Science

Researchers have created a publicly available dataset and evaluation framework to assess how well natural language processing and retrieval-augmented generation systems align with the requirements of the EU AI Act.

New research explores how large language models assess investor profiles, revealing both promise and pitfalls in using AI for personalized financial advice.

New research shows that analyzing the emotional tone of news articles, using advanced AI, can significantly improve the accuracy of aluminum price forecasting.

A new data-driven approach leveraging machine learning and combined data sources offers unprecedented accuracy in forecasting spatial spectrum demand.

As multi-agent systems become increasingly prevalent, a critical examination of their vulnerabilities and potential security failures is paramount.

A rigorous analytical approach, drawing on complex systems theory, can unlock more effective strategies for reducing the climate impact of our food supply.

A new framework reconstructs the complex web of production and financial relationships to reveal how vulnerabilities cascade through the modern economy.

A rigorous new framework assesses the performance of ensemble learning methods in detecting financial risk within Enterprise Resource Planning systems.

A new approach to detecting anomalies in chemical processes uses symbolic machine learning to build understandable models, offering a compelling alternative to ‘black box’ neural networks.

A new approach to forecasting U.S. recessions focuses on identifying economic indicators that shift from stable to ‘at-risk,’ proving surprisingly effective even with basic modeling techniques.