mlsanityData to 7 Oct 2026

Diffusion Models already have a Semantic Latent Space

arXiv 2210.10960 ↗First posted 20 Oct 2022cs.CVcited by 324

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

Diffusion models achieve outstanding generative performance in various domains. Despite their great success, they lack semantic latent space which is essential for controlling the generative process. To address the problem, we propose asymmetric reverse process (Asyrp) which discovers the semantic latent space in frozen pretrained diffusion models. Our semantic latent space, named h-space, has nice properties for accommodating semantic image manipulation: homogeneity, linearity, robustness, and consistency across timesteps. In addition, we introduce a principled design of the generative process for versatile editing and quality boost ing by quantifiable measures: editing strength of an interval and quality deficiency at a timestep. Our method is applicable to various architectures (DDPM++, iD- DPM, and ADM) and datasets (CelebA-HQ, AFHQ-dog, LSUN-church, LSUN- bedroom, and METFACES). Project page: https://kwonminki.github.io/Asyrp/

Key facts. Cited by 324 papers in the map (data through 7 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 37 of its 48 references that are in the map. First posted on arXiv 20 Oct 2022.

Builds on

37 of 48 references.

Closest ancestors · 13

Same direction, newest first

  1. 2022Elucidating the Design Space of Diffusion-Based Generative Models1 Jun
  2. 2022Perception Prioritized Training of Diffusion Models1 Apr
  3. 2022Generating High Fidelity Data from Low-density Regions using Diffusion Models31 Mar
  4. 2022Learning Fast Samplers for Diffusion Models by Differentiating Through Sample Quality11 Feb
  5. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  6. 2021Diffusion Autoencoders: Toward a Meaningful and Decodable Representation30 Nov
  7. 2021ILVR: Conditioning Method for Denoising Diffusion Probabilistic Models6 Aug
  8. 2021Score-based Generative Modeling in Latent Space10 Jun
  9. 2021Diffusion Models Beat GANs on Image Synthesis11 May
  10. 2021Improved Denoising Diffusion Probabilistic Models18 Feb
  11. 2020Score-Based Generative Modeling through Stochastic Differential Equations26 Nov
  12. 2020Denoising Diffusion Implicit Models6 Oct
1 more
  1. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 24

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  3. 2022Generative Adversarial Networkscited by 5,603
  4. 2013Auto-Encoding Variational Bayescited by 3,980
  5. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  6. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,595
  7. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
  8. 2022RePaint: Inpainting using Denoising Diffusion Probabilistic Modelscited by 1,419
  9. 2015LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loopcited by 1,350
  10. 2020Training Generative Adversarial Networks with Limited Datacited by 1,184
  11. 2019StarGAN v2: Diverse Image Synthesis for Multiple Domainscited by 1,095
  12. 2016Semantic Understanding of Scenes through the ADE20K Datasetcited by 1,063
12 more
  1. 2021StyleCLIP: Text-Driven Manipulation of StyleGAN Imagerycited by 990
  2. 2021Blended Diffusion for Text-driven Editing of Natural Imagescited by 924
  3. 2018BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentationcited by 645
  4. 2019MaskGAN: Towards Diverse and Interactive Facial Image Manipulationcited by 624
  5. 2020GANSpace: Discovering Interpretable GAN Controlscited by 612
  6. 2020InterFaceGAN: Interpreting the Disentangled Face Representation Learned by GANscited by 393
  7. 2021More Control for Free! Image Synthesis with Semantic Diffusion Guidancecited by 211
  8. 2021EditGAN: High-Precision Semantic Image Editingcited by 156
  9. 2019Style Transformer: Unpaired Text Style Transfer without Disentangled Latent Representationcited by 84
  10. 2021LatentCLR: A Contrastive Learning Approach for Unsupervised Discovery of Interpretable Directionscited by 52
  11. 2021Image-Based CLIP-Guided Essence Transfercited by 39
  12. 2021Predict, Prevent, and Evaluate: Disentangled Text-Driven Image Manipulation Empowered by Pre-Trained Vision-Language Modelcited by 34

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 8; 2023: 73; 2024: 110; 2025: 82; 2026: 51.

324 papers.

Sort:

By area:

