{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2605.29523", "kind": "paper", "url": "https://arxiv.org/abs/2605.29523"}], "authors": ["Eunbyeol Cho", "Yunseung Lee", "Mirae Kim", "Jeewon Yang", "Youngjun Kwak", "Edward Choi"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2605.29523", "source_url": "https://arxiv.org/abs/2605.29523"}, "categories": ["Finance", "Vision & 3D", "cs.LG"], "description": {"en": "Evaluates hallucination detection in multi-turn Korean financial RAG dialogues, with a taxonomy based on context answerability. Includes training and test splits for fine-tuning and benchmarking detectors."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2605.29523", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2605.29523", "title": "K-FinHallu"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2605.29523", "modality": null, "name": "K-FinHallu", "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": "unknown", "origin_record_id": "bm_k-finhallu_490348d5", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "692c27cc472fdbf1386c6683e2b1a3bd7f258003aaefc5e5d698aacdda95e419", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2605.29523"}, "publisher": null, "released": "2026-05-28", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2605.29523", "source_url": "https://arxiv.org/abs/2605.29523"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2605-29523", "source": "claire_radar", "source_benchmark_id": "2605.29523", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Finance"], "area": "Vision & 3D", "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/2605.29523", "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": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-21T04:30:40.934319Z", "inputHash": "1f9402de42a9fc715cd5223d519e4a885bbd401e306e81d9e08423a66ff64b36", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates hallucination detection in multi-turn Korean financial RAG dialogues, with a taxonomy based on context answerability. Includes training and test splits for fine-tuning and benchmarking detectors.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We introduce K-FinHallu, the first benchmark for hallucination detection in multi-turn Korean financial RAG."}, "familyId": "bmf_44667dfd90e2", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_k-finhallu_490348d5", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Financial Services"], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2605.29523", "paper": "https://arxiv.org/abs/2605.29523", "pdf": "https://arxiv.org/pdf/2605.29523", "project": null, "report": "https://arxiv.org/abs/2605.29523"}, "metrics": [], "motivation": "Large Language Models (LLMs) have advanced financial automation through Retrieval-Augmented Generation (RAG), yet hallucinations remain a critical barrier to deployment in high-stakes environments.", "name": "K-FinHallu", "oneLine": "Evaluates hallucination detection in multi-turn Korean financial RAG dialogues, with a taxonomy based on context answerability. Includes training and test splits for fine-tuning and benchmarking detectors.", "paperTitle": "K-FinHallu: A Hallucination Detection Benchmark for Multi-Turn RAG in Korean Finance", "primaryDomain": "Finance", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2605.29523", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "K-FinHallu Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://arxiv.org/abs/2605.29523"}], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-05-28", "source": {"authors": ["Eunbyeol Cho", "Yunseung Lee", "Mirae Kim", "Jeewon Yang", "Youngjun Kwak", "Edward Choi"], "categories": ["cs.LG"], "id": "2605.29523", "title": "K-FinHallu: A Hallucination Detection Benchmark for Multi-Turn RAG in Korean Finance", "type": "arxiv", "url": "https://arxiv.org/abs/2605.29523"}, "sourceUpdatedAt": "2026-05-28T07:40:19Z", "topics": ["cs.LG"], "venueAttempts": [], "whyItMatters": "Targets a high-stakes multilingual domain where existing benchmarks lack coverage. Provides a resource to improve hallucination detection and refusal behavior in financial applications."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
