模型能否免费捕捉自身幻觉？无标签置信信号在弃权任务中媲美标签数据集
原标题：Can a Model Catch Its Own Hallucinations for Free?: Label-Free Doubt Signals Hold Their Own Against a Labelled Dataset for Abstention
AI 摘要
该研究提出一种无标签的弃权微调方法，利用模型自身的冻结置信度（token概率）来教导模型在低置信度时回答“我不确定”，无需正确性标签。在六个开源模型（1B-8B）的短事实问答测试中，该方法与标签监督的弃权调优性能相当，无统计显著差异。研究还发现，该信号无法识别“自信但错误”的事实，是其盲点。
正文节选
Can a Model Catch Its Own Hallucinations for Free?: Label-Free Doubt Signals Hold Their Own Against a Labelled Dataset for Abstention Abstract Large language models state false facts as fluently as true ones, yet a model often “knows” internally when it is on shaky ground: the probability it assigns to its own answer tends to dip on the facts it gets wrong. The usual way to act on this, teaching a model to abstain rather than guess, requires a labelled dataset of right and wrong answers. We ask