PAIR: Bridging Perception and Action in Vision-Language-Action Models
K. Feng, G. Sun, Ang Li · arXiv, 2026
About Me
I am a student in Computer Science at the University of Maryland, College Park, advised by Prof. Ang Li and Prof. Furong Huang. My research focuses on embodied AI and representation learning.
News
ROCKET accepted to NeurIPS 2026.
Publications
DIVA: Dual-Space Intent-Aware Visual Attenuation for Vision-Language-Action Policies
K. Feng, G. Sun, et al. · arXiv, 2026
ROCKET: Residual-Oriented Multi-Layer Alignment for Spatially-Aware Vision-Language-Action Models
G. Sun, T. Du, K. Feng, et al. · arXiv, 2026
Drop-Then-Recovery: How Redundant Are Vision-Language-Action Models?
G. Sun, K. Feng, et al. · arXiv, 2026
Causeway: Restoring Task Accessibility for Instruction Switching in VLA Policies
Q. Wang, K. Feng, et al. · arXiv, 2026