Skip to main content

University of Michigan School of Information

Menu

Media Center

UMSI welcomes media inquiries

Adar: Can AI research assistants help scientists read a barrage of papers? The results are mixed

Quoted by Science. Associate professor Eytan Adar. As scientists read flood of papers, AI developers aim to help.

Wednesday, 11/29/2023

With the number of research papers from scientists expanding in recent years - nearly 3 million were published last year -  researchers and students are overwhelmed with the plethora of new information emerging each day and looking for ways to keep up with advances in their fields.

A host of new AI tools promising to rapidly summarize the contents of a research paper are proliferating. But do they work? Researchers are skeptical, reports Science. 

University of Michigan School of Information associate professor Eytan Adar has been examining the effectiveness and accuracy of these tools. For Science, he recommends looking into Skim, which provides “a quick diagnostic, a triage, about whether [a paper] is worth engaging with,” which “is very valuable,” he says. 

RELATED 

Read “As scientists read a flood of papers, AI developers aim to help” on Science. 

Visit Eytan Adar’s UMSI faculty profile to learn more about research interests and publications. 

 

— Noor Hindi, UMSI public relations specialist