no code implementations • 26 Jan 2024 • Yu-Shan Tai, An-Yeu, Wu
However, without considering the asymmetry in activations and relying on hand-crafted settings, these methods often struggle to maintain performance under low-bit quantization.
no code implementations • 22 May 2023 • Yu-Shan Tai, Ming-Guang Lin, An-Yeu, Wu
Due to the non-normally distributed values after Softmax and GeLU, post-training quantization on ViTs results in severe accuracy degradation.
no code implementations • 16 Jul 2022 • Yu-Shan Tai, Cheng-Yang Chang, Chieh-Fang Teng, AnYeu, Wu
Recently, deep convolutional neural networks (CNNs) have achieved many eye-catching results.
no code implementations • 17 Oct 2021 • Yu-Shan Tai, Chieh-Fang Teng, Cheng-Yang Chang, An-Yeu Wu
Our test results show that the proposed methods effectively reduce 2. 91x~5. 97x memory access with negligible accuracy drop on MobileNetV2/ResNet18/VGG16.