Abstract

We propose a method to determine whether a given article was written entirely by a generative language model or perhaps contains edits by a different author, possibly a human. Our process involves multiple tests for the origin of individual sentences or other pieces of text and combining these tests using a method that is sensitive to rare alternatives, i.e., non-null effects are few and scattered across the text in unknown locations. Interestingly, this method also identifies pieces of text suspected to contain edits. We demonstrate the effectiveness of the method in detecting edits through extensive evaluations using real data and provide an information-theoretic analysis of the factors affecting its success. In particular, we discuss optimality properties under a theoretical framework for text editing saying that sentences are generated mainly by the language model, except perhaps for a few sentences that might have originated via a different mechanism. Our analysis raises several interesting research questions at the intersection of information theory and data science.

Keywords: generative language model, authorship analysis, information theory, multiple comparisons, sparsity, higher criticism

08/15/2024: To preview this content, click below for the Just Accepted version of the article. This peer-reviewed version has been accepted for its content and is currently being copyedited to conform with HDSR’s style and formatting requirements.

©2024 Idan Kashtan and Alon Kipnis. This article is licensed under a Creative Commons Attribution (CC BY 4.0) International license, except where otherwise indicated with respect to particular material included in the article.