🧠 AI Ed Wiki

📄 Research Article

AICoFE: AI-Powered Feedback System

◐ medium · 2026-07-29ai-ed-evaluation, feedback-loop, student-experience, human-in-the-loop-ai, higher-ed, learning-analytics, formative-assessment

AICoFE (AI-based Collaborative Feedback) is a multi-LLM feedback generation system for higher education that combines independently fine-tuned language models with teacher-in-the-loop mediation, producing diverse feedback perspectives while preserving pedagogical authority through Learning Analytics dashboards.

Key Findings

1. Multi-LLM diversity improves feedback quality. AICoFE uses three independently fine-tuned LLMs (GPT-4.1-mini, Gemini 2.5 Flash, Llama 3.1) to generate diverse feedback perspectives from the same input data — quantitative rubric scores, validated qualitative observations, rubric level descriptions, and instructional materials. Each model produces independent drafts that the teacher can inspect and curate.

2. Teacher mediation is central, not an afterthought. Rather than automating feedback end-to-end, the system positions educators as active curators. The teacher dashboard enables instructors to compose final feedback by selecting individual sentences or paragraphs from AI outputs, with a visual legend indicating the proportion of content contributed by each LLM. This preserves pedagogical judgment while reducing feedback workload.

3. Dual-purpose transparency supports both practice and research. The system tracks the extent of teacher modification and curation, serving simultaneously as (a) a reflection tool for individual instructors examining their curation patterns, and (b) a dataset for large-scale analysis of the teacher mediation role in AI-assisted feedback.

4. Role-specific dashboards close the feedback loop. The student dashboard provides access to video recordings of presentations, self-evaluation rubrics, visual comparisons of self vs. external evaluations, teacher-curated AI feedback, and the ability to rate perceived agreement and usefulness — completing a full assessment-for-learning cycle.

Implications

AICoFE addresses a critical tension in AI Feedback Quality: the trade-off between efficiency and pedagogical authority. By adopting a Human In The Loop AI architecture rather than full automation, it demonstrates that AI can accelerate feedback workflows without displacing the teacher's interpretive role. This aligns with emerging best practices in AI Feedback Quality research, which increasingly emphasizes teacher agency over raw automation.

The multi-LLM pipeline is a practical response to the observation that different models excel at different dimensions of feedback — tone, specificity, actionability. Rather than selecting a single "best" LLM, AICoFE treats model diversity as a feature, surfacing complementary perspectives that the teacher can synthesize. This approach parallels work in AI Peer Feedback Systems that leverages multiple AI-generated perspectives for richer formative assessment.

The Learning Analytics component of AICoFE is notable for its role-specific design: teacher and student dashboards serve fundamentally different functions within the same platform, and the system's transparency features (source attribution, curation tracking) support both reflective practice and Learning Analytics at scale.

For Higher Ed institutions facing growing assessment loads, AICoFE models a path where AI-assisted feedback scales without sacrificing the relational and interpretive dimensions of effective Formative Assessment.

Connected Concepts

  • Formative Assessment
  • Higher Ed
  • Human In The Loop AI
  • Learning Analytics
  • AI Ed Evaluation
  • Student Experience

    • Connected Articles

  • AI Generated Feedback Higher Ed — Artificial intelligence and feedback in university education: effectiveness and student perceptions
  • AI Peer Feedback Systems — AI Peer Feedback Systems
  • A4l Analytics Pipeline — Generalizing a Highly Configurable Analytics Pipeline to Replicate and Support Educational Research Across Multiple D...
  • Aaai2026 Prompting Literacy K12 — Learning to Use AI for Learning: Teaching Responsible Use of AI Chatbot to K-12 Students Through an AI Literacy Module
  • Academiclaw Student Agent Benchmark — AcademiClaw: When Students Set Challenges for AI Agents
  • Access Not Enough AI Tutoring 2026 — Access is Not Enough: Human Support Improves Engagement with AI Tutoring
  • Adaptive Pretesting Retention — Do Gains from Generative AI-Enabled Adaptive Pretesting Persist? Evidence from a Retention Study
  • Affective Text Wearable Student Health — A Formative Study of Brief Affective Text as a Complement to Wearable Sensing for Longitudinal Student Health Monitoring
  • Agency Gap AI Writing — The agency gap in AI-supported writing: how reactive and proactive agent designs shape multimodal reasoning
  • Agent Voice Accents K12 Group Learning — Exploring How Agent Voice Accents Shape Human-AI Collaboration in K-12 Group Learning
  • Agentic AI Education Scoping Review — Agentic AI in Education: A Scoping Review of Research Landscape, Capabilities, and the Frontier Agent Paradigm
  • Agentic AI Pedagogical Best Practice 2026 — Agentic AI and Pedagogical Best Practice: The Tension Between Automation and Learning
  • Agentic Education Coding — Agentic Education with AI Coding Assistants
  • Agentic Literacy Debt — Agentic Literacy Debt: A Structural Problem the AI Literacy Field Has Not Yet Named
  • Agents That Teach Incidental Learning — Agents That Teach: Designing Incidental Learning Back into AI-Assisted Software Development
  • Agreement Not Quality LLM Coding Verification — Agreement Is Not Quality: Blind Expert Verification of Human and LLM Qualitative Coding When Human Consensus Is Not G...
  • AI Adult Learning Guidelines Dis2026 — Guidelines for Designing AI Technologies to Support Adult Learning
  • AI Agents Constructive Conflict Design Education 2026 — Enacting Constructive Conflicts with AI Agents to Enhance Reconsideration among Novice Interaction Designers
  • AI Agents Peer Learning Discourse — When AI Agents Teach Each Other: Discourse Patterns Resembling Peer Learning in the Moltbook Community
  • AI Assessment Human Tutors — AI-Driven Assessment of Human Tutors: Linking Training Performance to Real-Life Practice
  • AI Assessment Scale Reform — A bit of chaos and madness": The AI Assessment Scale and the work of assessment reform
  • AI Assistance Discretionary Feedback — AI Assistance for Discretionary Work: Increasing Feedback Provision in Higher Education
  • AI Assisted Learning Modes Eeg — An exploratory behavioral and electroencephalographic study of artificial intelligence-assisted learning modes in hig...
  • AI Assisted Se Curriculum Syllabus Analysis 2026 — Mapping the Emerging Curriculum for AI-Assisted Software Engineering via Syllabus Analysis
  • AI Assisted Writing Research Teams — Smaller, Younger, and More Impactful: How AI-Assisted Writing Transforms Research Teams

    • Citation

    Becerra, Á., Palma, A., & Cobos, R. (2026). AICoFe: AI-Based Collaborative Feedback System for Higher Education. arXiv:2605.04740.