Residualized Temporal Sparse Autoencoders for Interpreting Diffusion Models 事件

Residualized Temporal Sparse Autoencoders for Interpreting Diffusion Models arXiv:2605.27813v1 Announce Type: new Abstract: Text-to-image diffusion models generate images through an iterative denoising process, so internal neural layers produce trajectories of activations rather than single static representations. Sparse autoencoders (SAEs) have recently been used to decompose diffusion activations into interpretable feature directions, but most approaches analyze activations at individual time