{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:67C6XAD5RCZPJ4KMWMOWIYOALY","short_pith_number":"pith:67C6XAD5","schema_version":"1.0","canonical_sha256":"f7c5eb807d88b2f4f14cb31d6461c05e12736b32f1847dac5f70ac20336aed8a","source":{"kind":"arxiv","id":"2507.12232","version":1},"attestation_state":"computed","paper":{"title":"MGFFD-VLM: Multi-Granularity Prompt Learning for Face Forgery Detection with VLM","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chunlei Peng, Decheng Liu, Jingyi Zhang, Tao Chen","submitted_at":"2025-07-16T13:47:13Z","abstract_excerpt":"Recent studies have utilized visual large language models (VLMs) to answer not only \"Is this face a forgery?\" but also \"Why is the face a forgery?\" These studies introduced forgery-related attributes, such as forgery location and type, to construct deepfake VQA datasets and train VLMs, achieving high accuracy while providing human-understandable explanatory text descriptions. However, these methods still have limitations. For example, they do not fully leverage face quality-related attributes, which are often abnormal in forged faces, and they lack effective training strategies for forgery-awa"},"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":"2507.12232","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-16T13:47:13Z","cross_cats_sorted":[],"title_canon_sha256":"52760da07cc4d9fdf398012ba6b5a138e9d07eaa89986887a9f930e4e767f8a8","abstract_canon_sha256":"73eb4c62445804e91abbbdf1a23b3d95b276c351c2db108ea35708a2915328a2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:38:17.141470Z","signature_b64":"Fh2wENJHSPVtCHEc6XRtIa47igLIrBXDIM+iE5PaD0QRkst+68cYxBs4E0kZEOyCal9EsG3v7pSk4DChpRVQDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f7c5eb807d88b2f4f14cb31d6461c05e12736b32f1847dac5f70ac20336aed8a","last_reissued_at":"2026-07-05T11:38:17.140974Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:38:17.140974Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MGFFD-VLM: Multi-Granularity Prompt Learning for Face Forgery Detection with VLM","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chunlei Peng, Decheng Liu, Jingyi Zhang, Tao Chen","submitted_at":"2025-07-16T13:47:13Z","abstract_excerpt":"Recent studies have utilized visual large language models (VLMs) to answer not only \"Is this face a forgery?\" but also \"Why is the face a forgery?\" These studies introduced forgery-related attributes, such as forgery location and type, to construct deepfake VQA datasets and train VLMs, achieving high accuracy while providing human-understandable explanatory text descriptions. However, these methods still have limitations. For example, they do not fully leverage face quality-related attributes, which are often abnormal in forged faces, and they lack effective training strategies for forgery-awa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.12232","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/2507.12232/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":"2507.12232","created_at":"2026-07-05T11:38:17.141029+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.12232v1","created_at":"2026-07-05T11:38:17.141029+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.12232","created_at":"2026-07-05T11:38:17.141029+00:00"},{"alias_kind":"pith_short_12","alias_value":"67C6XAD5RCZP","created_at":"2026-07-05T11:38:17.141029+00:00"},{"alias_kind":"pith_short_16","alias_value":"67C6XAD5RCZPJ4KM","created_at":"2026-07-05T11:38:17.141029+00:00"},{"alias_kind":"pith_short_8","alias_value":"67C6XAD5","created_at":"2026-07-05T11:38:17.141029+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.17489","citing_title":"Employing Vision-Language Models for Face Image Quality Assessment","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/67C6XAD5RCZPJ4KMWMOWIYOALY","json":"https://pith.science/pith/67C6XAD5RCZPJ4KMWMOWIYOALY.json","graph_json":"https://pith.science/api/pith-number/67C6XAD5RCZPJ4KMWMOWIYOALY/graph.json","events_json":"https://pith.science/api/pith-number/67C6XAD5RCZPJ4KMWMOWIYOALY/events.json","paper":"https://pith.science/paper/67C6XAD5"},"agent_actions":{"view_html":"https://pith.science/pith/67C6XAD5RCZPJ4KMWMOWIYOALY","download_json":"https://pith.science/pith/67C6XAD5RCZPJ4KMWMOWIYOALY.json","view_paper":"https://pith.science/paper/67C6XAD5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.12232&json=true","fetch_graph":"https://pith.science/api/pith-number/67C6XAD5RCZPJ4KMWMOWIYOALY/graph.json","fetch_events":"https://pith.science/api/pith-number/67C6XAD5RCZPJ4KMWMOWIYOALY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/67C6XAD5RCZPJ4KMWMOWIYOALY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/67C6XAD5RCZPJ4KMWMOWIYOALY/action/storage_attestation","attest_author":"https://pith.science/pith/67C6XAD5RCZPJ4KMWMOWIYOALY/action/author_attestation","sign_citation":"https://pith.science/pith/67C6XAD5RCZPJ4KMWMOWIYOALY/action/citation_signature","submit_replication":"https://pith.science/pith/67C6XAD5RCZPJ4KMWMOWIYOALY/action/replication_record"}},"created_at":"2026-07-05T11:38:17.141029+00:00","updated_at":"2026-07-05T11:38:17.141029+00:00"}