Tail-Aware HiFloat4: W4A4 Post-Training Quantization for Wan2.2 事件

Tail-Aware HiFloat4: W4A4 Post-Training Quantization for Wan2.2 arXiv:2605.26628v1 Announce Type: new Abstract: This report describes Tail-Aware HiFloat4, our submission to the low-bit text-to-video generation quantization challenge. Our method adapts the public ViDiT-Q post-training quantization pipeline to Wan2.2 under the HiFloat4 numerical format. We quantize the main linear layers in both Wan2.2 transformer modules with W4A4 HiFloat4 fake quantization, keep numerically sensitive boundary m