The new Jev model looks really useful. Here is a paper evaluating Jev as a judge arxiv.org/abs/2609.26550 #jev #llms
arxiv.org
JEV-as-a-Judge: Accept When Confident, Escalate When Unsure
LLM-as-a-judge enables evaluation across diverse tasks, but inference cost and confidence reliability become critical at scale. We study whether a decision-only judge can provide an economical first p...