Solaris: A Foundation Model of the Sun
Harris Abdul Majid, Pietro Sittoni, Francesco Tudisco

TL;DR
Solaris is a novel large-scale 3D transformer model trained on 13 years of solar imagery, capable of forecasting the Sun's atmosphere and generalizing to new wavelengths with limited data.
Contribution
This paper introduces Solaris, the first foundation model for solar atmospheric forecasting, leveraging extensive multi-wavelength data and demonstrating strong generalization capabilities.
Findings
Solaris outperforms models trained from scratch on a new wavelength.
It effectively captures complex solar atmospheric dynamics.
Pre-training on multi-wavelength data improves forecasting accuracy.
Abstract
Foundation models have demonstrated remarkable success across various scientific domains, motivating our exploration of their potential in solar physics. In this paper, we present Solaris, the first foundation model for forecasting the Sun's atmosphere. We leverage 13 years of full-disk, multi-wavelength solar imagery from the Solar Dynamics Observatory, spanning a complete solar cycle, to pre-train Solaris for 12-hour interval forecasting. Solaris is built on a large-scale 3D Swin Transformer architecture with 109 million parameters. We demonstrate Solaris' ability to generalize by fine-tuning on a low-data regime using a single wavelength (1700 {\AA}), that was not included in pre-training, outperforming models trained from scratch on this specific wavelength. Our results indicate that Solaris can effectively capture the complex dynamics of the solar atmosphere and transform solar…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsSolar and Space Plasma Dynamics · Astro and Planetary Science
MethodsDense Connections · Label Smoothing · Dropout · Linear Layer · Stochastic Depth · Layer Normalization · Byte Pair Encoding · Adam · Residual Connection · Softmax
