SC
Sanghyuk Chun
cs.CVcs.LGcs.AIcs.SDeess.ASstat.MLcs.IRcs.CYAlgorithmsArtificial intelligence
On Valency
published · living versionsW_zydnf4v8·v1 · currentpublished
Multi-State Linear Attention with Generalized Second-Order Routing
with Hee Seung Hwang, Xindi Wu, Olga Russakovsky, Zhiwei Deng
1 version
Preprints & journals
53 papers in the corpus · 2015–2026In-Context Learning Can Help Vision-Language Models Overcome Training Priors2610.12942v3 · Kun Wang, Xindi Wu, Sanghyuk Chun et al.2026 · 0 citationsarXiv
Depth as Time in One-Step Generative Models2610.03626v1 · Arnold Caleb Asiimwe, William Yang, Sanghyuk Chun et al.2026 · 0 citationsarXiv
Do Counterfactually Fair Image Classifiers Satisfy Group Fairness? -- A Theoretical and Empirical Study2607.06603v1 · Sangwon Jung, Sumin Yu, Sanghyuk Chun et al.2026 · 1 citationarXiv
CoMet: Context and Multiplicity Decomposition for Multimodal Uncertainty Estimation2606.32012v1 · Sanghyuk Chun, William Yang, Amaya Dharmasiri et al.2026 · 0 citationsarXiv
Mitigating Cross-Image Information Leakage in Multi-Image Understanding with Large Vision-Language Models2508.13744v3 · Yeji Park, Minyoung Lee, Sanghyuk Chun et al.2025 · 0 citationsarXiv
Emergence of Text Readability in Vision Language Models2506.19389v1 · Jaeyoo Park, Sanghyuk Chun, Wonjae Kim et al.2025 · 0 citationsarXiv
Multiplicity is an Inevitable and Inherent Challenge in Multimodal Learning2505.19614v2 · Sanghyuk Chun, Olga Russakovsky2025 · 1 citationarXiv
LongProLIP: A Probabilistic Vision-Language Model with Long Context Text2503.08048v2 · Sanghyuk Chun, Sangdoo Yun2025 · 0 citationsarXiv
FairDRO: Group fairness regularization via classwise robust optimization.39561587 · Park, Taeeon, Jung, Sangwon, Chun, Sanghyuk et al.2024 · 7 citationsNeural networks : the official journal of the International Neural Network Society. 2025;182:106891
Probabilistic Language-Image Pre-Training2410.18857v4 · Sanghyuk Chun, Wonjae Kim, Song Park et al.2024 · 0 citationsarXiv
Read, Watch and Scream! Sound Generation from Text and Video2407.05551v2 · Yujin Jeong, Yunji Kim, Sanghyuk Chun et al.2024 · 14 citationsarXiv
An Efficient Post-hoc Framework for Reducing Task Discrepancy of Text Encoders for Composed Image Retrieval2406.09188v2 · Jaeseok Byun, Seokhyeon Jeong, Wonjae Kim et al.2024 · 2 citationsarXiv
HYPE: Hyperbolic Entailment Filtering for Underspecified Images and Texts2404.17507v2 · Wonjae Kim, Sanghyuk Chun, Taekyung Kim et al.2024 · 8 citationsarXiv
Toward Interactive Regional Understanding in Vision-Large Language Models2403.18260v1 · Jungbeom Lee, Sanghyuk Chun, Sangdoo Yun2024 · 1 citationarXiv
Few-Shot Font Generation With Weakly Supervised Localized Representations.35947564 · Park, Song, Chun, Sanghyuk, Cha, Junbum et al.2024 · 22 citationsIEEE transactions on pattern analysis and machine intelligence. 2024;46(3):1479-1495
Language-only Efficient Training of Zero-shot Composed Image Retrieval2312.01998v2 · Geonmo Gu, Sanghyuk Chun, Wonjae Kim et al.2023 · 66 citationsarXiv
Learning with Unmasked Tokens Drives Stronger Vision Learners2310.13593v3 · Taekyung Kim, Sanghyuk Chun, Byeongho Heo et al.2023 · 1 citationarXiv
Improved Probabilistic Image-Text Representations2305.18171v5 · Sanghyuk Chun2023 · 8 citationsarXiv
RoCOCO: Robustness Benchmark of MS-COCO to Stress-test Image-Text Matching Models2304.10727v5 · Seulki Park, Daeho Um, Hajung Yoon et al.2023 · 1 citationarXiv
Three Recipes for Better 3D Pseudo-GTs of 3D Human Mesh Estimation in the Wild2304.04875v1 · Gyeongsik Moon, Hongsuk Choi, Sanghyuk Chun et al.2023 · 9 citationsarXiv
CompoDiff: Versatile Composed Image Retrieval With Latent Diffusion2303.11916v4 · Geonmo Gu, Sanghyuk Chun, Wonjae Kim et al.2023 · 20 citationsarXiv
SeiT: Storage-Efficient Vision Training with Tokens Using 1% of Pixel Storage2303.11114v2 · Song Park, Sanghyuk Chun, Byeongho Heo et al.2023 · 7 citationsarXiv
Re-weighting Based Group Fairness Regularization via Classwise Robust Optimization2303.00442v1 · Sangwon Jung, Taeeon Park, Sanghyuk Chun et al.2023 · 5 citationsarXiv
Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasets.35468058 · Choe, Junsuk, Oh, Seong Joon, Chun, Sanghyuk et al.2023 · 21 citationsIEEE transactions on pattern analysis and machine intelligence. 2023;45(2):1732-1748
