Zixiang Chen

Department of Computer Science,

University of California, Los Angeles.

Contact: chenzx19 at cs dot ucla dot edu

chenzx997 at gmail dot com

[Google Scholar] [Linkedin] [CV] [Twitter]

Zixiang Chen is a Ph.D. student in Computer Science at UCLA, advised by Prof. Quanquan Gu. His research interests lie in the theoretical foundations and algorithm design of deep learning and reinforcement learning, with a recent focus on generative models and their domain adaptation. He was a visiting graduate student at the Simons Institute for the Theory of Computing and was awarded the UCLA dissertation fellowship. Before joining UCLA, he received his bachelor's degree in mathematics from Tsinghua University.

Recent News

[01/2025] A work on theoretical analysis of Discrete Diffusion Models is accepted to ICLR 2025.
[01/2025] A work on studying the Power of Multitask Representation Learning is accepted to AISTATS 2025.
[01/2025] Happy New Year!
[12/2024] Flying to and attending NeurIPS 2024. Have a good time in Vancouver!
[11/2024] Excited to join and give a lightning talk at Google's Theory and Practice of Foundation Models Workshop.
[09/2024] Three works on accelerating discrete diffusion inference, Sign SGD matching SQ lower bounds on k-sparse parity problems, and self-play finetuning diffusion models are accepted to NeurIPS 2024.
[06/2024] I am deeply honored and grateful to have been awarded the UCLA Dissertation Year Fellowship.
[05/2024] Excited to share that SPIN has been accepted to ICML 2024. Check out more here.
[01/2024] Two papers about multimodal Learning and in-context learning are accepted to ICLR 2024.
[01/2024] Happy new year! Release our first paper in 2024: "Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models." Let LLM Self Play fIne tuNing (SPIN) and get stronger!
[12/2023] Flying to and attending NeurIPS. Have a good time in New Orleans!

Prior News

[09/2023] Two papers get accepted to Neurips 2023. Prepare to revisit New Orleans.
[07/2023] Have a good time in Hawaii.
[05/2023] Two papers get accepted to ICML 2023. Looking forward to the trip to Hawaii.
[01/2023] Three papers are accepted to ICLR2023.
[12/2023] First time attending the conference in person. Enjoy the Neurips very much.
[09/2022] Two papers are accepted to Neurips 2022.
[08/2022] Talk on "Toward Understanding Mixture of Experts in Deep Learning”, in Representation Learning Theory TTIC Chicago Summer Workshop.
[06/2022] Passed my oral qualify exam. Thanks a lot to the committees for their valuable suggestions and guidance!.
[05/2022] Talk on “Benign Overfitting in Two-layer Convolutional Neural Networks” in Math Machine Learning seminar MPI MiS + UCLA.