Search

Exposure to risks associated with the production and usage of products, particularly oil, poses significant threats to both ecological systems and human health. Notable examples include the Gulf War Oil Spill (1991) and the Deepwater Horizon Oil Spill (2010). However, numerous smaller-scale oil spills, which collectively contribute to substantial oil releases, often remain overlooked. To fill this gap, our study first developed a detailed oil spill incidents database, covering 1967 to 2023. We quantified the released amount (RA) of oil spills recorded by the National Oceanic and Atmospheric Administration (NOAA). Subsequently, we utilized life cycle impact indicators in ReCiPe to evaluate the environmental impacts of these spills. Our findings reveal that approximately 6 million gallons of oil have been spilled in North America, equating to the volume of three Deepwater Horizon spills. The human toxicity of the annually spilled oil is comparable to the release of hundreds of tons of dichlorobenzene (DCB), while its ecotoxicities are equivalent to tens of thousands of tons DCB. Additionally, we observed that the global warming potential and ozone formation potential of spilled oil are highly dependent on the ratio of its light fraction in the spilled oil products.

In this research, we delve into the intricacies of viscous fluid flow with electric field coupling by employing the Finite Element Method (FEM) in tandem with the level set method. We generate a weak form for satisfying governing equations for electric field and fluid velocity while two phases are tracked by the level set function. The primary focus of this study is the complex interactions between free -falling jet and electric field, and the behavior of droplet encompassing deformation, fission, and fusion under the influence of electric field. The main contribution of this paper is given a new implement by using the P1/P1 scheme to directly solve the weak forms of coupled governing equations, which significantly improves calculation efficiency compared to the P2/P1 scheme, and we open source the code. This implement is verified by comparing with the experimental results of oil droplets deforming under an electric field. Computations are performed by FEniCS open -source packages. The phenomena documented underscore the multifaceted relationship between electrodynamic forces and fluid mechanics, accentuated distinctly under non -uniform electric field conditions.

Game learning analytics (GLA) is an emerging technology that facilitates teachers' evidencebased pedagogical design and assessments. Despite its affordances and potential in K-12 classrooms, teachers' integration of GLA in teaching practices remains largely unexplored. This study implemented an educational game on collaborative problem solving (CPS) and a GLA system for assisting K-12 teachers in evaluating students' CPS skills and processes and quest performance and engagement. Based on the integrative model of behavioural prediction, this study aimed to examine 1) the extent to which personal, environmental, and technological factors affected teachers' usage intention and behaviour towards the GLA system, 2) the effects of moderators on the intention-behaviour relationship, and 3) how the structural model relationships differed across teachers with various individual characteristics. Survey data from 300 in-service teachers from Chinese primary and secondary schools were collected and analysed using partial least squares structural equation modelling. Results indicated that our model demonstrated strong insample and out-of-sample predictive power. In particular, teachers' attitudes, subjective norms, and self-efficacy influenced their behavioural intention, while technological pedagogical content knowledge, school support, and behavioural intention predicted their actual behaviour. In addition, technostress acted as a significant moderator of the intention-behaviour relationship. Moreover, teachers' gaming preferences, teaching subjects, and years of teaching explained the heterogeneity of their GLA usage. This study contributes to a theoretical understanding of and methodological advancements in studying teachers' usage intention and behaviour on GLA and yields practical implications for the design and implementation of GLA in K-12 classrooms.

Diffusion models have achieved remarkable success in generating high quality image and video data. More recently, they have also been used for image compression with high perceptual quality. In this paper, we present a novel approach to extreme video compression leveraging the predictive power of diffusion-based generative models at the decoder. The conditional diffusion model takes several neural compressed frames and generates subsequent frames. When the reconstruction quality drops below the desired level, new frames are encoded to restart prediction. The entire video is sequentially encoded to achieve a visually pleasing reconstruction, considering perceptual quality metrics such as the learned perceptual image patch similarity (LPIPS) and the Frechet video distance (FVD), at bit rates as low as 0.02 bits per pixel (bpp). Experimental results demonstrate the effectiveness of the proposed scheme compared to standard codecs such as H.264 and H.265 in the low bpp regime. The results showcase the potential of exploiting the temporal relations in video data using generative models. Code is available at: https://github.com/ElesionKyrie/Extreme-Video-Compression-With-Prediction-Using-Pre-trainded-Diffusion-Models

Referring image segmentation (RIS) aims to locate the particular region corresponding to the language expression. Existing methods incorporate features from different modalities in a \emph{bottom-up} manner. This design may get some unnecessary image-text pairs, which leads to an inaccurate segmentation mask. In this paper, we propose a referring image segmentation method called HARIS, which introduces the Human-Like Attention mechanism and uses the parameter-efficient fine-tuning (PEFT) framework. To be specific, the Human-Like Attention gets a \emph{feedback} signal from multi-modal features, which makes the network center on the specific objects and discard the irrelevant image-text pairs. Besides, we introduce the PEFT framework to preserve the zero-shot ability of pre-trained encoders. Extensive experiments on three widely used RIS benchmarks and the PhraseCut dataset demonstrate that our method achieves state-of-the-art performance and great zero-shot ability.

Recently, diffusion-based purification (DBP) has emerged as a promising approach for defending against adversarial attacks. However, previous studies have used questionable methods to evaluate the robustness of DBP models, their explanations of DBP robustness also lack experimental support. We re-examine DBP robustness using precise gradient, and discuss the impact of stochasticity on DBP robustness. To better explain DBP robustness, we assess DBP robustness under a novel attack setting, Deterministic White-box, and pinpoint stochasticity as the main factor in DBP robustness. Our results suggest that DBP models rely on stochasticity to evade the most effective attack direction, rather than directly countering adversarial perturbations. To improve the robustness of DBP models, we propose Adversarial Denoising Diffusion Training (ADDT). This technique uses Classifier-Guided Perturbation Optimization (CGPO) to generate adversarial perturbation through guidance from a pre-trained classifier, and uses Rank-Based Gaussian Mapping (RBGM) to convert adversarial pertubation into a normal Gaussian distribution. Empirical results show that ADDT improves the robustness of DBP models. Further experiments confirm that ADDT equips DBP models with the ability to directly counter adversarial perturbations.

