KAUST researchers developed a machine learning algorithm to control a deformable mirror within the Subaru Telescope's exoplanet imaging camera, compensating for atmospheric turbulence. The algorithm, which computes a partial singular value decomposition (SVD), outperforms a standard SVD by a factor of four. The KAUST team received a best paper award at the PASC Conference for this work, which has already been deployed at the Subaru Telescope. Why it matters: This advancement enables sharper images of exoplanets, facilitating their identification and study, and showcases the impact of optimizing core linear algebra algorithms.
The AQABA project, a collaboration involving KAUST and international institutions, studies air quality and climate change in the Arabian Basin using a marine research vessel. The vessel traveled from France through the Suez Canal, around the Arabian Peninsula, and stopped at KAUST. Researchers presented findings on atmospheric dust, air pollution, and aerosol measurements, highlighting the impact of dust on renewable energy and air pollution on health. Why it matters: The project provides crucial data for understanding and addressing climate challenges and air quality issues in the Middle East.
KAUST collaborated with NASA's Langley Research Center to launch six weather balloons from KAUST's Coastal & Marine Laboratory, reaching an altitude of 35 kilometers. The balloons were equipped with instruments to measure meteorological properties and characterize the optical properties of aerosols, including a Compact Optical Backscatter Aerosol Detector (COBALD). The research focuses on understanding the impact of dust aerosols on the Arabian Peninsula, including their effects on climate, air quality, and solar energy. Why it matters: This collaboration advances understanding of atmospheric aerosols in the region, with implications for climate modeling, solar energy efficiency, and Red Sea ecosystems.
A new benchmark, LongShOTBench, is introduced for evaluating multimodal reasoning and tool use in long videos, featuring open-ended questions and diagnostic rubrics. The benchmark addresses the limitations of existing datasets by combining temporal length and multimodal richness, using human-validated samples. LongShOTAgent, an agentic system, is also presented for analyzing long videos, with both the benchmark and agent demonstrating the challenges faced by state-of-the-art MLLMs.