Talks

Tutorials

  • Tutorial at ChineseCSCW 2022: knowledge Tracing learning, Nov., 2022. [Website&Slides] [More]

Invited talks

  • knowledge Tracing learning models, at ChineseCSCW 2022, Taiyuan, China. Nov 2022.
  • Few-shot Learning for Quality Detection of Logistical Parcels, at ISCTech 2023, Qingdao, China, July 2023
  • NCDFSA: Neural Cognitive Diagnostic Focusing on Students’ Attention to Knowledge Concepts, at 2023 IEEE International Conference on Systems, Man, and Cybernetics (SMC), Oahu, HI, USA, October 2023, (Quanlong Guan and Guoxiong Wei).
  • A Practical Framework of Blockchain in IoT Information Management,at 2023 IEEE International Conference on Systems, Man, and Cybernetics (SMC), Oahu, HI, USA, October 2023,(Quanlong Guan and Jiawei Lei).
  • Exploiting the Potential Anomaly Detection in Automobile Safety Data with Multi-type Neural Network,at 20th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services (MobiQuitous 2023), Melbourne, Australia, November 2023,(Quanlong Guan and Tian Zhang).
  • Efficient Parcel Damage Detection via Faster R-CNN: A Deep Learning Approach for Logistical Parcels’ Automated Inspection, at EAI MobiQuitious 2023, Melbourne, Australia, Nov 2023,(Quanlong Guan and Zhi Chen).
  • A Retrospect to Multi-prompt Learning across Vision and Language, at ICCV 2023, Paris, France, October 2023.
  • KG4Ex: An Explainable Knowledge Graph-Based Approach for Exercise Recommendation, at CIKM ‘23: Proceedings of the 32nd ACM International Conference on Information and Knowledge Management, Birmingham, United Kingdom, October, 2023, (Quanlong Guan and Xinghe Cheng).
  • Deformation and Penetration Hybrid Detection-Net for Parcels Inspection in Industrial Supply Chain, at ICASSP 2024, Seoul, Korea, April 2024,(Accepted).
  • YTCNet: A Real-time Algorithm for Parcel Damage Detection with Rich Features and Attention, at CSCWD 2024, Tianjin, China, May 2024,(Accepted).
Hit Counter