Welcome to Assured and Intelligent Cyber-Physical Systems Laboratory at University of Notre Dame!
AICPS@ND is led by Dr. Fanxin Kong. Our research centers around assurance and intelligence of cyber-physical systems with applications to various robotic systems. Research areas include i) physical/embodied AI/SI: reinforcement learning, neuro-symbolic AI/SI, and foundation models, ii) security: attack detection, diagnosis, and recovery, and iii) safety: timing and functional correctness.
Recent News (More)
[Top] Hiring: we are looking for self-motivated PhD students. If you are interested in my research and working in our lab, please directly contact Dr. Fanxin Kong with your CV by fkong at nd dot edu.
[Oct. 2026] Invited Talk: I gave a talk on Assured and Intelligent Cyber-Physical Systems in Towards the Next Generation of CPS, a Workshop on the 20th Anniversary of the NSF CPS Program. Thanks NSF for the invitation.
[Oct. 2026] Pre-print: our pre-print "SLoSa: Small, Low-Power, and Safe Foundation Models for Robotic Task Planning", is available by arXiv:2610.28955.
[Oct. 2026] Pre-print: our pre-print "Vulnerability Analysis for Safe Multi-Agent Reinforcement Learning", is available by arXiv:2610.27293.
[Oct. 2026] Benchmark: while VLA benchmarks usually ask whether a robot finishes the task, our SafeVLA-Bench also asks whether it did so safely! Check more on https://safevla.org!
[Oct. 2026] Pre-print: our pre-print "SafeInferCom: Safe Inference-Time Compute via Verifier-Guided Mid-Generation Intervention for Robotic Task Planning", is available by arXiv:2610.11223.
[Oct. 2026] Invited Talk: I gave a talk on Assured and Intelligent Cyber-Physical Systems in Robotics Seminar Series at NC State University. Thanks Zhishan for the invitation.
[Sep. 2026] Paper: our paper "Exploiting Fine-Tuning Structures to Improve Adversarial Transferability on Downstream SAM", has been accepted by the Fortieth Annual Conference on Neural Information Processing Systems (NeurIPS). Congratulations to Shixiong and all!
[Jun. 2026] Pre-print: our pre-print "Safe-RULE: Safe Reinforcement UnLEarning", is available by arXiv:2606.09559.
[May 2026] Pre-print: our ore-print "SafeVLA-Bench: A Benchmark for the Success-Safety Gap in Vision-Language-Action Models", is available by arXiv:2606.00773.
[May 2026] Award: Jialiang has been awarded 2025-2026 Outstanding Teaching Assistant. Congratulations to Jialiang!
[April 2026] Paper: our paper "Deadline-Safe Control Synthesis for Cyber-Physical Systems with Reinforcement Learning", has been accepted by ACM Transactions on Embedded Computing Systems (TECS), 2026.
[Mar. 2026] Award: Zifan has received the 2026 All University Doctoral Prize. Big congratulations to Zifan!
[Mar. 2026] Pre-print: our pre-print "SafePilot: A Framework for Assuring LLM-enabled Cyber-Physical Systems", is available by arXiv:2603.21523.
[Feb. 2026] Pre-print: our pre-print "SafeGen-LLM: Enhancing Safety Generalization in Task Planning for Robotic Systems", is available by arXiv:2602.24235.
[Feb. 2026] Pre-print: our pre-print "Enhancing LLM-Based Test Generation by Eliminating Covered Code", is available by arXiv:2602.21997.
[Feb. 2026] Pre-print: our pre-print "Demonstration-Guided Observation Attacks on Black-Box Safe Reinforcement Learning Controllers for Robotic Systems", is available by arXiv:2602.16543.
[Jan. 2026] Paper: our paper "SafeNet: A Neural-Symbolic Network for Safe Planning in Robotic Systems using Formal Method-Guided LLM Fine-Tuning", has been accepted for presentation at the 2026 IEEE International Conference on Robotics and Automation (ICRA). Congratulations to Zifan and Jialiang!
