Date
Publisher
arXiv
Effective peer feedback is essential for developing critical reflection in higher education, yet its impact is often limited by the inconsistent quality of student-generated comments. This paper presents the implementation and deployment of AICoFe (AI-based Collaborative Feedback), a system designed to bridge this gap through a human-centered AI approach. We describe a modular architecture that orchestrates a multi-LLM pipeline, utilizing GPT-4.1-mini, Gemini 2.5 Flash, and Llama 3.1, to synthesize quantitative rubric data and qualitative observations into coherent, actionable feedback. Key to the system is a "teacher-in-the-loop" mediation workflow, where educators use specialized Learning Analytics dashboards to curate and refine AI-generated drafts before delivery. Furthermore, we detail the underlying data infrastructure, which employs a hybrid SQL and MongoDB strategy to ensure traceability and manage semi-structured feedback versions.
Study at a glance
- Question: How can a multi-model generative AI system support teacher-mediated collaborative feedback in higher education?
- Setting and data: Universidad Autónoma de Madrid, undergraduate and master's courses, approximately 80 students and 5 teachers.
- Method: System development and pilot deployment with fine-tuned large language models and Learning Analytics dashboards.
Why it matters
Peer feedback helps develop critical reflection, but its educational impact is often limited because students struggle to provide consistent, constructive, and actionable comments.
Key findings
- Students perceive the AI-assisted feedback as highly coherent and useful across both undergraduate and master's course levels.
- System usability evaluations indicate excellent user acceptance of the collaborative feedback platform across cohorts.
- Fine-tuning three distinct models (GPT-4.1-mini, Gemini 2.5 Flash, and Llama 3.1) on 50 evaluation instances aligned feedback production with specific course pedagogical objectives.
Summary drafted with AI from the full paper and may contain errors. See the full paper for details.
What is the application?
Who age?
Why use AI?
