Self-rankings as a predictor of scientific impact beyond peer review.

Authors' self-rankings of their AI conference submissions predict future citations twice as well as peer review scores, highlighting their value in assessing research impact.

• Why it matters: With the rapid growth of submissions, traditional peer review struggles to reliably identify high-impact work, risking overlooked innovative research. Understanding alternative indicators like self-assessment can improve impact prediction and resource allocation.
• What they did: Researchers conducted a large-scale experiment over a year at a leading AI conference, collecting authors' self-rankings and comparing them to peer review scores and citation outcomes for over 1,000 papers.
• The result: Self-rankings effectively identified highly cited papers and outperformed peer review scores in predicting future impact, suggesting they can serve as a valuable complement in research evaluation processes.