不要轻信前提：评估大语言模型的前提批判能力
Don't Take the Premise for Granted: Evaluating the Premise Critique Ability of Large Language Models
AI总结：
本研究提出前提批判基准，评估大语言模型在面对错误前提时的批判能力，揭示其在推理与前提批判上的差异及改进需求。
摘要：
大型语言模型（LLMs）已经经历了快速的发展，展现出显著的能力。然而，一个显著的弱点依然存在：LLMs往往不加批判地接受错误或矛盾的前提，导致推理效率低下和输出不可靠。这突显了LLMs具备前提批判能力的重要性，即能够主动识别和阐述输入前提中的错误。大多数现有研究评估LLMs在理想环境下的推理能力，主要忽略了面对错误前提时的脆弱性。因此，我们引入了前提批判基准（PCBench），通过在三个难度级别中整合四种错误类型，并配以多维评估指标进行设计。我们对15种代表性LLMs进行了系统评估。我们的发现显示：（1）大多数模型依赖显式提示来检测错误，自主批判能力有限；（2）前提批判能力取决于问题难度和错误类型，直接矛盾比复杂或程序性错误更容易检测；（3）推理能力与前提批判能力并不总是一致；（4）错误前提会促使推理模型过度思考，明显延长响应时间，因为反复尝试解决冲突。这些发现强调了增强LLMs对输入有效性的主动评估的紧迫性，将前提批判定位为开发可靠、以人为本系统的基础能力。代码可在https://github.com/MLGroupJLU/Premise_Critique获取。
英文摘要：
Large language models (LLMs) have witnessed rapid advancements, demonstrating remarkable capabilities. However, a notable vulnerability persists: LLMs often uncritically accept flawed or contradictory premises, leading to inefficient reasoning and unreliable outputs. This emphasizes the significance of possessing the \textbf{Premise Critique Ability} for LLMs, defined as the capacity to proactively identify and articulate errors in input premises. Most existing studies assess LLMs' reasoning ability in ideal settings, largely ignoring their vulnerabilities when faced with flawed premises. Thus, we introduce the \textbf{Premise Critique Bench (PCBench)}, designed by incorporating four error types across three difficulty levels, paired with multi-faceted evaluation metrics. We conducted systematic evaluations of 15 representative LLMs. Our findings reveal: (1) Most models rely heavily on explicit prompts to detect errors, with limited autonomous critique; (2) Premise critique ability depends on question difficulty and error type, with direct contradictions being easier to detect than complex or procedural errors; (3) Reasoning ability does not consistently correlate with the premise critique ability; (4) Flawed premises trigger overthinking in reasoning models, markedly lengthening responses due to repeated attempts at resolving conflicts. These insights underscore the urgent need to enhance LLMs' proactive evaluation of input validity, positioning premise critique as a foundational capability for developing reliable, human-centric systems. The code is available at https://github.com/MLGroupJLU/Premise_Critique.
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。
