Mathemyths: Leveraging Large Language Models to Teach Mathematical Language through Child-AI Co-Creative Storytelling

Mathemyths: использование больших языковых моделей для обучения математическому языку посредством совместного творчества детей и искусственного интеллекта при создании историй
Chao Zhang, X. Liu, Katherine Ziska, Soobin Jeon, Chi‐Lin Yu, Ying Xu
2024-05-11

child-AI co-creative storytellingeducational conversational agentslarge language modelsmathematical language learningprompt engineering
Mathematical language is a cornerstone of a child’s mathematical development, and children can effectively acquire this language through storytelling with a knowledgeable and engaging partner. In this study, we leverage the recent advances in large language models to conduct free-form, creative conversations with children. Consequently, we developed Mathemyths, a joint storytelling agent that takes turns co-creating stories with children while integrating mathematical terms into the evolving narrative. This paper details our development process, illustrating how prompt-engineering can optimize LLMs for educational contexts. Through a user study involving 35 children aged 4-8 years, our results suggest that when children interacted with Mathemyths, their learning of mathematical language was comparable to those who co-created stories with a human partner. However, we observed differences in how children engaged with co-creation partners of different natures. Overall, we believe that LLM applications, like Mathemyths, offer children a unique conversational experience pertaining to focused learning objectives.
1
Children engaged differently with AI and human co-creation partners, indicating that partner nature influences storytelling interaction.
2
In a user study with 35 children aged 4–8, mathematical-language learning with Mathemyths was comparable to learning through co-creative storytelling with a human partner.
3
LLM-based storytelling applications can provide children with distinctive conversational experiences tailored to specific learning objectives.
4
Mathemyths is a joint storytelling agent that co-creates free-form narratives with children while integrating mathematical terminology.
5
Prompt engineering was used to optimize a large language model for child-oriented educational storytelling and focused mathematical-language learning.

Mathemyths joint storytelling agent (LLM-based child–AI co-creative storytelling system)

Children’s acquisition of mathematical language and their co-creative engagement during child–AI storytelling

Publication Details
Publication Date
2024-05-11
Journal
Publisher
ISSN
Cited by
89
Access Type
Author Information
Authors
Chao Zhang
X. Liu
Katherine Ziska
Soobin Jeon
Chi‐Lin Yu
Ying Xu
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat
Make a presentation
100%