RectifiedHR: High-Resolution Diffusion via Energy Profiling and Adaptive Guidance Scheduling
This work addresses visual quality degradation in high-resolution image generation for AI and computer vision applications, presenting an incremental improvement through adaptive scheduling.
The paper tackled the problem of energy instabilities and guidance artifacts in high-resolution image synthesis with diffusion models, achieving superior stability scores of 0.9998 and consistency metrics of 0.9873 compared to fixed-guidance approaches.
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.