ℹ️ Short Bio
Hi, I’m Yuanzhe, a master student at University of Califronia, San Diego. I have the great honor of being collaborating with Prof. Yaoqing Yang, Prof. Julian McAuley, Prof. Zhiting Hu and Dr. Ren Pu.
My current research is focused on
- Understanding the mechanisms, dynamics and generalization of LLM and SciML models via numerical analysis. [2, 3, 4].
- Memory system in LLM & Agent [1, 5].
📖 Educations
|
University of California, San Diego (UCSD) M.S. in Computer Science and Engineering |
2024.09 - 2026.03 (Expected) |
|
Huazhong University of Science and Technology (HUST) BEng. in Artificial Intelligence, Innovation Experimental Honor Class, Qiming School GPA: 3.91/4.0 |
2020.09 - 2024.06 |
🔥 News
- 2025.07: 😁 We open-sourced the MemoryAgentBench. Thanks for the great help from Yu Wang!
- 2025.05: 🎉🎉 Two papers are accepted by ICML 2025 as Poster! See you at Vancouver.
- 2024.09: 🎉🎉 Excited to share that our work “Model Balancing Helps Low-data Training and Fine-tuning” is accepted by EMNLP 2024 as Oral Presentation!
- 2024.06: 😁 I graduated from HUST!
- 2024.06: 😄 I created my account on OpenReview!
📝 Writing Samples
(# denotes equal contribution)
Preprints & On-Working Prj
[1] Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions
{Yuanzhe Hu#, Yu Wang#}, Julian McAuley
HF Dataset | Code | Paper | 
Submitted to NeurIPs 2025 (D&B) / ICML 2025 LCFM Workshop
[2] Loss Landscape Analysis of Scientific Machine Learning Models
On-Working, aimming at ICLR 2026.
Published
[3] Eigenspectrum Analysis of Neural Networks without Aspect Ratio Bias
Yuanzhe Hu, Kinshuk Goel, Vlad Killiakov, Yaoqing Yang
Code | Paper | ICML Website | Review | 
ICML 2025
Last Update: 07/2025