{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:IZFBIH72MZHMPGBRSQHB5QPSJG","short_pith_number":"pith:IZFBIH72","schema_version":"1.0","canonical_sha256":"464a141ffa664ec79831940e1ec1f24983ca46cf49f21f9421219bca216ccd3b","source":{"kind":"arxiv","id":"2403.06951","version":2},"attestation_state":"computed","paper":{"title":"DEADiff: An Efficient Stylization Diffusion Model with Disentangled Representations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hongtao Xie, Jiawei Liu, Lang Chen, Qian He, Shancheng Fang, Tianhao Qi, Yanze Wu, Yongdong Zhang","submitted_at":"2024-03-11T17:35:23Z","abstract_excerpt":"The diffusion-based text-to-image model harbors immense potential in transferring reference style. However, current encoder-based approaches significantly impair the text controllability of text-to-image models while transferring styles. In this paper, we introduce DEADiff to address this issue using the following two strategies: 1) a mechanism to decouple the style and semantics of reference images. The decoupled feature representations are first extracted by Q-Formers which are instructed by different text descriptions. Then they are injected into mutually exclusive subsets of cross-attentio"},"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":"2403.06951","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-11T17:35:23Z","cross_cats_sorted":[],"title_canon_sha256":"48455ac1dba1648b7b6995fabc938a5c0d5d1b3888db06bc307118785e88e3cf","abstract_canon_sha256":"a88a6727802bb8e9ce39b977d8732567337821eca2bfe90f11c5d42758517477"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:54:57.480435Z","signature_b64":"vbN3CQZqOAvkfbukqJs+OZ/iBSoaIrGbea57zdxnD9zdSpliTui7kVOYVR4GgVbpiEf6cgUVg4loZlsDykbtAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"464a141ffa664ec79831940e1ec1f24983ca46cf49f21f9421219bca216ccd3b","last_reissued_at":"2026-07-05T07:54:57.479834Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:54:57.479834Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DEADiff: An Efficient Stylization Diffusion Model with Disentangled Representations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hongtao Xie, Jiawei Liu, Lang Chen, Qian He, Shancheng Fang, Tianhao Qi, Yanze Wu, Yongdong Zhang","submitted_at":"2024-03-11T17:35:23Z","abstract_excerpt":"The diffusion-based text-to-image model harbors immense potential in transferring reference style. However, current encoder-based approaches significantly impair the text controllability of text-to-image models while transferring styles. In this paper, we introduce DEADiff to address this issue using the following two strategies: 1) a mechanism to decouple the style and semantics of reference images. The decoupled feature representations are first extracted by Q-Formers which are instructed by different text descriptions. Then they are injected into mutually exclusive subsets of cross-attentio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.06951","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/2403.06951/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":"2403.06951","created_at":"2026-07-05T07:54:57.479916+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.06951v2","created_at":"2026-07-05T07:54:57.479916+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.06951","created_at":"2026-07-05T07:54:57.479916+00:00"},{"alias_kind":"pith_short_12","alias_value":"IZFBIH72MZHM","created_at":"2026-07-05T07:54:57.479916+00:00"},{"alias_kind":"pith_short_16","alias_value":"IZFBIH72MZHMPGBR","created_at":"2026-07-05T07:54:57.479916+00:00"},{"alias_kind":"pith_short_8","alias_value":"IZFBIH72","created_at":"2026-07-05T07:54:57.479916+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2412.12242","citing_title":"OmniPrism: Learning Disentangled Visual Concept for Image Generation","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08364","citing_title":"MegaStyle: Constructing Diverse and Scalable Style Dataset via Consistent Text-to-Image Style Mapping","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IZFBIH72MZHMPGBRSQHB5QPSJG","json":"https://pith.science/pith/IZFBIH72MZHMPGBRSQHB5QPSJG.json","graph_json":"https://pith.science/api/pith-number/IZFBIH72MZHMPGBRSQHB5QPSJG/graph.json","events_json":"https://pith.science/api/pith-number/IZFBIH72MZHMPGBRSQHB5QPSJG/events.json","paper":"https://pith.science/paper/IZFBIH72"},"agent_actions":{"view_html":"https://pith.science/pith/IZFBIH72MZHMPGBRSQHB5QPSJG","download_json":"https://pith.science/pith/IZFBIH72MZHMPGBRSQHB5QPSJG.json","view_paper":"https://pith.science/paper/IZFBIH72","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.06951&json=true","fetch_graph":"https://pith.science/api/pith-number/IZFBIH72MZHMPGBRSQHB5QPSJG/graph.json","fetch_events":"https://pith.science/api/pith-number/IZFBIH72MZHMPGBRSQHB5QPSJG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IZFBIH72MZHMPGBRSQHB5QPSJG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IZFBIH72MZHMPGBRSQHB5QPSJG/action/storage_attestation","attest_author":"https://pith.science/pith/IZFBIH72MZHMPGBRSQHB5QPSJG/action/author_attestation","sign_citation":"https://pith.science/pith/IZFBIH72MZHMPGBRSQHB5QPSJG/action/citation_signature","submit_replication":"https://pith.science/pith/IZFBIH72MZHMPGBRSQHB5QPSJG/action/replication_record"}},"created_at":"2026-07-05T07:54:57.479916+00:00","updated_at":"2026-07-05T07:54:57.479916+00:00"}