{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2605.29170", "kind": "paper", "url": "https://arxiv.org/abs/2605.29170"}, {"id": "hf:overthelex/ua-legal-bench", "kind": "dataset", "url": "https://huggingface.co/datasets/overthelex/ua-legal-bench"}], "authors": ["Volodymyr Ovcharov"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2605.29170", "source_url": "https://arxiv.org/abs/2605.29170"}, "categories": ["General AI", "Language & Knowledge", "Reasoning"], "description": {"en": "UA-Legal-Bench is a five-task benchmark for evaluating LLMs on Ukrainian legal reasoning using court decisions, covering classification, outcome prediction, and norm extraction."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2605.29170", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2605.29170", "title": "UA-Legal-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2605.29170", "modality": null, "name": "UA-Legal-Bench", "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_ua-legal-bench_a6ec4166", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "941062e64abc917c3288d8cab9d8082ee63df68281219ad1b5761270e569b570", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2605.29170"}, "publisher": null, "released": "2026-05-27", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2605.29170", "source_url": "https://arxiv.org/abs/2605.29170"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2605-29170", "source": "claire_radar", "source_benchmark_id": "2605.29170", "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": "overthelex/ua-legal-bench", "hfDatasetDownloads": 320, "hfDatasetLikes": 0, "hfPaperUpvotes": 0, "hfPaperUrl": "https://huggingface.co/papers/2605.29170", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-30", "state": "stale"}, "hfDatasetLikes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-30", "state": "stale"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-30", "state": "stale"}}, "status": "stale"}, "capabilities": ["Reasoning"], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "e978ff6afdfdcdf1dc47a20a0ed30c25f1ac1b7881c123b03f722f73e8691bd4", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "UA-Legal-Bench is a five-task benchmark for evaluating LLMs on Ukrainian legal reasoning using court decisions, covering classification, outcome prediction, and norm extraction.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "We introduce UA-Legal-Bench, a five-task benchmark for evaluating large language models on Ukrainian legal reasoning, built from the Unified State Register of Court Decisions (EDRSR) -- one of the world's largest open judicial corpora (99.5 million decisions)."}, "familyId": "bmf_2308fd0ed61a", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_ua-legal-bench_a6ec4166", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": "https://huggingface.co/datasets/overthelex/ua-legal-bench", "hfPaper": "https://huggingface.co/papers/2605.29170", "paper": "https://arxiv.org/abs/2605.29170", "pdf": "https://arxiv.org/pdf/2605.29170", "project": null, "report": "https://arxiv.org/abs/2605.29170"}, "metrics": [], "motivation": "Legal NLP benchmarks are overwhelmingly English-centric, leaving failure modes in morphologically rich, non-Latin-script languages undetected.", "name": "UA-Legal-Bench", "oneLine": "UA-Legal-Bench is a five-task benchmark for evaluating LLMs on Ukrainian legal reasoning using court decisions, covering classification, outcome prediction, and norm extraction.", "paperTitle": "UA-Legal-Bench: A Benchmark for Evaluating Large Language Models on Ukrainian Legal Reasoning", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2605.29170", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "overthelex", "organizationType": "community", "role": "benchmark-publisher", "sourceUrl": "https://huggingface.co/datasets/overthelex/ua-legal-bench"}], "ranking": {}, "readiness": "Inspectable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-05-27", "source": {"authors": ["Volodymyr Ovcharov"], "categories": ["cs.CL", "cs.AI"], "id": "2605.29170", "title": "UA-Legal-Bench: A Benchmark for Evaluating Large Language Models on Ukrainian Legal Reasoning", "type": "arxiv", "url": "https://arxiv.org/abs/2605.29170"}, "sourceUpdatedAt": "2026-05-27T23:12:20Z", "topics": ["Reasoning"], "venueAttempts": [], "whyItMatters": "Addresses the dearth of legal NLP benchmarks for non-English, morphologically rich languages, and provides insights into few-shot effects and model scaling in legal tasks."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
