Special-Character Adversarial Attacks on Open-Source Language Model
Large language models (LLMs) have achieved remarkable performance across diverse natural language processing tasks, yet their vulnerability to character-level adversarial manipulations presents significant security challenges for real-world deployments. This paper presents a study of different special character attacks including unicode, homoglyph, structural, and textual encoding attacks aimed at bypassing safety mechanisms. We evaluate seven prominent open-source models ranging from 3.8B to 32B parameters on 4,000+ attack attempts. These experiments reveal critical vulnerabilities across all model sizes, exposing failure modes that include successful jailbreaks, incoherent outputs, and unrelated hallucinations.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Every Character Counts: From Vulnerability to Defense in Phishing Detection
Phishing attacks targeting both organizations and individuals are becoming an increasingly significant threat as technology advances. Current automatic detection methods often lack explainability and robustness in detect…
Bluff: Interactively Deciphering Adversarial Attacks on Deep Neural Networks
Deep neural networks (DNNs) are now commonly used in many domains. However, they are vulnerable to adversarial attacks: carefully crafted perturbations on data inputs that can fool a model into making incorrect predictio…
Unsourced Adversarial CAPTCHA: A Bi-Phase Adversarial CAPTCHA Framework
With the rapid advancements in deep learning, traditional CAPTCHA schemes are increasingly vulnerable to automated attacks powered by deep neural networks (DNNs). Existing adversarial attack methods often rely on origina…
Adversarial AttackDiversityLanguage ModelingLanguage Modelling+1Adversarial Examples in the Physical World: A Survey
Deep neural networks (DNNs) have demonstrated high vulnerability to adversarial examples, raising broad security concerns about their applications. Besides the attacks in the digital world, the practical implications of …
SurveyAdversariaLib: An Open-source Library for the Security Evaluation of Machine Learning Algorithms Under Attack
We present AdversariaLib, an open-source python library for the security evaluation of machine learning (ML) against carefully-targeted attacks. It supports the implementation of several attacks proposed thus far in the …
BIG-bench Machine LearningGeneral Classification