1 paper · 1 filter
Gengluo Li, Shangpin Peng, Xingyu Wan +16
Vision Large Language Models (VLLMs) have achieved remarkable success in modern text-rich visual understanding. However, their perceptual robustness in the face of the continuous m…