让工具服务器自带记忆，跨任务跨框架复用，提升大模型调用效率。
ToolAtlas: Learning Once, Reusing Everywhere with Tool-Side Memory
- 工具端构建图谱化记忆，记录能力边界与组合关系。
- 在两个基准上，pass@1最高提升21.61%，pass@4提升18.61%。
- 无需重训或探索，即可跨环境、跨框架直接复用。
大型语言模型（LLM）代理越来越多依赖由共享提供者服务的外部工具，并由异构下游代理访问。现有方法通过代理端参数更新、提示优化或代理侧记忆来改进工具使用，导致工具知识难以共享且仅限于过往任务中的行为。我们主张可复用的工具知识应由工具提供者维护。本文提出ToolAtlas，一种基于图结构的框架，通过执行验证的探测，在工具提供方构建持久化的工具能力、失败边界及跨工具组合的记忆。推理时，代理通过自适应图遍历查询该工具记忆。在基于MCP的两个基准上，涵盖八项服务，ToolAtlas在pass@1上最高比现有工具端优化和代理端记忆基线提升21.61%，pass@4提升18.61%。同一工具记忆在不同环境实例和代理框架间迁移，无需重训练或任务期探索，分别带来高达24.16%/16.22%和17.49%/14.27%的相对提升。消融实验表明，这些增益源于将工具中心的记忆组织与能力引导的执行探测相结合。结果确立了提供方工具记忆作为一种有效且可复用的范式。代码已开源：https://github.com/PuppyKnightUniversity/ToolAtlas。
原文摘要 · Abstract (English)
Large language model (LLM) agents increasingly rely on external tools served by shared providers and accessed by heterogeneous downstream agents. Existing approaches improve tool use on the agent side through parameter updates, prompt refinement, or agent-side memory, making tool knowledge difficult to share and limited to behaviors observed in past tasks. We argue that reusable tool knowledge should instead be maintained by the tool provider. We introduce ToolAtlas, a graph-based framework that builds a persistent provider-side tool memory of tool capabilities, failure boundaries, and cross-tool compositions through execution-verified probing. At inference time, agents query the tool memory via adaptive graph traversal. Across two MCP-based benchmarks spanning eight services, ToolAtlas outperforms existing tool-side optimization and agent-side memory baselines by up to 21.61% in pass@1 and 18.61% in pass@4. The same tool memory also transfers across environment instances and agent frameworks without retraining or task-time exploration, yielding up to 24.16%/16.22% and 17.49%/14.27% relative gains in pass@1/pass@4, respectively. Ablation studies show that these gains arise from combining tool-centered memory organization with capability-guided execution probing. These results establish provider-side tool memory as an effective and reusable paradigm for tool servers. Our code is in: https://github.com/PuppyKnightUniversity/ToolAtlas.
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。