Xue Li

STAR: Decode-Phase Rescheduling for LLM Inference

Zhibin Wang, Zetao Hong, Xue Li, Shipeng Li et autres

Large Language Model (LLM) inference has emerged as a fundamental paradigm, however, variations in output length cause severe workload imbalance in the decode phase, particularly for long-output reasoning tasks. Existing systems, such as PD disaggregation architectures, rely on static prefill-to-decode scheduling, which …

cn (code pays fourni par la source)

Rethinking multi-stage calibration fusion network and loss function for steel surface defect detection

Xue Li, Shilong Zhao, Yuting Wang

Surface defects in steel severely impact product reliability and service life. However, achieving satisfactory defect detection results is challenging due to the need to balance detection accuracy and efficiency in practical applications, the wide variation in surface defect sizes, and the presence …

cn (code pays fourni par la source)

Correction: Identification and functional characteristics of a novel splice site variant in L1CAM caused X-linked hydrocephalus

Shijie Zhou, Hao Helen Zhang, Xue Li, Quan Chen et autres

• Please read through all the templates before choosing• Pick the most relevant text template(s) from the following page and delete all others.• Edit the text as necessary, ensuring that the original incorrect text is included for the record, please see the …

cn (code pays fourni par la source)

DNMDR: Dynamic networks and multi-view drug representations for safe medication recommendation

Guanlin Liu, Xiaomei Yu, Zihao Liu, Shucheng Liu et autres

cn (code pays fourni par la source)

Kinetic modeling study of autoignition, laminar burning velocity, and emission characteristics of n-heptane/ammonia mixtures

Zhenhua Ji, Zhongqiang Bao, Hao Chen, Xue Li et autres

Introduction The mixed combustion of ammonia (NH 3 ) and high activity fuel in internal combustion engines is an effective way to achieve global decarbonization. However, there remains a lack of detailed high-precision numerical models and basic theoretical research on this topic. …

cn (code pays fourni par la source)

Identification and functional characteristics of a novel splice site variant in L1CAM caused X-linked hydrocephalus

Shijie Zhou, Hao Helen Zhang, Xue Li, Quan Chen et autres

Background The L1CAM gene encodes an axonal glycoprotein belonging to the immunoglobulin supergene family that plays a crucial role in nervous system development. In this study, we reported a novel disease-causing mutation in the 3′ splice site of L1CAM and provided some …

cn (code pays fourni par la source)

Fabrication mechanism and high-temperature properties of bicontinuous Ti2AlN/TiAl composites: Experimental, DFT, and DL investigations

Zhang Zhexuan, Yang Zhou, Xue Li, Wu Yi Chong et autres

sg, cn (code pays fourni par la source)

DNMDR: Dynamic Networks and Multi-view Drug Representations for Safe Medication Recommendation

G. M. Liu, Xiaomei Yu, Zihao Liu, Xue Li et autres

Medication Recommendation (MR) is a promising research topic which booms diverse applications in the healthcare and clinical domains. However, existing methods mainly rely on sequential modeling and static graphs for representation learning, which ignore the dynamic correlations in diverse medical events of …

Study of Interface Adhesion Between Polyurethane and Aggregate Based on Surface Free Energy Theory and Molecular Dynamics Simulation

Zengping Zhang, Jiajun Ji, Xue Li, Zhongbao Shi et autres

In order to eliminate the negative effects caused by traditional pavements, permeable pavements are gradually being used in road construction. In recent years, polyurethane (PU) has been used as a new binder in permeable pavement mixtures. However, compared to traditional pavement mixtures, …

cn (code pays fourni par la source)

FSFDS: Enhancing Flight Sensor Fault Diagnosis via Diffusion and Self-Attention Networks

Tong Sun, Bei Hong, Xue Li, Ping Gao et autres

FSFDS: enhancing flight sensor fault diagnosis via diffusion and self-attention networks

Jiaojiao Gu, Ping Gao, Xue Li, Bei Hong et autres

Aircraft fault diagnosis relies on flight sensor data, with accuracy critical for safe operation. Deep neural networks (DNNs) have shown success in many fields, but their application in fault prediction faces challenges: 1) DNN-generated fault data differs significantly from real faults; 2) …

cn (code pays fourni par la source)

Cf-Cwsad: A Framework for Class Imbalance-Aware Semi-Supervised Learning in Skin Lesion Classification

Dandan Wang, Kang An, Xuan Sun, Haibo Zhang et autres

cn, au (code pays fourni par la source)