mlsanityData to 14 Oct 2026

GLASS Flows: Transition Sampling for Alignment of Flow and Diffusion Models

arXiv 2509.25170 ↗First posted 29 Sep 2025cs.LG cs.AI stat.MLcited by 17

On the map: Diffusion models → Flow matching, time series and science

Abstract (arXiv)

The performance of flow matching and diffusion models can be greatly improved at inference time using reward alignment algorithms, yet efficiency remains a major limitation. While several algorithms were proposed, we demonstrate that a common bottleneck is the sampling method these algorithms rely on: many algorithms require to sample Markov transitions via SDE sampling, which is significantly less efficient and often less performant than ODE sampling. To remove this bottleneck, we introduce GLASS Flows, a new sampling paradigm that simulates a "flow matching model within a flow matching model" to sample Markov transitions. As we show in this work, this "inner" flow matching model can be retrieved from a pre-trained model without any re-training, combining the efficiency of ODEs with the stochastic evolution of SDEs. On large-scale text-to-image models, we show that GLASS Flows eliminate the trade-off between stochastic evolution and efficiency. Combined with Feynman-Kac Steering, GLASS Flows improve state-of-the-art performance in text-to-image generation, making it a simple, drop-in solution for inference-time scaling of flow and diffusion models.

Key facts. Cited by 17 papers in the map (data through 14 Oct 2026). Placed in Flow matching, time series and science, one of 14 directions in Diffusion models. Builds on 59 of its 66 references that are in the map. First posted on arXiv 29 Sep 2025.

Builds on

59 of 66 references.

Closest ancestors · 10

Same direction, newest first

  1. 2025Feynman-Kac-Flow: Inference Steering of Conditional Flow Matching to an Energy-Tilted Posterior1 Sep
  2. 2025Transition Matching: Scalable and Flexible Generative Modeling30 Jun
  3. 2025Gaussian Mixture Flow Matching Models7 Apr
  4. 2025On the Guidance of Flow Matching4 Feb
  5. 2024Flow Matching Guide and Code9 Dec
  6. 2024D-Flow: Differentiating through Flows for Controlled Generation21 Feb
  7. 2023Training-free Linear Image Inverses via Flows25 Sep
  8. 2023Practical and Asymptotically Exact Conditional Sampling in Diffusion Models30 Jun
  9. 2022Flow Matching for Generative Modeling6 Oct
  10. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow7 Sep

Other references · 49

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2020Denoising Diffusion Implicit Modelscited by 9,781
  4. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  5. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  6. 2022Scalable Diffusion Models with Transformerscited by 6,054
  7. 2022Classifier-Free Diffusion Guidancecited by 5,847
  8. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,181
  9. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 3,045
  10. 2021CLIPScore: A Reference-free Evaluation Metric for Image Captioningcited by 2,424
  11. 2021SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equationscited by 1,937
  12. 2022Diffusion Posterior Sampling for General Noisy Inverse Problemscited by 1,416
37 more
  1. 2022Scaling Autoregressive Models for Content-Rich Text-to-Image Generationcited by 1,295
  2. 2023ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generationcited by 1,045
  3. 2023Pick-a-Pic: An Open Dataset of User Preferences for Text-to-Image Generationcited by 977
  4. 2023GenEval: An Object-Focused Framework for Evaluating Text-to-Image Alignmentcited by 972
  5. 2023Stochastic Interpolants: A Unifying Framework for Flows and Diffusionscited by 939
  6. 2023Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesiscited by 890
  7. 2023Diffusion Model Alignment Using Direct Preference Optimizationcited by 790
  8. 2024SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformerscited by 737
  9. 2025Flow-GRPO: Training Flow Matching Models via Online RLcited by 661
  10. 2023Universal Guidance for Diffusion Modelscited by 437
  11. 2025DanceGRPO: Unleashing GRPO on Visual Generationcited by 413
  12. 2023DreamSim: Learning New Dimensions of Human Visual Similarity using Synthetic Datacited by 337
  13. 2023FreeDoM: Training-Free Energy-Guided Conditional Diffusion Modelcited by 294
  14. 2021Score-Based Generative Modeling with Critically-Damped Langevin Diffusioncited by 233
  15. 2024Adjoint Matching: Fine-tuning Flow and Diffusion Generative Models with Memoryless Stochastic Optimal Controlcited by 233
  16. 2022Diffusion models as plug-and-play priorscited by 231
  17. 2025A General Framework for Inference-time Scaling and Steering of Diffusion Modelscited by 224
  18. 2025MixGRPO: Unlocking Flow-based GRPO Efficiency with Mixed ODE-SDEcited by 178
  19. 2023Manifold Preserving Guided Diffusioncited by 157
  20. 2024ReNO: Enhancing One-step Text-to-Image Models through Reward-based Noise Optimizationcited by 119
  21. 2024TFG: Unified Training-Free Guidance for Diffusion Modelscited by 107
  22. 2024Principled Probabilistic Imaging using Diffusion Models as Plug-and-Play Priorscited by 97
  23. 2025Feynman-Kac Correctors in Diffusion: Annealing, Guidance, and Product of Expertscited by 96
  24. 2025Inference-Time Alignment in Diffusion Models with Reward-Guided Generation: Tutorial and Reviewcited by 88
  25. 2023Restart Sampling for Improving Generative Processescited by 84
  26. 2023Diffusion Models for Black-Box Optimizationcited by 76
  27. 2023The Blessing of Randomness: SDE Beats ODE in General Diffusion-based Image Editingcited by 62
  28. 2025Inference-time Scaling of Diffusion Models through Classical Searchcited by 49
  29. 2023Deep Momentum Multi-Marginal Schr"odinger Bridgecited by 42
  30. 2023Bespoke Solvers for Generative Flow Modelscited by 37
  31. 2025Dynamic Search for Inference-Time Alignment in Diffusion Modelscited by 37
  32. 2024Training-free Diffusion Model Alignment with Sampling Demonscited by 31
  33. 2025Distributional Diffusion Models with Scoring Rulescited by 25
  34. 2025LATINO-PRO: LAtent consisTency INverse sOlver with PRompt Optimizationcited by 16
  35. 2024Learning from negative feedback, or positive feedback or bothcited by 13
  36. 2023Nested Diffusion Processes for Anytime Image Generationcited by 5
  37. 2025TADA: Improved Diffusion Sampling with Training-free Augmented Dynamicscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 17.

