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:2606. 30376v1 Announce Type: new Abstract: Aligning generative flow models on continuous spaces via online reinforcement learning is constrained by intractable trajectory likelihoods.
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:2608.24855v1 Announce Type: new Abstract: Latent world models are inherently strong encoders that transform image pixel to latent embedding, yet existing world models still rely on online traje...
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:2605. 29920v2 Announce Type: replace Abstract: We introduce Midpoint Generative Models (MGM), a principled framework for training one-step generative models.
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:2609. 22752v1 Announce Type: new Abstract: Diffusion models have demonstrated strong power in generative modeling tasks across multiple domains, exhibiting a remarkable capability of learning complex distributions from samples.
arXiv:2602. 17554v3 Announce Type: replace Abstract: Training large-scale generative models is resource-intensive and relies heavily on heuristic dataset weighting.
arXiv:2606. 06967v1 Announce Type: new Abstract: Generative policies provide expressive and multimodal action distributions, making them attractive for reinforcement learning (RL) in complex continuous-control tasks.
arXiv:2607. 19332v1 Announce Type: new Abstract: Generative models have undergone many generations of evolution, from VAEs/GANs to diffusion/flow matching.
arXiv:2601. 21542v3 Announce Type: replace-cross Abstract: Flow Matching (FM) models have emerged as a leading paradigm for high-fidelity synthesis.
The paper introduces a mesh‑free, self‑supervised neural operator—called the Normalizing Flow Invertible Solution Transformer (NFIST)—for stochastic mean‑field control (MFC). By reformulating the controlled Fokker–Planck dynamics as a deterministic continuity equation via a probability‑flow ODE and an invertible normalizing‑flow transformer, the authors enable closed‑form score evaluation with linear cost per particle. The resulting operator learns from task prompts (distribution parameters or particle clouds) and can solve unseen MFC tasks in a single forward pass, achieving zero‑shot generalization across applications such as stochastic optimal control, Schrödinger bridges, systemic‑risk control, and obstacle‑avoiding path planning.