Yizhong Wang
Research Scientist, ByteDance Seed
Incoming Assistant Professor, UT Austin
About
I am a Research Scientist at ByteDance Seed, where I work on post-training language models and general agents. I am also an incoming Assistant Professor in the Department of Computer Science at the University of Texas at Austin, and part of the UT Austin NLP community.
I received my PhD from the University of Washington in 2025, advised by Hannaneh Hajishirzi and Noah Smith. During my PhD, I was a long-term student researcher at the Allen Institute for AI (AI2), where I led or co-led projects including Self-Instruct, Tülu, and OLMo. Before UW, I received my master's degree from Peking University and my bachelor's degree from Shanghai Jiao Tong University.
Research Themes
- Open LLM ecosystem. I am interested in the full stack of open language model development, including data, training, infrastructure, and evaluation. This direction builds on my work on Tülu and OLMo.
- Learning algorithms. I study how models can learn effectively from scalable signals, in both large-scale and data-efficient settings. This started with my work on instruction tuning, synthetic data generation, and reinforcement learning with verifiable rewards.
- Multi-agent systems. I am exploring how multi-agent systems can be structured or emerge through training, and the impact of such systems in doing actual work.
- AI for scientific discovery. I am keen on developing general AI methods that can empower real scientific discovery. I believe this requires new recipes beyond the current LLM/agent paradigm.
If you are interested in working with me on these topics, please read the Working with Me section below.
Selected Publications
* indicates equal contribution. For a full list, see my Google Scholar page.
Working with Me
I plan to recruit PhD students to start in Fall 2027 at the University of Texas at Austin. If your research interests align with mine (or something new you think might excite me), I strongly encourage you to directly apply to the UT Austin CS PhD program. Please mention my name as a potential advisor in your application. You don't have to email me to express interest.
I also welcome interest from undergraduates, masters, and postdocs. But because my bandwidth for individual email conversations is limited, please use the interest form below to share your background and research interests. My team members and I will review submissions periodically, but may not be able to respond to every inquiry.