Membrane distillation (MD) is an emerging thermal desalination technology capable of desalinating waters of any salinity. During typical MD processes, the saline feedwater is heated and acts as the thermal energy carrier; however, temperature polarization (as well as thermal energy loss) contributes to low distillate fluxes, low single-pass water recovery and poor thermal efficiency. An alternative approach is to integrate an extra thermal energy carrier as part of the membrane and/or module assembly, which can channel externally provided heat directly to the membrane-feedwater interface and/or along the feed channel length. This direct-heat delivery has been demonstrated to increase single-pass water recovery and enhance the overall thermal efficiency. We developed a bench-scale direct-heated vacuum MD (DHVMD) process to desalinate pre-treated oil and gas "produced water" with an initial total dissolved solids of 115,500 ppm at a feed temperature ranging between 24 and 32 °C. We evaluated both water flux and specific energy consumption (SEC) as a function of water recovery. The system achieved a 50% water recovery without significant scaling, with an average flux >6 kg m-2 hr-1 and a SEC as low as 2,530 kJ kg-1. The major species of mineral scales (i.e., NaCl, CaSO4, and SrSO4) that limited the water recovery to 68% were modeled in terms of thermodynamics and identified by scanning electron microscopy and energy-dispersive X-ray spectroscopy. In addition, we further developed and employed a physics-based process model to estimate temperature, salinity, water transport and energy flows for full-scale vacuum MD and DHVMD modules. Model results show that a direct-heat input rate of 3,600 W can increase single-pass water recovery from 2.1% to 3.1% while lowering the thermal SEC from 7,800 kJ kg-1 to 6,517 kJ kg-1 in an unoptimized module. Finally, the scaling up potential of DHVMD process is briefly discussed.

In this work, parallel plate capacitors are numerically simulated by solving weak forms within the framework of the finite element method. Two different domains are studied. We study the infinite parallel plate capacitor problem and verify the implementation by deriving analytical solutions with a single layer and multiple layers between two plates. Furthermore, we study the finite parallel plate capacitor problem and verify it by Love’s potential equation and Xiang’s capacitance equation. Moreover, the fringing effect is considered and extended to problems with multiple dielectric layers, such a solution is not possible by means of the existing analytical solutions. Besides, we realize the possibility of choosing different boundary conditions (electric potential boundary conditions and charge density boundary conditions) by changing the weak form. Finally, a transient solution that includes dielectric loss and calculates the quality factor of a capacitor is presented, which may be used in capacitor design. Convergence and consistency of results are demonstrated by comparing the results between analytical and numerical solutions and also the results from different boundary conditions.

Especially at high frequencies, as they occur in 5G or relevant applications, electromagnetism in electronic components plays a significant role in system behavior. Moreover, dissipative effects are triggered by mechanical deformation. Indeed, dissipation is an energy loss and thus the industry tries to minimize this effect by optimizing the design. We examine this justification by using multiphysics simulations by means of the Finite Element Method (FEM) and solve deformation and dissipative dielectric material equations for a simple electronic component in a monolithic manner.

Gravitational waves can generate electromagnetic effects inside a strong electric or magnetic field within the Standard Model and general relativity. Here we propose using a quarterly split cavity and LC-resonance circuit to detect a high-frequency gravitational wave from 0.1 MHz to GHz. We perform a full 3D simulation of the cavity's signal for sensitivity estimate. Our sensitivity depends on the coherence time scale of the high-frequency gravitational wave sources and the volume size of the split cavity. We discuss the resonant measurement schemes for narrow-band gravitational wave sources and also a non-resonance scheme for broadband signals. For a meter-sized split cavity under a 14 Tesla magnetic field, the LC resonance enhanced sensitivity to the gravitational wave strain is expected to reach $h\sim 10^{-20}$ around $10$ MHz., Comment: 8 pages, 7 figures

Several pulsar timing array (PTA) experiments such as NANOGrav and PPTA reported evidence of a gravitational wave background at the nano-Hz frequency band recently. This signal can originate from scalar-induced gravitational waves (SIGW) generated by the enhanced curvature perturbation. Here we perform a joint likelihood inference on PTA datasets, and our results show that if the PTA signals were indeed of SIGW origin, the curvature perturbations amplitude required will produce primordial black holes (PBHs) in $[2 \times 10^{-5}, 2 \times 10^{-2}]\ m_\odot$ mass range. Mergers of these PBHs can leave a strong gravitational wave signature in the $[10^{-3}, 10^8]$ Hz frequency range, to be detectable at upcoming interferometers such as the Einstein Telescope, DECIGO and BBO, etc. This offers a multi-frequency opportunity to further scrutinize the source of the observed PTA signal., Comment: 7 pages, 5 figures

This article proposes an adaptive control for a flexible manipulator (FM) under the influence of distributed disturbances, unknown dead zones, and input quantization. First, the hybrid effect of the unknown dead zone and input quantization is formulated and represented based on some essential transformations. Then, an adaptive robust quantized control with online updating laws is developed to address the uncertainty of the dead zone, ensure robustness and angle position, and dampen the vibration in the FM system. Subsequently, the Lyapunov theoretical analysis is employed to ensure the bounded stability of the system. Finally, numerical simulations and experiments with a Quanser platform are given to further verify the feasibility and superiority of the designed scheme.

The very recent concept of metaverse highlights the importance of virtual reality (VR) and augmented reality (AR), which associates with a wide variety of applications in entertainment, medical treatment, and human-machine interfaces. The current VR/AR technologies mainly rely on visual interaction, while immersive experience in VR and AR highly demands sensational feedback, such as haptic and temperature with noticeable quality in wearable or even skin-integrated formarts. In this article, we report a wearable and flexible haptic interface based on electromagnetic vibrotactile actuators with high wearability and stability. By adopting double layers of copper (Cu) coils at the top and bottom of the magnetic disc, an enhanced electromagnetic field can be generated. Additionally, the intensity of the haptic feedback can be modulated according to sensed pressure in the virtual world by adjusting the value of power input and frequency. The actuator exhibits high stability and tolerance upon environmental, cyclic, and impact resistance tests. Finally, the actuators are developed into the soft VR interfaces for mounting on forearms, fingers, and hands to verify their superiority over conventional haptic actuators in the aspects of performance and applications. © 2014 IEEE.

