MIT researchers make language models scalable self-learners
The scientists used a natural language-based logical inference dataset to create smaller language models that outperformed much larger counterparts.
Computer Science and Artificial Intelligence Laboratory (CSAIL)
Scaling audio-visual learning without labels
A new multimodal technique blends major self-supervised learning methods to learn more similarly to humans.
A more effective way to train machines for uncertain, real-world situations
Researchers develop an algorithm that decides when a “student” machine should follow its teacher, and when it should learn on its own.
New tool helps people choose the right method for evaluating AI models
Selecting the right method gives users a more accurate picture of how their model is behaving, so they are better equipped to correctly interpret its predictions.
Researchers use AI to identify similar materials in images
This machine-learning method could assist with robotic scene understanding, image editing, or online recommendation systems.
Using data to write songs for progress
Senior Ananya Gurumurthy adds her musical talents to her math and computer science studies to advocate using data for social change.
Is medicine ready for AI? Doctors, computer scientists, and policymakers are cautiously optimistic
With the artificial intelligence conversation now mainstream, the 2023 MIT-MGB AI Cures conference saw attendance double from previous years.
3 Questions: Jacob Andreas on large language models
The CSAIL scientist describes natural language processing research through state-of-the-art machine-learning models and investigation of how language can enhance other types of artificial intelligence.
Study: AI models fail to reproduce human judgements about rule violations
Models trained using common data-collection techniques judge rule violations more harshly than humans would, researchers report.
Training machines to learn more like humans do
Researchers identify a property that helps computer vision models learn to represent the visual world in a more stable, predictable way.