用多模态大模型实现开放世界工具选择,支持新工具无需重训。
RaTA-Tool: Retrieval-based Tool Selection with Multimodal Large Language Models

- 通过检索匹配任务描述与工具语义,替代固定映射。
- 在开放世界多模态场景下准确率提升显著,优于现有方法。
- 适用于需动态调用外部工具的智能系统研发者。
基于基础模型的工具学习旨在赋予AI系统调用外部资源(如API、计算工具、专用模型)以解决超越单一语言生成能力的复杂任务。尽管大语言模型(LLMs)和多模态大语言模型(MLLMs)在推理与感知方面取得进展,现有工具使用方法仍主要局限于文本输入和封闭世界设置,难以理解多模态用户指令,也无法泛化到训练中未见的工具。本文提出RaTA-Tool框架,实现开放世界多模态工具选择。该方法不直接学习查询到工具标识符的映射,而是让MLLM将多模态查询转化为结构化任务描述,并通过匹配该表示与语义丰富的机器可读工具描述来检索最适工具。此检索范式天然支持新增工具而无需重新训练。为进一步提升任务描述与工具选择的一致性,引入基于直接偏好优化(DPO)的改进阶段。为推动该方向研究,我们还构建了首个开放世界多模态工具使用数据集,其工具描述标准化来源于Hugging Face模型卡。大量实验表明,该方法在开放世界多模态场景下显著提升工具选择性能。
原文摘要 · Abstract (English)
Tool learning with foundation models aims to endow AI systems with the ability to invoke external resources -- such as APIs, computational utilities, and specialized models -- to solve complex tasks beyond the reach of standalone language generation. While recent advances in Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs) have expanded their reasoning and perception capabilities, existing tool-use methods are predominantly limited to text-only inputs and closed-world settings. Consequently, they struggle to interpret multimodal user instructions and cannot generalize to tools unseen during training. In this work, we introduce RaTA-Tool, a novel framework for open-world multimodal tool selection. Rather than learning direct mappings from user queries to fixed tool identifiers, our approach enables an MLLM to convert a multimodal query into a structured task description and subsequently retrieve the most appropriate tool by matching this representation against semantically rich, machine-readable tool descriptions. This retrieval-based formulation naturally supports extensibility to new tools without retraining. To further improve alignment between task descriptions and tool selection, we incorporate a preference-based optimization stage using Direct Preference Optimization (DPO). To support research in this setting, we also introduce the first dataset for open-world multimodal tool use, featuring standardized tool descriptions derived from Hugging Face model cards. Extensive experiments demonstrate that our approach significantly improves tool-selection performance, particularly in open-world, multimodal scenarios.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。