Junsuk Choe

I am an associate professor in the Department of Computer Science and Engineering at Sogang University.  Before joining Sogang University in Fall 2021,  I spent two wonderful years as a Research Intern and then a Research Scientist at NAVER AI Lab, where I had the good fortune to work with Dongyoon Han, Sangdoo Yun, Seong Joon Oh, Sanghyuk Chun, Byeongho Heo, and Jung-Woo Ha.  I obtained my Ph.D. from the Yonsei University in 2020 under the supervision of Prof. Hyunjung Shim, and my B.S. also from the Yonsei University in 2013.   

For Prospective Students: We are currently recruiting one new lab member. Interested students are encouraged to email me.

News

In this year, I serve as an Area Chair for ICLR 2026, COLM 2026, and NeurIPS 2026. 

25.12: We have 1 AAAI, 1 ICML, 1 COLM, 1 NeurIPS, 4 Pattern Recognition Letters, and 1 Neurocomputing paper(s) in 2025.

25.09: Promoted to Associate Professor at Sogang University. Thanks to all my students, colleagues, and collaborators for their support! 🎉

25.08: Granted the '생성AI 선도인재양성사업’ (Co-PI, led by NC AI, funded by IITP) and the ‘독자 AI 파운데이션 모델 프로젝트’ (Co-PI, led by Upstage, funded by NIPA/NIA/IITP)." 

24.12: We have 1 AAAI, 1 WACV, 1 Pattern Recognition Letters, and 1 Pattern Recognition paper(s) in 2024.

24.09: Granted the Global Humanities and Social Convergence Research Program (1.8B KRW)  as a Co-PI from the NRF Korea (PI: Prof. Dasaem Jeong).

24.04: Granted the Outstanding Young Scientist Grant (900M KRW) as a PI from the NRF Korea (Co-PI: Prof. Buru Chang).

23.12: We have 1 TPAMI, 1 ICCV, and 1 Pattern Recognition Letters paper(s) in 2023.

22.12: We have 1 CVPR paper in 2022.

22.10: Received the Sogang College of Engineering Outstanding Lecture Award.

21.12: We have 1 CVPR, 4 ICCV, 1 TPAMI, and 1 Pattern Recognition paper(s) in 2021.

21.09: Joined the Dept. of Computer Science at Sogang University as Assistant Professor.

21.06: Recognized as an Outstanding Reviewer by CVPR 2021.

20.05: Defended Ph.D. thesis and joined NAVER AI Lab as Research Scientist.

Publications

J: Journal, C: Conference, W: Workshop,  P: Preprint / Under review

^Co-corresponding Authors, *Co-first Authors

Bold: Lab Members

[P5] Enhancing Cross-Modal Interactions in Audio–Visual LLMs via Training-Free Attention Injection

Minyoung Lee, Jimin Hong, Junsuk Choe

Under review, 2026.

[P4] Inserting Random Tokens at the Tail: A Simple Training-Free Input Perturbation for LVLMs

Jimin Hong, Minyoung Lee, Junsuk Choe

Under review, 2026.

[P3] Mitigating Cross-Image Information Leakage in Multi-Image Understanding with Large Vision–Language Models

Yeji Park, Minyoung Lee, Sanghyuk Chun, Junsuk Choe

Under review, 2026.

[P2] Knowledge Vector Weakening: Efficient Training-free Unlearning for Large Vision–Language Models

Yejin Kim, Dongjun Hwang, Sungmin Cha, Junsuk Choe

Under review, 2026. 

[P1] ALISE: Alignment-aware Data Selection for Unlearning in Contrastive Vision–Language Models

Dongjun Hwang, Yejin Kim, Beomyun Kwon, Junsuk Choe

Under review, 2026. (Extended version of the ICML 2026 MemFM workshop paper)

[W7] Alignment-aware Data Selection for Unlearning in Contrastive Vision-Language Models

Dongjun Hwang, Yejin Kim, Beomyun Kwon, Junsuk Choe

ICML 2026 Workshop on the Impact of Memorization on Trustworthy Foundation Models.

[W6] Behavioral Proxy Conditioning for Financial Stress Scenario Generation with a Pretrained Diffusion Model

Elena Kuular, Junsuk Choe

ICML 2026 Workshop on Foundation Models for Structured Data.

