The ACM Web Conference (WWW) is the top-tier academic conference on web technologies. Our lab had three papers presented in Sydney, Australia:  

• Jian Lang presented "Biting Off More Than You Can Detect: Retrieval-Augmented Multimodal Experts for Short Video Hate Detection."

• Rongpei Hong presented "Following Clues, Approaching the Truth: Explainable Micro-Video Rumor Detection via Chain-of-Thought Reasoning."

• Xinglin Lian presented "Facing Anomalies Head-On: Network Traffic Anomaly Detection via Uncertainty-Inspired Inter-Sample Differences."

At the 32nd ACM International Conference on Information and Knowledge Management, Birmingham, UK, our lab presents four papers, two oral presentations and two posters. 

• Ce Li: Simplifying Temporal Heterogeneous Network for Continuous-Time Link prediction

• Leyuan Liu: Towards Trustworthy Rumor Detection with Interpretable Graph Structural Learning

• Zhangtao Cheng: Enhancing Information Diffusion Prediction with Self-Supervised Disentangled User and Cascade Representations

• Xovee Xu: MDCC: A Multimodal Dynamic Dataset for Donation-based Crowdfunding Campaigns

At the 24th IEEE International Conference on Mobile Data Management in Singapore, Wenxin Tai showcased "RIPGeo: Robust Street-Level IP Geolocation," a novel approach to IP geolocation using a graph neural network model. 

This research addresses the challenge of accurately locating IP addresses at the street level, crucial for mobile computing and spatial data management. RIPGeo introduces innovative training strategies and a multi-task learning framework, significantly enhancing accuracy and stability in IP geolocation. This presentation highlights a breakthrough in geolocation technology, marking a substantial advancement in the field.

This paper also wins the Best Student Paper Award!

At the 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing in Rhodes Island, Greece, Xovee presents his paper "Diffusion Probabilistic Modeling for Fine-Grained Urban Traffic Flow Inference with Relaxed Structural Constraint," which proposes a new way to infer fine-grained urban flow maps using coarce-grained ones.