My on-going studies have shown that AI conversational systems hold promise to create a more enriching learning experiences for children. Yet the development of such systems relies on extensive manual labor of content creators, designers, and engineers, which may have made large-scale, cost-effective development and production of such learning resources quite difficult.
To enable scalable development of conversational AI systems, I am carrying out research to explore the feasibility of using automatic question-answer (QA) generation to facilitate the learning and assessment of narrative comprehension skills. There are three components for this long-term project:
- Creating a high-quality QA pair dataset for model training and evaluation (paper in ACL 2022);
- Developing AI models that expand the state-of-the-art deep neural network techniques (e.g., BERT) for machine comprehension tasks over the created dataset, and customizing the training to meet the unique requirements of an educational context (paper in ACL 2022);
- Building an interactive reading system with QA functionalities (i.e., the agent asking students questions and assessing their answers) to enhance and evaluate students’ comprehension (paper in CHI 2022).
Xu, Y., Wang, D., Yu, M., Ritchie, D., Yao, B., Wu, T., Zhang, Z., Li, T., Bradford, N., Sun, B., Hoang, T., Sang, Y., Hou, Y., Ma, X., Yang, D., Peng, N., Yu, Z., & Warschauer, M. (2022). Fantastic questions and where to find them: FairytaleQA— An authentic dataset for narrative comprehension. Association for Computational Linguistics. [Pre-print]
Zhang, Z., Xu, Y., Wang Y., Yao, B., Ritchie, D., Wu, T., Yu, M., Wang, D., & Li, T. (2022). Storybuddy: A human-AI collaborative agent for parent-child interactive storytelling. In Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems. [DOI]
Yao, B., Wang, D., Wu, T., Zhang, Z., Li, T., Yu, M., & Xu, Y. (2022). It is AI’s Turn to Ask Humans a Question: Question and Answer Pair Generation for Children’s Storybooks with FairytaleQA Dataset. Association for Computational Linguistics. [Pre-print]