GIST Scholar
Not a member yet
30271 research outputs found
Sort by
Sample-Incremental Influence Function: A Sensitivity Metric to detect Key Feature Replacement Ji-sung Son Gwangju Institute of Science and Technology
No full text인플루언스 함수는 모델의 민감도를 정량적으로 측정하는 지표로, 파라미터 변화의 정도를 측정하고, 모델 파라미터의 동작에 대한 통찰을 제공하며, 재학습 없이도 파라미 터 예측을 가능하게 한다. 전통적인 인플루언스 함수 기반 민감도는 샘플 제거 방식을 기반으로 하여 특정 샘플을 제거했을 때의 영향을 측정하는 데에만 초점이 맞춰져 있 다. 이러한 제한된 접근은 새로운 샘플이 추가될 때 모델의 동작을 분석하는 데 한계가 있으며, 이는 동적 데이터 환경에 적응하는 데 있어 중요한 제약으로 작용한다. 본 연 구에서는 추가 데이터를 기반으로 모델 민감도를 평가할 수 있도록 인플루언스 함수를 확장한 Sample-Incremental Influence Function을 제안한다. 상관 분석, CAM 기반 시각 화를 통해 본 방법의 유효성을 입증하였으며, 주요 특징 교체를 탐지할 수 있는 능력을 확인하였다. 제안된 민감도 지표는 미지의 샘플에 해당하는 특징과 관련된 다양한 응용 분야에서 강력한 잠재력을 보여준다. ©2025 손 지 성 ALL RIGHTS RESERVED|Influence function serves as a quantitative metric of model sensitivity, measuring the degree of parameter variation, offering insights into model parameter behavior, and enabling parameter predictions without the need for retraining. The traditional influence function-based sensitivity relies on a sample-decremental approach, which only measures the impact of removing specific samples. This narrow focus restricts its ability to analyze model behavior in response to newly introduced samples, highlighting a significant limitation in adapting to dynamic data environments. In this work, we present the Sample-Incremental Influence Function that leverages influence function to evaluate the impact of additional data on model sensitivity. Correlation analysis and CAM-based visualization demonstrate the validity of our method and reveal its ability to detect key feature replacement. Sensitivity show strong potential for various applications related to feature corresponding to unknown sample. ©2025 Ji-sung Son ALL RIGHTS RESERVEDMasterAbstract (English) i
Abstract (Korean) ii
List of Contents iii
List of Tables v
List of Figures vi
List of Algorithms ix
1 Introduction 1
2 Background 5
2.1 Influence Function 5
2.2 Gradient-weighted Class Activation Mapping (Grad-CAM) and Key fea-
ture 7
2.3 Out of Distribution detection 9
3 Method 11
3.1 Motivation 11
3.2 Proposed Method 13
3.2.1 Sample-incremental Influence function 13
4 Experiments 15
4.1 Validation of the proposed method 15
4.2 Difference between Confidence and Sensitivity 18
4.3 Characteristics of highly sensitive data 21
4.4 Out of Distribution Detection 23
5 Conclusion and Future work 27
5.1 Conclusion 27
5.2 Future work 28
Summary 29
– iii –
References 30
A Abbreviations 33
B Proof 34
B.1 Derivation of Inverse Proportionality 34
B.2 Derivation of minimizer form 35
B.3 Derivation of the Leave-One-Out (LOO) Deviation and the Include-One-
In (IOI) Deviation 37
C Application to various dataset 40
Acknowledgements 42
– iv
Enhancing anti-tumor immunity and immunotherapy efficacy through the downregulation of Acetyl-CoA carboxylase alpha in cancer cells
No full textCancer immunotherapy has revolutionized treatment for many cancers by harnessing the immune system’s ability to recognize and eliminate tumor cells. Among these therapies, immune checkpoint inhibitors (ICIs) have shown significant promise in clinical trials. However, the development of resistance to ICIs remains a major obstacle. In this context, the role of Acetyl-CoA carboxylase alpha (ACACA), an enzyme involved in fatty acid metabolism, in cancer progression and immune modulation remains underexplored. To investigate its impact, we utilized CRISPR-Cas9 to generate ACACA-knockdown (ACACA KD) cancer cells and examined their effects on tumor growth and immune responses. In vitro, ACACA KD cells displayed reduced proliferation and enhanced susceptibility to immune-mediated cytotoxicity. In vivo, ACACA KD tumors grew more slowly, showed greater infiltration of effector T cells and NK cells, and upregulated immune-related pathways. Moreover, combining anti-PD-1 therapy with ACACA knockdown resulted in improved antitumor efficacy. Our study highlights ACACA as a potential target for enhancing the effectiveness of cancer immunotherapy.MasterAbstract i
