mlsanityData to 14 Oct 2026

Coupling Models for One-Step Discrete Generation

arXiv 2605.07193 ↗First posted 8 May 2026cs.LGcited by 2

Not on the map. Why?

Abstract (arXiv)

Generative modeling over discrete structures underpins applications across deep learning, from biological sequence design and code generation to large language models, yet generation often remains sequential, relying on autoregressive decoding or iterative refinement. In this work, we introduce Coupling Models(Coupling Models), a one-step discrete generative model that learns a direct coupling between discrete sequences and Gaussian latents. Unlike recent distillation methods that compress a pretrained multi-step sampler into a few steps, Coupling Model trains a purpose-built decoder to invert this coupling and generate samples in a single step. The model also avoids complex continuous flows over the simplex and hand-specified data-to-noise couplings. Empirically,Coupling Model improves the strongest one-step baselines in each domain: it reduces LM1B text-generation perplexity by 33% at its lowest-perplexity operating point, Fly Brain enhancer-design FBD by 18%, and MNIST-Binary FID by 46%. These results suggest that effective one-step discrete generation depends strongly on how data and noise are coupled before decoding. Code is available at https://github.com/pengzhangzhi/Coupling-Models.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 52 of its 57 references that are in the map. First posted on arXiv 8 May 2026.

Builds on

52 of 57 references.

References · 52

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  4. 2022Classifier-Free Diffusion Guidancecited by 5,847
  5. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  6. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,719
  7. 2016The Concrete Distribution: A Continuous Relaxation of Discrete Random Variablescited by 1,148
  8. 2022MaskGIT: Masked Generative Image Transformercited by 1,099
  9. 2022Diffusion-LM Improves Controllable Text Generationcited by 1,027
  10. 2024Simple and Effective Masked Diffusion Language Modelscited by 848
  11. 2021Vector Quantized Diffusion Model for Text-to-Image Synthesiscited by 774
  12. 2023Discrete Diffusion Modeling by Estimating the Ratios of the Data Distributioncited by 745
38 more
  1. 2023Muse: Text-To-Image Generation via Masked Generative Transformerscited by 648
  2. 2021Argmax Flows and Multinomial Diffusion: Learning Categorical Distributionscited by 574
  3. 2017Non-Autoregressive Neural Machine Translationcited by 463
  4. 2022DiffuSeq: Sequence to Sequence Text Generation with Diffusion Modelscited by 450
  5. 2018Do Deep Generative Models Know What They Don't Know?cited by 443
  6. 2022Analog Bits: Generating Discrete Data using Diffusion Models with Self-Conditioningcited by 373
  7. 2024Discrete Flow Matchingcited by 317
  8. 2015A note on the evaluation of generative modelscited by 292
  9. 2024Scaling Diffusion Language Models via Adaptation from Autoregressive Modelscited by 274
  10. 2021Autoregressive Diffusion Modelscited by 210
  11. 2024Scaling up Masked Diffusion Models on Textcited by 199
  12. 2018Deterministic Non-Autoregressive Neural Sequence Modeling by Iterative Refinementcited by 195
  13. 2022SSD-LM: Semi-autoregressive Simplex-based Diffusion Language Model for Text Generation and Modular Controlcited by 194
  14. 2024Simple Guidance Mechanisms for Discrete Diffusion Modelscited by 155
  15. 2023A Reparameterized Discrete Diffusion Model for Text Generationcited by 152
  16. 2025The Diffusion Dualitycited by 135
  17. 2024Unlocking Guidance for Discrete State-Space Diffusion and Flow Modelscited by 129
  18. 2019Insertion Transformer: Flexible Sequence Generation via Insertion Operationscited by 125
  19. 2022Latent Diffusion for Language Generationcited by 115
  20. 2024Dirichlet Flow Matching with Applications to DNA Sequence Designcited by 115
  21. 2024Diffusion Language Models Are Versatile Protein Learnerscited by 98
  22. 2025Path Planning for Masked Diffusion Model Samplingcited by 80
  23. 2024Fine-Tuning Discrete Diffusion Models via Reward Optimization with Applications to DNA and Protein Designcited by 79
  24. 2024Beyond Autoregression: Fast LLMs via Self-Distillation Through Timecited by 78
  25. 2023Dirichlet Diffusion Score Model for Biological Sequence Generationcited by 75
  26. 2024Fisher Flow Matching for Generative Modeling over Discrete Datacited by 72
  27. 2026Flow Map Language Models: One-step Language Modeling via Continuous Denoisingcited by 66
  28. 2024DPLM-2: A Multimodal Diffusion Protein Language Modelcited by 64
  29. 2024Categorical Flow Matching on Statistical Manifoldscited by 47
  30. 2026Categorical Flow Mapscited by 44
  31. 2023DiffuSeq-v2: Bridging Discrete and Continuous Text Spaces for Accelerated Seq2Seq Diffusion Modelscited by 36
  32. 2023Diffusion-NAT: Self-Prompting Discrete Diffusion for Non-Autoregressive Text Generationcited by 21
  33. 2025DLM-One: Diffusion Language Models for One-Step Sequence Generationcited by 20
  34. 2025FS-DFM: Fast and Accurate Long Text Generation with Few-Step Diffusion Language Modelscited by 17
  35. 2025Ultra-Fast Language Generation via Discrete Diffusion Divergence Instructcited by 16
  36. 2025ReDi: Rectified Discrete Flowcited by 14
  37. 2025Planner Aware Path Learning in Diffusion Language Models Trainingcited by 14
  38. 2025Bidirectional Normalizing Flow: From Data to Noise and Backcited by 7

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Distilled Continuous Diffusion Language Models Can Write Code in Few Steps---or Onecited by 2
  2. 2026One-Step Generation via Riemannian Wasserstein Gradient Flowscited by 0