{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:5G6US73PVV5DOUC5UDVNUNPFHO","short_pith_number":"pith:5G6US73P","schema_version":"1.0","canonical_sha256":"e9bd497f6fad7a37505da0eada35e53ba67d3380085a8d1b5892125c72e69701","source":{"kind":"arxiv","id":"2607.09891","version":1},"attestation_state":"computed","paper":{"title":"What You Train Is What You Get: Gender Bias, Training Composition, and Post-Hoc Mitigation in Audio Deepfake Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SD","authors_text":"Aishwarya R. Fursule, Anderson R. Avila, Shruti Kshirsagar, Vamshi Nallaguntla","submitted_at":"2026-07-10T18:29:24Z","abstract_excerpt":"Audio deepfake detection models determine whether speech is genuine or artificially generated, but high overall accuracy can mask substantial performance disparities across demographic groups. In this work, we investigate gender bias in audio deepfake detection using the ASVspoof5 dataset. We use ASVspoof5 under a controlled custom split designed to isolate gender-composition effects. We train attack-specific models on nine training sets with different gender compositions, ranging from female-only to male-only. We use a ResNet18 classifier with LogSpectrogram and WavLM-Base+ features, and we e"},"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":"2607.09891","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2026-07-10T18:29:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"39121737c91f7dfe05db7e25755924a2258a086aacbfe68e9d929cbe3a8e9407","abstract_canon_sha256":"9d3b4d1c5131f3184cd50367b9be5775a60dc04bc398288af855fc0f5b7bc015"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T00:18:41.759270Z","signature_b64":"E3laEPtWeJTkf6Wg/fIoV0PXfTxLpmUd9F+7ap+e1KzAgIHmWfZXsfFkOMf9WdooFmgNjR4eajM9cH6olUlgAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e9bd497f6fad7a37505da0eada35e53ba67d3380085a8d1b5892125c72e69701","last_reissued_at":"2026-07-14T00:18:41.758461Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T00:18:41.758461Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"What You Train Is What You Get: Gender Bias, Training Composition, and Post-Hoc Mitigation in Audio Deepfake Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SD","authors_text":"Aishwarya R. Fursule, Anderson R. Avila, Shruti Kshirsagar, Vamshi Nallaguntla","submitted_at":"2026-07-10T18:29:24Z","abstract_excerpt":"Audio deepfake detection models determine whether speech is genuine or artificially generated, but high overall accuracy can mask substantial performance disparities across demographic groups. In this work, we investigate gender bias in audio deepfake detection using the ASVspoof5 dataset. We use ASVspoof5 under a controlled custom split designed to isolate gender-composition effects. We train attack-specific models on nine training sets with different gender compositions, ranging from female-only to male-only. We use a ResNet18 classifier with LogSpectrogram and WavLM-Base+ features, and we e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.09891","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/2607.09891/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":"2607.09891","created_at":"2026-07-14T00:18:41.758880+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.09891v1","created_at":"2026-07-14T00:18:41.758880+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.09891","created_at":"2026-07-14T00:18:41.758880+00:00"},{"alias_kind":"pith_short_12","alias_value":"5G6US73PVV5D","created_at":"2026-07-14T00:18:41.758880+00:00"},{"alias_kind":"pith_short_16","alias_value":"5G6US73PVV5DOUC5","created_at":"2026-07-14T00:18:41.758880+00:00"},{"alias_kind":"pith_short_8","alias_value":"5G6US73P","created_at":"2026-07-14T00:18:41.758880+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/5G6US73PVV5DOUC5UDVNUNPFHO","json":"https://pith.science/pith/5G6US73PVV5DOUC5UDVNUNPFHO.json","graph_json":"https://pith.science/api/pith-number/5G6US73PVV5DOUC5UDVNUNPFHO/graph.json","events_json":"https://pith.science/api/pith-number/5G6US73PVV5DOUC5UDVNUNPFHO/events.json","paper":"https://pith.science/paper/5G6US73P"},"agent_actions":{"view_html":"https://pith.science/pith/5G6US73PVV5DOUC5UDVNUNPFHO","download_json":"https://pith.science/pith/5G6US73PVV5DOUC5UDVNUNPFHO.json","view_paper":"https://pith.science/paper/5G6US73P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.09891&json=true","fetch_graph":"https://pith.science/api/pith-number/5G6US73PVV5DOUC5UDVNUNPFHO/graph.json","fetch_events":"https://pith.science/api/pith-number/5G6US73PVV5DOUC5UDVNUNPFHO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5G6US73PVV5DOUC5UDVNUNPFHO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5G6US73PVV5DOUC5UDVNUNPFHO/action/storage_attestation","attest_author":"https://pith.science/pith/5G6US73PVV5DOUC5UDVNUNPFHO/action/author_attestation","sign_citation":"https://pith.science/pith/5G6US73PVV5DOUC5UDVNUNPFHO/action/citation_signature","submit_replication":"https://pith.science/pith/5G6US73PVV5DOUC5UDVNUNPFHO/action/replication_record"}},"created_at":"2026-07-14T00:18:41.758880+00:00","updated_at":"2026-07-14T00:18:41.758880+00:00"}