S2E Lab

Paper: InstruMate: A Systematic Framework for Assessing Android App Repackaging Resilience

Thu, 01 Jan 2026 17:00:00 +0000

Our work, “InstruMate: A Systematic Framework for Assessing Android App Repackaging Resilience”, was published in 33rd IEEE International Conference on Software Analysis, Evolution and Reengineering.

Publication details, authors, citation information, and available resources are provided below.

Paper: Automated Detection of Configuration-Specific Security Vulnerabilities via Patch Analysis

Thu, 01 Jan 2026 17:00:00 +0000

Our work, “Automated Detection of Configuration-Specific Security Vulnerabilities via Patch Analysis”, was published in ACM International Conference on the Foundations of Software Engineering.

Publication details, authors, citation information, and available resources are provided below.

Paper: CodeGuard: Improving LLM Guardrails in CS Education

Thu, 01 Jan 2026 17:00:00 +0000

Our work, “CodeGuard: Improving LLM Guardrails in CS Education”, was published in Findings of the Association for Computational Linguistics: EACL 2026.

Publication details, authors, citation information, and available resources are provided below.

Paper: Cyber-AI Supply Chain Vulnerabilities

Thu, 01 Jan 2026 17:00:00 +0000

Our work, “Cyber-AI Supply Chain Vulnerabilities”, was published in AI for Cybersecurity: Research and Practice.

Publication details, authors, citation information, and available resources are provided below.

Paper: Challenges to Using Large Language Models in Code Generation and Repair

Wed, 01 Jan 2025 17:00:00 +0000

Our work, “Challenges to Using Large Language Models in Code Generation and Repair”, was published in IEEE Security & Privacy.

Publication details, authors, citation information, and available resources are provided below.

Paper: "Large Language Models in Computer Science Education: A Systematic Literature Review" accepted at SIGCSE TS 2025

Tue, 01 Oct 2024 16:00:00 +0000

Our paper, “Large Language Models in Computer Science Education: A Systematic Literature Review”, has been accepted to the prestigious conference on computer science education, 56th ACM Technical Symposium on Computer Science Education (SIGCSE TS 2025).

We systematically analyzed 125 papers on using LLMs in computer science education. Our findings indicate that most current research focuses on undergraduate education and introductory programming courses. They also indicate that most research applies case-based studies, while the most widely used PL is Python. Although students are usually positive about using LLMs, instructors are worried about learning effectiveness because of potential over-reliance on them. Our systemic literature review also indicates that educators are gradually adopting LLMs in their courses, but most CS curricula still need to be changed to accommodate recent advances in AI.

Paper: "SALLM: Security Assessment of Generated Code" accepted at ASYDE 2024 (ASE Workshop)

Sat, 07 Sep 2024 16:00:00 +0000

Our paper, “SALLM: Security Assesment of Generated Code”, has been accepted to the 6th International Workshop on Automated and verifiable Software sYstem Development (ASYDE) co-located with Automated Software Engineering conference (ASE 2024).

This is the first kind of paper to introduce a framework for automated security evaluation of the generated code using dynamic and static analysis. We have 100 Python prompts with unit tests for functionality and security. We benchmarked several models with the SALLM framework, and found GPT-3.5 balanced both functional and secure code generation.

Paper: "FRANC: A Lightweight Framework for High-Quality Code Generation" and "The Fault in our Stars: Quality Assessment of Code Generation Benchmarks " accepted at SCAM 2022

Thu, 29 Aug 2024 16:00:00 +0000

Two papers (” FRANC: A Lightweight Framework for High-Quality Code Generation” and "”The Fault in our Stars: Quality Assessment of Code Generation Benchmarks”) got accepted at the 24th IEEE International Conference on Source Code Analysis and Manipulation (SCAM 2024) in the research track.

Paper: "Using Large Language Models to Generate JUnit Tests: An Empirical Study" accepted at EASE 2024.

Thu, 07 Mar 2024 16:00:00 +0000

Our paper, Using Large Language Models to Generate JUnit Tests: An Empirical Study, got accepted for the 28th International Conference on Evaluation and Assessment in Software Engineering (EASE 2024) in the research track. In this work, we analyzed three models with different prompting techniques to generate JUnit tests for the HumanEval dataset and real-world software. We evaluated the LLMs’ generated tests using compilation rates, test correctness, test coverage, and test smell. We found that though the models have higher coverage for small programming problems from the HumanEval dataset, they lack good coverage for real-world software from the Evosuite dataset.

Paper: "Understanding Regular Expression Denial of Service (ReDoS): Insights from LLM-Generated Regexes and Developer Forums" accepted at ICPC 2024.

Thu, 01 Feb 2024 16:00:00 +0000

Our paper, Understanding Regular Expression Denial of Service (ReDoS): Insights from LLM-Generated Regexes and Developer Forums, got accepted for the 32nd ACM/IEEE International Conference on Program Comprehension (ICPC 2024). In this work, we analyzed the robustness in generating functionally correct and ReDoS-free RegEx by LLMs, how understandable the LLM-generated and real world ReDoS-vulnerable RegEx and what are the perspective of the developers for ReDoS vulnerability.