{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FBD5T2ACS423767DAPX2HE62VW","short_pith_number":"pith:FBD5T2AC","schema_version":"1.0","canonical_sha256":"2847d9e8029735bffbe303efa393daad94b5a48cbdb4fe3e85a0eb4c449943c3","source":{"kind":"arxiv","id":"2503.04721","version":3},"attestation_state":"computed","paper":{"title":"Full-Duplex-Bench: A Benchmark to Evaluate Full-duplex Spoken Dialogue Models on Turn-taking Capabilities","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Alexander H. Liu, Gopala Anumanchipalli, Guan-Ting Lin, Hung-yi Lee, Jiachen Lian, Qirui Wang, Tingle Li","submitted_at":"2025-03-06T18:59:16Z","abstract_excerpt":"Spoken dialogue modeling poses challenges beyond text-based language modeling, requiring real-time interaction, turn-taking, and backchanneling. While most Spoken Dialogue Models (SDMs) operate in half-duplex mode-processing one turn at a time - emerging full-duplex SDMs can listen and speak simultaneously, enabling more natural conversations. However, current evaluations remain limited, focusing mainly on turn-based metrics or coarse corpus-level analyses. To address this, we introduce Full-Duplex-Bench, a benchmark that systematically evaluates key interactive behaviors: pause handling, back"},"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":"2503.04721","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-06T18:59:16Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"4be8eb270ec5d1730f578c9da62a4f22a051bdcc354a3bd6e302c8b868ad47dd","abstract_canon_sha256":"191ad2b7989351e9cc506c3b422bad1165d4cbdbb23ac6b34a7d395e707d2163"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:54:44.962265Z","signature_b64":"nAF5OKCWcClyw1WHTHvPh00VLU1ravMCasod6Qbn7Q/zk51A2QqSCQcVIi/k5y59it1mmKc50/in7H0AQBHiBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2847d9e8029735bffbe303efa393daad94b5a48cbdb4fe3e85a0eb4c449943c3","last_reissued_at":"2026-07-05T11:54:44.961820Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:54:44.961820Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Full-Duplex-Bench: A Benchmark to Evaluate Full-duplex Spoken Dialogue Models on Turn-taking Capabilities","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Alexander H. Liu, Gopala Anumanchipalli, Guan-Ting Lin, Hung-yi Lee, Jiachen Lian, Qirui Wang, Tingle Li","submitted_at":"2025-03-06T18:59:16Z","abstract_excerpt":"Spoken dialogue modeling poses challenges beyond text-based language modeling, requiring real-time interaction, turn-taking, and backchanneling. While most Spoken Dialogue Models (SDMs) operate in half-duplex mode-processing one turn at a time - emerging full-duplex SDMs can listen and speak simultaneously, enabling more natural conversations. However, current evaluations remain limited, focusing mainly on turn-based metrics or coarse corpus-level analyses. To address this, we introduce Full-Duplex-Bench, a benchmark that systematically evaluates key interactive behaviors: pause handling, back"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.04721","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/2503.04721/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":"2503.04721","created_at":"2026-07-05T11:54:44.961878+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.04721v3","created_at":"2026-07-05T11:54:44.961878+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.04721","created_at":"2026-07-05T11:54:44.961878+00:00"},{"alias_kind":"pith_short_12","alias_value":"FBD5T2ACS423","created_at":"2026-07-05T11:54:44.961878+00:00"},{"alias_kind":"pith_short_16","alias_value":"FBD5T2ACS423767D","created_at":"2026-07-05T11:54:44.961878+00:00"},{"alias_kind":"pith_short_8","alias_value":"FBD5T2AC","created_at":"2026-07-05T11:54:44.961878+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":31,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2607.07985","citing_title":"A Reliability Assessment of LALM Audio Judges for Full-Duplex Voice Agents","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2607.06540","citing_title":"Hierarchical Acoustic-Semantic Modeling: Modality Separation and Semantic Coherence for Full-Duplex SLMs","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2606.19453","citing_title":"A Survey of Full-Duplex Spoken Dialogue Systems: Architectural Hierarchy, Interaction Ontology, and Decision State Machine","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17360","citing_title":"Omni-DuplexEval: Evaluating Real-time Duplex Omni-modal Interaction","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01345","citing_title":"TurnNat: Automatic Evaluation of Turn-Taking Naturalness in Dyadic Spoken