mlsanityData to 7 Oct 2026

CogView3: Finer and Faster Text-to-Image Generation via Relay Diffusion

arXiv 2403.05121 ↗First posted 8 Mar 2024cs.CVcited by 72

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

Abstract (arXiv)

Recent advancements in text-to-image generative systems have been largely driven by diffusion models. However, single-stage text-to-image diffusion models still face challenges, in terms of computational efficiency and the refinement of image details. To tackle the issue, we propose CogView3, an innovative cascaded framework that enhances the performance of text-to-image diffusion. CogView3 is the first model implementing relay diffusion in the realm of text-to-image generation, executing the task by first creating low-resolution images and subsequently applying relay-based super-resolution. This methodology not only results in competitive text-to-image outputs but also greatly reduces both training and inference costs. Our experimental results demonstrate that CogView3 outperforms SDXL, the current state-of-the-art open-source text-to-image diffusion model, by 77.0% in human evaluations, all while requiring only about 1/2 of the inference time. The distilled variant of CogView3 achieves comparable performance while only utilizing 1/10 of the inference time by SDXL.

Key facts. Cited by 72 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 26 of its 35 references that are in the map. First posted on arXiv 8 Mar 2024.

Builds on

26 of 35 references.

Closest ancestors · 4

Same direction, newest first

  1. 2023Wuerstchen: An Efficient Architecture for Large-Scale Text-to-Image Diffusion Models1 Jun
  2. 2023MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generation16 Feb
  3. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding23 May
  4. 2022Hierarchical Text-Conditional Image Generation with CLIP Latents13 Apr

Other references · 22

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2023GPT-4 Technical Reportcited by 18,132
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  4. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  5. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  6. 2020Denoising Diffusion Implicit Modelscited by 9,655
  7. 2022Classifier-Free Diffusion Guidancecited by 5,717
  8. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,290
  9. 2021Zero-Shot Text-to-Image Generationcited by 4,276
  10. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  11. 2022Scaling Autoregressive Models for Content-Rich Text-to-Image Generationcited by 1,287
  12. 2021Cascaded Diffusion Models for High Fidelity Image Generationcited by 1,155
10 more
  1. 2023Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesiscited by 847
  2. 2023Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inferencecited by 826
  3. 2023Adversarial Diffusion Distillationcited by 741
  4. 2022On Distillation of Guided Diffusion Modelscited by 721
  5. 2023CogVLM: Visual Expert for Pretrained Language Modelscited by 704
  6. 2021CogView: Mastering Text-to-Image Generation via Transformerscited by 691
  7. 2023Scaling up GANs for Text-to-Image Synthesiscited by 492
  8. 2023Emu: Enhancing Image Generation Models Using Photogenic Needles in a Haystackcited by 278
  9. 2023Relay Diffusion: Unifying diffusion process across resolutions for image synthesiscited by 78
  10. 2023Mixture of Diffusers for scene composition and high resolution image generationcited by 56

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 24; 2025: 28; 2026: 20.

72 papers.

Sort:

By area:

