GIT_FEED

composo-ai/llm-judge-criteria-ensembling

Criteria injection + ensembling for LLM-as-judge accuracy on RewardBench 2 — 83.6% with one-sentence criteria and k=8 ensembling.

View on GitHub
0Active

On the radar — signal detected

Stars
5
Forks
0
Contributors
1
Language
Python

Score updated Apr 7, 2026

// SUBSCRIBE

The repos that moved this week, why they matter, and what to watch next. One email. No noise.