What is it about?
Educators want their courses to do more than cover the basics of a single subject. Many want to connect their material to broader goals, such as the United Nations Sustainable Development Goals and UNESCO AI Competency Framework. However, this often requires educators to engage with domains beyond their primary expertise, a demand on time and effort many cannot meet. Consequently, these cross-cutting objectives are frequently overlooked. This work introduces Earthovia 2.0, a software designed to makes this integration possible and easier. Educators can upload their existing syllabus and select or add learning goals they would like to include. The software then helps them redesign the course using a method called backward design, where you start with learning objectives and work backward to the assessments and activities to achieve those goals. To keep its suggestions accurate and trustworthy, the tool does not rely on a general purpose Large language model (LLMs). Instead uses a technique called Retrieval augmented generation, which pulls in the actual text of official educational frameworks before generating any course content. This reduces the hallucinated and misleading answers that LLMs can produce. We conducted a preliminary study with three computer science faculty members. They found the tool useful in places, building homework and edits they were willing to adopt, while also noting that it sometimes forced connections that felt unnatural. The results point to real promise alongside clear room for improvement.
Featured Image
Photo by KOBU Agency on Unsplash
Why is it important?
Most AI tools built for education are aimed at students. Far fewer are built to support the educators who design courses, and almost none focus on the specific challenge of integrating interdisciplinary goals that fall outside a teacher's main field. This work targets that gap directly.
Perspectives
Earthovia 2.0 began with a simple observation: the goals we claim to value most, like Sustainable Development Goals and AI Competency Goals, are often overlooked because the additional time and effort required to add these objectives. Closing that gap is what drove this work. What surprised me was how much the faculty feedback reshaped my thinking. Earthovia 2.0 proved helpful in some moments and clumsy in others, which revealed the reasoning capabilities I need to improve in future work.
Ronak Wani
Worcester Polytechnic Institute
Read the Original
This page is a summary of: RAG-Powered Backward Curricula Design Software for Integrating Interdisciplinary Learning Objectives, March 2026, ACM (Association for Computing Machinery),
DOI: 10.1145/3748522.3785296.
You can read the full text:
Contributors
The following have contributed to this page
