Mehrab Tanjim, Yeonjun In, Xiang Chen, Victor Bursztyn, Ryan A. Rossi, Sungchul Kim, Guang-Jie Ren, Vaishnavi Muppala, Shun Jiang, Yongsung Kim, Chanyoung Park. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

The accurate representation of pavement in three-dimensional (3D) space is pivotal for road infrastructure management. However, the sparse measurement from affordable equipment, such as 2D LiDAR, has limited the fully utilization of 3D pavement data. This paper aims to address this challenge by exploring the point cloud upsampling for pavement area. Specifically, Pavement-PU, a novel data-driven model is designed to enhance the quality and efficiency of dense point cloud generation task. Utilizing a transformer-based feature extraction module coupled with an integrated point upsampling strategy, Pavement-PU significantly improves the density, uniformity, and accuracy of point clouds derived from sparse and irregular initial scans. Through rigorous testing on the public dataset and a specially curated Pavement3D dataset, the model demonstrates substantial improvements over existing methods in terms of both quantitative metrics and qualitative assessments. Ablation studies further validate the impact of our architectural choices, confirming the effectiveness of the innovative structures implemented within the network. Our research paves the way for more effective and efficient methods in pavement maintenance and monitoring, leveraging advanced techniques for practical, real-world applications in pavement management.

Thanks to the speedy development of science and technology, the service sector of economy has plays more and more important role in modern society. Its rapid growth has led it outstrip the other two sectors of economy, agriculture and industry. However the research of service, such as service science, management and engineering, comparing with agriculture and industry, is rare. In this paper, the service characteristics, definitions and classifications of services have been analyzed and compared. The modeling methods for services have been illustrated too. Based on this illustration and comparison, we suggest describing service by its components. By analyzing the relationship of each component, a value network model is provided to describe the service ecosystem in business environment. At last a simple example is given to analyze how the service providers work in a competitive business environment.

To add descriptive ability to knowledge representation (KR) system based on OWL, many extensions to OWL are proposed to augment its expressive power. Many of these extensions take the approach by changing the syntax and semantic of the description logic corresponding to OWL. A new approach is proposed in this paper by using a dynamic concept interpretation method without changing the structure of original description logic. This approach provides a flexible extension mechanism outside the description logic framework, which will not bring any additional logical computational complexity or uncertainty to the original description logic. And various implementations can be realized through this approach for different extension purposes.

Under current synchronous portal rendering mechanism, a user has to wait until all the portlets are ready for the next interaction. In the case of consuming remote portlets with Web services for remote portlets (WSRP) specification, the situation can be worse. To improve user experiences, an asynchronous mechanism is proposed in this paper leveraging popular Ajax technology. We also discussed several issues such as URL rewriting, cross-portlet interaction, security, etc. that have to be considered while applying asynchronous mechanism, and built up a prototype to prove that using Ajax in rendering remote portlets while providing cross-portlet interaction capability is feasible.