# ARES: Adaptive Red-Teaming and End-to-End Repair of Policy-Reward System

arXiv: https://arxiv.org/abs/2604.18789 · HTML: https://mlsanity.com/paper/2604.18789/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Safety alignment and jailbreak attacks, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 12 of its 59 references that are in the map.
- First posted on arXiv 20 Apr 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, 12 of 12)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [PKU-SafeRLHF: Towards Multi-Level Safety Alignment for LLMs with Human Preference](https://mlsanity.com/paper/2406.15513/) (2024)
- [MART: Improving LLM Safety with Multi-round Automatic Red-Teaming](https://mlsanity.com/paper/2311.07689/) (2023)
- [ALERT: A Comprehensive Benchmark for Assessing Large Language Models' Safety through Red Teaming](https://mlsanity.com/paper/2404.08676/) (2024)
- [FLIRT: Feedback Loop In-context Red Teaming](https://mlsanity.com/paper/2308.04265/) (2023)
- [A Mousetrap: Fooling Large Reasoning Models for Jailbreak with Chain of Iterative Chaos](https://mlsanity.com/paper/2502.15806/) (2025)
- [OpenAI's Approach to External Red Teaming for AI Models and Systems](https://mlsanity.com/paper/2503.16431/) (2025)
- [Ferret: Faster and Effective Automated Red Teaming with Reward-Based Scoring Technique](https://mlsanity.com/paper/2408.10701/) (2024)

## Abstract (arXiv)

Reinforcement Learning from Human Feedback (RLHF) is central to aligning Large Language Models (LLMs), yet it introduces a critical vulnerability: an imperfect Reward Model (RM) can become a single point of failure when it fails to penalize unsafe behaviors. While existing red-teaming approaches primarily target policy-level weaknesses, they overlook what we term systemic weaknesses cases where both the core LLM and the RM fail in tandem. We present ARES, a framework that systematically discovers and mitigates such dual vulnerabilities. ARES employs a ``Safety Mentor'' that dynamically composes semantically coherent adversarial prompts by combining structured component types (topics, personas, tactics, goals) and generates corresponding malicious and safe responses. This dual-targeting approach exposes weaknesses in both the core LLM and the RM simultaneously. Using the vulnerabilities gained, ARES implements a two-stage repair process: first fine-tuning the RM to better detect harmful content, then leveraging the improved RM to optimize the core model. Experiments across multiple adversarial safety benchmarks demonstrate that ARES substantially enhances safety robustness while preserving model capabilities, establishing a new paradigm for comprehensive RLHF safety alignment.