Fetching the paper…

LRP-QViT: Mixed-Precision Vision Transformer Quantization via Layer-wise Relevance Propagation · Around