# Shaping the Safety Boundaries: Understanding and Defending Against Jailbreaks in Large Language Models

arXiv: https://arxiv.org/abs/2412.17034 · HTML: https://mlsanity.com/paper/2412.17034/

## Key facts

- Cited by 23 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 58 references that are in the map.
- First posted on arXiv 22 Dec 2024.

## Placed in

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

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

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [Optuna: A Next-generation Hyperparameter Optimization Framework](https://mlsanity.com/paper/1907.10902/) (2019)
- [Jailbreaking Black Box Large Language Models in Twenty Queries](https://mlsanity.com/paper/2310.08419/) (2023)
- [Baseline Defenses for Adversarial Attacks Against Aligned Language Models](https://mlsanity.com/paper/2309.00614/) (2023)
- [SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks](https://mlsanity.com/paper/2310.03684/) (2023)
- [Detecting Language Model Attacks with Perplexity](https://mlsanity.com/paper/2308.14132/) (2023)
- [Justice or Prejudice? Quantifying Biases in LLM-as-a-Judge](https://mlsanity.com/paper/2410.02736/) (2024)
- [DeepInception: Hypnotize Large Language Model to Be Jailbreaker](https://mlsanity.com/paper/2311.03191/) (2023)
- [A Wolf in Sheep's Clothing: Generalized Nested Jailbreak Prompts can Fool Large Language Models Easily](https://mlsanity.com/paper/2311.08268/) (2023)
- [The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learning](https://mlsanity.com/paper/2312.01552/) (2023)
- [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)
- [ArtPrompt: ASCII Art-based Jailbreak Attacks against Aligned LLMs](https://mlsanity.com/paper/2402.11753/) (2024)
- [COLD-Attack: Jailbreaking LLMs with Stealthiness and Controllability](https://mlsanity.com/paper/2402.08679/) (2024)
- [How Alignment and Jailbreak Work: Explain LLM Safety through Intermediate Hidden States](https://mlsanity.com/paper/2406.05644/) (2024)
- [GradSafe: Detecting Jailbreak Prompts for LLMs via Safety-Critical Gradient Analysis](https://mlsanity.com/paper/2402.13494/) (2024)
- [JailbreakRadar: Comprehensive Assessment of Jailbreak Attacks Against LLMs](https://mlsanity.com/paper/2402.05668/) (2024)
- [Defending Large Language Models Against Jailbreak Attacks via Layer-specific Editing](https://mlsanity.com/paper/2405.18166/) (2024)
- [Intention Analysis Makes LLMs A Good Jailbreak Defender](https://mlsanity.com/paper/2401.06561/) (2024)
- [Towards Understanding Jailbreak Attacks in LLMs: A Representation Space Analysis](https://mlsanity.com/paper/2406.10794/) (2024)
- [Multi-Turn Context Jailbreak Attack on Large Language Models From First Principles](https://mlsanity.com/paper/2408.04686/) (2024)
- [Defending Jailbreak Prompts via In-Context Adversarial Game](https://mlsanity.com/paper/2402.13148/) (2024)
- [JailbreakLens: Interpreting Jailbreak Mechanism in the Lens of Representation and Circuit](https://mlsanity.com/paper/2411.11114/) (2024)

