From Handwriting to Feedback: Evaluating VLMs and LLMs for AI-Powered Assessment in Indonesian Classrooms
This work addresses the problem of AI-driven educational assessment in underrepresented classrooms, though it is incremental as it applies existing methods to new data.
The study evaluated VLMs and LLMs for AI-powered assessment in Indonesian classrooms using over 14K handwritten answers, finding that VLMs struggled with handwriting recognition, which affected LLM grading, but LLM feedback was still pedagogically useful despite limitations in personalization and contextual relevance.
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.