与LLM捉迷藏：隐秘错误生成与自我改进诊断的对抗博弈
Hide and Seek with LLMs: An Adversarial Game for Sneaky Error Generation and Self-Improving Diagnosis
AI总结：
提出捉迷藏博弈（HSG）对抗框架，通过生成与检测欺骗性错误的动态博弈，显著提升LLM在数学推理中的错误诊断能力，准确率较基线提高16.8%-31.4%。
摘要：
大型语言模型（LLM）在跨领域的推理和生成方面表现出色，但仍难以识别和诊断复杂错误。这主要源于训练目标优先考虑正确答案，限制了对错误的接触和学习。尽管近期研究开始通过引入错误信号来解决这一问题，但大多数依赖浅层、静态的错误，限制了深层诊断能力的提升。为克服这一局限，我们提出了捉迷藏博弈（HSG），一个用于错误生成和诊断的动态对抗框架，并在数学问题求解上进行了评估。HSG涉及两个对抗角色：Sneaky，通过生成微妙、具有欺骗性的推理错误来“隐藏”；以及Diagnosis，通过准确检测这些错误来“寻找”。通过对抗性共同进化，错误隐蔽性和诊断精度均得到增强。在多个数学推理任务上的实验表明，HSG显著提升了错误诊断性能，相比GPT-4o等基线，准确率提高了16.8%至31.4%。我们还发布了一个包含欺骗性错误和诊断注释的挑战性数据集，作为未来研究的基准。
英文摘要：
Large Language Models (LLMs) excel in reasoning and generation across domains, but still struggle with identifying and diagnosing complex errors. This stems mainly from training objectives that prioritize correct answers, limiting exposure to and learning from errors. While recent studies have begun to address this by introducing error signals, most rely on shallow, static errors, restricting improvement in deep diagnostic ability. To overcome this, we propose Hide and Seek Game (HSG), a dynamic adversarial framework for error generation and diagnosis, and evaluate it on mathematical problem-solving. HSG involves two adversarial roles: Sneaky, which "hides" by generating subtle, deceptive reasoning errors, and Diagnosis, which "seeks" to accurately detect them. Through adversarial co-evolution, both error stealth and diagnostic precision are enhanced. Experiments on several math reasoning tasks show that HSG significantly boosts error diagnosis, achieving 16.8\%--31.4\% higher accuracy than baselines like GPT-4o. We also release a challenging dataset of deceptive errors and diagnostic annotations as a benchmark for future research.
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。
