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{https://github.com/NUSTM/LLMs-Waver-In-Judgements}.

利用教育中的质问策略，我们提出了一种 “跟进质问机制” 以及两个评估指标，用于评估大型语言模型在面对质疑时的判断一致性，并对 ChatGPT、PaLM2-Bison 和 Vicuna-13B 进行了评估。实证结果表明，即使初始答案是正确的，当语言模型面临质疑、否定或误导等干扰时，判断一致性会大幅降低。此外，我们还研究了不同设置下模型的判断一致性，并观察了提示语气对此问题的影响，进行了深入的错误分析以获得更深入的行为洞察。此外，我们还探索了几种提示方法以缓解此问题，并展示了它们的有效性。