[W5] EvoVLM: Multimodal Evolutionary Feedback for Visual Symbolic Regression

Hyejin Lee, Junsuk Choe

ICML 2026 Workshop on AI for Science. 

[C20] Rarr: Real-Time Attention-Driven Rain Removal With Hierarchical Scale-Aware Efficient Network

Seungho Eum, Ihjjoon Cho, Jeonghyeon Kim, Junsuk Choe, Unsang Park

ICPR 2026. 

[C19] Training-Free Uncertainty-guided Logit Adjustment for Few-Shot Class-Incremental Learning [code]

Sungwon Woo, Dongjun Hwang, Shiwon Kim, Junsuk Choe, Jongho Nang

CVPR 2026 Findings. 

[W4] Grounding the "Not": Symbolic Representation of Negation for Logical Reasoning in VLMs

Inha Kang, Seonho Lee, Jiho Choi, Junsuk Choe, Hyunjung Shim

ICLR 2026 Workshop on Logical Reasoning of Large Language Models.

[C18] Enhancing Multi-Image Understanding through Delimiter Token Scaling [code]

Minyoung Lee, Yeji Park, Dongjun Hwang, Yejin Kim, Seong Joon Oh, Junsuk Choe 

ICLR 2026. 

[C17] What "Not" to Detect: Negation-Aware VLMs via Structured Reasoning and Token Merging [code]

Inha Kang, Youngsun Lim, Seonho Lee, Jiho Choi, Junsuk Choe, Hyunjung Shim

ICLR 2026. 

[C16] OVS Meets Continual Learning: Towards Sustainable Open-Vocabulary Segmentation [code]

Dongjun Hwang, Yejin Kim, Minyoung Lee, Seong Joon Oh, Junsuk Choe

NeurIPS 2025. 

[C15] Improving Fisher Information Estimation and Efficiency for LoRA-based LLM Unlearning [code]

Yejin Kim*, Eunwon Kim*, Buru Chang^, Junsuk Choe^

COLM 2025. 

[W3] LMLT : Low-to-high Multi-Level Vision Transformer for Lightweight Image Super-Resolution [code]

Jeongsoo Kim, Jongho Nang, Junsuk Choe

ICCV 2025 Workshop on Advances in Image Manipulation.

[J12] Small Object Matters in Weakly Supervised Object Localization

Dongjun Hwang, Seong Joon Oh, Junsuk Choe

Neurocomputing, 2025.  (IF: 5.5) 

[C14] NegMerge: Sign-Consensual Weight Merging for Machine Unlearning [code]

Hyo Seo Kim, Dongyoon Han^, Junsuk Choe^

ICML 2025. 

[J11] VHOIP: Video-based Human-Object Interaction recognition with CLIP Prior knowledge

Doyeol Baek, Junsuk Choe

Pattern Recognition Letters, 2025. (IF: 3.9) 

[J10] Fog-Free Training for Foggy Scene Understanding

Minyoung Lee, Kyungwoo Song, Junsuk Choe

Pattern Recognition Letters, 2025. (IF: 3.9) 

[J9] Curriculum Learning with Class-label Composition for Weakly Supervised Semantic Segmentation

Dongjun Hwang, Hyoseo Kim, Doyeol Baek, Hyunbin Kim, Inhye Kye, Junsuk Choe

Pattern Recognition Letters, 2025. (IF: 3.9) 

[C13] ConVis: Contrastive Decoding with Hallucination Visualization for Mitigating Hallucinations in Multimodal Large Language Models [code]

Yeji Park*, Deokyeong Lee*, Junsuk Choe^, Buru Chang^

AAAI 2025.

[J8] Improving ViT Interpretability with Patch-level Mask Prediction

Junyong Kang, Byeongho Heo, Junsuk Choe

Pattern Recognition Letters, 2025. (IF: 3.9) 

[J7] Discovering an Inference Recipe for Weakly-Supervised Object Localization

Sanghuk Lee*, Cheolhyun Mun*, Youngjung Uh, Junsuk Choe, Hyeran Byun

Pattern Recognition, 2024.  (IF: 7.5) 

[J6] Weakly-supervised Incremental learning for Semantic segmentation with Class Hierarchy

Hyoseo Kim, Junsuk Choe

Pattern Recognition Letters, 2024.  (IF: 3.9) 

[C12] Weakly Supervised Semantic Segmentation for Driving Scenes [code]

Dongseob Kim*, Seungho Lee*, Junsuk Choe, Hyunjung Shim

AAAI 2024.

