Fetching the paper…

VL-LTR: Learning Class-wise Visual-Linguistic Representation for Long-Tailed Visual Recognition · Around