Cover Image

AttentionDefense: Leveraging System Prompt Attention for Explainable Defense Against Novel Jailbreaks

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Siska, Charlotte, Sankaran, Anush
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Attention Slipping: A Mechanistic Understanding of Jailbreak Attacks and Defenses in LLMs
by: Hu, Xiaomeng, et al.
Published: (2025)

Defense Against the Dark Prompts: Mitigating Best-of-N Jailbreaking with Prompt Evaluation
by: Armstrong, Stuart, et al.
Published: (2025)

A Systematic Literature Review on LLM Defenses Against Prompt Injection and Jailbreaking: Expanding NIST Taxonomy
by: Correia, Pedro H. Barcha, et al.
Published: (2026)

Jailbreak Attacks and Defenses Against Large Language Models: A Survey
by: Yi, Sibo, et al.
Published: (2024)

Provable Defense Framework for LLM Jailbreaks via Noise-Augumented Alignment
by: Cheng, Zehua, et al.
Published: (2026)

IPIGuard: A Novel Tool Dependency Graph-Based Defense Against Indirect Prompt Injection in LLM Agents
by: An, Hengyu, et al.
Published: (2025)

Test-Time Immunization: A Universal Defense Framework Against Jailbreaks for (Multimodal) Large Language Models
by: Yu, Yongcan, et al.
Published: (2025)

How Jailbreak Defenses Work and Ensemble? A Mechanistic Investigation
by: Long, Zhuohang, et al.
Published: (2025)

Alert-ME: An Explainability-Driven Defense Against Adversarial Examples in Transformer-Based Text Classification
by: Sabir, Bushra, et al.
Published: (2023)

How Few-shot Demonstrations Affect Prompt-based Defenses Against LLM Jailbreak Attacks
by: Wang, Yanshu, et al.
Published: (2026)

Defensive Dual Masking for Robust Adversarial Defense
by: Yang, Wangli, et al.
Published: (2024)

May I have your Attention? Breaking Fine-Tuning based Prompt Injection Defenses using Architecture-Aware Attacks
by: Pandya, Nishit V., et al.
Published: (2025)

ShieldLearner: A New Paradigm for Jailbreak Attack Defense in LLMs
by: Ni, Ziyi, et al.
Published: (2025)

A Graph-Enhanced Defense Framework for Explainable Fake News Detection with LLM
by: Wang, Bo, et al.
Published: (2026)

THRD: A Training-Free Multi-Turn Defense Framework for Jailbreak Attacks on Large Language Models
by: Ma, Zhiqing, et al.
Published: (2026)

An Embarrassingly Simple Defense Against LLM Abliteration Attacks
by: Shairah, Harethah Abu, et al.
Published: (2025)

Dual Attention Guided Defense Against Malicious Edits
by: Zhang, Jie, et al.
Published: (2025)

Jatmo: Prompt Injection Defense by Task-Specific Finetuning
by: Piet, Julien, et al.
Published: (2023)

Layer-Level Self-Exposure and Patch: Affirmative Token Mitigation for Jailbreak Attack Defense
by: Ouyang, Yang, et al.
Published: (2025)

DistillGuard: Evaluating Defenses Against LLM Knowledge Distillation
by: Jiang, Bo
Published: (2026)

Scalable Defense against In-the-wild Jailbreaking Attacks with Safety Context Retrieval
by: Chen, Taiye, et al.
Published: (2025)

Audio Jailbreaks in Large Audio-Language Models: Taxonomy, Attack-Defense Analysis, and Cost-Aware Evaluation
by: Feng, Bo-Han, et al.
Published: (2026)

Semantic Mirror Jailbreak: Genetic Algorithm Based Jailbreak Prompts Against Open-source LLMs
by: Li, Xiaoxia, et al.
Published: (2024)

Formalizing and Benchmarking Prompt Injection Attacks and Defenses
by: Liu, Yupei, et al.
Published: (2023)

DOGe: Defensive Output Generation for LLM Protection Against Knowledge Distillation
by: Li, Pingzhi, et al.
Published: (2025)

MetaDefense: Defending Finetuning-based Jailbreak Attack Before and During Generation
by: Jiang, Weisen, et al.
Published: (2025)

Improved Few-Shot Jailbreaking Can Circumvent Aligned Language Models and Their Defenses
by: Zheng, Xiaosen, et al.
Published: (2024)

Spectral Attention Steering for Prompt Highlighting
by: Li, Weixian Waylon, et al.
Published: (2026)

A Survey on Proactive Defense Strategies Against Misinformation in Large Language Models
by: Liu, Shuliang, et al.
Published: (2025)

Stop Tracking Me! Proactive Defense Against Attribute Inference Attack in LLMs
by: Yan, Dong, et al.
Published: (2026)

Defense Against Indirect Prompt Injection via Tool Result Parsing
by: Yu, Qiang, et al.
Published: (2026)

Semantics as a Shield: Label Disguise Defense (LDD) against Prompt Injection in LLM Sentiment Classification
by: Li, Yanxi, et al.
Published: (2025)

E$^2$AT: Multimodal Jailbreak Defense via Dynamic Joint Optimization for Multimodal Large Language Models
by: Lu, Liming, et al.
Published: (2025)

X-Teaming: Multi-Turn Jailbreaks and Defenses with Adaptive Multi-Agents
by: Rahman, Salman, et al.
Published: (2025)

Feint and Attack: Attention-Based Strategies for Jailbreaking and Protecting LLMs
by: Pu, Rui, et al.
Published: (2024)

FRACTURED-SORRY-Bench: Framework for Revealing Attacks in Conversational Turns Undermining Refusal Efficacy and Defenses over SORRY-Bench (Automated Multi-shot Jailbreaks)
by: Priyanshu, Aman, et al.
Published: (2024)

Multilingual Collaborative Defense for Large Language Models
by: Li, Hongliang, et al.
Published: (2025)

Adversarial Attacks and Defense for Conversation Entailment Task
by: Yang, Zhenning, et al.
Published: (2024)

Next-Gen CAPTCHAs: Leveraging the Cognitive Gap for Scalable and Diverse GUI-Agent Defense
by: Liu, Jiacheng, et al.
Published: (2026)

Round Attention: A Novel Round-Level Attention Mechanism to Accelerate LLM Inference
by: Tang, Yaohua, et al.
Published: (2025)