Strong Stochastic Flow Maps learn the strong solution map of additive-noise SDEs via a pathwise-convergent polynomial Brownian approximation, generalizing deterministic flow maps and enabling simulation-free training that outperforms prior weak-convergence stochastic methods on image generation and
hub
Boffi, Michael S
22 Pith papers cite this work. Polarity classification is still indexing.
22
Pith papers citing it
hub tools
citation-role summary
background 3
method 1
citation-polarity summary
representative citing papers
AnyFlow enables any-step video diffusion by distilling flow-map transitions over arbitrary time intervals with on-policy backward simulation.
FMRG reformulates guidance as deterministic optimal control, deriving a single-trajectory method using the flow map that matches or exceeds baselines on reward-guided generation and inverse problems with 3 NFEs at text-to-image scale.
Flow models reach 99.2% Sudoku accuracy in 7 passes and 96.1% on out-of-distribution Sudoku-Extreme by selecting dynamically stable candidates and training with self-conditioning plus DPO to avoid failed outputs.
Isokinetic Flow Matching adds a lightweight regularization term to flow matching that penalizes acceleration along paths via self-guided finite differences, yielding straighter trajectories and large gains in few-step sampling quality on CIFAR-10.
Shortcut models enable high-quality single or few-step sampling in diffusion models with one network and training phase by conditioning on desired step size.
RMMD simultaneously distills diffusion models and optimizes rewards, yielding better FID-reward trade-offs on ImageNet than DI++, DRaFT and HyperNoise, and a 7.5x faster GenCast model that beats its teacher on 93% of weather variables while improving calibration.
FMLM+ with Posterior Refinement bridges masked diffusion and flow map models to match discrete baseline quality in language generation using 32x fewer neural function evaluations via posterior scoring and refinement.
BFQ enables single-step noise-to-action mapping in offline RL by dividing flow-path displacements into bootstrappable short-range components learned from marginal velocity.
Zero-parameter naive samplers achieve state-of-the-art generative perplexity while producing incoherent text, proving the metric is unsound; distributional divergences like MAUVE and energy distance correctly rank them below trained models.
Derives RAM, a reward-adjusted consistency loss extending diffusion pretraining regression to efficient KL-regularized RL post-training, achieving peak rewards up to 50x faster than Flow-GRPO on Stable Diffusion 3.5M.
By requiring and using highly discriminative LLM text features, the work enables the first effective one-step text-conditioned image generation with MeanFlow.
A framework learns fault-indexed Perron-Frobenius operators from trajectory data to provide certifiable 2-Wasserstein bounds for detecting actuator faults and enabling recovery via density propagation in nonlinear control-affine systems.
OT-NFM parameterizes the flow map directly with neural flows and uses optimal transport for consistent noise-data couplings to achieve ODE-free one-step generation while avoiding mean collapse.
Self-consistent distribution matching plus cache-aware mixed-step training improves 2–4 NFE video quality on Wan 2.1 and real-time autoregressive backbones without extra inference cost.
Drifting with Gaussian kernels exactly matches score-matching on smoothed distributions via Tweedie's formula, while Laplace kernels approximate this closely in high dimensions.
Continuous flows on token embeddings with flow-map distillation produce one-step language models whose quality exceeds recent 8-step discrete diffusion baselines on LM1B and OpenWebText.
MeanFlow uses a derived identity between average and instantaneous velocities to train one-step flow models, achieving FID 3.43 on ImageNet 256x256 with 1-NFE from scratch.
Formalizes nonlinear M2M regression and introduces transformer architectures as static maps and dynamic velocity fields between probability measures, tested on synthetic, particle, and organoid datasets.
DriftXpress approximates the attraction field of drifting models with a Nyström landmark projection, reducing training time by 2.6–6.7× at comparable FID.
DE-CM trains a flow-map consistency model on three sub-trajectories (coupling, instantaneous, noise-to-noisy) and reports 1.70 FID one-step on ImageNet 256.
A principled monograph showing that variational, score-based, and flow-based diffusion models are instances of one continuous-time transport backbone, with sampling equal to solving a differential equation governed by the Fokker–Planck equation.
citing papers explorer
-
Strong Stochastic Flow Maps
Strong Stochastic Flow Maps learn the strong solution map of additive-noise SDEs via a pathwise-convergent polynomial Brownian approximation, generalizing deterministic flow maps and enabling simulation-free training that outperforms prior weak-convergence stochastic methods on image generation and
-
AnyFlow: Any-Step Video Diffusion Model with On-Policy Flow Map Distillation
AnyFlow enables any-step video diffusion by distilling flow-map transitions over arbitrary time intervals with on-policy backward simulation.
-
How to Guide Your Flow: Few-Step Alignment via Flow Map Reward Guidance
FMRG reformulates guidance as deterministic optimal control, deriving a single-trajectory method using the flow map that matches or exceeds baselines on reward-guided generation and inverse problems with 3 NFEs at text-to-image scale.
