Jungsoo Lee

PhD Student
KAIST AI
bebeto[at]kaist[dot]ac[dot]kr

[Google Scholar] [LinkedIn] [Github] [CV]

About Me

I am a PhD student at Korea Advanced Institute of Science and Technology (KAIST) advised by Professor Jaegul Choo. I am mainly interested in robustness of AI, previously focused on debiasing and addressing domain shifts in image classification.

Personal News

[Dec 2023] I got promoted as a senior research engineer at Qualcomm Korea.
[Apr 2023] I started my regular position as a machine learning research engineer at Qualcomm Korea.
[Apr 2023] I finished my internship at Qualcomm Korea.
[Oct 2022] I started my internship at Qualcomm Korea.
[Aug 2022] I finished my internship at AI Lab, Kakao Enterprise.
[Aug 2021] I started my internship at AI Lab, Kakao Enterprise.

Publication News

[July 2023] Our "Wisdom of Crowds" paper was accepted at ICCV 2023.
[July 2023] Our "CAFA" paper was accepted at ICCV 2023.
[June 2023] Our "ReLD" paper was accepted at ECML/PKDD 2023.
[Feb 2023] Our "EcoTTA" paper was accepted at CVPR 2023.
[Nov 2022] Our "BiasEnsemble" paper was accepted as oral presentation at AAAI 2023.
[April 2022] Our "DASH" paper was accepted at EuroVis 2022 (Short paper).
[Oct 2021] Our "Inter-Prototype (Face)" paper was accepted at BMVC 2021.
[Sep 2021] Our "Debiasing" paper was accepted as oral presentation at NeurIPS 2021.
[July 2021] Our "SML" paper was accepted as oral presentation at ICCV 2021.

Publications

Towards Open-set Test-Time Adaptation Utilizing the Wisdom of Crowds in Entropy Minimization.
Jungsoo Lee, Debasmit Das, Jaegul Choo, and Sungha Choi.
ICCV 2023
[Arxiv]

CAFA: Class-Aware Feature Alignment for Test-Time Adaptation
Sanghun Jung, Jungsoo Lee, Nanhee Kim, Amirreza Shaban, Byron Boots, and Jaegul Choo
ICCV 2023
[Arxiv]

Deep Imbalanced Time-series Forecasting via Local Discrepancy Density.
Junwoo Park, Jungsoo Lee, Youngin Cho, Woncheol Shin, Dongmin Kim, Jaegul Choo, and Edward Choi.
ECML/PKDD 2023
[Arxiv]

EcoTTA: Memory-Efficient Continual Test-time Adaptation via Self-distilled Regularization.
Junha Song, Jungsoo Lee, In So Kweon, and Sungha Choi.
CVPR 2023
[Arxiv]

Revisiting the Importance of Amplifying Bias for Debiasing
Jungsoo Lee*, Jeonghoon Park*, Daeyoung Kim*, Juyoung Lee, Edward Choi, and Jaegul Choo (*: equal contributions)
AAAI 2023, Accepted as Oral Presentation
[Arxiv] [Github]

DASH: Visual Analytics for Debiasing Image Classification via User-Driven Synthetic Data Augmentation.
Bum Chul Kwon, Jungsoo Lee, Chaeyeon Chung, Nyoungwoo Lee, Ho-jin Choi, and Jaegul Choo.
EuroVis 2022 (Short paper)

Improving Face Recognition with Large Age Gaps by Learning to Distinguish Children
Jungsoo Lee*, Jooyeol Yun*, Sunghyun Park, Yonggyu Kim, and Jaegul Choo (*: equal contributions)
BMVC 2021
[Arxiv] [Github]

Learning Debiased Representation via Disentangled Feature Augmentation
Jungsoo Lee*, Eungyeup Kim*, Juyoung Lee, Jihyeon Lee, and Jaegul Choo (*: equal contributions)
NeurIPS 2021, Accepted as Oral Presentation (<1%)
[Arxiv] [Github]

Standardized Max Logits: A Simple yet Effective Approach for Identifying Unexpected Road Obstacles in Urban-Scene Segmentation
Sanghun Jung*, Jungsoo Lee*, Daehoon Gwak, Sungha Choi, and Jaegul Choo (*: equal contributions)
ICCV 2021, Accepted as Oral Presentation (<3%)
[Arxiv] [Github] [Video]

Understanding Human-side Impact of Sampling Image Batches in Subjective Attribute Labeling
Chaeyeon Chung*, Jungsoo Lee*, Kyungmin Park, Junsoo Lee, Minjae Kim, Mookyung Song, Yeonwoo Kim, Jaegul Choo, and Sungsoo (Ray) Hong (*: equal contributions)
CSCW 2021
[ACM DL] [Github]

Love in Lyrics: An Exploration of Supporting Textual Manifestation of Affection in Social Messaging
Taewook Kim, Jungsoo Lee, Zhenhui Peng, and Xiaojuan Ma
CSCW 2019
[ACM DL]