Enhancing Robustness of LLM-Driven Multi-Agent Systems through Randomized Smoothing
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912468237811712 |
|---|---|
| author | Hu, Jinwei Dong, Yi Ding, Zhengtao Huang, Xiaowei |
| author_facet | Hu, Jinwei Dong, Yi Ding, Zhengtao Huang, Xiaowei |
| contents | This paper presents a defense framework for enhancing the safety of large language model (LLM) empowered multi-agent systems (MAS) in safety-critical domains such as aerospace. We apply randomized smoothing, a statistical robustness certification technique, to the MAS consensus context, enabling probabilistic guarantees on agent decisions under adversarial influence. Unlike traditional verification methods, our approach operates in black-box settings and employs a two-stage adaptive sampling mechanism to balance robustness and computational efficiency. Simulation results demonstrate that our method effectively prevents the propagation of adversarial behaviors and hallucinations while maintaining consensus performance. This work provides a practical and scalable path toward safe deployment of LLM-based MAS in real-world, high-stakes environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_04105 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Enhancing Robustness of LLM-Driven Multi-Agent Systems through Randomized Smoothing Hu, Jinwei Dong, Yi Ding, Zhengtao Huang, Xiaowei Artificial Intelligence Multiagent Systems This paper presents a defense framework for enhancing the safety of large language model (LLM) empowered multi-agent systems (MAS) in safety-critical domains such as aerospace. We apply randomized smoothing, a statistical robustness certification technique, to the MAS consensus context, enabling probabilistic guarantees on agent decisions under adversarial influence. Unlike traditional verification methods, our approach operates in black-box settings and employs a two-stage adaptive sampling mechanism to balance robustness and computational efficiency. Simulation results demonstrate that our method effectively prevents the propagation of adversarial behaviors and hallucinations while maintaining consensus performance. This work provides a practical and scalable path toward safe deployment of LLM-based MAS in real-world, high-stakes environments. |
| title | Enhancing Robustness of LLM-Driven Multi-Agent Systems through Randomized Smoothing |
| topic | Artificial Intelligence Multiagent Systems |
| url | https://arxiv.org/abs/2507.04105 |