As online news consumption grows, personalized recommendation systems have become integral to digital journalism. However, these systems risk reinforcing filter bubbles and political polarization by failing to incorporate diverse perspectives. Stance detection—identifying a text’s position on a target—can help mitigate this by enabling viewpoint-aware recommendations and data-driven analyses of media bias. Yet, existing stance detection research remains largely limited to short texts and high-resource languages. To address these gaps, we introduce K-News-Stance, the first Korean dataset for article-level stance detection, comprising 2,000 news articles with article-level and 21,650 segment-level stance annotations across 47 societal issues. We also propose JoA-ICL, a Journalism-guided Agentic In-Context Learning framework that employs a language model agent to predict the stances of key structural segments (e.g., leads, quotes), which are then aggregated to infer the overall article stance. Experiments showed that JoA-ICL outperforms existing stance detection methods, highlighting the benefits of segment-level agency in capturing the overall position of long-form news articles. Two case studies further demonstrate its broader utility in promoting viewpoint diversity in news recommendations and uncovering patterns of media bias.

As AI becomes more integral in our lives, the need for transparency and responsibility grows. While natural language explanations (NLEs) are vital for clarifying the reasoning behind AI decisions, evaluating them through human judgments is complex and resource-intensive due to subjectivity and the need for fine-grained ratings. This study explores the alignment between ChatGPT and human assessments across multiple scales (i.e., binary, ternary, and 7-Likert scale). We sample 300 data instances from three NLE datasets and collect 900 human annotations for both informativeness and clarity scores as the text quality measurement. We further conduct paired comparison experiments under different ranges of subjectivity scores, where the baseline comes from 8,346 human annotations. Our results show that ChatGPT aligns better with humans in more coarse-grained scales. Also, paired comparisons and dynamic prompting (i.e., providing semantically similar examples in the prompt) improve the alignment. This research advances our understanding of large language models’ capabilities to assess the text explanation quality in different configurations for responsible AI development.

In this paper, we present a novel smoothing approach for ultra-wideband (UWB) aided unmanned aerial vehicle (UAV) positioning. Existing works based on smoothing or filtering estimate 3D position of UAV by updating a solution for each single 1D low-dimensional UWB range measurement. However, a low-dimensional single range measurement merely acts as a weak constraint in a solution space for UAV position estimation, and thus it can often lead to incorrect estimation in unfavorable conditions. Inspired by the idea that the multilateration outcome can be utilized as a measurement providing a strong constraint, we utilize two types of UWB-based measurements: (i) each single 1D range as a high-rate measurement with a weak constraint, and (ii) multilateration outcome as a low-rate measurement with a strong constraint. We propose an incremental smoothing-based method that seamlessly integrates these two types of UWB-based measurements and inertial measurement into a unified factor graph framework. Through experiments under a variety of scenarios, we demonstrate the effectiveness of the proposed method.

Some news headlines mislead readers with overrated or false information, and identifying them in advance will better assist readers in choosing proper news stories to consume. This research introduces million-scale pairs of news headline and body text dataset with incongruity label, which can uniquely be utilized for detecting news stories with misleading headlines. On this dataset, we develop two neural networks with hierarchical architectures that model a complex textual representation of news articles and measure the incongruity between the headline and the body text. We also present a data augmentation method that dramatically reduces the text input size a model handles by independently investigating each paragraph of news stories, which further boosts the performance. Our experiments and qualitative evaluations demonstrate that the proposed methods outperform existing approaches and efficiently detect news stories with misleading headlines in the real world.

As the world becomes more digitized and interconnected, information that was once considered to be private such as one's health status is now being shared publicly. To understand this new phenomenon better, it is crucial to study what types of health information are being shared on social media and why, as well as by whom. In this paper, we study the traits of users who share their personal health and fitness related information on social media by analyzing fitness status updates that MyFitnessPal users have shared via Twitter. We investigate how certain features like user profile, fitness activity, and fitness network in social media can potentially impact the long-term engagement of fitness app users. We also discuss implications of our findings to achieve a better retention of these users and to promote more sharing of their status updates.

