从谬误推理：通过逻辑谬误理解增强大型语言模型的逻辑推理能力
Reason from Fallacy: Enhancing Large Language Models' Logical Reasoning through Logical Fallacy Understanding
AI总结：
本文提出WHAT、WHY、HOW三维LFU任务，并构建基于GPT-4与少量人工校验的LFUD数据集，用于评估和微调LLM，显著提升其逻辑推理能力。
摘要：
大型语言模型（LLMs）已在许多推理任务中表现出良好性能，但在包括逻辑推理在内的一些复杂推理任务上仍存在困难。LLMs在逻辑推理上表现欠佳的一个不可忽视的原因，是它们忽视了对逻辑谬误的正确理解。为评估LLMs的逻辑谬误理解（LFU）能力，本文从WHAT、WHY和HOW三个认知维度提出了五项具体任务。针对这些LFU任务，我们基于GPT-4并辅以少量人工工作，成功构建了一个新数据集LFUD。大量实验证明，LFUD不仅可用于评估LLMs的LFU能力，还可用于微调LLMs，使其在逻辑推理上获得显著增强的性能。
英文摘要：
Large Language Models (LLMs) have demonstrated good performance in many reasoning tasks, but they still struggle with some complicated reasoning tasks including logical reasoning. One non-negligible reason for LLMs' suboptimal performance on logical reasoning is their overlooking of understanding logical fallacies correctly. To evaluate LLMs' capability of logical fallacy understanding (LFU), we propose five concrete tasks from three cognitive dimensions of WHAT, WHY, and HOW in this paper. Towards these LFU tasks, we have successfully constructed a new dataset LFUD based on GPT-4 accompanied by a little human effort. Our extensive experiments justify that our LFUD can be used not only to evaluate LLMs' LFU capability, but also to fine-tune LLMs to obtain significantly enhanced performance on logical reasoning.
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。
