Ask Again, Then Fail: Large Language Models' Vacillations in Judgment

Code

• github.com/nustm/llms-waver-in-judgments
  OfficialIn papernone★ 12
• github.com/nustm/llms-waver-in-judgements
  OfficialIn papernone★ 12

Abstract

We observe that current conversational language models often waver in their judgments when faced with follow-up questions, even if the original judgment was correct. This wavering presents a significant challenge for generating reliable responses and building user trust. To comprehensively assess this issue, we introduce a Follow-up Questioning Mechanism along with two metrics to quantify this inconsistency, confirming its widespread presence in current language models. To mitigate this issue, we explore various prompting strategies for closed-source models; moreover, we develop a training-based framework Unwavering-FQ that teaches language models to maintain their originally correct judgments through synthesized high-quality preference data. Our experimental results confirm the effectiveness of our framework and its ability to enhance the general capabilities of models.

Tasks

Reproductions