Search

Different from the flow semantics of natural languages, programming languages are inherently rigid in structure and grammar. Existing fine-tuning methodologies for code vulnerability detection generally treat code as long text sequences, stripping away structural elements such as newlines ('/n') and whitespace. However, this approach inadvertently results in the loss of crucial structural information, diminishing the distinct characteristics of code and impairing the accuracy of vulnerability detection. To address these challenges, we propose a novel network architecture method based on pre-trained code models, which incorporates structural information awareness. We propose an enhanced code text processing workflow that retains structural elements prior to modeling. This refinement allows the model to retain and exploit line-level structural information and semantic information during the modeling process. Furthermore, we introduce a new network architecture, the Code Structure-Aware Network through Line-level Semantic Learning (CSLS), which integrates three key components: global vulnerability awareness, line-structural awareness, and sensitive-line awareness. We have conducted comprehensive experiments using vulnerability detection datasets from real-world projects. Extensive experiments were conducted on vulnerability detection datasets derived from real-world projects. The results demonstrate that our new code pre-processing flow significantly improves existing baselines (e.g., a 3\% accuracy improvement on the Devign dataset when applied to popular models such as CoderBert and UniXcoder). The proposed network architecture also demonstrates superior accuracy in detecting vulnerabilities, surpassing newly established benchmarks. These findings underscore the importance of structural information in enhancing the efficacy of code vulnerability detection models.

Large Language Models (LLMs) have strong capabilities in code comprehension, but fine-tuning costs and semantic alignment issues limit their project-specific optimization; conversely, code models such CodeBERT are easy to fine-tune, but it is often difficult to learn vulnerability semantics from complex code languages. To address these challenges, this paper introduces the Multi-Model Collaborative Vulnerability Detection approach (M2CVD) that leverages the strong capability of analyzing vulnerability semantics from LLMs to improve the detection accuracy of code models. M2CVD employs a novel collaborative process: first enhancing the quality of vulnerability semantic description produced by LLMs through the understanding of project code by code models, and then using these improved vulnerability semantic description to boost the detection accuracy of code models. We demonstrated M2CVD's effectiveness on two real-world datasets, where M2CVD significantly outperformed the baseline. In addition, we demonstrate that the M2CVD collaborative method can extend to other different LLMs and code models to improve their accuracy in vulnerability detection tasks.

This paper presents a novel point cloud compression method COT-PCC by formulating the task as a constrained optimal transport (COT) problem. COT-PCC takes the bitrate of compressed features as an extra constraint of optimal transport (OT) which learns the distribution transformation between original and reconstructed points. Specifically, the formulated COT is implemented with a generative adversarial network (GAN) and a bitrate loss for training. The discriminator measures the Wasserstein distance between input and reconstructed points, and a generator calculates the optimal mapping between distributions of input and reconstructed point cloud. Moreover, we introduce a learnable sampling module for downsampling in the compression procedure. Extensive results on both sparse and dense point cloud datasets demonstrate that COT-PCC outperforms state-of-the-art methods in terms of both CD and PSNR metrics. Source codes are available at \url{https://github.com/cognaclee/PCC-COT}.

Individual objects, whether users or services, within a specific region often exhibit similar network states due to their shared origin from the same city or autonomous system (AS). Despite this regional network similarity, many existing techniques overlook its potential, resulting in subpar performance arising from challenges such as data sparsity and label imbalance. In this paper, we introduce the regional-based dual latent state learning network(R2SL), a novel deep learning framework designed to overcome the pitfalls of traditional individual object-based prediction techniques in Quality of Service (QoS) prediction. Unlike its predecessors, R2SL captures the nuances of regional network behavior by deriving two distinct regional network latent states: the city-network latent state and the AS-network latent state. These states are constructed utilizing aggregated data from common regions rather than individual object data. Furthermore, R2SL adopts an enhanced Huber loss function that adjusts its linear loss component, providing a remedy for prevalent label imbalance issues. To cap off the prediction process, a multi-scale perception network is leveraged to interpret the integrated feature map, a fusion of regional network latent features and other pertinent information, ultimately accomplishing the QoS prediction. Through rigorous testing on real-world QoS datasets, R2SL demonstrates superior performance compared to prevailing state-of-the-art methods. Our R2SL approach ushers in an innovative avenue for precise QoS predictions by fully harnessing the regional network similarities inherent in objects.

