Yiyang (Wendy) Wu | 吴抑扬

I am first year MIIS master student at Carnegie Mellon University. In my undergrad at University of Southern California, I worked in LIME Lab with Prof. Jieyu Zhao. I majored in Computer Science (CS) and double minored in Psychology and Economics (my previous majors before Computer Science).
My research focuses on advancing reinforcement learning frameworks to enhance the reasoning and alignment of large language models, while critically examining their gaps and inherent limitations compared to human cognition.

Email  /  Google Scholar  /  Github  /  LinkedIn  /  CV

Research

This paper presents AdaRFT, a lightweight extension to reinforcement finetuning that adaptively adjusts task difficulty based on reward signals, reducing training steps by up to 2× while boosting reasoning accuracy on competition-level math benchmarks.

Experience

Machine Learning Engineer Intern : TikTok Inc., Trust & Safety Team, San Jose, CA (May – Aug 2026)
Owned the post-training loop (data, SFT, RL, self-distillation, eval) for a model that moderates video and re-adjudicates the moderation system's own false positives.

Award

Academic Achievement Award : University of Southern California, 2023

Service

Course Producer : CSCI270: Introduction to Algorithms and the Theory of Computing (Sp24, Fa24, Sp25), CS170: Discrete Methods in Computer Science (Sp23) at USC

Inspired by the template here.