Search Results for author:
Found 4 papers, 0 papers with code
MPTQ-ViT: Mixed-Precision Post-Training Quantization for Vision Transformer
However, without considering the asymmetry in activations and relying on hand-crafted settings, these methods often struggle to maintain performance under low-bit quantization.
TSPTQ-ViT: Two-scaled post-training quantization for vision transformer
Due to the non-normally distributed values after Softmax and GeLU, post-training quantization on ViTs results in severe accuracy degradation.
Learnable Mixed-precision and Dimension Reduction Co-design for Low-storage Activation
Recently, deep convolutional neural networks (CNNs) have achieved many eye-catching results.
Compression-aware Projection with Greedy Dimension Reduction for Convolutional Neural Network Activations
Our test results show that the proposed methods effectively reduce 2. 91x~5. 97x memory access with negligible accuracy drop on MobileNetV2/ResNet18/VGG16.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
