arXiv subjects

Yujue Wang

Publications and source records attributed to Yujue Wang.

7 recordsLinked to original sources

arXiv2026

EAGLE: Edge-Aware Graph Learning for Proactive Delivery Delay Prediction in Smart Logistics Networks

Modern logistics networks generate rich operational data streams at every warehouse node and transportation lane -- from order timestamps and routing records to shipping manifests -- yet predicting delivery delays remains predominantly reactive. Existing predictive approaches typically treat this problem either as a tabular classification task, ignoring network topology, or as a time-series anomaly detection task, overlooking the spatial dependencies of the supply chain graph. To bridge this gap, we propose a hybrid deep learning framework for proactive supply chain risk management. The proposed method jointly models temporal order-flow dynamics via a lightweight Transformer patch encoder and inter-hub relational dependencies through an Edge-Aware Graph Attention Network (E-GAT), optimized via a multi-task learning objective. Evaluated on the real-world DataCo Smart Supply Chain dataset, our framework achieves consistent improvements over baseline methods, yielding an F1-score of 0.8762 and an AUC-ROC of 0.9773. Across four independent random seeds, the framework exhibits a cross-seed F1 standard deviation of only 0.0089 -- a 3.8 times improvement over the best ablated variant -- achieving the strongest balance of predictive accuracy and training stability among all evaluated models.

cs.AI↗

arXiv2026

Evaluating Privilege Usage of Agents with Real-World Tools

Equipping LLM agents with real-world tools can substantially improve productivity. However, granting agents autonomy over tool use also transfers the associated privileges to both the agent and the underlying LLM. Improper privilege usage may lead to serious consequences, including information leakage and infrastructure damage. While several benchmarks have been built to study agents' security, they often rely on pre-coded tools and restricted interaction patterns. Such crafted environments differ substantially from the real-world, making it hard to assess agents' security capabilities in critical privilege control and usage. Therefore, we propose GrantBox, a security evaluation sandbox for analyzing agent privilege usage. GrantBox automatically integrates real-world tools and allows LLM agents to invoke genuine privileges, enabling the evaluation of privilege usage under prompt injection attacks. Our results indicate that while LLMs exhibit basic security awareness and can block some direct attacks, they remain vulnerable to more sophisticated attacks, resulting in an average attack success rate of 84.80% in carefully crafted scenarios.

cs.CR↗

arXiv2026

LLM-Grounded Explainable AI for Supply Chain Risk Early Warning via Temporal Graph Attention Networks

Disruptions at critical logistics nodes pose severe risks to global supply chains, yet existing risk prediction systems typically prioritize forecasting accuracy without providing operationally interpretable early warnings. This paper proposes an evidence-grounded framework that jointly performs supply chain bottleneck prediction and faithful natural-language risk explanation by coupling a Temporal Graph Attention Network (TGAT) with a structured large language model (LLM) reasoning module. Using maritime hubs as a representative case study for global supply chain nodes, daily spatial graphs are constructed from Automatic Identification System (AIS) broadcasts, where inter-node interactions are modeled through attention-based message passing. The TGAT predictor captures spatiotemporal risk dynamics, while model-internal evidence -- including feature z-scores and attention-derived neighbor influence -- is transformed into structured prompts that constrain LLM reasoning to verifiable model outputs. To evaluate explanatory reliability, we introduce a directional-consistency validation protocol that quantitatively measures agreement between generated risk narratives and underlying statistical evidence. Experiments on six months of real-world logistics data demonstrate that the proposed framework outperforms baseline models, achieving a test AUC of 0.761, AP of 0.344, and recall of 0.504 under a strict chronological split while producing early warning explanations with 99.6\% directional consistency. Results show that grounding LLM generation in graph-model evidence enables interpretable and auditable risk reporting without sacrificing predictive performance. The framework provides a practical pathway toward operationally deployable explainable AI for supply chain risk early warning and resilience management.

