Layer Skip: Enabling Early Exit Inference and Self-Speculative Decoding Paper • 2404.16710 • Published 18 days ago • 54 • 8
Cross-Attention Makes Inference Cumbersome in Text-to-Image Diffusion Models Paper • 2404.02747 • Published Apr 3 • 11 • 1
Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction Paper • 2404.02905 • Published Apr 3 • 59 • 2
The Unreasonable Ineffectiveness of the Deeper Layers Paper • 2403.17887 • Published Mar 26 • 74 • 12