The Public Archive
The Catalogue of Errors Ltd is using large language models to perform an audit of the existing literature. It is inspired by Socrates' original injunction to know what we do not know, as well as Karl Popper's definition of the scientific method as "the search for and the elimination of errors in the service of truth."
The reports produced by "Reviewer 2"—our automated assessment system—examine the existing literature for errors of research design, execution, and interpretation. They are produced by large language models, overseen by a human editor on behalf of The Catalogue of Errors Ltd. They represent the honest opinion of The Catalogue of Errors Ltd, although their contents should be verified by a qualified expert. Any errors in the reports can be left in the private comments, and they will be checked and corrected if necessary in a revised version.
You can also nominate a text to add to the public archive.
And please consider making a donation to help us maintain and expand it.