Research
Advancing the science of graph learning
Peer-reviewed papers, technical reports, and open-source contributions from the Kumo AI research team.
Latest publications
Explore our latest research across graph neural networks, relational learning, and predictive AI.
Systems & InfrastructureFeb 5, 2026
PLUREL: Synthetic Data unlocks Scaling Laws for Relational Foundation Models
Read more→
Jure Leskovec
Graph TransformersNov 12, 2025
Smarter Adaptive Graph Sampling for More Accurate Graph Learning
Read more→
Vid Kocijan
Systems & InfrastructureNov 6, 2025
Accelerating Graph Construction for Production Graph Transformer Workloads
Read more→
Josh, Min, Jinu
Graph TransformersOct 1, 2025
Large-Scale Training of Graph Transformers - and How the Kumo Training Backend Works
Read more→
Zack Drach
Graph TransformersJun 17, 2025
Time Series Forecasting with Graph Transformers
Read more→
Jan Eric Lenssen, Matthias Fey
Systems & InfrastructureMay 20, 2025
KumoRFM: A Foundation Model for In-Context Learning on Relational Data
Read more→
Systems & InfrastructureApr 28, 2025
Speeding Up Graph Learning Models with PyG and torch.compile
Read more→
Akihiro Nitta, Matthias Fey
Graph TransformersApr 28, 2025
Relational Graph Transformers: A New Frontier in AI for Relational Data
Read more→
Federico, Matthias, Jure
Graph TransformersApr 22, 2025
An Introduction to Graph Transformers
Read more→
Federico, Matthias, Jure
Graph TransformersFeb 5, 2025
Introduction to Relational Deep Learning (RDL)
Read more→
Matthias Fey
Graph TransformersFeb 5, 2025
RelBench: A Benchmark for Deep Learning on Relational Database
Read more→
Matthias Fey
Graph TransformersSep 6, 2024
Improving recommendation systems with LLMs and Graph Transformers
Read more→
Zecheng, Alan, Vid, Matthias, Jure
Ship better AI products in a fraction of the time.
Use Kumo to build and run AI models on your relational data.
