skddj Goto Github PK

followers: 15.0 following: 93.0 repos: 60.0 gists: 1.0

Name: Yiming Shi

Type: User

Company: @University of Electronic Science and Technology of China

Bio: undergrad@UESTC -> Pursuing a Ph.D. in Multimodal

Twitter: uestcshiym

Location: Chengdu, SiChuan

Blog: https://academic.shiym.top

Hi there

Welcome to my page, where I aim to connect technology with business value and share my insights and experiences. Wishing you a fantastic day! 😊

🤺 About Me

Hi, I'm Yiming Shi 👋. I'm an undergraduate in the Internet+ Dual Degree Program at the University of Electronic Science and Technology of China (UESTC), majoring in Finance and Computer Science and Technology.

✊ Technologies

🔭 Current Position

TSAIL, Tsinghua University 📌 2024-7-1 — Present
- Position: Research Intern
- Co-advised by Prof. Jun Zhu and Dr. Zehua Chen
- Focus: parameter-efficient fine-tuning(PEFT), Diffusion, Weight2Weight

🚀 Previous Experience

Center for Future Media, UESTC 📌 2023-5-31 — Present
- Position: Research Intern
- Co-advised by Prof. Yang Yang and Assoc. Prof. Jiwei Wei
- Focus: Fine-tuning large multimodal models and investigating cross-modal interactions

Star Studio, UESTC 📌 2022-11-23 — Present
- Position: DevOps Team Member
- Responsibilities: Operations and maintenance for the in-house forum "清水河畔" at UESTC

🔬 Research Interests

My research focuses on:

Parameter-efficient fine-tuning (PEFT)
Diffusion
Multimodal

I'm deeply passionate about deep learning and artificial intelligence.

- 📫 How to reach me? Feel free to send me an email [email protected]

- 📱 My WeChat QRCode (scan to connect):

We're working towards making the world a better place by constructing elegant hierarchies for maximum code reuse and extensibility.

🏢 Education

University of Electronic Science and Technology of China 📌 2021-08-20 — Present
- Position: Undergraduate
- Major: Dual Degree in Finance & Computer Science and Technology

📃 Recent Blog

金融学基础 - 2022-12-27
数据挖掘与大数据分析 - 2022-12-27
概率论与数理统计 - 2022-12-24
计算机操作系统 - 2023-07-01
金融衍生工具 - 2023-07-01

📈 GitHub Statistics

Yiming Shi's Projects

adalora

AdaLoRA: Adaptive Budget Allocation for Parameter-Efficient Fine-Tuning (ICLR 2023).

awesome-langchain

😎 Awesome list of tools and projects with the awesome LangChain framework

b-lora

This is B-LoRA repo.

bert_score

BERT score for text generation

boardgame_match

chatgpt-next-web

One-Click to deploy well-designed ChatGPT web UI on Vercel. 一键拥有你自己的 ChatGPT 网页服务。

chinese-langchain

中文langchain项目|小必应，Q.Talk，强聊，QiangTalk

clip-interrogator

Process batch of images and save the prompts to captions directory. Image to prompt with BLIP and CLIP

consistencydecoder

Consistency Distilled Diff VAE

custom-diffusion

Custom Diffusion: Multi-Concept Customization of Text-to-Image Diffusion (CVPR 2023)

cv-arxiv-daily

🎓Automatically Update CV Papers Daily using Github Actions (Update Every 12th hours)

cvpapers

ArXiv Paper Collector for cs.CV

cvpr2023-twins

Official code for "TWINS: A Fine-Tuning Framework for Improved Transferability of Adversarial Robustness and Generalization", CVPR 2023

deepsparkhub

DeepSparkHub selects hundreds of application algorithms and models, covering various fields of AI and general-purpose computing, to support the mainstream intelligent computing scenarios.

difffit-pytorch

Implementation of "DiffFit: Unlocking Transferability of Large Diffusion Models via Simple Parameter-Efficient Fine-Tuning"

difflook_datasets

Creating datasets for DiffLook.

diffusers

更适合成电宝宝体质的 custom diffusion

everyone-can-use-english

人人都能用英语

feishu-openai

🎒 飞书 ×（GPT-4 + DALL·E + Whisper）= 飞一般的工作体验 🚀 语音对话、角色扮演、多话题讨论、图片创作、表格分析、文档导出 🚀

findfun

findfun is a web app to help ss to find fun.

ghm-greater-bay-ai-challenge-final-round

Our Final Round Repository for the Greater Bay Area AI Algorithm Competition. This hub encapsulates featuring algorithms like DDIM for speed optimization and image2image processing techniques.

ghm-greater-bay-ai-challenge-preliminary-round

Our Preliminary Round repository for the Greater Bay Area AI Algorithm Competition. This hub shows featuring Textual Inversion in CLIP Text Encoder, using of LoRA for UViT layers etc.

gpuhold

Python script performs matrix multiplication using GPUs. It utilizes the PyTorch library for GPU acceleration. The script takes command-line arguments to specify the number of GPUs to use, the size of the matrices, and the sleep interval between each multiplication operation.