Partial scan is a common approach to accelerate Magnetic Resonance Imaging (MRI) data acquisition in both 2D and 3D settings. However, accurately reconstructing images from partial scan data (i.e., incomplete k-space matrices) remains challenging due to lack of an effectively global receptive field in both spatial and k-space domains. To address this problem, we propose the following: (1) a novel convolutional operator called Faster Fourier Convolution (FasterFC) to replace the two consecutive convolution operations typically used in convolutional neural networks (e.g., U-Net, ResNet). Based on the spectral convolution theorem in Fourier theory, FasterFC employs alternating kernels of size 1 in 3D case) in different domains to extend the dual-domain receptive field to the global and achieves faster calculation speed than traditional Fast Fourier Convolution (FFC). (2) A 2D accelerated MRI method, FasterFC-End-to-End-VarNet, which uses FasterFC to improve the sensitivity maps and reconstruction quality. (3) A multi-stage 3D accelerated MRI method called FasterFC-based Single-to-group Network (FAS-Net) that utilizes a single-to-group algorithm to guide k-space domain reconstruction, followed by FasterFC-based cascaded convolutional neural networks to expand the effective receptive field in the dual-domain. Experimental results on the fastMRI and Stanford MRI Data datasets demonstrate that FasterFC improves the quality of both 2D and 3D reconstruction. Moreover, FAS-Net, as a 3D high-resolution multi-coil (eight) accelerated MRI method, achieves superior reconstruction performance in both qualitative and quantitative results compared with state-of-the-art 2D and 3D methods.

[EN]: The objective of this work is to identify the different names, positions, and roles of editorial board members (EBMs) of journals in the Information Science and Library Science (ISLS) field of the Journal Citation Reports (JCR) and to establish a classification of the different roles to improve communication between editors and researchers. Among the 86 journals included in the ISLS field, the first and last names, position, and institutional affiliation of each EBM were extracted from 84 journals that allowed access to committee information.The information was standardized, and the positions were classified by consensus into major groups according to the role they play in the journals.The 84 journals included 4,122 positions held by EBMs, distributed in 201 positions and classified into 11 broad categories.The majority of positions were included in the Editorial Board category (n = 1,516), followed by Editorial Advisory Board (n = 734) and Associate Editor (n = 566). The journals were published by 36 publishers, with Elsevier and Taylor & Francis contributing the most journals (11 each). The journals in which the EBMs had the highest percentage of foreign members were those publi-shed in the Netherlands. This study provides a picture of the broad distribution of the roles and terminological titles that EBMs have in Information Science and Library Science journals. Because of this diversity, it is advisable to develop a guide to good editorial practices that includes both a description of the complex range of roles performed by EBMs and a unified nomenclature for these universal positions., [ES]: El objetivo de este trabajo es identificar las diferentes denominaciones, puestos y papeles de los miembros de los comités editoriales (MCEs) de las revistas del área de Information Science and Library Science (ISLS) deJournal Citation Reports (JCR) y establecer una clasificación de los roles que contribuya a mejorar la comunicación entre editores e investigadores. De las 86 revistas incluidas del área ISLSse extrajeron los apellidos y nombres, puesto y afiliación institucional de cada MCE de un total de 84 revistas que permitían acceder a la información sobre los comités. Se normalizó la información y se clasificaron los puestos por consenso en grandes grupos según el papel que desempeñaban en las revistas. En las 84 revistas se inclu-yeron 4.122 MCEs distribuidos en 201 puestos y se clasificaron en 11 grandes categorías. La mayoría de puestos estaban incluidos en la categoría Editorial Board (n=1.516), seguida de Editorial Advisory Board (n=734) y Associate Editor (n=566). Las revistas se editaron en 36 editoriales, siendo Elsevier y Taylor & Francis las que aportaron más revistas (11 cada una). Las revistas en las que los MCEs presentaban un mayor porcentaje de miembros extranjeros son las editadas en los Países Bajos. El estudio proporciona una imagen de la amplia casuística de los papeles y denominaciones terminológicas que tienen los MCEs en las revistas del área Information Science & Library Science. Debido a esta diversidad, sería conveniente elaborar una guía de buenas prácticas editoriales que incluyera tanto la descripción del complejo abanico de funciones que desarrollan los MCEs como una denominación unificada de estos puestos de uso universal.

The adverse impacts of extremely large oil spill incidents such as Deepwater Horizon Oil Spill have been widely studied in recent years. Catastrophic oil spill disasters such as Deepwater Horizon Oil Spill have been widely studied. While vast smaller oil spills, which also could result in significant amount of oil release, still lack attention. This research assessed distributions and environmental impacts caused by historical oil spill incidents in U.S. We first analyzed the spatial and temporal characteristics of oil spills. Then the relationship between spillage amount and spill frequency is evaluated via statistical approaches. Last, based on a thorough literature review of environmental, ecological, and economic impacts of oil spill, we proposed a framework to estimate the impact of oil spill incidents.

Referring image segmentation (RIS) aims to segment a particular region based on a language expression prompt. Existing methods incorporate linguistic features into visual features and obtain multi-modal features for mask decoding. However, these methods may segment the visually salient entity instead of the correct referring region, as the multi-modal features are dominated by the abundant visual context. In this paper, we propose MARIS, a referring image segmentation method that leverages the Segment Anything Model (SAM) and introduces a mutual-aware attention mechanism to enhance the cross-modal fusion via two parallel branches. Specifically, our mutual-aware attention mechanism consists of Vision-Guided Attention and Language-Guided Attention, which bidirectionally model the relationship between visual and linguistic features. Correspondingly, we design a Mask Decoder to enable explicit linguistic guidance for more consistent segmentation with the language expression. To this end, a multi-modal query token is proposed to integrate linguistic information and interact with visual information simultaneously. Extensive experiments on three benchmark datasets show that our method outperforms the state-of-the-art RIS methods. Our code will be publicly available.

In the Standard Model Effective Field Theory (SMEFT), operators involving the top quark are generally difficult to probe, and can generate sizable loop contributions to the electroweak precision observables, measured by past and future lepton colliders. Could the high precision of the electroweak measurements compensate the loop suppression and provide competitive reaches on these operators? Would the inclusion of these contributions introduce too many additional parameters for a meaningful global electroweak analysis to be done? In this paper, we perform a detailed phenomenological study to address these two important questions. Focusing on eight dimension-6 operators that generate anomalous couplings between the electroweak gauge bosons and the third-generation quarks, we calculate their one loop contributions to the $e^+e^- \to f\bar{f}$ processes both on and off the Z-pole and the $e^-e^+ \to WW$ process. A global analysis is performed with these eight operators and the ones that contribute to the above processes at tree level, using the measurements at LEP, SLC and several low energy experiments. We find that, while the current electroweak precision measurements are sensitive to the one-loop effects of the top-quark operators, it is difficult to separate them from the operators that contribute at the tree level, making a global analysis rather challenging. Under more assumptions, competitive reaches could be obtained in a global fit. Another important finding of our study is that the two operators that generate dipole interactions of the bottom quark have significant impacts in the Z-pole measurements and should not be omitted. We also discuss the implication of the recently reported W-boson mass measurement at CDF to our results. Finally, we estimate the reaches of future lepton colliders in probing the top-quark operators with precision electroweak measurements., Comment: 38 pages, 14 figures

