MIT scientists investigate memorization risk in the age of clinical AI
New research demonstrates how AI models can be tested to ensure they don’t cause harm by revealing anonymized patient health data.
Computer Science and Artificial Intelligence Laboratory (CSAIL)
Guided learning lets “untrainable” neural networks realize their potential
CSAIL researchers find even “untrainable” neural nets can learn effectively when guided by another network’s built-in biases using their guidance method.
Read MoreA new way to increase the capabilities of large language models
MIT-IBM Watson AI Lab researchers developed an expressive architecture that provides better state tracking and sequential reasoning in LLMs over long texts.
Read MoreEnabling small language models to solve complex reasoning tasks
The “self-steering” DisCIPL system directs small models to work together on tasks with constraints, like itinerary planning and budgeting.
Read MoreThe technique can help scientists in economics, public health, and other fields understand whether to trust the results of their experiments.
Read MoreNew control system teaches soft robots the art of staying safe
MIT CSAIL and LIDS researchers developed a mathematically grounded system that lets soft robots deform, adapt, and interact with people and objects, without violating safety limits.
Read MoreMIT scientists debut a generative AI model that could create molecules addressing hard-to-treat diseases
BoltzGen generates protein binders for any biological target from scratch, expanding AI’s reach from understanding biology toward engineering it.
Read MoreUnderstanding the nuances of human-like intelligence
Associate Professor Phillip Isola studies the ways in which intelligent machines “think,” in an effort to safely integrate AI into human society.
Read MoreCharting the future of AI, from safer answers to faster thinking
MIT PhD students who interned with the MIT-IBM Watson AI Lab Summer Program are pushing AI tools to be more flexible, efficient, and grounded in truth.
Read MoreMIT researchers propose a new model for legible, modular software
The coding framework uses modular concepts and simple synchronization rules to make software clearer, safer, and easier for LLMs to generate.
Read More
You must be logged in to post a comment.