24 citations · 114 across the 37 of their papers we have counts for
1 paper · 2 filters
Zhiwen Li, Zhongjie Duan, Jinyan Ye +4
Replicating In-Context Learning (ICL) in computer vision remains challenging due to task heterogeneity. We propose \textbf{VIRAL}, a framework that elicits visual reasoning from a…