The relationship between the ecological environment and tourism development is extremely complex. Summarizing evolutionary trends, temporal patterns and spatial differences in time and space can contribute to the coordinated development of an ecological environment and tourism. Based on the Carrying capacity–Supporting capability–Attraction capability–Evolutional capability–Development capability model and the Pressure–State–Response model, this paper proposes a conceptual index system using the entropy weight method and the coupled coordination degree model. The coupling coordination relationship between the ecological environment and tourism development in seven prefectures in the Kyushu region of Japan from 2010 to 2019 was analyzed. Then, using the membership degree model, the coordinated development of each prefecture was further analyzed. The results mainly show that in the Kyushu region, the degree of coupling coordination between the ecological environment and the tourism development system changed from moderate coordination to marginal coordination during the study period, then steadily improved, with large differences between regions. Based on the analysis results, this paper puts forward specific suggestions for the prefectures with lagging tourism development and lagging ecological environment development, respectively, to provide a reference path for the sustainable development of the ecological environment and tourism in the Kyushu region. For prefectures in the ecological environment development lag-type stage, the following requirements were identified: (a) the need for the government to strengthen tourism infrastructures; (b) the need for the government to focus on developing regional ecotourism, focusing on the fragility of island ecosystems and the sustainability of resources in the Kyushu region; (c) the need for the government to scientifically formulate tourism-related policies and plan tourism-related investments. For prefectures in the touris, Environmental Research Letters, 18(1), art. no. 014004; 2022

The current study attempted to address an overarching question – why students often fail to learn from written corrective feedback (WCF). To answer this question, I investigated the nature of L2 writing teachers’ WCF and students’ responses to WCF, and probed into the factors shaping the WCF practices and responses. Two main theories guided this study. Sociocultural theory (SCT) helped evaluate the effectiveness of the WCF teachers provided – by exploring to what extent the feedback accorded with learners’ language development. Activity Theory allowed for a deeper understanding of teachers’ and students’ feedback-related behaviours by highlighting the interplay of individual and contextual factors shaping these behaviours. This study was situated in a private language institute (primarily an IELTS/TOFEL training institute). This is a context with a profit orientation and customer focus potentially leading to a teacher-student relationship different from that in university or school settings where many previous studies of WCF have been conducted. Six teachers, 23 students and one administrator participated in the study. Data collected included semi-structured interviews, retrospections and students’ drafts of essays along with teachers’ feedback. The essays were analysed quantitatively to investigate the nature of teachers’ WCF, students’ responses in revision (e.g., acceptance/rejection), and whether teachers’ feedback aligned with students’ accuracy development. The interviews and retrospections were analysed qualitatively to explore the factors shaping teachers’ and students’ feedback-related behaviours. The findings on the teachers’ WCF centre around the explicitness and quantity of WCF. In terms of feedback explicitness, the teachers tended to provide primarily direct WCF (i.e., direct corrections). A key reason for this, as reported by the teachers, was that indirect WCF (e.g., underlining errors) seemed more time-consuming due to the re-reading and extra clari

Magnetic resonance imaging plays an essential role in clinical diagnosis by acquiring the structural information of biological tissue. Recently, many multi-contrast MRI super-resolution networks achieve good effects. However, most studies ignore the impact of the inappropriate foreground scale and patch size of multi-contrast MRI, which probably leads to inappropriate feature alignment. To tackle this problem, we propose the Flexible Alignment Super-Resolution Network (FASR-Net) for multi-contrast MRI Super-Resolution. The Flexible Alignment module of FASR-Net consists of two modules for feature alignment. (1) The Single-Multi Pyramid Alignment(S-A) module solves the situation where low-resolution (LR) images and reference (Ref) images have different scales. (2) The Multi-Multi Pyramid Alignment(M-A) module solves the situation where LR and Ref images have the same scale. Besides, we propose the Cross-Hierarchical Progressive Fusion (CHPF) module aiming at fusing the features effectively, further improving the image quality. Compared with other state-of-the-art methods, FASR-net achieves the most competitive results on FastMRI and IXI datasets. Our code will be available at \href{https://github.com/yimingliu123/FASR-Net}{https://github.com/yimingliu123/FASR-Net}.

Intelligent edge network is maturing to enable smart and efficient transportation systems. In this letter, we consider unmanned aerial vehicle (UAV)-assisted vehicular networks where UAVs provide caching and computing services in complement with base station (BS). One major challenge is that vehicles need to obtain timely situational awareness via orchestration of ubiquitous caching and computing resources. Note that cached data for vehicles' perception tasks contains time-varying context information, thus freshness of cached data should be considered in conjunction with task execution to guarantee timeliness of obtained status updates. To this end, we propose a two-stage performance metric to quantify the impact of cache refreshing and computation offloading decisions on the age of status updates. We formulate an energy minimization problem by jointly considering cache refreshing, computation offloading and aging of status updates. To facilitate online decision making, we propose a deep deterministic policy gradient(DDPG)-based solution procedure and incorporate differentiated experience replay mechanism to accelerate convergence. Simulation results show that the performance of proposed solution is competitive in terms of energy consumption for obtaining fresh status updates.

Although robust PCA has been increasingly adopted to extract vessels from X-ray coronary angiography (XCA) images, challenging problems such as inefficient vessel-sparsity modelling, noisy and dynamic background artefacts, and high computational cost still remain unsolved. Therefore, we propose a novel robust PCA unrolling network with sparse feature selection for super-resolution XCA vessel imaging. Being embedded within a patch-wise spatiotemporal super-resolution framework that is built upon a pooling layer and a convolutional long short-term memory network, the proposed network can not only gradually prune complex vessel-like artefacts and noisy backgrounds in XCA during network training but also iteratively learn and select the high-level spatiotemporal semantic information of moving contrast agents flowing in the XCA-imaged vessels. The experimental results show that the proposed method significantly outperforms state-of-the-art methods, especially in the imaging of the vessel network and its distal vessels, by restoring the intensity and geometry profiles of heterogeneous vessels against complex and dynamic backgrounds.

