用模型自身置信度做拒绝回答，效果不输有标签数据。
Can a Model Catch Its Own Hallucinations for Free?: Label-Free Doubt Signals Hold Their Own Against a Labelled Dataset for Abstention
- 用模型自身置信度作为拒绝信号，无需标注数据。
- 在六种模型上，准确率与有标签训练相当。
- 适合想低成本提升模型可靠性的人使用。
大语言模型常以同样流畅的语调陈述错误事实，但其内部往往能感知自身不确定性：对错误事实的预测概率通常较低。传统上利用此特性让模型拒绝回答，需依赖正确/错误答案的标注数据。本文探究模型自身的置信度（免费且无需标签）能否替代标注数据实现该目标。我们采用LoRA微调，在六种开放权重模型（1B-8B，两家族）上，仅基于置信度高低决定回答或说“不确定”，不依赖任何正确性标签。通过独立裁判模型评估事实类问答的正确性，结果表明：在相同覆盖度下，该无标签方法与有监督的拒答训练无统计差异。对比仅强化难例的控制组未见提升，说明收益源于校准而非死记硬背。唯一盲区是自信出错的事实，无法被识别。因此，模型自身的怀疑信号可近乎免费替代标注数据用于教其何时拒答。代码与资源可应要求提供。
原文摘要 · Abstract (English)
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 whether the model's own confidence, which is free and needs no labels, can do that job instead. We fine-tune each model (with LoRA) to answer when its frozen confidence is high and to say "I'm not sure" when it is low, using the signal alone and no correctness labels. Across six open-weights models (1B-8B, two families) on short-form factual question answering, with correctness adjudicated by an independent judge model, this label-free recipe holds its own against label-supervised abstention-tuning: at matched coverage we find no statistically detectable difference between the two. A control that drills hard examples instead of abstaining does not help, indicating the gain comes from calibration, not rote memorization. The signal's one blind spot is confidently wrong facts, which it cannot flag. A model's own doubt is thus a near-free substitute for a labelled dataset when teaching it when to abstain. Code and artifacts are available on request.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。