Search papers, labs, and topics across Lattice.
Beijing Institute of Technology
3
0
5
A novel multi-agent framework enhances zero-shot 3D understanding by iteratively optimizing viewpoints and integrating fragmented observations, leading to significant performance gains.
Action chunk utilization triples and physical execution steps drop by over 50%, resulting in a 5.83x speedup in VLA model deployment without sacrificing performance.
PhysAgent achieves unprecedented levels of physical accuracy and diversity in 4D motion synthesis by leveraging trajectory-grounded feedback from multiple agents.