## Built on by (most cited first, 23 of 23)

- [Reasoning as an Adaptive Defense for Safety](https://mlsanity.com/paper/2507.00971/) (2025)
- [Audio Jailbreak: An Open Comprehensive Benchmark for Jailbreaking Large Audio-Language Models](https://mlsanity.com/paper/2505.15406/) (2025)
- [On Almost Surely Safe Alignment of Large Language Models at Inference-Time](https://mlsanity.com/paper/2502.01208/) (2025)
- [Mitigating the Safety Alignment Tax with Null-Space Constrained Policy Optimization](https://mlsanity.com/paper/2512.11391/) (2025)
- [Analysing the Safety Pitfalls of Steering Vectors](https://mlsanity.com/paper/2603.24543/) (2026)
- [Activation Approximations Can Incur Safety Vulnerabilities Even in Aligned LLMs: Comprehensive Analysis and Defense](https://mlsanity.com/paper/2502.00840/) (2025)
- [The Cylindrical Representation Hypothesis for Language Model Steering](https://mlsanity.com/paper/2605.01844/) (2026)
- [Probing the Robustness of Large Language Models Safety to Latent Perturbations](https://mlsanity.com/paper/2506.16078/) (2025)
- [MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks](https://mlsanity.com/paper/2605.11002/) (2026)
- [Test-Time Safety Alignment](https://mlsanity.com/paper/2604.26167/) (2026)
- [LLM in the Middle: A Systematic Review of Threats and Mitigations to Real-World LLM-based Systems](https://mlsanity.com/paper/2509.10682/) (2025)
- [Eyes-on-Me: Scalable RAG Poisoning through Transferable Attention-Steering Attractors](https://mlsanity.com/paper/2510.00586/) (2025)
- [GUARD-SLM: Token Activation-Based Defense Against Jailbreak Attacks for Small Language Models](https://mlsanity.com/paper/2603.28817/) (2026)
- [D-Judge: Disrupting Multi-Turn Jailbreaks using Semantics-Preserving Output Rewriting](https://mlsanity.com/paper/2606.02640/) (2026)
- [SAID: Safety-Aware Intent Defense via Prefix Probing for Large Language Models](https://mlsanity.com/paper/2510.20129/) (2025)
- [Mitigating Many-shot Jailbreak Attacks with One Single Demonstration](https://mlsanity.com/paper/2605.08277/) (2026)
- [Why Do Aligned LLMs Remain Jailbreakable: Refusal-Escape Directions, Operator-Level Sources, and Safety-Utility Trade-off](https://mlsanity.com/paper/2605.08878/) (2026)
- [EVA: Editing for Versatile Alignment against Jailbreaks](https://mlsanity.com/paper/2605.14750/) (2026)
- [Babel: Jailbreaking Safety Attention via Obfuscation Distribution Optimized Sampling](https://mlsanity.com/paper/2605.17971/) (2026)
- [SafeSpec: Fast and Safe LLM via Dynamic Reflective Sampling](https://mlsanity.com/paper/2606.19755/) (2026)
- [Security and Privacy in Retrieval-Augmented Generation: Architectures, Threats, Defenses, and Future Directions for Building Trustworthy Systems](https://mlsanity.com/paper/2606.25533/) (2026)
- [Moving the Safety Barrier: Dynamic Routing Adaptive Alignment Against White-Box Attacks](https://mlsanity.com/paper/2608.02674/) (2026)
- [Safety Cost of Steering Vectors Is Separable and Reducible](https://mlsanity.com/paper/2608.08383/) (2026)

## Abstract (arXiv)

Jailbreaking in Large Language Models (LLMs) is a major security concern as it can deceive LLMs to generate harmful text. Yet, there is still insufficient understanding of how jailbreaking works, which makes it hard to develop effective defense strategies. We aim to shed more light into this issue: we conduct a detailed large-scale analysis of seven different jailbreak methods and find that these disagreements stem from insufficient observation samples. In particular, we introduce textitsafety boundary, and we find that jailbreaks shift harmful activations outside that safety boundary, where LLMs are less sensitive to harmful information. We also find that the low and the middle layers are critical in such shifts, while deeper layers have less impact. Leveraging on these insights, we propose a novel defense called Activation Boundary Defense (ABD), which adaptively constrains the activations within the safety boundary. We further use Bayesian optimization to selectively apply the defense method to the low and the middle layers. Our experiments on several benchmarks show that ABD achieves an average DSR of over 98% against various forms of jailbreak attacks, with less than 2% impact on the model's general capabilities.