The AURA Lab

(A)I for (U)nderstandable and (R)esponsible (A)utomation in Software Engineering

prof_pic.png

The AURA Lab, founded in Fall 2024, advances research at the intersection of efficiency, automation, and explainability in software engineering. Our mission is to design innovative and responsible AI-driven solutions that improve developer productivity while addressing key challenges of transparency and resource efficiency. Through rigorous methods, open collaboration, and a commitment to impactful work, the AURA Lab contributes practical advances in AI-supported software engineering for academia, industry, and society.

news

Jun 15, 2025 🔥 Exciting update! Our paper, "Is Quantization a Deal-breaker? Empirical Insights from Large Code Models", is now available on arXiv 🎉. 👏 Shout out to: Saima Afrin, Bowen Xu, and Antonio Mastropaolo. 🚀
Apr 28, 2025 🌟 🔥 Fantastic news! The position paper co-authored with Denys Poshyvanyk, "A Path Less Traveled: Reimagining Software Engineering Automation via a Neurosymbolic Paradigm", has been accepted for publication in the AI-SDLC 2025
The workshop is co-located with the ACM International Conference on the Foundations of Software Engineering (FSE'25) -- Trondheim, Norway 🇳🇴
Apr 15, 2025 🎉 Congratulations to Ben Trembley for receiving W&M’s Charles Center Summer Research Grant! 👏 Ben’s research explores the intersection of software engineering, AI, and optimization, with a focus on cutting the cost of training large language models that enables software development automation (e.g., GitHub Copilot and ChatGPT). The work addresses the urgent need for resource‑efficient scaling as models grow to trillions of parameters.
Apr 15, 2025 🏆🏆🏆 Big Win!!! -- Thrilled to annource our NSF CRII SHF: Advancing Sustainable Software Engineering Practices with Energy-Efficient Large Language Models for Code (2451058)
PI: Antonio Mastropaolo. June 1, 2025 - May 31, 2027
Jan 13, 2025 🎉 Exciting news for Saima! She begins her PhD journey with a publication in the FORGE 2025 Research Track! 🚀 👏 Stay tuned for the pre-print, "Resource-Efficient & Effective Code Summarization". 🌟