{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VOVUNUCX5WLJ7EZRJ7GGS65CWE","short_pith_number":"pith:VOVUNUCX","schema_version":"1.0","canonical_sha256":"abab46d057ed969f93314fcc697ba2b133f292547c930d60a67a80ab9988eb5c","source":{"kind":"arxiv","id":"2507.22824","version":1},"attestation_state":"computed","paper":{"title":"Bi-Level Optimization for Self-Supervised AI-Generated Face Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Baosheng Yu, Kede Ma, Mian Zou, Nan Zhong, Yibing Zhan","submitted_at":"2025-07-30T16:38:29Z","abstract_excerpt":"AI-generated face detectors trained via supervised learning typically rely on synthesized images from specific generators, limiting their generalization to emerging generative techniques. To overcome this limitation, we introduce a self-supervised method based on bi-level optimization. In the inner loop, we pretrain a vision encoder only on photographic face images using a set of linearly weighted pretext tasks: classification of categorical exchangeable image file format (EXIF) tags, ranking of ordinal EXIF tags, and detection of artificial face manipulations. The outer loop then optimizes th"},"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":"2507.22824","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-30T16:38:29Z","cross_cats_sorted":[],"title_canon_sha256":"179ab79f997bd11ee2e385aff0b090be8d0acfcd67c56e84aa99a999455cb3f6","abstract_canon_sha256":"310463a58caead22b9512c52e9e6fee87009b6df29a5bb465f3e6eee4a20383e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:45:49.851028Z","signature_b64":"L8QHUYChDPNykWeM3c/w3NmYPa6ueyYLhRSZJCEZ05ky1/63BEa00ISH86PfKKEMWdtkOH/xxMYp4QySmw2eAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"abab46d057ed969f93314fcc697ba2b133f292547c930d60a67a80ab9988eb5c","last_reissued_at":"2026-07-05T11:45:49.850500Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:45:49.850500Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bi-Level Optimization for Self-Supervised AI-Generated Face Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Baosheng Yu, Kede Ma, Mian Zou, Nan Zhong, Yibing Zhan","submitted_at":"2025-07-30T16:38:29Z","abstract_excerpt":"AI-generated face detectors trained via supervised learning typically rely on synthesized images from specific generators, limiting their generalization to emerging generative techniques. To overcome this limitation, we introduce a self-supervised method based on bi-level optimization. In the inner loop, we pretrain a vision encoder only on photographic face images using a set of linearly weighted pretext tasks: classification of categorical exchangeable image file format (EXIF) tags, ranking of ordinal EXIF tags, and detection of artificial face manipulations. The outer loop then optimizes th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.22824","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/2507.22824/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":"2507.22824","created_at":"2026-07-05T11:45:49.850562+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.22824v1","created_at":"2026-07-05T11:45:49.850562+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.22824","created_at":"2026-07-05T11:45:49.850562+00:00"},{"alias_kind":"pith_short_12","alias_value":"VOVUNUCX5WLJ","created_at":"2026-07-05T11:45:49.850562+00:00"},{"alias_kind":"pith_short_16","alias_value":"VOVUNUCX5WLJ7EZR","created_at":"2026-07-05T11:45:49.850562+00:00"},{"alias_kind":"pith_short_8","alias_value":"VOVUNUCX","created_at":"2026-07-05T11:45:49.850562+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VOVUNUCX5WLJ7EZRJ7GGS65CWE","json":"https://pith.science/pith/VOVUNUCX5WLJ7EZRJ7GGS65CWE.json","graph_json":"https://pith.science/api/pith-number/VOVUNUCX5WLJ7EZRJ7GGS65CWE/graph.json","events_json":"https://pith.science/api/pith-number/VOVUNUCX5WLJ7EZRJ7GGS65CWE/events.json","paper":"https://pith.science/paper/VOVUNUCX"},"agent_actions":{"view_html":"https://pith.science/pith/VOVUNUCX5WLJ7EZRJ7GGS65CWE","download_json":"https://pith.science/pith/VOVUNUCX5WLJ7EZRJ7GGS65CWE.json","view_paper":"https://pith.science/paper/VOVUNUCX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.22824&json=true","fetch_graph":"https://pith.science/api/pith-number/VOVUNUCX5WLJ7EZRJ7GGS65CWE/graph.json","fetch_events":"https://pith.science/api/pith-number/VOVUNUCX5WLJ7EZRJ7GGS65CWE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VOVUNUCX5WLJ7EZRJ7GGS65CWE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VOVUNUCX5WLJ7EZRJ7GGS65CWE/action/storage_attestation","attest_author":"https://pith.science/pith/VOVUNUCX5WLJ7EZRJ7GGS65CWE/action/author_attestation","sign_citation":"https://pith.science/pith/VOVUNUCX5WLJ7EZRJ7GGS65CWE/action/citation_signature","submit_replication":"https://pith.science/pith/VOVUNUCX5WLJ7EZRJ7GGS65CWE/action/replication_record"}},"created_at":"2026-07-05T11:45:49.850562+00:00","updated_at":"2026-07-05T11:45:49.850562+00:00"}