mlsanityData to 14 Oct 2026

Continuous-Time Trajectory Generation from Discrete Observations with Stochasticity

arXiv 2609.32026 ↗First posted 25 Sep 2026stat.ML cs.LG math.DScited by 0

Not on the map. Why?

Abstract (arXiv)

Physical systems evolve continuously in time, yet their states are typically observed only at discrete times. Generating trajectories consistent with their probability densities from such observations therefore requires capturing the continuous-time evolution rather than only learning transition mappings between consecutive observations. We propose PhiBE-Flow, a framework that directly estimates the probability velocity field induced by the stochastic differential equation (SDE) which governs this continuous-time distributional evolution. PhiBE-Flow learns from discrete observations using a model-free approach requiring neither known SDE coefficients nor score estimation. We establish convergence guarantees for the method, accounting for both time-discretization and finite-sample errors. We evaluate PhiBE-Flow on systems of increasing complexity, from controlled stochastic numerical systems to Navier--Stokes dynamics and real-world videos. Our results show that PhiBE-Flow accurately recovers probability flows of stochastic dynamics, preserves multiscale physical statistics, and improves video generation performance over representative baselines. The code is available at https://github.com/R1fe/PhiBE-Flow.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 47 of its 56 references that are in the map. First posted on arXiv 25 Sep 2026.

Builds on

47 of 56 references.

References · 47

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  4. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  5. 2022Flow Matching for Generative Modelingcited by 5,702
  6. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,253
  7. 2014Sequence to Sequence Learning with Neural Networkscited by 2,986
  8. 2022Video Diffusion Modelscited by 2,305
  9. 2019Transformer-XL: Attentive Language Models Beyond a Fixed-Length Contextcited by 1,850
  10. 2015Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcastingcited by 1,436
  11. 2018Towards Accurate Generative Models of Video: A New Metric & Challengescited by 1,234
  12. 2022Building Normalizing Flows with Stochastic Interpolantscited by 995
35 more
  1. 2023Stochastic Interpolants: A Unifying Framework for Flows and Diffusionscited by 939
  2. 2023Improving and generalizing flow-based generative models with minibatch optimal transportcited by 879
  3. 2019Hamiltonian Neural Networkscited by 712
  4. 2021VideoGPT: Video Generation using VQ-VAE and Transformerscited by 658
  5. 2021Diffusion Schr"odinger Bridge with Applications to Score-Based Generative Modelingcited by 602
  6. 2021Machine learning accelerated computational fluid dynamicscited by 471
  7. 2015Unsupervised Learning of Video Representations using LSTMscited by 435
  8. 2019Machine Learning for Fluid Mechanicscited by 385
  9. 2022MCVD: Masked Conditional Video Diffusion for Prediction, Generation, and Interpolationcited by 309
  10. 2020Scalable Gradients for Stochastic Differential Equationscited by 271
  11. 2016Unsupervised Learning for Physical Interaction through Video Predictioncited by 260
  12. 2021Likelihood Training of Schr"odinger Bridge using Forward-Backward SDEs Theorycited by 219
  13. 2023PDE-Refiner: Achieving Accurate Long Rollouts with Neural PDE Solverscited by 216
  14. 2019Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learningcited by 212
  15. 2019Neural Stochastic Differential Equations: Deep Latent Gaussian Models in the Diffusion Limitcited by 176
  16. 2018Stochastic Adversarial Video Predictioncited by 173
  17. 2021Neural SDEs as Infinite-Dimensional GANscited by 161
  18. 2025Do generative video models understand physical principles?cited by 161
  19. 2018Data-Driven Forecasting of High-Dimensional Chaotic Systems with Long Short-Term Memory Networkscited by 155
  20. 2025VideoPhy-2: A Challenging Action-Centric Physical Commonsense Evaluation in Video Generationcited by 149
  21. 2020A Review on Deep Learning Techniques for Video Predictioncited by 114
  22. 2020Transformers for Modeling Physical Systemscited by 106
  23. 2014Video (language) modeling: a baseline for generative models of natural videoscited by 92
  24. 2024Probabilistic Forecasting with Stochastic Interpolants and F"ollmer Processescited by 66
  25. 2022Efficient Video Prediction via Sparsely Conditioned Flow Matchingcited by 55
  26. 2020Solving Inverse Stochastic Problems from Discrete Particle Observations Using the Fokker-Planck Equation and Physics-informed Neural Networkscited by 52
  27. 2025"PhyWorldBench": A Comprehensive Evaluation of Physical Realism in Text-to-Video Modelscited by 36
  28. 2025SDE Matching: Scalable and Simulation-Free Training of Latent Stochastic Differential Equationscited by 23
  29. 2020Identifying Latent Stochastic Differential Equationscited by 21
  30. 2023Amortized Reparametrization: Efficient and Scalable Variational Inference for Latent SDEscited by 18
  31. 2023Latent SDEs on Homogeneous Spacescited by 16
  32. 2023How to Learn and Generalize From Three Minutes of Data: Physics-Constrained and Uncertainty-Aware Neural Stochastic Differential Equationscited by 10
  33. 2023Variational Inference for SDEs Driven by Fractional Noisecited by 9
  34. 2025Geometry-Preserving Encoder/Decoder in Latent Generative Modelscited by 6
  35. 2023Decomposing Thermodynamic Dissipation of Linear Langevin Systems via Oscillatory Modes and Its Application to Neural Dynamicscited by 4

Built on by

Not cited yet.