Learning to summarize with human feedback

【RSS 原文 · 未经 AI 筛选或翻译】We’ve applied reinforcement learning from human feedback to train language models that are better at summarization.