Evaluating Vision-Language and Large Language Models for Automated Student Assessment in Indonesian Classrooms

TL;DR

This study evaluates the performance of vision-language and large language models on real-world handwritten student answers in Indonesian classrooms, highlighting their strengths and limitations in educational assessment tasks.

Contribution

It provides the first comprehensive evaluation of VLMs and LLMs on naturally handwritten student responses in a real-world educational setting, revealing practical challenges.

Findings

VLMs struggle with handwriting recognition, affecting grading accuracy.

LLMs can generate pedagogically useful feedback despite visual input errors.

The dataset presents realistic handwriting and linguistic challenges not covered in prior work.

Abstract

Despite rapid progress in vision-language and large language models (VLMs and LLMs), their effectiveness for AI-driven educational assessment in real-world, underrepresented classrooms remains largely unexplored. We evaluate state-of-the-art VLMs and LLMs on over 14K handwritten answers from grade-4 classrooms in Indonesia, covering Mathematics and English aligned with the local national curriculum. Unlike prior work on clean digital text, our dataset features naturally curly, diverse handwriting from real classrooms, posing realistic visual and linguistic challenges. Assessment tasks include grading and generating personalized Indonesian feedback guided by rubric-based evaluation. Results show that the VLM struggles with handwriting recognition, causing error propagation in LLM grading, yet LLM feedback remains pedagogically useful despite imperfect visual inputs, revealing limits in personalization and contextual relevance.