Yekyung Kim email scholar github
Toward faithful evaluation and alignment of long-context language models.
I am a third-year Ph.D. student at the University of Maryland, CLIP Lab, advised by Mohit Iyyer. My research in NLP asks how to evaluate and align language models as everything gets long — across long-context understanding, long-form generation, and long-horizon problem-solving. I began my Ph.D. at UMass NLP and later transferred to UMD along with my advisor.
Evaluating faithfulness & factuality — in long-context understanding (FABLES, OneRuler) and long-form generation (VeriScore), including recent work on argument collapse.
Aligning language models — post-training with synthetic data for instruction following (BLEUBERI) and compositional reasoning (ongoing work).
Before my Ph.D., I worked at Hyundai Motor Group and LG Electronics as a research engineer. I was selected as a specialist in AI and conducted research at CMU LTI as a visiting scientist mentored by Jaime Carbonell.
