{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PQ4444HRYYPUJIWKX5ZPDYDHDA","short_pith_number":"pith:PQ4444HR","schema_version":"1.0","canonical_sha256":"7c39ce70f1c61f44a2cabf72f1e06718014c7a53df3c04900a38f567027d1039","source":{"kind":"arxiv","id":"2404.08788","version":1},"attestation_state":"computed","paper":{"title":"Detecting AI-Generated Images via CLIP","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"A.G. Moskowitz, J. Peterson, T. Gaona","submitted_at":"2024-04-12T19:29:10Z","abstract_excerpt":"As AI-generated image (AIGI) methods become more powerful and accessible, it has become a critical task to determine if an image is real or AI-generated. Because AIGI lack the signatures of photographs and have their own unique patterns, new models are needed to determine if an image is AI-generated. In this paper, we investigate the ability of the Contrastive Language-Image Pre-training (CLIP) architecture, pre-trained on massive internet-scale data sets, to perform this differentiation. We fine-tune CLIP on real images and AIGI from several generative models, enabling CLIP to determine if an"},"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":"2404.08788","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-04-12T19:29:10Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4ccf5ed9b3d753ca95af507dd5631741b124533360f23c1263950dfb4b7d908f","abstract_canon_sha256":"150d32e34b28d864d936d2fd95c374d52bcadd64fee989d77ecbe80efc20c599"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:07:40.107209Z","signature_b64":"7VomjtpZ/vEGV4jt2UbQ0egatGPC2P329rd6pWWZxpw0tdgNid9LmVAr0mcbBBy0H324q3vqrPvwl31COwsrAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7c39ce70f1c61f44a2cabf72f1e06718014c7a53df3c04900a38f567027d1039","last_reissued_at":"2026-07-05T08:07:40.106675Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:07:40.106675Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Detecting AI-Generated Images via CLIP","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"A.G. Moskowitz, J. Peterson, T. Gaona","submitted_at":"2024-04-12T19:29:10Z","abstract_excerpt":"As AI-generated image (AIGI) methods become more powerful and accessible, it has become a critical task to determine if an image is real or AI-generated. Because AIGI lack the signatures of photographs and have their own unique patterns, new models are needed to determine if an image is AI-generated. In this paper, we investigate the ability of the Contrastive Language-Image Pre-training (CLIP) architecture, pre-trained on massive internet-scale data sets, to perform this differentiation. We fine-tune CLIP on real images and AIGI from several generative models, enabling CLIP to determine if an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.08788","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/2404.08788/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":"2404.08788","created_at":"2026-07-05T08:07:40.106728+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.08788v1","created_at":"2026-07-05T08:07:40.106728+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.08788","created_at":"2026-07-05T08:07:40.106728+00:00"},{"alias_kind":"pith_short_12","alias_value":"PQ4444HRYYPU","created_at":"2026-07-05T08:07:40.106728+00:00"},{"alias_kind":"pith_short_16","alias_value":"PQ4444HRYYPUJIWK","created_at":"2026-07-05T08:07:40.106728+00:00"},{"alias_kind":"pith_short_8","alias_value":"PQ4444HR","created_at":"2026-07-05T08:07:40.106728+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02224","citing_title":"Chroma Clues: Leveraging Color Statistics to Detect Synthetic Images","ref_index":44,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PQ4444HRYYPUJIWKX5ZPDYDHDA","json":"https://pith.science/pith/PQ4444HRYYPUJIWKX5ZPDYDHDA.json","graph_json":"https://pith.science/api/pith-number/PQ4444HRYYPUJIWKX5ZPDYDHDA/graph.json","events_json":"https://pith.science/api/pith-number/PQ4444HRYYPUJIWKX5ZPDYDHDA/events.json","paper":"https://pith.science/paper/PQ4444HR"},"agent_actions":{"view_html":"https://pith.science/pith/PQ4444HRYYPUJIWKX5ZPDYDHDA","download_json":"https://pith.science/pith/PQ4444HRYYPUJIWKX5ZPDYDHDA.json","view_paper":"https://pith.science/paper/PQ4444HR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.08788&json=true","fetch_graph":"https://pith.science/api/pith-number/PQ4444HRYYPUJIWKX5ZPDYDHDA/graph.json","fetch_events":"https://pith.science/api/pith-number/PQ4444HRYYPUJIWKX5ZPDYDHDA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PQ4444HRYYPUJIWKX5ZPDYDHDA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PQ4444HRYYPUJIWKX5ZPDYDHDA/action/storage_attestation","attest_author":"https://pith.science/pith/PQ4444HRYYPUJIWKX5ZPDYDHDA/action/author_attestation","sign_citation":"https://pith.science/pith/PQ4444HRYYPUJIWKX5ZPDYDHDA/action/citation_signature","submit_replication":"https://pith.science/pith/PQ4444HRYYPUJIWKX5ZPDYDHDA/action/replication_record"}},"created_at":"2026-07-05T08:07:40.106728+00:00","updated_at":"2026-07-05T08:07:40.106728+00:00"}