Researchers discover a shortcoming that makes LLMs less reliable
Large language models can learn to mistakenly link certain sentence patterns with specific topics — and may then repeat these patterns instead of reasoning.
Read More
Institute for Medical Engineering and Science (IMES)
Charting 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 MoreUsing generative AI, researchers design compounds that can kill drug-resistant bacteria
The team used two different AI approaches to design novel antibiotics, including one that showed promise against MRSA.
Read MoreCellLENS reveals hidden patterns in cell behavior within tissues, offering deeper insights into cell heterogeneity — vital for advancing cancer immunotherapy.
Read MoreThe Language/AI Incubator, an MIT Human Insight Collaborative project, is investigating how AI can improve communications among patients and practitioners.
Read MoreThe MIT-MGB Seed Program, launched with support from Analog Devices Inc., will fund joint research projects that advance technology and clinical research.
Read MoreResearchers find nonclinical information in patient messages — like typos, extra white space, and colorful language — reduces the accuracy of an AI model.
Read More3 Questions: How to help students recognize potential bias in their AI datasets
Courses on developing AI models for health care need to focus more on identifying and addressing bias, says Leo Anthony Celi.
Read MoreStudy shows vision-language models can’t handle queries with negation words
Words like “no” and “not” can cause this popular class of AI models to fail unexpectedly in high-stakes settings, such as medical diagnosis.
Read MoreCan deep learning transform heart failure prevention?
A deep neural network called CHAIS may soon replace invasive procedures like catheterization as the new gold standard for monitoring heart health.
Read More
You must be logged in to post a comment.