Problem
Research questions and friction points this paper is trying to address.
AI-enabled coding agents
code understanding
academic integrity
professional accountability
personalized assessment
Personalized Assessments from Personal Artifacts
🤖 AI Summary
The widespread adoption of AI-assisted programming risks undermining academic integrity and professional accountability, as developers may lack genuine understanding of the code they submit. To address this challenge, this work proposes the personalized probing puzzles ($p^3$) approach—the first method to leverage automated question generation for assessing an individual’s comprehension of their own code. By integrating static code analysis with the synthesis of concise, targeted puzzles and enabling asynchronous delivery, $p^3$ facilitates rapid and personalized verification of conceptual understanding. A pilot deployment in a graduate-level cloud computing course demonstrated that $p^3$ effectively uncovered students’ blind spots in code comprehension, confirming its feasibility and efficacy. This approach offers a novel paradigm for evaluating programming proficiency and upholding academic integrity in the era of AI-augmented development.
📝 Abstract
The rapid development and popularization of AI-enabled coding agents have meant software engineering students and professionals cannot be assumed to understand their own code, which risks academic integrity and professional accountability. We developed a method called Personalized Probing Puzzles ($p^3$) to evaluate students' understanding of their own code, and tested $p^3$ in a graduate-level cloud computing course. Our pilot study shows that $p^3$ can help identify potential gaps in students' understanding of their own code. The puzzles are automatically generated, asynchronously administered, and finished in minutes. Future work is needed to correlate puzzle results with code understanding and to embed $p^3$ in a professional code review process.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Personalized Probing Puzzles
code understanding assessment
AI-enabled coding agents
automated puzzle generation
academic integrity
🔎 Similar Papers
No similar papers found.
