Domain Gating Ensemble Networks for AI-Generated Text Detection

TL;DR

This paper introduces DoGEN, a domain gating ensemble method that enhances AI-generated text detection across unseen domains by combining expert models with domain classification, achieving state-of-the-art results.

Contribution

The paper proposes DoGEN, a novel ensemble approach that improves domain adaptation in AI-generated text detection, outperforming larger models on unseen domains.

Findings

Achieves state-of-the-art in-domain detection performance.

Outperforms larger models on out-of-domain detection.

Provides open-source code and models for future research.

Abstract

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.