Dong Seong Kim

Harnessing LLMs for Document-Guided Fuzzing of Python Libraries

Bin Duan, Tarek Mahmud, Meiru Che, Yan Yan et autres

Python libraries underpin deep learning, scientific computing, data analysis, and computer vision, making their reliability critical to downstream applications. Testing their APIs requires inputs that satisfy both per-parameter constraints and dependencies among parameters. Existing approaches either leave such constraints implicit in generated …

Testing Deep Learning Library APIs via Cross-Framework Differential Fuzzing

Bin Duan, Ruican Dong, Naipeng Dong, Dong Seong Kim et autres

Deep learning libraries underpin many safety- and reliability-critical applications, yet existing API-level testing techniques often rely on intra-library properties or CPU--GPU differential oracles and may miss defects that behave consistently across hardware backends. We present Xamt, a cross-framework differential fuzzing approach for …

Spatially robust and visually stealthy physical adversarial attacks on object detection systems

Zi Wang, Gang Min Kim, Hyunjae Kang, Huy Kang Kim et autres

The vulnerability of object detection systems to physical adversarial examples poses an important safety concern for autonomous driving systems. Prior physical attacks built on Expectation Over Transformation (EOT) demonstrate feasibility but often generate conspicuous high-frequency perturbations that are easily perceptible to human …

au, kr (code pays fourni par la source)

Continual learning for adaptive IoT network intrusion detection via domain-incremental learning methods

Oliver Delgado, Hyunjae Kang, Ulysses Lam, Jung Taek Seo et autres

The proliferation of Internet of Things (IoT) devices increases exposure to network intrusions and motivates the development of Network Intrusion Detection Systems (NIDS) that can adapt to newly emerging attacks. However, models retrained on new threats often suffer from catastrophic forgetting, resulting …

au, kr (code pays fourni par la source)

Towards Improving the Robustness of Deepfake Audio Detection With Denoising Methods

Yi-Xiang Wang, Dong Seong Kim

Exploring the enhancement of deepfake audio detection robustness through the integration of denoising methods into traditional machine learning pipelines. Motivated by the increasing misuse of deepfake audio in phishing and identity spoofing, the study focuses on improving detection accuracy under real-world noisy …

au (code pays fourni par la source)

Lightweight Preprocessing Defenses for Robust Deepfake Detection Against Adversarial Perturbations

A Zhang, Seonghoon Jeong, Hyunjae Kang, Dong Seong Kim

Deepfake detection models achieve high performance on benchmark datasets yet underperform against adversarial perturbations. This paper investigates lightweight, model-agnostic input preprocessing as a defense mechanism against such perturbations. We evaluate 17 preprocessing configurations—ranging from simple compression to multi-step transformations—against Statistical Consistency Attack …

au, kr (code pays fourni par la source)

APT-Agent: Automated Penetration Testing using Large Language Models

William Guanting Li, Alsharif Abuadbba, Kristen Moore, Dong Seong Kim

Penetration testing is essential to securing modern web infrastructures, yet traditional manual methods struggle to keep pace with their scale and complexity. Large Language Models (LLMs) offer new opportunities for automating these tasks, but existing approaches face two persistent challenges: hallucination of …

au (code pays fourni par la source)

A Black-Box Inversion Attack on Small Language Models and Differential Privacy-Based Defense

Xingyun Wang, Seonghoon Jeong, Hyunjae Kang, Dong Seong Kim

The proliferation of Small Language Models (SLMs) and distilled architectures has brought Large Language Model (LLM)-class capabilities to consumer-grade hardware. However, the privacy risks associated with these optimized models remain under-explored compared to their larger counterparts. In this work, we investigate the …

au, kr (code pays fourni par la source)

Feedback-Driven Fuzzing for COTS Robots via Packet Semantic Interpretation

Jihyeon Yu, Jinha Kim, Wooseong Jung, Dong Seong Kim et autres

kr, au (code pays fourni par la source)

Protecting federated learning from malicious attacks using consensus technique

Woocheol Kim, Jaehyoung Park, Jin-Hee Cho, Dong Seong Kim et autres

kr, au, us (code pays fourni par la source)

Robustness Evaluation Under RGB-Camera Attacks in CARLA: A Systematic Evaluation of Color Modes and Attack Types

Yu‐Feng Lin, Sangmin Park, Hyunjae Kang, Huy Kang Kim et autres

The robustness of YOLOv5-based camera perception for autonomous driving was systematically evaluated under diverse visual perturbations and spectral configurations using the CARLA simulation environment. An agent-camera framework decoupled perception from vehicle control, enabling consistent testing across 20 configurations and 200 trials (104,231 …

au, kr (code pays fourni par la source)

XAMT: Cross-Framework API Matching for Testing Deep Learning Libraries

Bin Duan, Naipeng Dong, Dong Seong Kim, Guowei Yang

Deep learning powers critical applications such as autonomous driving, healthcare, and finance, where the correctness of underlying libraries is essential. Bugs in widely used deep learning APIs can propagate to downstream systems, causing serious consequences. While existing fuzzing techniques detect bugs through …

au (code pays fourni par la source)