Wei Le  Associate Professor CV

Computer Science, Iowa State University

Research interests: trustworthy software and AI

     



If you are interested in AI for software engineering, AI for medicine, or analyzing AI models and code, please submit your inquiries here: research track (Ph.D students) and engineering track.

Recent work

Towards Causal Deep Learning for Vulnerability Detection, ICSE 2024

Dataflow Analysis-Inspired Deep Learning for Efficient Vulnerability Detection, ICSE 2024 (talk@TOOS, talk@Google Journal Club)

TRACED: Execution-aware Pre-training for Source Code, ICSE 2024

Towards Understanding and Enhancing Robustness of Deep Learning Models against Malicious Unlearning Attacks, KDD 2023

Reducing Branch Divergence to Speed up Parallel Execution of Unit Testing on GPUs, The Journal of Supercomputing 2023

Research Interests

   

My research area is at the intersection of software engineering and machine learning. I am interested in advancing AI and making it safe and responsible to deploy. We are currently working on AI for software engineering, AI for medicine/health care, various challenges that bring AI to human level intelligence, and reliability of AI implementations. Our team: Program Analysis & AI Lab at Iowa State University

Selected Research Papers  

    [ICSE23] An Empirical Study of Deep Learning Models for Vulnerability Detection [artifact]
    [ICSE23] GameRTS: A Regression Testing Framework for Video Games [artifact] [talk]
    [ASE22] FuzzerAid: Grouping Fuzzed Crashes Based On Fault Signatures
    [ICSE22] DeepDiagnosis: Automatically Diagnosing Faults and Recommending Actionable Fixes in Deep Learning Programs [artifact]
    [ICSE22] DeepStability: A Study of Unstable Numerical Methods and Their Solutions in Deep Learning [artifact]
    [ICML21] RNNRepair: Automatic RNN Repair via Model-based Analysis [artifact]
    [ISSTA21] Validating Static Warnings via Testing Code Fragments [artifact], [ISSTA21 talk]
    [ICSE21] DeepLocalize: Fault Localization for Deep Neural Networks [artifact]
    [TSE19] Specifying Callback Control Flow of Mobile Apps Using Finite Automata
    [TSE17] Automatic Loop Summarization via Path Dependency Analysis
    [FSE16] Proteus: Computing Disjunctive Loop Summary via Path Dependency Analysis (Distinguished Paper)
    [ICSE16] Generating Performance Distributions via Probabilistic Symbolic Execution
    [ICSE14] Patch Verification via Multiversion Interprocedural Control Flow Graphs [open source code]
    [ICSE13] Segmented Symbolic Analysis
    [TOSEM13] Marple: Detecting Faults in Path Segments Using Automatically Generated Analyses
    [FSE10] Path-Based Fault Correlations
    [FSE08] Marple, A Demand-Driven Path-Sensitive Buffer Overflow Detector (Best FSE presentation)

Teaching

    2016-2024 Fall/Spring: COM S 342 Principles of Programming Languages

    2016-2023 Fall/Spring: COM S 413, COM S/CPRE 513 Program Analysis

    2020 Spring: COM S 665A Advanced Topics -- the Intersection of Software Engineering and Machine Learning

Professional Activities

    Co-chairing: ASE 2024 Journal First Track, ISSTA 2021 Doctoral Symposium, ICSE 2020 Co-located Events, 2018 Midwest Big Data Summer School Software Analytics Track, ESEC/FSE 2017 Doctoral Symposium, ICST 2017 Workshop, ICST 2016 Testing Tool Demo

    PC/Journal Review: ICSE 2025 Area Chair, ICSE 2023/2022/2021/2019/2016, FSE 2024/2022/2020/2019/2018, ASE 2021/2018/2016, ISSTA 2024/2020/2015, SPLASH 2024, CAIN 2022, TSE 2021/2017/2016/2015, TOSEM 2019/2016/2015/2014, ICST 2016, ICSE 2015 Demonstration, ICPC 2014 ERA, MOBS 2013, WAM 2023

    Panelist: NSF 2023/2022/2019/2018/2016/2014/2013, DOE 2016/2015, NIH 2022

    STEM Volunteering: Instructor @ Software Carpentry and Data Carpentry 2017-2022, SIGSOFT CAPS coordinator: 2014-2015