Tag: clinical AI
Making it easier to verify an AI model’s responses
Despite their impressive capabilities, large language models are far from perfect. These artificial intelligence models sometimes “hallucinate” by generating incorrect or unsupported information in response to a query.Due to this hallucination problem, an LLM’s responses are often verified by human fact-checkers, especially if a model is deployed in a high-stakes setting like health care or finance. However, validation processes typically require people to read through long documents cited by the model, a task so onerous and error-prone it may prevent some users from deploying generative AI models in the first place.
To help human validators, MIT researchers created a user-friendly system that enables people to verify an LLM’s responses much more quickly. With this tool, called SymGen, an LLM generates responses with citations that point directly to the place in a source document, such as a given cell in a database. Learn more
Study: When allocating scarce resources with AI, randomization can improve fairness
Organizations are increasingly utilizing machine-learning models to allocate scarce resources or opportunities. For instance, such models can help companies screen resumes to choose job interview candidates or aid hospitals in ranking kidney transplant patients based on their likelihood of survival.When deploying a model, users typically strive to ensure its predictions are fair by reducing bias. This often involves techniques like adjusting the features a model uses to make decisions or calibrating the scores it generates.
However, researchers from MIT and Northeastern University argue that these fairness methods are not sufficient to address structural injustices and inherent uncertainties. In a new paper, they show how randomizing a model’s decisions in a structured way can improve fairness in certain situations. Learn more
Reverse Engineering Dementia With Human Computer Interaction
With many millions of Americans suffering from Alzheimer’s and dementia, cognitive decline is a major issue. The cost of dementia care is a rude awakening for many families, and patients experiencing the troubling symptoms of these difficulties might despair when they hear that there’s really no “cure,” just treatment. One of the problems is that dementia can look a lot like other forms of cognitive decline, like milder senility. So part of the process is diagnosis. We may not be able to cure dementia. But we can at least get help figuring out how to diagnose it with tools based on something called HCI. What is HCI? It stands for ‘human-computer interaction’. In some ways, it’s pretty much what it sounds like – the study of users and their behaviors in using computers. But it’s also a form of cognitive engineering, and may give us a window into the human mind. Looking at stylus-based interaction tasks, scientists are pondering quite a few metrics that reveal details on what people are thinking: eye fixation, blink rate, pupil size, etc. That in turn can help with the epidemic of cognitive impairment as we age (as presented by Randall Davis in this presentation. Davis also showed us some of the new technology coming down the pike). Learn more
Say Hello to Your Addiction Risk Score — Courtesy of the Tech Industry
MIT Assistant Professor of EECS and Jameel Clinic Principal Investigator Marzyeh Ghassemi spoke with New York Times Opinion Contributor Maia Szalavitz on how the task of addiction prediction and prevention could potentially perpetuate biases in medical decision making. Learn more