[Jan. 2026] Paper: our paper "Vulnerability Exploration of Safe Reinforcement Learning in Cyber-Physical Systems via STL Mining", has been accepted by the 17th ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS) 2026, part of CPS-IoT Week 2026. Congratulations to Jialiang!
[Jan. 2026] Award: I received the 2025 ACM SIGBED Early Career Researcher Award! Many thanks to ACM SIGBED! Many thanks to my nominator!
[Jan. 2026] Presentation: our abstract "LLM-Based Software Test Generation", has been accepted by the Flight Software Workshop (FSW) 2026. Congratulations to Weizhe!
[July 2025] Paper: our paper "Vulnerability Analysis for Safe Reinforcement Learning in Cyber-Physical Systems", has been accepted by Transactions on Cyber-Physical Systems. Congratulations to Shixiong!
[Apr. 2025] Award: I received the NSF CAREER Award on Security Foundations of Safe Learning Enabled Cyber-Physical Systems! Thanks NSF!
Black-Box Stealthy Sensor Attacks on Autonomous Vehicles
Vunlnerability analysis for safe reinforcement learning
No recovery vs real-time attack recovery
Recent Publications (Full Publications)
VLA/LLM/FM-Enabled Systems:
[NeurIPS'26] Shixiong Jiang, Jialiang Fan, Mengyu Liu, Pengfei Gu, Danny Z Chen, and Fanxin Kong, "Exploiting Fine-Tuning Structures to Improve Adversarial Transferability on Downstream SAM", in the Fortieth Annual Conference on Neural Information Processing Systems, 2026.
[ICRA'26] Zifan Wang, Jialiang Fan, Rui Zuo, Qinru Qiu, and Fanxin Kong, "SafeNet: A Neural-Symbolic Network for Safe Planning in Robotic Systems using Formal Method-Guided LLM Fine-Tuning", in the IEEE International Conference on Robotics and Automation, 2026.
[HCSS'25] Weizhe Xu and Fanxin Kong, "LLM-enabled Software Testing", in the 25th Annual High Confidence Software and Systems Conference, 2025.
[ICCPS'24] Weizhe Xu, Mengyu Liu, Steven Drager, Matthew Anderson, and Fanxin Kong, "Assuring LLM-Enabled Cyber-Physical Systems", in the 15th ACM/IEEE International Conference on Cyber-Physical Systems, Poster/Demo Track, part of CPS-IoT Week 2024.
[FMSys'24] Weizhe Xu, Mengyu Liu, Oleg Sokolsky, Insup Lee, and Fanxin Kong, "LLM-enabled Cyber-Physical Systems: Survey, Research Opportunities, and Challenges", International Workshop on Foundation Models for Cyber-Physical Systems & Internet of Things, co-located with CPS-IoT Week 2024.
Secure & Safe Reinforcement Learning:
[ICCPS'26] Jialiang Fan, Shixiong Jiang, Mengyu Liu, and Fanxin Kong, "Vulnerability Exploration of Safe Reinforcement Learning in Cyber-Physical Systems via STL Mining", in the 17th ACM/IEEE International Conference on Cyber-Physical Systems, 2026.
[EMSOFT'24] Shixiong Jiang, Mengyu Liu, and Fanxin Kong, "Backdoor Attacks on Safe Reinforcement Learning-Enabled Cyber-Physical Systems", in the ACM SIGBED International Conference on Embedded Software, part of Embedded Systems Week (ESWEEK) 2024.
[ICCPS'24] Shixiong Jiang, Mengyu Liu, and Fanxin Kong, "Vulnerability Analysis for Safe Reinforcement Learning in Cyber-Physical Systems", in the 15th ACM/IEEE International Conference on Cyber-Physical Systems, part of CPS-IoT Week 2024.
