Tag
该论文提出了一个用于评估大语言模型在闭环物理交互中表现的基准，模型需通过连接机械臂、摄像头和传感器的系统在真实环境中执行任务并根据物理反馈调整行动。实验揭示了LLM在物理行动中的常见失败模式，如幻觉状态变化、糟糕的异常恢复和不稳定的长程规划。
This paper introduces Agent as Policy (AGP), a framework that enables general-purpose agents to directly control physical robots for manipulation tasks without task-specific training, achieving high success rates across various real-world scenarios.
A general-purpose agent directly controls a physical robot by interpreting visuals, writing executable programs, and revising actions based on physical feedback across diverse manipulation tasks, achieving high success rates without task-specific training.
The article proposes a conceptual 3-tier embodied AI architecture to address the limitations of scaling LLMs for agency, emphasizing physical cost auditing and offline sleep cycles.
SoftVTBench presents a synchronized visuo-tactile dataset and deformation-aware benchmark for evaluating physical interaction quality in deformable-object manipulation, including 4,000 demonstrations and a new success metric.
This paper demonstrates that training a world model through random physical exploration leads to latent representations that encode spatial semantic structure (direction and position) without any linguistic supervision, highlighting physical geometry as the organizing principle.