This study investigates factors that may determine satisfaction in customer service operations. We utilized more than 170,000 online chat sessions between customers and agents to identify characteristics of chat sessions that incurred dissatisfying experience. Quantitative data analysis suggests that sentiments or moods conveyed in online conversation are the most predictive factor of perceived satisfaction. Conversely, other session related meta data (such as that length, time of day, and response time) has a weaker correlation with user satisfaction. Knowing in advance what can predict satisfaction allows customer service staffs to identify potential weaknesses and improve the quality of service for better customer experience.

Modern Massively Multiplayer Online Role-Playing Games (MMORPGs) provide lifelike virtual environments in which players can conduct a variety of activities including combat, trade, and chat with other players. While the game world and the available actions therein are inspired by their offline counterparts, the games' popularity and dedicated fan base are testaments to the allure of novel social interactions granted to people by allowing them an alternative life as a new character and persona. In this paper we investigate the phenomenon of "gender swapping," which refers to players choosing avatars of genders opposite to their natural ones. We report the behavioral patterns observed in players of Fairyland Online, a globally serviced MMORPG, during social interactions when playing as in-game avatars of their own real gender or gender-swapped. We also discuss the effect of gender role and self-image in virtual social situations and the potential of our study for improving MMORPG quality and detecting online identity frauds.

Video on demand services have been increasingly proliferated in the Internet. One popular way to disseminate video files among numerous users is to leverage peer-to-peer (P2P) systems (e.g., BitTorrent). However, BitTorrent is not designed with video streaming requirements and hence suffers from long setup delay. In this paper, the drawbacks of existing P2P-based streaming solutions are analyzed in terms of sequential delivery. Then we propose Waterfall that splits the whole swarm into multiple swarms, which are then cascaded by the scene sequence. In this way, peers in a swarm download the chunks of the same video scene from the peers in the same swarm as well as the ones in the preceding swarm that already moved on to the next scene. The average setup delay and maximum playback rate of Waterfall are analyzed. Experiments from a wide area network testbed reveal that Waterfall achieves two to three times higher playback rate and significantly low setup delay than the prior BitTorrent-based streaming solutions.

Since Internet routers are not aware of the contents being forwarded, the same content file is often delivered multiple times inefficiently. Similarly, users cannot exploit a nearby copy of the content of interest unless the content file is serviced by costly content delivery networks. Prior studies on the content-aware routing for efficient content delivery suffer from the scalability problem due to a large number of contents. We propose a scalable content routing, dubbed SCAN, which can exploit nearby and multiple content copies for the efficient delivery. SCAN exchanges the information of the cached contents using Bloom filter. Compared with IP routing, SCAN can offer reduced delivery latency, reduced traffic volume, and load balancing among links.

Peer-to-peer (P2P) live streaming services are getting more popular as the average link capacity in the Internet becomes greater for end hosts. In order to provide a high quality P2P live streaming service, it is essential to handle user dynamics effectively to mitigate streaming disruptions caused by user churn. We collect massive user traces from the largest P2P live streaming site in Korea, namely, Afreeca. Based on the trace results obtained, we analyze user dynamics in the service. In particular, we analyze the user behaviors in terms of three criteria: long-stay users, short-stay users, and user churn. (1) The results show that a certain amount of long-stay users exist in a live streaming session, and it is challenging to recognize the long-stay users online, which will help to provide more resilient live streaming service. (2) Around 20% of the users stay in a session shorter than 1% of the session length. Short-stay users incur high control overhead and we discuss how to efficiently handle the short-stay users. (3) By intentionally disrupting a video streaming, we closely observe the users' leaving behavior and suggest the acceptable recovery time to keep the users remaining the session. We believe our findings can be useful not only to P2P live streaming systems, but also to every live streaming systems.

In this paper, we investigate efficient data access algorithms in intelligent transportation system (ITS)-based networks with multi-hop wireless links. We introduce a proxy cache (PC) and propose PC-based poll-each-read (P-PER) and PC-based callback (P-CB) data access algorithms to reduce the transmission cost over the bottleneck wireless links. Extensive simulation results are given to demonstrate the performance of P-PER and P-CB. It is shown that P-PER and P-CB can improve the cache hit performance and reduce the transmission cost significantly. A tradeoff between P-PER and P-CB suggests the need to use a hybrid proxy-based approach to attain optimal performance of data access in ITS-based networks with multi-hop wireless links.