Naver Corporation

About: Naver Corporation is a company organization based out in Seongnam-si, South Korea. It is known for research contribution in the topics: Terminal (electronics) & Computer science. The organization has 4038 authors who have published 4294 publications receiving 35045 citations. The organization is also known as: NAVER Corporation & NAVER.

Design of a Ceiling Glass Installation Robot

Abstract: Building materials and components are much larger and heavier than many industrial materials. Ceiling glass is a type of building material for interior finishing. The demand for larger ceiling glass has increased along with the number of high-rise buildings and an increased interest in interior design. The objective of the study is to introduce robotic technology for installing ceiling glass on construction site. Robotically installed ceiling glass is receiving special attention because of the difficulties in moving to high installation positions and handling fragile building materials. Below, we describe the design of a ceiling glass installation robot. After analyzing a target project, we establish a design concept for a proposed robot. Finally, we describe the detailed design of the robot.

Hacking and Security of Encrypted Access Points in Wireless Network

Abstract: An increasing number of people who use a smart phone or tablet PC are accessing wireless networks in public facilities including cafes and shopping centers. For example, iPhones and Android Phones have been available since 2010. However, security incidents may occur through all sorts of malicious code infection of users’ personal information during the use of an insecure wireless network. In this paper, we will describe the Wi-Fi protected access (WPA) and WPA2 encryption systems used to access a wireless network from a smart phone and tablet PC, and demonstrate the access point (AP) hacking process in a wireless network to which a password is applied on the basis of the analyzed WPA and WPA2 passwords. We will analyze the method of successful AP hacking and propose an approach to enhancing wireless LAN security. This study will contribute to enhancing the security and stability of wireless networks.

Textbook Question Answering with Multi-modal Context Graph Understanding and Self-supervised Open-set Comprehension.

Abstract: In this work, we introduce a novel algorithm for solving the textbook question answering (TQA) task which describes more realistic QA problems compared to other recent tasks. We mainly focus on two related issues with analysis of the TQA dataset. First, solving the TQA problems requires to comprehend multi-modal contexts in complicated input data. To tackle this issue of extracting knowledge features from long text lessons and merging them with visual features, we establish a context graph from texts and images, and propose a new module f-GCN based on graph convolutional networks (GCN). Second, scientific terms are not spread over the chapters and subjects are split in the TQA dataset. To overcome this so called ‘out-of-domain’ issue, before learning QA problems, we introduce a novel self-supervised open-set learning process without any annotations. The experimental results show that our model significantly outperforms prior state-of-the-art methods. Moreover, ablation studies validate that both methods of incorporating f-GCN for extracting knowledge from multi-modal contexts and our newly proposed self-supervised learning process are effective for TQA problems.

The Ameliorative Effects of L-2-Oxothiazolidine-4-Carboxylate on Acetaminophen-Induced Hepatotoxicity in Mice

Abstract: The aim of the study was to investigate the ameliorative effects and the mechanism of action of L-2-oxothiazolidine-4-carboxylate (OTC) on acetaminophen (APAP)-induced hepatotoxicity in mice. Mice were randomly divided into six groups: normal control group, APAP only treated group, APAP + 25 mg/kg OTC, APAP + 50 mg/kg OTC, APAP + 100 mg/kg OTC, and APAP + 100 mg/kg N-acetylcysteine (NAC) as a reference control group. OTC treatment significantly reduced serum alanine aminotransferase and aspartate aminotransferase levels in a dose dependent manner. OTC treatment was markedly increased glutathione (GSH) production and glutathione peroxidase (GSH-px) activity in a dose dependent manner. The contents of malondialdehyde and 4-hydroxynonenal in liver tissues were significantly decreased by administration of OTC and the inhibitory effect of OTC was similar to that of NAC. Moreover, OTC treatment on APAP-induced hepatotoxicity significantly reduced the formation of nitrotyrosin and terminal deoxynucleotidyl transferase dUTP nick end labeling positive areas of liver tissues in a dose dependent manner. Furthermore, the activity of caspase-3 in liver tissues was reduced by administration of OTC in a dose dependent manner. The ameliorative effects of OTC on APAP-induced liver damage in mice was similar to that of NAC. These results suggest that OTC has ameliorative effects on APAP-induced hepatotoxicity in mice through anti-oxidative stress and anti-apoptotic processes.

The ins and outs of speaker recognition: lessons from VoxSRC 2020

Abstract: The VoxCeleb Speaker Recognition Challenge (VoxSRC) at Interspeech 2020 offers a challenging evaluation for speaker recognition systems, which includes celebrities playing different parts in movies. The goal of this work is robust speaker recognition of utterances recorded in these challenging environments. We utilise variants of the popular ResNet architecture for speaker recognition and perform extensive experiments using a range of loss functions and training parameters. To this end, we optimise an efficient training framework that allows powerful models to be trained with limited time and resources. Our trained models demonstrate improvements over most existing works with lighter models and a simple pipeline. The paper shares the lessons learned from our participation in the challenge.