Nayeong Kim
I am a Ph.D candidate in computer science and engineering department at POSTECH, advised by Suha Kwak. My long-term research goal is to work 'Toward ensuring the trustworthiness of AI models.' During my pursuit of a Ph.D., I aim to address the pressing issue of combating spurious correlations that undermine the integrity of AI models in real-world applications. I am a member of Computer Vision Lab at POSTECH. Previously, I received a B.S. in computer science and engineering department from POSTECH in 2017.
Experience
- Visiting Researcher, University of Tübingen, Host: Prof. Seongjoon Oh Mar. 2024 - May. 2024
- Research Intern, Samsung Electronics, Jan. 2018 - Feb. 2018
- Research Intern, Exem, Mar. 2017 - Sep. 2017
Publications
Exploiting Synthetic Data for Data Imbalance Problems: Baselines from a Data Perspective
ICML 2023 - Workshop on MMFM: What is Next in Multimodal Foundation Models? (MMFM)
- [
- paper
- ]
Removing Multiple Biases through the Lens of Multi-task Learning
ICML, 2023 - Workshop on Spurious Correlations, Invariance and Stability (SCIS)
- [
- paper
- ]
Awards
- Qualcomm Innovation Fellowship Winner , Qualcomm Korea Corp., “Leveraging Proxy of Training Data for Test-Time Adaptation” (2023).
- Samsung Strategy and Innovation Paper Award, “Leveraging Proxy of Training Data for Test-Time Adaptation” (2023).
- BK21 Best Paper Award from POSTECH CSE, “Learning Debiased Classifier with Biased Committee” (2023).
Talks
- Doctoral Colloquium: “Debiased Training by Discovering and Mitigating Spurious Correlations”, Korean Conference on Computer Vision (KCCV), Aug. 2024
- “Learning Debiased Classifier with Biased Committee”, Korean Artificial Intelligence Association Fall conference, NAVER and KAIA, Nov. 2022
Services
- Reviewer in NeurIPS (2023, 2024), ICLR (2024, 2025), Tiny Papers@ICLR (2023, 2024), AAAI (2024, 2025), CVPR(2024), ECCV (2024), WACV (2024, 2025), ACCV (2024), ICCV (2023), ICML (2022)