ADAPT Researcher Claire McNamara has paper accepted to ACM Intelligent User Interfaces Conference (IUI) 2026

ADAPT Researcher and Trinity School of Computer Science and Statistics Ph.D. student Claire McNamara’s paper “Filling the Gap: LLMs as Scaffolds for Competency Question Instantiation” has been formally accepted to the main track of ACM Intelligent User Interfaces Conference (IUI) 2026.

The paper examines the use of large language models (LLMs) as scaffolds rather than autonomous generators for structured question-answer instantiation over large knowledge graphs, with a focus on accuracy, provenance, robustness, and their subsequent potential impact on interface design.  In the process, Claire undertook a large-scale evaluation of The Virtual Record Treasury of Ireland knowledge graph (24,900 generated question-answer pairs across four LLMs) and her natural language question and answer curation framework, Tús Maith, designed and implemented as part of her Ph.D. research.

The ACM Conference on Intelligent User Interfaces (ACM IUI) is the annual premiere venue where researchers and practitioners meet and discuss state-of-the-art advances at the intersection of Artificial Intelligence (AI) and Human-Computer Interaction (HCI). The 2026 conference will be held in Paphos, Cyprus.

Claire McNamara is part of ADAPT as a Ph.D. student under the supervision of Declan O’Sullivan at Trinity College Dublin. Her research consists of opening up the possibilities of Linked Data to domain experts. Claire graduated from Trinity College Dublin in 2021.