Sungbin Shin

prof_pic.jpeg

Hi, I am a second-year Ph.D student in computer science and engineering at POSTECH, advised by Prof. Namhoon Lee.

My vision is to make machine learning models more accessible to everyone. As of now, I am interested in building efficient models through compression and understanding generalization of large-scale neural networks.

Contact me via ssbin4@postech.ac.kr if you have any questions.

news

Sep 2024 Our paper on LLM pruning has been accepted at EMNLP 2024!
Nov 2023 Our paper on sharpness-aware minimization won the best paper award at JKAIA 2023.
Check out the related news!
Jul 2023 I am attending ICML 2023 in person! See you in Hawaii!
Jun 2023 2 papers have been accepted at ICML workshops!
May 2023 I will be working as a student researcher at Google starting from this month!

selected publications

2024

  1. Rethinking Pruning Large Language Models: Benefits and Pitfalls of Reconstruction Error Minimization
    Sungbin Shin, Wonpyo Park, Jaeho Lee, Namhoon Lee
    EMNLP, 2024
    arXiv Code Poster Thread

2023

  1. Critical Influence of Overparameterization on Sharpness-aware Minimization
    Sungbin Shin*, Dongyeop Lee*, Maksym Andriushchenko, Namhoon Lee
    arXiv preprint, 2023; ICML Workshop on High-dimensional Learning Dynamics, 2023
    Best paper award at JKAIA 2023
    arXiv Code Thread
  2. A Closer Look at the Intervention Procedure of Concept Bottleneck Models
    Sungbin Shin, Yohan Jo, Sungsoo Ahn, Namhoon Lee
    ICML, 2023; NeurIPS Workshop on TSRML, 2022
    arXiv Code Poster Thread