mlsanityData to 14 Oct 2026

How far can we go with ImageNet for Text-to-Image generation?

arXiv 2502.21318 ↗First posted 28 Feb 2025cs.CVcited by 10

Not on the map. Why?

Abstract (arXiv)

Recent text-to-image (T2I) generation models have achieved remarkable sucess by training on billion-scale datasets, following a `bigger is better' paradigm that prioritizes data quantity over availability (closed vs open source) and reproducibility (data decay vs established collections). We challenge this established paradigm by demonstrating that one can achieve capabilities of models trained on massive web-scraped collections, using only ImageNet enhanced with well-designed text and image augmentations. With this much simpler setup, we reach the performance of FLUX and achieve a +5 overall score over SD3 on GenEval and +12 on DPGBench over SDXL while using just 1/1000th the training images and 3x to 10x less parameters. This opens the way for more reproducible research as ImageNet is widely available and the proposed standardized training setup only requires 500 hours of H100 to train a text-to-image model.

Key facts. Cited by 10 papers in the map (data through 14 Oct 2026). Builds on 82 of its 103 references that are in the map. First posted on arXiv 28 Feb 2025.

Builds on

82 of 103 references.

References · 82

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  4. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  5. 2020Denoising Diffusion Implicit Modelscited by 9,781
  6. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  7. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  8. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  9. 2023Segment Anythingcited by 8,234
  10. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  11. 2023Visual Instruction Tuningcited by 7,869
  12. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
38 more
  1. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,585
  2. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,355
  3. 2022Scalable Diffusion Models with Transformerscited by 6,054
  4. 2022Flow Matching for Generative Modelingcited by 5,702
  5. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
  6. 2017Neural Discrete Representation Learningcited by 4,712
  7. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,356
  8. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,181
  9. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,152
  10. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,761
  11. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,700
  12. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 3,045
  13. 2022Scaling Instruction-Finetuned Language Modelscited by 3,024
  14. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,884
  15. 2021CLIPScore: A Reference-free Evaluation Metric for Image Captioningcited by 2,424
  16. 2022Scaling Autoregressive Models for Content-Rich Text-to-Image Generationcited by 1,295
  17. 2021Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Conceptscited by 1,216
  18. 2023ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generationcited by 1,045
  19. 2023Pick-a-Pic: An Open Dataset of User Preferences for Text-to-Image Generationcited by 977
  20. 2023GenEval: An Object-Focused Framework for Evaluating Text-to-Image Alignmentcited by 972
  21. 2019Improved Precision and Recall Metric for Assessing Generative Modelscited by 944
  22. 2023Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesiscited by 890
  23. 2021Going deeper with Image Transformerscited by 764
  24. 2024SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformerscited by 737
  25. 2024Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Thinkcited by 703
  26. 2023DataComp: In search of the next generation of multimodal datasetscited by 644
  27. 2020Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentationcited by 596
  28. 2024ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignmentcited by 475
  29. 2024Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generationcited by 457
  30. 2020Reliable Fidelity and Diversity Metrics for Generative Modelscited by 434
  31. 2024One Step Diffusion via Shortcut Modelscited by 391
  32. 2023DeepCache: Accelerating Diffusion Models for Freecited by 359
  33. 2024SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformerscited by 340
  34. 2024PixArt-Sigma: Weak-to-Strong Training of Diffusion Transformer for 4K Text-to-Image Generationcited by 329
  35. 2023SemDeDup: Data-efficient learning at web-scale through semantic deduplicationcited by 303
  36. 2020Query-Key Normalization for Transformerscited by 280
  37. 2025Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Stepscited by 234
  38. 2023Stable Bias: Analyzing Societal Representations in Diffusion Modelscited by 226

50 most cited of 82 All 82 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 8.

10 papers.

  1. 2026Better Source, Better Flow: Learning Condition-Dependent Source Distribution for Flow Matchingcited by 9
  2. 2026Diffusion Reinforcement Learning via Centered Reward Distillationcited by 5
  3. 2025Training-Free Synthetic Data Generation with Dual IP-Adapter Guidancecited by 3
  4. 2026Ambient Dataloops: Generative Models for Dataset Refinementcited by 2
  5. 2026Test-Time Conditioning with Representation-Aligned Visual Featurescited by 2
  6. 2026IdEst: Assessing Self-Supervised Learning Representations via Intrinsic Dimensioncited by 2
  7. 2025MIRO: MultI-Reward cOnditioned pretraining improves T2I quality and efficiencycited by 1
  8. 2026Diffusion Image Generation with Explicit Modeling of Data Manifold Geometrycited by 0
  9. 2026From SRA to Self-Flow: Data Augmentation or Self-Supervision?cited by 0
  10. 2026Diffusable Latents from Structure-Agnostic Distillationcited by 0