Ruihan Yang (杨睿涵)

I'm a Ph.D. student in Statistics at Fudan University, advised by Prof. Deqing Yang. Previously, I received my Bachelor's degree in Mathematics and Applied Mathematics from Fudan University in 2021. I am a recipient of China National Scholarship for Graduate Students (研究生国家奖学金, Top 1%).

Currently, I am a Research Intern at Moonshot AI RL Team, working on agentic intelligence for Kimi K2.5.

👀 I am on the job market! Seeking research roles in LLM Agents and Reinforcement Learning. Open to full-time positions and research collaborations. Let's connect!

📧 Email  ·  📄 CV  ·  🎓 Google Scholar  ·  💻 Github

profile photo

Research: LLM Agents and Reinforcement Learning

I am dedicated to advancing LLM agents through reinforcement learning and uncertainty quantification:

News

  • 2025-10: Started at Moonshot AI RL Team! Contributing to Tool-Integrated Reasoning for Kimi K2.5 (first model to reach HLE 50+)!
  • 2025-09: Two papers accepted to NeurIPS 2025! ARIA received Spotlight!
  • 2025-08: UNCLE accepted to EMNLP 2025!
  • 2025-05: LoGU accepted to ACL 2025!
  • 2025-01: SelfGoal accepted to NAACL 2025!
  • 2022-10: Awarded National Scholarship (Top 1%)!

Selected Publications


A full list of publications is here. (* indicates equal contribution.)
Kimi K2.5: Visual Agentic Intelligence
Ruihan Yang (contributor), contributed to Tool-Integrated Reasoning
Technical Report, 2025.   (First HLE 50+)
Homepage
Think Fast and Slow: Step-Level Cognitive Depth Adaptation for LLM Agents
Ruihan Yang, Fanghua Ye, Xiang Wei, Xinbo Xu, Xiaqiang Tang, Bo Zhao, Shanyi Wang, Zhaopeng Tu, Xiaolong Li, Deqing Yang
Preprint, 2025.
Paper
ARIA: Training Language Agents with Intention-Driven Reward Aggregation
Ruihan Yang*, Yikai Zhang*, Aili Chen, Siyu Yuan, Jiangjie Chen, Deqing Yang, Yanghua Xiao
NeurIPS, 2025.   (Spotlight)
Homepage / Code
The Lighthouse of Language: Enhancing LLM Agents via Critique-Guided Improvement
Ruihan Yang, Fanghua Ye, Jian Li, Siyu Yuan, Yikai Zhang, Zhaopeng Tu, Xiaolong Li, Deqing Yang
NeurIPS, 2025.
Homepage
UNCLE: Benchmark Uncertainty in Long-form Generation
Ruihan Yang*, Caiqi Zhang*, Zhisong Zhang, Xinting Huang, Nigel Collier, Dong Yu, Deqing Yang
EMNLP, 2025.
Homepage / Dataset
LoGU: Long-form Generation with Uncertainty Expressions
Ruihan Yang*, Caiqi Zhang*, Zhisong Zhang, Xinting Huang, Sen Yang, Nigel Collier, Dong Yu, Deqing Yang
ACL, 2025.
Homepage
SelfGoal: Your Language Agents Already Know How to Achieve High-level Goals
Ruihan Yang, Jiangjie Chen, Yikai Zhang, Siyu Yuan, Aili Chen, Kyle Richardson, Yanghua Xiao, Deqing Yang
NAACL, 2025.
Homepage / Code

Experience
  • Moonshot AI RL Team - Research Intern (Oct. 2025 - Present)
    Supervisor: Longhui Yu · Topic: Agentic Intelligence (Kimi K2.5 post-train)
  • Tencent Hunyuan AI Digital Human - Research Intern (Oct. 2024 - Oct. 2025)
    Supervisors: Fanghua Ye, Jian Li · Topic: Agent Learning
  • Tencent AI Lab - Research Intern (Mar. 2024 - Oct. 2024)
    Supervisors: Leyang Cui, Zhisong Zhang · Topic: Long-Form Uncertainty Expression

Education
  • Ph.D. in Statistics (2023 - 2027 expected)
    Fudan University · Supervisor: Deqing Yang
  • Ph.D. in Applied Mathematics (2021 - 2023)
    Fudan University
  • B.S. in Mathematics and Applied Mathematics (2017 - 2021)
    Fudan University

Awards & Service
  • National Scholarship for Graduate Students (Top 1%), 2022
  • Fudan University Excellent Student Award, 2021-2025
  • Reviewer: ACL 2024, 2025 · EMNLP 2024, 2025