{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CBX6EKEYOYF5ZD3PH5S7OTKPLC","short_pith_number":"pith:CBX6EKEY","schema_version":"1.0","canonical_sha256":"106fe22898760bdc8f6f3f65f74d4f58a92626000d2396b840262d759e98f05c","source":{"kind":"arxiv","id":"2605.18409","version":1},"attestation_state":"computed","paper":{"title":"EnvTriCascade: An Environment-Aware Tri-Stage Cascaded Framework for ESDD2 2026 Challenge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A tri-stage cascaded detector first checks mix consistency then fuses two five-class classifiers via attention to reach 0.8266 Macro-F1 on the ESDD2 test set.","cross_cats":[],"primary_cat":"cs.SD","authors_text":"Haonan Cheng, Hengyan Huang, Jian Liu, Jiayi Zhou, Long Ye, Qin Zhang, Xiaoxuan Guo, Yuankun Xie","submitted_at":"2026-05-18T13:48:52Z","abstract_excerpt":"ADD in real-world scenarios has evolved from speech-only spoofing to more challenging component-level settings, where speech and environmental sounds may be independently manipulated. To tackle this, we propose EnvTriCascade, an Environment-Aware Tri-Stage Cascaded framework for the ESDD2 Challenge. First, a mix-consistency detector provides a binary prior to distinguish original recordings from manipulated mixtures, which calibrates the final decisions. Next, two complementary five-class detectors, leveraging SSLAM+XLS-R and EAT-large+XLS-R representations, extract robust multi-branch feature"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.18409","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2026-05-18T13:48:52Z","cross_cats_sorted":[],"title_canon_sha256":"547fd1155152450b3fe83913e91d13c2e898a754874b802f7a56aa931715bfde","abstract_canon_sha256":"a09748cfbb26a1e33a50d3e67fda17f7e9c68e6d9ef77d08343ef8d86ae8c4a4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:05:59.345153Z","signature_b64":"wqorMbfQxn+TlcefoaoXsUbOTJCH2o0yp7e8czTz2YHqduQUNzAHT01rggv4mKjL3NXGFQ+JcOgQn4DRtRp6Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"106fe22898760bdc8f6f3f65f74d4f58a92626000d2396b840262d759e98f05c","last_reissued_at":"2026-05-20T00:05:59.344384Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:05:59.344384Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EnvTriCascade: An Environment-Aware Tri-Stage Cascaded Framework for ESDD2 2026 Challenge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A tri-stage cascaded detector first checks mix consistency then fuses two five-class classifiers via attention to reach 0.8266 Macro-F1 on the ESDD2 test set.","cross_cats":[],"primary_cat":"cs.SD","authors_text":"Haonan Cheng, Hengyan Huang, Jian Liu, Jiayi Zhou, Long Ye, Qin Zhang, Xiaoxuan Guo, Yuankun Xie","submitted_at":"2026-05-18T13:48:52Z","abstract_excerpt":"ADD in real-world scenarios has evolved from speech-only spoofing to more challenging component-level settings, where speech and environmental sounds may be independently manipulated. To tackle this, we propose EnvTriCascade, an Environment-Aware Tri-Stage Cascaded framework for the ESDD2 Challenge. First, a mix-consistency detector provides a binary prior to distinguish original recordings from manipulated mixtures, which calibrates the final decisions. Next, two complementary five-class detectors, leveraging SSLAM+XLS-R and EAT-large+XLS-R representations, extract robust multi-branch feature"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Trained exclusively on the official CompSpoofV2 dataset, our system achieves a Macro-F1 score of 0.8266 on the test set, significantly outperforming the official baseline and ranking second in the challenge.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The mix-consistency detector and the two complementary five-class detectors (SSLAM+XLS-R and EAT-large+XLS-R) provide independent and complementary information that, when fused via cross-branch attention, yields robust generalization to the unseen test mixing conditions.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"EnvTriCascade is a tri-stage cascaded framework using mix-consistency detection followed by dual SSL-based five-class classifiers with cross-branch attention and RawBoost augmentation, achieving 0.8266 Macro-F1 on the ESDD2 2026 challenge test set.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A tri-stage cascaded detector first checks mix consistency then fuses two five-class classifiers via attention to reach 0.8266 Macro-F1 on the ESDD2 test set.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"2d25e28f2c67cc4331c926df81944ac2254816e220633ef28f6722341d8a9c41"},"source":{"id":"2605.18409","kind":"arxiv","version":1},"verdict":{"id":"5212a764-8342-4533-86fb-307b26e35452","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T23:39:52.488274Z","strongest_claim":"Trained exclusively on the official CompSpoofV2 dataset, our system achieves a Macro-F1 score of 0.8266 on the test set, significantly outperforming the official baseline and ranking second in the challenge.","one_line_summary":"EnvTriCascade is a tri-stage cascaded framework using mix-consistency detection followed by dual SSL-based five-class classifiers with cross-branch attention and RawBoost augmentation, achieving 0.8266 Macro-F1 on the ESDD2 2026 challenge test set.