arXiv subjects

Jonghyun Kim

Publications and source records attributed to Jonghyun Kim.

At least 19 recordsLinked to original sources

arXiv2026

MINT: Molecularly Informed Training with Spatial Transcriptomics Supervision for Pathology Foundation Models

Pathology foundation models learn morphological representations through self-supervised pretraining on large-scale whole-slide images, yet they do not explicitly capture the underlying molecular state of the tissue. Spatial transcriptomics technologies bridge this gap by measuring gene expression in situ, offering a natural cross-modal supervisory signal. We propose MINT (Molecularly Informed Training), a fine-tuning framework that incorporates spatial transcriptomics supervision into pretrained pathology Vision Transformers. MINT appends a learnable ST token to the ViT input to encode transcriptomic information separately from the morphological CLS token, preventing catastrophic forgetting through DINO self-distillation and explicit feature anchoring to the frozen pretrained encoder. Gene expression regression at both spot-level (Visium) and patch-level (Xenium) resolutions provides complementary supervision across spatial scales. Trained on 577 publicly available HEST samples, MINT achieves the best overall performance on both HEST-Bench for gene expression prediction (mean Pearson r = 0.440) and EVA for general pathology tasks (0.803), demonstrating that spatial transcriptomics supervision complements morphology-centric self-supervised pretraining.

cs.CV↗

arXiv2025

EXAONE Path 2.5: Pathology Foundation Model with Multi-Omics Alignment

Cancer progression arises from interactions across multiple biological layers, especially beyond morphological and across molecular layers that remain invisible to image-only models. To capture this broader biological landscape, we present EXAONE Path 2.5, a pathology foundation model that jointly models histologic, genomic, epigenetic and transcriptomic modalities, producing an integrated patient representation that reflects tumor biology more comprehensively. Our approach incorporates three key components: (1) multimodal SigLIP loss enabling all-pairwise contrastive learning across heterogeneous modalities, (2) a fragment-aware rotary positional encoding (F-RoPE) module that preserves spatial structure and tissue-fragment topology in WSI, and (3) domain-specialized internal foundation models for both WSI and RNA-seq to provide biologically grounded embeddings for robust multimodal alignment. We evaluate EXAONE Path 2.5 against six leading pathology foundation models across two complementary benchmarks: an internal real-world clinical dataset and the Patho-Bench benchmark covering 80 tasks. Our framework demonstrates high data and parameter efficiency, achieving on-par performance with state-of-the-art foundation models on Patho-Bench while exhibiting the highest adaptability in the internal clinical setting. These results highlight the value of biologically informed multimodal design and underscore the potential of integrated genotype-to-phenotype modeling for next-generation precision oncology.

cs.LG↗

arXiv2025

Real-time 3D Visualization of Radiance Fields on Light Field Displays

Radiance fields, including their recent efficient forms such as 3D Gaussian Splatting and Sparse Voxels, have revolutionized photorealistic 3D scene visualization by enabling high-fidelity reconstruction of complex environments, making them a natural match for light field displays. However, integrating these technologies presents significant computational challenges, as light field displays require many high-resolution renderings from slightly shifted viewpoints, while radiance fields rely on computationally intensive volume rendering, which is intractable to achieve real-time speeds even with efficient scene representations. In this paper, we propose a unified and efficient framework for real-time radiance field rendering on light field displays. Rather than re-rendering each view independently, our method converts the input radiance field into shared intermediate sweeping planes that can be efficiently composited into dense light-field views in a single pass. Our method prioritizes shared, non-directional plane caching for real-time performance, trading fine view-dependent color effects for a modest increase in intermediate memory usage. Our framework generalizes across different scene representations without retraining and avoids repeated computation across views. We further demonstrate a real-time interactive application on a Looking Glass display, achieving 200+ FPS at 512p across 45 rendered views and enabling seamless, immersive 3D interactive viewing experiences. On standard benchmarks, our method achieves up to 22x speedup compared to independently rendering each view, while largely preserving image quality.

