Hi! I am a PhD student at Gaoling School of Artificial Intelligence, Renmin University of China, advised by Prof. Yankai Lin. Previously, I obtained my Master's degree from Peking University as a member of LANCO, advised by Prof. Xu Sun.

My research focuses on scalable oversight for Large Language Models, including alignment, safety, and self-evolution.

News

• 2026/07 Our paper CURE received 🏆 Outstanding Paper Award at ACL 2026!
• 2026/05 Received ICML 2026 Gold Reviewer Award.
• 2026/04 One first-author paper (CURE) accepted at ACL 2026.
• 2026/03 Excited to join DeepSeek-AI as an intern!
• 2026/01 One first-author paper (LaSeR) accepted at ICLR 2026.
• 2025/09 One first-author paper (TOPS) accepted at NeurIPS 2025.
• 2025/07 Started research internship at Tencent Hunyuan LLM Department!
• 2025/01 One first-author paper (Weak-to-Strong Deception) accepted at ICLR 2025.
• 2024/11 Two first-author papers accepted at COLING 2025.
• 2024/09 My first PhD first-author paper (BadAgents) accepted at NeurIPS 2024!
• 2024/09 Started research internship at Microsoft Research Asia!
• 2023/09 Started my PhD journey at Renmin University of China!
• 2023/06 Graduated from Peking University!

Education

• PhD Student, Gaoling School of AI, Renmin University of China · Sept. 2023 – Now
• MSc in Data Science (Statistics), Peking University · Sept. 2020 – July 2023
• Visiting Student, UC Berkeley · Jan. 2019 – May 2019
• BE in Mathematics (Honors), Xi'an Jiaotong University · Sept. 2016 – July 2020

Experience

• DeepSeek-AI, Intern · Mar. 2026 – Now Mentor: Yu Wu
• Hunyuan Team, Tencent, Research Intern · July 2025 – Mar. 2026 Mentor: Weijie Liu & Ruobing Xie
• Microsoft Research Asia, Research Intern · Sept. 2024 – May 2025 Mentor: Shuming Ma & Furu Wei
• WeChat AI, Tencent, Research Intern · Jan. 2021 – Aug. 2024 Mentor: Yankai Lin & Peng Li

Preprints

• Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolation
  Wenkai Yang, Weijie Liu, Ruobing Xie, Kai Yang, Saiyong Yang, Yankai Lin
  Paper Code
• DeepCritic: Deliberate Critique with Large Language Models
  Wenkai Yang#, Jingwen Chen#, Yankai Lin, Ji-Rong Wen
  Paper Code

Selected Publications

(# = Equal Contribution) · → Full Publication List

LLM Alignment & RL

• CURE: Critique-Driven Unified Reinforcement Learning for Test-Time Self-Improvement
  Guirong Chen#, Shuqi Ye#, Wenkai Yang# (Project Lead), Shiqi Shen, Guangyao Shen, Yankai Lin
  ACL 2026 🏆 Outstanding Paper Award
  Paper Code
• LaSeR: Reinforcement Learning with Last-Token Self-Rewarding
  Wenkai Yang, Weijie Liu, Ruobing Xie, Yiju Guo, Lulu Wu, Saiyong Yang, Yankai Lin
  ICLR 2026
  Paper Code
• Towards Thinking-Optimal Scaling of Test-Time Compute for LLM Reasoning
  Wenkai Yang, Shuming Ma, Yankai Lin, Furu Wei
  NeurIPS 2025
  Paper Code
• Super(ficial)-alignment: Strong Models May Deceive Weak Models in Weak-to-Strong Generalization
  Wenkai Yang, Shiqi Shen, Guangyao Shen, Wei Yao, Yong Liu, Zhi Gong, Yankai Lin, Ji-Rong Wen
  ICLR 2025
  Paper Code
• Distilling Rule-based Knowledge into Large Language Models
  Wenkai Yang, Yankai Lin, Jie Zhou, Ji-Rong Wen
  COLING 2025
  Paper Code

LLM Safety

• Exploring Backdoor Vulnerabilities of Chat Models
  Wenkai Yang#, Yunzhuo Hao#, Yankai Lin
  COLING 2025
  Paper Code
• Watch Out for Your Agents! Investigating Backdoor Threats to LLM-Based Agents
  Wenkai Yang#, Xiaohan Bi#, Yankai Lin, Sishuo Chen, Jie Zhou, Xu Sun
  NeurIPS 2024
  Paper Code
• Towards Codable Text Watermarking for Large Language Models
  Lean Wang#, Wenkai Yang#, Deli Chen#, Hao Zhou, Yankai Lin, Fandong Meng, Jie Zhou, Xu Sun
  ICLR 2024
  Paper Code
• RAP: Robustness-Aware Perturbations for Defending against Backdoor Attacks on NLP Models
  Wenkai Yang, Yankai Lin, Peng Li, Jie Zhou, Xu Sun
  EMNLP 2021
  Paper Code
• Rethinking Stealthiness of Backdoor Attack against NLP Models
  Wenkai Yang, Yankai Lin, Peng Li, Jie Zhou, Xu Sun
  ACL 2021
  Paper Code
• Be Careful about Poisoned Word Embeddings: Exploring the Vulnerability of Embedding Layers in NLP Models
  Wenkai Yang, Lei Li, Zhiyuan Zhang, Xuancheng Ren, Xu Sun, Bin He
  NAACL 2021
  Paper Code

Federated Learning

• Decentralized Decoupled Training for Federated Long-Tailed Learning
  Wenkai Yang, Deli Chen, Hao Zhou, Fandong Meng, Jie Zhou, Xu Sun
  TMLR
  Paper Code
• When to Trust Aggregated Gradients: Addressing Negative Client Sampling in Federated Learning
  Wenkai Yang, Yankai Lin, Guangxiang Zhao, Peng Li, Jie Zhou, Xu Sun
  TMLR
  Paper Code

Awards

• ACL Outstanding Paper Award, 2026
• ICML Gold Reviewer Award, 2026
• PhD Special Program of Young Talent Cultivation Project, CAST, 2025
• Excellent Graduate of Beijing, 2023
• Excellent Graduate of Peking University, 2023
• National Scholarship of China (Highest graduate scholarship), 2021
• Pacemaker to Merit Student (Highest graduate honor), 2021

↑ Back to top