Xin's Homepage

What lies behind us and what lies before us are tiny matters compared to what lies within us.

Ralph Waldo Emerson

I am actively seeking research internships or funded PhD opportunities that align with my passion and expertise, and a healthy environment where I can work alongside like-minded individuals who value integrity, knowledge, and creativity, and embrace modern civilization.

My research interests include deep generative models, multimodal large language models, reinforcement learning, time series analysis, and computer vision.

news

Jan 25, 2026	I began my PhD program with a scholarship at the UniTrento, on November 1, 2024. Then, I left the program around the end of December 2024, with approval received on January 13, 2025, and appreciated the understanding from the university staff.
Dec 01, 2023	I officially left my PhD program at AIRC, UU, and appreciated the understanding from the staff at UU.

latest posts

Oct 29, 2025	Reconstructing the Scaffolding A Learner’s Notes on Ordinary Differential Equations
Oct 29, 2025	Reconstructing the Scaffolding A Learner’s Notes on Ordinary Differential Equations
Oct 29, 2025	Reconstructing the Scaffolding A Learner’s Notes on Ordinary Differential Equations

selected publications

One Framework to Rule Them All: Unifying RL-Based and RL-Free Methods in RLHF

Xin Cai

2025

arXiv Bib

@article{cai2025frameworkruleallunifying,
  title = {One Framework to Rule Them All: Unifying RL-Based and RL-Free Methods in RLHF},
  author = {Cai, Xin},
  year = {2025},
  eprint = {2503.19523},
  archiveprefix = {arXiv},
  primaryclass = {cs.LG},
  url = {https://arxiv.org/abs/2503.19523},
}