Elucidating Representation Degradation Problem in Diffusion Model Training
Diffusion models have achieved remarkable success, yet their training remains inefficient due to a severe optimization bottleneck, which we term Representation Degradation. As noise levels increase, the outputs of the trained model exhibit progressive structural distortion, which can destabilize training and impair generation quality. Our analysis suggests that this instability is driven by mismatched target recoverability, which is associated with Neural Tangent Kernel (NTK) spectral weakening and effective low-rank behavior. To address this, we propose Elucidated Representation Diffusion (ERD), a plug-and-play framework that dynamically reallocates optimization effort according to effective recoverability. By stabilizing representation learning without external supervision, ERD accelerates convergence and achieves strong empirical performance across diffusion backbones.
Code (0)
등록된 구현이 없습니다.
Tasks
Representation LearningSimilar Papers 제목 키워드 기반
LLDiffusion: Learning Degradation Representations in Diffusion Models for Low-Light Image Enhancement
Current deep learning methods for low-light image enhancement (LLIE) typically rely on pixel-wise mapping learned from paired data. However, these methods often overlook the importance of considering degradation represen…
Image EnhancementImage GenerationLow-Light Image EnhancementDR-BFR: Degradation Representation with Diffusion Models for Blind Face Restoration
Blind face restoration (BFR) is fundamentally challenged by the extensive range of degradation types and degrees that impact model generalization. Recent advancements in diffusion models have made considerable progress i…
Blind Face RestorationContrastive LearningDenoisingElucidating and Endowing the Diffusion Training Paradigm for General Image Restoration
While diffusion models demonstrate strong generative capabilities in image restoration (IR) tasks, their complex architectures and iterative processes limit their practical application compared to mainstream reconstructi…
Image RestorationElucidating the Exposure Bias in Diffusion Models
Diffusion models have demonstrated impressive generative capabilities, but their \textit{exposure bias} problem, described as the input mismatch between training and sampling, lacks in-depth exploration. In this paper, w…
AttributeImage GenerationDeeDSR: Towards Real-World Image Super-Resolution via Degradation-Aware Stable Diffusion
Diffusion models, known for their powerful generative capabilities, play a crucial role in addressing real-world super-resolution challenges. However, these models often focus on improving local textures while neglecting…
Contrastive LearningDenoisingImage Super-ResolutionSuper-Resolution