STAR-1: Safer Alignment of Reasoning LLMs with 1K Data
Paper: https://
arxiv.org/pdf/2504.01903
Project Page: https://
ucsc-vlaa.github.io/STAR-1

→ View original post on X — @jiqizhixin