Timing-Safe Learning Enabled Systems:
[RTSS'24] Mengyu Liu, Pengyuan Lu, Xin Chen, Oleg Sokolsky, Insup Lee, and Fanxin Kong, "Deadline-Safe Reach-Avoid Control Synthesis for Cyber-Physical Systems with Reinforcement Learning", in the 45th IEEE Real-Time Systems Symposium, 2024.
[MEMCODE'24] Mengyu Liu, Pengyuan Lu, Xin Chen, Oleg Sokolsky, Insup Lee and Fanxin Kong, "Model-free PAC Time-Optimal Control Synthesis with Reinforcement Learning", in the 22nd International Symposium on Formal Methods and Models for System Design, colocate with Embedded Systems Week, 2024.
Attack Recovery:
[RTAS'24] Lin Zhang, Luis Burbano, Xin Chen, Alvaro A. Cardenas, Steve Drager, Matthew Adderson, and Fanxin Kong, "Fast Attack Recovery for Stochastic Cyber-Physical Systems", in the 30th IEEE Real-Time and Embedded Technology and Applications Symposium, 2024.
[RTSS'23] Mengyu Liu, Lin Zhang, Vir Phoha, and Fanxin Kong, "Learn-to-Respond: Sequence-Predictive Recovery from Sensor Attacks in Cyber-Physical Systems", in the 44th IEEE Real-Time Systems Symposium, 2023.
[RTAS'23] Lin Zhang, Kaustubh Sridhar, Mengyu Liu, Pengyuan Lu, Xin Chen, Fanxin Kong, Oleg Sokolsky and Insup Lee, "Real-Time Data-Predictive Attack-Recovery for Complex Cyber-Physical Systems", in the 29th IEEE Real-Time and Embedded Technology and Applications Symposium, 2023.
[EMSOFT'21] Lin Zhang, Pengyuan Lu, Fanxin Kong, Xin Chen, Oleg Sokolsky and Insup Lee, "Real-Time Attack-Recovery for Cyber-Physical Systems using Linear-Quadratic Regulator", in the 21st ACM SIGBED International Conference on Embedded Software, 2021.
[RTSS'20] Lin Zhang, Xin Chen, Fanxin Kong, and Alvaro A. Cardenas, "Real-Time Recovery for Cyber-Physical Systems using Linear Approximations", in the 41st IEEE Real-Time Systems Symposium, 2020.
[ICCPS'18] Fanxin Kong, Meng Xu, James Weimer, Oleg Sokolsky and Insup Lee, "Cyber-Physical System Checkpointing and Recovery", in the 9th ACM/IEEE International Conference on Cyber-Physical Systems, 2018.
Attack Detection & Diagnosis:
[RTAS'25] Weizhe Xu, Xin Chen, Matthew Anderson, Steven Drager, and Fanxin Kong, "Recovery-Guaranteed Sensor Attack Detection for Cyber-Physical Systems", in the 31st IEEE Real-Time and Embedded Technology and Applications Symposium, 2025.
[RTSS'23] Zifan Wang, Lin Zhang, Qinru Qiu, and Fanxin Kong, "Catch You If Pay Attention: Temporal Sensor Attack Diagnosis Using Attention Mechanisms for Cyber-Physical Systems", in the 44th IEEE Real-Time Systems Symposium, 2023.
[DAC'22] Lin Zhang, ZifanWang, Mengyu Liu, and Fanxin Kong, "Adaptive Window-Based Sensor Attack Detection for Cyber-Physical Systems", in the 59th Design Automation Conference, 2022.
[RTAS'21] Francis Akowuah and Fanxin Kong, "Real-Time Adaptive Sensor Attack Detection in Autonomous Cyber-Physical Systems", in the 27th IEEE Real-Time and Embedded Technology and Applications Symposium, 2021.
[DAC'20] Tianjia He, Lin Zhang, Fanxin Kong, and Asif Salekin, "Exploring Inherent Sensor Redundancy for Automotive Anomaly Detection", in the 57th Design Automation Conference, 2020.