1 citations · 1 across the 7 of their papers we have counts for
1 paper · 1 filter
Mengyuan Fan, Bokai Huang, JiaMing Pan +4
Vision Transformers (ViTs) achieve strong performance on image recognition and mobile vision applications, but their high-dimensional linear projections and attention computations…