Welcome!

I am currently a graduate student at the School of Data Science, Fudan University, and a member of the Knowledge Works Research Lab.

My research interests primarily focus on large language models (LLM), including but not limited to reinforcement learning, data selection, post-training and agent-based systems. I have contributed to several research papers in these areas. I hope to continue exploring research with curiosity, dedication, and an open mind.

Besides research, I live with eight cats at home. Two of them are my own cats: Pidan, a Siamese cat, and Christmas, a Chinese Li Hua cat. I will also share occasional updates and news about my cats here.

🔥 News

• 2026.07:   Our paper entitled “PIQUE: Planning and Invoking Questions via Unified Exploration” was accepted to ACM MM 2026.
• 2026.04:   Our paper entitled “What Makes an Ideal Quote? Recommending “Unexpected yet Rational” Quotations via Novelty” was accepted to ACL 2026 Oral.
• 2026.04:   Our paper entitled “Instructions are All You Need: Self-supervised Reinforcement Learning for Instruction Following” was accepted to ACL 2026.
• 2026.04:   Our paper entitled “ChemAmp: Amplified Chemistry Tools via Composable Agents” was accepted to ACL 2026 Findings.
• 2026.02:   Our paper entitled “GeoRK2: Geometry-Guided Runge–Kutta Integration for Diffusion Transformer Acceleration” was accepted to CVPR 2026 Highlight.
• 2026.01:   Our paper entitled “SPICE: Submodular Penalized Information-Conflict Selection for Efficient Large Language Model Training” was accepted to ICLR 2026.
• 2025.08:  🎉🎉 We welcomed Pidan, the most mischievous Siamese cat, into our home.
• 2024.12:  🎉🎉 We adopted Christmas, the most handsome Chinese Li Hua cat.
• 2024.09:  I started my graduate studies at Fudan University.
• 2020.09:  I started my undergraduate studies at Xiamen University.
• 2001.09:  I started my journey in life.

📝 Publications

• What Makes an Ideal Quote? Recommending “Unexpected yet Rational” Quotations via Novelty
  Powei Chang, Jin Xiao, Guanglei Yue, Qianyu He, Yanghua Xiao, Deqing Yang, Jiaqing Liang
  ACL 2026 Oral

• Instructions are all you need: Self-supervised Reinforcement Learning for Instruction Following
  Qingyu Ren^*, Qianyu He^*, Powei Chang, Jie Zeng, Zeye Sun, Fei Yu, Jiaqing Liang, Yanghua Xiao
  ACL 2026

• ChemAmp: Amplified Chemistry Tools via Composable Agents
  Zhucong Li^*, Powei Chang^*, Jin Xiao, Zhijian Zhou, Qianyu He, Jiaqing Liang, Fenglei Cao, Xu Yinghui, Yuan Qi
  ACL 2026 Findings

• GeoRK2: Geometry-Guided Runge-Kutta Integration for Diffusion Transformer Acceleration
  Chaoqun Sun^*, Zongjing Fu^*, Powei Chang, Jinpeng Zhang, Jianxiang Xiang, Yukang Gao, Chenyu Wang
  CVPR 2026 Highlight

• SPICE: Submodular Penalized Information-Conflict Selection for Efficient Large Language Model Training
  Powei Chang^*, Jinpeng Zhang^*, Bowen Chen, Chenyu Wang, Chenlu Guo, Yixing Zhang, Yukang Gao, JianXiang Xiang, Yue Gao, Chaoqun Sun, Yiyi Chen, Dongying Kong
  ICLR 2026

• SALT: When More Rollouts Don’t Help in Group-Based Policy Optimization and How to Make Them Matter
  Powei Chang^*, Jinpeng Zhang^*, Chaoqun Sun, MiniWell Tsao, Lianrui Li, Jianxiang Xiang, Chenyu Wang, Yukang Gao, Dongying Kong
  arXiv preprint

• QUILL: Quotation Generation Enhancement of Large Language Models
  Jin Xiao, Powei Chang, Qianyu He, Jiaqing Liang, Feng Wei, Jinglei Chen, Zujie Liang, Deqing Yang, Yanghua Xiao
  arXiv preprint

📖 Educations

• 2024.09 - 2027.06, Master of Science in Statistics, School of Data Science, Fudan University, Shanghai, China
  • Supervised by Prof. Jiaqing Liang
• 2020.09 - 2024.06, Bachelor of Economics in Public Finance and Taxation, School of Economics, Xiamen University, Xiamen, China
  • Xiamen University is, in my heart, the most beautiful university :)

If you need anything, please contact my cat butlers: @Pidan and @Christmas.