Dual-Modality Multi-Stage Adversarial Safety Training: Robustifying Multimodal Web Agents Against Cross-Modal Attacks

arXiv:2603.04364v2 Announce Type: replace-cross Abstract: Multimodal web agents that process both screenshots and accessibility trees are increasingly deployed to interact with web interfaces, yet their dual-stream architecture opens an underexplored attack surface: an adversary who injects content into the webpage DOM simultaneously corrupts both observation channels with a consistent deceptive narrative. Our vulnerability analysis on MiniWob++ reveals that attacks including a visual component far outperform text-only injections, exposing critical gaps in text-centric VLM safety training. Motivated by this finding, we propose Dual-Modality Multi-Stage Adversarial Safety Training (DMAST), a framework that formalizes the agent-attacker interaction as a two-player general-sum Markov game and co-trains both players through a three-stage pipeline: (1) imitation learning from a strong teacher model, (2) oracle-guided supervised fine-tuning that uses a novel zero-acknowledgment strategy to instill task-focused reasoning under adversarial noise, and (3) adversarial reinforcement learning via Group Relative Policy Optimization (GRPO) self-play. On out-of-distribution tasks, DMAST nearly halves the attack success rate (41.2%rightarrow21.4%) while raising task completion by over 60% relative (6.2%rightarrow10.2%). Our approach outperforms established training-based defenses and complements prompt-based defenses, demonstrating genuine co-evolutionary progress and robust generalization to complex, unseen environments. Code is available at https://github.com/huajianduzhuo-code/DMAST_official.

Related

• EVA: Evolving Semantic Adversaries for Red-Teaming GUI Agents Against Environmental Injection Attacks
• Analyzing Defensive Misdirection Against Model-Guided Automated Attacks on Agentic AI Systems
• COMIC: Reference-Aware Safety Gating for Multimodal Large Language Models

Source: arXiv cs.AI | 2026-10-07