The paper proposes a method called JudgeLM to evaluate large language models (LLMs) in open-ended scenarios. They fine-tune LLMs as scalable judges and introduce techniques to address biases. JudgeLM achieves state-of-the-art performance and high agreement with human judges.

https://arxiv.org/abs//2310.17631

YouTube: https://www.youtube.com/@ArxivPapers

TikTok: https://www.tiktok.com/@arxiv_papers

Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016

Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers