Skip to content
星际流动

Online Distributionally Robust LLM Alignment via Regression to Relative Reward

发布
采集
学术前沿 6.3 分 — DRO-REBEL addressing overoptimization in offline RLHF
原文: cs.LG updates on arXiv.org

评分 6.3 · 来源:cs.LG updates on arXiv.org · 发布于 2026-04-20

评分依据:DRO-REBEL addressing overoptimization in offline RLHF