Caution for the Environment: Multimodal LLM Agents are Susceptible to Environmental Distractions

This paper highlights the vulnerability of multimodal agents to environmental distractions. The researchers demonstrate that these agents, which process multiple types of input (e.g., text, images, audio), can be significantly impacted by irrelevant or misleading environmental cues. The study provides insights into the limitations of current multimodal systems and emphasizes the need for more robust architectures that can filter out distractions and maintain focus on relevant information in complex, real-world environments.

• GUI-Perturbed: Domain Randomization Reveals Systematic Brittleness in GUI Grounding Models
  April 15, 2026 · arXiv
• The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents
  April 12, 2026 · arXiv
• CORA: Conformal Risk-Controlled Agents for Safeguarded Mobile GUI Automation
  April 10, 2026 · arXiv
• Preference Redirection via Attention Concentration: An Attack on Computer Use Agents
  April 9, 2026 · arXiv
• Are GUI Agents Focused Enough? Automated Distraction via Semantic-level UI Element Injection
  April 9, 2026 · arXiv
• Dual-Modality Multi-Stage Adversarial Safety Training: Robustifying Multimodal Web Agents Against Cross-Modal Attacks
  March 4, 2026 · arXiv