Adversarial attacks and defenses for large language models (LLMs): methods, frameworks & challenges

被引:1
|
作者
Kumar, Pranjal [1 ]
机构
[1] Lovely Profess Univ, Sch Comp Sci & Engn, Dept Intelligent Syst, Phagwara 144411, Punjab, India
关键词
Adversarial attacks; Artificial intelligence; Natural language processing; Machine learning; Neural networks; Large language models; ChatGPT; GPT; COMPUTER VISION; EXAMPLES;
D O I
10.1007/s13735-024-00334-8
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Large language models (LLMs) have exhibited remarkable efficacy and proficiency in a wide array of NLP endeavors. Nevertheless, concerns are growing rapidly regarding the security and vulnerabilities linked to the adoption and incorporation of LLM. In this work, a systematic study focused on the most up-to-date attack and defense frameworks for the LLM is presented. This work delves into the intricate landscape of adversarial attacks on language models (LMs) and presents a thorough problem formulation. It covers a spectrum of attack enhancement techniques and also addresses methods for strengthening LLMs. This study also highlights challenges in the field, such as the assessment of offensive or defensive performance, defense and attack transferability, high computational requirements, embedding space size, and perturbation. This survey encompasses more than 200 recent papers concerning adversarial attacks and techniques. By synthesizing a broad array of attack techniques, defenses, and challenges, this paper contributes to the ongoing discourse on securing LM against adversarial threats.
引用
收藏
页数:28
相关论文
共 50 条
  • [21] A Critical Review of Methods and Challenges in Large Language Models
    Moradi, Milad
    Yan, Ke
    Colwell, David
    Samwald, Matthias
    Asgari, Rhona
    CMC-COMPUTERS MATERIALS & CONTINUA, 2025, 82 (02): : 1681 - 1698
  • [22] Recommender Systems in the Era of Large Language Models (LLMs)
    Zhao, Zihuai
    Fan, Wenqi
    Li, Jiatong
    Liu, Yunqing
    Mei, Xiaowei
    Wang, Yiqi
    Wen, Zhen
    Wang, Fei
    Zhao, Xiangyu
    Tang, Jiliang
    Li, Qing
    IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2024, 36 (11) : 6889 - 6907
  • [23] Large language models (LLMs) as agents for augmented democracy
    Gudino, Jairo F.
    Grandi, Umberto
    Hidalgo, Cesar
    PHILOSOPHICAL TRANSACTIONS OF THE ROYAL SOCIETY A-MATHEMATICAL PHYSICAL AND ENGINEERING SCIENCES, 2024, 382 (2285):
  • [24] Are Large Language Models (LLMs) Ready for Agricultural Applications?
    Shende, Ketan
    Resource: Engineering and Technology for Sustainable World, 2025, 32 (01): : 28 - 30
  • [25] A survey on augmenting knowledge graphs (KGs) with large language models (LLMs): models, evaluation metrics, benchmarks, and challenges
    Ibrahim, Nourhan
    Aboulela, Samar
    Ibrahim, Ahmed
    Kashef, Rasha
    Discover Artificial Intelligence, 2024, 4 (01):
  • [26] Computing Architecture for Large-Language Models (LLMs) and Large Multimodal Models (LMMs)
    Liang, Bor-Sung
    PROCEEDINGS OF THE 2024 INTERNATIONAL SYMPOSIUM ON PHYSICAL DESIGN, ISPD 2024, 2024, : 233 - 234
  • [27] Context is everything in regulatory application of large language models (LLMs)
    Tong, Weida
    Renaudin, Michael
    DRUG DISCOVERY TODAY, 2024, 29 (04)
  • [28] Operating Conversational Large Language Models (LLMs)in the Presenceof Errors
    Gao, Zhen
    Deng, Jie
    Reviriego, Pedro
    Liu, Shanshan
    Pozo, Alejando
    Lombardi, Fabrizio
    IEEE NANOTECHNOLOGY MAGAZINE, 2025, 19 (01) : 31 - 37
  • [29] A Survey on the Use of Large Language Models (LLMs) in Fake News
    Papageorgiou, Eleftheria
    Chronis, Christos
    Varlamis, Iraklis
    Himeur, Yassine
    FUTURE INTERNET, 2024, 16 (08)
  • [30] Addressing digital inequities in the age of large language models (LLMs)
    Ng, Olivia
    Han, Siew Ping
    MEDICAL EDUCATION, 2024, 58 (12) : 1545 - 1546