# Understanding and Enhancing the Transferability of Jailbreaking Attacks

arXiv: https://arxiv.org/abs/2502.03052 · HTML: https://mlsanity.com/paper/2502.03052/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- Placed in Jailbreaking, one of 12 directions in LLM safety and security.
- Builds on 52 of its 67 references that are in the map.
- First posted on arXiv 5 Feb 2025.

## Placed in

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

## Builds on (most cited first, 30 of 52)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [Jailbreaking Black Box Large Language Models in Twenty Queries](https://mlsanity.com/paper/2310.08419/) (2023)
- [Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations](https://mlsanity.com/paper/2312.06674/) (2023)
- [Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!](https://mlsanity.com/paper/2310.03693/) (2023)
- [Red Teaming Language Models with Language Models](https://mlsanity.com/paper/2202.03286/) (2022)
- [AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models](https://mlsanity.com/paper/2310.04451/) (2023)
- [Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned](https://mlsanity.com/paper/2209.07858/) (2022)
- [Tree of Attacks: Jailbreaking Black-Box LLMs Automatically](https://mlsanity.com/paper/2312.02119/) (2023)
- [Baseline Defenses for Adversarial Attacks Against Aligned Language Models](https://mlsanity.com/paper/2309.00614/) (2023)
- [How Johnny Can Persuade LLMs to Jailbreak Them: Rethinking Persuasion to Challenge AI Safety by Humanizing LLMs](https://mlsanity.com/paper/2401.06373/) (2024)
- [GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Prompts](https://mlsanity.com/paper/2309.10253/) (2023)
- [JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models](https://mlsanity.com/paper/2404.01318/) (2024)
- [Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study](https://mlsanity.com/paper/2305.13860/) (2023)
- [SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks](https://mlsanity.com/paper/2310.03684/) (2023)
- [Catastrophic Jailbreak of Open-source LLMs via Exploiting Generation](https://mlsanity.com/paper/2310.06987/) (2023)
- [Detecting Language Model Attacks with Perplexity](https://mlsanity.com/paper/2308.14132/) (2023)
- [DeepInception: Hypnotize Large Language Model to Be Jailbreaker](https://mlsanity.com/paper/2311.03191/) (2023)
- [A Holistic Approach to Undesired Content Detection in the Real World](https://mlsanity.com/paper/2208.03274/) (2022)
- [Low-Resource Languages Jailbreak GPT-4](https://mlsanity.com/paper/2310.02446/) (2023)
- [Shadow Alignment: The Ease of Subverting Safely-Aligned Language Models](https://mlsanity.com/paper/2310.02949/) (2023)

## Built on by (most cited first, 22 of 22)

- [We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems](https://mlsanity.com/paper/2506.13666/) (2025)
- [FORCE: Transferable Visual Jailbreaking Attacks via Feature Over-Reliance CorrEction](https://mlsanity.com/paper/2509.21029/) (2025)
- [Jailbreaking the Non-Transferable Barrier via Test-Time Data Disguising](https://mlsanity.com/paper/2503.17198/) (2025)
- [NEXUS: Network Exploration for eXploiting Unsafe Sequences in Multi-Turn LLM Jailbreaks](https://mlsanity.com/paper/2510.03417/) (2025)
- [Is Gradient Ascent Really Necessary? Memorize to Forget for Machine Unlearning](https://mlsanity.com/paper/2602.06441/) (2026)
- [Enhancing Instruction Following of LLMs via Activation Steering with Dynamic Rejection](https://mlsanity.com/paper/2603.06745/) (2026)
- [RouteHijack: Routing-Aware Attack on Mixture-of-Experts LLMs](https://mlsanity.com/paper/2605.02946/) (2026)
- [Jailbreak Distillation: Renewable Safety Benchmarking](https://mlsanity.com/paper/2505.22037/) (2025)
- [sudoLLM: On Multi-role Alignment of Language Models](https://mlsanity.com/paper/2505.14607/) (2025)
- [MetaCipher: A Time-Persistent and Universal Multi-Agent Framework for Cipher-Based Jailbreak Attacks for LLMs](https://mlsanity.com/paper/2506.22557/) (2025)
- [The Tower of Babel Revisited: Multilingual Jailbreak Prompts on Closed-Source Large Language Models](https://mlsanity.com/paper/2505.12287/) (2025)
- [Jinx: Unlimited LLMs for Probing Alignment Failures](https://mlsanity.com/paper/2508.08243/) (2025)
- [Exposing Long-Tail Safety Failures in Large Language Models through Efficient Diverse Response Sampling](https://mlsanity.com/paper/2603.14355/) (2026)
- [Persuasion Attacks Can Decrease Effectiveness of CoT Monitoring](https://mlsanity.com/paper/2607.08066/) (2026)
- [Exploring the Secondary Risks of Large Language Models](https://mlsanity.com/paper/2506.12382/) (2025)
- [SoK: A Comprehensive Security Analysis of Jailbreak Resilience in GPT and DeepSeek Models](https://mlsanity.com/paper/2506.18543/) (2025)
- [Activation-Guided Local Editing for Jailbreaking Attacks](https://mlsanity.com/paper/2508.00555/) (2025)
- [One Leak Away: How Pretrained Model Exposure Amplifies Jailbreak Risks in Finetuned LLMs](https://mlsanity.com/paper/2512.14751/) (2025)
- [ShallowJail: Steering Jailbreaks against Large Language Models](https://mlsanity.com/paper/2602.07107/) (2026)
- [Model-Agnostic Lifelong LLM Safety via Externalized Attack-Defense Co-Evolution](https://mlsanity.com/paper/2605.13411/) (2026)
- [Babel: Jailbreaking Safety Attention via Obfuscation Distribution Optimized Sampling](https://mlsanity.com/paper/2605.17971/) (2026)
- [SAFEGuard: Detect Optimization-Based Jailbreak Attacks Through Harmful Semantic Analysis and Fluency Measurement](https://mlsanity.com/paper/2609.05850/) (2026)

## Abstract (arXiv)

Jailbreaking attacks can effectively manipulate open-source large language models (LLMs) to produce harmful responses. However, these attacks exhibit limited transferability, failing to disrupt proprietary LLMs consistently. To reliably identify vulnerabilities in proprietary LLMs, this work investigates the transferability of jailbreaking attacks by analysing their impact on the model's intent perception. By incorporating adversarial sequences, these attacks can redirect the source LLM's focus away from malicious-intent tokens in the original input, thereby obstructing the model's intent recognition and eliciting harmful responses. Nevertheless, these adversarial sequences fail to mislead the target LLM's intent perception, allowing the target LLM to refocus on malicious-intent tokens and abstain from responding. Our analysis further reveals the inherent distributional dependency within the generated adversarial sequences, whose effectiveness stems from overfitting the source LLM's parameters, resulting in limited transferability to target LLMs. To this end, we propose the Perceived-importance Flatten (PiF) method, which uniformly disperses the model's focus across neutral-intent tokens in the original input, thus obscuring malicious-intent tokens without relying on overfitted adversarial sequences. Extensive experiments demonstrate that PiF provides an effective and efficient red-teaming evaluation for proprietary LLMs.