Software Technology Track: AI-Driven Program Analysis and Software Synthesis: Transforming Modern Software Engineering

The integration of artificial intelligence (AI) into program analysis and software synthesis is reshaping the landscape of software engineering. AI-powered techniques are revolutionizing defect detection, performance optimization, security assurance, and software generation, enabling more intelligent, efficient, and scalable approaches. As modern software systems grow in complexity—spanning distributed architectures, microservices, and AI/ML-driven components—traditional program analysis techniques struggle to maintain scalability, precision, and adaptability. AI methodologies, including machine learning (ML), deep learning, and large language models (LLMs), are emerging as transformative forces that enhance program analysis frameworks and drive the automation of software development, modification, and evolution.

This minitrack provides a platform for discussing the latest advances at the intersection of AI and program analysis, with a focus on innovative methodologies, automated software synthesis techniques, practical implementations, and real-world case studies. The goal is to foster collaboration between academia and industry, bridging theoretical research with applied solutions to advance the state of the art in AI-augmented software engineering.

Key Themes include:
1. AI-Powered Static Analysis: Leveraging machine learning and deep learning models to improve the accuracy, efficiency, and scalability of static code analysis, including vulnerability detection, type inference, and performance profiling.
2. Machine Learning for Dynamic Analysis: Using AI techniques for runtime monitoring, anomaly detection, predictive debugging, and intelligent test case generation.
3. AI-Augmented Software Synthesis and Modification: Exploring AI-driven techniques for automated code generation, program transformation, refactoring, and patch synthesis.
4. AI-Driven Software Generation: Investigating generative AI models for automated software creation, including domain-specific code generation, AI-assisted software design, and automated system architecture synthesis.
5. Integration of AI and Program Analysis Frameworks: Investigating how AI models can be seamlessly integrated into traditional program analysis tools, enhancing their adaptability and effectiveness.
6. Applications of AI-Augmented Analysis and Software Synthesis: Examining use cases in software verification, automated repair, security assessment, and performance optimization.
7. Benchmarks and Datasets for AI-Augmented Analysis: Discussing the need for standardized datasets, benchmark suites, and evaluation metrics to advance research in AI-driven program analysis.
8. Real-World Case Studies and Challenges: Showcasing industry implementations, lessons learned, and the challenges of deploying AI-powered program analysis tools at scale.
9. Emerging Trends and Future Directions: Identifying open research problems, novel AI techniques, and future possibilities for enhancing software engineering practices through AI.

We welcome contributions that explore theoretical advancements, algorithmic innovations, tool and framework development, empirical evaluations, and case studies related to AI-augmented program analysis and software synthesis. Topics include, but are not limited to:
1. AI-assisted bug detection, vulnerability identification, and security assurance
2. Large language models (LLMs) for software analysis, synthesis, and code generation
3. AI-driven program transformation, refactoring, and automated patching
4. Intelligent test case generation and automated debugging techniques
5. Neural program synthesis and reinforcement learning for software engineering
6. AI-enhanced compiler optimizations and performance tuning
7. Automated reverse engineering and decompilation using AI techniques
8. Hybrid approaches combining formal methods with AI-driven analysis
9. Ethical, interpretability, and reliability concerns in AI-augmented program analysis
10. Tools, datasets, and benchmark suites for evaluating AI-driven software engineering solutions
11. AI-driven software generation and evolution, including automated code completion, software design pattern synthesis, and program synthesis for specialized domains
12. AI-powered automated software architecture design, enabling the generation of modular and scalable software systems
13. Code generation techniques leveraging transformer-based models, generative adversarial networks (GANs), and other deep learning approaches

This minitrack is designed to facilitate cross-disciplinary dialogue, encouraging collaboration between researchers, software engineers, and industry professionals who are leveraging AI to transform software engineering. By bridging AI research with practical software development challenges, we aim to advance the capabilities of AI-powered program analysis and software synthesis, ultimately improving the reliability, security, and efficiency of modern software systems.

Important Dates for Paper Submission
June 15, 2025 | 11:59 pm HST: Paper Submission Deadline
August 17, 2025 | 11:59 pm HST: Notification of Acceptance/Rejection
September 22, 2025|11:59 pm HST: Deadline for Authors to Submit Final Manuscript for Publication
October 1, 2025 | 11:59 pm HST: Conference registration deadline for at least one author of each paper

Contacts
Ryan Karl | Software Engineering Institute, Carnegie Mellon University | rmkarl@sei.cmu.edu
Shen Zhang | Software Engineering Institute, Carnegie Mellon University | szhang@sei.cmu.edu
Yash Hindka | Software Engineering Institute, Carnegie Mellon University | yhindka@sei.cmu.edu

Copyright 2025 Carnegie Mellon University.

The view, opinions, and/or findings contained in this material are those of the author(s) and should not be construed as an official Government position, policy, or decision, unless designated by other documentation.

NO WARRANTY. THIS CARNEGIE MELLON UNIVERSITY AND SOFTWARE ENGINEERING INSTITUTE MATERIAL IS FURNISHED ON AN "AS-IS" BASIS. CARNEGIE MELLON UNIVERSITY MAKES NO WARRANTIES OF ANY KIND, EITHER EXPRESSED OR IMPLIED, AS TO ANY MATTER INCLUDING, BUT NOT LIMITED TO, WARRANTY OF FITNESS FOR PURPOSE OR MERCHANTABILITY, EXCLUSIVITY, OR RESULTS OBTAINED FROM USE OF THE MATERIAL. CARNEGIE MELLON UNIVERSITY DOES NOT MAKE ANY WARRANTY OF ANY KIND WITH RESPECT TO FREEDOM FROM PATENT, TRADEMARK, OR COPYRIGHT INFRINGEMENT.

[DISTRIBUTION STATEMENT A] This material has been approved for public release and unlimited distribution. Please see Copyright notice for non-US Government use and distribution.

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. Requests for permission for non-licensed uses should be directed to the Software Engineering Institute at permission@sei.cmu.edu.

DM25-0402