# xJailbreak: Representation Space Guided Reinforcement Learning for Interpretable LLM Jailbreaking

arXiv: https://arxiv.org/abs/2501.16727 · HTML: https://mlsanity.com/paper/2501.16727/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in Safety alignment and jailbreak attacks, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 21 of its 29 references that are in the map.
- First posted on arXiv 28 Jan 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Safety alignment and jailbreak attacks](https://mlsanity.com/alignment/safety-alignment-and-jailbreak-attacks/)
- LLM safety and security / [Jailbreaking](https://mlsanity.com/safety/jailbreaking/)

## Builds on (most cited first, 21 of 21)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [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)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Jailbroken: How Does LLM Safety Training Fail?](https://mlsanity.com/paper/2307.02483/) (2023)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [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)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Prompts](https://mlsanity.com/paper/2309.10253/) (2023)
- [Safety Alignment Should Be Made More Than Just a Few Tokens Deep](https://mlsanity.com/paper/2406.05946/) (2024)
- [Catastrophic Jailbreak of Open-source LLMs via Exploiting Generation](https://mlsanity.com/paper/2310.06987/) (2023)
- [GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher](https://mlsanity.com/paper/2308.06463/) (2023)
- [AmpleGCG: Learning a Universal and Transferable Generative Model of Adversarial Suffixes for Jailbreaking Both Open and Closed LLMs](https://mlsanity.com/paper/2404.07921/) (2024)
- [RL-JACK: Reinforcement Learning-powered Black-box Jailbreaking Attack against LLMs](https://mlsanity.com/paper/2406.08725/) (2024)
- [Constrain Alignment with Sparse Autoencoders](https://mlsanity.com/paper/2411.07618/) (2024)
- [PathSeeker: Exploring LLM Security Vulnerabilities with a Reinforcement Learning-Based Jailbreak Approach](https://mlsanity.com/paper/2409.14177/) (2024)
- [Quantification of Large Language Model Distillation](https://mlsanity.com/paper/2501.12619/) (2025)

## Built on by (most cited first, 11 of 11)

- [Refusal Falls off a Cliff: How Safety Alignment Fails in Reasoning?](https://mlsanity.com/paper/2510.06036/) (2025)
- [Using Mechanistic Interpretability to Craft Adversarial Attacks against Large Language Models](https://mlsanity.com/paper/2503.06269/) (2025)
- [Probing the Difficulty Perception Mechanism of Large Language Models](https://mlsanity.com/paper/2510.05969/) (2025)
- [Jailbreaking LLMs & VLMs: Mechanisms, Evaluation, and Unified Defense](https://mlsanity.com/paper/2601.03594/) (2026)
- [Your Mailbox Is Mine: Prompt Injection Attacks Against Real-World LLM Email Agents](https://mlsanity.com/paper/2507.02699/) (2025)
- [Scaling Patterns in Adversarial Alignment: Evidence from Multi-LLM Jailbreak Experiments](https://mlsanity.com/paper/2511.13788/) (2025)
- [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)
- [TrailBlazer: History-Guided Reinforcement Learning for Black-Box LLM Jailbreaking](https://mlsanity.com/paper/2602.06440/) (2026)
- [Safety Geometry Collapse in Multimodal LLMs and Adaptive Drift Correction](https://mlsanity.com/paper/2605.18104/) (2026)
- [Can MLLMs Detect Phishing? A Comprehensive Security Benchmark Suite Focusing on Dynamic Threats and Multimodal Evaluation in Academic Environments](https://mlsanity.com/paper/2511.15165/) (2025)

## Abstract (arXiv)

Safety alignment mechanism are essential for preventing large language models (LLMs) from generating harmful information or unethical content. However, cleverly crafted prompts can bypass these safety measures without accessing the model's internal parameters, a phenomenon known as black-box jailbreak. Existing heuristic black-box attack methods, such as genetic algorithms, suffer from limited effectiveness due to their inherent randomness, while recent reinforcement learning (RL) based methods often lack robust and informative reward signals. To address these challenges, we propose a novel black-box jailbreak method leveraging RL, which optimizes prompt generation by analyzing the embedding proximity between benign and malicious prompts. This approach ensures that the rewritten prompts closely align with the intent of the original prompts while enhancing the attack's effectiveness. Furthermore, we introduce a comprehensive jailbreak evaluation framework incorporating keywords, intent matching, and answer validation to provide a more rigorous and holistic assessment of jailbreak success. Experimental results show the superiority of our approach, achieving state-of-the-art (SOTA) performance on several prominent open and closed-source LLMs, including Qwen2.5-7B-Instruct, Llama3.1-8B-Instruct, and GPT-4o-0806. Our method sets a new benchmark in jailbreak attack effectiveness, highlighting potential vulnerabilities in LLMs. The codebase for this work is available at https://github.com/Aegis1863/xJailbreak.