Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note2 min read

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

Source

arXiv

arxiv.org/abs/2607.02432v1

Read the source

Source

Plain-Language Brief

This research evaluates the effectiveness of four large language models in grading Linux/bash command responses from second-year Computer Engineering students' exams. The study uses a four-level cognitive taxonomy approach to grade the responses, with Gemini~3.0 Pro achieving the highest human-AI agreement when using rubric-guided prompts.

What the Document Says

  • The paper evaluates Large Language Models (LLMs) for grading Linux/bash command responses in computing education.
  • A four-level cognitive taxonomy approach combines cognitive complexity and operational impact, ranging from information retrieval to advanced system management.
  • Gemini~3.0 Pro with rubric-guided prompting achieved the highest human-AI agreement among the tested models.
  • The study used 1200 real responses from second-year Computer Engineering students.
  • Three expert instructors independently graded the responses manually for comparison.
  • Agreement declined consistently as taxonomy level increased, with the largest discrepancies at higher levels.
  • Rubric quality had a larger effect than provider choice in improving agreement.
  • Structured prompts consistently improved agreement across all models.
  • The results show that question complexity is a reliable predictor of the difficulty LLMs face in grading accurately.

Key Concepts Mentioned

  • Large Language Models
  • Linux/bash command responses
  • Computing education
  • Cognitive taxonomy
  • Grading
  • Question complexity
  • Agreement
  • Rubric-guided prompting
  • Gemini~3.0 Pro

Claims Made by the Document

  • LLMs can approximate expert judgment when grading Linux/bash command responses.
  • Question complexity is a reliable predictor of the difficulty LLMs face in grading accurately.

Evidence Used

  • The study reports agreement scores and bias between human and AI grading.
  • The results show that agreement declined consistently as taxonomy level increased.

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.