Understanding at scale

Semantic Knowledge Graphs

As the volume of human knowledge continues to explode, we need better tools for organizing, connecting, and reasoning about information. Our research develops semantic knowledge graph techniques that enable both humans and AI systems to navigate complex knowledge landscapes.

We're working on methods for automated knowledge extraction, relationship inference, and temporal evolution of knowledge. This includes graph neural networks that can reason over knowledge structures, and interfaces that help users explore and contribute to growing knowledge bases.

Research Topics

Graph neural networks for knowledge reasoningAutomated ontology construction and maintenanceCross-domain knowledge transfer and alignmentTemporal knowledge evolution and versioningCollaborative knowledge curation interfacesKnowledge graph question answering

Key Findings

Research in progress, findings coming soon

Future Directions

Scaling knowledge graphs to web-scale while maintaining quality
Integrating knowledge graphs with large language models
Developing tools for community-driven knowledge curation

Interested in this research area?

GET IN TOUCH