Yu Ying Chiu (Kelly Chiu)
MS student at University of Washington, Research Associate at New York University.
kellycyy [at] uw [dot] edu
I’m a final-year master’s student in Computational Linguistics (NLP) at the University of Washington, advised by Prof. Yejin Choi and Prof. Tim Althoff. I also worked closely with Prof. Shane Steinert-Threlkeld.
I was a ML Alignment & Theory Scholar (MATS), worked with Evan Hubinger from Anthropic on AI values evaluations (LitmusValues).
Currently, I’m a Research Associate at New York University, working with Prof. Sydney Levine from NYU/Google DeepMind and Prof. Mitchell Gordon from MIT/OpenAI. Check out our recent work, MoReBench (a PhD-level, rubric-based moral reasoning benchmark), in collaboration with Scale AI!
My research focuses on the intersection of Human-centered AI and AI Safety. Broadly, I’m intrigued by how humans and AI understand the world, think and make important decisions, and how we can use these insights to better support humans using AI and ensure AI can benefit humanity . My goal is to build and enhance moral, empathetic, and culturally sensitive AI, with a current focus on real-world data collection, human-AI collaboration and value-alignment evaluation.
Always happy to connect with people who are excited to chat or collaborate on AI value alignment and AI safety research!