cs.GR↗

arXiv2025

EXAONE Path 2.0: Pathology Foundation Model with End-to-End Supervision

In digital pathology, whole-slide images (WSIs) are often difficult to handle due to their gigapixel scale, so most approaches train patch encoders via self-supervised learning (SSL) and then aggregate the patch-level embeddings via multiple instance learning (MIL) or slide encoders for downstream tasks. However, patch-level SSL may overlook complex domain-specific features that are essential for biomarker prediction, such as mutation status and molecular characteristics, as SSL methods rely only on basic augmentations selected for natural image domains on small patch-level area. Moreover, SSL methods remain less data efficient than fully supervised approaches, requiring extensive computational resources and datasets to achieve competitive performance. To address these limitations, we present EXAONE Path 2.0, a pathology foundation model that learns patch-level representations under direct slide-level supervision. Using only 37k WSIs for training, EXAONE Path 2.0 achieves state-of-the-art average performance across 10 biomarker prediction tasks, demonstrating remarkable data efficiency.

cs.CV↗

arXiv2025

Effect of Electrode Array Position on Electric Field Intensity in Glioblastoma Patients Undergoing Electric Field Therapy

Background: The intensity of the electric field applied to a brain tumor by electric field therapy is influenced by the position of the electrode array, which should be optimized based on the patient's head shape and tumor characteristics. This study assessed the effects of varying electrode positions on electric field intensity in glioblastoma multiforme (GBM) patients. Methods: This study enrolled 13 GBM patients. The center of the MR slice corresponding to the center of the tumor was set as the reference point for the electrodes, creating pairs of electrode arrays in the top-rear and left-right positions. Based on this reference plan, four additional treatment plans were generated by rotating three of the four electrode arrays, all except the top electrode array, by 15$^\circ$ and 30$^\circ$ from their reference positions, resulting in a total of five treatment plans per patient. Electric field frequency was set at 200 kHz, and current density at 31 mArms/cm$^2$. The minimum and mean electric field intensities, homogeneity index (HI), and coverage index (CovI) were calculated and compared. Results: The optimal plans showed differences ranging from-0.39% to 24.20% for minimum intensity and -14.29% to 16.67% for mean intensity compared to reference plans. HI and CovI varied from 0.00% to 48.65% and 0.00% to 95.3%, respectively. The average improvements across all patients were 8.96% for minimum intensity, 5.11% for mean intensity, 15.65% for HI, and 17.84% for CovI. Conclusions: Optimizing electrode angle improves electric field therapy outcomes in GBM patients by maximizing field intensity and coverage. Keywords: electric field therapy; glioblastoma multiforme (GBM); treatment planning system (TPS); electrode array position; tumor coverage

physics.med-ph↗

arXiv2024

Large Étendue 3D Holographic Display with Content-adaptive Dynamic Fourier Modulation

Emerging holographic display technology offers unique capabilities for next-generation virtual reality systems. Current holographic near-eye displays, however, only support a small étendue, which results in a direct tradeoff between achievable field of view and eyebox size. Étendue expansion has recently been explored, but existing approaches are either fundamentally limited in the image quality that can be achieved or they require extremely high-speed spatial light modulators. We describe a new étendue expansion approach that combines multiple coherent sources with content-adaptive amplitude modulation of the hologram spectrum in the Fourier plane. To generate time-multiplexed phase and amplitude patterns for our spatial light modulators, we devise a pupil-aware gradient-descent-based computer-generated holography algorithm that is supervised by a large-baseline target light field. Compared with relevant baseline approaches, our method demonstrates significant improvements in image quality and étendue in simulation and with an experimental holographic display prototype.

cs.GR↗

arXiv2024

Redefining Urban Centrality: Integrating Economic Complexity Indices into Central Place Theory

This study introduces a metric designed to measure urban structures through the economic complexity lens, building on the foundational theories of urban spatial structure, the Central Place Theory (CPT) (Christaller, 1933). Despite the significant contribution in the field of urban studies and geography, CPT has limited in suggesting an index that captures its key ideas. By analyzing various urban big data of Seoul, we demonstrate that PCI and ECI effectively identify the key ideas of CPT, capturing the spatial structure of a city that associated with the distribution of economic activities, infrastructure, and market orientation in line with the CPT. These metrics for urban centrality offer a modern approach to understanding the Central Place Theory and tool for urban planning and regional economic strategies without privacy issues.

