{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2LDMO7XU36XRQBSZPY7ARSCL7G","short_pith_number":"pith:2LDMO7XU","schema_version":"1.0","canonical_sha256":"d2c6c77ef4dfaf1806597e3e08c84bf9bec20e2fe8312b75eff46a2d4e3c264f","source":{"kind":"arxiv","id":"2503.06142","version":1},"attestation_state":"computed","paper":{"title":"VLForgery Face Triad: Detection, Localization and Attribution via Multimodal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bing Fan, Bin Li, Feng Ding, Guopu Zhu, Xinan He, Yue Zhou","submitted_at":"2025-03-08T09:55:19Z","abstract_excerpt":"Faces synthesized by diffusion models (DMs) with high-quality and controllable attributes pose a significant challenge for Deepfake detection. Most state-of-the-art detectors only yield a binary decision, incapable of forgery localization, attribution of forgery methods, and providing analysis on the cause of forgeries. In this work, we integrate Multimodal Large Language Models (MLLMs) within DM-based face forensics, and propose a fine-grained analysis triad framework called VLForgery, that can 1) predict falsified facial images; 2) locate the falsified face regions subjected to partial synth"},"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":"2503.06142","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-08T09:55:19Z","cross_cats_sorted":[],"title_canon_sha256":"11a245509fbe8bfec153c96f9f84f1a4f5ef8613e8afdfb7ca1518d9dac6d29b","abstract_canon_sha256":"c9fed4662f0d50dfc866acfd61eb1525b5fa877fbca08bfc10445a82d3e71110"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:26:55.137486Z","signature_b64":"l+zif33IIazeIFIS6uPYoPnWS8N8tKIi+ptYOjHTZFL3WwTD881/wMUHrw1QjN7WQq6/Asjp7wRQTI4QJ1pFAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d2c6c77ef4dfaf1806597e3e08c84bf9bec20e2fe8312b75eff46a2d4e3c264f","last_reissued_at":"2026-07-05T10:26:55.136843Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:26:55.136843Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VLForgery Face Triad: Detection, Localization and Attribution via Multimodal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bing Fan, Bin Li, Feng Ding, Guopu Zhu, Xinan He, Yue Zhou","submitted_at":"2025-03-08T09:55:19Z","abstract_excerpt":"Faces synthesized by diffusion models (DMs) with high-quality and controllable attributes pose a significant challenge for Deepfake detection. Most state-of-the-art detectors only yield a binary decision, incapable of forgery localization, attribution of forgery methods, and providing analysis on the cause of forgeries. In this work, we integrate Multimodal Large Language Models (MLLMs) within DM-based face forensics, and propose a fine-grained analysis triad framework called VLForgery, that can 1) predict falsified facial images; 2) locate the falsified face regions subjected to partial synth"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.06142","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/2503.06142/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":"2503.06142","created_at":"2026-07-05T10:26:55.136916+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.06142v1","created_at":"2026-07-05T10:26:55.136916+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.06142","created_at":"2026-07-05T10:26:55.136916+00:00"},{"alias_kind":"pith_short_12","alias_value":"2LDMO7XU36XR","created_at":"2026-07-05T10:26:55.136916+00:00"},{"alias_kind":"pith_short_16","alias_value":"2LDMO7XU36XRQBSZ","created_at":"2026-07-05T10:26:55.136916+00:00"},{"alias_kind":"pith_short_8","alias_value":"2LDMO7XU","created_at":"2026-07-05T10:26:55.136916+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12294","citing_title":"Bridging the Modality Gap in Forensic Image Retrieval","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08709","citing_title":"UniShield: Unified Face Attack Detection via KG-Informed Multimodal Reasoning","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2LDMO7XU36XRQBSZPY7ARSCL7G","json":"https://pith.science/pith/2LDMO7XU36XRQBSZPY7ARSCL7G.json","graph_json":"https://pith.science/api/pith-number/2LDMO7XU36XRQBSZPY7ARSCL7G/graph.json","events_json":"https://pith.science/api/pith-number/2LDMO7XU36XRQBSZPY7ARSCL7G/events.json","paper":"https://pith.science/paper/2LDMO7XU"},"agent_actions":{"view_html":"https://pith.science/pith/2LDMO7XU36XRQBSZPY7ARSCL7G","download_json":"https://pith.science/pith/2LDMO7XU36XRQBSZPY7ARSCL7G.json","view_paper":"https://pith.science/paper/2LDMO7XU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.06142&json=true","fetch_graph":"https://pith.science/api/pith-number/2LDMO7XU36XRQBSZPY7ARSCL7G/graph.json","fetch_events":"https://pith.science/api/pith-number/2LDMO7XU36XRQBSZPY7ARSCL7G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2LDMO7XU36XRQBSZPY7ARSCL7G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2LDMO7XU36XRQBSZPY7ARSCL7G/action/storage_attestation","attest_author":"https://pith.science/pith/2LDMO7XU36XRQBSZPY7ARSCL7G/action/author_attestation","sign_citation":"https://pith.science/pith/2LDMO7XU36XRQBSZPY7ARSCL7G/action/citation_signature","submit_replication":"https://pith.science/pith/2LDMO7XU36XRQBSZPY7ARSCL7G/action/replication_record"}},"created_at":"2026-07-05T10:26:55.136916+00:00","updated_at":"2026-07-05T10:26:55.136916+00:00"}