PulseAugur
EN
LIVE 08:24:18

Volterra Generative Models Introduce Path-Dependent Noise for Enhanced AI Generation

By PulseAugur Editorial · [2 sources] ·

Researchers have introduced Volterra generative models, a new framework for continuous-time score-based generative models. Unlike traditional models that use memoryless Brownian perturbations, Volterra models incorporate path-dependent noise through fractional kernels. To manage the resulting non-Markovian dynamics, the researchers developed finite-dimensional Markovian lifts and a Gaussian-bridge reconstruction sampler for stability. Experiments on MNIST and CIFAR-10 datasets demonstrated that these persistent fractional perturbations can enhance generation quality. AI

IMPACT Introduces a novel approach to generative modeling by incorporating path-dependent noise, potentially improving image generation quality.

RANK_REASON The cluster contains an academic paper detailing a new generative model framework published on arXiv.

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

Volterra Generative Models Introduce Path-Dependent Noise for Enhanced AI Generation

COVERAGE [2]

  1. arXiv cs.AI TIER_1 Deutsch(DE) · Yusen Jia, Bingyan Han ·

    Volterra Generative Models

    arXiv:2606.18071v1 Announce Type: cross Abstract: Score-based diffusion models typically use Brownian perturbations, which provide tractable reverse-time dynamics but impose memoryless noising. We introduce Volterra generative models, a continuous-time score-based framework whose…

  2. arXiv cs.AI TIER_1 Deutsch(DE) · Bingyan Han ·

    Volterra Generative Models

    Score-based diffusion models typically use Brownian perturbations, which provide tractable reverse-time dynamics but impose memoryless noising. We introduce Volterra generative models, a continuous-time score-based framework whose forward process injects path-dependent noise thro…

RELATED ENTITIES

RELATED TOPICS