Don’t worry, this article will be short enough to not create a time-consuming task for readers looking to absorb its key points. Researchers don’t always have that luxury when they’re trying to keep up with the latest scientific papers and publications.
The Allen Institute for AI has created a tool that uses artificial intelligence to help readers skim more effectively.
Semantic Scholar, a research and product development team within AI2, has released a new tool as part of its Semantic Reader that enhances skimming. It’s an AI-powered tool that automatically pre-highlights and color codes a research paper to show the goals, methods, and key results, helping scholars focus their reading, and improve their research skills over time.
Cassidy Trier, a senior product designer at AI2, designs human and AI interactions. In a new blog post, Trier highlights the design steps taken to arrive at the latest iteration of AI2’s augmented scientific reading application.
Trier said the skimming feature started as an internship project by Raymond Fok, a PhD student at the University of Washington, who interned at Semantic Scholar.
Fok researched and developed an early PDF reader with automatically-generated highlights called Scim, according to Trier. The Scim prototype was tested with 458 papers and 31 participants across two studies. The study results suggest that the system “helps study participants gain a high-level understanding of papers and draws their attention to details in the paper that might otherwise be skipped. Participants noted this is particularly helpful in dense passages and with papers in unfamiliar domains,” Trier wrote.
Early research and design of the tool was passed to the product team at Semantic Scholar, which worked on subsequent iterations and features, while gaining feedback from users.
Users can try Semantic Reader’s skimming feature by visiting this example paper, or another of the roughly 480,000 computer science papers available from ArXiv sources. You must be on a desktop device to use Semantic Reader.