Goal-Aware Identification and Rectification of Misinformation in Multi-Agent Systems
This addresses security risks in multi-agent AI systems, though it appears incremental as it builds on existing defense concepts with a new dataset and framework.
The paper tackles misinformation injection vulnerabilities in Large Language Model-based Multi-Agent Systems by introducing MisinfoTask, a dataset for evaluating robustness, and ARGUS, a defense framework that reduces misinformation toxicity by approximately 28.17% and improves task success rates under attack by approximately 10.33%.
Large Language Model-based Multi-Agent Systems (MASs) have demonstrated strong advantages in addressing complex real-world tasks. However, due to the introduction of additional attack surfaces, MASs are particularly vulnerable to misinformation injection. To facilitate a deeper understanding of misinformation propagation dynamics within these systems, we introduce MisinfoTask, a novel dataset featuring complex, realistic tasks designed to evaluate MAS robustness against such threats. Building upon this, we propose ARGUS, a two-stage, training-free defense framework leveraging goal-aware reasoning for precise misinformation rectification within information flows. Our experiments demonstrate that in challenging misinformation scenarios, ARGUS exhibits significant efficacy across various injection attacks, achieving an average reduction in misinformation toxicity of approximately 28.17% and improving task success rates under attack by approximately 10.33%. Our code and dataset is available at: https://github.com/zhrli324/ARGUS.