mlsanityData to 7 Oct 2026

SVG-T2I: Scaling Up Text-to-Image Latent Diffusion Model Without Variational Autoencoder

arXiv 2512.11749 ↗First posted 12 Dec 2025cs.CVcited by 12

On the map: Diffusion models → Image generation, text-to-image and editing

Abstract (arXiv)

Visual generation grounded in Visual Foundation Model (VFM) representations offers a highly promising unified pathway for integrating visual understanding, perception, and generation. Despite this potential, training large-scale text-to-image diffusion models entirely within the VFM representation space remains largely unexplored. To bridge this gap, we scale the SVG (Self-supervised representations for Visual Generation) framework, proposing SVG-T2I to support high-quality text-to-image synthesis directly in the VFM feature domain. By leveraging a standard text-to-image diffusion pipeline, SVG-T2I achieves competitive performance, reaching 0.75 on GenEval and 85.78 on DPG-Bench. This performance validates the intrinsic representational power of VFMs for generative tasks. We fully open-source the project, including the autoencoder and generation model, together with their training, inference, evaluation pipelines, and pre-trained weights, to facilitate further research in representation-driven visual generation.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models. Builds on 38 of its 53 references that are in the map. First posted on arXiv 12 Dec 2025.

Builds on

38 of 53 references.

Closest ancestors · 3

Same direction, newest first

  1. 2024Hunyuan-DiT: A Powerful Multi-Resolution Diffusion Transformer with Fine-Grained Chinese Understanding14 May
  2. 2024Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction3 Apr
  3. 2023GenEval: An Object-Focused Framework for Evaluating Text-to-Image Alignment17 Oct

Other references · 35

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2020Denoising Diffusion Implicit Modelscited by 9,655
  5. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  6. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  7. 2022Scalable Diffusion Models with Transformerscited by 5,931
  8. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,094
  9. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
  10. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,550
  11. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
  12. 2017Improved Training of Wasserstein GANscited by 2,223
23 more
  1. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,048
  2. 2022MaskGIT: Masked Generative Image Transformercited by 1,082
  3. 2023PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesiscited by 854
  4. 2024Autoregressive Model Beats Diffusion: Llama for Scalable Image Generationcited by 760
  5. 2024SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformerscited by 707
  6. 2024Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Thinkcited by 666
  7. 2024Autoregressive Image Generation without Vector Quantizationcited by 644
  8. 2021Score-based Generative Modeling in Latent Spacecited by 634
  9. 2022StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasetscited by 489
  10. 2021Diffusion Autoencoders: Toward a Meaningful and Decodable Representationcited by 447
  11. 2025Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Modelscited by 376
  12. 2025Diffusion Transformers with Representation Autoencoderscited by 275
  13. 2025Transfer between Modalities with MetaQueriescited by 231
  14. 2025REPA-E: Unlocking VAE for End-to-End Tuning with Latent Diffusion Transformerscited by 212
  15. 2023Fast Training of Diffusion Models with Masked Transformerscited by 172
  16. 2025HiDream-I1: A High-Efficient Image Generative Foundation Model with Sparse Diffusion Transformercited by 131
  17. 2023Return of Unconditional Generation: A Self-supervised Representation Generation Methodcited by 95
  18. 2025Representation Entanglement for Generation: Training Diffusion Transformers Is Much Easier Than You Thinkcited by 95
  19. 2025EQ-VAE: Equivariance Regularized Latent Space for Improved Generative Image Modelingcited by 85
  20. 2025Improving the Diffusability of Autoencoderscited by 74
  21. 2018Spherical Latent Spaces for Stable Variational Autoencoderscited by 66
  22. 2025UniLiP: Adapting CLIP for Unified Multimodal Understanding, Generation and Editingcited by 43
  23. 2025Boosting Generative Image Modeling via Joint Image-Feature Synthesiscited by 42

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 11.

12 papers.

  1. 2026Scaling Text-to-Image Diffusion Transformers with Representation Autoencoderscited by 49
  2. 2026Beyond Language Modeling: An Exploration of Multimodal Pretrainingcited by 32
  3. 2026EponaV2: Driving World Model with Comprehensive Future Reasoningcited by 12
  4. 2025RecTok: Reconstruction Distillation along Rectified Flowcited by 4
  5. 2026Research on World Models Is Not Merely Injecting World Knowledge into Specific Taskscited by 3
  6. 2026Omni123: Exploring 3D Native Foundation Models with Limited 3D Data by Unifying Text to 2D and 3D Generationcited by 3
  7. 2026Steering Visual Generation in Unified Multimodal Models with Understanding Supervisioncited by 3
  8. 2026RAE-AR: Taming Autoregressive Models with Representation Autoencoderscited by 1
  9. 2026Cohort-Scale Neural Atlases of Ultrasound Videocited by 0
  10. 2026SeFi-Image: A Text-to-Image Foundation Model with Semantic-First Diffusioncited by 0
  11. 2026WALT: Learning World-Model-Aligned Latent Trajectories for Autonomous Drivingcited by 0
  12. 2026Seeing the Heat: Synthesizing High-Resolution Wood Thermal Responses from Optical Imagerycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Latent Diffusion Model without Variational Autoencoder
  2. 2022VectorFusion: Text-to-SVG by Abstracting Pixel-Based Diffusion Models
  3. 2024SVGFusion: A VAE-Diffusion Transformer for Vector Graphic Generation
  4. 2024Chat2SVG: Vector Graphics Generation with Large Language Models and Image Diffusion Models
  5. 2024MaxFusion: Plug&Play Multi-Modal Generation in Text-to-Image Diffusion Models
  6. 2025Style Customization of Text-to-Vector Generation with Image Diffusion Priors
  7. 2023VGDiffZero: Text-to-image Diffusion Models Can Be Zero-shot Visual Grounders
  8. 2026High-Fidelity Text-to-Image Generation from Pre-Trained Vision-Language Models via Distribution-Conditioned Diffusion Decoding