# Adversary-Aware DPO: Enhancing Safety Alignment in Vision Language Models via Adversarial Training

arXiv: https://arxiv.org/abs/2502.11455 · HTML: https://mlsanity.com/paper/2502.11455/

## Key facts

- Cited by 12 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 23 of its 38 references that are in the map.
- First posted on arXiv 17 Feb 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 / [Safety alignment, refusal and guardrails](https://mlsanity.com/safety/safety-alignment-refusal-and-guardrails/)

## Builds on (most cited first, 23 of 23)

- [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)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [Towards Evaluating the Robustness of Neural Networks](https://mlsanity.com/paper/1608.04644/) (2016)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilities](https://mlsanity.com/paper/2308.02490/) (2023)
- [Are We on the Right Way for Evaluating Large Vision-Language Models?](https://mlsanity.com/paper/2403.20330/) (2024)
- [A General Theoretical Paradigm to Understand Learning from Human Preferences](https://mlsanity.com/paper/2310.12036/) (2023)
- [GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Prompts](https://mlsanity.com/paper/2309.10253/) (2023)
- [FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts](https://mlsanity.com/paper/2311.05608/) (2023)
- [JailBreakV: A Benchmark for Assessing the Robustness of MultiModal Large Language Models against Jailbreak Attacks](https://mlsanity.com/paper/2404.03027/) (2024)
- [Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs](https://mlsanity.com/paper/2407.15549/) (2024)
- [Eyes Closed, Safety On: Protecting Multimodal LLMs via Image-to-Text Transformation](https://mlsanity.com/paper/2403.09572/) (2024)
- [Efficient Adversarial Training in LLMs with Continuous Attacks](https://mlsanity.com/paper/2405.15589/) (2024)
- [Towards Understanding Jailbreak Attacks in LLMs: A Representation Space Analysis](https://mlsanity.com/paper/2406.10794/) (2024)
- [VLSBench: Unveiling Visual Leakage in Multimodal Safety](https://mlsanity.com/paper/2411.19939/) (2024)
- [Don't Say No: Jailbreaking LLM by Suppressing Refusal](https://mlsanity.com/paper/2404.16369/) (2024)

## Built on by (most cited first, 12 of 12)

- [GuardReasoner-VL: Safeguarding VLMs via Reinforced Reasoning](https://mlsanity.com/paper/2505.11049/) (2025)
- [Rethinking Jailbreak Detection of Large Vision Language Models with Representational Contrastive Scoring](https://mlsanity.com/paper/2512.12069/) (2025)
- [SaFeR-VLM: Toward Safety-aware Fine-grained Reasoning in Multimodal Models](https://mlsanity.com/paper/2510.06871/) (2025)
- [Misaligned Roles, Misplaced Images: Structural Input Perturbations Expose Multimodal Alignment Blind Spots](https://mlsanity.com/paper/2504.03735/) (2025)
- [WARD: Adversarially Robust Defense of Web Agents Against Prompt Injections](https://mlsanity.com/paper/2605.15030/) (2026)
- [Harmonious Parameter Adaptation in Continual Visual Instruction Tuning for Safety-Aligned MLLMs](https://mlsanity.com/paper/2511.20158/) (2025)
- [Empowering Multimodal LLMs with External Tools: A Comprehensive Survey](https://mlsanity.com/paper/2508.10955/) (2025)
- [Agentic Moderation: Multi-Agent Design for Safer Vision-Language Models](https://mlsanity.com/paper/2510.25179/) (2025)
- [SaFeR-ToolKit: Structured Reasoning via Virtual Tool Calling for Multimodal Safety](https://mlsanity.com/paper/2603.02635/) (2026)
- [Evolving Safety Landscape of Multi-modal Large Language Models: A Survey of Emerging Threats and Safeguards](https://mlsanity.com/paper/2608.07535/) (2026)
- [The Uncontrolled Variable: Vision-Language Model Refusal Responds to Image Presence in Ways Risk Cannot Explain](https://mlsanity.com/paper/2609.26174/) (2026)
- [SafeMol: Dual-Modality Safety Alignment for Molecular Multimodal Models](https://mlsanity.com/paper/2609.33640/) (2026)

## Abstract (arXiv)

Safety alignment is critical in pre-training large language models (LLMs) to generate responses aligned with human values and refuse harmful queries. Unlike LLM, the current safety alignment of VLMs is often achieved with post-hoc safety fine-tuning. However, these methods are less effective to white-box attacks. To address this, we propose textitAdversary-aware DPO (ADPO), a novel training framework that explicitly considers adversarial. textitAdversary-aware DPO (ADPO) integrates adversarial training into DPO to enhance the safety alignment of VLMs under worst-case adversarial perturbations. textitADPO introduces two key components: (1) an adversarial-trained reference model that generates human-preferred responses under worst-case perturbations, and (2) an adversarial-aware DPO loss that generates winner-loser pairs accounting for adversarial distortions. By combining these innovations, textitADPO ensures that VLMs remain robust and reliable even in the presence of sophisticated jailbreak attacks. Extensive experiments demonstrate that textitADPO outperforms baselines in the safety alignment and general utility of VLMs.