The conversational recommender systems (CRSs) have received extensive attention in recent years. However, most of the existing works focus on various deep learning models, which are largely limited by the requirement of large-scale human-annotated datasets. Such methods are not able to deal with the cold-start scenarios in industrial products. To alleviate the problem, we propose FORCE, a Framework Of Rule-based Conversational Recommender system that helps developers to quickly build CRS bots by simple configuration. We conduct experiments on two datasets in different languages and domains to verify its effectiveness and usability., Comment: AAAI 2022 (Demonstration Track)

Purpose: Long scan time in phase encoding for forming complete K-space matrices is a critical drawback of MRI, making patients uncomfortable and wasting important time for diagnosing emergent diseases. This paper aims to reducing the scan time by actively and sequentially selecting partial phases in a short time so that a slice can be accurately reconstructed from the resultant slice-specific incomplete K-space matrix. Methods: A transformer based deep reinforcement learning framework is proposed for actively determining a sequence of partial phases according to reconstruction-quality based Q-value (a function of reward), where the reward is the improvement degree of reconstructed image quality. The Q-value is efficiently predicted from binary phase-indicator vectors, incomplete K-space matrices and their corresponding undersampled images with a light-weight transformer so that the sequential information of phases and global relationship in images can be used. The inverse Fourier transform is employed for efficiently computing the undersampled images and hence gaining the rewards of selecting phases. Results: Experimental results on the fastMRI dataset with original K-space data accessible demonstrate the efficiency and accuracy superiorities of proposed method. Compared with the state-of-the-art reinforcement learning based method proposed by Pineda et al., the proposed method is roughly 150 times faster and achieves significant improvement in reconstruction accuracy. Conclusions: We have proposed a light-weight transformer based deep reinforcement learning framework for generating high-quality slice-specific trajectory consisting of a small number of phases. The proposed method, called TITLE (Transformer Involved Trajectory LEarning), has remarkable superiority in phase-encode selection efficiency and image reconstruction accuracy., Comment: 12 pages, 7 figures

Intuitive and efficient physical human-robot collaboration relies on the mutual observability of the human and the robot, i.e. the two entities being able to interpret each other's intentions and actions. This is remedied by a myriad of methods involving human sensing or intention decoding, as well as human-robot turn-taking and sequential task planning. However, the physical interaction establishes a rich channel of communication through forces, torques and haptics in general, which is often overlooked in industrial implementations of human-robot interaction. In this work, we investigate the role of haptics in human collaborative physical tasks, to identify how to integrate physical communication in human-robot teams. We present a task to balance a ball at a target position on a board either bimanually by one participant, or dyadically by two participants, with and without haptic information. The task requires that the two sides coordinate with each other, in real-time, to balance the ball at the target. We found that with training the completion time and number of velocity peaks of the ball decreased, and that participants gradually became consistent in their braking strategy. Moreover we found that the presence of haptic information improved the performance (decreased completion time) and led to an increase in overall cooperative movements. Overall, our results show that humans can better coordinate with one another when haptic feedback is available. These results also highlight the likely importance of haptic communication in human-robot physical interaction, both as a tool to infer human intentions and to make the robot behaviour interpretable to humans.

A combined treatment using medication and electro-stimulation increases its effectiveness in comparison with one treatment alone. However, the organic integration of two strategies in one miniaturized system for practical usage has seldom been reported. This article reports an implantable electronic medicine based on bioresorb-able microneedle devices that is activated wirelessly for electro-stimulation and sustainable delivery of anti-inflammatory drugs. The electronic medicine is composed of a radio frequency wireless power transmission system and a drug-loaded microneedle structure, all fabricated with bioresorbable materials. In a rat skeletal muscle injury model, periodic electrostimulation regulates cell behaviors and tissue regeneration while the anti-inflammatory drugs prevent inflammation, which ultimately enhance the skeletal muscle regeneration. Finally, the electronic medicine is fully bioresorbable, excluding the second surgery for device removal.

Yajun Chen, Yiming Liu, Menghao Wang Department of Hepatobiliary Surgery, the Second Affiliated Hospital of Chongqing Medical University, Chongqing, People’s Republic of ChinaCorrespondence: Menghao Wang; Yiming Liu, Email 305849@hospital.cqmu.edu.cn; doctorliu@hospital.cqmu.edu.cnBackground: Pancreatic adenocarcinoma (PAAD) is a highly malignant tumor with an extremely poor prognosis. Pyroptosis has been demonstrated to play an important role in tumor prognosis. However, the expression of pyroptosis-related genes in PAAD and their correlations with prognosis remains unclear.Methods: In this study, the 36 pyroptosis-related genes that were differentially expressed between normal pancreatic tissues and PAAD tissues were identified via the “limma” R package. Based on these differentially expressed genes (DEGs), a five-gene signature was established by applying the least absolute shrinkage and selection operator Cox regression in the TCGA cohort and was validated in the GEO cohort. The Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analyses of DEGs based on the risk model indicated that immune-associated biological processes and pathways were enriched. In vivo, we detected the expressions of CASP4 and CHMP4C by immunohistochemistry in tumor tissues and adjacent normal tissues. In vitro, we silenced CASP4 and CHMP4C to explore their effects on pancreatic cancer cells.Results: PAAD patients in the low-risk group showed significantly higher survival possibilities than those in the high-risk group. The expressions of CASP4 and CHMP4C in tumor tissue were higher than those in the adjacent normal tissues in vivo. The knockdown of CASP4 significantly inhibited the invasion and migration but not the proliferation of PANC-1 cells. The knockdown of CHMP4C obviously inhibited the proliferation, migration, and invasion of PANC-1 cells.Conclusion: Pyroptosis-related genes play important roles in predicting the prognosis of PAAD, and CASP4 and CHMP4C affect the met

