mlsanityData to 14 Oct 2026

ROMS-IMLE: A Minimalist Approach to Competitive Single-Step Generative Modelling

arXiv 2607.19332 ↗First posted 21 Jul 2026cs.LG cs.CVcited by 0

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

Generative models have undergone many generations of evolution, from VAEs/GANs to diffusion/flow matching. Along the way, the underlying techniques have become more complicated and various beliefs about what drives strong empirical performance have taken hold. Due to the success of diffusion models and flow matching, one of the more common beliefs is the importance of transforming the noise distribution to the data distribution gradually through many small transformations. We ask whether this is truly necessary, and take a minimalist approach to designing a competitive generative model. We start with the bare-bones essentials, namely just a training objective and a model. We purposefully make both simple. For the training objective, we choose Implicit Maximum Likelihood Estimation (IMLE), and eschew more complicated alternatives such as variational inference, adversarial training and numerical integration. For the model, we eschew transformers and instead choose a moderately sized convolutional network. Then we judiciously added elements that are truly essential, which surprisingly do not include iterative denoising. The result is a single-step parameter-efficient generative model that produces high quality samples at fast speed: it achieves an FID of 2.56 on ImageNet 256 and simultaneously attains good precision and recall.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 41 of its 55 references that are in the map. First posted on arXiv 21 Jul 2026.

Builds on

41 of 55 references.

Closest ancestors · 14

Same direction, newest first

  1. 2025Inductive Moment Matching10 Mar
  2. 2024One Step Diffusion via Shortcut Models16 Oct
  3. 2024SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers16 Jan
  4. 2023Consistency Models2 Mar
  5. 2022Scalable Diffusion Models with Transformers19 Dec
  6. 2022Classifier-Free Diffusion Guidance26 Jul
  7. 2022Elucidating the Design Space of Diffusion-Based Generative Models1 Jun
  8. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  9. 2021Tackling the Generative Learning Trilemma with Denoising Diffusion GANs15 Dec
  10. 2021Score-based Generative Modeling in Latent Space10 Jun
  11. 2021Diffusion Models Beat GANs on Image Synthesis11 May
  12. 2020Score-Based Generative Modeling through Stochastic Differential Equations26 Nov
2 more
  1. 2020Denoising Diffusion Implicit Models6 Oct
  2. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 27

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  3. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  4. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  5. 2022Generative Adversarial Networkscited by 5,627
  6. 2013Auto-Encoding Variational Bayescited by 4,053
  7. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,988
  8. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,626
  9. 2017Arbitrary Style Transfer in Real-time with Adaptive Instance Normalizationcited by 2,010
  10. 2024Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Predictioncited by 1,023
  11. 2022Building Normalizing Flows with Stochastic Interpolantscited by 995
  12. 2019Improved Precision and Recall Metric for Assessing Generative Modelscited by 944
15 more
  1. 2025Mean Flows for One-step Generative Modelingcited by 561
  2. 2017Towards Principled Methods for Training Generative Adversarial Networkscited by 464
  3. 2023Rethinking FID: Towards a Better Evaluation Metric for Image Generationcited by 310
  4. 2017A General and Adaptive Robust Loss Functioncited by 236
  5. 2021StyleSwin: Transformer-based GAN for High-resolution Image Generationcited by 152
  6. 2019Seeing What a GAN Cannot Generatecited by 148
  7. 2025Improved Mean Flows: On the Challenges of Fastforward Generative Modelscited by 110
  8. 2025EQ-VAE: Equivariance Regularized Latent Space for Improved Generative Image Modelingcited by 89
  9. 2024An Undetectable Watermark for Generative Image Modelscited by 81
  10. 2018Implicit Maximum Likelihood Estimationcited by 52
  11. 2020Combating Mode Collapse in GAN training: An Empirical Analysis using Hessian Eigenvaluescited by 27
  12. 2021Styleformer: Transformer based Generative Adversarial Networks with Style Vectorcited by 27
  13. 2025IMLE Policy: Fast and Sample Efficient Visuomotor Policy Learning via Implicit Maximum Likelihood Estimationcited by 13
  14. 2021Multimodal Shape Completion via IMLEcited by 5
  15. 2026Implicit Maximum Likelihood Estimation for Real-time Generative Model Predictive Controlcited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Rejection Sampling IMLE: Designing Priors for Better Few-Shot Image Synthesis
  2. 2023The Score-Difference Flow for Implicit Generative Modeling
  3. 2026Are We Overthinking One-Step Generation?
  4. 2025Revisiting Diffusion Models: From Generative Pre-training to One-Step Generation
  5. 2025Generative AI in Depth: A Survey of Recent Advances, Model Variants, and Real-World Applications
  6. 2023DiffFlow: A Unified SDE Framework for Score-Based Diffusion Models and Generative Adversarial Networks
  7. 2022Progressive Distillation for Fast Sampling of Diffusion Models
  8. 2024GUD: Generation with Unified Diffusion