Showing 2025Show all
2 papers · 1 filter
cs.CV2025
Your Latent Mask is Wrong: Pixel-Equivalent Latent Compositing for Diffusion Models
Rowan Bradbury, Dazhi Zhong
Latent inpainting in diffusion models still relies almost universally on linearly interpolating VAE latents under a downsampled mask. We propose a key principle for compositing ima…
cs.LG2025
Deterministic Continuous Replacement: Fast and Stable Module Replacement in Pretrained Transformers
Rowan Bradbury, Aniket Srinivasan Ashok, Sai Ram Kasanagottu +2
Replacing modules in pretrained models, especially swapping quadratic self-attention for efficient attention alternatives, poses a hard optimization problem: cold-start reinitializ…