Understanding Latent Diffusability via Fisher Geometry

arXiv:2604.02751 · cs.LG · Submitted 2026-04-03 · Read on arXiv

cs.LG

Submitted: 2026-04-03

Updated: 2026-10-01

Code: https://github.com/tanelp/tiny-diffusion

Terminology

• Fisher Information — 2× in this paper · explained in 3 episodes
• Alignment — 1× in this paper · explained in 1 episode
• Diffusion Model — 1× in this paper · explained in 9 episodes
• Diffusion Models — 1× in this paper · explained in 6 episodes
• Fisher Information Rate — 1× in this paper
• Latent Space — 1× in this paper · explained in 7 episodes
• Minimum Mean Squared Error — 1× in this paper

Sources

• Latent Forcing: Reordering the Diffusion Trajectory for Pixel-Space Image Generation
• Unified Latents (UL): How to train your latents
• Enhancing Variational Autoencoders with Smooth Robust Latent Encoding
• The Intrinsic Dimension of Images and Its Impact on Learning
• Latent Diffusion Model without Variational Autoencoder
• Improved Baselines with Representation Autoencoders
• Diffusion Transformers with Representation Autoencoders
• Diffusing in the Right Space: A Systematic Study of Latent Diffusability

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks