{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:NNJKLWKTWR4GLXCMA35AJTQNNC","short_pith_number":"pith:NNJKLWKT","schema_version":"1.0","canonical_sha256":"6b52a5d953b47865dc4c06fa04ce0d68950af50fd01125e18089a092583fe27d","source":{"kind":"arxiv","id":"2304.13949","version":2},"attestation_state":"computed","paper":{"title":"UCF: Uncovering Common Features for Generalizable Deepfake Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Baoyuan Wu, Yanbo Fan, Yong Zhang, Zhiyuan Yan","submitted_at":"2023-04-27T04:07:29Z","abstract_excerpt":"Deepfake detection remains a challenging task due to the difficulty of generalizing to new types of forgeries. This problem primarily stems from the overfitting of existing detection methods to forgery-irrelevant features and method-specific patterns. The latter has been rarely studied and not well addressed by previous works. This paper presents a novel approach to address the two types of overfitting issues by uncovering common forgery features. Specifically, we first propose a disentanglement framework that decomposes image information into three distinct components: forgery-irrelevant, met"},"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":"2304.13949","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-04-27T04:07:29Z","cross_cats_sorted":[],"title_canon_sha256":"62e78fe9e3b0104e6805f755ac3967a2a3ced34e85bcd657bb2083c631c97f43","abstract_canon_sha256":"90fecfd3e375a24ff3cc5f7b3b28017b311559df0e04489fd8bcdd712f24a8ed"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:06:26.161290Z","signature_b64":"bWrl6z2CjSvDYOvft8RkV29OQOSpj991iiQPuoJsdZpB992e2awsX9QrNKPFHDEZPkTD0W1IanOkuSA8FjKIDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6b52a5d953b47865dc4c06fa04ce0d68950af50fd01125e18089a092583fe27d","last_reissued_at":"2026-07-05T07:06:26.160766Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:06:26.160766Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UCF: Uncovering Common Features for Generalizable Deepfake Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Baoyuan Wu, Yanbo Fan, Yong Zhang, Zhiyuan Yan","submitted_at":"2023-04-27T04:07:29Z","abstract_excerpt":"Deepfake detection remains a challenging task due to the difficulty of generalizing to new types of forgeries. This problem primarily stems from the overfitting of existing detection methods to forgery-irrelevant features and method-specific patterns. The latter has been rarely studied and not well addressed by previous works. This paper presents a novel approach to address the two types of overfitting issues by uncovering common forgery features. Specifically, we first propose a disentanglement framework that decomposes image information into three distinct components: forgery-irrelevant, met"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.13949","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/2304.13949/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":"2304.13949","created_at":"2026-07-05T07:06:26.160830+00:00"},{"alias_kind":"arxiv_version","alias_value":"2304.13949v2","created_at":"2026-07-05T07:06:26.160830+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.13949","created_at":"2026-07-05T07:06:26.160830+00:00"},{"alias_kind":"pith_short_12","alias_value":"NNJKLWKTWR4G","created_at":"2026-07-05T07:06:26.160830+00:00"},{"alias_kind":"pith_short_16","alias_value":"NNJKLWKTWR4GLXCM","created_at":"2026-07-05T07:06:26.160830+00:00"},{"alias_kind":"pith_short_8","alias_value":"NNJKLWKT","created_at":"2026-07-05T07:06:26.160830+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09881","citing_title":"Toward Calibrated, Fair, and accurate Deepfake Detection","ref_index":82,"is_internal_anchor":false},{"citing_arxiv_id":"2411.15633","citing_title":"Orthogonal Subspace Decomposition for Generalizable AI-Generated Image Detection","ref_index":95,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14570","citing_title":"Deepfake Detection Generalization with Diffusion Noise","ref_index":76,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NNJKLWKTWR4GLXCMA35AJTQNNC","json":"https://pith.science/pith/NNJKLWKTWR4GLXCMA35AJTQNNC.json","graph_json":"https://pith.science/api/pith-number/NNJKLWKTWR4GLXCMA35AJTQNNC/graph.json","events_json":"https://pith.science/api/pith-number/NNJKLWKTWR4GLXCMA35AJTQNNC/events.json","paper":"https://pith.science/paper/NNJKLWKT"},"agent_actions":{"view_html":"https://pith.science/pith/NNJKLWKTWR4GLXCMA35AJTQNNC","download_json":"https://pith.science/pith/NNJKLWKTWR4GLXCMA35AJTQNNC.json","view_paper":"https://pith.science/paper/NNJKLWKT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2304.13949&json=true","fetch_graph":"https://pith.science/api/pith-number/NNJKLWKTWR4GLXCMA35AJTQNNC/graph.json","fetch_events":"https://pith.science/api/pith-number/NNJKLWKTWR4GLXCMA35AJTQNNC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NNJKLWKTWR4GLXCMA35AJTQNNC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NNJKLWKTWR4GLXCMA35AJTQNNC/action/storage_attestation","attest_author":"https://pith.science/pith/NNJKLWKTWR4GLXCMA35AJTQNNC/action/author_attestation","sign_citation":"https://pith.science/pith/NNJKLWKTWR4GLXCMA35AJTQNNC/action/citation_signature","submit_replication":"https://pith.science/pith/NNJKLWKTWR4GLXCMA35AJTQNNC/action/replication_record"}},"created_at":"2026-07-05T07:06:26.160830+00:00","updated_at":"2026-07-05T07:06:26.160830+00:00"}