FIMA-Q: Post-Training Quantization for Vision Transformers by Fisher Information Matrix Approximation | IEEE Conference Publication | IEEE Xplore