Research & Publications | Xiaobing Chen

My research is centered on making AI systems collaborative, efficient, and resource-aware. Below is a curated selection of my work, categorized by core research themes.

🌐 Federated & Distributed Learning

Optimizing how decentralized devices work together to train high-performance models under constraints.

• DualGFL: Federated Learning with a Dual-Level Coalition-Auction Game
  AAAI 2025 Link
• Joint Device and Training Scheduling for Wireless Federated Learning
  IEEE Internet of Things Journal 2025 Link
• Cost-Effective Federated Learning: A Unified Approach to Device and Training Scheduling
  IEEE International Conference on Communications (ICC) 2024 Link
• Client Selection for Wireless Federated Learning With Data and Latency Heterogeneity
  IEEE Internet of Things Journal 2024 Link

🧠 Foundation Models & Multi-Modal Reasoning

Bridging the gap between Large Language Models and numerical or temporal data domains.

• Enhancing Time Series Forecasting via Multi-Level Text Alignment with Large Language Models
  DASFAA 2025 Link
• DPAL-BERT: A Faster and Lighter Question Answering Model
  CMES-Computer Modeling in Engineering & Sciences 2024 Link
• Dynamic Multi-Granularity Translation System: DAG-Structured Multi-Granularity Representation and Self-Attention
  Systems 2024 Link

👁️ Computer Vision & Intelligent Perception

Efficient architectures for detection, estimation, and visual reasoning.

• A Novel Self-Supervised Learning Network for Binocular Disparity Estimation
  CMES-Computer Modeling in Engineering & Sciences 2025 Link
• CGBi_YOLO: Lightweight Land Target Detection Network
  Land 2024 Link
• Convolution Transformer for Image Feature Extraction
  CMES-Computer Modeling in Engineering & Sciences 2024 Link
• Knowledge-base Graph Embedding Module Design for Visual Question Answering Model
  Pattern Recognition 2021 Link

⌚ Efficient Systems & Sensor Intelligence

Applying AI to real-world sensing and control problems.

• Temporal Contrastive Learning for Sensor-based Human Activity Recognition: A Self-Supervised Approach
  IEEE Sensors Journal 2025 Link
• Deep Learning-Based Geomorphic Feature Identification in Dredge Pit Marine Environment
  Journal of Marine Science and Engineering 2024 Link
• Temporal Evolution Characteristics of PM2.5 Concentration Based on Continuous Wavelet Transform
  Science of The Total Environment 2020 Link