# CHASE: Adversarial Red-Blue Teaming for Improving LLM Safety using Reinforcement Learning

arXiv: https://arxiv.org/abs/2606.05523 · HTML: https://mlsanity.com/paper/2606.05523/

## Key facts

- Cited by 3 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 19 of its 31 references that are in the map.
- First posted on arXiv 4 Jun 2026.

## 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 / [Prompt injection and agent security](https://mlsanity.com/safety/prompt-injection-and-agent-security/)

## Builds on (most cited first, 19 of 19)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [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)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Dataset](https://mlsanity.com/paper/2307.04657/) (2023)
- [Safe RLHF: Safe Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2310.12773/) (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)
- [Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks](https://mlsanity.com/paper/2404.02151/) (2024)
- [Safety Alignment Should Be Made More Than Just a Few Tokens Deep](https://mlsanity.com/paper/2406.05946/) (2024)
- [Great, Now Write an Article About That: The Crescendo Multi-Turn LLM Jailbreak Attack](https://mlsanity.com/paper/2404.01833/) (2024)
- [DeepInception: Hypnotize Large Language Model to Be Jailbreaker](https://mlsanity.com/paper/2311.03191/) (2023)
- [Low-Resource Languages Jailbreak GPT-4](https://mlsanity.com/paper/2310.02446/) (2023)
- [Scalable and Transferable Black-Box Jailbreaks for Language Models via Persona Modulation](https://mlsanity.com/paper/2311.03348/) (2023)
- [Mitigating the Alignment Tax of RLHF](https://mlsanity.com/paper/2309.06256/) (2023)
- [MART: Improving LLM Safety with Multi-round Automatic Red-Teaming](https://mlsanity.com/paper/2311.07689/) (2023)
- [Efficient Adversarial Training in LLMs with Continuous Attacks](https://mlsanity.com/paper/2405.15589/) (2024)
- [Panacea: Pareto Alignment via Preference Adaptation for LLMs](https://mlsanity.com/paper/2402.02030/) (2024)
- [MixAT: Combining Continuous and Discrete Adversarial Training for LLMs](https://mlsanity.com/paper/2505.16947/) (2025)

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

- [Agent Hacks Agent: Autoresearch for Production-Agent Red-Teaming](https://mlsanity.com/paper/2607.11698/) (2026)
- [Co-Evolution in Agentic Systems: Toward Self-Directed Evolution Beyond Human Design](https://mlsanity.com/paper/2608.10299/) (2026)
- [ReSI: Recursive Safety Improvement toward Resistant and Resilient AI](https://mlsanity.com/paper/2610.12233/) (2026)

## Abstract (arXiv)

Despite advances in safety alignment, prompt-rewriting attacks such as persona modulation, fictional framing and persuasion-based reformulation, can bypass safety filters even on frontier models. Existing defenses either rely on non-scalable human curation or white-box optimisation that overfits to specific model internals, leaving aligned models brittle against the very class of adaptive black-box adversaries they will face in deployment. To address this gap, we introduce CHASE (Co-evolutionary Hardening through Adversarial Safety-Escalation), a closed-loop red-blue teaming framework in which a black-box attacker and a safety-aligned defender co-evolve. The attacker is trained via Group Relative Policy Optimization (GRPO) under a multiplicative reward that jointly enforces bypass effectiveness and intent fidelity, while the defender is hardened on the harvested adversarial rewrites through a two-stage GRPO + rejection-sampled SFT pipeline balanced with benign data. Evaluated on BeaverTails and JailbreakBench against five held-out attack families (PAIR, TAP, AutoDAN, PAP, Translation), CHASE cuts mean StrongREJECT score by 43.2% with 0% false-refusal on benign prompts. Beyond the headline result, CHASE shows that template-free RL exploration recovers latent attack primitives that transfer across mechanistically distinct attack families, suggesting a path toward LLM safety hardening that generalises beyond the narrow distributions achieved thus far in adversarial training.