
TX
Tongda Xu, Yu Zhang, Jun Hu, Huijuan Huang
· 1 min read
ResearcharXiv cs.CV
Lossless Latent Diffusion Models
arXiv:2611.25454v1 Announce Type: new
Abstract: We study the problem of training image diffusion models while preserving pixel-exact reconstruction. Prior works typically address this problem by directly modeling images in pixel space, requiring modifications to the diffusion training objective or model architecture. In this work, we introduce a different paradigm: rather than modifying diffusion models, we construct a lossless transformation that maps pixels into a representation more amenable to diffusion modeling. We term our approach the Lossless Latent Diffusion Model (LLDM). We construct LLDM using residual latent transformation and uneven latent scaling, and explain its effectiveness through a proposed metric, semantic rank priority (SRP). Empirically, LLDM substantially outperforms prior pixel-space diffusion models in generation quality and even surpasses several widely used lossy VAEs. Furthermore, we show that SRP is strongly correlated with downstream generative performance.
Original source
This story was published by arXiv cs.CV and written by Tongda Xu, Yu Zhang, Jun Hu, Huijuan Huang. SyncAI.news shows a preview; the complete article is on the publisher's site.
Read the full story on arxiv.org


