Jingyi Wang王竟亦

Assistant professor, Zhejiang University · leading the IS2 (Intelligent System Security) Lab, Hangzhou, China

Jingyi Wang

I am interested in building formal methods that make complex intelligent systems (software, AI models and control) verifiably trustworthy.

Much of this work happens with industry. We collaborate closely with Alibaba, Ant Group and Huawei, which lets us take verification and security analysis from the paper into systems running at production scale, and brings the problems that matter back into the lab.

I am looking for PhD students, postdocs and research interns. Collaborations from academia and industry are warmly welcome — see openings.

Recent

Aug 2026
The deadline for the second round of SETTA 2026 is approaching (Aug 24, 2026), consider submitting your work!
Feb 2026
Rounding-Guided Backdoor Injection in Deep Learning Model Quantization appears at NDSS 2026.
Jan 2026
Serving on the program committees of ICSE, FSE and USENIX Security 2027, and as SVT track co-chair at SAC 2027.
Nov 2025
Two papers at ASE 2025, including an industry case study on end-to-end formal verification of decentralized identities in the Alipay Super App.
Oct 2025
Generalized Security-Preserving Refinement for Concurrent Systems and Provable Repair of Deep Neural Network Defects through Pre-image Synthesis and Property Refinement appears at ACM CCS 2025.

Research themes

Each theme links to the matching papers. See all publications.

Experience

2020–now
Assistant professor, College of Control Science and Engineering, Zhejiang University, leading the IS2 Lab.
2019–20
Research fellow with Prof. Jin Song Dong, School of Computing, National University of Singapore.
2018–19
Research fellow with Prof. Jun Sun, Information Systems Technology and Design, Singapore University of Technology and Design.
2018
Ph.D., Singapore University of Technology and Design.
2013
B.E. in information engineering, Xi’an Jiaotong University.

Awards

Outside work

I play badminton regularly, and DOTA2 and PES when there is time.