  1. 2023MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generationcited by 529
  2. 2023Pix2Video: Video Editing using Image Diffusioncited by 318
  3. 2023Diffusion Self-Guidance for Controllable Image Generationcited by 305
  4. 2024MACE: Mass Concept Erasure in Diffusion Modelscited by 298
  5. 2023Text-to-image Diffusion Models in Generative AI: A Surveycited by 264
  6. 2023Guided Motion Diffusion for Controllable Human Motion Synthesiscited by 221
  7. 2022SpaText: Spatio-Textual Representation for Controllable Image Generationcited by 217
  8. 2024FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Modelscited by 214
  9. 2024Diffusion Model-Based Image Editing: A Surveycited by 212
  10. 2023Style Injection in Diffusion: A Training-free Approach for Adapting Large-scale Diffusion Models for Style Transfercited by 194
  11. 2023TF-ICON: Diffusion-Based Training-Free Cross-Domain Image Compositioncited by 182
  12. 2022SINE: SINgle Image Editing with Text-to-Image Diffusion Modelscited by 159
38 more
  1. 2023Concept Sliders: LoRA Adaptors for Precise Control in Diffusion Modelscited by 148
  2. 2023Let 2D Diffusion Model Know 3D-Consistency for Robust Text-to-3D Generationcited by 133
  3. 2023A Neural Space-Time Representation for Text-to-Image Personalizationcited by 123
  4. 2023Understanding the Latent Space of Diffusion Models through the Lens of Riemannian Geometrycited by 121
  5. 2022Uncovering the Disentanglement Capability in Text-to-Image Diffusion Modelscited by 116
  6. 2022Disentangled Representation Learningcited by 111
  7. 2023Diffusion Model as Representation Learnercited by 95
  8. 2023AVID: Any-Length Video Inpainting with Diffusion Modelcited by 81
  9. 2023Divide & Bind Your Attention for Improved Generative Semantic Nursingcited by 80
  10. 2025SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoderscited by 75
  11. 2023Restoration-Degradation Beyond Linear Diffusions: A Non-Asymptotic Analysis For DDIM-Type Samplerscited by 73
  12. 2023Self-Discovering Interpretable Diffusion Latent Directions for Responsible Text-to-Image Generationcited by 69
  13. 2023StyleDiffusion: Prompt-Embedding Inversion for Text-Based Editingcited by 68
  14. 2023LooseControl: Lifting ControlNet for Generalized Depth Conditioningcited by 68
  15. 2024Diffusion Models, Image Super-Resolution And Everything: A Surveycited by 63
  16. 2024Visual Style Prompting with Swapping Self-Attentioncited by 62
  17. 2025Set You Straight: Auto-Steering Denoising Trajectories to Sidestep Unwanted Conceptscited by 61
  18. 2023Discovering Interpretable Directions in the Semantic Latent Space of Diffusion Modelscited by 60
  19. 2023Concept Algebra for (Score-Based) Text-Controlled Generative Modelscited by 59
  20. 2023On the Design Fundamentals of Diffusion Models: A Surveycited by 59
  21. 2024Zero-Shot Unsupervised and Text-Based Audio Editing Using DDPM Inversioncited by 58
  22. 2024Smoothed Energy Guidance: Guiding Diffusion Models with Reduced Energy Curvature of Attentioncited by 55
  23. 2022Unsupervised Representation Learning from Pre-trained Diffusion Probabilistic Modelscited by 53
  24. 2024Diffusion Models and Representation Learning: A Surveycited by 53
  25. 2024Explainable and Interpretable Multimodal Large Language Models: A Comprehensive Surveycited by 52
  26. 2024On the Trajectory Regularity of ODE-based Diffusion Samplingcited by 51
  27. 2024Balancing Act: Distribution-Guided Debiasing in Diffusion Modelscited by 50
  28. 2023NoiseCLR: A Contrastive Learning Approach for Unsupervised Discovery of Interpretable Directions in Diffusion Modelscited by 46
  29. 2024Continuous, Subject-Specific Attribute Control in T2I Models by Identifying Semantic Directionscited by 39
  30. 2024Efficient Diffusion Models: A Comprehensive Survey from Principles to Practicescited by 39
  31. 2024Schedule Your Edit: A Simple yet Effective Diffusion Noise Schedule for Image Editingcited by 38
  32. 2023Training-free Content Injection using h-space in Diffusion Modelscited by 36
  33. 2024Exploring Diffusion Time-steps for Unsupervised Representation Learningcited by 36
  34. 2024Interpreting the Weight Space of Customized Diffusion Modelscited by 35
  35. 2023Faster Diffusion: Rethinking the Role of the Encoder for Diffusion Model Inferencecited by 33
  36. 2023All but One: Surgical Concept Erasing with Model Preservation in Text-to-Image Diffusion Modelscited by 33
  37. 2024Emergence of Hidden Capabilities: Exploring Learning Dynamics in Concept Spacecited by 33
  38. 2023Latent Space Editing in Transformer-Based Flow Matchingcited by 32

50 most cited of 324 All 324 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2023Non-Cross Diffusion for Semantic Consistency
  2. 2022The Stable Artist: Steering Semantics in Diffusion Latent Space
  3. 2023Smooth Diffusion: Crafting Smooth Latent Spaces in Diffusion Models
  4. 2022Unifying Diffusion Models' Latent Space, with Applications to CycleDiffusion and Guidance
  5. 2022Let us Build Bridges: Understanding and Extending Diffusion Generative Models
  6. 2025Exploring Representation-Aligned Latent Space for Better Generation
  7. 2024Contextualized Diffusion Models for Text-Guided Image and Video Generation
  8. 2026Diffusing in the Right Space: A Systematic Study of Latent Diffusability