We propose a level- and domain-adaptive explanation framework for controllable explanation generation in LLMs. A generator-verifier loop with structured outputs reduces hallucination by 50%, and a user-adaptive prompting mechanism enables multi-dimensional control of explanation difficulty. A user study (n=41) demonstrates improved clarity, usability, and domain-specific understanding over baseline tools.
@inproceedings{adapdict2026,title={AdapDict: A Level- and Domain-Adaptive Educational Dictionary and Encyclopedia System},author={Han, Manhwi and Back, SeungYeon and Chae, Dong-Kyu},booktitle={Proceedings of the ACM on Human-Computer Interaction (CSCW)},year={2026},note={Accepted; Co-first author},}
IEEE Access
An Empirical Study of the Clustering of Mountain Hiking GPS Trajectory Data
SeungYeon Back, Manhwi Han, Dong-Kyu Chae, and 1 more author
We propose a novel trajectory similarity measure integrating spatial, temporal, and velocity components for irregular spatio-temporal data, along with an epsilon-ball-based spatial similarity metric and a data-adaptive weighting framework (alpha, beta, gamma). Hierarchical clustering is applied to large-scale GPS trajectory data to uncover latent behavioral patterns.
@article{gpshiking2026,title={An Empirical Study of the Clustering of Mountain Hiking GPS Trajectory Data},author={Back, SeungYeon and Han, Manhwi and Chae, Dong-Kyu and Park, Seoncheol},journal={IEEE Access},year={2026},note={Under revision; Co-first author},}