GuardAD: Safeguarding Autonomous Driving MLLMs via Markovian Safety Logic
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909032794554368 |
|---|---|
| author | Zhang, Tianyuan Yue, Peng Peng, Zihao Liu, Jiangfan Ying, Zonghao Wang, Jiakai Li, Tianlin Yang, Jian Yang, Yaodong Liu, Aishan Liu, Xianglong |
| author_facet | Zhang, Tianyuan Yue, Peng Peng, Zihao Liu, Jiangfan Ying, Zonghao Wang, Jiakai Li, Tianlin Yang, Jian Yang, Yaodong Liu, Aishan Liu, Xianglong |
| contents | Multimodal large language models (MLLMs) are increasingly integrated into autonomous driving (AD) systems; however, they remain vulnerable to diverse safety threats, particularly in accident-prone scenarios. Recent safeguard mechanisms have shown promise by incorporating logical constraints, yet most rely on static formulations that lack temporally grounded safety reasoning over evolving traffic interactions, resulting in limited robustness in dynamic driving environments. To address these limitations, we propose GuardAD, a model-agnostic safeguard that formulates AD safety as an evolving Markovian logical state. GuardAD introduces Neuro-Symbolic Logic Formalization, which represents safety predicates over heterogeneous traffic participants and continuously induces them via n-th order Markovian Logic Induction. This design enables the inference of emerging and latent hazards beyond single-step observations. Rather than simply vetoing unsafe actions, GuardAD performs Logic-Driven Action Revision, where inferred safety states actively guide action refinement without modifying the underlying MLLM. Extensive experiments on multiple benchmarks and AD-MLLMs demonstrate that GuardAD substantially reduces accident rates (-32.07%) while slightly improving task performance (+6.85%). Moreover, closed-loop simulation evaluations, together with physical-world vehicle studies, further validate the effectiveness and potential of GuardAD. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_10386 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | GuardAD: Safeguarding Autonomous Driving MLLMs via Markovian Safety Logic Zhang, Tianyuan Yue, Peng Peng, Zihao Liu, Jiangfan Ying, Zonghao Wang, Jiakai Li, Tianlin Yang, Jian Yang, Yaodong Liu, Aishan Liu, Xianglong Artificial Intelligence Multimodal large language models (MLLMs) are increasingly integrated into autonomous driving (AD) systems; however, they remain vulnerable to diverse safety threats, particularly in accident-prone scenarios. Recent safeguard mechanisms have shown promise by incorporating logical constraints, yet most rely on static formulations that lack temporally grounded safety reasoning over evolving traffic interactions, resulting in limited robustness in dynamic driving environments. To address these limitations, we propose GuardAD, a model-agnostic safeguard that formulates AD safety as an evolving Markovian logical state. GuardAD introduces Neuro-Symbolic Logic Formalization, which represents safety predicates over heterogeneous traffic participants and continuously induces them via n-th order Markovian Logic Induction. This design enables the inference of emerging and latent hazards beyond single-step observations. Rather than simply vetoing unsafe actions, GuardAD performs Logic-Driven Action Revision, where inferred safety states actively guide action refinement without modifying the underlying MLLM. Extensive experiments on multiple benchmarks and AD-MLLMs demonstrate that GuardAD substantially reduces accident rates (-32.07%) while slightly improving task performance (+6.85%). Moreover, closed-loop simulation evaluations, together with physical-world vehicle studies, further validate the effectiveness and potential of GuardAD. |
| title | GuardAD: Safeguarding Autonomous Driving MLLMs via Markovian Safety Logic |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2605.10386 |