Adversarial Attacks on Large Language Model-Based System and Mitigating Strategies: A Case Study on ChatGPT
Security and Communication Networks · 10.06.2023 · 2023 · 1-10
DOI: 10.1155/2023/8691095 ↗Kurz gefasst
Fachartikel in der Zeitschrift Security and Communication Networks, erschienen am 10.06.2023. Verfasst von Bowen Liu und 5 weiteren Autor:innen. Die Arbeit wurde bisher 39-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Guangdong Laboratory of Artificial Intelligence Digital Economy, Dou Wanchun Expert Workstation of Yunnan Province.
Abstract (Originalsprache)
Machine learning algorithms are at the forefront of the development of advanced information systems. The rapid progress in machine learning technology has enabled cutting-edge large language models (LLMs), represented by GPT-3 and ChatGPT, to perform a wide range of NLP tasks with a stunning performance. However, research on adversarial machine learning highlights the need for these intelligent systems to be more robust. Adversarial machine learning aims to evaluate attack and defense mechanisms to prevent the malicious exploitation of these systems. In the case of ChatGPT, adversarial induction prompt can cause the model to generate toxic texts that could pose serious security risks or propagate false information. To address this challenge, we first analyze the effectiveness of inducing attacks on ChatGPT. Then, two effective mitigating mechanisms are proposed. The first is a training-free prefix prompt mechanism to detect and prevent the generation of toxic texts. The second is a RoBERTa-based mechanism that identifies manipulative or misleading input text via external detection models. The availability of this method is demonstrated through experiments.
Förderung
- Guangdong Laboratory of Artificial Intelligence Digital Economy
- Dou Wanchun Expert Workstation of Yunnan Province
Zitieren
APA
Liu, B., Xiao, B., Jiang, X., Cen, S., He, X., & Dou, W. (2023). Adversarial Attacks on Large Language Model-Based System and Mitigating Strategies: A Case Study on ChatGPT. Security and Communication Networks, 2023, 1-10. https://doi.org/10.1155/2023/8691095
Vancouver
Liu B, Xiao B, Jiang X, Cen S, He X, Dou W. Adversarial Attacks on Large Language Model-Based System and Mitigating Strategies: A Case Study on ChatGPT. Security and Communication Networks. 2023;2023:1-10. doi:10.1155/2023/8691095
BibTeX
@article{Liu2023,
author = {Liu, Bowen and Xiao, Boao and Jiang, Xutong and Cen, Siyuan and He, Xin and Dou, Wanchun},
title = {Adversarial Attacks on Large Language Model-Based System and Mitigating Strategies: A Case Study on ChatGPT},
journal = {Security and Communication Networks},
year = {2023},
volume = {2023},
pages = {1--10},
doi = {10.1155/2023/8691095}
}
EUVIMED