  1. 2024CogVideoX: Text-to-Video Diffusion Models with An Expert Transformercited by 2,198
  2. 2024ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Toolscited by 1,096
  3. 2025Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformercited by 247
  4. 2024Lumina-T2X: Transforming Text into Any Modality, Resolution, and Duration via Flow-based Large Diffusion Transformerscited by 137
  5. 2024Simpler Diffusion (SiD2): 1.5 FID on ImageNet512 with pixel-space diffusioncited by 74
  6. 2024ControlVAR: Exploring Controllable Visual Autoregressive Modelingcited by 58
  7. 2024VideoTetris: Towards Compositional Text-to-Video Generationcited by 51
  8. 2025Pixel-Perfect Depth with Semantics-Prompted Diffusion Transformerscited by 47
  9. 2024Diffusion Feedback Helps CLIP See Bettercited by 37
  10. 2025FlashVideo: Flowing Fidelity to Detail for Efficient High-Resolution Video Generationcited by 32
  11. 2025FlowTok: Flowing Seamlessly Across Text and Image Tokenscited by 31
  12. 2024Dimba: Transformer-Mamba Diffusion Modelscited by 26
38 more
  1. 2024ResMaster: Mastering High-Resolution Image Generation via Structural and Fine-Grained Guidancecited by 21
  2. 2026AutoFigure: Generating and Refining Publication-Ready Scientific Illustrationscited by 20
  3. 2024Improving Long-Text Alignment for Text-to-Image Diffusion Modelscited by 18
  4. 2024Owl-1: Omni World Model for Consistent Long Video Generationcited by 17
  5. 2024Self-Cross Diffusion Guidance for Text-to-Image Synthesis of Similar Subjectscited by 16
  6. 2025Enhancing Intent Understanding for Ambiguous prompt: A Human-Machine Co-Adaption Strategycited by 16
  7. 2025MMIG-Bench: Towards Comprehensive and Explainable Evaluation of Multi-Modal Image Generation Modelscited by 16
  8. 2025T2I-RiskyPrompt: A Benchmark for Safety Evaluation, Attack, and Defense on Text-to-Image Modelcited by 16
  9. 2024Is One GPU Enough? Pushing Image Generation at Higher-Resolutions with Foundation Modelscited by 14
  10. 2025MMMG: A Massive, Multidisciplinary, Multi-Tier Generation Benchmark for Text-to-Image Reasoningcited by 14
  11. 2025Audio Super-Resolution with Latent Bridge Modelscited by 13
  12. 2025Ultra-Resolution Adaptation with Easecited by 12
  13. 2025Patch Matters: Training-free Fine-grained Image Caption Enhancement via Local Perceptioncited by 12
  14. 2024PrimeDepth: Efficient Monocular Depth Estimation with a Stable Diffusion Preimagecited by 11
  15. 2025Twin Co-Adaptive Dialogue for Progressive Image Generationcited by 11
  16. 2024Slight Corruption in Pre-training Data Makes Better Diffusion Modelscited by 9
  17. 2024Text-to-Image Synthesis: A Decade Surveycited by 9
  18. 2025Red-Teaming Text-to-Image Systems by Rule-based Preference Modelingcited by 8
  19. 2025MIRAGE: Towards AI-Generated Image Detection in the Wildcited by 8
  20. 2026PosterVerse: A Full-Workflow Framework for Commercial-Grade Poster Generation with HTML-Based Scalable Typographycited by 8
  21. 2026Leveraging Verifier-Based Reinforcement Learning in Image Editingcited by 8
  22. 2025Towards Self-Improvement of Diffusion Models via Group Preference Optimizationcited by 7
  23. 2024TIPO: Text to Image with Text Presampling for Prompt Optimizationcited by 6
  24. 2025UTDesign: A Unified Framework for Stylized Text Editing and Generation in Graphic Design Imagescited by 6
  25. 2026WeEdit: A Dataset, Benchmark and Glyph-Guided Framework for Text-centric Image Editingcited by 6
  26. 2024Ethical-Lens: Curbing Malicious Usages of Open-Source Text-to-Image Modelscited by 4
  27. 2025DreamPainter: Image Background Inpainting for E-commerce Scenarioscited by 4
  28. 2026BitDance: Scaling Autoregressive Generative Models with Binary Tokenscited by 4
  29. 2026The Unreasonable Effectiveness of Text Embedding Interpolation for Continuous Image Steeringcited by 4
  30. 2024From Noise to Nuance: Advances in Deep Generative Image Modelscited by 3
  31. 2025Omni-Dish: Photorealistic and Faithful Image Generation and Editing for Arbitrary Chinese Dishescited by 3
  32. 2025Multimodal Benchmarking and Recommendation of Text-to-Image Generation Modelscited by 3
  33. 2025FlagEvalMM: A Flexible Framework for Comprehensive Multimodal Model Evaluationcited by 3
  34. 2024E4C: Enhance Editability for Text-Based Image Editing by Harnessing Efficient CLIP Guidancecited by 2
  35. 2024Tex4D: Zero-shot 4D Scene Texturing with Video Diffusion Modelscited by 2
  36. 2024Artificial Intelligence for Biomedical Video Generationcited by 2
  37. 2025Scale-Wise VAR is Secretly Discrete Diffusioncited by 2
  38. 2025TechImage-Bench: Rubric-Based Evaluation for Technical Image Generationcited by 2

50 most cited of 72 All 72 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2022Shifted Diffusion for Text-to-image Generation
  2. 2023PanGu-Draw: Advancing Resource-Efficient Text-to-Image Synthesis with Time-Decoupled Training and Reusable Coop-Diffusion
  3. 2025Reusing Computation in Text-to-Image Diffusion for Efficient Generation of Image Sets
  4. 2025Muddit: Liberating Generation Beyond Text-to-Image with a Unified Discrete Diffusion Model
  5. 2024Controllable Generation with Text-to-Image Diffusion Models: A Survey
  6. 2023Text-to-image Diffusion Models in Generative AI: A Survey
  7. 2026i1: A Simple and Fully Open Recipe for Strong Text-to-Image Models
  8. 2022Implementing and Experimenting with Diffusion Models for Text-to-Image Generation