Automated grading of Linux/bash examinations using large language models: a four-level cognitive taxonomy approach

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Manuel Alonso-Carracedo, Ruben Fernandez-Boullon, Pedro Celard, Francisco J. Rodriguez-Martinez, Lorena Otero-Cerdeira
• Venue: arXiv: cs.AI
• Published: 2026-07-02
• Collected: 2026-07-03
• URL: http://arxiv.org/abs/2607.02432v1
• PDF: https://arxiv.org/pdf/2607.02432v1
• DOI: UNKNOWN

Plain-Language Brief

The paper evaluates if four large language models can accurately grade Linux/bash command responses in a way that approximates expert judgment. The authors use a four-level cognitive taxonomy to test the models on real responses from second-year Computer Engineering students, with Gemini~3.0 Pro showing the highest agreement with human graders. The study found that question complexity is a reliable predictor of difficulty for the large language models and provides a framework for determining when AI-assisted grading is suitable.

What the Document Says

• The paper evaluates four Large Language Models (GPT, Claude Opus, Gemini, and GLM) for grading Linux/bash command responses.
• A four-level cognitive taxonomy is used to test the models on real student responses.
• Gemini~3.0 Pro with rubric-guided prompting had the highest human-AI agreement.
• Agreement declined as taxonomy level increased and was larger at higher levels.
• Rubric quality had a larger effect than provider choice.
• Structured prompts improved agreement across all models.

Key Concepts Mentioned

• Large Language Models
• Linux/bash command responses
• cognitive taxonomy
• human-AI agreement
• rubric
• prompt

Claims Made by the Document

• Gemini~3.0 Pro with rubric-guided prompting achieved the highest human-AI agreement.
• Question complexity is a reliable predictor of difficulty for large language models.

Evidence Used

• ICC(3,1) = 0.888, MAE = 0.10, Bland-Altman bias = -0.014

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Scalable and reliable grading of command-line examinations remains a challenge in computing education, where rising enrolments make manual marking difficult and rule-based autograders cannot handle partial credit, equivalent solutions, or syntactic variation. This paper evaluates whether four frontier Large Language Models (GPT, Claude Opus, Gemini, and GLM) can approximate expert judgment when grading short Linux/bash command responses. The study adopts a four-level cognitive taxonomy that combines cognitive complexity and operational impact, ranging from information retrieval (L1) and basic file manipulation (L2) to structural operations (L3) and advanced system management (L4). The models were tested with two prompt variants, a minimal baseline and a rubric-enhanced version, on 1200 real responses from second-year Computer Engineering students independently graded by three expert instructors. Gemini~3.0 Pro with rubric-guided prompting achieved the highest human-AI agreement (ICC(3,1) = 0.888, MAE = 0.10, Bland-Altman bias = -0.014). Agreement declined consistently as taxonomy level increased, with the largest discrepancies at higher levels. Across all models, rubric quality had a larger effect than provider choice, with structured prompts consistently improving agreement. These results show that question complexity is a reliable predictor of the difficulty LLMs face in grading accurately, and they establish a principled, taxonomy-based framework for determining which questions are suitable for AI-assisted grading and which require human review, while also providing a transferable evaluation protocol and prompt templates.