class Jiayou:
name = "Jiayou (Sam) Zhong"
currently = [
"Master of Mathematical Finance student @ University of Toronto",
"NLP Research Assistant",
]
incoming = "Quantitative Developer @ BMO"
previously = [
"Bachelor of Computer Science @ University of Waterloo",
"Machine Learning Engineer Intern @ Health Canada",
"Software Engineer Intern @ Health Canada",
]
skills = ["Python", "C++", "SQL", "Machine Learning", "DevOps", "Cloud Platform"]
interested_in = ["Quant Development", "Quant Research", "Software Engineering"]
favorite_domain = "Fintech"Bachelor of Computer Science @ U of Waterloo · Master of Mathematical Finance @ U of Toronto
AI alignment is the work of making AI systems behave in ways that reflect human goals and values. My research focuses on pluralistic alignment: helping language models account for different people's perspectives, especially when there is no single answer that works for everyone.
EMNLP 2025 · Pluralistic Alignment for Healthcare: A Role-Driven Framework
Introduces ETHOSAGENTS, a framework that simulates different roles and values to help language models respond to the diverse perspectives found in healthcare.
EMNLP 2026 · VISPA: Pluralistic Alignment via Automatic Value Selection and Activation
Introduces VISPA, a training-free approach that selects relevant values and steers a model's internal activations to better reflect them in its responses.
