paper-with-me

홈 › Papers

Does Aurora Encode Atmospheric Structure? Latent Regime Analysis and Attribution

2026-06-24 · Emma Kasteleyn, Ana Lucic arxiv

ML foundation models are able to emulate atmospheric dynamics accurately and efficiently but operate as opaque ``black boxes''. We investigate the internal representations of the Aurora model using spatially pooled PCA and layer-wise relevance propagation (LRP). We find evidence that Aurora's latent space is primarily organized by seasonal cycles, whereas extreme storm events do not form a linearly separable cluster. LRP indicates that the model attends to features consistent with the 3D vertical structure of the Great Storm of 1987. Perturbation tests show masking relevant regions degrades forecasts $3.31\times$ more than random masking. These findings suggest that Aurora learns meteorological coherence and vertical structure without explicit instruction.

📄 PDF Abstract BibTeX arXiv:2606.26361

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Physical Consistency of Aurora's Encoder: A Quantitative Study

2025-11-11 · Benjamin Richards, Pushpa Kumar Balan arxiv

The high accuracy of large-scale weather forecasting models like Aurora is often accompanied by a lack of transparency, as their internal representations remain largely opaque. This "black box" nature hinders their adopt…

Weather Forecasting

Toward Mechanistic Interpretability of an AI Foundation Model Fine-Tuned for Atmospheric Chemistry

2026-07-22 · Jason Y. Hu, Ivan Higuera-Mendieta, Patrick Obin Sturm, Makoto M. Kelp arxiv

Weather forecasting foundation models (FMs) are increasingly fine-tuned to predict air quality, offering fast global pollution forecasts at lower computational cost than conventional chemical transport models. These FMs …

Weather Forecasting

AURORA-LM: Autoencoding Unified Representation for Continuous-Latent Diffusion Language Modeling

2026-08-03 · Jiajun Liang, Yucheng Liao, Yukang Cao, Jiazhe Wei 외 hf

Language remains an outlier in generative modeling: while images, video, and audio are increasingly modeled in continuous latent spaces, text generation still relies predominantly on discrete tokens. Existing continuous …

Text Generation

Correlation of Auroral Dynamics and GNSS Scintillation with an Autoencoder

2019-10-04 · Kara Lamb, Garima Malhotra, Athanasios Vlontzos, Edward Wagstaff 외

High energy particles originating from solar activity travel along the the Earth's magnetic field and interact with the atmosphere around the higher latitudes. These interactions often manifest as aurora in the form of v…

Integrating GNSS-Derived Zenith Wet Delay into a Weather Foundation Model Improves Precipitation Forecasting

2026-07-06 · Leonardo Trentini, Fanny Lehmann, Laura Crocetti, Benedikt Soja arxiv

Global Navigation Satellite Systems (GNSS), best known for positioning, also serve weather science, as atmospheric water vapour delays their signals. This delay, the Zenith Wet Delay (ZWD), is a direct, all-weather measu…

Precipitation Forecasting