Published in ECCV, 2024
A robust test-time adaptation method using batch norm with statistics memory.
Recommended citation: Kang, J., Kim, N., Ok, J., & Kwak, S. (2024). "MemBN: Robust Test-Time Adaptation via Batch Norm with Statistics Memory." ECCV.
Download Paper
Published in ICML, 2024
Improving robustness to multiple spurious correlations via multi-objective optimization.
Recommended citation: Kim, N., Kang, J., Ahn, S., Ok, J., & Kwak, S. (2024). "Improving Robustness to Multiple Spurious Correlations by Multi-Objective Optimization." ICML.
Download Paper
Published in ICCV 2023 - Workshop on MMFM: What is Next in Multimodal Foundation Models? (MMFM), 2023
Exploiting synthetic data to address data imbalance problems.
Recommended citation: Ye-Bin, M.*, Hyeon-Woo, N.*, Choi, W., Kim, N., Kwak, S., & Oh, T.-H. (2023). "SYNAuG: Exploiting Synthetic Data for Data Imbalance Problems." ICCV 2023 Workshop on MMFM. (* equal contribution)
Download Paper
Published in ICML 2023 - Workshop on Spurious Correlations, Invariance and Stability (SCIS), 2023
Removing multiple biases through the lens of multi-task learning.
Recommended citation: Kim, N., Kang, J., Ahn, S., Ok, J., & Kwak, S. (2023). "Removing Multiple Biases through the Lens of Multi-task Learning." ICML Workshop on SCIS.
Download Paper
Published in ICML, 2023
Leveraging a proxy of training data for test-time adaptation.
Recommended citation: Kang, J., Kim, N., Kwon, D., Ok, J., & Kwak, S. (2023). "Leveraging Proxy of Training Data for Test-Time Adaptation." ICML.
Download Paper
Published in NeurIPS, 2022
Learning a debiased classifier with a biased committee.
Recommended citation: Kim, N., Hwang, S., Ahn, S., Park, J., & Kwak, S. (2022). "Learning Debiased Classifier with Biased Committee." NeurIPS.
Download Paper
Published in BMVC, 2021
A robust crowd counting method using image enhancement and dynamic feature selection.
Recommended citation: Kim, N., & Kwak, S. (2021). "Robust Crowd Counting via Image Enhancement and Dynamic Feature Selection." BMVC.
Download Paper
Published in arXiv preprint, 2025
A group prompt learning approach for group-robust fine-tuning.
Recommended citation: Kim, N., Oh, S. J., & Kwak, S. (2025). "GroupCoOp: Group-robust Fine-tuning via Group Prompt Learning." arXiv preprint.
Download Paper
Published in arXiv preprint, 2025
A domain-invariant learning approach for continual test-time adaptation.
Recommended citation: Lee, S., Kim, N., Kang, J., Oh, S. J., & Kwak, S. (2025). "DiCoTTA: Domain-invariant Learning for Continual Test-time Adaptation." arXiv preprint.
Download Paper