{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6CNFGZ4ZDAGG7LSN5GQF5S6IPC","short_pith_number":"pith:6CNFGZ4Z","schema_version":"1.0","canonical_sha256":"f09a536799180c6fae4de9a05ecbc878afa03e0e870708e63b0c53632b0738c4","source":{"kind":"arxiv","id":"2506.11532","version":1},"attestation_state":"computed","paper":{"title":"From Sharpness to Better Generalization for Speech Deepfake Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Junichi Yamagishi, Wen Huang, Xin Wang, Xuechen Liu, Yanmin Qian","submitted_at":"2025-06-13T07:36:31Z","abstract_excerpt":"Generalization remains a critical challenge in speech deepfake detection (SDD). While various approaches aim to improve robustness, generalization is typically assessed through performance metrics like equal error rate without a theoretical framework to explain model performance. This work investigates sharpness as a theoretical proxy for generalization in SDD. We analyze how sharpness responds to domain shifts and find it increases in unseen conditions, indicating higher model sensitivity. Based on this, we apply Sharpness-Aware Minimization (SAM) to reduce sharpness explicitly, leading to be"},"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":"2506.11532","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2025-06-13T07:36:31Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"af2e7eb875518cd5f6f6e1b3fb6fb4a0db7f9300edc1ed3ed17ae33872ab9e54","abstract_canon_sha256":"40f55d1412865abaf9b68b395e6a0c8bdc9c8be9b8fed59442a5af8bd83ed0c0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:06.197721Z","signature_b64":"692YOq1kTAlCdAQKIYHd9VVJVO5kclViZ/UUbLsmu52fENh21h34cJIYPDwSsQ/rmIefiaoTa0jajbYOiq6UCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f09a536799180c6fae4de9a05ecbc878afa03e0e870708e63b0c53632b0738c4","last_reissued_at":"2026-07-05T11:21:06.197178Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:06.197178Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Sharpness to Better Generalization for Speech Deepfake Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Junichi Yamagishi, Wen Huang, Xin Wang, Xuechen Liu, Yanmin Qian","submitted_at":"2025-06-13T07:36:31Z","abstract_excerpt":"Generalization remains a critical challenge in speech deepfake detection (SDD). While various approaches aim to improve robustness, generalization is typically assessed through performance metrics like equal error rate without a theoretical framework to explain model performance. This work investigates sharpness as a theoretical proxy for generalization in SDD. We analyze how sharpness responds to domain shifts and find it increases in unseen conditions, indicating higher model sensitivity. Based on this, we apply Sharpness-Aware Minimization (SAM) to reduce sharpness explicitly, leading to be"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.11532","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/2506.11532/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":"2506.11532","created_at":"2026-07-05T11:21:06.197238+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.11532v1","created_at":"2026-07-05T11:21:06.197238+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.11532","created_at":"2026-07-05T11:21:06.197238+00:00"},{"alias_kind":"pith_short_12","alias_value":"6CNFGZ4ZDAGG","created_at":"2026-07-05T11:21:06.197238+00:00"},{"alias_kind":"pith_short_16","alias_value":"6CNFGZ4ZDAGG7LSN","created_at":"2026-07-05T11:21:06.197238+00:00"},{"alias_kind":"pith_short_8","alias_value":"6CNFGZ4Z","created_at":"2026-07-05T11:21:06.197238+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.28662","citing_title":"Closed-Form Steepest Descent Direction toward Flat Minima: Reducing Upper Bounds on the Loss Hessian Eigenspectrum in Neural Networks","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10202","citing_title":"Wolkowicz-Styan Upper Bound on the Hessian Eigenspectrum for Cross-Entropy Loss in Nonlinear Smooth Neural Networks","ref_index":41,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6CNFGZ4ZDAGG7LSN5GQF5S6IPC","json":"https://pith.science/pith/6CNFGZ4ZDAGG7LSN5GQF5S6IPC.json","graph_json":"https://pith.science/api/pith-number/6CNFGZ4ZDAGG7LSN5GQF5S6IPC/graph.json","events_json":"https://pith.science/api/pith-number/6CNFGZ4ZDAGG7LSN5GQF5S6IPC/events.json","paper":"https://pith.science/paper/6CNFGZ4Z"},"agent_actions":{"view_html":"https://pith.science/pith/6CNFGZ4ZDAGG7LSN5GQF5S6IPC","download_json":"https://pith.science/pith/6CNFGZ4ZDAGG7LSN5GQF5S6IPC.json","view_paper":"https://pith.science/paper/6CNFGZ4Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.11532&json=true","fetch_graph":"https://pith.science/api/pith-number/6CNFGZ4ZDAGG7LSN5GQF5S6IPC/graph.json","fetch_events":"https://pith.science/api/pith-number/6CNFGZ4ZDAGG7LSN5GQF5S6IPC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6CNFGZ4ZDAGG7LSN5GQF5S6IPC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6CNFGZ4ZDAGG7LSN5GQF5S6IPC/action/storage_attestation","attest_author":"https://pith.science/pith/6CNFGZ4ZDAGG7LSN5GQF5S6IPC/action/author_attestation","sign_citation":"https://pith.science/pith/6CNFGZ4ZDAGG7LSN5GQF5S6IPC/action/citation_signature","submit_replication":"https://pith.science/pith/6CNFGZ4ZDAGG7LSN5GQF5S6IPC/action/replication_record"}},"created_at":"2026-07-05T11:21:06.197238+00:00","updated_at":"2026-07-05T11:21:06.197238+00:00"}