SENTRA: Selected-Next-Token Transformer for LLM Text Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Plyler, Mitchell, Zhang, Yilun, Tuzhilin, Alexander, Khalifah, Saoud, Tian, Sen
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915502983479296
author Plyler, Mitchell
Zhang, Yilun
Tuzhilin, Alexander
Khalifah, Saoud
Tian, Sen
author_facet Plyler, Mitchell
Zhang, Yilun
Tuzhilin, Alexander
Khalifah, Saoud
Tian, Sen
contents LLMs are becoming increasingly capable and widespread. Consequently, the potential and reality of their misuse is also growing. In this work, we address the problem of detecting LLM-generated text that is not explicitly declared as such. We present a novel, general-purpose, and supervised LLM text detector, SElected-Next-Token tRAnsformer (SENTRA). SENTRA is a Transformer-based encoder leveraging selected-next-token-probability sequences and utilizing contrastive pre-training on large amounts of unlabeled data. Our experiments on three popular public datasets across 24 domains of text demonstrate SENTRA is a general-purpose classifier that significantly outperforms popular baselines in the out-of-domain setting.
format Preprint
id arxiv_https___arxiv_org_abs_2509_12385
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SENTRA: Selected-Next-Token Transformer for LLM Text Detection
Plyler, Mitchell
Zhang, Yilun
Tuzhilin, Alexander
Khalifah, Saoud
Tian, Sen
Computation and Language
Machine Learning
LLMs are becoming increasingly capable and widespread. Consequently, the potential and reality of their misuse is also growing. In this work, we address the problem of detecting LLM-generated text that is not explicitly declared as such. We present a novel, general-purpose, and supervised LLM text detector, SElected-Next-Token tRAnsformer (SENTRA). SENTRA is a Transformer-based encoder leveraging selected-next-token-probability sequences and utilizing contrastive pre-training on large amounts of unlabeled data. Our experiments on three popular public datasets across 24 domains of text demonstrate SENTRA is a general-purpose classifier that significantly outperforms popular baselines in the out-of-domain setting.
title SENTRA: Selected-Next-Token Transformer for LLM Text Detection
topic Computation and Language
Machine Learning
url https://arxiv.org/abs/2509.12385