Contents ii
I. Introduction 1
II. Materials and methods 2
1. Cell lines and culture conditions 2
2. Knockdown of ACACA with CRISPR-Cas9 2
3. RNA isolation and real-time PCR 3
4. Cell proliferation assay 4
5. Cancer cell apoptosis and cytotoxicity assay 4
6. Animal model 4
7. Immune population profiling 5
8. Mouse transcriptome analysis (Mouse RNA-seq) 5
9. Human bulk transcriptome analysis (human RNA-seq) 6
10. Statistical analysis 6
III. Results 7
1. ACACA downregulation inhibits tumor growth and enhances immune response. 7
2. ACACA knockdown enhances cytotoxic effector T cell function. 12
3. ACACA downregulation inhibits tumor growth and enhances immune response. 20
IV. Discussion 24
V. Summary 26
1. Impact of ACACA expression on TME immunity 26
VI. References 27
VII. Acknowledgements 30
VIII. Curriculum Vitae 3
Effect of Tryptophan in Enhancing CH4 Enrichment from Hydrogen-Natural Gas Blends via Semi-unstirred Hydrate Formation
No full textPhotobiomodulation Promotes Early Recovery of Olfactory Function and Modulates Neuroprotective Gene Expression in a Mouse Model of Ischemic Stroke
No full textIschemic stroke often leads to neurological deficits, including olfactory dysfunction, which can significantly diminish quality of life. Photobiomodulation (PBM) has emerged as a promising therapeutic strategy for enhancing post-stroke recovery, although the molecular mechanisms, particularly regarding gene expression change, are not yet fully understood. This study investigates the long-term effects of photothrombosis (PT) on olfactory function and the olfactory bulb (OB) microenvironment, with a focus on PBM's efficacy during both early and late phases. In a mouse OB PT stroke model, PBM therapy (808-nm laser, 40 J/cm2 fluence, 325 mW/cm2, 2 min daily) was applied from day 2 to day 7 post-PT. Olfactory function was monitored from pre-stroke through day 28 using the buried food test (BFT), and MRI scans were performed on days 7 and 28 to assess tissue damage. RNA sequencing (RNA-seq) and reverse transcription quantitative PCR (RT-qPCR) were conducted on day 7 to evaluate gene expression changes, with additional RT-qPCR analyses performed on day 28. PBM significantly accelerated olfactory function recovery by day 14, with full recovery maintained through day 28. Despite functional recovery, MRI results indicated persistent infarction at 28 days. RNA-seq identified upregulation of neuroprotective genes, including Gpr39 and Or4m1, following PBM treatment, suggesting enhanced gene expression related to acute-phase recovery. However, the impact of PBM on gene expression and functional recovery appeared to wane in the later stages of recovery. These findings underscore PBM's potential to enhance early-stage recovery in ischemic stroke, though its benefits may be more limited in the chronic phase.FALSEsciescopu
Spatiotemporal Anomaly Detection in Distributed Acoustic Sensing Using a GraphDiffusion Model
No full textDistributed acoustic sensing (DAS), which can provide dense spatial and temporal measurements using optical fibers, is quickly becoming critical for large-scale infrastructure monitoring. However, anomaly detection in DAS data is still challenging owing to the spatial correlations between sensing channels and nonlinear temporal dynamics. Recent approaches often disregard the explicit sensor layout and instead handle DAS data as two-dimensional images or flattened sequences, eliminating the spatial topology. This work proposes GraphDiffusion, a novel generative anomaly-detection model that combines a conditional denoising diffusion probabilistic model (DDPM) and a graph neural network (GNN) to overcome these limitations. By