Zihan Fang

Dynamic Uncertainty-Aware Multimodal Fusion for Outdoor Health Monitoring

Zihan Fang, Lin Zheng, Senkang Hu, Yihang Tao et autres

Outdoor health monitoring is essential to detect early abnormal health status for safeguarding human health and safety. Conventional outdoor monitoring relies on static multimodal deep learning frameworks, which requires extensive data training from scratch and fails to capture subtle health status changes. …

hk (code pays fourni par la source)

DPO-F+: Aligning Code Repair Feedback with Developers' Preferences

Zihan Fang, Yifan Zhang, Yueke Zhang, Kevin Leach et autres

Large Language Models (LLMs) are increasingly used in software engineering tasks, especially code repair. However, developers often struggle to interpret model outputs, limiting effective human--AI teaming, where humans and AI work toward a shared objective. Prior work mainly optimizes generated code, giving …

LEO-Split: A Semi-Supervised Split Learning Framework Over LEO Satellite Networks

Zhe Chen, Zihan Fang, Xianhao Chen, Yue Gao et autres

Recently, the increasing deployment of LEO satellite systems has enabled various space analytics (e.g., crop and climate monitoring), which heavily relies on the advancements in deep learning (DL). However, the intermittent connectivity between LEO satellites and ground station (GS) significantly hinders the …

cn, hk, sg (code pays fourni par la source)

SL-ACC: A Communication-Efficient Split Learning Framework with Adaptive Channel-wise Compression

Zehang Lin, Zheng Lin, Yang Miao, Jianhao Huang et autres

The increasing complexity of neural networks poses a significant barrier to the deployment of distributed machine learning (ML) on resource-constrained devices, such as federated learning (FL). Split learning (SL) offers a promising solution by offloading the primary computing load from edge devices …

SatFed: A Resource-Efficient LEO-Satellite-Assisted Heterogeneous Federated Learning Framework

Y. Zhang, Zheng Lin, Zhe Chen, Zihan Fang et autres

Traditional federated learning (FL) frameworks rely heavily on terrestrial networks, whose coverage limitations and increasing bandwidth congestion significantly hinder model convergence. Fortunately, the advancement of low-Earth-orbit (LEO) satellite networks offers promising new communication avenues to augment traditional terrestrial FL. Despite this potential, …

cn, hk (code pays fourni par la source)

Task-Aware Parameter-Efficient Fine-Tuning of Large Pre-Trained Models at the Edge

Senkang Hu, Yanan Ma, Yihang Tao, Zhengru Fang et autres

Large language models (LLMs) have achieved remarkable success in various tasks, such as decision-making, reasoning, and question answering. They have been widely used in edge devices. However, fine-tuning LLMs to specific tasks at the edge is challenging due to the high computational …

LEO-Split: A Semi-Supervised Split Learning Framework over LEO Satellite Networks

Zheng Lin, Yuxin Zhang, Zhe Chen, Zihan Fang et autres

Recently, the increasing deployment of LEO satellite systems has enabled various space analytics (e.g., crop and climate monitoring), which heavily relies on the advancements in deep learning (DL). However, the intermittent connectivity between LEO satellites and ground station (GS) significantly hinders the …

ESL-LEO: An Efficient Split Learning Framework over LEO Satellite Networks

Zheng Lin, Y. Zhang, Zhe Chen, Zihan Fang et autres

cn, hk (code pays fourni par la source)

FedSN: A Federated Learning Framework over Heterogeneous LEO Satellite Networks

Zheng Lin, Zhe Chen, Zihan Fang, Xianhao Chen et autres

Recently, a large number of Low Earth Orbit (LEO) satellites have been launched and deployed successfully in space. Due to multimodal sensors equipped by the LEO satellites, they serve not only for communications but also for various machine learning applications. However, a …

SatFed: A Resource-Efficient LEO Satellite-Assisted Heterogeneous Federated Learning Framework

Yuxin Zhang, Zheng Lin, Zhe Chen, Zihan Fang et autres

Traditional federated learning (FL) frameworks rely heavily on terrestrial networks, where coverage limitations and increasing bandwidth congestion significantly hinder model convergence. Fortunately, the advancement of low-Earth orbit (LEO) satellite networks offers promising new communication avenues to augment traditional terrestrial FL. Despite this …

Seeing beyond words: Visualizing autism spectrum disorder biomarker insights

Xinyue Xie, Rongyi Zhou, Zihan Fang, Yongting Zhang et autres

Objective: This study employs bibliometric and visual analysis to elucidate global research trends in Autism Spectrum Disorder (ASD) biomarkers, identify critical research focal points, and discuss the potential integration of diverse biomarker modalities for precise ASD assessment. Methods: A comprehensive bibliometric analysis …

cn (code pays fourni par la source)