National University of Defense Technology

Cation‐Framework‐Preserving Oxygen‐Vacancy Engineering for Broadband Infrared Emission in High‐Entropy Spinel Oxides

ABSTRACT High‐temperature radiative ceramics require broadband infrared emission and stable structures, but the defects that enhance emission can also weaken oxide lattices. Here, we regulate oxygen vacancies in (MnCrFeNiCu) 3 O 4 , a high‐entropy spinel oxide (HEO), by vacuum an...

LOAD: Enabling Low-Bit Group Quantization for Vision Transformers via Adaptive Data Formats

Vision Transformers (ViTs) achieve state-of-the-art performance across various vision tasks but pose significant deployment challenges due to their high computational complexity and memory footprint. Post-training quantization (PTQ) offers an efficient, retraining-free compressio...

The Role of Supply Chain Innovation Support Policies as a Determinant of the Interdependence Between Digitalization and Greening

This study examines the association between supply chain innovation support policies and the dynamic interdependence between firms’ digitalization and greening. Using a panel of 320 A-share listed firms from 11 provinces in the Yangtze River Basin over 2010–2024, we construct a d...

Constrained Value Decomposition for Communication-Gated Task Allocation Under Asynchronous UAV Swarm Communication

Consensus-based distributed task allocation in UAV swarms resolves conflicts through repeated neighbor communication, saturating the ad hoc channel with packet loss and latency; reducing transmission frequency without sacrificing allocation quality is the key challenge. Prior com...

SDU-G2A: Self-Derived Uncertainty Guidance for Ground-to-Aerial Remote-Sensing Image Synthesis with Probabilistic BEV Footprints

Urban planning and land-use monitoring require not only semantic land-cover maps, but also image-like overhead observations that preserve the spatial organization of urban scenes. Conventional remote-sensing segmentation and image-generation methods usually operate by obtaining a...

BASG: A Behavior-Augmented Semantic Generalization Framework for Zero-Shot Cross-Domain Sequential Recommendation

Training sequential recommenders for interaction-disjoint target domains is challenging: no target interaction data are available for training, model selection, or calibration. Semantic item representations support transfer across non-overlapping catalogs but do not encode query-...

LSR: Eliminating Loop Control Overhead for Energy-Efficient Processors

Loops constitute a small fraction of static code but dominate dynamic execution time and control overhead, posing significant challenges for energy-efficient mobile and embedded processors. This paper introduces Loop Semantic Recognition (LSR) , a microarchitectural technique tha...

BiTTP: Bidirectional Consistency-Enhanced Framework for Low-Resource APT Tactic and Technique Attribution

Attributing advanced persistent threat (APT) activity to tactics and techniques in the MITRE ATT&CK knowledge base supports threat hunting and incident response, but annotated attack data are scarce. We present BiTTP, a consistency-verified self-training framework for low-resourc...

From granularity to consensus: Mitigating sample defects for Remote Sensing Image–Text Retrieval

Remote Sensing Image–Text Retrieval (RSITR) aims to retrieve remote sensing images or text descriptions when only one modality is available. Despite recent progress, existing RSITR methods often suffer from sample defects, which cause them to mistakenly treat semantically related...

Differential-linear cryptanalysis against ChaCha based on automated tools

Abstract ChaCha is a widely deployed ARX stream cipher used in modern secure-communication protocols. Reduced-round cryptanalysis assesses its security margin and tests automated methods without implying a threat to the full 20-round cipher. We combine MILP search, GPU correlatio...

Network Topology Obfuscation: Challenges, Solutions, and Comparisons

Attackers infer the network topology to locate critical components and to plan cost-effective attacks with high success rates. Network topology obfuscation (NTO) counters this by exposing a well-designed fake topology that misleads the inference and raises the cost of attack. Exi...

PLUME: Staged Fusion of FY-4A AGRI and Rain-Gauge Observations for Bidirectional Refinement of Short-Duration Heavy-Rainfall Warnings

Local short-duration heavy-rainfall warning requires early assessment of hazardous rainfall within a specified neighborhood. Geostationary meteorological satellites provide frequent, spatially continuous observations of cloud systems and storm evolution, but only indirectly refle...

Distributionally-robust heterogeneous graph neural networks via relation-aware grouping

Out-of-distribution (OOD) generalization remains a key challenge for heterogeneous graph neural networks (HGNNs), which often exhibit substantial performance degradation under real-world distribution shifts. While Group Distributionally Robust Optimization (GroupDRO) has proven e...

Spatial Oxygen-Vacancy Engineering for Enhanced Ferroelectricity in Flexible Hf0.5Zr0.5O2 Devices

HfO2-based ferroelectrics are promising for scalable memories, neuromorphic systems, and flexible electronics, but their performance is governed by oxygen vacancies that affect phase stability, leakage, wake-up behavior, and domain switching. Existing strategies mainly focus on t...

Small bias, large errors: a dual-perspective evaluation of numerical model near-surface wind direction over beijing megacity

Abstract Accurate surface wind fields are essential for understanding moisture transport and heavy rainfall forecasts, yet the ability of kilometer-scale numerical models to capture wind direction over heterogeneous surfaces remains poorly quantified. Using the EARS1km reanalysis...

Nanocomposite‐Enhanced Iontronic Tactile Sensor With High Sensitivity and Wide‐Range Linearity for Underwater Robotic Manipulation

ABSTRACT Hydrostatic preload increases response nonlinearity and limits weak‐force detection in underwater tactile sensing. Here, we present a nanocomposite‐enhanced iontronic tactile sensor (NeITS) that combines high sensitivity with wide‐range linearity through a coupled mechan...

IGPC: Multi-domain multimodal fake news detection via Info-Gini Weight Adaptation and Prototypical Multi-level Contrastive Learning

Multi-domain multimodal fake news detection requires models to integrate textual and visual evidence while generalizing across diverse domains. Existing methods still struggle with multi-level data imbalance, weak class geometry in data-scarce domains, and transfer to held-out ta...

LLM See, LLM Do: Enhancing Repository-Aware Code Generation in LLMs Through Logical Context and Usage Knowledge

Large language models (LLMs) have significantly advanced automatic code generation. Although they have demonstrated remarkable ability in writing standalone code, they still struggle with real-world repository-level coding tasks due to the lack of repository-specific knowledge. E...

Comparing LLM and Human Expert Coding in Qualitative Research: A Case Study on Emotional Intelligence

Large language models (LLMs) are increasingly used for qualitative coding, but their ability to recover human-defined construct boundaries remains unclear. Existing studies rarely distinguish thematic granularity from dimension-level semantic alignment. We compare a 12-dimension...

16-channel ultrashort fiber laser coherent polarization beam combination with active spatio-temporal error control

Coherent beam combination (CBC) has proven to be an effective approach to overcoming the power limitations of single-channel lasers. As a representative implementation of CBC, coherent polarization beam combination (CPBC) becomes increasingly complex with increasing channel count...