Overview

One paper, “Interaction Graph Anomaly Detection via Semantic Edge Augmentation with Large Language Models,” has been accepted to the 23rd IEEE International Conference on Mobile Ad-Hoc and Smart Systems (IEEE MASS 2026).

About the Work

This work investigates anomaly detection in interaction graphs by leveraging large language models to enrich graph structure with semantic information. The proposed approach introduces semantic edge augmentation to complement observed interactions, enabling the model to capture relationships that may not be directly represented in the original graph and improving the detection of anomalous interaction patterns.

Key Highlights

• Introduces LLM-based semantic information into interaction graph anomaly detection
• Uses semantic edge augmentation to enrich observed graph structures
• Combines structural interaction patterns with semantic relationships for anomaly detection
• Accepted to IEEE MASS 2026

Learn More

The paper will appear in the proceedings of the 23rd IEEE International Conference on Mobile Ad-Hoc and Smart Systems (IEEE MASS 2026).