AGACCI : Affiliated Grading Agents for Criteria-Centric Interface in Educational Coding Contexts

TL;DR

AGACCI introduces a multi-agent system that enhances the accuracy, interpretability, and consistency of code assessment in education by distributing evaluation tasks among specialized agents, outperforming single-model baselines.

Contribution

This paper presents AGACCI, a novel multi-agent framework for structured, criteria-centric evaluation of programming assignments in educational contexts.

Findings

AGACCI outperforms GPT-based baseline in accuracy and consistency.

The system maintains instructional intent and evaluative depth.

Performance varies across different task types.

Abstract

Recent advances in AI-assisted education have encouraged the integration of vision-language models (VLMs) into academic assessment, particularly for tasks that require both quantitative and qualitative evaluation. However, existing VLM based approaches struggle with complex educational artifacts, such as programming tasks with executable components and measurable outputs, that require structured reasoning and alignment with clearly defined evaluation criteria. We introduce AGACCI, a multi-agent system that distributes specialized evaluation roles across collaborative agents to improve accuracy, interpretability, and consistency in code-oriented assessment. To evaluate the framework, we collected 360 graduate-level code-based assignments from 60 participants, each annotated by domain experts with binary rubric scores and qualitative feedback. Experimental results demonstrate that AGACCI outperforms a single GPT-based baseline in terms of rubric and feedback accuracy, relevance, consistency, and coherence, while preserving the instructional intent and evaluative depth of expert assessments. Although performance varies across task types, AGACCI highlights the potential of multi-agent systems for scalable and context-aware educational evaluation.