Hi, my name is

Mufan Qiu.

I am an incoming PhD student at UNITES @ UNC, advised by Prof. Tianlong Chen.

I have a strong interest in generative models, and I’ve participated in research related to diffusion models and neural radiance fields. I am exploring their applications and potential improvements.

About Me

I’m a senior student majoring in Computational Mathematics and minoring in Computer Science at the University of Science and Technology of China. Currently, I’m preparing my admission for PhD programs abroad. I have a deep interest in algorithms, programming, and theoretical research, and have devoted a substantial amount of time to these fields. I hold a high level of interest in the theory and application of optimization as well as image generation and I aim at making it my future research direction.

I am currently doing an onstie internship at the science group in MSRA Beijing, and I am researching the possible applications of optimization and diffusion models in this field.

Here are a few technologies I've been working with recently:
  • C/C++
  • Python
  • Mathematica
  • Matlab
  • PyTorch
  • Git
  • Linux


Science Team Intern - Microsoft Research Asia (Beijing)
Oct 2023 - Jun 2024 (expected)
  • Alleviating the problem of inconsistent distributions between the initial state and the end state during the generation process by improving the diffusion process
  • Attempting to enhance the model’s generalization ability during the generation stage through positional encoding and sliding window techniques
Summer Research Intern (remote) - EIC Lab of GaTech
Apr 2023 - Oct 2023
  • In paper Factorized NeRF Bench, we constructed a unified framework for benchmarking NeRF models, which has the following features:
    • Accurately reproduced the testing results of previous NeRF papers
    • Implemented several NeRF variants as combinations of fundamental fields within the framework
    • Observed the performance of different fundamental fields in capturing high-frequency and low-frequency details, achieving improved rendering quality through carefully selected combinations
Generative AI Researcher - USTC
Dec 2022 - present
  • Analyzed the sources of content diversity generated by diffusion models and improved the generation quality by addressing the issue of dataset memorization in diffusion models
  • Employed distillation techniques and probabilistic principles to alleviate dataset memorization in diffusion models
Research in Computational Economics - USTC
Feb 2022 - Jun 2022
  • Quantitatively studying the relationship between cognitive hierarchies, cognitive capacity, and learning ability
  • Utilizing transfer matrices to depict the process of cognitive hierarchy evolution, providing a convenient means for quantitative analysis
  • Applying novel quantitative analysis techniques such as eigenvalue analysis to examine the evolution of cognitive hierarchies in the Beauty Contest game and the corresponding transition matrices


2020 - present
Undergraduate Student Majoring in Computational Mathematics
University of Science and Technology of China
GPA: 3.77 out of 4.3, rank 24 out of 163

Selected Awards

  • Outstanding College Student Award of China Computer Federation (Aug 2023)
  • Gold Medal, Scholarship for Outstanding Students of USTC (Oct 2023)
  • First Prize, 2021 Anhui Collegiate Programming Contest (Sept 2021)
  • Silver Medal, The 2022 ICPC Asia Nanjing Regional Contest (Dec 2022)
  • Silver Medal, The 2021 ICPC Asia Shanghai Regional Contest (Nov 2021)


  • Network Systems Experiment | Teaching Assistant (Mar 2023 - Jun 2023)
    • Similiar to the Stanford CS144 series experiments, assisted students to implement a fully functional TCP protocol based on the experiment framework
    • Responsible for answering questions, refining the experiment framework and testing platform
  • Microsoft Student Club | President (Sept 2022 - Jun 2023)
    • Assisted in organizing events such as the USTC-Microsoft Joint Doctoral Program Presentation, Innovation Practice Project Closing Ceremony, and Ada Workshop
  • Hackergame | Referee (Oct 2022)
    • Assisted in organizing the competition (with 4023 registered participants)
    • Main responsibilities included handling duplicate submissions, detecting cheating behaviors, and adjudicating other violations
  • Linux User Group | Lead of Technology Department (Sept 2021 - Jul 2022)
    • Routine maintenance of USTC open source software mirror
  • Computer Programming Club | Question Setter for School Programming Contest (Mar 2022)
    • Assisted in organizing a contest with approximately 100 students from the university
    • Designed some of the contest questions


Highly Optimized Arm Backend Mini C Compiler
C++ Clang
Highly Optimized Arm Backend Mini C Compiler
Developed an arm backend compiler optimized for a mini C syntax subset, surpassing clang O2 optimization in tests, and contributed to syntax tree parsing and IR generation, along with implementing advanced optimizations like sparse conditional constant propagation and aggressive dead code elimination.
Ray Tracing Framework Implemented with CUDA
Ray Tracing Framework Implemented with CUDA
Completed a basic ray tracing framework with CUDA acceleration, and enhanced it with additional textures, materials, objects, and scene configuration file support for final video rendering.

Get in Touch

My inbox is always open. Whether you have a question or just want to say hi, I’ll try my best to get back to you!