Although large language models (LLMs) have achieved excellent performance in
a variety of evaluation benchmarks, they still struggle in complex reasoning
tasks which require specific knowledge and multi-hop reasoning. To improve the
reasoning abilities, we propose \textbf{ChatCoT}, a tool-augmented
chain-of-thought reasoning framework for chat-based LLMs. In ChatCoT, we model
the chain-of-thought~(CoT) reasoning as multi-turn conversations, to utilize
tools in a more natural way through chatting. At each turn, LLMs can either
interact with tools or perform the reasoning. %framework to decompose reasoning
and manipulate tools for complex reasoning tasks via multi-round dialogue. %Our
idea is to decompose the complex problems into several sub-problems and solve
these sub-problems with the help of external tools. Our approach can
effectively leverage the multi-turn conversation ability of chat-based LLMs,
and integrate the thought chain following and tools manipulation in a unified
way. Specially, we initialize the early turns of the conversation by the tools,
tasks and reasoning format, and propose an iterative \emph{tool-augmented
reasoning} step to perform step-by-step tool-augmented reasoning. The
experiment results on two complex reasoning datasets (MATH and HotpotQA) have
shown the effectiveness of ChatCoT on complex reasoning tasks, achieving a
6.8\% relative improvement over the state-of-the-art baseline. %and ChatCoT can
achieve performance on Math. Our code and data are available at:
https://github.com/RUCAIBOX/ChatCoT.

ChatCoT 是一种用于聊天式大语言模型的工具辅助思维链推理框架，大幅提高了在复杂推理任务上的表现。