Automated Grading of Linux/Bash Examinations Using Large Language Models

Manuel Alonso-Carracedo, Ruben Fernandez-Boullon, Pedro Celard, Francisco J. Rodriguez-Martinez, Lorena Otero-Cerdeira (2026) 📄 Full text (arXiv)

This paper presents an llm-based grading system for Linux/bash command-line examinations, applying a four-level cognitive taxonomy to assess student work in programming courses. The system addresses the scalability challenge of rising enrolments by providing automated-grading that captures both partial correctness and conceptual understanding. The approach demonstrates high agreement with human graders, suggesting a viable path toward formative-assessment at scale. Results indicate that LLMs can evaluate command-line proficiency more nuancedly than traditional rule-based autograders, offering detailed ai-feedback-quality feedback for students.

Related Pages

• automated-grading — Automated assessment approaches in computing education

• llm — Large language models and their applications

• k-12 — K-12 education context

• ai-literacy — AI literacy frameworks and assessment

• student-ai-interaction — How students interact with AI systems

• scaffolding — Instructional scaffolding techniques

• ai-generated-content — AI-produced educational materials

• learning-analytics — Data-driven analysis of learning behaviors

• formative-assessment — Formative assessment in AI-enhanced education

• active-learning — Active learning approaches

Citation

APA: Manuel Alonso-Carracedo, Ruben Fernandez-Boullon, Pedro Celard, Francisco J. Rodriguez-Martinez, Lorena Otero-Cerdeira (2026). Automated Grading of Linux/Bash Examinations Using Large Language Models. arXiv:2607.02432.