mlsanityData to 7 Oct 2026

Isometric Representation Learning for Disentangled Latent Space of Diffusion Models

arXiv 2407.11451 ↗First posted 16 Jul 2024cs.LG cs.CVcited by 22

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

The latent space of diffusion model mostly still remains unexplored, despite its great success and potential in the field of generative modeling. In fact, the latent space of existing diffusion models are entangled, with a distorted mapping from its latent space to image space. To tackle this problem, we present Isometric Diffusion, equipping a diffusion model with a geometric regularizer to guide the model to learn a geometrically sound latent space of the training data manifold. This approach allows diffusion models to learn a more disentangled latent space, which enables smoother interpolation, more accurate inversion, and more precise control over attributes directly in the latent space. Our extensive experiments consisting of image interpolations, image inversions, and linear editing show the effectiveness of our method.

Key facts. Cited by 22 papers in the map (data through 7 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 38 of its 62 references that are in the map. First posted on arXiv 16 Jul 2024.

Builds on

38 of 62 references.

Closest ancestors · 9

Same direction, newest first

  1. 2023Understanding the Latent Space of Diffusion Models through the Lens of Riemannian Geometry24 Jul
  2. 2022Scalable Diffusion Models with Transformers19 Dec
  3. 2022Diffusion Models already have a Semantic Latent Space20 Oct
  4. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  5. 2021Diffusion Autoencoders: Toward a Meaningful and Decodable Representation30 Nov
  6. 2021Diffusion Models Beat GANs on Image Synthesis11 May
  7. 2020Score-Based Generative Modeling through Stochastic Differential Equations26 Nov
  8. 2020Denoising Diffusion Implicit Models6 Oct
  9. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 29

  1. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  2. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  3. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  4. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  5. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  6. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  7. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,595
  8. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
  9. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,286
  10. 2022Prompt-to-Prompt Image Editing with Cross Attention Controlcited by 2,409
  11. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,048
  12. 2022Imagen Video: High Definition Video Generation with Diffusion Modelscited by 1,765
17 more
  1. 2023Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Modelscited by 1,465
  2. 2022Imagic: Text-Based Real Image Editing with Diffusion Modelscited by 1,172
  3. 2016InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Netscited by 858
  4. 2019Interpreting the Latent Space of GANs for Semantic Face Editingcited by 691
  5. 2018Disentangling by Factorisingcited by 614
  6. 2020GANSpace: Discovering Interpretable GAN Controlscited by 612
  7. 2020Closed-Form Factorization of Latent Semantics in GANscited by 407
  8. 2020InterFaceGAN: Interpreting the Disentangled Face Representation Learned by GANscited by 393
  9. 2020StyleFlow: Attribute-conditioned Exploration of StyleGAN-Generated Images using Conditional Continuous Normalizing Flowscited by 348
  10. 2017Latent Space Oddity: on the Curvature of Deep Generative Modelscited by 169
  11. 2018IntroVAE: Introspective Variational Autoencoders for Photographic Image Synthesiscited by 85
  12. 2020Learning identifiable and interpretable latent models of high-dimensional neural activity using pi-VAEcited by 48
  13. 2020Learning Flat Latent Manifolds with VAEscited by 33
  14. 2021Do Not Escape From the Manifold: Discovering the Local Coordinates on the Latent Space of GANscited by 26
  15. 2017GLSR-VAE: Geodesic Latent Space Regularization for Variational AutoEncoder Architecturescited by 22
  16. 2018Disentangling Latent Space for VAE by Label Relevant/Irrelevant Dimensionscited by 19
  17. 2016Knowledge Guided Disambiguation for Large-Scale Scene Classification with Multi-Resolution CNNscited by 16

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 2; 2025: 16; 2026: 4.

22 papers.

Sort:

By area:

  1. 2025SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoderscited by 75
  2. 2025TIDE : Temporal-Aware Sparse Autoencoders for Interpretable Diffusion Transformers in Image Generationcited by 16
  3. 2025Latent Diffusion Models with Masked AutoEncoderscited by 11
  4. 2025SAEmnesia: Erasing Concepts in Diffusion Models with Supervised Sparse Autoencoderscited by 8
  5. 2025Latent Wavelet Diffusion For Ultra-High-Resolution Image Synthesiscited by 6
  6. 2025SummDiff: Generative Modeling of Video Summarization with Diffusioncited by 4
  7. 2025Is There a Better Source Distribution than Gaussian? Exploring Source Distributions for Image Flow Matchingcited by 4
  8. 2024Latent Space Disentanglement in Diffusion Transformers Enables Precise Zero-shot Semantic Editingcited by 3
  9. 2025CLAD: Constrained Latent Action Diffusion for Vision-Language Procedure Planningcited by 2
  10. 2024Latent Space Disentanglement in Diffusion Transformers Enables Zero-shot Fine-grained Semantic Editingcited by 1
  11. 2025SC-Pro: Training-Free Framework for Defending Unsafe Image Synthesis Attackcited by 1
  12. 2025SPF-Portrait: Towards Pure Text-to-Portrait Customization with Semantic Pollution-Free Fine-Tuningcited by 1
10 more
  1. 2025Latent Manifold Reconstruction and Representation with Topological and Geometrical Regularizationcited by 1
  2. 2025Be Tangential to Manifold: Discovering Riemannian Metric for Diffusion Modelscited by 1
  3. 2025Contrastive Diffusion Alignment: Learning Structured Latents for Controllable Generationcited by 1
  4. 2026Discovering Data Manifold Geometry through Geometric Propertiescited by 1
  5. 2025Expressive Score-Based Priors for Distribution Matching with Geometry-Preserving Regularizationcited by 0
  6. 2025Style Composition within Distinct LoRA modules for Traditional Artcited by 0
  7. 2025FLUID: Training-Free Face De-identification via Latent Identity Substitutioncited by 0
  8. 2026Equivariant Latent Alignment via Flow Matching under Group Symmetriescited by 0
  9. 2026From Noise to Control: Parameterized Diffusion Policiescited by 0
  10. 2026Interpreting Object-Dependent Concept Brittleness in Text-to-Image Diffusion Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Disentanglement in T-space for Faster and Distributed Training of Diffusion Models with Fewer Latent-states
  2. 2026What Matters for Diffusion-Friendly Latent Manifold? Prior-Aligned Autoencoders for Latent Diffusion
  3. 2022Image Embedding for Denoising Generative Models
  4. 2025REGLUE Your Latents with Global and Local Semantics for Entangled Diffusion
  5. 2026Intrinsic-Hybrid Latent Diffusion Models for Generative Modeling on Unknown Manifolds
  6. 2024Breaking the Curse of Dimensionality: Diffusion Models Efficiently Learn Low-Dimensional Distributions
  7. 2022Unifying Diffusion Models' Latent Space, with Applications to CycleDiffusion and Guidance
  8. 2026Boosting Latent Diffusion Models via Disentangled Representation Alignment