AI Generated Text
AI-generated text detection focuses on distinguishing computer-generated text from human-written content, primarily to combat misinformation and academic dishonesty. Current research emphasizes developing robust detection methods, often employing deep learning architectures like transformers, and exploring techniques to improve the accuracy and generalizability of these methods across various languages and domains, including evaluating the effectiveness of different features like stylometry and semantic analysis. The ability to reliably detect AI-generated text is crucial for maintaining the integrity of scientific research, ensuring the authenticity of online information, and addressing ethical concerns surrounding the use of large language models.
Papers
Deepfake tweets automatic detection
Adam Frej, Adrian Kaminski, Piotr Marciniak, Szymon Szmajdzinski, Soveatin Kuntur, Anna Wroblewska
Investigating the Influence of Prompt-Specific Shortcuts in AI Generated Text Detection
Choonghyun Park, Hyuhng Joon Kim, Junyeob Kim, Youna Kim, Taeuk Kim, Hyunsoo Cho, Hwiyeol Jo, Sang-goo Lee, Kang Min Yoo