Tong Zhao, Dongdong Zhang, Yiming Liu, Mingming Feng, Ze Xu, Haoyue Huang, Luoya Zhang, Wenjie Li, Xing Li Department of Nutrition and Food Hygiene, College of Public Health, Zhengzhou University, Zhengzhou, 450001, Henan, People’s Republic of ChinaCorrespondence: Xing LiDepartment of Nutrition and Food Hygiene, College of Public Health, Zhengzhou University, 100 Kexue Avenue, Zhengzhou, 450001, Henan, People’s Republic of ChinaTel +86 371 6778 1305Email lixing530@zzu.edu.cnBackground: GC (group-specific component globulin) encoding VDBP (Vitamin D binding protein) polymorphisms have been associated with susceptibility to some diseases such as diabetes, obesity, osteoporosis, and polycystic ovary syndrome, but the evidence for metabolic syndrome (MetS) in the Chinese rural population is inconclusive. Therefore, we investigated the relationship between GC variants (rs7041, rs4588, rs2282679, and rs705117) and MetS risk as well as VDBP levels in the Chinese rural population.Patients and Methods: The participants (range of age: 20– 90 years) of this case–control study were recruited from the northern Chinese Han rural population. We matched 445 MetS cases with non-MetS controls in a 1:1 ratio by sex, age (within 5 years). Real-time PCR technology was carried out by TaqMan assays to examine the four variants of rs7041, rs4588, rs2282679, and rs705117 within the GC gene. To identify the association of GC gene polymorphisms with MetS, we calculated ORs using a conditional logistic regression model adjusted for potential confounding factors.Results: We observed inverse associations of CA and AA genotypes of rs4588 with risk of MetS (OR = 0.678, 95% CI 0.505– 0.910, P = 0.010; 0.603, 95% CI 0.373– 0.973, P = 0.039, respectively) compared with carriers of CC genotype. A similar relationship was also found between rs2282679 and MetS, showing that carrying AC genotype of rs2282679 can decrease the risk of MetS (OR = 0.683, 95% CI 0.509– 0.917, P = 0.011) compared

Neural networks have provided powerful approaches to solve various scientific problems. Many of them are even difficult for human experts who are good at accessing the physical laws from experimental data. We investigate whether neural networks can assist us in exploring the fundamental laws of classical mechanics from data of planetary motion. Firstly, we predict the orbits of planets in the geocentric system using the gate recurrent unit, one of the common neural networks. We find that the precision of the prediction is obviously improved when the information of the Sun is included in the training set. This result implies that the Sun is particularly important in the geocentric system without any prior knowledge, which inspires us to gain Copernicus' heliocentric theory. Secondly, we turn to the heliocentric system and make successfully mutual predictions between the position and velocity of planets. We hold that the successful prediction is due to the existence of enough conserved quantities (such as conservations of mechanical energy and angular momentum) in the system. Our research provides a new way to explore the existence of conserved quantities in mechanics system based on neural networks., Comment: 6 pages, 5 figures

Governments across the world have implemented restrictive policies to slow the spread of COVID-19. Recommended face mask use has been a controversially discussed policy, among others, due to potential adverse effects on physical distancing. Using a randomized field experiment (N=300), we show that individuals keep a significantly larger distance from someone wearing a face mask than from an unmasked person. According to an additional survey experiment (N=456), masked individuals are not perceived as being more infectious than unmasked ones, but they are believed to prefer more distancing. This result suggests that, in times where mask use is voluntary, wearing a mask serves as a social signal for a preferred greater distance that is respected by others. Our findings provide strong evidence against the claim that mask use creates a false sense of security that would negatively affect physical distancing.

With the intensification of global warming, the development of low-carbon tourism is not only an inevitable requirement for human development, but also an important way to enhance tourism, improve the quality of tourism, and obtain greater economic and social benefits from tourism. Scientifically and effectively constructing a low-carbon tourism development level evaluation index system is conducive to the healthy and sustainable development of the tourism industry. However, the current evaluation of the development level of low-carbon tourism mainly focuses on low-carbon development, energy conservation, and emission reduction. There are few studies and a lack of specific empirical research evaluating the development of low-carbon tourism. In this research, the Delphi method and analytic hierarchy process were used for data collection and corresponding weight analysis. This research refers to the results of existing studies on low-carbon tourism evaluation index systems, and constructs an objective and quantitative evaluation model. It establishes a low-carbon tourism development evaluation model with 34 specific indicators from the three dimensions of the tourism economic support level, the tourism low-carbon development level, and the tourism policy support level. The research results were applied to the Daxinganling region for verification. The results show that the development of tourism in the Daxinganling region has changed from a mode dominated by economic interests to a mode in which tourism-related parties begin to seek effective low-carbon methods to maintain low-carbon tourism activities. In addition, the Daxinganling region has the lowest level of low-carbon tourism development and the highest level of financial support. Based on the research conclusions, corresponding suggestions for the development of low-carbon tourism in the Daxinganling region are provided. The results are also intended to provide a reference point in the optimization of low-carbon, Sustainability, 13(21), art. no. 12026; 2021

The tourism environment is the basis of sustainable development in the tourism economy. Exploring the coupling relationship between tourism economy and ecological environment systems can promote not only ecology-based tourism, but also contribute to the sustainable development of tourism economy. Based on data from Nagasaki Prefecture, Japan, from 2010–2019, this paper aims to introduce an indicator system and develop an integrated approach to assess the coupling and coordination between the tourism economy and the environment. The indicator system consists of two levels, six aspects, and eighteen indicators, based on entropy method. A Pressure-State-Response (PSR) model framework of the coupling and coordination mechanism of tourism economy and ecological environment was constructed based on the development status of Nagasaki Prefecture in Japan. Then, the degree of coupled coordination of its tourism economy and ecological environment is evaluated, providing a comprehensive evaluation index of the system. In conclusion, suggestions for promoting the sustainable development of tourism and environment in Nagasaki Prefecture, Japan, are proposed. The purpose of this research is to reveal dynamic trends that exist between the coupled development of tourism economy and the ecological environment. A further aim of this paper is to provide a reference for macro policy formulation in small and medium-sized cities regarding the sustainable development of the tourism economic system and ecological environment system., International Journal of Environmental Research and Public Health, 18(23), art. no. 12818; 2021

After the Belt and Road Initiative (BRI) was put forward, the tourism industry between Azerbaijan and China developed rapidly, and gradually became an important area of cooperation between the two countries to promote economic cooperation, non-governmental exchange, and cultural exchange. However, there is currently a lack of research on the development strategy of tourism cooperation between the two countries under the BRI. This paper aims to explore the development strategy of sustainable tourism cooperation between Azerbaijan and China under BRI through a hybrid model of SWOT (Strengths, Weaknesses, Opportunities, and Threats) and AHP (Analytic Hierarchy Process). The factor elements of the SWOT matrix introduce ambiguity due to the complex layering of elemental importance. Therefore, the AHP method was used to systematically evaluate the importance of strategic factors and reduce the subjective bias caused by the SWOT analysis method. By constructing the Delphi-AHP-SWOT strategic quadrilateral and calculating the azimuth of the strategy type, the type and intensity map of the tourism cooperation development strategy between Azerbaijan and China under the BRI is drawn. Finally, it is concluded that the development strategy of tourism cooperation between Azerbaijan and China is a pioneering development strategy. Suggestions are given for future development. This paper provides a valuable experience for the sustainable development of tourism cooperation among countries under the BRI., Sustainability, 14(1), art. no. 187; 2022

Expectations of higher income and better living conditions are the major causes behind migration. Whether migration is temporary or permanent is a consequence of both these expectations as well as the policies and the economic development of the host area. Migration comes with risks and migrants are more willing to take risks than nonmigrants. Entrepreneurship, like migration, also requires risk-taking for potentially high returns. The theme that connects the three chapters in the dissertation is the risks people take as they decide to migrate, to start new businesses, and the hedging strategies they adopt to mitigate these risks. This paper studies internal migration in China and India as well as the risk hedging strategies of trading companies in the US markets.

