{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.21143", "kind": "paper", "url": "https://arxiv.org/abs/2607.21143"}], "authors": ["Minh Ngoc Ta", "My Anh Tran Nguyen", "Duong D. Nguyen", "Yuxia Wang", "Preslav Nakov"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.21143", "source_url": "https://arxiv.org/abs/2607.21143"}, "categories": ["General AI", "Language & Knowledge", "cs.CL"], "description": {"en": "RegretBench evaluates clarification policies in multi-turn conversational LLMs, using hidden-intent tasks and a regret-based objective to measure value loss relative to a reference policy."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.21143", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.21143", "title": "RegretBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Trường Đại học Khoa học và Công nghệ Hà Nội", "Hanoi University of Science and Technology", "Sofia University \"St. Kliment Ohridski\""], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2607.21143", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2607.21143", "source_url": "https://openalex.org/W7170693974"}, "key": "claire-radar:2607.21143", "modality": null, "name": "RegretBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "Medium", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_regretbench_a3a93900", "origin_source": "arxiv", "recognition_confidence": "0.85", "record_sha256": "f9cd8e3c198f65ea56d2fd3cc1a3ebf76ec6e2f8176c98199d32079b399426c3", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.21143"}, "publisher": null, "released": "2026-07-23", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.21143", "source_url": "https://arxiv.org/abs/2607.21143"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-21143", "source": "claire_radar", "source_benchmark_id": "2607.21143", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2607.21143", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "unavailable"}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "707046653d174b7998a50f86e7495e269011a32e152dc0eefb32c930442e8746", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "RegretBench evaluates clarification policies in multi-turn conversational LLMs, using hidden-intent tasks and a regret-based objective to measure value loss relative to a reference policy.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence"], "snippet": "We introduce RegretBench, a multi-turn benchmark that evaluates clarification as policy behavior rather than isolated question quality."}, "familyId": "bmf_7ae5aa690b6f", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_regretbench_a3a93900", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2607.21143", "paper": "https://arxiv.org/abs/2607.21143", "pdf": "https://arxiv.org/pdf/2607.21143", "project": null, "report": "https://arxiv.org/abs/2607.21143"}, "metrics": [], "motivation": "Ambiguous user requests make clarification a sequential decision problem for conversational LLM assistants: they must decide whether to ask, what to ask, when to stop, and when to answer.", "name": "RegretBench", "oneLine": "RegretBench evaluates clarification policies in multi-turn conversational LLMs, using hidden-intent tasks and a regret-based objective to measure value loss relative to a reference policy.", "paperTitle": "One More Turn, Less Regret: A Regret-Based Multi-Turn Benchmark for LLMs' Clarification Policies", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2607.21143", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {"90d": {"confidence": "Low", "coverage": 0.0, "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": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v13", "rank": null, "score": null}}, "readiness": "Paper only", "recognitionConfidence": 0.85, "relation": "introduces", "releasedAt": "2026-07-23", "source": {"authors": ["Minh Ngoc Ta", "My Anh Tran Nguyen", "Duong D. Nguyen", "Yuxia Wang", "Preslav Nakov"], "categories": ["cs.CL", "cs.AI"], "id": "2607.21143", "title": "One More Turn, Less Regret: A Regret-Based Multi-Turn Benchmark for LLMs' Clarification Policies", "type": "arxiv", "url": "https://arxiv.org/abs/2607.21143"}, "sourceUpdatedAt": "2026-07-24T08:22:37Z", "topics": ["cs.CL"], "venueAttempts": [], "whyItMatters": "It addresses the evaluation gap in conversational AI by jointly measuring intent resolution, interaction cost, and stopping decisions, offering a more comprehensive assessment of clarification behavior."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
