Huawei Technologies (China)

ChronoSteer: Bridging Large Language Model and Time Series Foundation Model via Synthetic Cross-Modal Alignment Dataset

Conventional forecasting methods are trained end-to-end on unimodal time series, which limits their ability to exploit textual information and undermines their generalization in data-scarce scenarios. Recently, large language models (LLMs) and time series foundation models (TSFMs...

Analytical Prediction of Critical Transitions in Oscillator Networks With Non‐Local Links

ABSTRACT A wide range of dynamical systems can be modeled as oscillators coupled through an interaction network. Such frameworks have been widely used to explore the emergence of collective phenomena across diverse applications. Here, we develop an analytical framework for oscill...

Lingxi: Repository-Level Issue Resolution Framework Enhanced by Procedural Knowledge Guided Scaling

Driven by the advancements of Large Language Models (LLMs), LLM-powered agents are making significant improvements in software engineering tasks, yet struggle with complex, repository-level issue resolution. Existing agent-based methods have two key limitations. First, they lack...

XSearch: Explainable Code Search via Concept-to-Code Alignment

With the emergence of deep learning, semantic code search has been widely adopted in both academia and industry. These approaches embed natural-language queries and code snippets into a shared embedding space and retrieve results based on vector similarity. Despite their strong p...

From Static to Dynamic: Benchmarking Real-World Code Review with MCR-Bench

In real-world software development, code review typically involves iterative interactions between developers and reviewers to improve software quality, making the process costly and time-consuming. Although recent work explores large language models (LLMs) for automated code revi...

Lattice Distortion‐Triggered Cu─O─Ce Synergistic Sites Coupled With Machine Learning for Intelligent Multi‐Pesticide Recognition

ABSTRACT Non‐enzymatic electrochemical sensors have emerged as highly promising platforms for pesticide monitoring, favored for their rapid response, cost‐effectiveness, and robust environmental stability. However, conventional sensors struggle with simultaneous multi‐target reco...

TestART: Improving LLM-based Unit Test ing via Co-evolution of A utomated Generation and R epair I t eration

Unit testing is crucial for detecting bugs in individual program units but consumes time and effort. Recently, large language models (LLMs) have demonstrated remarkable capabilities in generating unit test cases. However, several problems limit their ability to generate high-qual...

Ammonium-Nitrate Ratio and Soil Moisture Jointly Regulate Early-Spring Nitrogen Acquisition Through Root Architectural Plasticity in Apple

Abstract Early-spring nitrogen (N) acquisition in perennial fruit trees represents a critical physiological bottleneck determining vegetative-reproductive balance, as remobilized woody-tissue N supports initial growth but excessive depletion compromises reproductive performance....

Research on the dynamic evolution of competitive and water-film-mediated wear mechanisms of the carbon strips in the pantograph-catenary system at low temperature

Abstract To investigate the evolution of wear mechanisms in carbon strips at low temperature, this study conducted comparative friction tests on copper-impregnated carbon strips at room temperature (25 °C) and low temperature (-20 °C). To replicate operating conditions, the exper...