Hi! I’m Michael, a fourth year Computer Science PhD student at MIT advised by Armando Solar-Lezama and supported by an NSF fellowship. I’m currently working on program analysis for dynamic languages, specifically Javascript. I’m interested finding different ways to augment traditional program analysis with language models. Previously, I was a security researcher at MIT Lincoln Laboratory. I graduated with my undergraduate from Northeastern University in 2018.

News

• 2025_03: ABSINT-AI accepted to VerifAI 2025 at ICLR.
• 2023.05: Switched to the Computer-Aided Programming group advised by Armando Solar-Lezama.
• 2023.05: Finished my TA requirement by TA’ing 6.5660.
• 2023.04: RaceInjector is accepted at SOAP 2023.
• 2022.12: Finished my TQE requirement.
• 2022.06: NeuDep is accepted to ESEC/FSE 2022.
• 2022.05: CVE-2022-29968 found and presented ICML. Joseph was the real expert here - I just rode his coattails.
• 2022.05: COBRA-GCN was accepted and presented at DIMVA.
• 2022.04: Received the NSF GRFP.
• 2021.09: Started my PhD at MIT under Una-May O’Reilly in the ALFA group.
• 2019.01: First real job at MIT Lincoln Laboratory.
• 2018.12: Graduated from Northeastern University.

Publications

• ABSINT-AI: Language Models for Abstract Interpretation Michael Wang, Kexin Pei, Armando Solar-Lezama. VerifAI 2025.
• CALLME: Call Graph Augmentation with Large Language Models for Javascript. Michael Wang, Kexin Pei, Armando Solar-Lezama. In Submission.
• RaceInjector: Injecting Races to Evaluate and Learn Dynamic Race Detection Algorithms Michael Wang, Shashank Srikant, Malavika Samak, Una-May O’Reilly. SOAP 2023.
• NeuDep: Neural Binary Memory Dependence Analysis Kexin Pei, Dongdong She, Michael Wang, Scott Geng, Zhou Xuan, Yaniv David, Junfeng Yang, Suman Jana, Baishakhi Ray. ESEC/FSE 2022.
• COBRA-GCN: Contrastive Learning to Optimize Binary Representation Analysis with Graph Convolutional Networks Michael Wang, Alexander Interrante-Grant, Ryan Whelan, Tim Leek. DIMVA 2022.