Fetching the paper…

LAVT: Language-Aware Vision Transformer for Referring Image Segmentation · Around