DetectRL-X: Towards Reliable Multilingual and Real-World LLM-Generated Text Detection

TL;DR

DetectRL-X is a new multilingual benchmark for evaluating the reliability of LLM-generated text detectors across diverse languages, domains, and real-world modifications.

Contribution

It introduces a comprehensive multilingual benchmark with diverse datasets, attack simulations, and evaluation protocols to assess detector robustness in real-world scenarios.

Findings

Current detectors show varied performance across languages and domains.

Domain, generator type, and attack strategies significantly affect detection accuracy.

DetectRL-X effectively reveals strengths and weaknesses of state-of-the-art detectors.

Abstract

The effective detection and governance of Large Language Model (LLM) generated content has become increasingly critical due to the growing risk of misuse. Despite the impressive performance of existing detectors, their reliability and potential in multilingual, real-world scenarios remain largely underexplored. In this study, we introduce DetectRL-X, a comprehensive multilingual benchmark designed to evaluate advanced detectors across 8 dimensions. The benchmark encompasses 8 languages commonly used in commercial contexts and collects human-written texts from 6 domains highly susceptible to LLM misuse. To better aligned with real-world applications, We create LLM-generated texts using 4 popular commercial LLMs, and include typical AI-assisted writing operations such as polishing, expanding, and condensing to capture authentic usage patterns. Furthermore, we develop a multilingual framework for paraphrasing and perturbation attacks to simulate diverse human modifications and writing noise, enabling stress testing of detectors across languages. Experimental results on DetectRL-X reveal the strengths and limitations of current state-of-the-art detectors when applied to diverse linguistic resources. We further analyze how domains, generators, attack strategies, text length, and refinement operations influence performance in different languages, underscoring DetectRL-X as an effective benchmark for strengthening multilingual and language-specific detectors.