Author name cluster

Jungsoo Lee

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

1 author row

NeurIPS Conference 2025 Conference Paper

Generalized Contrastive Learning for Universal Multimodal Retrieval

Jungsoo Lee
Janghoon Cho
Hyojin Park
Durga Malladi
Kyuwoong Hwang
Fatih Porikli
Sungha Choi

Despite their consistent performance improvements, cross-modal retrieval models (e. g. , CLIP) show degraded performances with retrieving keys composed of fused image-text modality (e. g. , Wikipedia pages with both images and text). To address this critical challenge, multimodal retrieval has been recently explored to develop a unified single retrieval model capable of retrieving keys across diverse modality combinations. A common approach involves constructing new composed sets of image-text triplets (e. g. , retrieving a pair of image and text given a query image). However, such an approach requires careful curation to ensure the dataset quality and fails to generalize to unseen modality combinations. To overcome these limitations, this paper proposes Generalized Contrastive Learning (GCL), a novel loss formulation that improves multimodal retrieval performance without the burdensome need for new dataset curation. Specifically, GCL operates by enforcing contrastive learning across all modalities within a mini-batch, utilizing existing image-caption paired datasets to learn a unified representation space. We demonstrate the effectiveness of GCL by showing consistent performance improvements on off-the-shelf multimodal retrieval models (e. g. , VISTA, CLIP, and TinyCLIP) using the M-BEIR, MMEB, and CoVR benchmarks.

PDF Details

AAAI Conference 2023 Conference Paper

Revisiting the Importance of Amplifying Bias for Debiasing

Jungsoo Lee
Jeonghoon Park
Daeyoung Kim
Juyoung Lee
Edward Choi
Jaegul Choo

In image classification, debiasing aims to train a classifier to be less susceptible to dataset bias, the strong correlation between peripheral attributes of data samples and a target class. For example, even if the frog class in the dataset mainly consists of frog images with a swamp background (i.e., bias aligned samples), a debiased classifier should be able to correctly classify a frog at a beach (i.e., bias conflicting samples). Recent debiasing approaches commonly use two components for debiasing, a biased model fB and a debiased model fD. fB is trained to focus on bias aligned samples (i.e., overfitted to the bias) while fD is mainly trained with bias conflicting samples by concentrating on samples which fB fails to learn, leading fD to be less susceptible to the dataset bias. While the state of the art debiasing techniques have aimed to better train fD, we focus on training fB, an overlooked component until now. Our empirical analysis reveals that removing the bias conflicting samples from the training set for fB is important for improving the debiasing performance of fD. This is due to the fact that the bias conflicting samples work as noisy samples for amplifying the bias for fB since those samples do not include the bias attribute. To this end, we propose a simple yet effective data sample selection method which removes the bias conflicting samples to construct a bias amplified dataset for training fB. Our data sample selection method can be directly applied to existing reweighting based debiasing approaches, obtaining consistent performance boost and achieving the state of the art performance on both synthetic and real-world datasets.

PDF Details DOI

NeurIPS Conference 2021 Conference Paper

Learning Debiased Representation via Disentangled Feature Augmentation

Jungsoo Lee
Eungyeup Kim
Juyoung Lee
Jihyeon Lee
Jaegul Choo

Image classification models tend to make decisions based on peripheral attributes of data items that have strong correlation with a target variable (i. e. , dataset bias). These biased models suffer from the poor generalization capability when evaluated on unbiased datasets. Existing approaches for debiasing often identify and emphasize those samples with no such correlation (i. e. , bias-conflicting) without defining the bias type in advance. However, such bias-conflicting samples are significantly scarce in biased datasets, limiting the debiasing capability of these approaches. This paper first presents an empirical analysis revealing that training with "diverse" bias-conflicting samples beyond a given training set is crucial for debiasing as well as the generalization capability. Based on this observation, we propose a novel feature-level data augmentation technique in order to synthesize diverse bias-conflicting samples. To this end, our method learns the disentangled representation of (1) the intrinsic attributes (i. e. , those inherently defining a certain class) and (2) bias attributes (i. e. , peripheral attributes causing the bias), from a large number of bias-aligned samples, the bias attributes of which have strong correlation with the target variable. Using the disentangled representation, we synthesize bias-conflicting samples that contain the diverse intrinsic attributes of bias-aligned samples by swapping their latent features. By utilizing these diversified bias-conflicting features during the training, our approach achieves superior classification accuracy and debiasing results against the existing baselines on both synthetic and real-world datasets.

PDF Details

YNICL Journal 2017 Journal Article

Synergistic but independent: The role of corticospinal and alternate motor fibers for residual motor output after stroke

Robert Schulz
Eunhee Park
Jungsoo Lee
Won Hyuk Chang
Ahee Lee
Yun-Hee Kim
Friedhelm C. Hummel

BACKGROUND: Brain imaging has shown that not only the cortico-spinal tract (CST), but also alternate corticofugal motor fibers (aMF), such as the cortico-rubro-spinal and cortico-reticulo-spinal tract, influence residual motor output after stroke. So far, studies mainly have investigated each tract separately. A combined analysis of CST and aMF with assessment of their interactive role, i.e., that structural integrity of one tract influences the functional role of the structural integrity of the other, is pending. METHODS: 39 late subacute stroke patients (aged 59.4 ± 12.0 years, 100 ± 11 days after stroke) were included. Probabilistic tractography was used to reconstruct CST and aMF. Fractional anisotropy (FA) was calculated as a measure of microstructural integrity. Multiple-linear-regression analysis was used to associate tract-related FA with residual motor output and to determine interactions between CST and aMF. RESULTS: = 0.882). Thus, these data suggest that aMF and CST explain residual motor output in stroke patients in a synergistic, but mainly independent manner. CONCLUSIONS: The structural states of the CST and also - to a smaller degree - of the aMF correlate with residual motor output in late subacute stroke patients. Based on this statistical modeling with all inherent limitations, the novel finding of an absence of a significant interaction between both tracts in regard of their functional role, suggests that both corticofugal pathways act synergistically but largely independently. These findings add to the understanding of the functional role of different corticofugal motor fibers and their interactions for motor output after stroke.

Details DOI