},"attestation_state":"computed","paper":{"title":"Gradient Inversion of Federated Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","cs.CV"],"primary_cat":"cs.AI","authors_text":"Chi Hong, Jiyue Huang, Lydia Y. Chen, Stefanie Roos","submitted_at":"2024-05-30T18:00:03Z","abstract_excerpt":"Diffusion models are becoming defector generative models, which generate exceptionally high-resolution image data. Training effective diffusion models require massive real data, which is privately owned by distributed parties. Each data party can collaboratively train diffusion models in a federated learning manner by sharing gradients instead of the raw data. In this paper, we study the privacy leakage risk of gradient inversion attacks. First, we design a two-phase fusion optimization, GIDM, to leverage the well-trained generative model itself as prior knowledge to constrain the inversion se"},"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":"2405.20380","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-05-30T18:00:03Z","cross_cats_sorted":["cs.CR","cs.CV"],"title_canon_sha256":"e4bbbc56c4524c6d79b28e72f52a8397971878b10d4e09f2794113f0b69940e2","abstract_canon_sha256":"0950936665a29ef28847d4af9a86c2fe1c5e306545ce85272bd60d037b8bff32"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:25:36.681835Z","signature_b64":"78mASzJGNoUeoUQqPOSQRXyVNL7gZuHsyPXAQC9IiK9IRXoDrCuy3fzt5Krib+1kOzvYX9xjkT/WIqF7p9P8DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"362358169cfa4c6b1bdd823256c4d794177485bd73931275cf91f1bd2b462e8f","last_reissued_at":"2026-07-05T08:25:36.681306Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:25:36.681306Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Gradient Inversion of Federated Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","cs.CV"],"primary_cat":"cs.AI","authors_text":"Chi Hong, Jiyue Huang, Lydia Y. Chen, Stefanie Roos","submitted_at":"2024-05-30T18:00:03Z","abstract_excerpt":"Diffusion models are becoming defector generative models, which generate exceptionally high-resolution image data. Training effective diffusion models require massive real data, which is privately owned by distributed parties. Each data party can collaboratively train diffusion models in a federated learning manner by sharing gradients instead of the raw data. In this paper, we study the privacy leakage risk of gradient inversion attacks. First, we design a two-phase fusion optimization, GIDM, to leverage the well-trained generative model itself as prior knowledge to constrain the inversion se"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.20380","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/2405.20380/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":"2405.20380","created_at":"2026-07-05T08:25:36.681385+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.20380v1","created_at":"2026-07-05T08:25:36.681385+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.20380","created_at":"2026-07-05T08:25:36.681385+00:00"},{"alias_kind":"pith_short_12","alias_value":"GYRVQFU47JGG","created_at":"2026-07-05T08:25:36.681385+00:00"},{"alias_kind":"pith_short_16","alias_value":"GYRVQFU47JGGWG65","created_at":"2026-07-05T08:25:36.681385+00:00"},{"alias_kind":"pith_short_8","alias_value":"GYRVQFU4","created_at":"2026-07-05T08:25:36.681385+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.18934","citing_title":"Deep Learning Model Inversion Attacks and Defenses: A Comprehensive Survey","ref_index":42,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GYRVQFU47JGGWG65QIZFNRGXSQ","json":"https://pith.science/pith/GYRVQFU47JGGWG65QIZFNRGXSQ.json","graph_json":"https://pith.science/api/pith-number/GYRVQFU47JGGWG65QIZFNRGXSQ/graph.json","events_json":"https://pith.science/api/pith-number/GYRVQFU47JGGWG65QIZFNRGXSQ/events.json","paper":"https://pith.science/paper/GYRVQFU4"},"agent_actions":{"view_html":"https://pith.science/pith/GYRVQFU47JGGWG65QIZFNRGXSQ","download_json":"https://pith.science/pith/GYRVQFU47JGGWG65QIZFNRGXSQ.json","view_paper":"https://pith.science/paper/GYRVQFU4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.20380&json=true","fetch_graph":"https://pith.science/api/pith-number/GYRVQFU47JGGWG65QIZFNRGXSQ/graph.json","fetch_events":"https://pith.science/api/pith-number/GYRVQFU47JGGWG65QIZFNRGXSQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GYRVQFU47JGGWG65QIZFNRGXSQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GYRVQFU47JGGWG65QIZFNRGXSQ/action/storage_attestation","attest_author":"https://pith.science/pith/GYRVQFU47JGGWG65QIZFNRGXSQ/action/author_attestation","sign_citation":"https://pith.science/pith/GYRVQFU47JGGWG65QIZFNRGXSQ/action/citation_signature","submit_replication":"https://pith.science/pith/GYRVQFU47JGGWG65QIZFNRGXSQ/action/replication_record"}},"created_at":"2026-07-05T08:25:36.681385+00:00","updated_at":"2026-07-05T08:25:36.681385+00:00"}