大语言模型无数据自改进新法

AK· @_akhaliq · X·2026-10-26 22:46· 3 小时前AI 评分23

AI 导读

On-Policy Self-Training 大语言模型的无数据自我改进论文：https://huggingface.co/papers/2610.32770

正文

On-Policy Self-Training

Data-Free Self-Improvement in Large Language Models

paper: https://huggingface.co/papers/2610.32770

来源：AK · x.com