What's Inside Your Diffusion Model? A Score-Based Riemannian Metric to Explore the Data Manifold

• URL: http://arxiv.org/abs/2505.11128v2
• Date: Mon, 19 May 2025 09:31:19 GMT
• Title: What's Inside Your Diffusion Model? A Score-Based Riemannian Metric to Explore the Data Manifold
• Authors: Simone Azeglio, Arianna Di Bernardo,
• Abstract summary: We introduce a score-based Riemannian metric to characterize the intrinsic geometry of a data manifold.<n>Our approach creates a geometry where geodesics naturally follow the manifold's contours.<n>We show that our score-based geodesics capture meaningful perpendicular transformations that respect the underlying data distribution.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Recent advances in diffusion models have demonstrated their remarkable ability to capture complex image distributions, but the geometric properties of the learned data manifold remain poorly understood. We address this gap by introducing a score-based Riemannian metric that leverages the Stein score function from diffusion models to characterize the intrinsic geometry of the data manifold without requiring explicit parameterization. Our approach defines a metric tensor in the ambient space that stretches distances perpendicular to the manifold while preserving them along tangential directions, effectively creating a geometry where geodesics naturally follow the manifold's contours. We develop efficient algorithms for computing these geodesics and demonstrate their utility for both interpolation between data points and extrapolation beyond the observed data distribution. Through experiments on synthetic data with known geometry, Rotated MNIST, and complex natural images via Stable Diffusion, we show that our score-based geodesics capture meaningful transformations that respect the underlying data distribution. Our method consistently outperforms baseline approaches on perceptual metrics (LPIPS) and distribution-level metrics (FID, KID), producing smoother, more realistic image transitions. These results reveal the implicit geometric structure learned by diffusion models and provide a principled way to navigate the manifold of natural images through the lens of Riemannian geometry.

Related papers

• Follow the Energy, Find the Path: Riemannian Metrics from Energy-Based Models [35.5088111343308]
  We propose a method for deriving Riemannian metrics directly from pretrained Energy-Based Models.<n>These metrics define spatially varying distances, enabling the computation of geodesics.<n>We show that EBM-derived metrics consistently outperform established baselines.
  arXiv  Detail & Related papers  (2025-05-23T12:18:08Z)
• Learning over von Mises-Fisher Distributions via a Wasserstein-like Geometry [0.0]
  We introduce a geometry-aware distance metric for the family of von Mises-Fisher (vMF) distributions.<n>Motivated by the theory of optimal transport, we propose a Wasserstein-like distance that decomposes the discrepancy between two vMF distributions into two interpretable components.
  arXiv  Detail & Related papers  (2025-04-19T03:38:15Z)
• Score-based pullback Riemannian geometry [10.649159213723106]
  We propose a framework for data-driven Riemannian geometry that is scalable in both geometry and learning. We produce high-quality geodesics through the data support and reliably estimates the intrinsic dimension of the data manifold. Our framework can naturally be used with anisotropic normalizing flows by adopting isometry regularization during training.
  arXiv  Detail & Related papers  (2024-10-02T18:52:12Z)
• Improving embedding of graphs with missing data by soft manifolds [51.425411400683565]
  The reliability of graph embeddings depends on how much the geometry of the continuous space matches the graph structure. We introduce a new class of manifold, named soft manifold, that can solve this situation. Using soft manifold for graph embedding, we can provide continuous spaces to pursue any task in data analysis over complex datasets.
  arXiv  Detail & Related papers  (2023-11-29T12:48:33Z)
• Scaling Riemannian Diffusion Models [68.52820280448991]
  We show that our method enables us to scale to high dimensional tasks on nontrivial manifold. We model QCD densities on $SU(n)$ lattices and contrastively learned embeddings on high dimensional hyperspheres.
  arXiv  Detail & Related papers  (2023-10-30T21:27:53Z)
• A Heat Diffusion Perspective on Geodesic Preserving Dimensionality Reduction [66.21060114843202]
  We propose a more general heat kernel based manifold embedding method that we call heat geodesic embeddings. Results show that our method outperforms existing state of the art in preserving ground truth manifold distances. We also showcase our method on single cell RNA-sequencing datasets with both continuum and cluster structure.
  arXiv  Detail & Related papers  (2023-05-30T13:58:50Z)
• VTAE: Variational Transformer Autoencoder with Manifolds Learning [144.0546653941249]
  Deep generative models have demonstrated successful applications in learning non-linear data distributions through a number of latent variables. The nonlinearity of the generator implies that the latent space shows an unsatisfactory projection of the data space, which results in poor representation learning. We show that geodesics and accurate computation can substantially improve the performance of deep generative models.
  arXiv  Detail & Related papers  (2023-04-03T13:13:19Z)
• Shape And Structure Preserving Differential Privacy [70.08490462870144]
  We show how the gradient of the squared distance function offers better control over sensitivity than the Laplace mechanism. We also show how using the gradient of the squared distance function offers better control over sensitivity than the Laplace mechanism.
  arXiv  Detail & Related papers  (2022-09-21T18:14:38Z)
• Geometric Scattering on Measure Spaces [15.819230791757906]
  We introduce a general, unified model for geometric scattering on measure spaces.<n>We consider finite measure spaces that are obtained from randomly sampling an unknown manifold.<n>We propose two methods for constructing a data-driven graph on which the associated graph scattering transform approximates the scattering transform on the underlying manifold.
  arXiv  Detail & Related papers  (2022-08-17T22:40:09Z)
• GELATO: Geometrically Enriched Latent Model for Offline Reinforcement Learning [54.291331971813364]
  offline reinforcement learning approaches can be divided into proximal and uncertainty-aware methods. In this work, we demonstrate the benefit of combining the two in a latent variational model. Our proposed metrics measure both the quality of out of distribution samples as well as the discrepancy of examples in the data.
  arXiv  Detail & Related papers  (2021-02-22T19:42:40Z)
• Uniform Interpolation Constrained Geodesic Learning on Data Manifold [28.509561636926414]
  Along the learned geodesic, our method can generate high-qualitys between two given data samples. We provide a theoretical analysis of our model and use image translation as an example to demonstrate the effectiveness of our method.
  arXiv  Detail & Related papers  (2020-02-12T07:47:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.