BinaryDM: Towards Accurate Binarization of Diffusion Model

arXiv:2404.05662v1 Announce Type: new
Abstract: With the advancement of diffusion models (DMs) and the substantially increased computational requirements, quantization emerges as a practical solution to obtain compact and efficient low-bit DMs. However, the highly discrete representation leads to severe accuracy degradation, hindering the quantization of diffusion models to ultra-low bit-widths. In this paper, we propose BinaryDM, a novel accurate quantization-aware training approach to push the weights of diffusion models towards the limit of 1-bit. Firstly, we present a Learnable Multi-basis …

abstract accuracy advancement arxiv compact computational cs.cv diffusion diffusion model diffusion models however leads low paper practical quantization representation requirements solution type