{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.25218", "kind": "paper", "url": "https://arxiv.org/abs/2608.25218"}, {"kind": "website", "url": "https://turnbench.sesame.com"}], "authors": ["Freeman Jiang", "Ramon Sanabria", "Soham Deshmukh", "Bandhav Veluri", "Simon Michael Vuch Williams", "Elliott K. Suen", "Garreth Lee", "Kevin Yoonho Choi", "Takuya Umeki", "Riku Kubo", "Sathvik Udupa", "Chien-yu Huang", "Shih-Yun Shan Kuan", "Zhuoyan Tao", "Satyapriya Krishna", "Sefik Emre Eskimez", "Yu Tsao", "Hung-yi Lee", "Shinji Watanabe"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.25218", "source_url": "https://arxiv.org/abs/2608.25218"}, "categories": ["General AI", "Vision & 3D", "eess.AS"], "description": {"en": "Evaluates end-of-turn and interruption detection in dyadic human conversation across six interaction styles using a 30-hour hand-labeled corpus and standardized metrics."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.25218", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.25218", "title": "TurnBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Brno University of Technology", "Carnegie Mellon University", "National Taiwan University", "Academia Sinica"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.25218", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.25218", "source_url": "https://openalex.org/W7204453739"}, "key": "claire-radar:2608.25218", "modality": null, "name": "TurnBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "true", "origin_record_id": "bm_turnbench_084201bf", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "b79481cbad457017c27eee2da0f8f5776bb82e10eda98556d592ee9b44184548", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-08-27T04:14:52.836173Z", "source_url": "https://arxiv.org/abs/2608.25218"}, "publisher": null, "released": "2026-08-25", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.25218", "source_url": "https://arxiv.org/abs/2608.25218"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-25218", "source": "claire_radar", "source_benchmark_id": "2608.25218", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Vision & 3D", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/SesameAILabs/turnbench", "githubScope": "benchmark_repo", "githubStars": 49, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 1, "hfPaperUrl": "https://huggingface.co/papers/2608.25218", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-25", "state": "stale"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "Spoken dialogue benchmarking is a growing area with a clear leaderboard and multi-domain scope, likely to attract moderate early interest from speech and NLP communities.", "score": 65}, "capabilities": [], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-27T04:14:52.836173Z", "inputHash": "2b00ba9b372106bee99f6a1046797422aebe846738d80f44af424a86e9e81348", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "TurnBench: A Multi-Domain Benchmark for Turn-Taking Dynamics in Spoken Dialogue", "canonicalNameSource": "paper_title", "decisionReason": "Includes a released corpus, training set, public leaderboard, and standardized evaluation protocol; meets repeatable benchmarking and public comparison criteria.", "model": "deepseek-v4-pro", "reviewedAt": "2026-08-27T04:14:52.836173Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates end-of-turn and interruption detection in dyadic human conversation across six interaction styles using a 30-hour hand-labeled corpus and standardized metrics.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "score_submission", "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "To address this, we present TurnBench, a multi-domain benchmark that pairs a 30-hour, hand-labeled corpus of dyadic human conversation with a standardized evaluation protocol for end-of-turn and interruption detection."}, "familyId": "bmf_2b94493a4968", "firstSeenAt": "2026-08-27", "heat": null, "id": "bm_turnbench_084201bf", "indexedAt": "2026-08-27T04:12:10.575570Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2608.25218", "pdf": "https://arxiv.org/pdf/2608.25218", "project": "https://turnbench.sesame.com", "report": "https://arxiv.org/abs/2608.25218"}, "metrics": [], "motivation": "Speakers in natural conversation take turns speaking and listening, deciding in real time when to take, hold, or yield the floor.", "name": "TurnBench", "oneLine": "Evaluates end-of-turn and interruption detection in dyadic human conversation across six interaction styles using a 30-hour hand-labeled corpus and standardized metrics.", "paperTitle": "TurnBench: A Multi-Domain Benchmark for Turn-Taking Dynamics in Spoken Dialogue", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.25218", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-27T04:12:10.575570Z"}, "publications": [], "publishers": [{"name": "Sesame", "organizationType": "company-research-lab", "role": "benchmark-publisher", "sourceUrl": "https://turnbench.sesame.com"}], "ranking": {"90d": {"confidence": "Medium", "coverage": 0.7, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 90}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": 0.5221847482253589, "value": 49}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.12247537038455211, "value": 1}}, "confidence": "Medium", "coverage": 0.7, "rank": 101, "score": 46}, "method": "attention-ranking-v13", "rank": 101, "score": 46}}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-25", "source": {"authors": ["Freeman Jiang", "Ramon Sanabria", "Soham Deshmukh", "Bandhav Veluri", "Simon Michael Vuch Williams", "Elliott K. Suen", "Garreth Lee", "Kevin Yoonho Choi", "Takuya Umeki", "Riku Kubo", "Sathvik Udupa", "Chien-yu Huang", "Shih-Yun Shan Kuan", "Zhuoyan Tao", "Satyapriya Krishna", "Sefik Emre Eskimez", "Yu Tsao", "Hung-yi Lee", "Shinji Watanabe"], "categories": ["eess.AS", "cs.CL"], "id": "2608.25218", "title": "TurnBench: A Multi-Domain Benchmark for Turn-Taking Dynamics in Spoken Dialogue", "type": "arxiv", "url": "https://arxiv.org/abs/2608.25218"}, "sourceUpdatedAt": "2026-08-25T23:14:36Z", "topics": ["eess.AS"], "venueAttempts": [], "whyItMatters": "Provides the first multi-domain, triple-annotated corpus for turn-taking, enabling consistent comparison across conversation types and revealing type-dependent interruption errors."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