econ.GN↗

arXiv2024

Population Concentration in High-Complexity Regions within City during the heat wave

This study investigates the impact of the 2018 summer heat wave on urban mobility in Seoul and the role of economic complexity in the region's resilience. Findings from subway and mobile phone data indicate a significant decrease in the floating population during extreme heat wave, underscoring the thermal vulnerability of urban areas. However, urban regions with higher complexity demonstrate resilience, attracting more visitors despite high temperatures. Our results suggest the centrality of economic complexity in urban resilience against climate-induced stressors. Additionally, it implies that high-complexity small businesses' clusters can serve as focal points for sustaining urban vitality in the face of thermal shocks within city. In the long run perspective, our results imply the possibility that people are more concentrated in high complexity region in the era of global warming.

econ.GN↗

arXiv2024

HyperCLOVA X Technical Report

We introduce HyperCLOVA X, a family of large language models (LLMs) tailored to the Korean language and culture, along with competitive capabilities in English, math, and coding. HyperCLOVA X was trained on a balanced mix of Korean, English, and code data, followed by instruction-tuning with high-quality human-annotated datasets while abiding by strict safety guidelines reflecting our commitment to responsible AI. The model is evaluated across various benchmarks, including comprehensive reasoning, knowledge, commonsense, factuality, coding, math, chatting, instruction-following, and harmlessness, in both Korean and English. HyperCLOVA X exhibits strong reasoning capabilities in Korean backed by a deep understanding of the language and cultural nuances. Further analysis of the inherent bilingual nature and its extension to multilingualism highlights the model's cross-lingual proficiency and strong generalization ability to untargeted languages, including machine translation between several language pairs and cross-lingual inference tasks. We believe that HyperCLOVA X can provide helpful guidance for regions or countries in developing their sovereign LLMs.

cs.CL↗

arXiv2023

Mesh Represented Recycle Learning for 3D Hand Pose and Mesh Estimation

In general, hand pose estimation aims to improve the robustness of model performance in the real-world scenes. However, it is difficult to enhance the robustness since existing datasets are obtained in restricted environments to annotate 3D information. Although neural networks quantitatively achieve a high estimation accuracy, unsatisfied results can be observed in visual quality. This discrepancy between quantitative results and their visual qualities remains an open issue in the hand pose representation. To this end, we propose a mesh represented recycle learning strategy for 3D hand pose and mesh estimation which reinforces synthesized hand mesh representation in a training phase. To be specific, a hand pose and mesh estimation model first predicts parametric 3D hand annotations (i.e., 3D keypoint positions and vertices for hand mesh) with real-world hand images in the training phase. Second, synthetic hand images are generated with self-estimated hand mesh representations. After that, the synthetic hand images are fed into the same model again. Thus, the proposed learning strategy simultaneously improves quantitative results and visual qualities by reinforcing synthetic mesh representation. To encourage consistency between original model output and its recycled one, we propose self-correlation loss which maximizes the accuracy and reliability of our learning strategy. Consequently, the model effectively conducts self-refinement on hand pose estimation by learning mesh representation from its own output. To demonstrate the effectiveness of our learning strategy, we provide extensive experiments on FreiHAND dataset. Notably, our learning strategy improves the performance on hand pose and mesh estimation without any extra computational burden during the inference.

