Xueru Wen 温学儒

Biography

I am a Ph.D. student at the University of Chinese Academy of Sciences (UCAS), advised by Prof. Ben He. I conduct research in the Chinese Information Processing Laboratory at the Institute of Software, Chinese Academy of Sciences (ISCAS). I expect to graduate in 2028.

I received my B.S. degree from Jilin University in 2023, where I was in the Tang Aoqing Honors Program in Science, Computer Science Track. Since September 2023, I have been interning at Xiaohongshu (RedNote), and became an Ace Top Intern in July 2025.

Research Interests

My research focuses on LLM post-training, particularly reward modeling, agents, and recursive self-improvement (RSI). I study how reliable reward signals and reinforcement learning improve agent reasoning and tool use. For RSI, I explore how agents can automate data generation, evaluation, and training to support iterative self-improvement. As models become more capable, I am interested in scalable oversight that keeps this process reliable and aligned.

Internship & Project Experience

Publications

My name is in bold. BibTeX · Google Scholar

1. Coupled Variational Reinforcement Learning for Language Model General Reasoning Paper Code
   Xueru Wen, Jie Lou, Yanjiang Liu, Hongyu Lin, Ben He, Xianpei Han, Le Sun, Yaojie Lu, Debing Zhang.
   ICML 2026.
2. Decoupling Reasoning and Confidence: Resurrecting Calibration in Reinforcement Learning from Verifiable Rewards Paper
   Zhengzhao Ma, Xueru Wen, Boxi Cao, Yaojie Lu, Hongyu Lin, Jinglin Yang, Min He, Xianpei Han, Le Sun.
   ICML 2026.
3. Learning from Failures: Correction-Oriented Policy Optimization with Verifiable Rewards Paper
   Mengjie Ren, Jie Lou, Boxi Cao, Xueru Wen, Hongyu Lin, Xianpei Han, Le Sun, Xing Yu, Yaojie Lu.
   COLM 2026.
4. Rethinking Reward Model Evaluation: Are We Barking up the Wrong Tree? Paper
   Xueru Wen, Jie Lou, Yaojie Lu, Hongyu Lin, Xing Yu, Xinyu Lu, Ben He, Xianpei Han, Debing Zhang, Le Sun.
   ICLR 2025 Spotlight
5. Cheems: A Practical Guidance for Building and Evaluating Chinese Reward Models from Scratch Paper Code
   Xueru Wen, Jie Lou, Zichao Li, Yaojie Lu, Xing Yu, Yuqiu Ji, Guohai Xu, Hongyu Lin, Ben He, Xianpei Han, Le Sun, Debing Zhang.
   ACL 2025.
6. On-Policy Self-Alignment with Fine-grained Knowledge Feedback for Hallucination Mitigation Paper
   Xueru Wen, Jie Lou, Xinyu Lu, Yuqiu Ji, Xinyan Guan, Yaojie Lu, Hongyu Lin, Ben He, Xianpei Han, Debing Zhang, Le Sun.
   ACL 2025.
7. Transferable Post-training via Inverse Value Learning Paper
   Xinyu Lu, Xueru Wen, Yaojie Lu, Bowen Yu, Hongyu Lin, Haiyang Yu, Le Sun, Xianpei Han, Yongbin Li.
   NAACL 2025.
8. AutoAlign: Get Your LLM Aligned with Minimal Annotations Paper
   Xinyu Lu, Dong Xu, Chunkang Zhang, Xinyan Guan, Junxiang Wang, Qingyu Zhang, Pengbo Wang, Yingzhi Mao, Hao Xiang, Xueru Wen, Zichao Li, Yaojie Lu, Hongyu Lin, Le Sun, Xianpei Han.
   ACL 2025.
9. Critic-CoT: Boosting the Reasoning Abilities of Large Language Model via Chain-of-Thought Critic Paper
   Xin Zheng, Jie Lou, Boxi Cao, Xueru Wen, Yuqiu Ji, Hongyu Lin, Yaojie Lu, Xianpei Han, Debing Zhang, Le Sun.
   ACL 2025.
10. The Devil Is in the Details: Tackling Unimodal Spurious Correlations for Generalizable Multimodal Reward Models Paper Code
    Zichao Li, Xueru Wen, Jie Lou, Yuqiu Ji, Yaojie Lu, Xianpei Han, Debing Zhang, Le Sun.
    ICML 2025.
11. Offline Pseudo Relevance Feedback for Efficient and Effective Single-Pass Dense Retrieval Paper
    Xueru Wen, Xiaoyang Chen, Xuanang Chen, Ben He, Le Sun.
    SIGIR 2023.
12. End-to-End Entity Detection with Proposer and Regressor Paper
    Xueru Wen, Changjiang Zhou, Haotian Tang, Luguang Liang, Hong Qi, Yu Jiang.
    Neural Processing Letters 2023.
13. Type-supervised Sequence Labeling Based on the Heterogeneous Star Graph for Named Entity Recognition Paper
    Xueru Wen, Changjiang Zhou, Haotian Tang, Luguang Liang, Yu Jiang, Hong Qi.
    arXiv, 2022.

Tools & Resources

A few small tools I’ve developed that you might find useful. Feel free to have your agent give them a try.

Awards & Honors

• 中国科学院大学院长奖奖学金优秀奖, Aug. 2026 (Chinese Academy of Sciences system)
• CAST Young Talent Support Program, Doctoral Student Special Track, 2025.

Contact

Email: wenxueru2022@iscas.ac.cn
Chinese Information Processing Laboratory
Institute of Software, Chinese Academy of Sciences, Beijing, China