Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2506.00509 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909630464000000 |
|---|---|
| author | Li, Zherui Mi, Yan Zhou, Zhenhong Jiang, Houcheng Zhang, Guibin Wang, Kun Fang, Junfeng |
| author_facet | Li, Zherui Mi, Yan Zhou, Zhenhong Jiang, Houcheng Zhang, Guibin Wang, Kun Fang, Junfeng |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_00509 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Goal-Aware Identification and Rectification of Misinformation in Multi-Agent Systems Li, Zherui Mi, Yan Zhou, Zhenhong Jiang, Houcheng Zhang, Guibin Wang, Kun Fang, Junfeng Computation and Language 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. |
| title | Goal-Aware Identification and Rectification of Misinformation in Multi-Agent Systems |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2506.00509 |