The development of high-performance sodium (Na) ion batteries requires improved electrode materials. The energy and power densities of Na superionic conductor (NaSICON) electrode materials are promising for large-scale energy storage applications. However, several practical issues limit the full utilization of the theoretical energy densities of NaSICON electrodes. A pressing challenge lies in the limited sodium extraction in low Na content NaSICONs, e.g., $\rm Na_1V^{IV}V^{IV}(PO_4)_3 \leftrightarrow V^{V}V^{IV}(PO_4)_3 + 1e^- + 1Na^+$. Hence, it is important to quantify the Na-ion mobility in a broad range of NaSICON electrodes. Using a kinetic Monte Carlo approach bearing the accuracy of first-principles calculations, we elucidate the variability of Na-ion transport vs. Na content in three important NaSICON electrodes, Na$_{\rm x}$Ti$_{2}$(PO$_{4}$)$_{3}$, Na$_{\rm x}$V$_{2}$(PO$_{4}$)$_{3}$, and Na$_{\rm x}$Cr$_{2}$(PO$_{4}$)$_{3}$. Our study suggests that Na$^+$ transport in NaSICON electrodes is almost entirely determined by the local electrostatic and chemical environment set by the transition metal and the polyanionic scaffold. The competition with the ordering-disordering phenomena of Na-vacancies also plays a role in influencing Na-transport. We link the variations in the Na$^+$ kinetic properties by analyzing the competition of ligand field stabilization transition metal ions and their ionic radii. We interpret the limited Na-extraction at $x = 1$ observed experimentally by gaining insights into the local Na-vacancy interplay. We propose that targeted chemical substitutions of transition metals disrupting local charge arrangements will be critical to reducing the occurrence of strong Na$^+$-vacancy orderings at low Na concentrations, thus, expanding the accessible capacities of these electrode materials.

Quality of Service (QoS) prediction is an essential task in recommendation systems, where accurately predicting unknown QoS values can improve user satisfaction. However, existing QoS prediction techniques may perform poorly in the presence of noise data, such as fake location information or virtual gateways. In this paper, we propose the Probabilistic Deep Supervision Network (PDS-Net), a novel framework for QoS prediction that addresses this issue. PDS-Net utilizes a Gaussian-based probabilistic space to supervise intermediate layers and learns probability spaces for both known features and true labels. Moreover, PDS-Net employs a condition-based multitasking loss function to identify objects with noise data and applies supervision directly to deep features sampled from the probability space by optimizing the Kullback-Leibler distance between the probability space of these objects and the real-label probability space. Thus, PDS-Net effectively reduces errors resulting from the propagation of corrupted data, leading to more accurate QoS predictions. Experimental evaluations on two real-world QoS datasets demonstrate that the proposed PDS-Net outperforms state-of-the-art baselines, validating the effectiveness of our approach.

Sodium (Na) vanadium (V) fluorophosphate $\mathrm{Na_xV_2(PO_4)_2F_3}$ (NVPF) is a highly attractive intercalation electrode material due to its high operation voltage, large capacity, and long cycle life. However, several issues limit the full utilization of NVPF's energy density: 1) the high voltage plateau associated with extracting the "third" Na ion in the reaction N$_1$VPF $\rightarrow$ VPF (~4.9 V vs Na/Na$^+$) appears above the electrochemical stability window of most practical electrolytes (~4.5 V); 2) a sudden drop in Na-ion diffusivity is observed near composition $\mathrm{Na_1V_2(PO_4)_2F_3}$. Therefore, it is important to investigate the potential substitution of V by other transition metals in NVPF derivatives, which can access the extraction of the third Na-ion. In this work, we investigate the partial substitution of V with molybdenum (Mo), niobium (Nb), or tungsten (W) in NVPF to improve its energy density. We examine the structural and electrochemical behaviors of $\mathrm{Na_xV_{2-y}Mo_y(PO_4)_2F_3}$, $\mathrm{Na_xV_{2-y}Nb_y(PO_4)_2F_3}$, and $\mathrm{Na_xW_{2}(PO_4)_2F_3}$ across the whole Na composition region of 0 $\leq$ x $\leq$ 4, and at various transition metal substitution levels, namely, y=0.5, 1.0, 1.5, 2.0 for Mo, and y=1.0, 2.0 for Nb. We find that partial substitution of 50% V by Mo in NVPF reduces the voltage plateau for extracting the third Na ion by 0.6 Volts, which enables further Na extraction from $\mathrm{Na_1Mo_{2}(PO_4)_2F_3}$ and increases the theoretical gravimetric capacity from ~128 to ~174 mAh/g. Analysis of the migration barriers for Na-ions in $\mathrm{Na_xVMo(PO_4)_2F_3}$ unveils improved kinetic properties over NVPF. The proposed $\mathrm{Na_xVMo(PO_4)_2F_3}$ material provides an optimal gravimetric energy density of ~577.3 Wh/kg versus ~507 Wh/kg for the pristine NVPF, which amounts to an increase of ~13.9%.

