Domain Gating Ensemble Networks for AI-Generated Text Detection

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Tripathi, Arihant, Dugan, Liam, Gao, Charis, Huan, Maggie, Jin, Emma, Zhang, Peter, Zhang, David, Zhao, Julia, Callison-Burch, Chris
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916745256632320
author Tripathi, Arihant
Dugan, Liam
Gao, Charis
Huan, Maggie
Jin, Emma
Zhang, Peter
Zhang, David
Zhao, Julia
Callison-Burch, Chris
author_facet Tripathi, Arihant
Dugan, Liam
Gao, Charis
Huan, Maggie
Jin, Emma
Zhang, Peter
Zhang, David
Zhao, Julia
Callison-Burch, Chris
contents As state-of-the-art language models continue to improve, the need for robust detection of machine-generated text becomes increasingly critical. However, current state-of-the-art machine text detectors struggle to adapt to new unseen domains and generative models. In this paper we present DoGEN (Domain Gating Ensemble Networks), a technique that allows detectors to adapt to unseen domains by ensembling a set of domain expert detector models using weights from a domain classifier. We test DoGEN on a wide variety of domains from leading benchmarks and find that it achieves state-of-the-art performance on in-domain detection while outperforming models twice its size on out-of-domain detection. We release our code and trained models to assist in future research in domain-adaptive AI detection.
format Preprint
id arxiv_https___arxiv_org_abs_2505_13855
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Domain Gating Ensemble Networks for AI-Generated Text Detection
Tripathi, Arihant
Dugan, Liam
Gao, Charis
Huan, Maggie
Jin, Emma
Zhang, Peter
Zhang, David
Zhao, Julia
Callison-Burch, Chris
Computation and Language
Artificial Intelligence
Machine Learning
As state-of-the-art language models continue to improve, the need for robust detection of machine-generated text becomes increasingly critical. However, current state-of-the-art machine text detectors struggle to adapt to new unseen domains and generative models. In this paper we present DoGEN (Domain Gating Ensemble Networks), a technique that allows detectors to adapt to unseen domains by ensembling a set of domain expert detector models using weights from a domain classifier. We test DoGEN on a wide variety of domains from leading benchmarks and find that it achieves state-of-the-art performance on in-domain detection while outperforming models twice its size on out-of-domain detection. We release our code and trained models to assist in future research in domain-adaptive AI detection.
title Domain Gating Ensemble Networks for AI-Generated Text Detection
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2505.13855