飞行猪、FaR及更广泛的领域：评估LLM在反事实世界中的推理能力
Flying Pigs, FaR and Beyond: Evaluating LLM Reasoning in Counterfactual Worlds
AI总结：
本文研究了LLM在反事实场景下的推理能力，提出FaR方法以增强模型的元认知，减少逻辑与知识冲突带来的性能下降。
摘要：
推理领域的一个基本挑战是导航假设性的反事实世界，其中逻辑可能与固有知识冲突。我们通过研究大型语言模型（LLM）在这一前沿问题上的表现，探讨了当上下文与模型参数知识冲突时，LLM是否能进行逻辑推理。为了系统分析，我们首先引入了CounterLogic基准测试，专门用于分离逻辑有效性与知识对齐。对11个LLM在六个多样化推理数据集上的评估显示，模型在反事实场景中的准确率平均下降14%，相比知识对齐的场景。我们假设这一差距并非源于逻辑处理的缺陷，而是源于无法管理上下文与知识之间的认知冲突。受人类元认知启发，我们提出了一种简单但强大的干预方法：Flag & Reason（FaR），即在推理前提示模型首先标记潜在的知识冲突。这一元认知步骤非常有效，将性能差距缩小到仅7%，并提高整体准确率4%。我们的发现诊断并研究了现代LLM推理中的关键限制，并展示了元认知意识如何使它们成为更稳健和可靠的思考者。
英文摘要：
A fundamental challenge in reasoning is navigating hypothetical, counterfactual worlds where logic may conflict with ingrained knowledge. We investigate this frontier for Large Language Models (LLMs) by asking: Can LLMs reason logically when the context contradicts their parametric knowledge? To facilitate a systematic analysis, we first introduce CounterLogic, a benchmark specifically designed to disentangle logical validity from knowledge alignment. Evaluation of 11 LLMs across six diverse reasoning datasets reveals a consistent failure: model accuracy plummets by an average of 14% in counterfactual scenarios compared to knowledge-aligned ones. We hypothesize that this gap stems not from a flaw in logical processing, but from an inability to manage the cognitive conflict between context and knowledge. Inspired by human metacognition, we propose a simple yet powerful intervention: Flag & Reason (FaR), where models are first prompted to flag potential knowledge conflicts before they reason. This metacognitive step is highly effective, narrowing the performance gap to just 7% and increasing overall accuracy by 4%. Our findings diagnose and study a critical limitation in modern LLMs' reasoning and demonstrate how metacognitive awareness can make them more robust and reliable thinkers.
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。