We elucidate the thermodynamics of sodium (Na) intercalation into the sodium super-ionic conductor (NaSICON)-type electrode, Na$_x$V$_2$(PO$_4$)$_3$, for promising Na-ion batteries with high-power density. This is the first report of a computational temperature-composition phase diagram of the NaSICON-type electrode Na$_x$V$_2$(PO$_4$)$_3$. We identify two thermodynamically stable phases at the compositions Na$_2$V$_2$(PO$_4$)$_3$ and Na$_{3.5}$V$_2$(PO$_4$)$_3$, and their structural features are described for the first time based on our computational analysis. We unveil the crystal-structure and the electronic-structure origins of the ground-state compositions associated with specific Na/vacancy arrangements, which are driven by charge orderings on the vanadium sites. These results are significant for the optimization of high-energy and power densities electrodes for sustainable Na-ion batteries, Comment: 5 figures

Real-time 3D human pose estimation is crucial for human-computer interaction. It is cheap and practical to estimate 3D human pose only from monocular video. However, recent bone splicing based 3D human pose estimation method brings about the problem of cumulative error. In this paper, the concept of virtual bones is proposed to solve such a challenge. The virtual bones are imaginary bones between non-adjacent joints. They do not exist in reality, but they bring new loop constraints for the estimation of 3D human joints. The proposed network in this paper predicts real bones and virtual bones, simultaneously. The final length of real bones is constrained and learned by the loop constructed by the predicted real bones and virtual bones. Besides, the motion constraints of joints in consecutive frames are considered. The consistency between the 2D projected position displacement predicted by the network and the captured real 2D displacement by the camera is proposed as a new projection consistency loss for the learning of 3D human pose. The experiments on the Human3.6M dataset demonstrate the good performance of the proposed method. Ablation studies demonstrate the effectiveness of the proposed inter-frame projection consistency constraints and intra-frame loop constraints., Comment: 10 pages, 7 figures. Accepted by TCDS 2022

In recent years, the number of online services has grown rapidly, invoke the required services through the cloud platform has become the primary trend. How to help users choose and recommend high-quality services among huge amounts of unused services has become a hot issue in research. Among the existing QoS prediction methods, the collaborative filtering(CF) method can only learn low-dimensional linear characteristics, and its effect is limited by sparse data. Although existing deep learning methods could capture high-dimensional nonlinear features better, most of them only use the single feature of identity, and the problem of network deepening gradient disappearance is serious, so the effect of QoS prediction is unsatisfactory. To address these problems, we propose an advanced probability distribution and location-aware ResNet approach for QoS Prediction(PLRes). This approach considers the historical invocations probability distribution and location characteristics of users and services, and first use the ResNet in QoS prediction to reuses the features, which alleviates the problems of gradient disappearance and model degradation. A series of experiments are conducted on a real-world web service dataset WS-DREAM. The results indicate that PLRes model is effective for QoS prediction and at the density of 5%-30%, which means the data is sparse, it significantly outperforms a state-of-the-art approach LDCF by 12.35%-15.37% in terms of MAE.