Preprints now cited as legitimate sources by 1 in 5 journal articles

A new study by Chiaki Miura and Ichiro Sakata, titled "Preprints Without Curation Are Increasingly Cited by Journals" (arXiv:2607.25220), reveals a seismic shift in scholarly citation behavior. Analyzing metadata from six major preprint servers (including bioRxiv, arXiv, medRxiv) linked to the world's largest bibliographic database, the researchers found that the proportion of journal articles citing preprints has grown exponentially since the mid-2010s. Key inflection points coincided with bioRxiv's launch in 2013 and the introduction of a preprint content type in Crossref's DOI system in 2016. By 2021, nearly one in five journal articles cited at least one preprint, and an increasing share of those citations refer to non-curated (never peer-reviewed) versions.

The growth is not explained by confounding factors like the increasing stock of preprints, faster publishing cycles, or a few highly-cited outliers. Instead, it represents a structural change: preprints are now treated as legitimate, autonomous objects of citation, independent of their later publication status. Surprisingly, the COVID-19 pandemic did not amplify the trend but coincided with a deceleration in preprint citation growth. The authors argue this motivates renewed attention to citation conventions and journal policies as preprints become firmly embedded in scholarly communication.