DetectRL: Benchmarking LLM-Generated Text Detection in Real-World Scenarios
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916650581753856 |
|---|---|
| author | Wu, Junchao Zhan, Runzhe Wong, Derek F. Yang, Shu Yang, Xinyi Yuan, Yulin Chao, Lidia S. |
| author_facet | Wu, Junchao Zhan, Runzhe Wong, Derek F. Yang, Shu Yang, Xinyi Yuan, Yulin Chao, Lidia S. |
| contents | Detecting text generated by large language models (LLMs) is of great recent interest. With zero-shot methods like DetectGPT, detection capabilities have reached impressive levels. However, the reliability of existing detectors in real-world applications remains underexplored. In this study, we present a new benchmark, DetectRL, highlighting that even state-of-the-art (SOTA) detection techniques still underperformed in this task. We collected human-written datasets from domains where LLMs are particularly prone to misuse. Using popular LLMs, we generated data that better aligns with real-world applications. Unlike previous studies, we employed heuristic rules to create adversarial LLM-generated text, simulating various prompts usages, human revisions like word substitutions, and writing noises like spelling mistakes. Our development of DetectRL reveals the strengths and limitations of current SOTA detectors. More importantly, we analyzed the potential impact of writing styles, model types, attack methods, the text lengths, and real-world human writing factors on different types of detectors. We believe DetectRL could serve as an effective benchmark for assessing detectors in real-world scenarios, evolving with advanced attack methods, thus providing more stressful evaluation to drive the development of more efficient detectors. Data and code are publicly available at: https://github.com/NLP2CT/DetectRL. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_23746 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | DetectRL: Benchmarking LLM-Generated Text Detection in Real-World Scenarios Wu, Junchao Zhan, Runzhe Wong, Derek F. Yang, Shu Yang, Xinyi Yuan, Yulin Chao, Lidia S. Computation and Language Artificial Intelligence Detecting text generated by large language models (LLMs) is of great recent interest. With zero-shot methods like DetectGPT, detection capabilities have reached impressive levels. However, the reliability of existing detectors in real-world applications remains underexplored. In this study, we present a new benchmark, DetectRL, highlighting that even state-of-the-art (SOTA) detection techniques still underperformed in this task. We collected human-written datasets from domains where LLMs are particularly prone to misuse. Using popular LLMs, we generated data that better aligns with real-world applications. Unlike previous studies, we employed heuristic rules to create adversarial LLM-generated text, simulating various prompts usages, human revisions like word substitutions, and writing noises like spelling mistakes. Our development of DetectRL reveals the strengths and limitations of current SOTA detectors. More importantly, we analyzed the potential impact of writing styles, model types, attack methods, the text lengths, and real-world human writing factors on different types of detectors. We believe DetectRL could serve as an effective benchmark for assessing detectors in real-world scenarios, evolving with advanced attack methods, thus providing more stressful evaluation to drive the development of more efficient detectors. Data and code are publicly available at: https://github.com/NLP2CT/DetectRL. |
| title | DetectRL: Benchmarking LLM-Generated Text Detection in Real-World Scenarios |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2410.23746 |