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The mix-consistency detector and the two complementary five-class detectors (SSLAM+XLS-R and EAT-large+XLS-R) provide independent and complementary information that, when fused via cross-branch attention, yields robust generalization to the unseen test mixing conditions.","pith_extraction_headline":"A tri-stage cascaded detector first checks mix consistency then fuses two five-class classifiers via attention to reach 0.8266 Macro-F1 on the ESDD2 test set."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.18409/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-20T00:01:20.303360Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"cited_work_retraction","ran_at":"2026-05-19T23:51:58.376600Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T23:51:35.632984Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"citation_quote_validity","ran_at":"2026-05-19T23:49:59.955040Z","status":"completed","version":"0.1.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T23:33:29.260774Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"external_links","ran_at":"2026-05-19T23:31:38.763042Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T23:21:58.706375Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"98549cbc941e121db2ba3b1b8d3e74287241a38dc2b1a38dfbcc25046c2a5fc3"},"references":{"count":30,"sample":[{"doi":"","year":2026,"title":"MAGIC-TTS: Fine-Grained Controllable Speech Synthesis with Explicit Local Duration and Pause Control","work_id":"b1e631c7-b602-4df0-8c51-fb603c32f4a7","ref_index":1,"cited_arxiv_id":"2604.21164","is_internal_anchor":true},{"doi":"","year":2026,"title":"Text-To-Speech with Chain-of-Details: modeling temporal dynamics in speech generation","work_id":"4b59cf81-23ff-457b-83a1-5be2c8f3d221","ref_index":2,"cited_arxiv_id":"2604.19330","is_internal_anchor":true},{"doi":"","year":2026,"title":"X-VC: Zero-shot Streaming Voice Conversion in Codec Space","work_id":"99481ebe-cad0-4ce9-ab20-26bd94e3d83a","ref_index":3,"cited_arxiv_id":"2604.12456","is_internal_anchor":true},{"doi":"","year":2026,"title":"Emotion-aware prefix: Towards explicit emotion control in voice conversion models,","work_id":"43fb5df9-97cb-4fa4-b2e7-f9daec942154","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Speechfake: A large-scale multilingual speech deepfake dataset incorporating cutting-edge generation methods,","work_id":"0df1c1ff-3953-4a7a-b20f-efffa6876864","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":30,"snapshot_sha256":"0521bc1e3d026ccfb826a92eca88ca82b2a4d7b00df21375fd64c0a8c68c1e2d","internal_anchors":5},"formal_canon":{"evidence_count":2,"snapshot_sha256":"0d4cb13c20ece12be3a1ba094d5e468d8ad65a0011ae6c9ef8cffd90e12db9a3"},"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":"2605.18409","created_at":"2026-05-20T00:05:59.344495+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.18409v1","created_at":"2026-05-20T00:05:59.344495+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18409","created_at":"2026-05-20T00:05:59.344495+00:00"},{"alias_kind":"pith_short_12","alias_value":"CBX6EKEYOYF5","created_at":"2026-05-20T00:05:59.344495+00:00"},{"alias_kind":"pith_short_16","alias_value":"CBX6EKEYOYF5ZD3P","created_at":"2026-05-20T00:05:59.344495+00:00"},{"alias_kind":"pith_short_8","alias_value":"CBX6EKEY","created_at":"2026-05-20T00:05:59.344495+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CBX6EKEYOYF5ZD3PH5S7OTKPLC","json":"https://pith.science/pith/CBX6EKEYOYF5ZD3PH5S7OTKPLC.json","graph_json":"https://pith.science/api/pith-number/CBX6EKEYOYF5ZD3PH5S7OTKPLC/graph.json","events_json":"https://pith.science/api/pith-number/CBX6EKEYOYF5ZD3PH5S7OTKPLC/events.json","paper":"https://pith.science/paper/CBX6EKEY"},"agent_actions":{"view_html":"https://pith.science/pith/CBX6EKEYOYF5ZD3PH5S7OTKPLC","download_json":"https://pith.science/pith/CBX6EKEYOYF5ZD3PH5S7OTKPLC.json","view_paper":"https://pith.science/paper/CBX6EKEY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.18409&json=true","fetch_graph":"https://pith.science/api/pith-number/CBX6EKEYOYF5ZD3PH5S7OTKPLC/graph.json","fetch_events":"https://pith.science/api/pith-number/CBX6EKEYOYF5ZD3PH5S7OTKPLC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CBX6EKEYOYF5ZD3PH5S7OTKPLC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CBX6EKEYOYF5ZD3PH5S7OTKPLC/action/storage_attestation","attest_author":"https://pith.science/pith/CBX6EKEYOYF5ZD3PH5S7OTKPLC/action/author_attestation","sign_citation":"https://pith.science/pith/CBX6EKEYOYF5ZD3PH5S7OTKPLC/action/citation_signature","submit_replication":"https://pith.science/pith/CBX6EKEYOYF5ZD3PH5S7OTKPLC/action/replication_record"}},"created_at":"2026-05-20T00:05:59.344495+00:00","updated_at":"2026-05-20T00:05:59.344495+00:00"}