Qiyuan Zhang

I am currently a third‑year Ph.D. student advised by Prof. Kede Ma and Prof. Chen Ma. Previously, I completed my B.Sc. and M.Sc. in Computer Science at the University of Electronic Science and Technology of China and spent time at Singapore Management University working with Jing Jiang. Soon, I will join Prof. Xue Liu’s group at MBZUAI as a visiting student.

My research interests lie in auto‑evaluation, reward modeling, preference modeling, and improved scaling strategies such as test‑time scaling for large language models. I am always excited about new collaborations—if you share these interests or see potential synergies, feel free to reach out via email!

Now, I am interning with Hunyuan-X team@Tencent, where I am focusing my efforts on advancing generative reward modeling. I am also seeking visiting or research‑intern opportunities to further explore frontier research topics.

In addition, I regularly post self-reflections on Medium—feel free to take a look if you’re interested!

Current Research Areas

• LLM‑as‑a‑Judge / Generative Reward Models
• Methods for Test‑Time Scaling
• LLM Performance Prediction
• Automatic Benchmark Construction