Image
“Rather than needing to use all these external tools to try and understand what is going on under the hood, maybe we can train models to talk about themselves in a way that is accurate,” says Jacob Andreas (Credit: Bryce Vickmark).
CSAIL article

When deploying deep learning models in high-stakes scenarios, researchers would like to understand how they work. Today, this requires unraveling an enormously complex network of internal processing nodes to understand what the model knows and how it will likely behave in practice.

Image
null
CSAIL article

On Monday, November 9th, 2026, over 1,000 people joined the MIT CSAIL Alliances Agentic AI Summit. With 250 people gathered on campus and 900+ online, this event drew widespread attention from industry professionals and academics in more than 40 countries and a range of business sectors. 

Image
Regina Barzilay (left) and James Collins are the recipients of this year’s 2026 Pritzker Prize for AI in Science Research Excellence, which is jointly awarded by the University of Chicago and Caltech (Credits: Photos: Max Rousseau (Barzilay) and M. Scott Brauer (Collins)).
CSAIL article

Regina Barzilay, MIT’s School of Engineering Distinguished Professor of AI and Health, and James Collins, MIT’s Termeer Professor of Medical Engineering and Science, are the recipients of the 2026 Pritzker Prize for AI in Science Research Excellence. 

Stata Center exterior

Join us for a CSAIL Alliances Tech Talk!

This talk will explore the portfolio construction process at Trexquant and the practical application of deep learning methods in quantitative trading.

Denis will introduce Mixture of Experts (MoE) models and demonstrate their behavior through a toy example using synthetic data. He will then discuss how neural networks are constructed and applied within Trexquant's real-world trading strategies, including aspects of model setup and expert construction that are specific to quantitative trading.

CSAIL article

There’s a delicate art to teaching robots, even when you’re preparing them for predictable environments like factories, where they’ll repeat the same tasks a little differently depending on the obstacles they face. Whether a human is suddenly in their way or there’s new clutter, the machine must closely mimic its operator’s actions by staying on a trajectory (or motion path).

Image
null
CSAIL article

In 2026, the hype for artificial intelligence (AI) agents is louder than ever before. These semi-autonomous programs can “think” and execute well-defined tasks in areas like customer service and software development, typically using language models (LMs). But fields like medical diagnosis and scientific discovery require them to inquire about a vast range of solutions in uncertain environments, which LMs struggle with.