Towards Accurate Post-training Quantization for Diffusion Models

arXiv:2305.18723v4 Announce Type: replace
Abstract: In this paper, we propose an accurate data-free post-training quantization framework of diffusion models (ADP-DM) for efficient image generation. Conventional data-free quantization methods learn shared quantization functions for tensor discretization regardless of the generation timesteps, while the activation distribution differs significantly across various timesteps. The calibration images are acquired in random timesteps which fail to provide sufficient information for generalizable quantization function learning. Both issues cause sizable quantization errors with obvious image generation performance degradation. …

abstract arxiv cs.cv data diffusion diffusion models distribution framework free functions image image generation images learn paper quantization tensor training type while