Study: Some language reward models exhibit political bias
Research from the MIT Center for Constructive Communication finds this effect occurs even when reward models are trained on factual data.
Research from the MIT Center for Constructive Communication finds this effect occurs even when reward models are trained on factual data.
MIT and Sierra Leone’s Ministry of Health launch specialized program to train future clinicians on the fabrication of artificial limbs and braces for people with disabilities.
Associate Professor Catherine D’Ignazio thinks carefully about how we acquire and display data — and why we lack it for many things.
The method could help communities visualize and prepare for approaching storms.
Acclaimed keyboardist Jordan Rudess’s collaboration with the MIT Media Lab culminates in live improvisation between an AI “jam_bot” and the artist.
A new design tool uses UV and RGB lights to change the color and textures of everyday objects. The system could enable surfaces to display dynamic patterns, such as health data and fashion designs.
By snugly wrapping around neurons, these devices could help scientists probe subcellular regions of the brain, and might even help restore some brain function.
Organized by the MIT Museum, the 2024 celebration of science, technology, and culture was the largest in its history, with over 50,000 visitors at 300 events.
By enabling users to chat with an older version of themselves, Future You is aimed at reducing anxiety and guiding young people to make better choices.
The program will invite students to investigate new vistas at the intersection of music, computing, and technology.
Elemind, founded by researchers from MIT, has developed a headband that uses acoustic stimulation to move people into a sleep state.
MIT researchers identify facility-level factors that could worsen heat impacts for incarcerated people.
The interlocking bricks, which can be repurposed many times over, can withstand similar pressures as their concrete counterparts.
“We are adding a new layer of control between the world of computers and what your eyes see,” says Barmak Heshmat, co-founder of Brelyon and a former MIT postdoc.
Researchers developed an easy-to-use tool that enables an AI practitioner to find data that suits the purpose of their model, which could improve accuracy and reduce bias.