The paper introduces Bridge Graphical Models (BGMs), a framework that decomposes continuous‑time generative models into independent design choices: endpoint coupling, bridge law, Markovian projection, and current‑preserving dynamics. It defines the Markovization gap as the time‑integrated conditional variance of bridge velocity given the Markov state, quantifying an irreducible loss before training. Experiments on synthetic, latent, and pixel‑space tasks (CIFAR‑10 and Fashion‑MNIST) show that a feature‑space proxy of this gap, estimated quickly before training, predicts downstream training loss and FID in the same direction as full training results.
arXiv:2607. 27372v1 Announce Type: new Abstract: The deep learning revolution, kicked off by AlexNet, taught us that end-to-end training beats decomposing a problem into hand-designed stages.
arXiv:2607. 23026v1 Announce Type: cross Abstract: Mean-field games (MFGs) offer a unifying lens on continuous-time generative modeling: a cost tuple recovering twelve prominent models---Continuous Normalizing Flows, OT-Flow, Score-based Models, Schr\"{o}dinger Bridges, and more---as special cases of one variational problem.
arXiv:2605. 00941v4 Announce Type: replace Abstract: Flow matching has become a leading framework for generative modeling, but quantifying the uncertainty of its samples remains an open problem.
Neural Bridge Processes (NBPs) replace the input‑independent forward kernel of Neural Diffusion Processes with an input‑anchored bridge trajectory, allowing conditioning inputs to influence the noisy training states. When input and output dimensions differ, NBPs learn an output‑space anchor that guides the generative path without altering the denoising backbone. Theoretical analysis shows that this anchoring yields pathwise input distinguishability, injects input information into noisy states, and provides a direct gradient pathway, leading to consistent performance gains across synthetic regression, EEG, CylinderFlow, and image regression tasks.
Flow-based generative models are typically sampled by solving a deterministic ordinary differential equation (ODE), whereas online reinforcement learning requires stochastic rollouts for policy exploration and optimization. Existing GRPO methods for flow models therefore replace the inference-time ODE with a stochastic differential equation (SDE) during training.
arXiv:2607. 19332v1 Announce Type: new Abstract: Generative models have undergone many generations of evolution, from VAEs/GANs to diffusion/flow matching.
arXiv:2605. 18931v2 Announce Type: replace-cross Abstract: Heavy-tailed distributions are prevalent in performance evaluation, network traffic, and risk modeling.
arXiv:2608. 05600v1 Announce Type: cross Abstract: Flow-based generative models are typically sampled by solving a deterministic ordinary differential equation (ODE), whereas online reinforcement learning requires stochastic rollouts for policy exploration and optimization.
arXiv:2606. 08672v1 Announce Type: cross Abstract: Diffusion and flow generative models sample by integrating a learned ODE, but high quality still requires many sequential model evaluations.
arXiv:2608.29335v1 Announce Type: new Abstract: Latent generative models typically follow a two-stage pipeline, training a variational autoencoder for reconstruction and then a generative model on th...
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.
arXiv:2602. 00797v2 Announce Type: replace-cross Abstract: Flow-based methods have achieved significant success in various generative modeling tasks, capturing nuanced details within complex data distributions.