{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SFQM7E335PYXIYYSZKTELEOCUH","short_pith_number":"pith:SFQM7E33","schema_version":"1.0","canonical_sha256":"9160cf937bebf1746312caa64591c2a1d7a43dc0a85bf7d3c0006cc99aa769b1","source":{"kind":"arxiv","id":"2503.18032","version":1},"attestation_state":"computed","paper":{"title":"Anomaly Detection and Localization for Speech Deepfakes via Feature Pyramid Matching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.MM"],"primary_cat":"cs.SD","authors_text":"Daniele Ugo Leonzio, Davide Salvi, Emma Coletta, Paolo Bestagini, Viola Negroni","submitted_at":"2025-03-23T11:15:22Z","abstract_excerpt":"The rise of AI-driven generative models has enabled the creation of highly realistic speech deepfakes - synthetic audio signals that can imitate target speakers' voices - raising critical security concerns. Existing methods for detecting speech deepfakes primarily rely on supervised learning, which suffers from two critical limitations: limited generalization to unseen synthesis techniques and a lack of explainability. In this paper, we address these issues by introducing a novel interpretable one-class detection framework, which reframes speech deepfake detection as an anomaly detection task."},"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":"2503.18032","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2025-03-23T11:15:22Z","cross_cats_sorted":["cs.CV","cs.MM"],"title_canon_sha256":"a9603e53f2345d33abef9a49182d94f427269e182bb0960c4d165f0609a75656","abstract_canon_sha256":"632d07905e7dc5e0644cee4e6ff064bb0813c8ee5ac587f7317fb978fa310e85"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:38:00.976843Z","signature_b64":"xSLsykOCQJBocNJuWnH/t05m29obr69FdfieMvETcwzp7uYz5ZneS/yKsM0hk4jwtdky+sGlvtNeEEVV49KOBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9160cf937bebf1746312caa64591c2a1d7a43dc0a85bf7d3c0006cc99aa769b1","last_reissued_at":"2026-07-05T10:38:00.976130Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:38:00.976130Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Anomaly Detection and Localization for Speech Deepfakes via Feature Pyramid Matching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.MM"],"primary_cat":"cs.SD","authors_text":"Daniele Ugo Leonzio, Davide Salvi, Emma Coletta, Paolo Bestagini, Viola Negroni","submitted_at":"2025-03-23T11:15:22Z","abstract_excerpt":"The rise of AI-driven generative models has enabled the creation of highly realistic speech deepfakes - synthetic audio signals that can imitate target speakers' voices - raising critical security concerns. Existing methods for detecting speech deepfakes primarily rely on supervised learning, which suffers from two critical limitations: limited generalization to unseen synthesis techniques and a lack of explainability. In this paper, we address these issues by introducing a novel interpretable one-class detection framework, which reframes speech deepfake detection as an anomaly detection task."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.18032","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/2503.18032/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":"2503.18032","created_at":"2026-07-05T10:38:00.976232+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.18032v1","created_at":"2026-07-05T10:38:00.976232+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.18032","created_at":"2026-07-05T10:38:00.976232+00:00"},{"alias_kind":"pith_short_12","alias_value":"SFQM7E335PYX","created_at":"2026-07-05T10:38:00.976232+00:00"},{"alias_kind":"pith_short_16","alias_value":"SFQM7E335PYXIYYS","created_at":"2026-07-05T10:38:00.976232+00:00"},{"alias_kind":"pith_short_8","alias_value":"SFQM7E33","created_at":"2026-07-05T10:38:00.976232+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.08626","citing_title":"Phoneme-Level Analysis for Person-of-Interest Speech Deepfake Detection","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SFQM7E335PYXIYYSZKTELEOCUH","json":"https://pith.science/pith/SFQM7E335PYXIYYSZKTELEOCUH.json","graph_json":"https://pith.science/api/pith-number/SFQM7E335PYXIYYSZKTELEOCUH/graph.json","events_json":"https://pith.science/api/pith-number/SFQM7E335PYXIYYSZKTELEOCUH/events.json","paper":"https://pith.science/paper/SFQM7E33"},"agent_actions":{"view_html":"https://pith.science/pith/SFQM7E335PYXIYYSZKTELEOCUH","download_json":"https://pith.science/pith/SFQM7E335PYXIYYSZKTELEOCUH.json","view_paper":"https://pith.science/paper/SFQM7E33","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.18032&json=true","fetch_graph":"https://pith.science/api/pith-number/SFQM7E335PYXIYYSZKTELEOCUH/graph.json","fetch_events":"https://pith.science/api/pith-number/SFQM7E335PYXIYYSZKTELEOCUH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SFQM7E335PYXIYYSZKTELEOCUH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SFQM7E335PYXIYYSZKTELEOCUH/action/storage_attestation","attest_author":"https://pith.science/pith/SFQM7E335PYXIYYSZKTELEOCUH/action/author_attestation","sign_citation":"https://pith.science/pith/SFQM7E335PYXIYYSZKTELEOCUH/action/citation_signature","submit_replication":"https://pith.science/pith/SFQM7E335PYXIYYSZKTELEOCUH/action/replication_record"}},"created_at":"2026-07-05T10:38:00.976232+00:00","updated_at":"2026-07-05T10:38:00.976232+00:00"}