Search papers, labs, and topics across Lattice.
Nanjing University, Robotics
2
0
5
Tactile reinforcement learning unlocks a 24.7% performance boost over GPT-4o in tactile reasoning, revealing that MLLMs can learn to ground reasoning in physical contact.
VLA models struggle not with *what* they see, but with *how* they see it: FocusVLA reveals that directing attention to task-relevant visual regions unlocks significant performance gains in robotic manipulation.
