Divide and Translate: Compositional First-Order Logic Translation and Verification for Complex Logical Reasoning

ICLR 2025 Divide and Translate: Compositional First-Order Logic Translation and Verification for Complex Logical Reasoning Hyun Ryu, Gyeongman Kim, Hyemin S. Lee, Eunho Yang  Complex logical reasoning tasks require a long sequence of reasoning, which a large language model (...

Towards Precise Prediction Uncertainty in GNNs: Refining GNNs with Topology-grouping Strategy

AAAI 2024 Towards Precise Prediction Uncertainty in GNNs: Refining GNNs with Topology-grouping Strategy  Hyunjin Seo, Kyusung Seo, Joonhyung Park, Eunho Yang  Recent advancements in graph neural networks (GNNs) have highlighted the critical need of calibrating model predicti...

TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias

ECCV 2024 TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias Sanghyun Jo, Soohyun Ryu, Sungyub Kim, Eunho Yang, Kyungsu Kim  We identify a critical bias in contemporary CLIP-based models, which we denote as single tag bias. This bi...

CloudFixer: Test-Time Adaptation for 3D Point Clouds via Diffusion-Guided Geometric Transformation

ECCV 2024 CloudFixer: Test-Time Adaptation for 3D Point Clouds via Diffusion-Guided Geometric Transformation  Hajin Shim*, Changhun Kim*, Eunho Yang (*: equal contribution)   3D point clouds captured from real-world sensors frequently encompass noisy points due to ...

PruNeRF: Segment-Centric Dataset Pruning via 3D Spatial Consistency

ICML 2024 PruNeRF: Segment-Centric Dataset Pruning via 3D Spatial Consistency  Yeonsung Jung, Heecheol Yun, Joonhyung Park, Jin-Hwa Kim, Eunho Yang  Neural Radiance Fields (NeRF) have shown remarkable performance in learning 3D scenes. However, NeRF exhibits vulnerability wh...

Data-Efficient Unsupervised Interpolation Without Any Intermediate Frame for 4D Medical Images

CVPR 2024 Data-Efficient Unsupervised Interpolation Without Any Intermediate Frame for 4D Medical Images JungEun Kim, Hangyul Yoon, Geondo Park, Kyungsu Kim, Eunho Yang  4D medical images, which represent 3D images with temporal information, are crucial in clinical practice for c...

TEDDY: Trimming Edges with Degree-based Graph Discrimination strategY

ICLR 2024 TEDDY: Trimming Edges with Degree-based Graph Discrimination strategY  Hyunjin Seo*, Jihun Yun* and Eunho Yang (*: equal contribution)   Since the pioneering work on the lottery ticket hypothesis for graph neural networks (GNNs) was proposed in Chen et al...

Language-Interfaced Tabular Oversampling via Progressive Imputation and Self-Authentication

ICLR 2024 Language-Interfaced Tabular Oversampling via Progressive Imputation and Self-Authentication   Geondo Park, June Yong Yang, Joowon kim, Hyeongwon Jang and Eunho Yang   Tabular data in the wild are frequently afflicted with class-imbalance, biasing mac...

COMPACT AND DE-BIASED NEGATIVE INSTANCE EMBEDDING FOR MULTI-INSTANCE LEARNING ON WHOLE-SLIDE IMAGE CLASSIFICATION

ICASSP 2024 COMPACT AND DE-BIASED NEGATIVE INSTANCE EMBEDDING FOR MULTI-INSTANCELEARNING ON WHOLE-SLIDE IMAGE CLASSIFICATION  Joohyung Lee, Heejeong Nam, Kwanhyung Lee, Sangchul Hahn   Whole-slide image (WSI) classification is a challenging task because 1) patches ...

SeamsTalk: Seamless Talking Face Generation via Flow-Guided Inpainting

IEEE 2024 SeamsTalk: Seamless Talking Face Generation via Flow-Guided Inpainting Yeongho Jeong, Gyeongman Kim, Doohyuk Jang, Jaeryong Hwang, Eunho Yang  Talking face generation aims to generate a face video to speak according to a given audio or driving video. Despite the im...