Recent advances in generative models for language have enabled the creation of convincing synthetic text or deepfake text. Prior work has demonstrated the potential for misuse of deepfake text to mislead content consumers. Therefore, deepfake text detection, the task of discriminating between human and machine-generated text, is becoming increasingly critical. Several defenses have been proposed for deepfake text detection. However, we lack a thorough understanding of their real-world applicability. In this paper, we collect deepfake text from 4 online services powered by Transformer-based tools to evaluate the generalization ability of the defenses on content in the wild. We develop several low-cost adversarial attacks, and investigate the robustness of existing defenses against an adaptive attacker. We find that many defenses show significant degradation in performance under our evaluation scenarios compared to their original claimed performance. Our evaluation shows that tapping into the semantic information in the text content is a promising approach for improving the robustness and generalization performance of deepfake text detection schemes.
How Generalizable are Deepfake Image Detectors? An Empirical Study
School of Computing and Information Systems, Singapore Management University(新加坡管理大学计算与信息系统学院);College of Computer Science and Technology, Harbin Engineering University(哈尔滨工程大学计算机科学与技术学院)
ASVspoof 2021: Towards Spoofed and Deepfake Speech Detection in the Wild
Université de Lorraine(洛林大学);CNRS(法国国家科学研究中心);Inria(法国国家信息与自动化研究所);LORIA(洛林计算机科学及其应用实验室);National Institute of Informatics(国立信息学研究所);Institute for Advancing Intelligence, TCG CREST(TCG CREST 智能推进研究所);EURECOM(欧洲通信学院);Cerence Inc.(赛轮思公司);Avignon Université(阿维尼翁大学);Nuance Communications(Nuance 通信公司);University of Eastern Finland(东芬兰大学)
Testing Storage-System Correctness: Challenges, Fuzzing Limitations, and AI-Augmented Opportunities
Institute of Computing Technology, Chinese Academy of Sciences (ICT, CAS)(中国科学院计算技术研究所);University of Chinese Academy of Sciences(中国科学院大学);Zhongguancun Laboratory(中关村实验室)