KAIST
Computer Vision & Machine Learning Lab

News

[Aug, 2025] The paper "3D-Aware Vision-Language Models Fine-Tuning with Geometric Distillation" is accepted to EMNLP 2025 Findings.

[Jun, 2025] The paper "I0T: Embedding Standardization Method Towards Zero Modality Gap" has been selected as oral presentation in ACL 2025!

[May, 2025] We have a paper accepted in ACL 2025.

[Feb, 2025] 3 papers accepted to CVPR 2025. 

[Jan, 2025] 2 papers accepted to ICLR 2025. 

[Jan, 2025] The paper "Label-Augmented Dataset Distillation" has been selected as oral presentation in WACV 2025.

[Jan, 2025] The paper "Evaluating Image Hallucination in Text-to-Image Generation with Question-Answering" has been selected as oral presentation in AAAI 2025.

[Dec, 2024] The paper "Evaluating Image Hallucination in Text-to-Image Generation with Question-Answering" is accepted to AAAI 2025.

[Sep, 2024] The paper "Understanding Multi-Granularity for Open-Vocabulary Part Segmentation" is accepted to NeurIPS 2024.

[Aug, 2024] The paper "Label-Augmented Dataset Distillation" is accepted to WACV 2025.

[Aug, 2024] The paper "Rethinking Data Augmentation for Robust LiDAR Semantic Segmentation in Adverse Weather" has been selected as oral presentation in ECCV 2024.

[Jul, 2024] Prof. Shim will serve as Program Chair of ACCV 2026.

[Jul, 2024] We have three papers accepted in ECCV 2024 and one paper accepted in IROS 2024. 

[Dec, 2023] The paper "Weakly Supervised Semantic Segmentation for Driving Scenes" is accepted to Association for the Advancement of Artificial Intelligence (AAAI).

[Apr. 2023] The paper "Saliency as Pseudo-pixel Supervision for Weakly and Semi-supervised Semantic Segmentation"  is accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI, IF=24.31).

[Feb, 2023] Youngjun and Kyungjune  have granted Ph.D degree and has graduated SIT, Yonsei.

[Jul, 2022] The paper "Few-shot Font Generation with Weakly Supervised Localized Representations"  is accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI, IF=16.39).

[Jul, 2022] News. We moved to the The Kim Jaechul Graduate School of AI at KAIST (KAIST 김재철 AI 대학원).
If you are interested in joining our lab, you should apply to the graduate program at KAIST AI. 

[Apr, 2022] The paper "Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasets"  is accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI, IF=16.39).

[Apr, 2022] The paper "Logit Mixing Training for More Reliable and Accurate Prediction" is accepted to International Joint Conference on Artificial Intelligence (IJCAI).

[Mar, 2022] The paper "Commonality in Natural Images Rescues GANs: Pretraining GANs with Generic and Privacy-free Synthetic Data" is accepted to IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[Mar, 2022] The paper "Threshold Matters in WSSS: Manipulating the Activation for the Robust and Accurate Segmentation Model Against Thresholds" is accepted to IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[Mar, 2022] The paper "Learning from Better Supervision: Self-distillation for Learning with Noisy Labels " is accepted to International Conference on Pattern Recognition (ICPR).

[Sep, 2021] Good news. Dr. Junsuk Choe starts his new career at Sogang University (Computer Science and Engineering) as a faculty member. Congratulation! 

[Aug, 2021] The paper "Distilling from Professors: Enhancing the Knowledge Distillation of Teachers" is accepted to Information Sciences (IF=6.795).

[Jul, 2021] The paper "Rethinking the Truly Unsupervised Image-to-Image Translation" is accepted to International Conference on Computer Vision (ICCV).

[Jul, 2021] The paper "Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts" is accepted to International Conference on Computer Vision (ICCV).

[Mar, 2021] The paper "Region-based Dropout with Attention Prior for Weakly Supervised Object Localization" is accepted to Pattern Recognition (IF=7.196).

[Mar, 2021] The paper "Railroad is not a Train: Saliency as Pseudo-pixel Supervision for Weakly Supervised Semantic Segmentation " is accepted to IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[Jan 16, 2021] We have one paper "Session-aware Linear Item-Item Models for Session-based Recommendation" accepted to The Web Conference (WWW).

[Dec, 2020] The paper "Few-shot Font Generation with Localized Style Representations and Factorization" is accepted to Association for the Advancement of Artificial Intelligence (AAAI).

[Oct, 2020] The paper "Rigid and non-rigid motion artifact reduction in X-ray CT using attention module" is accepted to Medical Image Analysis (IF=11.148).

[Aug, 2020] Congratulations! Junsuk Choe has granted Ph.D degree and has graduated SIT, Yonsei.

[Aug, 2020] The paper "GridMix: Strong Regularization Through Local Context Mapping" is accepted to Pattern Recognition (IF=7.196).

[May, 2020] The paper "Attention-based Dropout Layer for Weakly Supervised Single Object Localization and Semantic Segmentation" is accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence (IF=17.730).

[Feb, 2020] Congratulations! Duhyeon Bang has granted Ph.D degree and has graduated SIT, Yonsei.

[Feb, 2020] The paper "Discriminator Feature-based Inference by Recycling the Discriminator of GANs" is accepted to International Journal of Computer Vision (IF=6.071).