arXiv Machine Learning
Aug 21
Continuous Adversarial MeanFlow Transfer
arXiv:2608. 19540v1 Announce Type: new Abstract: Training fast generators on new domains with limited data remains challenging for two reasons.
Continuous Adversarial MeanFlow Transfer
Training fast generators on new domains with limited data remains challenging for two reasons. First, adapting a pretrained diffusion or flow model to a new domain leaves its costly multi-step sampling unaddressed, and existing acceleration methods are tied to the source parameterization--$ε$, $x$, $v$, or $u$--leaving heterogeneous pretrained models with no common acceleration target.
Related stories
arXiv Machine Learning
Jul 17
MeanFlowNFT: Bringing Forward-Process RL to Average-Velocity Generators
arXiv:2607. 15273v1 Announce Type: cross Abstract: MeanFlow generators achieve fast few-step sampling by predicting average velocities over time intervals, making them attractive for efficient generation.
arXiv Computer Vision
Oct 2
Smoother Flow Matching via Contrastive Trajectory Repulsion
arXiv:2610.01408v1 Announce Type: new Abstract: Trajectory crossing remains a critical bottleneck in Flow Matching (FM), and previous works typically view these crossings from a theoretical optimizat...
arXiv AI
Jul 14
MixFlow Training: Alleviating Exposure Bias with Slowed Interpolation Mixture
arXiv:2512. 19311v2 Announce Type: replace-cross Abstract: This paper studies the training-testing discrepancy (a.
arXiv Machine Learning
Aug 27
Continuous Adversarial Flow Models
The paper introduces continuous adversarial flow models, a continuous-time flow framework trained with an adversarial objective that replaces the fixed mean-squared-error criterion of flow matching. By incorporating a learned discriminator, the method guides training toward a different generalized distribution, yielding samples more closely aligned with the target data distribution. Applied as a post‑training step, it markedly improves ImageNet 256px generation metrics—reducing the guidance‑free FID of latent‑space SiT from 8.26 to 3.63 and of pixel‑space JiT from 7.17 to 3.57—and also enhances guided generation and text‑to‑image benchmarks.
arXiv Computer Vision
Oct 1
Enhancing Autoregressive Video Generation via Representation Adversarial Distillation
arXiv:2609.40037v1 Announce Type: new Abstract: Few-step autoregressive video generation enables efficient streaming synthesis, but errors introduced in early temporal blocks are reused as context an...
arXiv Machine Learning
Sep 22
When Does Adversarial Refinement Help? A Negative Result and Open Problem in Adapting R3GAN to Time Series Imputation
The paper investigates whether the stable GAN architecture R3GAN can improve time‑series imputation when adapted to 1‑D temporal data. Using a coarse‑to‑fine refinement framework and a frequency‑domain discriminator, the authors evaluate 14 saved configurations across three datasets and find a negative result: most configurations either show negligible improvement or degrade performance compared to baseline methods. The study highlights that the usual argument—GANs optimize distributional objectives rather than point‑wise ones—does not fully explain the lack of benefit, and it poses an open problem regarding why a learned discriminator fails to provide useful refinement gradients while diffusion denoisers succeed, offering practical guidance on when adversarial refinement may be worthwhile.
arXiv Machine Learning
Oct 7
Adaptive Mean Estimation by In-Context Learning: A Gradient-Flow Analysis
arXiv:2610.07804v1 Announce Type: new Abstract: Prior Fitted Networks (PFNs) such as TabPFN now rival established statistical procedures across prediction and estimation tasks. A natural explanation...
arXiv AI
4d ago
From Suboptimal to Optimal: One-Step Latent Flow Policies for Offline Reinforcement Learning
arXiv:2610.16984v1 Announce Type: cross Abstract: Offline reinforcement learning aims to learn effective policies from fixed datasets without environment interaction. Although generative models such...
arXiv Machine Learning
3d ago
On the Stability and Convergence of Flow Map Training
The paper investigates the training dynamics of flow map models, which convert diffusion trajectories into one‑step samplers for efficient probabilistic generation. It provides a theoretical analysis of the functional gradient flow induced by MeanFlow and Lagrangian losses, proving their convergence to the true flow map and identifying the risk of unbounded updates. The authors show that adaptive normalization and a specific reweighting scheme not only bound updates but also guarantee convergence to a neighborhood of the solution, and they confirm these findings with synthetic experiments and image‑generation benchmarks.
arXiv AI
Sep 30
MeanFlowAdvantage: Stable Reward Fine-Tuning for Few-Step Average-Velocity Generators
arXiv:2609.37670v1 Announce Type: new Abstract: MeanFlow enables efficient few-step generation by predicting interval-average velocities, but this representation creates a mismatch for reward fine-tu...
arXiv AI
Jun 2
Heterogeneous Decentralized Diffusion Models
arXiv:2603. 06741v2 Announce Type: replace-cross Abstract: Training frontier-scale diffusion models often requires substantial computational resources concentrated in tightly-coupled clusters, limiting participation to well-resourced institutions.