cs.CV↗

arXiv2023

Hybrid model for Single-Stage Multi-Person Pose Estimation

In general, human pose estimation methods are categorized into two approaches according to their architectures: regression (i.e., heatmap-free) and heatmap-based methods. The former one directly estimates precise coordinates of each keypoint using convolutional and fully-connected layers. Although this approach is able to detect overlapped and dense keypoints, unexpected results can be obtained by non-existent keypoints in a scene. On the other hand, the latter one is able to filter the non-existent ones out by utilizing predicted heatmaps for each keypoint. Nevertheless, it suffers from quantization error when obtaining the keypoint coordinates from its heatmaps. In addition, unlike the regression one, it is difficult to distinguish densely placed keypoints in an image. To this end, we propose a hybrid model for single-stage multi-person pose estimation, named HybridPose, which mutually overcomes each drawback of both approaches by maximizing their strengths. Furthermore, we introduce self-correlation loss to inject spatial dependencies between keypoint coordinates and their visibility. Therefore, HybridPose is capable of not only detecting densely placed keypoints, but also filtering the non-existent keypoints in an image. Experimental results demonstrate that proposed HybridPose exhibits the keypoints visibility without performance degradation in terms of the pose estimation accuracy.

cs.CV↗

arXiv2023

Referenceless User Controllable Semantic Image Synthesis

Despite recent progress in semantic image synthesis, complete control over image style remains a challenging problem. Existing methods require reference images to feed style information into semantic layouts, which indicates that the style is constrained by the given image. In this paper, we propose a model named RUCGAN for user controllable semantic image synthesis, which utilizes a singular color to represent the style of a specific semantic region. The proposed network achieves reference-free semantic image synthesis by injecting color as user-desired styles into each semantic layout, and is able to synthesize semantic images with unusual colors. Extensive experimental results on various challenging datasets show that the proposed method outperforms existing methods, and we further provide an interactive UI to demonstrate the advantage of our approach for style controllability.

cs.CV↗

arXiv2022

5G-NIDD: A Comprehensive Network Intrusion Detection Dataset Generated over 5G Wireless Network

With a plethora of new connections, features, and services introduced, the 5th generation (5G) wireless technology reflects the development of mobile communication networks and is here to stay for the next decade. The multitude of services and technologies that 5G incorporates have made modern communication networks very complex and sophisticated in nature. This complexity along with the incorporation of Machine Learning (ML) and Artificial Intelligence (AI) provides the opportunity for the attackers to launch intelligent attacks against the network and network devices. These attacks often traverse undetected due to the lack of intelligent security mechanisms to counter these threats. Therefore, the implementation of real-time, proactive, and self-adaptive security mechanisms throughout the network would be an integral part of 5G as well as future communication systems. Therefore, large amounts of data collected from real networks will play an important role in the training of AI/ML models to identify and detect malicious content in network traffic. This work presents 5G-NIDD, a fully labeled dataset built on a functional 5G test network that can be used by those who develop and test AI/ML solutions. The work further analyses the collected data using common ML models and shows the achieved accuracy levels.

cs.CR↗

arXiv2022

BigColor: Colorization using a Generative Color Prior for Natural Images

For realistic and vivid colorization, generative priors have recently been exploited. However, such generative priors often fail for in-the-wild complex images due to their limited representation space. In this paper, we propose BigColor, a novel colorization approach that provides vivid colorization for diverse in-the-wild images with complex structures. While previous generative priors are trained to synthesize both image structures and colors, we learn a generative color prior to focus on color synthesis given the spatial structure of an image. In this way, we reduce the burden of synthesizing image structures from the generative prior and expand its representation space to cover diverse images. To this end, we propose a BigGAN-inspired encoder-generator network that uses a spatial feature map instead of a spatially-flattened BigGAN latent code, resulting in an enlarged representation space. Our method enables robust colorization for diverse inputs in a single forward pass, supports arbitrary input resolutions, and provides multi-modal colorization results. We demonstrate that BigColor significantly outperforms existing methods especially on in-the-wild images with complex structures.

cs.CV↗

arXiv2022

Time-multiplexed Neural Holography: A flexible framework for holographic near-eye displays with fast heavily-quantized spatial light modulators

