# Rewrite to Jailbreak: Discover Learnable and Transferable Implicit Harmfulness Instruction

arXiv: https://arxiv.org/abs/2502.11084 · HTML: https://mlsanity.com/paper/2502.11084/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Jailbreaking, one of 12 directions in LLM safety and security.
- First posted on arXiv 16 Feb 2025.

## Placed in

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

## Built on by (most cited first, 7 of 7)

- [Probe-Rewrite-Evaluate: A Workflow for Reliable Benchmarks and Quantifying Evaluation Awareness](https://mlsanity.com/paper/2509.00591/) (2025)
- [StealthGraph: Exposing Domain-Specific Risks in LLMs through Knowledge-Graph-Guided Harmful Prompt Generation](https://mlsanity.com/paper/2601.04740/) (2026)
- [Evolving Skill-Structured Attack Memory Enhances LLM Jailbreaking](https://mlsanity.com/paper/2605.29237/) (2026)
- [Jailbreaking LLMs via Semantically Relevant Nested Scenarios with Targeted Toxic Knowledge](https://mlsanity.com/paper/2510.01223/) (2025)
- [The Plot Twist: Jailbreaking Unified Multimodal Models with a Three-Act NarrativeAttack](https://mlsanity.com/paper/2509.26473/) (2025)
- [Overlooked Safety Vulnerability in LLMs: Malicious Intelligent Optimization Algorithm Request and its Jailbreak](https://mlsanity.com/paper/2601.00213/) (2026)
- [Automated jailbreak attack targeting multiple defense strategies](https://mlsanity.com/paper/2606.16751/) (2026)

## Abstract (arXiv)

As Large Language Models (LLMs) are widely applied in various domains, the safety of LLMs is increasingly attracting attention to avoid their powerful capabilities being misused. Existing jailbreak methods create a forced instruction-following scenario, or search adversarial prompts with prefix or suffix tokens to achieve a specific representation manually or automatically. However, they suffer from low efficiency and explicit jailbreak patterns, far from the real deployment of mass attacks to LLMs. In this paper, we point out that simply rewriting the original instruction can achieve a jailbreak, and we find that this rewriting approach is learnable and transferable. We propose the Rewrite to Jailbreak (R2J) approach, a transferable black-box jailbreak method to attack LLMs by iteratively exploring the weakness of the LLMs and automatically improving the attacking strategy. The jailbreak is more efficient and hard to identify since no additional features are introduced. Extensive experiments and analysis demonstrate the effectiveness of R2J, and we find that the jailbreak is also transferable to multiple datasets and various types of models with only a few queries. We hope our work motivates further investigation of LLM safety. The code can be found at https://github.com/ythuang02/R2J/.