← Search

Byungjai Kim

2 accepted papers

2025

Test-Time Ensemble via Linear Mode Connectivity: A Path to Better Adaptation

ICLR 2025poster

Test-time adaptation updates pretrained models on the fly to handle distribution shifts in test data. While existing research has focused on stable optimization during adaptation, less attention has been given to enhancing model representations for adaptation capability. To address this gap, we prop…

Cited by 0SourcePDFScholar
2023

BiasAdv: Bias-Adversarial Augmentation for Model Debiasing

CVPR 2023poster

Neural networks are often prone to bias toward spurious correlations inherent in a dataset, thus failing to generalize unbiased test criteria. A key challenge to resolving the issue is the significant lack of bias-conflicting training data (i.e., samples without spurious correlations). In this paper…

Cited by 30SourcePDFScholar