James Yang is a PhD candidate and researcher in computer and human vision based in Bethesda, Maryland, with eight years of experience bridging vision-language modeling, interpretability, and multimodal AI. His work at the University of Minnesota focuses on building more robust LVLMs and creating benchmarks to diagnose model distractions and hallucinations, informed by several large-scale human attention datasets he helped develop (IQVA, AiR, VisualHow). Trained in both computer science and statistics, he brings quantitative rigor to model diagnostics and has practical experience packaging research into reproducible tools, including an R package published during earlier research. James combines academic depth with hands-on teaching and research-assistant roles, making him equally comfortable explaining algorithms in classrooms and designing experiments that probe human–machine perception. Notably, his project portfolio emphasizes interpretable model learning and real-world evaluation strategies that prioritize reliability over raw performance.
9 years of coding experience
Bachelor of Arts - BA, Double major in Computer Science and Statistics, Bachelor of Arts - BA, Double major in Computer Science and Statistics at Carleton College
High School Degree, High School Degree at Breck School
Contributions:11 commits, 1 PR, 762 pushes in 4 months
reactcssjavascriptfoodcalculator
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.