With the emergence of generative conversational large language models (LLMs) like ChatGPT, serving as virtual assistants in various fields, the stability and reliability of their responses have become crucial. However, during usage, it has been observed that these models tend to waver in their judgements when confronted with follow-up questions from users expressing skepticism or disagreement. In this work, we draw inspiration from questioning strategies in education and propose a \textsc{Follow-up Questioning Mechanism} along with two evaluation metrics to assess the judgement consistency of LLMs before and after exposure to disturbances. We evaluate the judgement consistency of ChatGPT, PaLM2-Bison, and Vicuna-13B under this mechanism across eight reasoning benchmarks. Empirical results show that even when the initial answers are correct, judgement consistency sharply decreases when LLMs face disturbances such as questioning, negation, or misleading. Additionally, we study these models' judgement consistency under various settings (sampling temperature and prompts) to validate this issue further, observing the impact of prompt tone and conducting an in-depth error analysis for deeper behavioral insights. Furthermore, we also explore several prompting methods to mitigate this issue and demonstrate their effectiveness\footnote{\url{https://github.com/NUSTM/LLMs-Waver-In-Judgements}}.

翻译：随着ChatGPT等生成式对话大语言模型作为各领域虚拟助手的涌现，其响应的稳定性和可靠性变得至关重要。然而，在使用过程中发现，当面对用户表达怀疑或不同意见的后续问题时，这些模型往往会在判断中摇摆不定。本研究受教育领域提问策略的启发，提出了一种“后续提问机制”及两项评估指标，用于衡量大语言模型在受到干扰前后的判断一致性。我们基于该机制在八个推理基准上评估了ChatGPT、PaLM2-Bison和Vicuna-13B的判断一致性。实验结果表明，即使初始答案正确，当大语言模型面临质疑、否定或误导等干扰时，其判断一致性会急剧下降。此外，我们研究了这些模型在不同设置（采样温度和提示）下的判断一致性以进一步验证该问题，观察到提示语气的影响，并进行了深入的错误分析以获取更深层次的行为洞察。最后，我们还探索了多种提示方法来缓解这一问题，并展示了其有效性\footnote{\url{https://github.com/NUSTM/LLMs-Waver-In-Judgements}}。