{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PGFCI7B5ZXCYQBFKMD7QXISQ7E","short_pith_number":"pith:PGFCI7B5","schema_version":"1.0","canonical_sha256":"798a247c3dcdc58804aa60ff0ba250f91fde3c1caf1f0273758c7d4de7ab27da","source":{"kind":"arxiv","id":"2502.03559","version":2},"attestation_state":"computed","paper":{"title":"Comprehensive Layer-wise Analysis of SSL Models for Audio Deepfake Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Sebastian M\\\"oller, Suraj Maharjan, Tim Polzehl, Yassine El Kheir, Youness Samih","submitted_at":"2025-02-05T19:17:24Z","abstract_excerpt":"This paper conducts a comprehensive layer-wise analysis of self-supervised learning (SSL) models for audio deepfake detection across diverse contexts, including multilingual datasets (English, Chinese, Spanish), partial, song, and scene-based deepfake scenarios. By systematically evaluating the contributions of different transformer layers, we uncover critical insights into model behavior and performance. Our findings reveal that lower layers consistently provide the most discriminative features, while higher layers capture less relevant information. Notably, all models achieve competitive equ"},"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":"2502.03559","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2025-02-05T19:17:24Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"752e151efeae1f19ba8071ea13251e1ab09debaff1b0bdce144ad44632535193","abstract_canon_sha256":"fd831b71f908d0c67b43ebc5e42fff7f6451e4bab5a58a76d050dc626d566dc0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:10:49.331552Z","signature_b64":"QsnlQctPYReCa533xpMw8og2UTUk18MqUuLDmNWg33dhqtOMJbco99c+6jWVTov7STJm5n5Xwp7u4a3dephQDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"798a247c3dcdc58804aa60ff0ba250f91fde3c1caf1f0273758c7d4de7ab27da","last_reissued_at":"2026-07-05T10:10:49.330985Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:10:49.330985Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Comprehensive Layer-wise Analysis of SSL Models for Audio Deepfake Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Sebastian M\\\"oller, Suraj Maharjan, Tim Polzehl, Yassine El Kheir, Youness Samih","submitted_at":"2025-02-05T19:17:24Z","abstract_excerpt":"This paper conducts a comprehensive layer-wise analysis of self-supervised learning (SSL) models for audio deepfake detection across diverse contexts, including multilingual datasets (English, Chinese, Spanish), partial, song, and scene-based deepfake scenarios. By systematically evaluating the contributions of different transformer layers, we uncover critical insights into model behavior and performance. Our findings reveal that lower layers consistently provide the most discriminative features, while higher layers capture less relevant information. Notably, all models achieve competitive equ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.03559","kind":"arxiv","version":2},"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/2502.03559/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":"2502.03559","created_at":"2026-07-05T10:10:49.331049+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.03559v2","created_at":"2026-07-05T10:10:49.331049+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.03559","created_at":"2026-07-05T10:10:49.331049+00:00"},{"alias_kind":"pith_short_12","alias_value":"PGFCI7B5ZXCY","created_at":"2026-07-05T10:10:49.331049+00:00"},{"alias_kind":"pith_short_16","alias_value":"PGFCI7B5ZXCYQBFK","created_at":"2026-07-05T10:10:49.331049+00:00"},{"alias_kind":"pith_short_8","alias_value":"PGFCI7B5","created_at":"2026-07-05T10:10:49.331049+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.10758","citing_title":"Anchoring the Unknown: Open-Set Model Attribution via Proxy-Anchor Learning","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08184","citing_title":"AT-ADD: All-Type Audio Deepfake Detection Challenge Evaluation Plan","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PGFCI7B5ZXCYQBFKMD7QXISQ7E","json":"https://pith.science/pith/PGFCI7B5ZXCYQBFKMD7QXISQ7E.json","graph_json":"https://pith.science/api/pith-number/PGFCI7B5ZXCYQBFKMD7QXISQ7E/graph.json","events_json":"https://pith.science/api/pith-number/PGFCI7B5ZXCYQBFKMD7QXISQ7E/events.json","paper":"https://pith.science/paper/PGFCI7B5"},"agent_actions":{"view_html":"https://pith.science/pith/PGFCI7B5ZXCYQBFKMD7QXISQ7E","download_json":"https://pith.science/pith/PGFCI7B5ZXCYQBFKMD7QXISQ7E.json","view_paper":"https://pith.science/paper/PGFCI7B5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.03559&json=true","fetch_graph":"https://pith.science/api/pith-number/PGFCI7B5ZXCYQBFKMD7QXISQ7E/graph.json","fetch_events":"https://pith.science/api/pith-number/PGFCI7B5ZXCYQBFKMD7QXISQ7E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PGFCI7B5ZXCYQBFKMD7QXISQ7E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PGFCI7B5ZXCYQBFKMD7QXISQ7E/action/storage_attestation","attest_author":"https://pith.science/pith/PGFCI7B5ZXCYQBFKMD7QXISQ7E/action/author_attestation","sign_citation":"https://pith.science/pith/PGFCI7B5ZXCYQBFKMD7QXISQ7E/action/citation_signature","submit_replication":"https://pith.science/pith/PGFCI7B5ZXCYQBFKMD7QXISQ7E/action/replication_record"}},"created_at":"2026-07-05T10:10:49.331049+00:00","updated_at":"2026-07-05T10:10:49.331049+00:00"}