News
|
Xinyuan Zhu
朱心远
Lab for Data Science
100, Fuxing Road, Hefei, China
Email:
|
I am a final-year Ph.D. student at the University of Science and Technology of China (USTC)
, co-supervised by Prof. Xiangnan He and Prof. Fuli Feng since September 2021.
My research focuses on machine learning for biology and medicine (AI for Health), with particular interests in LLM agents, automated scientific discovery, and drug discovery. Previously, I worked on recommender systems and causal inference.
I obtained B.Eng in Artificial Intelligence and Information Security in USTC.
I was also affiliated with the School of the Gifted Young at USTC, a selective undergraduate program.
Publications
| Hierarchical Progressive Learning for Zero-Shot Peptide-HLA Binding Prediction and Automated Antigenic Peptide Design
Xinyuan Zhu, Jiadong Lu, Xinting Hu, Tengchuan Jin, Shan Lu and Fuli Feng Cell Reports |
| Machine Learning Dataset and Benchmark for Accurate T Cell Receptor-pHLA Binding Prediction
Xinyuan Zhu, Jiadong Lu, Yeqing Lu, Yuyan Zhang and Fuli Feng Under reivew of Nature Computational Science |
| Generative Design of High-affinity T-cell Receptors by Progressive Learning with Structural Confidence
Xinyuan Zhu, Qingshuo Jin, Jiadong Lu, Jun Wu, Lixia Zhu, Yuhua Shang, Tengchuan Jin and Fuli Feng Under reivew of Nature Communications |
| Improving Prostate Cancer Risk Prediction through Partial AUC Optimization
Xinyuan Zhu, Xiaohan Ren, Wentao Shi, Changming Wang, Xuehan Liu, Yuqing Liu, Tao Tao, Fuli Feng The Web Conference (WWW) 2024 - Health Day |
| Benchmarking TCR-pMHC Structure Prediction: a Unified Evaluation and CDR3-based Functional Insights
Jiadong Lu, Xinyuan Zhu, Xinting Hu, Cheng Zhang and Fuli Feng Under reivew |
| DrugTrail: Interpretable Drug Discovery via Structured Reasoning and Druggability-Tailored Preference Optimization
Yurou Liu, Mingyang Li, Xinyuan Zhu, Rui Jiao, Yiming Dong, Xinyu Tang, Yang Liu, Jieping Ye, Bing Su, Zheng Wang The Fourteenth International Conference on Learning Representations (ICLR 2026) |
| Probing RLVR Training Instability through the Lens of Objective-level Hacking
Yiming Dong, Kun Fu, Haoyu Li, Xinyuan Zhu, Yurou Liu, Lijing Shao, Jieping Ye, Zheng Wang Preprint. Under review. |
| Machine Learning-enhanced Preoperative Assessment of Spontaneous Ventilation Risk in Diverse Video-assisted Thoracoscopic Surgeries
Xiaohan Ren, Fengfa Yang, Jiayu Li, Shujun Wu, Xinyuan Zhu, Fuli Feng, Xinting Hu, Lan Lan The Journal of Thoracic and Cardiovascular Surgery Open |
| Mitigating Hidden Confounding Effects for Causal Recommendation
Xinyuan Zhu, Yang Zhang, Fuli Feng, Xun Yang, Dingxian Wang, Xiangnan He IEEE Transactions on Knowledge and Data Engineering (TKDE) |
Education
| University of Science and Technology of China Ph.D. student in Artificial Intelligence Sep 2021 - , Hefei, Anhui, China Advisor: Prof. Xiangnan He and Prof. Fuli Feng. |
| University of Science and Technology of China B.Eng. in Information Security and Artificial Intelligence Sep 2017 - June 2021, Hefei, Anhui, China Advisor: Prof.Xiangnan He and Prof.Fuli Feng. |
Industry Experiences
| Research Intern, AI for Science Lab, Alibaba Cloud, Apr 2025 - now • My research during internship in Alibaba Cloud is focused on the design and implementation of LLM agent system for virtual screening and drug discovery. |
| Research Intern, BioGeometry, Apr 2023 - Jul 2023 Mentor: Dr. Meng Qu and Prof. Jian Tang • During my internship at BioGeometry, a pioneering AI Drug Discovery startup, I played a key role in designing and developing the advanced RAG system for biomedicine. This state-of-the-art system leveraged the power of LLMs and vast repositories of biomedical information, including texts, patents, molecular data, and protein information, to greatly accelerate the drug discovery process. |
| Research Intern, Theory Dept., 2012 Lab, Huawei Technologies Co., Ltd, Jan 2020 - June 2020 Mentor: Dr. Sen Wang and Dr. Peng Wang • Design privacy-preserving learning mechanism with (local) differential privacy for edge-cloud collaborative learning. • Use model inversion attack and membership inference attack to test the performance of the system. • The built prototype Mistnet has been transferred to Plato (a federated learning framework) and Sedna (AI Toolkit over KubeEdge). |
Selected Awards
|
USTC Chuying Fund Support, 2024 USTC Outstanding Student Scholarship, 2021-2024 USTC Outstanding Graduate 2021 Qiangwei Scholarship, 2021 Cyrus Tang Scholarship, 2017 - 2021 Challenge Cup, National Third Prize, 2020 Neural Network Intelligence (NNI) by MSRA, Outstanding Project 2020 Robotics Competition of USTC, Bronze Medal, 2019 |
Activities/Services
|
Reviewer ICLR '26, ICML '26, KDD '26, etc. |
|
Teaching Assistant Probability and Statistics / Advanced Computer Programming / Introduction to Cryptography. 2020.1 - 2021.6 @USTC |