批判还是顺从？思维链解释中推理的双刃剑效应
Critical or Compliant? The Double-Edged Sword of Reasoning in Chain-of-Thought Explanations
AI总结：
本研究通过扰动推理链、操控表述语气，探究多模态道德场景下CoT解释的双刃剑效应，发现用户易因结果认同、自信语气忽视VLM推理错误，呼吁设计引导批判性思考的解释机制。
摘要：
解释常被推崇为提升透明度的工具，但也会助长确认偏差：只要输出看起来可接受，用户就可能默认推理过程是正确的。我们通过系统性扰动推理链、操控表述语气，研究了多模态道德场景下思维链（Chain-of-Thought，CoT）解释的这种双刃剑作用。具体而言，我们分析了视觉语言模型（vision language models，VLMs）中的推理错误，以及这类错误如何影响用户信任与错误检测能力。研究结果揭示了两个关键效应：（1）用户常将信任等同于结果认同，即便推理存在缺陷仍持续依赖模型；（2）自信的表述语气会抑制用户的错误检测能力，同时维持用户对模型的依赖，这表明表述风格能够凌驾于正确性之上。这些结果凸显CoT解释既能澄清问题也会产生误导，强调NLP系统需要提供鼓励审视与批判性思考而非盲目信任的解释。所有代码将公开发布。
英文摘要：
Explanations are often promoted as tools for transparency, but they can also foster confirmation bias; users may assume reasoning is correct whenever outputs appear acceptable. We study this double-edged role of Chain-of-Thought (CoT) explanations in multimodal moral scenarios by systematically perturbing reasoning chains and manipulating delivery tones. Specifically, we analyze reasoning errors in vision language models (VLMs) and how they impact user trust and the ability to detect errors. Our findings reveal two key effects: (1) users often equate trust with outcome agreement, sustaining reliance even when reasoning is flawed, and (2) the confident tone suppresses error detection while maintaining reliance, showing that delivery styles can override correctness. These results highlight how CoT explanations can simultaneously clarify and mislead, underscoring the need for NLP systems to provide explanations that encourage scrutiny and critical thinking rather than blind trust. All code will be released publicly.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
