Language-Aware Vision Transformer for Referring Segmentation.

Referring segmentation is a fundamental vision-language task that aims to segment out an object from an image or video in accordance with a natural language description. One of the key challenges behind this task is leveraging the referring expression for highlighting relevant positions in the image or video frames. A paradigm for tackling this problem in both the image and the video domains is to leverage a powerful vision-language ("cross-modal") decoder to fuse features independently extracted from a vision encoder and a language encoder.

Tunable Room-Temperature Ferromagnetism in Two-Dimensional CrTe.

The manipulation of magnetism provides a unique opportunity for the development of data storage and spintronic applications. Until now, electrical control, pressure tuning, stacking structure dependence, and nanoscale engineering have been realized. However, as the dimensions are decreased, the decrease of the ferromagnetism phase transition temperature () is a universal trend in ferromagnets.