Blur was naturally analyzed in the frequency domain, by estimating the latent sharp image and the blur kernel given a blurry image. Recent progress on image deblurring always designs end-to-end architectures and aims at learning the difference between blurry and sharp image pairs from pixel-level, which inevitably overlooks the importance of blur kernels. This paper reveals an intriguing phenomenon that simply applying ReLU operation on the frequency domain of a blur image followed by inverse Fourier transform, i.e., frequency selection, provides faithful information about the blur pattern (e.g., the blur direction and blur level, implicitly shows the kernel pattern). Based on this observation, we attempt to leverage kernel-level information for image deblurring networks by inserting Fourier transform, ReLU operation, and inverse Fourier transform to the standard ResBlock. 1x1 convolution is further added to let the network modulate flexible thresholds for frequency selection. We term our newly built block as Res FFT-ReLU Block, which takes advantages of both kernel-level and pixel-level features via learning frequency-spatial dual-domain representations. Extensive experiments are conducted to acquire a thorough analysis on the insights of the method. Moreover, after plugging the proposed block into NAFNet, we can achieve 33.85 dB in PSNR on GoPro dataset. Our method noticeably improves backbone architectures without introducing many parameters, while maintaining low computational complexity. Code is available at https://github.com/DeepMed-Lab/DeepRFT-AAAI2023., Comment: AAAI 2023

The worldwide spread of Covid-19 pandemic caused a dramatic cutback of human activities and triggered a large-scale atmospheric composition experiment. This unfortunate situation provides the opportunity to investigate the response of atmospheric composition to the shutdown measures. Our focus will be on China, where the pandemic emerged in January 2020, and thence strict lockdowns were enforced. Substantial, large-scale decreases in pollutants levels over China and subsequent recovery were revealed by spaceborne observations from TROPOMI instrument on board Sentinel-5 Precursor, as well as by in situ measurements. Most published work on this topic relied on observed changes in column abundances of nitrogen dioxide (NO2), a predominantly anthropogenic compound and an important precursor for ozone production andsecondary aerosol formation. Our work adds to this picture by studing the evolution of two other satellite-derived compounds, formaldehyde (HCHO) and peroxyacylnitrate (PAN), observed by TROPOMI and IASI, respectively. HCHO is an intermediate product in the chemical processing of volatile organic compounds (VOCs) of anthropogenic and natural origin. PAN is formed in the oxidation of anthropogenic and biogenic VOCs, and constitute the principal tropospheric NOx reservoir, enabling the transport and release of NOx away from the sources. Chemistry-transportsimulations of PAN are challenging due to large uncertainties in formation mechanisms and precursor emissions. We will evaluate and analyze the observed variability of NO2, HCHO, and PAN columns using model simulations with the MAGRITTE v1.1 regional CTM run at 0.5ox0.5o resolution over China for 2019 and 2020. The model uses updated anthropogenic emissions from the CONFORM dataset, which takes into account the reductions during the shutdowns based on traffic and other economic activity data., info:eu-repo/semantics/nonPublished

China was the first country to undergo large‐scale lockdowns in response to the pandemic in early 2020 and a progressive return to normalization after April 2020. Substantial decreases in pollutants levels and their subsequent recovery were revealed by spaceborne observations and in situ measurements of nitrogen dioxide (NO2), a predominantly anthropogenic compound and an important precursor for ozone and aerosol formation. This study adds to this picture with the analysis of satellite data of oxygenated volatile organic compounds (OVOCs), namely formaldehyde (HCHO), glyoxal (CHOCHO) and peroxyacetyl nitrate (PAN). The observations reveal important changes in pollutants levels in response to the pandemic‐induced shutdowns and subsequent drop in pollutant emissions. In February 2020, when the shutdowns were at their peak, the observed declines in the OVOC levels were generally weaker (less than 20%) than the substantial NO2 reductions (‐40%). In May 2020, the observations reveal moderate decreases for NO2 (‐15%) and PAN (‐21%), and small changes for CHOCHO (‐3%) and HCHO (6%). Simulations with a regional atmospheric model using anthropogenic emissions taking into account the reductions during the shutdowns based on activity data, explain to a large extent the observed decreases between 2020 and 2019 in regions affected by the lockdowns. In areas where biomass burning and biogenic sources are dominant, the observed changes reflect the interannual variability of these sources, and are well captured by the model simulation., info:eu-repo/semantics/nonPublished

China was the first country to undergo large-scale lockdowns in response to the pandemic in early 2020 and a progressive return to normalization after April 2020. Spaceborne observations of atmospheric nitrogen dioxide (NO2) and oxygenated volatile organic compounds (OVOCs), including formaldehyde (HCHO), glyoxal (CHOCHO), and peroxyacetyl nitrate (PAN), reveal important changes over China in 2020, relative to 2019, in response to the pandemic-induced shutdown and the subsequent drop in pollutant emissions. In February, at the peak of the shutdown, the observed declines in OVOC levels were generally weaker (less than 20%) compared to the observed NO2 reductions (-40%). In May 2020, the observations reveal moderate decreases in NO2 (-15%) and PAN (-21%), small changes in CHOCHO (-3%) and HCHO (6%). Model simulations using the regional model MAGRITTEv1.1 with anthropogenic emissions accounting for the reductions due to the pandemic explain to a large extent the observed changes in lockdown-affected regions. The model results suggest that meteorological variability accounts for a minor but non-negligible part (~-5%) of the observed changes for NO2, whereas it is negligible for CHOCHO but plays a more substantial role for HCHO and PAN, especially in May. The interannual variability of biogenic and biomass burning emissions also contribute to the observed variations, explaining e.g. the important column increases of NO2 and OVOCs in February 2020, relative to 2019. These changes are well captured by the model simulations., SCOPUS: ar.j, info:eu-repo/semantics/published

