Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

synthetic abstract detection

Synthetic abstract detection is the process of identifying whether the summary or abstract of an academic or scientific publication was generated by artificial intelligence rather than composed by human researchers. This task uses natural language processing techniques, statistical text analysis, and machine learning classifiers to evaluate stylistic patterns, token probability distributions, and linguistic anomalies typical of large language models. As generative text systems become increasingly capable of producing fluent and convincing technical prose, synthetic abstract detection serves as an essential mechanism for preserving the integrity of scholarly communication, aiding peer review screening, and preventing machine-generated or fraudulent research summaries from entering academic literature.

1 item

Cross-Domain Detection of GPT-2-Generated Technical Text

Cross-Domain Detection of GPT-2-Generated Technical Text

Juan Diego Rodriguez, Todd Hay, David Gros, Zain Shamsi, Ravi Srinivasan

OrganizationsEn SoluciónUniversity of California, DavisUniversity of Texas at Austin

Why you should read this

Demonstrates that machine-generated technical text detectors can successfully transfer across distinct scientific domains with only a few hundred labeled examples, enabling effective identification of synthetic content and paragraph tampering in full-length research papers.

Machine-generated text presents a potential threat not only to the public sphere, but also to the scientific enterprise, whereby genuine research is undermined by convincing, synthetic text. In this paper we examine the problem of detecting GPT-2-generated technical research text. We first consider the realistic scenario where the defender does not have full information about the adversary's text generation pipeline, but is able to label small amounts of in-domain genuine and synthetic text in order to adapt to the target distribution. Even in the extreme scenario of adapting a physics-domain detector to a biomedical detector, we find that only a few hundred labels are sufficient for good performance. Finally, we show that paragraph-level detectors can be used to detect the tampering of full-length documents under a variety of threat models.

Added

2026-09-26