Indeed, when we as researchers first look at a paper, we look at where it was published, who are the authors and where are they from, and some metrics like downloads, reads, and of course, citations. Most of this information is superficial, contributing no real useful information to understanding the research. Even citations, as used today, are mostly used as a number to make a quick assessment of the work, where the higher the number of citations an article has, the better.
However, citations represent a wealth of information. Behind each of the 41 articles that cite my work are years of directly related research and many thousands, if not millions, of dollars of research funding. But if I want to learn what these articles say about my work, I would need to read each of them. This is so impractical that it is effectively never fully done.
We’re changing that at scite, a new platform that uses deep learning to show how an article has been cited and, specifically, if it has been supported or contradicted, where the citations appear in the citing paper, and if it is a self-cite or a citation from a review or article. In short, we want to make citations smart–citations that not merely tell how many times an article is cited, but also provide the context for each citation and the citation meaning, such as whether it provides supporting or contradicting evidence for the cited claim.