VAST Challenge
Coordinated-view systems built by our student teams for the IEEE VAST Challenge
Every year since 2024 our student teams enter the IEEE VAST Challenge, building visual analytics systems that uncover hidden behavior in large, heterogeneous event logs and knowledge graphs: suspicious vessels in fishing records, unreliable sources in a news corpus, covert coordination in maritime radio communications, partiality in a committee's decisions, or a worm-like instruction propagating through a multi-agent system. The challenge changes every year, but the method does not: coordinated views that combine temporal pattern analysis, network and topic modeling, anomaly detection, and, increasingly, large language models that extract structured evidence from text. The 2025 entry received the Innovative Display of Daily Communication Patterns Award. The challenge is the lab's training ground for visual analytics design and teamwork.
RESEARCHERS
-
Jorge Poco FGV EMAp -
Felipe Moreno-Vera FGV EMAp -
Juanpablo Heredia FGV EMAp -
Mauro Diaz FGV EMAp -
Andres De La Puente FGV EMAp -
Leighton L. Estrada FGV EMAp -
Joel Perca FGV EMAp -
Luis Sante FGV EMAp -
Fabricio Venturim FGV EMAp -
-
João Rulff FGV EMAp
PUBLICATIONS
Interactive Platform for the Visual Analysis of Suspicious Communication Patterns
Andres De La Puente, Juliano Genari, Marcelo Baez, Felipe A. Moreno-Vera, Jorge Poco
ClearWater: Interactive platform for bias investigation
Ana Luiza Nunes, Danilo Cardoso, Tomas Ferranti, Felipe A. Moreno-Vera, Jorge Poco
HopViz: Tracing Anomalous Behaviour Hop-By-Hop Through a Tangled Multi-Agent System
Joel Perca, Luis Sante, Leighton Estrada-Rayme, Mauro Diaz, Juanpablo Heredia, Felipe A. Moreno-Vera, João Rulff, Jorge Poco
AgentAudit: Visual Analytics for Tracing Behavioral Escalation in Multi-Agent Systems
Luis Sante, Mauro Diaz, Leighton Estrada-Rayme, Joel Perca, Juanpablo Heredia, Felipe A. Moreno-Vera, João Rulff, Jorge Poco