Enhancing LLM collaboration for smarter, more efficient solutions

“Co-LLM” algorithm helps a general-purpose AI model collaborate with an expert large language model by combining the best parts of both answers, leading to more factual responses.

Alex Shipps | MIT CSAIL • mit
Sept. 16, 2024 ~8 min

A fast and flexible approach to help doctors annotate medical scans

“ScribblePrompt” is an interactive AI framework that can efficiently highlight anatomical structures across different medical scans, assisting medical workers to delineate regions of interest and abnormalities.

Alex Shipps | MIT CSAIL • mit
Sept. 9, 2024 ~8 min


A framework for solving parabolic partial differential equations

A new algorithm solves complicated partial differential equations by breaking them down into simpler problems, potentially guiding computer graphics and geometry processing.

Alex Shipps | MIT CSAIL • mit
Aug. 28, 2024 ~7 min

Toward a code-breaking quantum computer

Building on a landmark algorithm, researchers propose a way to make a smaller and more noise-tolerant quantum factoring circuit for cryptography.

Adam Zewe | MIT News • mit
Aug. 23, 2024 ~8 min

LLMs develop their own understanding of reality as their language abilities improve

In controlled experiments, MIT CSAIL researchers discover simulations of reality developing deep within LLMs, indicating an understanding of language beyond simple mimicry.

Alex Shipps | MIT CSAIL • mit
Aug. 14, 2024 ~9 min

MIT researchers use large language models to flag problems in complex systems

The approach can detect anomalies in data recorded over time, without the need for any training.

Adam Zewe | MIT News • mit
Aug. 14, 2024 ~7 min

A new approach to fine-tuning quantum materials

An MIT-led group shows how to achieve precise control over the properties of Weyl semimetals and other exotic substances.

Steve Nadis | Department of Nuclear Science and Engineering • mit
Aug. 12, 2024 ~9 min

A new model offers robots precise pick-and-place solutions

SimPLE learns to pick, regrasp, and place objects using the objects’ computer-aided design model.

Anne Wilson | Department of Mechanical Engineering • mit
Aug. 9, 2024 ~5 min


Helping robots practice skills independently to adapt to unfamiliar environments

New algorithm helps robots practice skills like sweeping and placing objects, potentially helping them improve at important tasks in houses, hospitals, and factories.

Alex Shipps | MIT CSAIL • mit
Aug. 8, 2024 ~7 min

Method prevents an AI model from being overconfident about wrong answers

More efficient than other approaches, the “Thermometer” technique could help someone know when they should trust a large language model.

Adam Zewe | MIT News • mit
July 31, 2024 ~7 min

/

109