Vol. 52, No. 4, Apr. 2025
Digital Library


u-INJECTOR: Unicorn-based Firmware Dynamic Analysis Tool

Youngbeen Yoo, Hanbit Kim, Junghyung Park, Jinsung Cho

http://doi.org/10.5626/JOK.2025.52.4.269

As proliferation of IoT devices fuels growth of the embedded market and expands application areas of embedded devices, the risk of vulnerability exploitation is increasing, particularly as many IoT device manufacturers does not allow security guidelines. This has led to an increased risk of exploitation through vulnerability attacks. Consequently, it's crucial to analyze and address vulnerabilities in embedded firmware beforehand to ensure system safety. While dynamic analysis techniques are crucial for assessing such vulnerabilities, unique development environments of embedded devices, as opposed to host PCs, along with the diversity in hardware architectures, pose challenges in setting up a firmware analysis environment using virtualization tools like QEMU alone. To overcome these challenges, this study introduced an embedded firmware vulnerability analysis tool, u-INJECTOR, utilizing the open-source CPU virtualization tool, Unicorn. u-INJECTOR can significantly reduce environment construction costs compared to QEMU by automatically analyzing symbols of executable files and establishing a virtualization environment for embedded firmware. The u-INJECTOR described in this research is expected to serve as a valuable tool for detecting vulnerabilities against side-channel and fault injection attacks.

Development of a Rover Swarm Autonomous Driving System with Lateral Control Based on the L1 Controller

Gyeonghun Lim, Sungtae Moon

http://doi.org/10.5626/JOK.2025.52.4.276

With the increasing use of unmanned vehicles in various fields, services such as logistics, surveillance, and reconnaissance are being actively provided. In particular, swarm driving, which involves multiple unmanned vehicles, is gaining attention due to its advantages over single-vehicle operation, such as reduced task time, expanded operational range, and improved system reliability. However, previous studies have often neglected lateral control, resulting in reduced precision in swarm driving, and due to the complexity and high cost of development, most have been conducted in simulation environments. This paper implements precise and stable swarm control by utilizing the lightweight and easy-to-implement L1 controller for lateral control. Furthermore, the proposed swarm driving system was developed using the low-cost, open-source PX4-ROS2 platform, and a Data Distribution Service based communication method was employed for communication between unmanned vehicles. The system was validated in real-world environments, confirming its performance.

Attention Map-Based Automatic Masking for Object Swapping in Diffusion Models

Soohyun Lee, Jongyoul Park

http://doi.org/10.5626/JOK.2025.52.4.284

latent diffusion model, stable diffusion, text-to-image model, object swapping, automatic masking AbstractDiffusion models have gained significant traction in the realm of text-to-image generation. The advent of Null-Text Inversion techniques has opened up new avenues for image editing by inverting real images into noise and applying modifications. However, most image editing methods, particularly those involving object manipulation, require user-defined masks, necessitating incorporation of an additional masking model into the pipeline. This complicates the inference process, which ideally should be streamlined within a single model. This paper proposed AutoMask, an attention-based automatic object masking method utilizing attention maps inherent in diffusion models to generate masks during the inference process. Unlike conventional approaches, AutoMask could leverage information obtained from the inversion step, eliminating the need for user intervention in masking. Experiments demonstrated the effectiveness of AutoMask in generating novel objects.

Research on Action Selection Techniques and Dynamic Dense Reward Application for Efficient Exploration in Policy-Based Reinforcement Learning

Junhyuk Kim, Junoh Kim, Kyungeun Cho

http://doi.org/10.5626/JOK.2025.52.4.293

Nowadays, reinforcement learning is being studied and utilized in various fields, including autonomous driving, robotics, and gaming. The goal of reinforcement learning is to find the optimal policy for an agent to interact with its environment. Depending on the environment and the specific problem, either a policy-based algorithm or a value-based algorithm is selected for use. Policy-based algorithms can effectively learn in continuous and high-dimensional action spaces, but they face challenges such as the influence of learning rate parameters on the learning process and increased difficulty in converging to an optimized policy in complex environments. To address these issues, this paper proposes a behavior selection technique and a dynamic dense reward design based on a simulated annealing algorithm. The proposed method is applied to two different environments, and experimental results show that the policy-based reinforcement learning algorithms utilizing this method outperform the standard reinforcement learning algorithms.

Enhancing Retrieval-Augmented Generation Through Zero-Shot Sentence-Level Passage Refinement with LLMs

Taeho Hwang, Soyeong Jeong, Sukmin Cho, Jong C. Park

http://doi.org/10.5626/JOK.2025.52.4.304

This study presents a novel methodology designed to enhance the performance and effectiveness of Retrieval-Augmented Generation (RAG) by utilizing Large Language Models (LLMs) to eliminate irrelevant content at the sentence level from retrieved documents. This approach refines the content of passages exclusively through LLMs, avoiding the need for additional training or data, with the goal of improving the performance in knowledge-intensive tasks. The proposed method was tested in an open-domain question answering (QA) environment, where it demonstrated its ability to effectively remove unnecessary content and outperform over traditional RAG methods. Overall, our approach has proven effective in enhancing performance compared to conventional RAG techniques and has shown the capability to improve RAG's accuracy in a zero-shot setting without requiring additional training data.