treating each channel as a graph node and building edges based on Euclidean proximity, the GNN explicitly models the spatial arrangement of DAS sensors, allowing the network to capture local interchannel dependencies. The conditional DDPM uses iterative denoising to model the temporal dynamics of standard signals, enabling the system to detect deviations without the need for anomalies. The performance evaluations based on real-world DAS datasets reveal that GraphDiffusion achieves 98.2% and 98.0% based on the area under the curve (AUC) of the F1-score at K different levels (F1K-AUC), an AUC of receiver operating characteristic (ROC) at K different levels (ROCK-AUC), outperforming other comparative models. © 2025 Elsevier B.V., All rights reserved.TRUEsciescopu
Vascular and glymphatic dysfunction as drivers of cognitive impairment in Alzheimer's disease: Insights from computational approaches
No full textAlzheimer's disease (AD) is driven by complex interactions between vascular dysfunction, glymphatic system impairment, and neuroinflammation. Vascular aging, characterized by arterial stiffness and reduced cerebral blood flow (CBF), disrupts the pulsatile forces necessary for glymphatic clearance, exacerbating amyloid-beta (Aβ) accumulation and cognitive decline. This review synthesizes insights into the mechanistic crosstalk between these systems and explores their contributions to AD pathogenesis. Emerging machine learning (ML) tools, such as DeepLabCut and Motion sequencing (MoSeq), offer innovative solutions for analyzing multimodal data and enhancing diagnostic precision. Integrating ML with imaging and behavioral analyses bridges gaps in understanding vascular-glymphatic dysfunction. Future research must prioritize these interactions to develop early diagnostics and targeted interventions, advancing our understanding of neurovascular health in AD. © 2025TRUEsciescopu
Mitigating Instability in High Residual Adaptive Sampling for PINNs via Langevin Dynamics
No full textRecently, physics-informed neural networks (PINNs) have gained attention in the scientific community for their potential to solve partial differential equations (PDEs). However, they face challenges related to resource efficiency and slow convergence. Adaptive sampling methods, which prioritize collocation points with high residuals, improve both efficiency and accuracy. However, these methods often neglect points with medium or low residuals, which can affect stability as the complexity of the model increases. In this paper, we investigate this limitation and show that high residual-based approaches require stricter learning rate bounds to ensure stability. To address this, we propose a Langevin dynamics-based Adaptive Sampling (LAS) framework that is robust to various learning rates and model complexities. Our experiments demonstrate that the proposed method outperforms existing approaches in terms of relative
error, and stability across a range of environments, including high-dimensional PDEs where Monte Carlo integration-based methods typically suffer from instability
Occurrence and Spatial Characterization of PFAS in River Sediments Using an Optimized Analytical Method
No full textAdvancing Image Captioning with Regional Attention in Vision-Language Transformers and Multimodal Learning Zubia Naz School of Electrical Engineering and Computer Science Gwangju Institute of Science and Technology
