{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:JQDI5RQYTU64JSCDLL5SBRYBJ4","short_pith_number":"pith:JQDI5RQY","schema_version":"1.0","canonical_sha256":"4c068ec6189d3dc4c8435afb20c7014f3b5c8b369d1b48c18940e0c52bfcdbbb","source":{"kind":"arxiv","id":"2605.27190","version":1},"attestation_state":"computed","paper":{"title":"Learning When to Think While Listening in Large Audio-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.SD"],"primary_cat":"cs.CL","authors_text":"Cheng Zhu, Jiatao Gu, Suhao Yu, Weici Zhao, Yang Xiao, Zhiyuan Song","submitted_at":"2026-05-26T15:43:11Z","abstract_excerpt":"Recent advances in Large Audio-Language Models (LALMs) have made real-time, streaming spoken interaction increasingly practical. In this setting, reasoning quality and responsiveness are tightly coupled: delaying reasoning until the speech endpoint can improve answer quality but moves deliberation into user-visible response delay, while answering too early risks committing before decisive evidence arrives. We introduce a learnable wait-think-answer control formulation for LALMs. Motivated by the incremental nature of human conversation, the controller decides under partial audio evidence when "},"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":"2605.27190","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-26T15:43:11Z","cross_cats_sorted":["cs.AI","cs.LG","cs.SD"],"title_canon_sha256":"ab0a5e1ddeb9bc5aff30f785e92ae19bf670d10e4f42f8bd6bda5bebd5009b06","abstract_canon_sha256":"78344c9761936a88d74ac0d2740d0a3294153bb24b3c5742ea744974c019eb68"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-27T02:05:47.303176Z","signature_b64":"9b4uJjWYedebhEamYqsdxHH0FcWk0aHI39suMbDJMBdjQ23YPfUVFcyldURsX8XlwT/EPU/7IgW5G3NNOwFuDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4c068ec6189d3dc4c8435afb20c7014f3b5c8b369d1b48c18940e0c52bfcdbbb","last_reissued_at":"2026-05-27T02:05:47.302432Z","signature_status":"signed_v1","first_computed_at":"2026-05-27T02:05:47.302432Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning When to Think While Listening in Large Audio-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.SD"],"primary_cat":"cs.CL","authors_text":"Cheng Zhu, Jiatao Gu, Suhao Yu, Weici Zhao, Yang Xiao, Zhiyuan Song","submitted_at":"2026-05-26T15:43:11Z","abstract_excerpt":"Recent advances in Large Audio-Language Models (LALMs) have made real-time, streaming spoken interaction increasingly practical. In this setting, reasoning quality and responsiveness are tightly coupled: delaying reasoning until the speech endpoint can improve answer quality but moves deliberation into user-visible response delay, while answering too early risks committing before decisive evidence arrives. We introduce a learnable wait-think-answer control formulation for LALMs. Motivated by the incremental nature of human conversation, the controller decides under partial audio evidence when "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27190","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/2605.27190/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":"2605.27190","created_at":"2026-05-27T02:05:47.302566+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.27190v1","created_at":"2026-05-27T02:05:47.302566+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27190","created_at":"2026-05-27T02:05:47.302566+00:00"},{"alias_kind":"pith_short_12","alias_value":"JQDI5RQYTU64","created_at":"2026-05-27T02:05:47.302566+00:00"},{"alias_kind":"pith_short_16","alias_value":"JQDI5RQYTU64JSCD","created_at":"2026-05-27T02:05:47.302566+00:00"},{"alias_kind":"pith_short_8","alias_value":"JQDI5RQY","created_at":"2026-05-27T02:05:47.302566+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/JQDI5RQYTU64JSCDLL5SBRYBJ4","json":"https://pith.science/pith/JQDI5RQYTU64JSCDLL5SBRYBJ4.json","graph_json":"https://pith.science/api/pith-number/JQDI5RQYTU64JSCDLL5SBRYBJ4/graph.json","events_json":"https://pith.science/api/pith-number/JQDI5RQYTU64JSCDLL5SBRYBJ4/events.json","paper":"https://pith.science/paper/JQDI5RQY"},"agent_actions":{"view_html":"https://pith.science/pith/JQDI5RQYTU64JSCDLL5SBRYBJ4","download_json":"https://pith.science/pith/JQDI5RQYTU64JSCDLL5SBRYBJ4.json","view_paper":"https://pith.science/paper/JQDI5RQY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.27190&json=true","fetch_graph":"https://pith.science/api/pith-number/JQDI5RQYTU64JSCDLL5SBRYBJ4/graph.json","fetch_events":"https://pith.science/api/pith-number/JQDI5RQYTU64JSCDLL5SBRYBJ4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JQDI5RQYTU64JSCDLL5SBRYBJ4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JQDI5RQYTU64JSCDLL5SBRYBJ4/action/storage_attestation","attest_author":"https://pith.science/pith/JQDI5RQYTU64JSCDLL5SBRYBJ4/action/author_attestation","sign_citation":"https://pith.science/pith/JQDI5RQYTU64JSCDLL5SBRYBJ4/action/citation_signature","submit_replication":"https://pith.science/pith/JQDI5RQYTU64JSCDLL5SBRYBJ4/action/replication_record"}},"created_at":"2026-05-27T02:05:47.302566+00:00","updated_at":"2026-05-27T02:05:47.302566+00:00"}