A Similarity-Based Multi-Knowledge Transfer Algorithm for Enhancing Learning Efficiency of Reinforcement Learning-Based Autonomous Agent

Yeryeong Cho, Soohyun Park, Joongheon Kim

http://doi.org/10.5626/JOK.2025.52.4.310

This paper proposed a similarity-based multi-knowledge transfer algorithm (SMTRL) to enhance the learning efficiency of autonomous agents in reinforcement learning. SMTRL can calculates the similarity between pre-trained models and the current model and dynamically adjust the knowledge transfer ratio based on this similarity to maximize learning efficiency. In complex environments, autonomous agents face significant challenges when learning independently, as this process can be time-consuming and inefficient, making knowledge transfer essential. However, differences between pre-trained models and actual environments can result in negative transfer, leading to diminished learning performance. To tackle this issue, SMTRL dynamically can adjusts the ratio of knowledge transfer from highly similar pre-trained models, thereby accelerating learning stability. Furthermore, experimental results demonstrated that the proposed algorithm outperformed traditional reinforcement learning and traditional knowledge transfer learning in terms of convergence speed. Therefore, this paper introduces a novel approach to efficient knowledge transfer for autonomous agents and discusses its applicability to complex mobility environments and directions for future research.

Tailored Sentiment Analysis of Economic News Based on a Mixture of Quotation and Attribute Encoders

Seo-In Choi, Dae-Min Park, Byung-Won On

http://doi.org/10.5626/JOK.2025.52.4.319

News articles provide information on various topics such as politics, economics, society, and culture. Their neutral tone often limits the ability of traditional sentiment analysis models to effectively capture emotions. To address this issue, we proposed a novel sentiment analysis model that combined quotations with article attribute values. For sentiment analysis, we employed deep learning-based models such as BERT, KoBERT (optimized for Korean), and KLUE. Embedding results from these models were integrated using a Mixture of Experts (MoE) structure to simultaneously learn the emotional information in quotations and the attribute information of articles. Experimental results demonstrated that the proposed models, including the attribute-based phrase and attribute group embedding models, achieved higher accuracy and reliability than conventional quotation-only analysis and traditional machine learning models. In particular, the KLUE model optimized for Korean data showed improved performance. Incorporating diverse attribute information significantly enhanced predictive accuracies of sentiment analysis models. These findings suggest that effectively combining quotation data with article attribute information enables more sophisticated sentiment analysis, even for neutral news articles.

Researcher Recommendation Index Considering Recency and Impact

Sanghyeuk Kim, Insoo Jung, Jongwoo Jeon, Sangho Song, Christopher Retiti Diop Emane, Dojin Choi, Jaesoo Yoo

http://doi.org/10.5626/JOK.2025.52.4.331

This paper proposes a new researcher recommendation index to address limitations of traditional academic performance metrics such as h-index, g-index, and i10-index, which fail to adequately reflect recent research achievements. The proposed index incorporates two key elements: recency and research impact. By assigning weights based on the recency of a researcher’s publications, this index is designed to evaluate academic performance in alignment with current research trends. Additionally, the index allows for user-adjustable weighting of recency and research impact, enabling tailored researcher recommendations according to specific evaluation goals. Experiments analyzing correlations between the proposed index and existing indices, show that the proposed index more effectively identifies researchers with strong recent achievements. While it maintains a certain level of correlation with traditional indices, it also demonstrates unique characteristics due to its emphasis on recency. Specifically, when the weight for recency is increased, researchers with recent significant contributions rank higher. Conversely, when the weight for research impact is emphasized, the index shows strong correlations with traditional indices, providing stable evaluation results.

Root Cause Analysis for Microservice Systems Using Anomaly Propagation by Resource Sharing

Junho Park, Joyce Jiyoung Whang

http://doi.org/10.5626/JOK.2025.52.4.341

Identifying root causes of failures in microservice systems remains a critical challenge due to intricate interactions among resources and propagation of errors. We propose AnoProp, a novel model for root cause analysis to address challenges by capturing inter-resource interactions and the resulting propagation of anomalies. AnoProp incorporates two core techniques: the anomaly score measurement for metrics using regression models and the root cause score evaluation for resources based on the propagation rate of these anomalies. Experimental results using an Online Boutique dataset demonstrated that AnoProp surpassed existing models across various evaluation metrics, validating its ability to provide balanced performance for different types of root causes. This study underscores the potential of AnoProp to enhance system stability and boost operational efficiency in microservice environments.

Effective Detection of Generated Images Using Frequency Transform

Hyoungwon Seo, Dongsu Kim, Seoyoen Oh, Jisang Lee, Haneol Jang

http://doi.org/10.5626/JOK.2025.52.4.350

In today's digital era, advanced image generation techniques have produced counterfeit images that are nearly indistinguishable from real ones, thereby undermining the trustworthiness of digital information. Conventional machine learning and deep learning methods have shown limitations when confronting these evolving generative algorithms. This study introduces a novel approach that can analyze characteristics of generated images in the frequency domain. Specifically, we independently applied the Fast Fourier Transform (FFT) and the Discrete Cosine Transform (DCT) to evaluate the effectiveness of each method for detecting generated images. Experimental results revealed that the FFT-based model improved the test accuracy by approximately 12.8%, while the DCT-based model demonstrated a performance enhancement of about 22.2%. These findings confirm that a frequency domain approach outperforms traditional spatial domain-based detection techniques. It is expected to make a substantial contribution to enhancing image reliability in digital forensics.


Search




Journal of KIISE

  • ISSN : 2383-630X(Print)
  • ISSN : 2383-6296(Electronic)
  • KCI Accredited Journal

Editorial Office

  • Tel. +82-2-588-9240
  • Fax. +82-2-521-1352
  • E-mail. chwoo@kiise.or.kr