DART: An AIGT Detector using AMR of Rephrased Text
This addresses the challenge of AIGT detection for practical applications, though it appears incremental as it builds on existing detection methods.
The paper tackles the problem of detecting AI-generated text from black-box LLMs in real-world scenarios where the origin is unknown, proposing DART, which achieved discrimination of multiple black-box LLMs without relying on probabilistic features.
As large language models (LLMs) generate more human-like texts, concerns about the side effects of AI-generated texts (AIGT) have grown. So, researchers have developed methods for detecting AIGT. However, two challenges remain. First, the performance of detecting black-box LLMs is low because existing models focus on probabilistic features. Second, most AIGT detectors have been tested on a single-candidate setting, which assumes that we know the origin of an AIGT and which may deviate from the real-world scenario. To resolve these challenges, we propose DART, which consists of four steps: rephrasing, semantic parsing, scoring, and multiclass classification. We conducted three experiments to test the performance of DART. The experimental result shows that DART can discriminate multiple black-box LLMs without probabilistic features and the origin of AIGT.