Paper-Conference

Extremely Low-Bit Convolution Optimization for Quantized Neural Network on Modern Computer Architectures

Differentiable Soft Quantization: Bridging Full-Precision and Low-Bit Neural Networks