{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CYJ5AYXJ25OJOIUQBPRTCC35WW","short_pith_number":"pith:CYJ5AYXJ","schema_version":"1.0","canonical_sha256":"1613d062e9d75c9722900be3310b7db5a036069461795098442c418be4a1b074","source":{"kind":"arxiv","id":"2506.07294","version":3},"attestation_state":"computed","paper":{"title":"Towards Generalized Source Tracing for Codec-Based Deepfake Speech","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CR","cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Haibin Wu, Hung-yi Lee, I-Ming Lin, Jyh-Shing Roger Jang, Lin Zhang, Xuanjun Chen","submitted_at":"2025-06-08T21:36:10Z","abstract_excerpt":"Recent attempts at source tracing for codec-based deepfake speech (CodecFake), generated by neural audio codec-based speech generation (CoSG) models, have exhibited suboptimal performance. However, how to train source tracing models using simulated CoSG data while maintaining strong performance on real CoSG-generated audio remains an open challenge. In this paper, we show that models trained solely on codec-resynthesized data tend to overfit to non-speech regions and struggle to generalize to unseen content. To mitigate these challenges, we introduce the Semantic-Acoustic Source Tracing Networ"},"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":"2506.07294","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SD","submitted_at":"2025-06-08T21:36:10Z","cross_cats_sorted":["cs.CR","cs.LG","eess.AS"],"title_canon_sha256":"2405ed2775b38bd7b81250cc427bc6d57850049041266fc85d9c9a37294a712e","abstract_canon_sha256":"d38a6f776006fb4ee58437e483f0d06b99be76f338dfc76259ccfda8bf813564"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:54:48.091147Z","signature_b64":"wxidGZF8+sthN2OIlms6O17cMEsyC73pbz+Ct8Q29lacn7CCtL+SkCaPb+M8iyditn/W1ltzlHFuy+QhWlc5AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1613d062e9d75c9722900be3310b7db5a036069461795098442c418be4a1b074","last_reissued_at":"2026-07-05T11:54:48.090665Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:54:48.090665Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Generalized Source Tracing for Codec-Based Deepfake Speech","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CR","cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Haibin Wu, Hung-yi Lee, I-Ming Lin, Jyh-Shing Roger Jang, Lin Zhang, Xuanjun Chen","submitted_at":"2025-06-08T21:36:10Z","abstract_excerpt":"Recent attempts at source tracing for codec-based deepfake speech (CodecFake), generated by neural audio codec-based speech generation (CoSG) models, have exhibited suboptimal performance. However, how to train source tracing models using simulated CoSG data while maintaining strong performance on real CoSG-generated audio remains an open challenge. In this paper, we show that models trained solely on codec-resynthesized data tend to overfit to non-speech regions and struggle to generalize to unseen content. To mitigate these challenges, we introduce the Semantic-Acoustic Source Tracing Networ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.07294","kind":"arxiv","version":3},"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/2506.07294/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":"2506.07294","created_at":"2026-07-05T11:54:48.090724+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.07294v3","created_at":"2026-07-05T11:54:48.090724+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.07294","created_at":"2026-07-05T11:54:48.090724+00:00"},{"alias_kind":"pith_short_12","alias_value":"CYJ5AYXJ25OJ","created_at":"2026-07-05T11:54:48.090724+00:00"},{"alias_kind":"pith_short_16","alias_value":"CYJ5AYXJ25OJOIUQ","created_at":"2026-07-05T11:54:48.090724+00:00"},{"alias_kind":"pith_short_8","alias_value":"CYJ5AYXJ","created_at":"2026-07-05T11:54:48.090724+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07494","citing_title":"Mitigating Proxy-to-Wild Domain Gap in Deepfake Speech","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04841","citing_title":"Joint Fullband-Subband Modeling for High-Resolution SingFake Detection","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17642","citing_title":"HCFD: A Benchmark for Audio Deepfake Detection in Healthcare","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19949","citing_title":"Indic-CodecFake meets SATYAM: Towards Detecting Neural Audio Codec Synthesized Speech Deepfakes in Indic Languages","ref_index":155,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CYJ5AYXJ25OJOIUQBPRTCC35WW","json":"https://pith.science/pith/CYJ5AYXJ25OJOIUQBPRTCC35WW.json","graph_json":"https://pith.science/api/pith-number/CYJ5AYXJ25OJOIUQBPRTCC35WW/graph.json","events_json":"https://pith.science/api/pith-number/CYJ5AYXJ25OJOIUQBPRTCC35WW/events.json","paper":"https://pith.science/paper/CYJ5AYXJ"},"agent_actions":{"view_html":"https://pith.science/pith/CYJ5AYXJ25OJOIUQBPRTCC35WW","download_json":"https://pith.science/pith/CYJ5AYXJ25OJOIUQBPRTCC35WW.json","view_paper":"https://pith.science/paper/CYJ5AYXJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.07294&json=true","fetch_graph":"https://pith.science/api/pith-number/CYJ5AYXJ25OJOIUQBPRTCC35WW/graph.json","fetch_events":"https://pith.science/api/pith-number/CYJ5AYXJ25OJOIUQBPRTCC35WW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CYJ5AYXJ25OJOIUQBPRTCC35WW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CYJ5AYXJ25OJOIUQBPRTCC35WW/action/storage_attestation","attest_author":"https://pith.science/pith/CYJ5AYXJ25OJOIUQBPRTCC35WW/action/author_attestation","sign_citation":"https://pith.science/pith/CYJ5AYXJ25OJOIUQBPRTCC35WW/action/citation_signature","submit_replication":"https://pith.science/pith/CYJ5AYXJ25OJOIUQBPRTCC35WW/action/replication_record"}},"created_at":"2026-07-05T11:54:48.090724+00:00","updated_at":"2026-07-05T11:54:48.090724+00:00"}