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

arXiv: https://arxiv.org/abs/2504.12321 · HTML: https://mlsanity.com/paper/2504.12321/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Jailbreaking, one of 12 directions in LLM safety and security.
- Builds on 24 of its 34 references that are in the map.
- First posted on arXiv 10 Apr 2025.

## Placed in

- LLM safety and security / [Jailbreaking](https://mlsanity.com/safety/jailbreaking/)

## Builds on (most cited first, 24 of 24)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [Jailbroken: How Does LLM Safety Training Fail?](https://mlsanity.com/paper/2307.02483/) (2023)
- [Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection](https://mlsanity.com/paper/2302.12173/) (2023)
- [Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!](https://mlsanity.com/paper/2310.03693/) (2023)
- [Ignore Previous Prompt: Attack Techniques For Language Models](https://mlsanity.com/paper/2211.09527/) (2022)
- [The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions](https://mlsanity.com/paper/2404.13208/) (2024)
- ["Do Anything Now": Characterizing and Evaluating In-The-Wild Jailbreak Prompts on Large Language Models](https://mlsanity.com/paper/2308.03825/) (2023)
- [A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications](https://mlsanity.com/paper/2402.07927/) (2024)
- [Safety-Tuned LLaMAs: Lessons From Improving the Safety of Large Language Models that Follow Instructions](https://mlsanity.com/paper/2309.07875/) (2023)
- [Generating Wikipedia by Summarizing Long Sequences](https://mlsanity.com/paper/1801.10198/) (2018)
- [TrustLLM: Trustworthiness in Large Language Models](https://mlsanity.com/paper/2401.05561/) (2024)
- [SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding](https://mlsanity.com/paper/2402.08983/) (2024)
- [LLM Self Defense: By Self Examination, LLMs Know They Are Being Tricked](https://mlsanity.com/paper/2308.07308/) (2023)
- [Removing RLHF Protections in GPT-4 via Fine-Tuning](https://mlsanity.com/paper/2311.05553/) (2023)
- [AutoDefense: Multi-Agent LLM Defense against Jailbreak Attacks](https://mlsanity.com/paper/2403.04783/) (2024)
- [Phi-3 Safety Post-Training: Aligning Language Models with a "Break-Fix" Cycle](https://mlsanity.com/paper/2407.13833/) (2024)
- [Creativity Has Left the Chat: The Price of Debiasing Language Models](https://mlsanity.com/paper/2406.05587/) (2024)
- [Decoding In-Context Learning: Neuroscience-inspired Analysis of Representations in Large Language Models](https://mlsanity.com/paper/2310.00313/) (2023)
- [Defending Large Language Models Against Attacks With Residual Stream Activation Analysis](https://mlsanity.com/paper/2406.03230/) (2024)

## Built on by (most cited first, 3 of 3)

- [Interpretation Meets Safety: A Survey on Interpretation Methods and Tools for Improving LLM Safety](https://mlsanity.com/paper/2506.05451/) (2025)
- [Metaphor Is Not All Attention Needs](https://mlsanity.com/paper/2605.12128/) (2026)
- [Measuring the Wrong Thing: Internal Harmfulness Scores Anti-Rank Successful Jailbreaks](https://mlsanity.com/paper/2608.09624/) (2026)

## Abstract (arXiv)

In the past few years, Language Models (LMs) have shown par-human capabilities in several domains. Despite their practical applications and exceeding user consumption, they are susceptible to jailbreaks when malicious input exploits the LM's weaknesses, causing it to deviate from its intended behavior. Current defensive strategies either classify the input prompt as adversarial or prevent LMs from generating harmful outputs. However, it is challenging to explain the reason behind the malicious nature of the jailbreak, which results in a wide variety of closed-box approaches. In this research, we propose and demonstrate that system-prompt attention from Small Language Models (SLMs) can be used to characterize adversarial prompts, providing a novel, explainable, and cheaper defense approach called AttentionDefense. Our research suggests that the attention mechanism is an integral component in understanding and explaining how LMs respond to malicious input that is not captured in the semantic meaning of text embeddings. The proposed AttentionDefense is evaluated against existing jailbreak benchmark datasets. Ablation studies show that SLM-based AttentionDefense has equivalent or better jailbreak detection performance compared to text embedding-based classifiers and GPT-4 zero-shot detectors.To further validate the efficacy of the proposed approach, we generate a dataset of novel jailbreak variants of the existing benchmark dataset using a closed-loop LLM-based multi-agent system. We demonstrate that the proposed AttentionDefense approach performs robustly on this novel jailbreak dataset while existing approaches suffer in performance. Additionally, for practical purposes AttentionDefense is an ideal solution as it has the computation requirements of a small LM but the performance of a LLM detector.