The paper introduces Contrastive Noise Alignment (CNA), a training-time method for generative flow models that dynamically aligns Gaussian noise with data samples using a cross-modal InfoNCE objective. By modeling noise as an interacting particle system and regularizing with angular entropy and radial norm penalties, CNA reduces arbitrary data-noise couplings and flow curvature. Empirical results show that CNA improves generation quality, lowering FID by over 50% for few-step pixel-space generation compared to standard rectified flow and outperforming optimal transport baselines by at least 24%.
arXiv:2510.15038v2 Announce Type: replace-cross Abstract: Flow-based Generative Models (FGMs) effectively transform noise into complex data distributions. Incorporating Optimal Transport (OT) to coup...
arXiv:2606. 04092v1 Announce Type: cross Abstract: Flow matching models learn to transport samples from a simple prior distribution to a complex data distribution.
The paper introduces ZeNOVA, a gradient‑free method for aligning initial noise in generative models. It uses annealed soft‑value guidance, manifold‑constrained hyperspherical Langevin dynamics, and Metropolis‑Hastings jumps to address instability in black‑box reward settings. Experiments on image and video models show ZeNOVA outperforms existing zeroth‑order baselines by more stably optimizing noise toward higher rewards.
arXiv:2610.13994v1 Announce Type: cross Abstract: Flow matching learns how to transport noise toward data, but its mean prediction does not directly report the endpoint uncertainty remaining at an in...
arXiv:2607. 19332v1 Announce Type: new Abstract: Generative models have undergone many generations of evolution, from VAEs/GANs to diffusion/flow matching.
Recent advances in distillation and flow-map models have enabled deterministic one- or few-step generation for high-quality data, facilitating a new branch of reward alignment approaches that directly...
arXiv:2602.19600v2 Announce Type: replace Abstract: Many high-dimensional datasets concentrate near a low-dimensional structure embedded in the ambient space. Generative models for such data must con...
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.
The paper investigates training objectives for denoising-based generative models, focusing on loss weighting and output parameterization such as noise-, clean image-, and velocity-based formulations. It conducts a systematic numerical study across synthetic datasets with controlled geometry and real image data, evaluating denoising accuracy via PSNR and generative quality via FID. The goal is to disentangle how training choices interact with data manifold dimensionality, model architecture, and dataset size, offering practical design insights rather than proposing a new method.
arXiv:2607. 12171v1 Announce Type: cross Abstract: In rectified-flow-based generative models, the neural network can be trained to predict two different targets, such as the instantaneous velocity or the data endpoint, to perform denoising.
arXiv:2602.15396v2 Announce Type: replace Abstract: Diffusion models often yield highly curved trajectories and noisy score targets due to an uninformative, memoryless forward process that induces in...
The paper introduces a total‑variance/signal‑to‑noise‑ratio (TV/SNR) framework that allows independent control of TV and SNR in diffusion models. By replacing exponentially exploding TV schedules with constant TV while keeping the SNR schedule, the authors achieve consistent improvements in generation efficiency. The approach generalizes the SNR schedule from optimal transport flow matching and works across various reverse diffusion solvers, reducing the search space to two interpretable parameters for robust sampling in scientific applications such as molecular structure generation.