15.06.2025
Xiaoxiang Zhu Receives Best Paper Award at EarthVision Workshop @ CVPR 2025
Advancing Foundation Models for Earth Observation
MCML PI Xiaoxiang Zhu and collaborators received the Best Paper Award at the EarthVision Workshop, held in conjunction with CVPR 2025, for their paper "Panopticon: Advancing Any-Sensor Foundation Models for Earth Observation." The work introduces a foundation model designed to process data from diverse Earth observation sensors within a unified framework.
Panopticon extends foundation models beyond fixed sensor configurations by enabling flexible integration of optical and radar imagery with varying spectral characteristics. By supporting sensor-agnostic learning, the approach improves performance across a wide range of Earth observation tasks and paves the way for more versatile AI systems for environmental monitoring and geospatial analysis.
Related
29.10.2026
Teaching Language Models to Express Uncertainty
An ICLR 2026 paper shows how reinforcement learning improves confidence estimates from language models.
27.10.2026
Runner-Up Prize at AutoML 2026 for Tabular-DC-Bench
Tabular-DC-Bench, by MCML Junior Member Mamdouh Aljoud and MCML Director Thomas Seidl, among others, won a runner-up prize at AutoML 2026.
27.10.2026
Welcoming Ewa Szczurek to MCML
Ewa Szczurek joins MCML as Professor of AI in Life Sciences at LMU, focusing on probabilistic and generative models for medicine.
