judge-jev
carlaiau/judge-jev
Reproducible experiments evaluating Jev as an automated judge across benchmarks and tasks.
- Stars
- 0
- Forks
- 0
Why it is listed
Research evaluates Jev Choice/Score/Noul as passage-relevance judges against human labels and published LLMJudge submissions; the public-label comparison is retrospective, not a blind challenge.