Instance-Aware Group Quantization for Vision Transformers

arXiv:2404.00928v1 Announce Type: cross
Abstract: Post-training quantization (PTQ) is an efficient model compression technique that quantizes a pretrained full-precision model using only a small calibration set of unlabeled samples without retraining. PTQ methods for convolutional neural networks (CNNs) provide quantization results comparable to full-precision counterparts. Directly applying them to vision transformers (ViTs), however, incurs severe performance degradation, mainly due to the differences in architectures between CNNs and ViTs. In particular, the distribution of activations for each channel vary drastically according …

abstract arxiv cnns compression convolutional neural networks cs.cv cs.lg however instance networks neural networks precision quantization results retraining samples set small them training transformers type vision vision transformers