The polarization of optical fields is a crucial degree of freedom in the all-optical analogue of electromagnetically induced transparency (EIT). However, the physical origins of EIT and polarization induced phenomena have not been well distinguished, which can lead to confusion in associated applications such as slow light and optical/quantum storage. Here we study the polarization effects in various optical EIT systems. We find that a polarization mismatch between whispering gallery modes in two indirectly coupled resonators can induce a narrow transparency window in the transmission spectrum resembling the EIT lineshape. However, such polarization induced transparency (PIT) is distinct from EIT: it originates from strong polarization rotation effects and shows unidirectional feature. The coexistence of PIT and EIT provides new routes for the manipulation of light flow in optical resonator systems., Comment: 8 pages, 4 figures, 57 references. The published version can be found via ULR: https://www.pnas.org/content/118/3/e2012982118

In this letter, we study the outage probability of intelligent reflecting surface (IRS) assisted full duplex two-way communication systems, which characterizes the performance of overcoming the transmitted data loss caused by long deep fades. To this end, we first derive the probability distribution of the cascaded end-to-end equivalent channel with an arbitrarily given IRS beamformer. Our analysis shows that deriving such probability distribution in the considered case is more challenging than the case with the phase-matched IRS beamformer. Then, with the derived probability distribution of the equivalent channel, we obtain the closed-form expression of the outage probability performance. It theoretically shows that the reflecting element number has a conspicuous effect on the improvement of the system reliability. Extensive numerical results verify the correctness of the derived results and confirm the superiority of the considered IRS assisted two-way communication system comparing to the one-way counterpart.

We present a novel spectral machine learning (SML) method in screening for pancreatic mass using CT imaging. Our algorithm is trained with approximately 30,000 images from 250 patients (50 patients with normal pancreas and 200 patients with abnormal pancreas findings) based on public data sources. A test accuracy of 94.6 percents was achieved in the out-of-sample diagnosis classification based on a total of approximately 15,000 images from 113 patients, whereby 26 out of 32 patients with normal pancreas and all 81 patients with abnormal pancreas findings were correctly diagnosed. SML is able to automatically choose fundamental images (on average 5 or 9 images for each patient) in the diagnosis classification and achieve the above mentioned accuracy. The computational time is 75 seconds for diagnosing 113 patients in a laptop with standard CPU running environment. Factors that influenced high performance of a well-designed integration of spectral learning and machine learning included: 1) use of eigenvectors corresponding to several of the largest eigenvalues of sample covariance matrix (spike eigenvectors) to choose input attributes in classification training, taking into account only the fundamental information of the raw images with less noise; 2) removal of irrelevant pixels based on mean-level spectral test to lower the challenges of memory capacity and enhance computational efficiency while maintaining superior classification accuracy; 3) adoption of state-of-the-art machine learning classification, gradient boosting and random forest. Our methodology showcases practical utility and improved accuracy of image diagnosis in pancreatic mass screening in the era of AI., Comment: 17 pages, 3 figures

Rule-based dialogue management is still the most popular solution for industrial task-oriented dialogue systems for their interpretablility. However, it is hard for developers to maintain the dialogue logic when the scenarios get more and more complex. On the other hand, data-driven dialogue systems, usually with end-to-end structures, are popular in academic research and easier to deal with complex conversations, but such methods require plenty of training data and the behaviors are less interpretable. In this paper, we propose a method to leverages the strength of both rule-based and data-driven dialogue managers (DM). We firstly introduce the DM of Carina Dialog System (CDS, an advanced industrial dialogue system built by Microsoft). Then we propose the "model-trigger" design to make the DM trainable thus scalable to scenario changes. Furthermore, we integrate pre-trained models and empower the DM with few-shot capability. The experimental results demonstrate the effectiveness and strong few-shot capability of our method., Comment: AAAI 2021 Demo Paper

The triple-based knowledge in large-scale knowledge bases is most likely lacking in structural logic and problematic of conducting knowledge hierarchy. In this paper, we introduce the concept of metaknowledge to knowledge engineering research for the purpose of structural knowledge construction. Therefore, the Metaknowledge Extraction Framework and Document Structure Tree model are presented to extract and organize metaknowledge elements (titles, authors, abstracts, sections, paragraphs, etc.), so that it is feasible to extract the structural knowledge from multi-modal documents. Experiment results have proved the effectiveness of metaknowledge elements extraction by our framework. Meanwhile, detailed examples are given to demonstrate what exactly metaknowledge is and how to generate it. At the end of this paper, we propose and analyze the task flow of metaknowledge applications and the associations between knowledge and metaknowledge.

The paths leading to future networks are pointing towards a data-driven paradigm to better cater to the explosive growth of mobile services as well as the increasing heterogeneity of mobile devices, many of which generate and consume large volumes and variety of data. These paths are also hampered by significant challenges in terms of security, privacy, services provisioning, and network management. Blockchain, which is a technology for building distributed ledgers that provide an immutable log of transactions recorded in a distributed network, has become prominent recently as the underlying technology of cryptocurrencies and is revolutionizing data storage and processing in computer network systems. For future data-driven networks (DDNs), blockchain is considered as a promising solution to enable the secure storage, sharing, and analytics of data, privacy protection for users, robust, trustworthy network control, and decentralized routing and resource managements. However, many important challenges and open issues remain to be addressed before blockchain can be deployed widely to enable future DDNs. In this article, we present a survey on the existing research works on the application of blockchain technologies in computer networks, and identify challenges and potential solutions in the applications of blockchains in future DDNs. We identify application scenarios in which future blockchain-empowered DDNs could improve the efficiency and security, and generally the effectiveness of network services.

Recently, deep learning based image deblurring has been well developed. However, exploiting the detailed image features in a deep learning framework always requires a mass of parameters, which inevitably makes the network suffer from high computational burden. To solve this problem, we propose a lightweight multiinformation fusion network (LMFN) for image deblurring. The proposed LMFN is designed as an encoder-decoder architecture. In the encoding stage, the image feature is reduced to various smallscale spaces for multi-scale information extraction and fusion without a large amount of information loss. Then, a distillation network is used in the decoding stage, which allows the network benefit the most from residual learning while remaining sufficiently lightweight. Meanwhile, an information fusion strategy between distillation modules and feature channels is also carried out by attention mechanism. Through fusing different information in the proposed approach, our network can achieve state-of-the-art image deblurring result with smaller number of parameters and outperforms existing methods in model complexity.