A reinforcement learning approach that guides reasoning models toward shorter, more efficient chains of thought while preserving or improving accuracy.
Justin's Homepage
I am Jiazheng (Justin) Wang, an M.S. student in Data Science at the University of Chicago. My work sits around LLM reasoning, reinforcement learning, and data-centric AI.
Papers and projects.
A controlled study showing that strategically diverse instruction data is central to generalization on unseen tasks.
A proof-oriented programming system trained with synthetic and repository-level data to improve formal code generation and repair under severe data scarcity.
Education, internships, and teaching.
2024–2026
University of Chicago
M.S. in Data Science
2024–2025
Data Science Clinic Teaching Assistant
Co-mentored undergraduate students on cell segmentation, annotation, and model fine-tuning for biological imaging.
2023
2021–2024
University of Illinois Urbana-Champaign
B.S. in Mathematics and Computer Science, with highest distinction.
Outside research.
Outside research, I enjoy playing guitar, photography, and music production.
I occasionally share music sketches online.
- Guitar
- Photography
- Music production
