VLALight: Lightweight Vision-Language-Action Models for Emergency-Aware Traffic Signal Control
VLALight is proposed, a lightweight end-to-end vision-language-action framework that directly maps intersection observations and signal-phase information to discrete signal actions and enables direct action prediction with a compact 0.5 B-parameter model.
Ke-Mou Jiang, Mao-Nan Wang, Xingchen Zou et al.
· 0 citations