paper-with-me

Papers

Adaptive Attacks Break Defenses Against Indirect Prompt Injection Attacks on LLM Agents

2025-02-27 · Qiusi Zhan, Richard Fang, Henil Shalin Panchal, Daniel Kang

Large Language Model (LLM) agents exhibit remarkable performance across diverse applications by using external tools to interact with environments. However, integrating external tools introduces security risks, such as indirect prompt injection (IPI) attacks. Despite defenses designed for IPI attacks, their robustness remains questionable due to insufficient testing against adaptive attacks. In this paper, we evaluate eight different defenses and bypass all of them using adaptive attacks, consistently achieving an attack success rate of over 50%. This reveals critical vulnerabilities in current defenses. Our research underscores the need for adaptive attack evaluation when designing defenses to ensure robustness and reliability. The code is available at https://github.com/uiuc-kang-lab/AdaptiveAttackAgent.

📄 PDF Abstract BibTeX arXiv:2503.00061

Code (1)

uiuc-kang-lab/adaptiveattackagent 공식 구현 pytorch

Tasks

Language ModelingLanguage ModellingLarge Language Model

Similar Papers 제목 키워드 기반

Black-box, Adaptive, Efficient, Transferable, Harmful, Applicable... Attacks Are All You Need to Break LLMs

2026-06-02 · Vincent Limbach, Jonas Dornbusch, David Lüdke, Stephan Günnemann 외 arxiv

Accurately evaluating adversarial robustness is a longstanding challenge. A flawed attack design can inflate robustness estimates, making deployment risk assessment and defense comparison unreliable. Historically, standa…

Adversarial Robustness

The Attacker Moves Second: Stronger Adaptive Attacks Bypass Defenses Against Llm Jailbreaks and Prompt Injections

2025-10-10 · Milad Nasr, Nicholas Carlini, Chawin Sitawarin, Sander V. Schulhoff 외 arxiv

How should we evaluate the robustness of language model defenses? Current defenses against jailbreaks and prompt injections (which aim to prevent an attacker from eliciting harmful knowledge or remotely triggering malici…

Reinforcement Learning

Robust Prompt Optimization for Defending Language Models Against Jailbreaking Attacks

2024-01-30 · Andy Zhou, Bo Li, Haohan Wang

Despite advances in AI alignment, large language models (LLMs) remain vulnerable to adversarial attacks or jailbreaking, in which adversaries can modify prompts to induce unwanted behavior. While some defenses have been …

SelfDefend: LLMs Can Defend Themselves against Jailbreaking in a Practical Manner

2024-06-08 · Xunguang Wang, Daoyuan Wu, Zhenlan Ji, Zongjie Li 외

Jailbreaking is an emerging adversarial attack that bypasses the safety alignment deployed in off-the-shelf large language models (LLMs) and has evolved into multiple categories: human-based, optimization-based, generati…

Adversarial AttackLLM JailbreakSafety Alignment

One Step to the Side: Why Defenses Against Malicious Finetuning Fail Under Adaptive Adversaries

2026-05-14 · Itay Zloczower, Eyal Lenga, Gilad Gressel, Yisroel Mirsky arxiv

Model providers increasingly release open weights or allow users to fine-tune foundation models through APIs. Although these models are safety-aligned before release, their safeguards can often be removed by fine-tuning …