avatar

Yu Yang

Ph.D. Student
Duke University
yu.yang (at) duke.edu


About Me

I am a fourth-year Ph.D. student at Duke University advised by Prof. Pan Xu. My primary research interests focus on reinforcement learning (RL), especially reinforcement learning with verifiable rewards (RLVR) for improving the reasoning capabilities of large language models. I also work on off-dynamics RL and RL applications in healthcare. More broadly, I am interested in incorporating RL techniques into foundation models to improve their reasoning, adaptation, and alignment with human preferences.

News

Older news

Publications

  1. Localized Dynamics-Aware Domain Adaptation for Off-Dynamics Offline Reinforcement Learning.
    Zhangjie Xia*, Yu Yang*, Pan Xu. (* Equal contribution)
    In Proc. of the 40th Annual Conference on Neural Information Processing Systems (NeurIPS), Sydney, Australia, 2026. [Spotlight, 292/30709 (0.95%)]
    Paper

  2. MOBODY: Model-Based Off-Dynamics Offline Reinforcement Learning.
    Yihong Guo, Yu Yang, Pan Xu, Anqi Liu.
    In Proc. of the 14th International Conference on Learning Representations (ICLR), Rio de Janeiro, Brazil, 2026.
    Paper Code

  3. Return Augmented Decision Transformer for Off-Dynamics Reinforcement Learning.
    Ruhan Wang*, Yu Yang*, Zhishuai Liu, Dongruo Zhou, Pan Xu. (* Equal contribution)
    Transactions on Machine Learning Research (TMLR), 2026.
    Paper

  4. rECMOmender: Reinforcement Learning for Decision Support in Venovenous Extracorporeal Membrane Oxygenation Management.
    Jiafeng Song, Sagar B Dave, Yu Yang, Henry Foote, Ronald Moore, Pulakesh Upadhyaya, Lisa Lima, Christina Creel, Pan Xu, Craig S. Jabaley, Rishikesan Kamaleswaran.
    Critical Care Explorations, 2026.
    Paper

  5. Pre-trained Language Models Improve the Few-shot Prompt Ability of Decision Transformer.
    Yu Yang, Pan Xu.
    Transactions on Machine Learning Research (TMLR), 2025.
    Paper Code

  6. An Interactive Framework for Generating Clinical Data with Human Feedback.
    Yu Yang*, Jiafeng Song*, Zhishuai Liu, Henry P Foote, Rishikesan Kamaleswaran, Pan Xu (* Equal contribution)
    In Proc. of IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI), Atlanta, Georgia, USA, 2025.
    Paper

  7. D2Fed: Federated Semi-Supervised Learning With Dual-Role Additive Local Training and Dual-Perspective Global Aggregation.
    Jingxin Mao, Yu Yang, Zhiwei Wei, Yanlong Bi, Rongqing Zhang.
    IEEE Transactions on Neural Networks and Learning Systems (TNNLS), 2025.
    Paper

  8. Optimal batched best arm identification.
    Tianyuan Jin, Yu Yang, Jing Tang, Xiaokui Xiao, Pan Xu.
    In Proc. of the 38th Annual Conference on Neural Information Processing Systems (NeurIPS), Vancouver, Canada, 2024.
    Paper Code

  9. More Efficient Randomized Exploration for Reinforcement Learning via Approximate Sampling.
    Haque Ishfaq, Yixin Tan, Yu Yang, Qingfeng Lan, Jianfeng Lu, A Rupam Mahmood, Doina Precup, Pan Xu.
    Presented at the 1st Reinforcement Learning Conference (RLC), Amherst, Massachusetts, USA, 2024.
    Paper Code

  10. On the stability of expressive positional encodings for graphs.
    Yinan Huang, William Lu, Joshua Robinson, Yu Yang, Muhan Zhang, Stefanie Jegelka, Pan Li.
    In Proc. of the 12th International Conference on Learning Representations (ICLR), Vienna, Austria, 2024.
    Paper Code

  11. Generative character inpainting guided by structural information.
    Haolong Li, Zizheng Zhong, Wei Guan, Chenghao Du, Yu Yang, Yuxiang Wei, Chen Ye.
    The Visual Computer, 2021.
    Paper

Preprints

  1. Inference-Time Alignment of Diffusion Models via Trust-Region Iterative Twisted Sequential Monte Carlo
    Weixin Wang*, Yu Yang*, Wei Deng, Pan Xu. (* Equal contribution)
    arXiv preprint, 2026.
    Paper

  2. Cross-Domain Energy-Guided Diffusion Generation for Off-Dynamics Reinforcement Learning
    Yu Yang, Yihong Guo, Anqi Liu, Pan Xu.
    arXiv preprint, 2026.
    Paper

  3. Diffusion Posterior Sampling for Nonlinear Contextual Bandits
    Weixin Wang*, Yu Yang*, Pan Xu. (* Equal contribution)
    Under Review.

  4. How to Provably Improve Return Conditioned Supervised Learning?
    Zhishuai Liu, Yu Yang, Ruhan Wang, Pan Xu, Dongruo Zhou.
    NeurIPS 2025 Workshop on ARLET.
    Paper

Services

Conference Reviewers


Powered by Jekyll and Minimal Light theme.