-
Flow Reasoning Models: Scaling Reasoning Through Iterative Self-Refinement
Flow models reach 99.2% Sudoku accuracy in 7 passes and 96.1% on out-of-distribution Sudoku-Extreme by selecting dynamically stable candidates and training with self-conditioning plus DPO to avoid failed outputs.
-
Isokinetic Flow Matching for Pathwise Straightening of Generative Flows
Isokinetic Flow Matching adds a lightweight regularization term to flow matching that penalizes acceleration along paths via self-guided finite differences, yielding straighter trajectories and large gains in few-step sampling quality on CIFAR-10.
-
One Step Diffusion via Shortcut Models
Shortcut models enable high-quality single or few-step sampling in diffusion models with one network and training phase by conditioning on desired step size.
-
Diffusion Fine-tuning with Rewarded Moment Matching Distillation
RMMD simultaneously distills diffusion models and optimizes rewards, yielding better FID-reward trade-offs on ImageNet than DI++, DRaFT and HyperNoise, and a 7.5x faster GenCast model that beats its teacher on 93% of weather variables while improving calibration.
-
Posterior Refinement: Fast Language Generation via Any-Order Flow Maps
FMLM+ with Posterior Refinement bridges masked diffusion and flow map models to match discrete baseline quality in language generation using 32x fewer neural function evaluations via posterior scoring and refinement.
-
Fast and Highly Expressive Policy Learning for Offline Reinforcement Learning via Bootstrapped Flow Q-Learning
BFQ enables single-step noise-to-action mapping in offline RL by dividing flow-path displacements into bootstrappable short-range components learned from marginal velocity.
-
Hacking Generative Perplexity: Why Unconditional Text Evaluation Needs Distributional Metrics
Zero-parameter naive samplers achieve state-of-the-art generative perplexity while producing incoherent text, proving the metric is unsound; distributional divergences like MAUVE and energy distance correctly rank them below trained models.
-
Reinforce Adjoint Matching: Scaling RL Post-Training of Diffusion and Flow-Matching Models
Derives RAM, a reward-adjusted consistency loss extending diffusion pretraining regression to efficient KL-regularized RL post-training, achieving peak rewards up to 50x faster than Flow-GRPO on Stable Diffusion 3.5M.
-
Extending One-Step Image Generation from Class Labels to Text via Discriminative Text Representation
By requiring and using highly discriminative LLM text features, the work enables the first effective one-step text-conditioned image generation with MeanFlow.
-
Perron-Frobenius Contractive Operator Matching for Data-Driven Reachable Fault Identification and Recovery
A framework learns fault-indexed Perron-Frobenius operators from trajectory data to provide certifiable 2-Wasserstein bounds for detecting actuator faults and enabling recovery via density propagation in nonlinear control-affine systems.
-
ODE-free Neural Flow Matching for One-Step Generative Modeling
OT-NFM parameterizes the flow map directly with neural flows and uses optimal transport for consistent noise-data couplings to achieve ODE-free one-step generation while avoiding mean collapse.
-
Salt: Self-Consistent Distribution Matching with Cache-Aware Training for Fast Video Generation
Self-consistent distribution matching plus cache-aware mixed-step training improves 2–4 NFE video quality on Wan 2.1 and real-time autoregressive backbones without extra inference cost.
-
A Unified View of Score-Based and Drifting Models
Drifting with Gaussian kernels exactly matches score-matching on smoothed distributions via Tweedie's formula, while Laplace kernels approximate this closely in high dimensions.
-
Flow Map Language Models: One-step Language Modeling via Continuous Denoising
Continuous flows on token embeddings with flow-map distillation produce one-step language models whose quality exceeds recent 8-step discrete diffusion baselines on LM1B and OpenWebText.
-
Mean Flows for One-step Generative Modeling
MeanFlow uses a derived identity between average and instantaneous velocities to train one-step flow models, achieving FID 3.43 on ImageNet 256x256 with 1-NFE from scratch.
-
Measure-to-measure Regression with Transformers
Formalizes nonlinear M2M regression and introduces transformer architectures as static maps and dynamic velocity fields between probability measures, tested on synthetic, particle, and organoid datasets.
-
DriftXpress: Faster Drifting Models via Projected RKHS Fields
DriftXpress approximates the attraction field of drifting models with a Nyström landmark projection, reducing training time by 2.6–6.7× at comparable FID.
-
Dual-End Consistency Model
DE-CM trains a flow-map consistency model on three sub-trajectories (coupling, instantaneous, noise-to-noisy) and reports 1.70 FID one-step on ImageNet 256.
-
The Principles of Diffusion Models
A principled monograph showing that variational, score-based, and flow-based diffusion models are instances of one continuous-time transport backbone, with sampling equal to solving a differential equation governed by the Fokker–Planck equation.