Benchmark RadarAI BENCHMARK PROFILE
CanLegalRAGBench
CanLegalRAGBench evaluates retrieval-augmented generation for Canadian case law with realistic queries and expert-annotated answers. It tests retrieval and answer generation grounded in legal documents.
- Released
- 2026-05-28
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Fills a gap in legal RAG evaluation for Canadian law, providing realistic scenarios and revealing limitations in automatic metrics and hallucination in generated answers.
Motivation
RAG-based legal assistants have been growing in popularity, but LLM hallucinations remain a key issue and potentially undermines justice.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.