Tianyi Wang

Tianyi Wang

Undergraduate student in Computer Science and Technology at BUPT, focused on post-training RL and optimizer implementation.

Tianyi Wang

Hi!👋🏻 I am Tianyi Wang, an undergraduate student majored in Computer Science at Beijing University of Posts and Telecommunications. My current focus is post-training algorithms, especially reinforcement learning. I am also highly interested in optimizer design, implementation, and reproduction.

If you want a quick overview of my work, start with these pages:

News

  • 2026.04 🎉 SPPO was accepted to ACL 2026
  • 2026.04 🎉 APO was accepted to ICML 2026
  • 2026.05 🎉 SPPO was selected as Oral presentation in ACL

Selected Works

SPPO paper

SPPO: Sequence-Level PPO for Long-Horizon Reasoning

ACL 2026 Oral · Co-first Author
Tianyi Wang*, Yixia Li*, Long Li, Yibiao Chen, Shaohan Huang, Yun Chen, Peng Li, Yang Liu, Guanhua Chen
Reformulates long-horizon reasoning as a sequence-level contextual bandit with a decoupled scalar value function to improve stability and reduce memory cost.
arXiv GitHub
APO paper

Anchored Policy Optimization: Mitigating Exploration Collapse

ICML 2026 · Co-first Author
Tianyi Wang*, Long Li*, Hongcan Guo, Yibiao Chen, Yixia Li, Yong Wang, Yun Chen, Guanhua Chen
Introduces support-constrained rectification for RLVR training to mitigate exploration collapse and improve both Pass@1 and response diversity.
arXiv GitHub
DyJR paper

DyJR: Preserving Diversity in Reinforcement Learning with Verifiable Rewards via Dynamic Jensen-Shannon Replay

Under Review
Long Li, Zhijian Zhou, Tianyi Wang, Weidi Xu, Zuming Huang, Wei Chu, Zhe Wang, Shirui Pan, Chao Qu, Yuan Qi
Introduces Dynamic Jensen-Shannon Replay to preserve diversity in reinforcement learning through a verifiable rewards mechanism.
arXiv

Education

Beijing University of Posts and Telecommunications

Bachelor Degree

Beijing University of Posts and Telecommunications
B.S. in Computer Science and Technology
2023.09 -- expected 2027.07

  • GPA 3.76/4.00, ranked 20/360 in the major
  • CET-6 503, CET-4 571
  • Focused on post-training algorithms, reinforcement learning, and optimizer engineering

Experience

Youtu internship

Internship

Youtu
Research Intern
Expected start: 2026.07.01
End date: TBD

  • Internship focused on machine learning and reinforcement learning related research
  • Details will be updated once the internship officially starts