Jiahao Li

: A Tool-Augmented Multi-Agent Framework for Long Visual Document Understanding

PhysX-CoT: Structured Physical Reasoning from a Single Image to Simulation-Ready 3D Assets

Simulation-ready 3D assets are central to robotics and embodied AI. Generating them from a single image is usually framed as a vision-language model that emits a serialized asset for a decoder to turn into geometry and physical fields, leaving the image-to-3D reasoning implicit. We argue the limiting factor is this out...

Lingjing: A Simulation Testbed for Multi-Agent Embodied Tasks in Open-Ended Cities

Lingjing provides a unified testbed that enables reproducible end-to-end evaluation and systematic failure diagnosis in urban multi-agent embodied intelligence and exposes shared physical and structured urban state to agents.