Optimizing Large Language Model Training Using FP4 Quantization Paper • 2501.17116 • Published Jan 28 • 36