59 citations · 136 across the 21 of their papers we have counts for
23 papers
Balancing Saliency and Coverage: Semantic Prominence-Aware Budgeting for Visual Token Compression in VLMs
Jaehoon Lee, Mingi Jung, Soohyuk Jang +3
Large Vision-Language Models (VLMs) achieve strong multimodal understanding capabilities by leveraging high-resolution visual inputs, but the resulting large number of visual token…
Know "No" Better: A Data-Driven Approach for Enhancing Negation Awareness in CLIP
Junsung Park, Jungbeom Lee, Jongyoon Song +3
While CLIP has significantly advanced multimodal understanding by bridging vision and language, the inability to grasp negation - such as failing to differentiate concepts like "pa…
Textual Training for the Hassle-Free Removal of Unwanted Visual Data: Case Studies on OOD and Hateful Image Detection
Saehyung Lee, Jisoo Mok, Sangha Park +3
In our study, we explore methods for detecting unwanted content lurking in visual datasets. We provide a theoretical analysis demonstrating that a model capable of successfully par…
Normality Addition via Normality Detection in Industrial Image Anomaly Detection Models
Jihun Yi, Dahuin Jung, Sungroh Yoon
The task of image anomaly detection (IAD) aims to identify deviations from normality in image data. These anomalies are patterns that deviate significantly from what the IAD model…
Disentangled Motion Modeling for Video Frame Interpolation
Jaihyun Lew, Jooyoung Choi, Chaehun Shin +2
Video Frame Interpolation (VFI) aims to synthesize intermediate frames between existing frames to enhance visual smoothness and quality. Beyond the conventional methods based on th…
Entropy is not Enough for Test-Time Adaptation: From the Perspective of Disentangled Factors
Jonghyun Lee, Dahuin Jung, Saehyung Lee +4
Test-time adaptation (TTA) fine-tunes pre-trained deep neural networks for unseen test data. The primary challenge of TTA is limited access to the entire test dataset during online…
Efficient Diffusion-Driven Corruption Editor for Test-Time Adaptation
Yeongtak Oh, Jonghyun Lee, Jooyoung Choi +3
Test-time adaptation (TTA) addresses the unforeseen distribution shifts occurring during test time. In TTA, performance, memory consumption, and time consumption are crucial consid…
On the Powerfulness of Textual Outlier Exposure for Visual OoD Detection
Sangha Park, Jisoo Mok, Dahuin Jung +2
Successful detection of Out-of-Distribution (OoD) data is becoming increasingly important to ensure safe deployment of neural networks. One of the main challenges in OoD detection…
PUCA: Patch-Unshuffle and Channel Attention for Enhanced Self-Supervised Image Denoising
Hyemi Jang, Junsung Park, Dahuin Jung +3
Although supervised image denoising networks have shown remarkable performance on synthesized noisy images, they often fail in practice due to the difference between real and synth…
Improving Visual Prompt Tuning for Self-supervised Vision Transformers
Seungryong Yoo, Eunji Kim, Dahuin Jung +2
Visual Prompt Tuning (VPT) is an effective tuning method for adapting pretrained Vision Transformers (ViTs) to downstream tasks. It leverages extra learnable tokens, known as promp…
Probabilistic Concept Bottleneck Models
Eunji Kim, Dahuin Jung, Sangha Park +2
Interpretable models are designed to make decisions in a human-interpretable manner. Representatively, Concept Bottleneck Models (CBM) follow a two-step process of concept predicti…
Diffusion-Stego: Training-free Diffusion Generative Steganography via Message Projection
Daegyu Kim, Chaehun Shin, Jooyoung Choi +2
Generative steganography is the process of hiding secret messages in generated images instead of cover images. Existing studies on generative steganography use GAN or Flow models t…
Sample-efficient Adversarial Imitation Learning
Dahuin Jung, Hyungyu Lee, Sungroh Yoon
Imitation learning, in which learning is performed by demonstration, has been studied and advanced for sequential decision-making tasks in which a reward function is not predefined…
New Insights for the Stability-Plasticity Dilemma in Online Continual Learning
Dahuin Jung, Dongjin Lee, Sunwon Hong +3
The aim of continual learning is to learn new tasks continuously (i.e., plasticity) without forgetting previously learned knowledge from old tasks (i.e., stability). In the scenari…
FedClassAvg: Local Representation Learning for Personalized Federated Learning on Heterogeneous Neural Networks
Jaehee Jang, Heonseok Ha, Dahuin Jung +1
Personalized federated learning is aimed at allowing numerous clients to train personalized models while participating in collaborative training in a communication-efficient manner…
Confidence Score for Source-Free Unsupervised Domain Adaptation
Jonghyun Lee, Dahuin Jung, Junho Yim +1
Source-free unsupervised domain adaptation (SFUDA) aims to obtain high performance in the unlabeled target domain using the pre-trained source model, not the source data. Existing…
Stein Latent Optimization for Generative Adversarial Networks
Uiwon Hwang, Heeseung Kim, Dahuin Jung +3
Generative adversarial networks (GANs) with clustered latent spaces can perform conditional generation in a completely unsupervised manner. In the real world, the salient attribute…
iCaps: An Interpretable Classifier via Disentangled Capsule Networks
Dahuin Jung, Jonghyun Lee, Jihun Yi +1
We propose an interpretable Capsule Network, iCaps, for image classification. A capsule is a group of neurons nested inside each layer, and the one in the last layer is called a cl…
PixelSteganalysis: Destroying Hidden Information with a Low Degree of Visual Degradation
Dahuin Jung, Ho Bae, Hyun-Soo Choi +1
Steganography is the science of unnoticeably concealing a secret message within a certain image, called a cover image. The cover image with the secret message is called a stego ima…
PixelSteganalysis: Pixel-wise Hidden Information Removal with Low Visual Degradation
Dahuin Jung, Ho Bae, Hyun-Soo Choi +1
Recently, the field of steganography has experienced rapid developments based on deep learning (DL). DL based steganography distributes secret information over all the available bi…
HexaGAN: Generative Adversarial Nets for Real World Classification
Uiwon Hwang, Dahuin Jung, Sungroh Yoon
Most deep learning classification studies assume clean data. However, when dealing with the real world data, we encounter three problems such as 1) missing data, 2) class imbalance…
AnomiGAN: Generative adversarial networks for anonymizing private medical data
Ho Bae, Dahuin Jung, Sungroh Yoon
Typical personal medical data contains sensitive information about individuals. Storing or sharing the personal medical data is thus often risky. For example, a short DNA sequence…
Security and Privacy Issues in Deep Learning
Ho Bae, Jaehee Jang, Dahuin Jung +4
To promote secure and private artificial intelligence (SPAI), we review studies on the model security and data privacy of DNNs. Model security allows system to behave as intended w…