GenAI Content Detection Task 3: Cross-Domain Machine-Generated Text Detection Challenge

TL;DR

This paper presents a shared task on detecting machine-generated text across multiple domains using the RAID benchmark, demonstrating high accuracy and robustness of detectors trained on seen domains and models.

Contribution

It introduces a new cross-domain detection challenge with the RAID benchmark and evaluates the effectiveness of current detectors in this setting.

Findings

Detectors achieved over 99% accuracy on RAID data.

High robustness of detectors across multiple domains and models.

Low false positive rate of 5% maintained.

Abstract

Recently there have been many shared tasks targeting the detection of generated text from Large Language Models (LLMs). However, these shared tasks tend to focus either on cases where text is limited to one particular domain or cases where text can be from many domains, some of which may not be seen during test time. In this shared task, using the newly released RAID benchmark, we aim to answer whether or not models can detect generated text from a large, yet fixed, number of domains and LLMs, all of which are seen during training. Over the course of three months, our task was attempted by 9 teams with 23 detector submissions. We find that multiple participants were able to obtain accuracies of over 99% on machine-generated text from RAID while maintaining a 5% False Positive Rate -- suggesting that detectors are able to robustly detect text from many domains and models simultaneously. We discuss potential interpretations of this result and provide directions for future research.