MAGE: Machine-generated Text Detection in the Wild
This addresses the need for AI-generated text detection to mitigate risks like fake news and plagiarism, but it is incremental as it builds on existing methods with a more comprehensive testbed.
The paper tackles the problem of detecting machine-generated text in real-world scenarios where sources are unknown, finding that distinguishing such text from human writing is challenging, especially out-of-distribution, but a top detector achieves 86.54% accuracy on out-of-domain texts from a new LLM.
Large language models (LLMs) have achieved human-level text generation, emphasizing the need for effective AI-generated text detection to mitigate risks like the spread of fake news and plagiarism. Existing research has been constrained by evaluating detection methods on specific domains or particular language models. In practical scenarios, however, the detector faces texts from various domains or LLMs without knowing their sources. To this end, we build a comprehensive testbed by gathering texts from diverse human writings and texts generated by different LLMs. Empirical results show challenges in distinguishing machine-generated texts from human-authored ones across various scenarios, especially out-of-distribution. These challenges are due to the decreasing linguistic distinctions between the two sources. Despite challenges, the top-performing detector can identify 86.54% out-of-domain texts generated by a new LLM, indicating the feasibility for application scenarios. We release our resources at https://github.com/yafuly/MAGE.