LRA-QViT: Integrating Low-Rank Approximation and Quantization for Robust and Efficient Vision Transformers | Read Paper on Bytez