Michael Schwarz

PhD student in Static Analysis @ TUM

prof_pic.jpg
m.schwarz@tum.de

I am a PhD candidate in Helmut Seidl’s group and an alumnus of the DFG Research Training Group ConVeY (Continuous Verification of Cyber-Physical Systems). Having submitted my dissertation in late 2024, I currently expect to graduate some time in late Spring 2025.

I am interested in Static Analysis and related techniques and currently work within the field of Sound Static Analysis by Abstract Interpretation. There, we have, e.g., proposed techniques for efficient abstract interpretation of multi-threaded programs, worked on increasing usability by making static analysis incremental, and designed novel analyses for sometimes overlooked features of the C programming language.

I contribute to and currently co-maintain (with Simmo Saan and Karoliine Holter from UTartu) the static analysis framework Goblint for multi-threaded C programs, which serves as the testbed for many of our ideas.

I received a B.Sc. from TUM in 2016, and an M.Sc. from TUM in 2019. During my Master’s, I spent one term on exchange at the University of Waterloo and interned at NTNU as a software engineer for one term.

news

Apr 06, 2025 Just accepted: Our paper Taking out the Toxic Trash: Recovering Precision in Mixed Flow-Sensitive Static Analyses will be presented at PLDI this year. See you in South Korea! 🇰🇷
Apr 02, 2025 Our team won a Recognition Award from Huawei for the performance of Goblint on industrial benchmarks submitted to SV-COMP’25 by their Dresden Research Center. 🏆
Mar 26, 2025 Our static analzyer Goblint won the data race category at SV-COMP’25 🏅
Jan 22, 2025 Looking forward to participating in the Dagstuhl Seminar Sound Static Program Analysis in Modern Software Engineering this October! 🤓

selected publications

(Click above for a full list)
  1. PLDI
    Taking out the Toxic Trash: Recovering Precision in Mixed Flow-Sensitive Static Analyses
    Fabian Stemmler, Michael Schwarz, Julian Erhard, Sarah Tilscher, and 1 more author
    Proc. ACM Program. Lang., Jun 2025
    To appear
    arXiv Code
  2. VMCAI
    Correctness Witness Validation by Abstract Interpretation
    Simmo Saan, Michael Schwarz, Julian Erhard, Helmut Seidl, and 2 more authors
    In Verification, Model Checking, and Abstract Interpretation, Jan 2024
    Best Paper Award arXiv HTML Code
  3. SAS
    Octagons Revisited - Elegant Proofs and Simplified Algorithms
    Michael Schwarz, and Helmut Seidl
    In Static Analysis Symposium, Oct 2023
    HTML
  4. SOAP
    When Long Jumps Fall Short: Control-Flow Tracking and Misuse Detection for Non-local Jumps in C
    Michael Schwarz, Julian Erhard, Vesal Vojdani, Simmo Saan, and 1 more author
    In State Of the Art in Program Analysis, Jun 2023
    HTML Code
  5. ESOP
    Clustered Relational Thread-Modular Abstract Interpretation with Local Traces
    Michael Schwarz, Simmo Saan, Helmut Seidl, Julian Erhard, and 1 more author
    In Programming Languages and Systems - 32nd European Symposium on Programming, Apr 2023
    arXiv HTML Code

service

PC Member SOAP '25, NSAD '24
Reviewer STTT Special Section CSV '23
Artifact Evaluation Committee SAS '24, ESOP/FASE/FoSSaCS '24, TACAS' 22
Outreach TEDxTUM (student-organized popular science conference) Licensee 2019-2022