LLMs tested for automated assessment of student programming explanations

A new research paper explores the use of Large Language Models (LLMs) for automatically assessing student self-explanations in programming education. The study compares LLM-based scoring methods against traditional semantic similarity techniques, aiming to determine the most effective approach for evaluating student-generated content. This research addresses the challenge of accurately judging the correctness of student explanations, a crucial component for enhancing learning through worked examples. AI

IMPACT This research could lead to more efficient and scalable methods for evaluating student understanding in programming courses.

RANK_REASON The cluster contains an academic paper detailing a comparison of LLMs against existing methods for a specific educational task. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Arun Balajiee Lekshmi Narayanan
• Large Language Models
• LLMs
• Programming Education

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →