When Geometry Aligns: Dihedral Hidden-State Transformations in UNet, ViT, and DiT Architectures
This paper explores the effects of geometric perturbations on the stability of diffusion architectures, specifically Diffusion Transformers and Convolutional UNets, using a unified framework and controlled internal interventions.
Explored the impact of geometric perturbations on the stability of diffusion architectures using a unified framework and controlled interventions
Before reading this…
Applications
- →Vision models
- →Diffusion models
To understand this paper, make sure you know these concepts first:
- Deep learning basicsfind papers →
- Understanding of diffusion architecturesfind papers →
Abstract
More Like ThisDiffusion architectures now encompass convolutional UNets as well as transformer-based designs such as Diffusion Transformers (DiTs), inspired by Vision Transformers (ViTs), yet the effects of structured geometric perturbations within these architectures remain poorly understood. We study this question through a unified framework that applies reflection-based elements of the dihedral group to intermediate hidden states as controlled internal interventions, contrasting geometrically consistent and inconsistent variants. Using activation-level diagnostics, including Self-Consistency Shift (SCS), Activation Mass Scatter (AMS), and Drift, we analyze feature stability and geometric drift. We find that consistent transformations improve stability, while inconsistent ones induce predictable, architecture-specific failures. In the main Stable Diffusion 2.1 U-Net study, we evaluate seven intervention modes over three seeds and complement the internal diagnostics with image-level FID, KID, CLIP score, and LPIPS diversity. Taken together with supporting ViT and controlled DiT analyses, these results establish geometric consistency as a key principle for stable hidden-state interventions in spatially structured vision and diffusion models.