{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3DYL27DS2QNHIOYMYZ4LMKSTHK","short_pith_number":"pith:3DYL27DS","schema_version":"1.0","canonical_sha256":"d8f0bd7c72d41a743b0cc678b62a533a8cae261f261b6d89ded324a119f9a9f0","source":{"kind":"arxiv","id":"2601.09413","version":2},"attestation_state":"computed","paper":{"title":"Speech-Hands: A Self-Reflection Voice Agentic Approach to Speech Recognition and Audio Reasoning with Omni Perception","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.MA","eess.AS"],"primary_cat":"cs.SD","authors_text":"Ankita Pasad, Arushi Goel, Boris Ginsburg, Chao-Han Huck Yang, Chenhui Chu, Hanrong Ye, Jinchuan Tian, Kunal Dhawan, Rafael Valle, Ryo Hachiuma, Shinji Watanabe, Shizhe Diao, Sreyan Ghosh, Szu-Wei Fu, Yu-Chiang Frank Wang, Yusuke Hirota, Zhehuai Chen, Zhen Wan","submitted_at":"2026-01-14T12:06:50Z","abstract_excerpt":"We introduce a voice-agentic framework that learns one critical omni-understanding skill: knowing when to trust itself versus when to consult external audio perception. Our work is motivated by a crucial yet counterintuitive finding: naively fine-tuning an omni-model on both speech recognition and external sound understanding tasks often degrades performance, as the model can be easily misled by noisy hypotheses. To address this, our framework, Speech-Hands, recasts the problem as an explicit self-reflection decision. This learnable reflection primitive proves effective in preventing the model"},"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":"2601.09413","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SD","submitted_at":"2026-01-14T12:06:50Z","cross_cats_sorted":["cs.AI","cs.CL","cs.MA","eess.AS"],"title_canon_sha256":"b86d87f75aebdd879c383f6c60cda290b868f254574258d9f69f6dfe6ced8966","abstract_canon_sha256":"cf3d0e52ed88500de1bd0c7767fc6083560d005120c84a4741f6963206578047"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:05:40.490934Z","signature_b64":"B0uhQ5Xh0TiJzz77VOuMV41qe8l1pLAvlb7gTdoN6se6Oug2LJi5FAZjP3C4CsQ8XJmMYkFYjfNNEibrpytNAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d8f0bd7c72d41a743b0cc678b62a533a8cae261f261b6d89ded324a119f9a9f0","last_reissued_at":"2026-05-20T00:05:40.490250Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:05:40.490250Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Speech-Hands: A Self-Reflection Voice Agentic Approach to Speech Recognition and Audio Reasoning with Omni Perception","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.MA","eess.AS"],"primary_cat":"cs.SD","authors_text":"Ankita Pasad, Arushi Goel, Boris Ginsburg, Chao-Han Huck Yang, Chenhui Chu, Hanrong Ye, Jinchuan Tian, Kunal Dhawan, Rafael Valle, Ryo Hachiuma, Shinji Watanabe, Shizhe Diao, Sreyan Ghosh, Szu-Wei Fu, Yu-Chiang Frank Wang, Yusuke Hirota, Zhehuai Chen, Zhen Wan","submitted_at":"2026-01-14T12:06:50Z","abstract_excerpt":"We introduce a voice-agentic framework that learns one critical omni-understanding skill: knowing when to trust itself versus when to consult external audio perception. Our work is motivated by a crucial yet counterintuitive finding: naively fine-tuning an omni-model on both speech recognition and external sound understanding tasks often degrades performance, as the model can be easily misled by noisy hypotheses. To address this, our framework, Speech-Hands, recasts the problem as an explicit self-reflection decision. This learnable reflection primitive proves effective in preventing the model"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.09413","kind":"arxiv","version":2},"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/2601.09413/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":"2601.09413","created_at":"2026-05-20T00:05:40.490353+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.09413v2","created_at":"2026-05-20T00:05:40.490353+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.09413","created_at":"2026-05-20T00:05:40.490353+00:00"},{"alias_kind":"pith_short_12","alias_value":"3DYL27DS2QNH","created_at":"2026-05-20T00:05:40.490353+00:00"},{"alias_kind":"pith_short_16","alias_value":"3DYL27DS2QNHIOYM","created_at":"2026-05-20T00:05:40.490353+00:00"},{"alias_kind":"pith_short_8","alias_value":"3DYL27DS","created_at":"2026-05-20T00:05:40.490353+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2605.15104","citing_title":"From Text to Voice: A Reproducible and Verifiable Framework for Evaluating Tool Calling LLM Agents","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21008","citing_title":"A Survey of Audio Reasoning in Multimodal Foundation Models","ref_index":105,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3DYL27DS2QNHIOYMYZ4LMKSTHK","json":"https://pith.science/pith/3DYL27DS2QNHIOYMYZ4LMKSTHK.json","graph_json":"https://pith.science/api/pith-number/3DYL27DS2QNHIOYMYZ4LMKSTHK/graph.json","events_json":"https://pith.science/api/pith-number/3DYL27DS2QNHIOYMYZ4LMKSTHK/events.json","paper":"https://pith.science/paper/3DYL27DS"},"agent_actions":{"view_html":"https://pith.science/pith/3DYL27DS2QNHIOYMYZ4LMKSTHK","download_json":"https://pith.science/pith/3DYL27DS2QNHIOYMYZ4LMKSTHK.json","view_paper":"https://pith.science/paper/3DYL27DS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.09413&json=true","fetch_graph":"https://pith.science/api/pith-number/3DYL27DS2QNHIOYMYZ4LMKSTHK/graph.json","fetch_events":"https://pith.science/api/pith-number/3DYL27DS2QNHIOYMYZ4LMKSTHK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3DYL27DS2QNHIOYMYZ4LMKSTHK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3DYL27DS2QNHIOYMYZ4LMKSTHK/action/storage_attestation","attest_author":"https://pith.science/pith/3DYL27DS2QNHIOYMYZ4LMKSTHK/action/author_attestation","sign_citation":"https://pith.science/pith/3DYL27DS2QNHIOYMYZ4LMKSTHK/action/citation_signature","submit_replication":"https://pith.science/pith/3DYL27DS2QNHIOYMYZ4LMKSTHK/action/replication_record"}},"created_at":"2026-05-20T00:05:40.490353+00:00","updated_at":"2026-05-20T00:05:40.490353+00:00"}