用多智能体让大模型自动建模任务，无需人工定义环境
Adaptive Domain Modeling with Language Models: A Multi-Agent Approach to Task Planning
- 大模型智能体协作生成和调整领域模型与目标
- 在虚拟家庭环境中实现复杂任务规划，表现优于基线
- 适合需要动态适应新任务的机器人系统研究者
我们提出TAPAS（基于任务的适应与规划智能体），一种将大型语言模型（LLMs）与符号规划相结合的多智能体框架，可在无需手动定义环境模型的情况下解决复杂任务。TAPAS通过基于语言模型的专用智能体，利用结构化工具调用机制协作生成并动态调整领域模型、初始状态和目标规范。下游智能体可通过工具调用向上游智能体请求修改，从而在不重新定义领域的情况下适应新属性与约束。一个基于ReAct（思考+行动）风格的执行智能体，结合自然语言计划翻译，实现了动态生成计划与真实机器人能力之间的衔接。TAPAS在基准规划领域及虚拟家庭（VirtualHome）模拟的真实世界环境中均展现出优异性能。
原文摘要 · Abstract (English)
We introduce TAPAS (Task-based Adaptation and Planning using AgentS), a multi-agent framework that integrates Large Language Models (LLMs) with symbolic planning to solve complex tasks without the need for manually defined environment models. TAPAS employs specialized LLM-based agents that collaboratively generate and adapt domain models, initial states, and goal specifications as needed using structured tool-calling mechanisms. Through this tool-based interaction, downstream agents can request modifications from upstream agents, enabling adaptation to novel attributes and constraints without manual domain redefinition. A ReAct (Reason+Act)-style execution agent, coupled with natural language plan translation, bridges the gap between dynamically generated plans and real-world robot capabilities. TAPAS demonstrates strong performance in benchmark planning domains and in the VirtualHome simulated real-world environment.
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。