LLM Powered Teacher
Supporting Teachers and Students in Personalized Learning to Enhance Recall, Retention, and Memorization.
Project Details
Client: Saudi Data and AI Authority (SDAIA)
Start Day: 01/09/2024
Tags: Education
Project Duration: 2 Month
Client Website: https://allam.tuwaiq.edu.sa/en
Let’s Work Together for Development
Call us directly, submit a sample or email us!
The Story
In July 2024, SDAIA announced a challenge aimed at supporting the Arabic LLM (ALLaM) through innovative ideas. At MozaicAI, we saw this as an exciting opportunity to expand our work into the Middle Eastern market. In response, we submitted a proposal for an LLM-powered teacher designed to help educators transfer their teaching styles into our LLM-based system and support students in engaging with the learning process by considering cognitive structures.
Our approach focuses on three key pillars:
- Understanding: Simplifying text based on students’ interests to enhance comprehension.
- Memorization: Leveraging text-to-speech technology to aid retention.
- Recall: Using visual mnemonics, such as text-to-image illustrations, to make ideas memorable and easily retrievable.
Our proposal received positive feedback and was officially accepted. Development began in September and October 2024, during which we worked diligently to bring our ideas to life. From November 7th to 9th, we were honored to present our work in Riyadh at the Crown Plaza meeting rooms.
Evaluation: User Study with End Users
The Results
We conducted both qualitative and quantitative analyses to assess the significant impact of our LLM-powered teacher. The quantitative results include performance evaluations through quizzes and a System Usability Scale (SUS) questionnaire. On the other hand, the qualitative findings highlight what students enjoyed most about the application and their suggestions for improvements in future versions.
