CSC News

MerryQuery Wins Prestigious AAAI-2025 Demonstration Award for Advancing AI in Education

MerryQuery, an innovative AI-powered educational platform developed at NC State University’s Department of Computer Science, has been honored with the Best Technical Demonstration Award at the 39th Annual Association for the Advancement of Artificial Intelligence (AAAI) Conference on Artificial Intelligence (AAAI-2025). The award recognizes MerryQuery’s impact in enhancing trustworthy AI-driven support for educators and students through Retrieval-Augmented Generation (RAG), multimodal data integration and personalized learning assistance.

Advancing AI in Education with MerryQuery

MerryQuery is designed to bridge the gap between large language models (LLMs) and effective classroom integration, addressing concerns about academic integrity, misinformation and student over-reliance. Developed by Benyamin Tabarsi, Aditya Basarkar, Xukun Liu, Dongkuan (DK) Xu and Tiffany Barnes, MerryQuery leverages RAG to provide contextually relevant responses grounded in course-specific materials.

The platform stands out due to its guided dialogues, source citations and instructor-controlled response granularity, allowing educators to monitor student interactions while ensuring AI-generated content aligns with pedagogical goals. Additionally, MerryQuery’s multimodal data processing capabilities enable it to analyze complex learning resources, such as PDFs with embedded images and tables, ensuring accuracy and fidelity in AI-driven educational support.

Recognition at AAAI-2025

AAAI, one of the world’s leading conferences in artificial intelligence, hosted the Best Technical Demonstration Awards to celebrate innovations pushing the boundaries of AI research and application. The MerryQuery demonstration received widespread attention for its ability to enhance learning experiences while upholding academic integrity.

“MerryQuery represents a transformative step toward responsible AI integration in education,” said Barnes. “By balancing AI assistance with pedagogical oversight, it empowers both students and educators.”

“This recognition at AAAI-2025 underscores the importance of trust and reliability in AI-driven education,” said Xu. “MerryQuery is an example of how intelligent systems can adapt to educational needs while maintaining rigor and accountability.” 

A Collaborative Effort with Future Impact

The success of MerryQuery is a testament to the collaborative efforts of faculty, students and interdisciplinary research at NC State. Benny Tabarsi, first author and a Ph.D. student co-advised by Barnes and Xu, presented the demonstration at AAAI-2025, engaging with AI and education researchers from around the world.

With this recognition, the team aims to further enhance MerryQuery by integrating reinforcement learning from human feedback (RLHF), automated grading features and adaptive learning pathways. The platform is set to expand its reach beyond computing education to other disciplines, reinforcing its position as a leading AI-powered educational tool.

Collaborating Faculty

MerryQuery's development was further enriched by the contributions of collaborating faculty members who bring diverse expertise to the project:

• Lina Battestilli, full teaching professor, NC State Department of Computer Science: Specializing in computing education and pedagogical innovation, Battestilli brings extensive experience in designing technology-driven learning environments.

• Joey Huang, assistant professor, Learning Design and Technology, NC State College of Education: With expertise in educational technology and digital learning strategies, Huang contributes insights into enhancing AI-driven learning experiences.

• Shiyan Jiang, assistant professor, Learning Design and Technology, NC State College of Education: Focused on AI-driven personalized learning, Jiang’s research helps shape adaptive and intelligent learning methodologies within MerryQuery.

For more details, visit the MerryQuery Platform.

###

Adapted for the NC State Department of Computer Science