> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/sdgm/research/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/sdgm/_mcp/server.

# Research Overview

> Research from Kumo on relational learning, graph models, and predictive AI

Kumo Research advances machine learning for relational and graph-structured data.
Explore practical guides, papers, and systems work from Kumo and the broader graph-learning community.

Explore the collection by area:

#### [Concepts & Models](/research/relational-deep-learning)

Learn the foundations of Predict, GNNs, graph transformers, and relational foundation models.

#### [Data & Benchmarks](/research/relbench)

Explore relational learning benchmarks and synthetic relational data for model research.

#### [Applications](/research/recsys-with-llms)

See how relational and graph learning support recommendations, forecasting, and churn prediction.

#### [Engineering Guides](/research/pyg-practitioners-guide)

Build and optimize graph learning systems with PyG, efficient graph construction, and adaptive sampling.

> Research from Kumo on relational learning, graph models, and predictive AI