aiqbench analysis
RD-JEPA demonstrates a significant advancement in machine learning by enabling data-efficient adaptation to new reaction-diffusion systems through self-supervised pretraining on latent trajectories. This approach outperforms supervised baselines and architecture-matched controls, proving that predictive latent representations can generalize across parameterized partial differential equations without requiring new simulation corpora for every operator change.
RD-JEPA: Predictive latent pretraining for few-trajectory transfer across reaction--diffusion equations
arXiv:2609.29403v1 Announce Type: cross Abstract: Learning surrogates for time-dependent partial differential equations often requires a new simulation corpus when the governing operator changes. We introduce RD-JEPA, a joint-embedding predictive architecture for self-supervised pretraining on reaction-diffusion trajectories. A single model is pretrained on five parameterized systems and then adapted to three held-out systems whose reaction operators and trajectories are excluded from pretraining. Using one, five, or ten complete trajectories from a held-out system, RD-JEPA achieves lower mean relative discrete $\ell^2$ field error and mean absolute spatial first-difference error than five supervised surrogate baselines, an independently trained control that removes the trajectory-dependent predictive latent pathway, and an architecture-matched model trained from scratch. Within the evaluated equations, output resolution, forecast horizons, and choices of adaptation trajectories, the results indicate that prediction of future-state representations can support data-efficient adaptation across related reaction-diffusion systems.
Abstract:Learning surrogates for time-dependent partial differential equations often requires a new simulation corpus when the governing operator changes. We introduce RD-JEPA, a joint-embedding predictive architecture for self-supervised pretraining on reaction-diffusion trajectories. A single model is pretrained on five parameterized systems and then adapted to three held-out systems whose reaction operators and trajectories are excluded from pretraining. Using one, five, or ten complete trajectories from a held-out system, RD-JEPA achieves lower mean relative discrete $\ell^2$ field error and mean absolute spatial first-difference error than five supervised surrogate baselines, an independently trained control that removes the trajectory-dependent predictive latent pathway, and an architecture-matched model trained from scratch. Within the evaluated equations, output resolution, forecast horizons, and choices of adaptation trajectories, the results indicate that prediction of future-state representations can support data-efficient adaptation across related reaction-diffusion systems.
Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
arXiv:2609.29403 [cs.AI]
(or arXiv:2609.29403v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.29403 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 231 words
Related coverage
Juno: Taming Predictive Latents for Vision-Language-Action Models
arXiv · Oct 8, 2026
EEG-JEPA: Structured Latent Prediction for EEG Foundation Models
arXiv · Aug 5, 2026
Clin-JEPA: A Multi-Phase Co-Training Framework for Joint-Embedding Predictive Pretraining on EHR Patient Trajectories
arXiv · Sep 18, 2026
J++ Lens: Jacobian Filtering Enables More Faithful Workspace Lenses
LessWrong · Oct 8, 2026