No full text비주얼대형언어모델(Visual Large Language Models, VLLMs)은복잡한의료이미 지를해석하고정확하며맥락에맞는텍스트설명을생성하는문제를해결함으로써의료 이미지 캡셔닝을 혁신적으로 변화시키고 있습니다. 본 연구에서는 Swin Transformer와 BART의 성능을 강화하기 위해 의료 이미지의 특정 영역에 동적으로 집중할 수 있는 메커니즘인 **지역 주의(regional attention)**를 도입했습니다. 이 혁신은 캡션 생성 과정에서 중요한 세부사항을 강조하도록 보장하며, 의료 이미지의 복잡하고 국지적인 특징을 포착하는 데 종종 어려움을 겪는 기존 모델의 한계를 극복합니다. 지역 주의를 활용하여 본 프레임워크는 의료적으로 중요한 세부사항을 식별하는 데 있어 더 높은 정밀도를 달성하여, 임상 의사결정을 개선하고 방사선 전문의가 의료 스캔에서 상태를 진단하는 데 있어 더 신뢰할 수 있고 상세한 캡션을 생성할 수 있도록 지원합니다. ROCO데이터셋을기반으로평가된본모델은 ROUGE및 BERTScore지표에서유 의미한개선을보여주었습니다. ROUGE는생성된캡션과참조텍스트간의주요단어의 겹침을 측정하여 언어적 정확성을 반영하며, BERTScore는 의미적 유사성을 평가하여 생성된 캡션의 맥락적 관련성과 임상적 유용성을 나타냅니다. 지역 주의의 통합은 세밀 한시각적세부사항과맥락적뉘앙스를효과적으로포착하도록모델을지원하여이러한 – iii – 발전에 직접 기여합니다. BLEU와 CIDEr와 같은 다른 지표들은 일관성을 유지하여 접 근 방식의 강건함을 확인시켰습니다. 지역 주의에 의해 주도된 의료 이미지 캡셔닝의 이 진보는 자동화된 의료 진단을 개선하고 방사선 전문의를 보다 정밀하고 해석 가능한 방식으로 지원하는 데 있어 상당한 이점을 제공합니다. ©2025 주 비 아 나즈 ALL RIGHTS RESERVED|Visual Large Language Models (VLLMs) are transforming medical image captioning by addressing the challenges of interpreting complex medical images and generating accurate, context-aware textual descriptions. In this work, we enhance the capabilities of Swin Transformer and BART by incorporating regional attention, a mechanism that dynamically focuses on specific regions of medical images. This innovation ensures that critical details are emphasized during caption generation, overcoming limitations of traditional models that often struggle to capture intricate and localized features of medical images. By leveraging regional attention, our framework achieves a higher precision in identifying medically significant details, thereby enhancing clinical decision-making and generating more reliable, detailed captions to support radiologists in diagnosing conditions from medical scans. Evaluated on the ROCO dataset, our model demonstrates significant improvements in ROUGE and BERTScore metrics. ROUGE measures the overlap of essential words between generated captions and reference texts, reflecting linguistic accuracy, while BERTScore evaluates semantic similarity, indicating the con- textual relevance and clinical utility of the generated captions. The integration of regional attention directly contributes to these advancements by enabling the model to effectively capture fine-grained visual details and contextual nuances. Other metrics, such as BLEU and CIDEr, remained consistent, confirming the robustness of the
approach. This advancement in medical image captioning, driven by regional attention, offers substantial benefits for improving automated medical diagnostics and supporting radiologists with enhanced precision and interpretability. ©2025 Zubia Naz ALL RIGHTS RESERVEDMasterAbstract (English) i
Abstract (Korean) iii
List of Contents v
List of Tables vii
List of Figures viii
List of Algorithms ix
1 Introduction 1
1.1 Introduction 1
1.2 Motivation 2
2 Preliminary 4
2.1 Visual Language Models(VLMs) 4
2.1.1 Image Features Representation 4
2.1.2 Natural Language Representation 6
2.2 Major Architectures 7
2.2.1 CLIP Architecture in Image Captioning 8
2.2.2 BLIP Architecture in Image Captioning 8
2.3 Limitations of Unified Approach 9
3 Baseline Method 10
3.1 Overview of GIT Architecture 10
3.1.1 Image Encoder 10
3.1.2 Text Decoder 11
3.1.3 Vision-Language Interaction Mechanism 11
3.1.4 Training and Fine Tuning 12
3.2 Similar Approaches Following GIT Architecture 12
3.2.1 CMRE-UoG team at ImageCLEFmedical Caption 2022: Concept
Detection and Image Captioning 12
3.2.2 Transferring Pre-Trained Large Language-Image Model for Med-
ical Image Captioning 14
– v –
4 Proposed Methodology 16
4.1 Architectural Overview 16
4.1.1 Image Encoder: Swin Transformer with Regional Attention 17
4.1.2 Text Decoder: BART-Based Model with Biomedical Embeddings
and Regional Features 20
4.1.3 Visual Interaction and Training Strategy 22
4.1.4 Dataset and Data Processing 22
4.1.5 Advantages of Model Choices 23
4.1.6 Limitations and Challenges 23
4.1.7 Practical Implications and Future Directions 24
5 Results and Conclusions 25
5.1 Overview of Evaluation Metrics 25
5.2 Results and Performance Analysis 26
5.2.1 Comparative Analysis 27
5.2.2 Qualitative Analysis 28
5.3 Limitations and Future Directions 29
5.4 Conclusion 30
Summary 32
References 34
A Abbreviations 38
Acknowledgements 39
– vi