cs.AI↗

arXiv2022

Short-term shock, long-lasting payment: Evidence from the Lushan Earthquake

Abrupt catastrophic events bring business risks into firms. The paper introduces the Great Lushan Earthquake in 2013 in China as an unexpected shock to explore the causal effects on public firms in both the long and short term. DID-PSM methods are conducted to examine the robustness of causal inference. The identifications and estimations indicate that catastrophic shock significantly negatively impacts cash flow liquidity and profitability in the short term. Besides, the practical influences on firms' manufacturing and operation emerge in the treated group. Firms increase non-business expenditures and retained earnings as a financial measure to resist series risk during the shock period. As the long-term payment, the decline in production factors, particularly in employment level and the loss in fixed assets, are permanent. The earthquake's comprehensive interactions are also reflected. The recovery from the disaster would benefit the companies by raising the growth rate of R\&D and enhancing competitiveness through increasing market share, though these effects are temporary. PSM-DID and event study methods are implemented to investigate the general effects of specific strong earthquakes on local public firms nationwide. Consistent with the Lushan Earthquake, the ratio of cash flow to sales dropped drastically and recovered in 3 subsequent semesters. The shock on sales was transitory, only in the current semester.

econ.GN↗

arXiv2022

Compacter networks as a defensive mechanism: How firms clustered during 2015 Financial Crisis in China

The stock market's reaction to the external risk shock is closely related to the cross-shareholding network structure. This paper takes the public information of listed companies in the A-share securities market as the primary sample to study the relationship between the stock return rate, market performance, and network topology before and after China's stock market crash in 2015. Data visualization and empirical analysis demonstrate that the return rate of stocks is related to the company's traditional business ability and the social capital brought by cross-holding. Several heteroscedasticity tests and endogeneity tests with IV are conducted to support the robustness. The structure of the cross-shareholding network experienced upheaval after the shock, even distorting the effects of market value, and assets holding on the return rate. The enterprises in the entire shareholding network are connected more firmly to overcome systematic external risks. The number of enterprise clusters is significantly reduced during the process. Besides, the number of newly established cross-shareholding relationships shows an outbreak, which may explain the rapid maintenance of stability in the financial system. When stable clustering is formed before and after a stock crash (rather than when it occurs), the clustering coefficient of clear clustering still has an apparent positive influence on the return rate of stocks. To sum up, the compacted network may prevent the firms from pursuing aggressive earning before the financial crisis, but would protect firms from suffering relatively high losses during and after the shock.

econ.GN↗

arXiv2022

Nebula Graph: An open source distributed graph database

This paper introduces the recent work of Nebula Graph, an open-source, distributed, scalable, and native graph database. We present a system design trade-off and a comprehensive overview of Nebula Graph internals, including graph data models, partitioning strategies, secondary indexes, optimizer rules, storage-side transactions, graph query languages, observability, graph processing frameworks, and visualization tool-kits. In addition, three sets of large-scale graph b

cs.DB↗

arXiv2019

MagneticSpy: Exploiting Magnetometer in Mobile Devices for Website and Application Fingerprinting

Recent studies have shown that aggregate CPU usage and power consumption traces on smartphones can leak information about applications running on the system or websites visited. In response, access to such data has been blocked for mobile applications starting from Android 8. In this work, we explore a new source of side-channel leakage for this class of attacks. Our method is based on the fact that electromagnetic activity caused by mobile processors leads to noticeable disturbances in magnetic sensor measurements on mobile devices, with the amplitude being proportional to the CPU workload. Therefore, recorded sensor data can be analyzed to reveal information about ongoing activities. The attack works on a number of devices: we evaluated 80 models of modern smartphones and tablets and observed the reaction of the magnetometer to the CPU activity on 56 of them. On selected devices we were able to successfully identify which application has been opened (with up to 90% accuracy) or which web page has been loaded (up to 91% accuracy). The presented side channel poses a significant risk to end users' privacy, as the sensor data can be recorded from native apps or even from web pages without user permissions. Finally, we discuss possible countermeasures to prevent the presented information leakage.

cs.CR↗