Sanghyuk Chun

papers

2,494

total citations

papers (19)

Domain Generalization by Mutual-Information Regularization with Pre-trained Models

ECCV 2022arXiv

174

citations

Re-Labeling ImageNet: From Single to Multi-Labels, From Global to Localized Labels

CVPR 2021arXiv

164

citations

Few-shot Compositional Font Generation with Dual Memory

ECCV 2020arXiv

citations

Multiple Heads Are Better Than One: Few-Shot Font Generation With Multiple Localized Experts

ICCV 2021arXiv

citations

ECCV Caption: Correcting False Negatives by Collecting Machine-and-Human-Verified Image-Caption Associations for MS-COCO

ECCV 2022arXiv

citations

Learning Fair Classifiers With Partially Annotated Group Labels

CVPR 2022arXiv

citations

Improved Probabilistic Image-Text Representations

ICLR 2024arXiv

citations

A Unified Analysis of Mixed Sample Data Augmentation: A Loss Function Perspective

NEURIPS 2022arXiv

citations

Read, Watch and Scream! Sound Generation from Text and Video

AAAI 2025arXiv

citations

Neural Hybrid Automata: Learning Dynamics With Multiple Modes and Stochastic Transitions

NEURIPS 2021arXiv

citations

SeiT: Storage-Efficient Vision Training with Tokens Using 1% of Pixel Storage

ICCV 2023arXiv

citations

Learning with Unmasked Tokens Drives Stronger Vision Learners

ECCV 2024arXiv

citations

An Efficient Post-hoc Framework for Reducing Task Discrepancy of Text Encoders for Composed Image Retrieval

ICCV 2025arXiv

citations

Language-only Training of Zero-shot Composed Image Retrieval

CVPR 2024

citations

What does automatic differentiation compute for neural networks?

ICLR 2024

citations

Sanghyuk Chun

papers (19)

Rethinking Spatial Dimensions of Vision Transformers

SWAD: Domain Generalization by Seeking Flat Minima

Probabilistic Embeddings for Cross-Modal Retrieval

Evaluating Weakly Supervised Object Localization Methods Right

Domain Generalization by Mutual-Information Regularization with Pre-trained Models

Re-Labeling ImageNet: From Single to Multi-Labels, From Global to Localized Labels

Few-shot Compositional Font Generation with Dual Memory

Multiple Heads Are Better Than One: Few-Shot Font Generation With Multiple Localized Experts

ECCV Caption: Correcting False Negatives by Collecting Machine-and-Human-Verified Image-Caption Associations for MS-COCO

Learning Fair Classifiers With Partially Annotated Group Labels

Improved Probabilistic Image-Text Representations

A Unified Analysis of Mixed Sample Data Augmentation: A Loss Function Perspective

Read, Watch and Scream! Sound Generation from Text and Video

Neural Hybrid Automata: Learning Dynamics With Multiple Modes and Stochastic Transitions

SeiT: Storage-Efficient Vision Training with Tokens Using 1% of Pixel Storage

Learning with Unmasked Tokens Drives Stronger Vision Learners

An Efficient Post-hoc Framework for Reducing Task Discrepancy of Text Encoders for Composed Image Retrieval

Language-only Training of Zero-shot Composed Image Retrieval

What does automatic differentiation compute for neural networks?

papers (19)

Rethinking Spatial Dimensions of Vision Transformers

SWAD: Domain Generalization by Seeking Flat Minima

Probabilistic Embeddings for Cross-Modal Retrieval

Evaluating Weakly Supervised Object Localization Methods Right

Domain Generalization by Mutual-Information Regularization with Pre-trained Models

Re-Labeling ImageNet: From Single to Multi-Labels, From Global to Localized Labels

Few-shot Compositional Font Generation with Dual Memory

Multiple Heads Are Better Than One: Few-Shot Font Generation With Multiple Localized Experts

ECCV Caption: Correcting False Negatives by Collecting Machine-and-Human-Verified Image-Caption Associations for MS-COCO

Learning Fair Classifiers With Partially Annotated Group Labels

Improved Probabilistic Image-Text Representations

A Unified Analysis of Mixed Sample Data Augmentation: A Loss Function Perspective

Read, Watch and Scream! Sound Generation from Text and Video

Neural Hybrid Automata: Learning Dynamics With Multiple Modes and Stochastic Transitions

SeiT: Storage-Efficient Vision Training with Tokens Using 1% of Pixel Storage

Learning with Unmasked Tokens Drives Stronger Vision Learners

An Efficient Post-hoc Framework for Reducing Task Discrepancy of Text Encoders for Composed Image Retrieval

Language-only Training of Zero-shot Composed Image Retrieval

What does automatic differentiation compute for neural networks?