PeopleofSuperintelligence

Researcher

Chongxuan Li

Also known as 李崇轩

Chongxuan Li is a researcher in deep generative models whose work spans diffusion sampling, text-to-3D generation and language models. He earned degrees at Tsinghua, visited the University of Amsterdam and completed postdoctoral research at Tsinghua in 2021. Analytic-DPM received ICLR's 2022 Outstanding Paper Award, while DPM-Solver generated high-quality samples in 10 to 20 function evaluations. In 2023 he coauthored ProlificDreamer, and in 2025 he coauthored papers on masked diffusion models and LLaDA. As of October 2026, he is a tenure-track associate professor at Renmin University's Gaoling School of AI, where his group focuses on deep generative models.

Timeline

  1. 2010–2019

    Tsinghua degrees

    He earns a B.E. from Tsinghua's Institute for Interdisciplinary Information and Sciences in 2014 and completes a Ph.D. in computer science and technology in 2019 under Bo Zhang.

  2. September 2017–September 2018

    Guest researcher in Amsterdam

    From September 2017 to September 2018, he works as a Ph.D. guest researcher at the University of Amsterdam under Max Welling.

  3. July 2019–July 2021

    Postdoctoral researcher at Tsinghua

    From July 2019 to July 2021, he works as a postdoctoral researcher at Tsinghua under Jun Zhu.

  4. 2022

    Analytic-DPM wins an ICLR award

    In 2022, he coauthors Analytic-DPM, a training-free inference framework that estimates the optimal reverse variance in diffusion probabilistic models. The paper receives ICLR's 2022 Outstanding Paper Award.

  5. 2022

    DPM-Solver for diffusion

    In 2022, he coauthors DPM-Solver, a high-order solver for diffusion ODEs that generates high-quality samples in 10 to 20 function evaluations without further training.

  6. 2023

    ProlificDreamer for text-to-3D

    In 2023, he coauthors ProlificDreamer, which introduces variational score distillation for text-to-3D generation and improves diversity and sample quality.

  7. 2025

    Masked diffusion and LLaDA

    At ICLR 2025, he coauthors Scaling up Masked Diffusion Models on Text; the study establishes the first scaling law for masked diffusion and trains models with up to 1.1 billion parameters. At NeurIPS 2025, he coauthors Large Language Diffusion Models, which introduces LLaDA, a diffusion model trained from scratch under the pre-training and supervised fine-tuning paradigm.

  8. As of October 2026

    Associate professor at Renmin

    He is a tenure-track associate professor at Gaoling School of AI, Renmin University of China. His group focuses on deep generative models.

Career

  1. NowRenmin University of ChinaTenure-track associate professor
  2. From 2019Tsinghua UniversityPostdoctoral researcher
  3. 2017–2018University of AmsterdamGuest researcher

Education

  1. 2019Tsinghua UniversityPh.D.
  2. 2014Tsinghua UniversityB.E.

From the biography.

  1. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other universityIncludes Chongxuan Li. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other university. We compare doctorate and 5+ year employment links, career timing and current workplaces for 187 major-prize and 1,111 best-paper winners.
  2. 84% of people shaping AI with recorded undergraduate and doctoral schools changed institutions for their doctorateIncludes Chongxuan Li. 84% of people shaping AI with recorded undergraduate and doctoral schools changed institutions for their doctorate. Across 2,116 people with both an undergraduate degree and a doctorate recorded, we trace which undergraduate schools feed which doctoral programs, how often students stay at the same school, and how paths differ by birth-country region.
  3. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes Renmin University of China. We estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.
  4. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes Renmin University of China. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.

Colleagues

People with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Chongxuan Li yet. Send us a free-licensed photo.

  1. Chongxuan Li ai.ruc.edu.cn. Checked 22 October 2026.
  2. Analytic-DPM: an Analytic Estimate of the Optimal Reverse Variance in Diffusion Probabilistic Models arxiv.org. Checked 22 October 2026.
  3. Chongxuan Li ai.ruc.edu.cn. Checked 22 October 2026.
  4. DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps arxiv.org. Checked 22 October 2026.
  5. ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillation proceedings.neurips.cc. Checked 22 October 2026.
  6. Scaling up Masked Diffusion Models on Text arxiv.org. Checked 22 October 2026.
  7. Chongxuan Li's Homepage zhenxuan00.github.io. Checked 22 October 2026.
  8. Large Language Diffusion Models proceedings.neurips.cc. Checked 22 October 2026.