Yiming Shi (史一鸣)

**Pursuing a PhD journey in multimodal research! If you're intrigued and considering collaboration or have insights, reach out via email. Let's innovate together!**

Hi, I’m Yiming Shi, an undergraduate at the University of Electronic Science and Technology of China (UESTC) enrolled in the Internet+ Dual Degree Program. I’m pursuing a BSc in Computer Science and a BEc in Finance.

Current Position

I’m currently an intern at TSAIL at Tsinghua University, co-advised by Prof. Jun Zhu and Dr. Zehua Chen. Currently I’m working with Dr. Duo Su in the area of dataset distillation, with a focus on Efficient AI.

News

  • [24/10/18] Our paper “LoLDU: Low-Rank Adaptation via Lower-Diag-Upper Decomposition for Parameter-Efficient Fine-Tuning” has been published on arXiv. You can read the full paper here. Our code can be accessed at here.

  • [24/08/11] Our paper “DiffLoRA: Generating Personalized Low-Rank Adaptation Weights with Diffusion” has been published on arXiv. You can read the full paper here.

News Archived

Previous Experience

Research Interests

My research primarily focuses on the following areas:

  • Parameter-Efficient Fine-Tuning (PEFT)
  • Diffusion Models
  • Dataset Distillation
  • Schrodinger Bridge
  • Multimodal

I’m deeply passionate about deep learning and its applications in these fields.

For a curated list of the latest research papers related to my interests, please visit my ArXiv daily digest. This resource leverages the ArXiv API to aggregate and present relevant topics in an easily digestible format.

Email / GitHub / Wechat / Blog