Holographic near-eye displays offer unprecedented capabilities for virtual and augmented reality systems, including perceptually important focus cues. Although artificial intelligence--driven algorithms for computer-generated holography (CGH) have recently made much progress in improving the image quality and synthesis efficiency of holograms, these algorithms are not directly applicable to emerging phase-only spatial light modulators (SLM) that are extremely fast but offer phase control with very limited precision. The speed of these SLMs offers time multiplexing capabilities, essentially enabling partially-coherent holographic display modes. Here we report advances in camera-calibrated wave propagation models for these types of holographic near-eye displays and we develop a CGH framework that robustly optimizes the heavily quantized phase patterns of fast SLMs. Our framework is flexible in supporting runtime supervision with different types of content, including 2D and 2.5D RGBD images, 3D focal stacks, and 4D light fields. Using our framework, we demonstrate state-of-the-art results for all of these scenarios in simulation and experiment.

cs.GR↗

arXiv2021

SuperStyleNet: Deep Image Synthesis with Superpixel Based Style Encoder

Existing methods for image synthesis utilized a style encoder based on stacks of convolutions and pooling layers to generate style codes from input images. However, the encoded vectors do not necessarily contain local information of the corresponding images since small-scale objects are tended to "wash away" through such downscaling procedures. In this paper, we propose deep image synthesis with superpixel based style encoder, named as SuperStyleNet. First, we directly extract the style codes from the original image based on superpixels to consider local objects. Second, we recover spatial relationships in vectorized style codes based on graphical analysis. Thus, the proposed network achieves high-quality image synthesis by mapping the style codes into semantic labels. Experimental results show that the proposed method outperforms state-of-the-art ones in terms of visual quality and quantitative measurements. Furthermore, we achieve elaborate spatial style editing by adjusting style codes.

cs.CV↗

arXiv2021

Adaptive Prototype Learning and Allocation for Few-Shot Segmentation

Prototype learning is extensively used for few-shot segmentation. Typically, a single prototype is obtained from the support feature by averaging the global object information. However, using one prototype to represent all the information may lead to ambiguities. In this paper, we propose two novel modules, named superpixel-guided clustering (SGC) and guided prototype allocation (GPA), for multiple prototype extraction and allocation. Specifically, SGC is a parameter-free and training-free approach, which extracts more representative prototypes by aggregating similar feature vectors, while GPA is able to select matched prototypes to provide more accurate guidance. By integrating the SGC and GPA together, we propose the Adaptive Superpixel-guided Network (ASGNet), which is a lightweight model and adapts to object scale and shape variation. In addition, our network can easily generalize to k-shot segmentation with substantial improvement and no additional computational cost. In particular, our evaluations on COCO demonstrate that ASGNet surpasses the state-of-the-art method by 5% in 5-shot segmentation.

cs.CV↗

arXiv2021

Edge and Identity Preserving Network for Face Super-Resolution

Face super-resolution (SR) has become an indispensable function in security solutions such as video surveillance and identification system, but the distortion in facial components is a great challenge in it. Most state-of-the-art methods have utilized facial priors with deep neural networks. These methods require extra labels, longer training time, and larger computation memory. In this paper, we propose a novel Edge and Identity Preserving Network for Face SR Network, named as EIPNet, to minimize the distortion by utilizing a lightweight edge block and identity information. We present an edge block to extract perceptual edge information, and concatenate it to the original feature maps in multiple scales. This structure progressively provides edge information in reconstruction to aggregate local and global structural information. Moreover, we define an identity loss function to preserve identification of SR images. The identity loss function compares feature distributions between SR images and their ground truth to recover identities in SR images. In addition, we provide a luminance-chrominance error (LCE) to separately infer brightness and color information in SR images. The LCE method not only reduces the dependency of color information by dividing brightness and color components but also enables our network to reflect differences between SR images and their ground truth in two color spaces of RGB and YUV. The proposed method facilitates the proposed SR network to elaborately restore facial components and generate high quality 8x scaled SR images with a lightweight network structure. Furthermore, our network is able to reconstruct an 128x128 SR image with 215 fps on a GTX 1080Ti GPU. Extensive experiments demonstrate that our network qualitatively and quantitatively outperforms state-of-the-art methods on two challenging datasets: CelebA and VGGFace2.

cs.CV↗