About me

I am a Wallenberg-NTU Presidential Postdoctoral Fellow in School of Computer Science and Engineering, Nanyang Technological University, Singapore in Prof.Yang Liu’s group. I received my Ph.D (2014-2018), MSc (2011-2014) and B.E. (2007-2011) degrees from Tianjin University , China.

My current research mainly focuses on program analysis, software testing, verification and AI security.

  • Program Analysis. We are working at program analysis especially on the challenging task, i.e., automatic loop analysis. We have conducted a series of works on loop analysis and summarisation, including the loop classification and disjunctive loop summarisation (FSE’16 and TSE’18), string loop summarisation (ISSTA’15) and loop termination analysis (ESEC/FSE’17).
  • Deep Learning Security. Currently, my major focus is on the security and quality analysis of deep learning systems, which is a new research problem. We have conducted several works including an empricy study about develop and deployment issues of deep learning system (ASE’19), coverage-guided fuzz testing for feedfoward neural networks (ISSTA’19, ASE’19), quantitative analysis of recurrent neural networks (ESEC/FSE’19, ASE’19), evaluating the data quality and robustness of models via mutation testing (ASE’19) and deep learning deployment issues detection (IJCAI’19).
  • Software Testing and Verification. We are working at several testing and security topics including adaptive fuzzing for vulnerability detection (ESEC/FSE’19), directed fuzzing (CCS’18), hybrid testing based on symbolic execution and fuzzing (JOS’19), root cause analysis of buffer overflow (ESEC/FSE’19) and automatic testing for online combat games (ASE’19). In addition, we also did some works about security protocol verification (JDCTA’12, ICFEM’14).


  1. (5/2021) Our paper “Automatic RNN Repair via Model-based Analysis” is accepted by ICML 2021
  2. (5/2021) Our paper “AVA: Adversarial Vignetting Attack against Visual Recognition” is accepted by IJCAI 2021
  3. (4/2021) Our paper “Breaking Neural Reasoning Architectures with Metamorphic Relation-Based Adversarial Examples” is accepted by TNNLS
  4. (3/2021) Our paper “Bias Field Poses a Threat to DNN-based X-Ray Recognition!” is accepted by ICME 2021
  5. (2/2021) Our paper “Stealing Deep Reinforcement Learning Models for Fun and Profit” is accepted by AsiaCCS 2021
  6. (1/2021) Our paper “Retrieval-Augmented Generation for Code Summarization via Hybrid GNN” is accepted by ICLR 2021 (spotlight)
  7. (12/2020) Our paper “Automatic Web Testing using Curiosity-Driven Reinforcement Learning” is accepted by ICSE 2021
  8. (12/2020) Two papers are accepted by AAAI 2021
  9. (11/2020) Our paper “ An Empirical Study on Robustness of DNNs with Out-of-Distribution Awareness” won the Best Paper Award in APSEC 2020
  10. (09/2020) Our paper “Watch out! Motion is Blurring the Vision of Your Deep Neural Networks” is accepted by NeurIPS 2020.
  11. (09/2020) Our paper “A Performance-Sensitive Malware Detection System Using Deep Learning on Mobile Devices” is accepted by TIFS.
  12. (08/2020) Our paper “Can We Trust Your Explanations? Sanity Checks for Interpreters in Android Malware Analysis” is accepted by TIFS.
  13. (07/2020) Four papers are accepted by ASE 2020.
  14. (07/2020) Four papers are accepted by ACM MM 2020.
  15. (07/2020) Our paper “Regression Testing of Massively Multiplayer Online Role-Playing Games” is accepted at ICSME 2020, NIER Track.
  16. (07/2020) Our paper “JSCSP: a Novel Policy-Based XSS Defense Mechanism for Browsers” is accepted at TDSC.
  17. (07/2020) Our paper “SPARK: Spatial-aware Online Incremental Attack Against Visual Tracking” is accepted at ECCV 2020.
  18. (07/2020) I was awarded the ACM Tianjin Doctoral Dissertation Award 2019.
  19. (05/2020) Our paper “FakeSpotter: A Simple yet Robust Baseline for Spotting AI-Synthesized Fake Faces” is accepted at IJCAI 2020.
  20. (03/2020) Our paper “How are Deep Learning Models Similar? An Empirical Study on Clone Analysis of Deep Learning Software” is accepted at ICPC 2020.
  21. (12/2019) Three papers “Towards Characterising Adversarial Defects of Deep Learning Software from the Lens of Uncertainty”, “Typestate-Guided Fuzzer for Discovering Use-after-Free Vulnerabilities” and “MemLock: Memory Usage Guided Fuzzing” are accepted at ICSE’2020.
  22. (11/2019) Our papers “Stealthy and Efficient Adversarial Attacks against Deep Reinforcement Learning” is accepted at AAAI 2020.
  23. (10/2019) Our paper “Wuji: Automatic Online Combat Game Testing Using Evolutionary Deep Reinforcement Learning” receives the ACM SIGSOFT Distinguished Paper Award in ASE’19.
  24. (10/2019) I was awarded ACM SIGSOFT CAPS Travel Funds for ASE 2019.
  25. (08/2019) Three papers including DeepHunter, DeepStellar and DeepMutation++ are accepted by ASE 2019.
  26. (08/2019) Two papers “An Empirical Study towards Characterizing Deep Learning Development and Deployment across Different Frameworks and Platforms” and “Wuji: Automatic Online Combat Game Testing Using Evolutionary Deep Reinforcement Learning”” are accepted by ASE 2019.
  27. (07/2019) Two papers “Safe Inputs Generation for Black-box Systems” and “MobiDroid: A Performance-Sensitive Malware Detection System on Mobile Platform” are accepted by ICECCS 2019.
  28. (05/2019) Three papers “DeepStellar: Model-Based Quantitative Analysis of Stateful Deep Learning Systems”, “Locating Vulnerabilities in Binaries via Memory Layout Recovering” and “Cerebro: Context-aware Adaptive Fuzzing for EffectiveVulnerability Detection” are accepted by ESEC/FSE 2019.
  29. (05/2019) Our paper “DiffChaser: Detecting Disagreements for Deep Neural Networks” is accepted by IJCAI 2019.
  30. (05/2019) Our paper “DeepHunter: A Coverage-Guided Fuzz Testing Framework for Deep Neural Networks” is accepted by ISSTA 2019.