FactionFormer: Context-Driven Collaborative Vision Transformer Models for Edge Intelligence | IEEE Conference Publication | IEEE Xplore