paper-with-me

홈 › Papers

RectifiedHR: High-Resolution Diffusion via Energy Profiling and Adaptive Guidance Scheduling

2025-07-13 · Ankit Sanjyal arxiv

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.

📄 PDF Abstract BibTeX arXiv:2507.09441

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

RectifiedHR: Enable Efficient High-Resolution Image Generation via Energy Rectification

2025-03-04 · Zhen Yang, Guibao Shen, Liang Hou, Mushui Liu 외

Diffusion models have achieved remarkable advances in various image generation tasks. However, their performance notably declines when generating images at resolutions higher than those used during the training period. D…

Image Generation

EnergyDiff: Universal Time-Series Energy Data Generation using Diffusion Models

2024-07-18 · Nan Lin, Peter Palensky, Pedro P. Vergara

High-resolution time series data are crucial for the operation and planning of energy systems such as electrical power systems and heating systems. Such data often cannot be shared due to privacy concerns, necessitating …

DenoisingTime Series

PlatformX: An End-to-End Transferable Platform for Energy-Efficient Neural Architecture Search

2025-10-10 · Xiaolong Tu, Dawei Chen, Kyungtae Han, Onur Altintas 외 arxiv

Hardware-Aware Neural Architecture Search (HW-NAS) has emerged as a powerful tool for designing efficient deep neural networks (DNNs) tailored to edge devices. However, existing methods remain largely impractical for rea…

Neural Architecture Search

Measurement of Generative AI Workload Power Profiles for Whole-Facility Data Center Infrastructure Planning

2026-04-08 · Roberto Vercellino, Jared Willard, Gustavo Campos, Weslley da Silva Pereira 외 arxiv

The rapid growth of generative artificial intelligence (AI) has introduced unprecedented computational demands, driving significant increases in the energy footprint of data centers. However, existing power consumption d…

Comparative Study of Hardware and Software Power Measurements in Video Compression

2023-12-19 · Angeliki Katsenou, Xinyi Wang, Daniel Schien, David Bull

The environmental impact of video streaming services has been discussed as part of the strategies towards sustainable information and communication technologies. A first step towards that is the energy profiling and asse…

Video Compression