ChatCoT: Tool-Augmented Chain-of-Thought Reasoning on Chat-based Large Language Models
This addresses the challenge of enhancing reasoning abilities in LLMs for tasks requiring specific knowledge and multi-step reasoning, representing an incremental advancement.
The paper tackles the problem of improving complex reasoning in chat-based large language models by proposing ChatCoT, a tool-augmented chain-of-thought framework, achieving a 7.9% relative improvement over state-of-the-art baselines on datasets like MATH and HotpotQA.
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 ChatCoT, a tool-augmented chain-of-thought reasoning framework for chat-based LLMs (e.g., ChatGPT). 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. 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 knowledge about tools, tasks, and reasoning format, and propose an iterative 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 7.9% relative improvement over the state-of-the-art baseline. Our code and data are available at: \url{https://github.com/RUCAIBOX/ChatCoT}.