Dialogue","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17542","citing_title":"Evaluating Large Language Models Abilities for Addressee, Turn-change, and Next Speaker Prediction in Meetings","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13450","citing_title":"Endpoint Anticipation for Low-Latency Spoken Dialogue","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11386","citing_title":"Overcoming State Inertia in Full-Duplex Spoken Language Models via Activation Steering","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01804","citing_title":"SpeechEditBench: A Bilingual Multi-Attribute Benchmark for Instruction-Guided Speech Editing","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13841","citing_title":"EVA-Bench: A New End-to-end Framework for Evaluating Voice Agents","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11484","citing_title":"Engagement Process: Rethinking the Temporal Interface of Action and Observation","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20755","citing_title":"DuplexSLA: A Full-Duplex Spoken Language Model with Synchronized Speech, Language, and Action","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30145","citing_title":"FacePlex: Full-Duplex Joint Speech-Facial Motion Generation for Conversational Avatars","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25343","citing_title":"Toward Native Multimodal Modeling: A Roadmap","ref_index":281,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30256","citing_title":"VideoFDB: Evaluating Full-Duplex Vision-Speech Capabilities in Conversational Agents","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11167","citing_title":"Multi-Faceted Interactivity Alignment in Full-Duplex Speech Models","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2505.15957","citing_title":"Towards Holistic Evaluation of Large Audio-Language Models: A Comprehensive Survey","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2603.17837","citing_title":"The Silent Thought: Modeling Internal Cognition in Full-Duplex Spoken Dialogue Models via Latent Reasoning","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20356","citing_title":"Synchronization and Turn-Taking in Full-Duplex Speech Dialogue Models","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20266","citing_title":"A Survey of Large Audio Language Models: Generalization, Trustworthiness, and Outlook","ref_index":96,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20755","citing_title":"DuplexSLA: A Full-Duplex Spoken Language Model with Synchronized Speech, Language, and Action","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17360","citing_title":"Omni-DuplexEval: Evaluating Real-time Duplex Omni-modal Interaction","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2509.26388","citing_title":"Game-Time: Evaluating Temporal Dynamics in Spoken Language Models","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2603.09643","citing_title":"MM-tau-p$^2$: Persona-Adaptive Prompting for Robust Multi-Modal Agent Evaluation in Dual-Control Settings","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2603.17837","citing_title":"The Silent Thought: Modeling Internal Cognition in Full-Duplex Spoken Dialogue Models via Latent Reasoning","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FBD5T2ACS423767DAPX2HE62VW","json":"https://pith.science/pith/FBD5T2ACS423767DAPX2HE62VW.json","graph_json":"https://pith.science/api/pith-number/FBD5T2ACS423767DAPX2HE62VW/graph.json","events_json":"https://pith.science/api/pith-number/FBD5T2ACS423767DAPX2HE62VW/events.json","paper":"https://pith.science/paper/FBD5T2AC"},"agent_actions":{"view_html":"https://pith.science/pith/FBD5T2ACS423767DAPX2HE62VW","download_json":"https://pith.science/pith/FBD5T2ACS423767DAPX2HE62VW.json","view_paper":"https://pith.science/paper/FBD5T2AC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.04721&json=true","fetch_graph":"https://pith.science/api/pith-number/FBD5T2ACS423767DAPX2HE62VW/graph.json","fetch_events":"https://pith.science/api/pith-number/FBD5T2ACS423767DAPX2HE62VW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FBD5T2ACS423767DAPX2HE62VW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FBD5T2ACS423767DAPX2HE62VW/action/storage_attestation","attest_author":"https://pith.science/pith/FBD5T2ACS423767DAPX2HE62VW/action/author_attestation","sign_citation":"https://pith.science/pith/FBD5T2ACS423767DAPX2HE62VW/action/citation_signature","submit_replication":"https://pith.science/pith/FBD5T2ACS423767DAPX2HE62VW/action/replication_record"}},"created_at":"2026-07-05T11:54:44.961878+00:00","updated_at":"2026-07-05T11:54:44.961878+00:00"}