Autoregression-Free Neural Operators for Time-Dependent PDEs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Jiaquan, Qin, Caiyan, Bian, Haoyu, Cai, Libin, Lu, Yi, Zhang, Chaoning, Dong, Wei, Guo, Yuanfang, Yang, Yang, Shen, Hen Tao
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911725987561472
author Zhang, Jiaquan
Qin, Caiyan
Bian, Haoyu
Cai, Libin
Lu, Yi
Zhang, Chaoning
Dong, Wei
Guo, Yuanfang
Yang, Yang
Shen, Hen Tao
author_facet Zhang, Jiaquan
Qin, Caiyan
Bian, Haoyu
Cai, Libin
Lu, Yi
Zhang, Chaoning
Dong, Wei
Guo, Yuanfang
Yang, Yang
Shen, Hen Tao
contents Neural operators learn mappings from function-dependent inputs to solutions, providing an effective framework for solving partial differential equations (PDEs). For time-dependent PDEs, existing methods typically perform long-horizon prediction through autoregressive rollout directly in high-dimensional physical field spaces, where each predicted state is recursively fed back as the input for the next step. Although effective for short-term prediction, this autoregressive rollout and the lack of continuous-time modeling lead to progressive error accumulation over long-horizon rollouts. In this work, we propose Autoregression-Free Neural Operators (AFNO), which map the time evolution of PDEs into a latent space and model continuous-time vector fields within it. AFNO uses flow matching to learn the latent vector field, thereby enabling continuous evolution over extended horizons, avoiding autoregressive rollout and capturing dynamics under varying parameter configurations through explicit conditioning on physical parameters. Theoretical analysis and extensive experiments on six PDEs demonstrate that AFNO improves long-horizon prediction stability and consistently reduces rollout errors compared with the baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2605_25413
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Autoregression-Free Neural Operators for Time-Dependent PDEs
Zhang, Jiaquan
Qin, Caiyan
Bian, Haoyu
Cai, Libin
Lu, Yi
Zhang, Chaoning
Dong, Wei
Guo, Yuanfang
Yang, Yang
Shen, Hen Tao
Machine Learning
Artificial Intelligence
Numerical Analysis
Neural operators learn mappings from function-dependent inputs to solutions, providing an effective framework for solving partial differential equations (PDEs). For time-dependent PDEs, existing methods typically perform long-horizon prediction through autoregressive rollout directly in high-dimensional physical field spaces, where each predicted state is recursively fed back as the input for the next step. Although effective for short-term prediction, this autoregressive rollout and the lack of continuous-time modeling lead to progressive error accumulation over long-horizon rollouts. In this work, we propose Autoregression-Free Neural Operators (AFNO), which map the time evolution of PDEs into a latent space and model continuous-time vector fields within it. AFNO uses flow matching to learn the latent vector field, thereby enabling continuous evolution over extended horizons, avoiding autoregressive rollout and capturing dynamics under varying parameter configurations through explicit conditioning on physical parameters. Theoretical analysis and extensive experiments on six PDEs demonstrate that AFNO improves long-horizon prediction stability and consistently reduces rollout errors compared with the baselines.
title Autoregression-Free Neural Operators for Time-Dependent PDEs
topic Machine Learning
Artificial Intelligence
Numerical Analysis
url https://arxiv.org/abs/2605.25413