多智能体视觉语言导航：形式化、基准与方法
原标题：Systematic Multi-Agent Vision-and-Language Navigation: Formulation, Benchmark, and Method
AI 摘要
该论文首次系统形式化多智能体视觉语言导航（VLN）为带约束的协调问题，任务包含依赖与资源约束（presence locks、holding chains）。作者构建了 MAVLN 基准，含 145 个场景、11,724 个 episode、最多四智能体团队及三种指令模式，并提出 TRISS 系统，结合 LLM 子任务调度器、共享拓扑记忆与冲突感知执行机制。实验表明 TRISS 可作为综合基线，但在调度、规划与执行上仍有较大提升空间。
正文节选
[Page 1] 1 Systematic Multi-Agent Vision-and-Language Navigation: Formulation, Benchmark, and Method Yunzhe Xu and Zhe Liu Abstract—Vision-and-Language Navigation (VLN) has largely focused on a single agent following a single instruction, yet many real-world applications require teams of robots to tackle tasks beyond the capabilities of any individual agent. We present Systematic Multi-Agent V