Foundation models, such as Large language Models (LLMs), have attracted significant amount of interest due to their large number of applications. Existing works show that appropriate prompt design, such as Chain-of-Thoughts, can unlock LLM's powerful capacity in diverse areas. However, when handling tasks involving repetitive sub-tasks and/or deceptive contents, such as arithmetic calculation and article-level fake news detection, existing prompting strategies either suffers from insufficient expressive power or intermediate errors triggered by hallucination. To make LLM more discerning to such intermediate errors, we propose to guide LLM with a Divide-and-Conquer program that simultaneously ensures superior expressive power and disentangles task decomposition, sub-task resolution, and resolution assembly process. Theoretic analysis reveals that our strategy can guide LLM to extend the expressive power of fixed-depth Transformer. Experiments indicate that our proposed method can achieve better performance than typical prompting strategies in tasks bothered by intermediate errors and deceptive contents, such as large integer multiplication, hallucination detection and misinformation detection.

翻译：基础模型，尤其是大型语言模型（LLMs），因其广泛的应用而备受关注。现有研究表明，适当的提示设计（如思维链）能够激发LLMs在多个领域的强大能力。然而，在处理涉及重复性子任务和/或欺骗性内容的任务（如算术计算和文章级假新闻检测）时，现有提示策略要么表达能力不足，要么因幻觉引发中间错误。为了让LLM对这些中间错误更具辨别力，我们提出用分治程序引导LLM，该方法同时确保强大的表达能力，并将任务分解、子任务求解与结果整合过程解耦。理论分析表明，我们的策略能引导LLM扩展固定深度Transformer的表达能力。实验表明，在受中间错误和欺骗性内容困扰的任务（如大整数乘法、幻觉检测和虚假信息检测）中，我们的方法能够取得优于典型提示策略的性能。