通过能量分析优化扩散模型生成高清图像，减少伪影并提升质量。
RectifiedHR: High-Resolution Diffusion via Energy Profiling and Adaptive Guidance Scheduling
- 基于能量轨迹动态调节无分类器引导强度
- 稳定度达0.9998，一致性达0.9873，优于固定引导方法
- 适合追求高质量图像生成与模型调试的研究者
高分辨率图像生成中，扩散模型常因能量不稳和引导伪影导致视觉质量下降。我们分析了采样过程中的潜在能量景观，提出自适应无分类器引导（CFG）调度策略，保持能量轨迹稳定。该方法引入能量感知调度机制，随时间动态调节引导强度，在DPM++ 2M上采用线性递减的CFG调度实现最优性能，显著提升图像清晰度与真实性，同时减少伪影。实验显示，本方法在稳定度（0.9998）和一致性（0.9873）指标上均优于固定引导方案。所提出的能量分析框架可作为理解与改进扩散模型行为的强大诊断工具。
原文摘要 · Abstract (English)
High-resolution image synthesis with diffusion models often suffers from energy instabilities and guidance artifacts that degrade visual quality. We analyze the latent energy landscape during sampling and propose adaptive classifier-free guidance (CFG) schedules that maintain stable energy trajectories. Our approach introduces energy-aware scheduling strategies that modulate guidance strength over time, achieving superior stability scores (0.9998) and consistency metrics (0.9873) compared to fixed-guidance approaches. We demonstrate that DPM++ 2M with linear-decreasing CFG scheduling yields optimal performance, providing sharper, more faithful images while reducing artifacts. Our energy profiling framework serves as a powerful diagnostic tool for understanding and improving diffusion model behavior.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。