通过解耦表征对齐提升潜在扩散模型
Boosting Latent Diffusion Models via Disentangled Representation Alignment
AI总结：
本文提出Semantics-Disentangled VAE，通过非线性映射架构提升VAE的语义解耦能力，实现高质量图像生成，FID达1.21。
摘要：
潜在扩散模型（LDMs）依赖变分自编码器（VAEs）提供的压缩潜在空间进行高质量图像生成。最近的研究尝试通过直接采用LDM训练中的对齐策略来获得生成友好的VAEs，利用视觉基础模型（VFMs）作为表征对齐目标。然而，这种对齐范式忽略了LDMs和VAEs在表征需求上的根本差异。简单的局部块到高维语义的特征映射可能引起语义崩溃，导致细粒度属性的丢失。本文揭示关键见解：不同于LDMs受益于高层全局语义，生成友好的VAE必须具备强语义解耦能力，以结构化方式保留细粒度属性信息。为解决这一差距，我们提出Semantics-Disentangled VAE（Send-VAE）。不同于以往浅层对齐方法，Send-VAE引入非线性映射架构，有效连接VAEs的局部结构与VFMs的密集语义，从而在潜在空间中鼓励涌现的解耦特性，无需显式正则化。大量实验建立了一种通过低级属性分离评估VAE潜在空间的新范式，并证明Send-VAE在ImageNet 256x256上实现了最先进的生成质量（FID为1.21）
英文摘要：
Latent Diffusion Models (LDMs) rely heavily on the compressed latent space provided by Variational Autoencoders (VAEs) for high-quality image generation. Recent studies have attempted to obtain generation-friendly VAEs by directly adopting alignment strategies from LDM training, leveraging Vision Foundation Models (VFMs) as representation alignment targets. However, such alignment paradigms overlook the fundamental differences in representational requirements between LDMs and VAEs. Simple feature mapping from local patches to high-dimensional semantics can induce semantic collapse, leading to the loss of fine-grained attributes. In this paper, we reveal a key insight: unlike LDMs that benefit from high-level global semantics, a generation-friendly VAE must possess strong semantic disentanglement capabilities to preserve fine-grained, attribute-level information in a structured manner. To address this discrepancy, we propose the Semantic-Disentangled VAE (Send-VAE). Deviating from previous shallow alignment approaches, Send-VAE introduces a non-linear mapping architecture to effectively bridge the local structures of VAEs and the dense semantics of VFMs, thereby encouraging emergent disentangled properties in the latent space without explicit regularization. Extensive experiments establish a new paradigm for evaluating VAE latent spaces via low-level attribute separability and demonstrate that Send-VAE achieves state-of-the-art generation quality (FID of 1.21) on ImageNet 256x256.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
