StepTool: Enhancing Multi-Step Tool Usage in LLMs through Step-Grained Reinforcement Learning (2410.07745v3)

Abstract: Despite powerful text generation capabilities, LLMs still need to learn how to utilize external tools to solve complex tasks, a process known as tool learning. Existing methods primarily rely on supervised fine-tuning to enhance tool-use capabilities, treating tool learning as a text-generation task while overlooking the decision-making complexities inherent in multi-step contexts. In this work, we propose modeling tool learning as a dynamic decision-making task and introduce StepTool, a novel step-grained reinforcement learning framework that enhances the multi-step tool use capabilities of LLMs. StepTool consists of two main components: Step-grained Reward Shaping, which assigns rewards at each tool interaction based on the success of tool invocation and its contribution to the task; and Step-grained Optimization, which uses policy gradient methods to optimize the model in a multi-step manner. Experimental results demonstrate that StepTool significantly outperforms existing methods in multi-step, tool-based tasks, offering a robust solution for tool learning.