Benchmark RadarAI BENCHMARK PROFILE
UA-Legal-Bench
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.
- Released
- 2026-05-27
- Readiness
- Inspectable
- Primary field
- General AI
Why it matters
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.
Motivation
Legal NLP benchmarks are overwhelmingly English-centric, leaving failure modes in morphologically rich, non-Latin-script languages undetected.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.