activity
20242026
collaborators

6 papers

cs.LG2026

Does Aurora Encode Atmospheric Structure? Latent Regime Analysis and Attribution

Emma Kasteleyn, Ana Lucic

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 A…

cs.LG2026

PhysMetrics.Weather: An Evaluation Framework for Physical Consistency in ML Weather Models

Emma Kasteleyn, Timo Maier, Axel Lauer +3

Machine learning weather prediction (MLWP) models have achieved impressive forecasting performance at a small fraction of the computational costs required for traditional physics-b…

cs.LG2026

(Sparse) Attention to the Details: Preserving Spectral Fidelity in ML-based Weather Forecasting Models

Maksim Zhdanov, Ana Lucic, Max Welling +1

We introduce Mosaic, a probabilistic weather forecasting model that addresses three failure modes of spectral degradation in ML-based weather prediction: spectral damping (statisti…

cs.AI2026

Same Content, Different Answers: Cross-Modal Inconsistency in MLLMs

Angela van Sprang, Laurens Samson, Ana Lucic +3

We introduce two new benchmarks REST and REST+ (Render-Equivalence Stress Tests) to enable systematic evaluation of cross-modal inconsistency in multimodal large language models (M…

cs.LG2025

Equivariant Sparse Autoencoders: Mechanistic Interpretability of Neural Networks on Symmetric Data

Ege Erdogan, Ana Lucic

Machine learning (ML) models achieve remarkable performance but remain hard to interpret due to their scale and complexity. In particular, their activations entangle many concepts…

physics.ao-ph2024

A Foundation Model for the Earth System

Cristian Bodnar, Wessel P. Bruinsma, Ana Lucic +15

Reliable forecasts of the Earth system are crucial for human progress and safety from natural disasters. Artificial intelligence offers substantial potential to improve prediction…