MIT researchers advance automated interpretability in AI models

MAIA is a multimodal agent that can iteratively design experiments to better understand various components of AI systems.

Rachel Gordon | MIT CSAIL • mit
yesterday ~10 min

Proton-conducting materials could enable new green energy technologies

Analysis and materials identified by MIT engineers could lead to more energy-efficient fuel cells, electrolyzers, batteries, or computing devices.

David L. Chandler | MIT News • mit
yesterday ~7 min


Large language models don’t behave like people, even though we may expect them to

A new study shows someone’s beliefs about an LLM play a significant role in the model’s performance and are important for how it is deployed.

Adam Zewe | MIT News • mit
yesterday ~8 min

AI model identifies certain breast tumor stages likely to progress to invasive cancer

The model could help clinicians assess breast cancer stage and ultimately help in reducing overtreatment.

Adam Zewe | MIT News • mit
July 22, 2024 ~6 min

China-based emissions of three potent climate-warming greenhouse gases spiked in past decade

Two studies pinpoint their likely industrial sources and mitigation opportunities.

Mark Dwortzan | MIT Joint Program on the Science and Policy of Global Change • mit
July 18, 2024 ~5 min

Machine learning unlocks secrets to advanced alloys

An MIT team uses computer models to measure atomic patterns in metals, essential for designing custom materials for use in aerospace, biomedicine, electronics, and more.

Poornima Apte | Department of Materials Science and Engineering • mit
July 18, 2024 ~9 min

Astronomers spot a highly “eccentric” planet on its way to becoming a hot Jupiter

The planet’s wild orbit offers clues to how such large, hot planets take shape.

Jennifer Chu | MIT News • mit
July 17, 2024 ~9 min

Creating and verifying stable AI-controlled systems in a rigorous and flexible way

Neural network controllers provide complex robots with stability guarantees, paving the way for the safer deployment of autonomous vehicles and industrial machines.

Alex Shipps | MIT CSAIL • mit
July 17, 2024 ~7 min


AI method radically speeds predictions of materials’ thermal properties

The approach could help engineers design more efficient energy-conversion systems and faster microelectronic devices, reducing waste heat.

Adam Zewe | MIT News • mit
July 16, 2024 ~8 min

How to assess a general-purpose AI model’s reliability before it’s deployed

A new technique enables users to compare several large models and choose the one that works best for their task.

Adam Zewe | MIT News • mit
July 16, 2024 ~7 min

/

599