ToDo: Token Downsampling for Efficient Generation of High-Resolution Images

Ethan Smith, Nayan Saxena, Aninda Saha

ToDo: Token Downsampling for Efficient Generation of High-Resolution Images: 10 upvotes on Hugging Face Daily Papers, #8 of 12 papers on 2024-02-22. Day-by-day upvote history.

Attention mechanism has been crucial for image diffusion models, however, their quadratic computational complexity limits the sizes of images we can process within reasonable time and memory constraints. This paper investigates the importance of dense attention in generative image models, which often contain redundant features, making them suitable for sparser attention mechanisms. We propose a novel training-free method ToDo that relies on token downsampling of key and value tokens to accelerate Stable Diffusion inference by up to 2x for common sizes and up to 4.5x or more for high resolutions like 2048x2048. We demonstrate that our approach outperforms previous methods in balancing efficient throughput and fidelity.

Paper page on Hugging Face · arXiv

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.