Xi'an Jiaotong University

About: Xi'an Jiaotong University is a education organization based out in Xi'an, China. It is known for research contribution in the topics: Heat transfer & Dielectric. The organization has 85440 authors who have published 99682 publications receiving 1579683 citations. The organization is also known as: '''Xi'an Jiaotong University''' & Xi'an Jiao Tong University.

A general salt-resistant hydrophilic/hydrophobic nanoporous double layer design for efficient and stable solar water evaporation distillation

Abstract: A novel and elegant hydrophilic/hydrophobic nanoporous double layer structure was designed and developed for efficient long-term water desalination. It contained a hydrophobic salt-resistant hierarchical layer of well-defined Cu2SnSe3 (or Cu2ZnSnSe4) nanosphere arrays for broad solar harvesting and water vapor evaporation, and a hydrophilic filter membrane for continuous water supply and vapor generation. The as-fabricated self-floatable devices achieve remarkable solar water evaporation performances (average evaporation rate: 1.657 kg m−2 h−1 and solar thermal conversion efficiency: 86.6% under one sun) with super stability for water distillation from seawater and wastewater containing organic dyes, heavy metals and bacteria.

Current progress in black phosphorus materials and their applications in electrochemical energy storage

Abstract: Recently, a new two-dimensional material, single- or few-layered black phosphorus (BP), has attracted considerable attention for applications in electronics, optoelectronics, and batteries due to its unique properties, including large specific surface area, anisotropy, and tunable and direct band gaps. In particular, contributions to electrochemical energy storage devices, such as lithium and sodium ion batteries and supercapacitors, have emerged. However, critical issues remain to be explored before scaled-up commercial production of BP, such as preparation, stability, and performance. Herein, we present the first review of recent progress in BP-based electrochemical energy storage device. The preparation and electrochemical properties of black phosphorus, recent advances, potential challenges, and relevant perspectives in electrochemical energy storage, and the potential of BP are discussed in this work.

Memetic algorithm for community detection in networks.

Abstract: Community structure is one of the most important properties in networks, and community detection has received an enormous amount of attention in recent years. Modularity is by far the most used and best known quality function for measuring the quality of a partition of a network, and many community detection algorithms are developed to optimize it. However, there is a resolution limit problem in modularity optimization methods. In this study, a memetic algorithm, named Meme-Net, is proposed to optimize another quality function, modularity density, which includes a tunable parameter that allows one to explore the network at different resolutions. Our proposed algorithm is a synergy of a genetic algorithm with a hill-climbing strategy as the local search procedure. Experiments on computer-generated and real-world networks show the effectiveness and the multiresolution ability of the proposed method.

Propagate Yourself: Exploring Pixel-Level Consistency for Unsupervised Visual Representation Learning

Abstract: Contrastive learning methods for unsupervised visual representation learning have reached remarkable levels of transfer performance. We argue that the power of contrastive learning has yet to be fully unleashed, as current methods are trained only on instance-level pretext tasks, leading to representations that may be sub-optimal for downstream tasks requiring dense pixel predictions. In this paper, we introduce pixel-level pretext tasks for learning dense feature representations. The first task directly applies contrastive learning at the pixel level. We additionally propose a pixel-to-propagation consistency task that produces better results, even surpassing the state-of-the-art approaches by a large margin. Specifically, it achieves 60.2 AP, 41.4 / 40.5 mAP and 77.2 mIoU when transferred to Pascal VOC object detection (C4), COCO object detection (FPN / C4) and Cityscapes semantic segmentation using a ResNet-50 backbone network, which are 2.6 AP, 0.8 / 1.0 mAP and 1.0 mIoU better than the previous best methods built on instance-level contrastive learning. Moreover, the pixel-level pretext tasks are found to be effective for pre-training not only regular backbone networks but also head networks used for dense downstream tasks, and are complementary to instance-level contrastive methods. These results demonstrate the strong potential of defining pretext tasks at the pixel level, and suggest a new path forward in unsupervised visual representation learning. Code is available at \url{this https URL}.