[C11] Small Objects Matters in Weakly-supervised Semantic Segmentation

Cheolhyun Mun*, Sanghuk Lee*, Youngjung Uh, Junsuk Choe, Hyeran Byun

WACV 2024.

[C10] Neglected Free Lunch - Learning Image Classifiers Using Annotation Byproducts [code]

Dongyoon Han*, Junsuk Choe*, Seonghyeok Chun, John Chung, Minsuk Chang, Sangdoo Yun, Jean Song, Seong Joon Oh

ICCV 2023.

[J5] Entropy Regularization for Weakly Supervised Object Localization

Dongjun Hwang, Jung-Woo Ha, Hyunjung Shim, Junsuk Choe

Pattern Recognition Letters, 2023.  (IF: 3.9) 

[J4] Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasets [arxiv][code]

Junsuk Choe*, Seong Joon Oh*, Sanghyuk Chun, Seungho Lee, Zeynep Akata, Hyunjung Shim

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023.  (IF: 20.8) 

[C9] Weakly Supervised Semantic Segmentation using Out-of-Distribution Data [code]

Jungbeom Lee, Seong Joon Oh, Sangdoo Yun, Junsuk Choe, Eunji Kim, Sungroh Yoon

CVPR 2022.

[J3] Attention-based Dropout Layer for Weakly Supervised Single Object Localization and Semantic Segmentation [code]

Junsuk Choe*, Seungho Lee*, and Hyunjung Shim

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021.  (IF: 24.314)

[C8] Keep CALM and Improve Visual Feature Attribution [code]

Jae Myung Kim*, Junsuk Choe*, Zeynep Akata, Seong Joon Oh

ICCV 2021.

[C7] Normalization Matters in Weakly Supervised Object Localization [code]

Jeesoo Kim, Junsuk Choe, Sangdoo Yun, Nojun Kwak

ICCV 2021.

[C6] Contrastive Attention Maps for Self-supervised Co-localization

Minsong Ki, Youngjung Uh, Junsuk Choe, Hyeran Byun

ICCV 2021.

[C5] Rethinking Spatial Dimensions of Vision Transformers [code]

Byeongho Heo, Sangdoo Yun, Dongyoon Han, Sanghyuk Chun, Junsuk Choe, Seong Joon Oh

ICCV 2021.

[J2] Region-based Dropout with Attention Prior for Weakly Supervised Object Localization 

Junsuk Choe, Dongyoon Han, Sangdoo Yun, Jung-Woo Ha, Seong Joon Oh, Hyunjung Shim

Pattern Recognition, 2021. (Q1; IF: 8.518) 

[C4] Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels [code]

Sangdoo Yun, Seong Joon Oh, Byeongho Heo, Dongyoon Han, Junsuk Choe, Sanghyuk Chun

CVPR 2021. 

[C3] Evaluating Weakly Supervised Object Localization Methods Right [code]

Junsuk Choe*, Seong Joon Oh*, Seungho Lee, Sanghyuk Chun, Zeynep Akata, Hyunjung Shim

CVPR 2020. 

[C2] CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features [code]

Sangdoo Yun, Dongyoon Han, Seong Joon Oh, Sanghyuk Chun, Junsuk Choe, Youngjoon Yoo

ICCV 2019.  (Oral presentation)

[C1] Attention-based Dropout Layer for Weakly Supervised Object Localization [code]

Junsuk Choe, Hyunjung Shim

CVPR 2019.  (Oral presentation)

[W2] An Empirical Evaluation on Robustness and Uncertainty of Regularization Methods

Sanghyuk Chun, Seong Joon Oh, Sangdoo Yun, Dongyoon Han, Junsuk Choe, Youngjoon Yoo

ICML 2019 Workshop on Uncertainty & Robustness in Deep Learning.

[J1] Robust approach to inverse lighting using RGB-D images

Junsuk Choe, Hyunjung Shim

Information Sciences, 2018. (IF: 5.524) 

[W1] Face generation for low-shot learning using generative adversarial networks

Junsuk Choe*, Song Park*, Kyungmin Kim*, Joo Hyun Park*, Dongseob Kim*, Hyunjung Shim

ICCV 2017 Workshop on MS-Celeb-1M Challenge. (Oral presentation)