17 papers.

Sort:

By area:

  1. 2026Transition Matching Distillation for Fast Video Generationcited by 30
  2. 2026Diamond Maps: Efficient Reward Alignment via Stochastic Flow Mapscited by 29
  3. 2026Enhanced Diffusion Sampling: Efficient Rare Event Sampling and Free Energy Calculation with Diffusion Modelscited by 13
  4. 2026How to Guide Your Flow: Few-Step Alignment via Flow Map Reward Guidancecited by 6
  5. 2026Aligning Few-Step Generative Models by Amortizing Sample-based Variational Inferencecited by 3
  6. 2026Advances in GRPO for Generation Models: A Surveycited by 2
  7. 2026The tractability landscape of diffusion alignment: regularization, rewards, and computational primitivescited by 2
  8. 2026Stitched Value Model for Diffusion Alignmentcited by 2
  9. 2026Exact Posterior Score Estimation for Solving Linear Inverse Problemscited by 2
  10. 2026BFMT: Enhancing Search Capabilities of Tree Sampler via Bootstrap Flow-Map Treecited by 1
  11. 2026WTF?! Simulation-Free Reinforcement Learning with Wasserstein-Tilted Flow Mapscited by 1
  12. 2026FluxLite: Inference-Time Proposal Control for Discrete Diffusion Modelscited by 1
5 more
  1. 2026Conflict-Aware Additive Guidance for Flow Models under Compositional Rewardscited by 0
  2. 2026PG-MAP: Joint MAP Optimization for Inference-Time Alignment of Diffusion and Flow-Matching Modelscited by 0
  3. 2026Sequentially-Controlled Interactive Multi-Particle Flow-Maps for Online Feedback-Driven Searchcited by 0
  4. 2026Direct Conditional Transition Sampling for Diffusion Inverse Problemscited by 0
  5. 2026Steepest Guidance: A Practical and Principled Approach to Inference-Time Alignment of Flow and Diffusion-based Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Flow map matching with stochastic interpolants: A mathematical framework for consistency models
  2. 2026Alignment of Diffusion Model and Flow Matching for Text-to-Image Generation
  3. 2025Score Distillation of Flow Matching Models
  4. 2025Diff2Flow: Training Flow Matching Models via Diffusion Model Alignment
  5. 2024Discrete Flow Matching
  6. 2026Reflective Flow Sampling Enhancement
  7. 2025Exploring the Design Space of Transition Matching
  8. 2025Diffusion Tree Sampling: Scalable inference-time alignment of diffusion models