Group Generalized Mean Pooling for Vision Transformer2212.04114v1 · Byungsoo Ko, Han-Gyu Kim, Byeongho Heo et al.2022 · 4 citationsarXiv
Similarity of Neural Architectures using Adversarial Attack Transferability2210.11407v4 · Jaehui Hwang, Dongyoon Han, Byeongho Heo et al.2022 · 2 citationsarXiv
An Extendable, Efficient and Effective Transformer-based Object Detector2204.07962v1 · Hwanjun Song, Deqing Sun, Sanghyuk Chun et al.2022 · 11 citationsarXiv
ECCV Caption: Correcting False Negatives by Collecting Machine-and-Human-verified Image-Caption Associations for MS-COCO2204.03359v5 · Sanghyuk Chun, Wonjae Kim, Song Park et al.2022 · 32 citationsarXiv
Domain Generalization by Mutual-Information Regularization with Pre-trained Models2203.10789v2 · Junbum Cha, Kyungjae Lee, Sungrae Park et al.2022 · 129 citationsarXiv
Dataset Condensation with Contrastive Signals2202.02916v3 · Saehyung Lee, Sanghyuk Chun, Sangwon Jung et al.2022 · 37 citationsarXiv
Learning Fair Classifiers with Partially Annotated Group Labels2111.14581v2 · Sangwon Jung, Sanghyuk Chun, Taesup Moon2021 · 26 citationsarXiv
ViDT: An Efficient and Effective Fully Transformer-based Object Detector2110.03921v2 · Hwanjun Song, Deqing Sun, Sanghyuk Chun et al.2021 · 62 citationsarXiv
Which Shortcut Cues Will DNNs Choose? A Study from the Parameter-Space Perspective2110.03095v2 · Luca Scimeca, Seong Joon Oh, Sanghyuk Chun et al.2021 · 10 citationsarXiv
StyleAugment: Learning Texture De-biased Representations by Style Augmentation without Pre-defined Textures2108.10549v1 · Sanghyuk Chun, Song Park2021 · 3 citationsarXiv
Neural Hybrid Automata: Learning Dynamics with Multiple Modes and Stochastic Transitions2106.04165v1 · Michael Poli, Stefano Massaroli, Luca Scimeca et al.2021 · 8 citationsarXiv
Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts2104.00887v1 · Song Park, Sanghyuk Chun, Junbum Cha et al.2021 · 107 citationsarXiv
Rethinking Spatial Dimensions of Vision Transformers2103.16302v2 · Byeongho Heo, Sangdoo Yun, Dongyoon Han et al.2021 · 680 citationsarXiv
SWAD: Domain Generalization by Seeking Flat Minima2102.08604v4 · Junbum Cha, Sanghyuk Chun, Kyungjae Lee et al.2021 · 209 citationsarXiv
Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels2101.05022v2 · Sangdoo Yun, Seong Joon Oh, Byeongho Heo et al.2021 · 115 citationsarXiv
Probabilistic Embeddings for Cross-Modal Retrieval2101.05068v2 · Sanghyuk Chun, Seong Joon Oh, Rafael Sampaio de Rezende et al.2021 · 270 citationsarXiv
Few-shot Font Generation with Localized Style Representations and Factorization2009.11042v2 · Song Park, Sanghyuk Chun, Junbum Cha et al.2020 · 113 citationsarXiv
AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights2006.08217v3 · Byeongho Heo, Sanghyuk Chun, Seong Joon Oh et al.2020 · 110 citationsarXiv
Few-shot Compositional Font Generation with Dual Memory2005.10510v2 · Junbum Cha, Sanghyuk Chun, Gayoung Lee et al.2020 · 108 citationsarXiv
An Empirical Evaluation on Robustness and Uncertainty of Regularization Methods2003.03879v1 · Sanghyuk Chun, Seong Joon Oh, Sangdoo Yun et al.2020 · 37 citationsarXiv
Evaluating Weakly Supervised Object Localization Methods Right2001.07437v2 · Junsuk Choe, Seong Joon Oh, Seungho Lee et al.2020 · 203 citationsarXiv
Visualizing and Understanding Self-attention based Music Tagging1911.04385v1 · Minz Won, Sanghyuk Chun, Xavier Serra2019 · 2 citationsarXiv
Neural Approximation of an Auto-Regressive Process through Confidence Guided Sampling1910.06705v1 · YoungJoon Yoo, Sanghyuk Chun, Sangdoo Yun et al.2019 · 0 citationsarXiv
Learning De-biased Representations with Biased Representations1910.02806v3 · Hyojin Bahng, Sanghyuk Chun, Sangdoo Yun et al.2019 · 147 citationsarXiv
Toward Interpretable Music Tagging with Self-Attention1906.04972v1 · Minz Won, Sanghyuk Chun, Xavier Serra2019 · 35 citationsarXiv
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features1905.04899v2 · Sangdoo Yun, Dongyoon Han, Seong Joon Oh et al.2019 · 5,386 citationsarXiv
Career total: 64 works. 53 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.