Jiangmiao Pang

The paper proposes Astra, an agentic framework that equips Vision-Language Models (VLMs) with the ability to perform spatial reasoning by actively generating and utilizing imagined visual evidence from a world simulator.

The paper proposes Astra, an agentic framework that equips Vision-Language Models (VLMs) with the ability to perform spatial reasoning by actively generating and utilizing imagined visual evidence from a world simulator.

The paper proposes Astra, an agentic framework that equips Vision-Language Models (VLMs) with the ability to perform spatial reasoning by actively generating and utilizing imagined visual evidence from a world simulator.

The paper presents EBench, a simulation benchmark for diagnosing generalist mobile manipulation policies.

The paper introduces MemoryWAM, a world action model with efficient persistent memory for robotic manipulation tasks.

This paper introduces the Kinematic-Aware Articulation Interface (KAI), a representation that captures the kinematic structure of articulated objects, improving sample efficiency and generalization in robot manipulation.