Benedikt Stroebl
@benediktstroebl
Agents & Evals. PhD @Princeton, YC
가입 November 2020
1.7K 팔로잉 중 1K 팬
Jev-as-a-Judge in Harbor Rewardkit is pretty amazing. It agrees with Fable and GPT and is very fast. When we test it on a real eval that uses an LLM judge, such as the @harvey LAB benchmark, Jev achieves perfect agreement but is 50x faster.
더 보기
