Science

As multi-agent systems become more complex, pinpointing the source of failures requires innovative approaches to observability and control.

Researchers have developed a parametric framework that combines landscape characteristics with weather forecasts to predict flash floods up to 72 hours in advance.

New research explores how quantifying uncertainty in neural networks can build more reliable and ethically sound question-answering systems.

New machine learning techniques are unlocking deeper insights into the structure and dynamics of complex networks, from social systems to biological pathways.

A new deep learning approach combines accurate disease classification with explainable AI, offering farmers and agronomists a transparent understanding of diagnostic decisions.

New research demonstrates how deep learning can predict and optimize robot speed reductions during collaboration with humans, boosting efficiency and safety.

New research reveals how carefully crafted prompts can mislead genomic foundation models, highlighting the need for robust security auditing.

A new approach to semi-decentralized spatio-temporal graph neural networks intelligently reduces communication overhead without sacrificing prediction quality, especially during critical traffic incidents.

A new analysis digs into the complex causal factors that can lead to AI systems operating outside of intended parameters.

Researchers have released a comprehensive aerial imagery dataset designed to help artificial intelligence better understand and respond to flooded environments.