aifollow.news
返回 Rohan Paul
Rohan Paul· @rohanpaul_ai · X· · 原发布时间 AI 评分60

微软研究提出自我争辩训练,减少模型受错误质疑后改掉正确答案

AI 导读

微软一篇新论文发现,模型面对错误反驳时常会放弃原本正确的答案,单靠提高准确率的训练无法解决。研究提出让模型先回答、再自行提出反对理由并裁定;在错误质疑测试中,Qwen3-8B 和 Gemma-2 9B 保持正确答案的比例从 53.3% 和 25.1% 升至约 89%。

原帖引用 · 原文

该语言的正文暂不可用,当前显示已有版本。

@rohanpaul_ai

New Microsoft paper finds that LLMs give up correct answers when challenged with wrong arguments, and accuracy training does not fix it.

When a user or another agent disputes an answer, models often abandon correct answers even when the challenge is wrong. More accuracy training, more sampling, or self-reflection does not fix it.

Self-Disagreement trains one model to state an answer, write the strongest case against it, then rule on that case and answer again. A standard answer checker rewards the final answer and the ruling.

When challengers were wrong, base Qwen3-8B kept its correct answer 53.3% of the time and Gemma-2 9B 25.1%. After training, both kept it about 89% of the time, while accuracy-only training added no such skill.

If your model faces pushback from users or other agents, test how often it drops correct answers, since accuracy scores will not show it.

来源:Rohan Paul · x.com

论文
发现内容有误?提交纠错