{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TKFFUWTVWQ5MHRFOOED4T55V5I","short_pith_number":"pith:TKFFUWTV","schema_version":"1.0","canonical_sha256":"9a8a5a5a75b43ac3c4ae7107c9f7b5ea1ec7cffda37a692a4ec0f95dfd92c9b7","source":{"kind":"arxiv","id":"2606.26384","version":1},"attestation_state":"computed","paper":{"title":"What Do Deepfake Benchmarks Measure? An Audit Using Frozen Self-Supervised Representations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Feng Liu, Samuel Pagon, Vishal Asnani, Yixuan Shen","submitted_at":"2026-06-24T21:10:42Z","abstract_excerpt":"As deepfake generators approach perceptual indistinguishability, reliable detection becomes critical. Yet, detectors that score well on benchmarks routinely fail in the wild. A concerning feedback loop has emerged: benchmarks drive increasingly complex, engineered detectors, yet if those benchmarks do not reflect real-world deepfakes, this complexity may be solving the wrong problem entirely. This raises a prior question: what are these benchmarks actually measuring? We conduct an audit of video, image, and audio deepfake benchmarks using a deliberately simple diagnostic. If a linear probe on "},"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":"2606.26384","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-06-24T21:10:42Z","cross_cats_sorted":[],"title_canon_sha256":"f1efe5908452af106ee3add98ad25afcf18f1b2777658a4f506a5449abe27d4c","abstract_canon_sha256":"884b1f8badadbb1d5a8deff129f126a041739167c87245181d10cf13b97a2e24"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-26T00:15:38.881084Z","signature_b64":"9RapbDYR6r/NdEE0Pl0gtVu8suRsfVE/J4s9FQrDXkvH6/Neq3MSEorY6HDRD49Y0W7sQMHN1obvdTo48w6jAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9a8a5a5a75b43ac3c4ae7107c9f7b5ea1ec7cffda37a692a4ec0f95dfd92c9b7","last_reissued_at":"2026-06-26T00:15:38.880673Z","signature_status":"signed_v1","first_computed_at":"2026-06-26T00:15:38.880673Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"What Do Deepfake Benchmarks Measure? An Audit Using Frozen Self-Supervised Representations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Feng Liu, Samuel Pagon, Vishal Asnani, Yixuan Shen","submitted_at":"2026-06-24T21:10:42Z","abstract_excerpt":"As deepfake generators approach perceptual indistinguishability, reliable detection becomes critical. Yet, detectors that score well on benchmarks routinely fail in the wild. A concerning feedback loop has emerged: benchmarks drive increasingly complex, engineered detectors, yet if those benchmarks do not reflect real-world deepfakes, this complexity may be solving the wrong problem entirely. This raises a prior question: what are these benchmarks actually measuring? We conduct an audit of video, image, and audio deepfake benchmarks using a deliberately simple diagnostic. If a linear probe on "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.26384","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/2606.26384/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":"2606.26384","created_at":"2026-06-26T00:15:38.880732+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.26384v1","created_at":"2026-06-26T00:15:38.880732+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.26384","created_at":"2026-06-26T00:15:38.880732+00:00"},{"alias_kind":"pith_short_12","alias_value":"TKFFUWTVWQ5M","created_at":"2026-06-26T00:15:38.880732+00:00"},{"alias_kind":"pith_short_16","alias_value":"TKFFUWTVWQ5MHRFO","created_at":"2026-06-26T00:15:38.880732+00:00"},{"alias_kind":"pith_short_8","alias_value":"TKFFUWTV","created_at":"2026-06-26T00:15:38.880732+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/TKFFUWTVWQ5MHRFOOED4T55V5I","json":"https://pith.science/pith/TKFFUWTVWQ5MHRFOOED4T55V5I.json","graph_json":"https://pith.science/api/pith-number/TKFFUWTVWQ5MHRFOOED4T55V5I/graph.json","events_json":"https://pith.science/api/pith-number/TKFFUWTVWQ5MHRFOOED4T55V5I/events.json","paper":"https://pith.science/paper/TKFFUWTV"},"agent_actions":{"view_html":"https://pith.science/pith/TKFFUWTVWQ5MHRFOOED4T55V5I","download_json":"https://pith.science/pith/TKFFUWTVWQ5MHRFOOED4T55V5I.json","view_paper":"https://pith.science/paper/TKFFUWTV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.26384&json=true","fetch_graph":"https://pith.science/api/pith-number/TKFFUWTVWQ5MHRFOOED4T55V5I/graph.json","fetch_events":"https://pith.science/api/pith-number/TKFFUWTVWQ5MHRFOOED4T55V5I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TKFFUWTVWQ5MHRFOOED4T55V5I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TKFFUWTVWQ5MHRFOOED4T55V5I/action/storage_attestation","attest_author":"https://pith.science/pith/TKFFUWTVWQ5MHRFOOED4T55V5I/action/author_attestation","sign_citation":"https://pith.science/pith/TKFFUWTVWQ5MHRFOOED4T55V5I/action/citation_signature","submit_replication":"https://pith.science/pith/TKFFUWTVWQ5MHRFOOED4T55V5I/action/replication_record"}},"created_at":"2026-06-26T00:15:38.880732+00:00","updated_at":"2026-06-26T00:15:38.880732+00:00"}