Relational Foundation Models for Enterprise Data [Jure Leskovec] – 768

In this episode, Jure Leskovec, co-founder and chief scientist at Kumo and professor of computer science at Stanford, joins us to explore two fronts of his work: AI for science and relational deep learning. We begin with AI Virtual Cell, a multiscale effort to learn data-driven representations from proteins to cells to patients using single-cell RNA-seq data, protein language models like ESM, and structure models like AlphaFold—without hand-encoding biology. Jure then dives into relational deep learning, reframing enterprise databases as graphs and training neural networks directly on raw multi-table data. He explains Kumo’s Relational Foundation Model (RFM2), which performs in-context learning over subgraphs to make accurate predictions on new databases and tasks with no training, and how this approach benchmarks against RelBench and other multi-table datasets. We also discuss real-world deployments at companies like Reddit, DoorDash, and Coinbase, explainability via attention over tables and columns, integration with agentic systems, deployment options, and practical limitations.

🗒️ For the full list of resources for this episode, visit the show notes page: https://twimlai.com/go/768.

🔔 Subscribe to our channel for more great content just like this: https://youtube.com/twimlai?sub_confirmation=1

🗣️ CONNECT WITH US!
===============================
Subscribe to the TWIML AI Podcast: https://twimlai.com/podcast/twimlai/
Follow us on Twitter: https://twitter.com/twimlai
Follow us on LinkedIn: https://www.linkedin.com/company/twimlai/
Join our Slack Community: https://twimlai.com/community/
Subscribe to our newsletter: https://twimlai.com/newsletter/
Want to get in touch? Send us a message: https://twimlai.com/contact/

📖 CHAPTERS
===============================
00:00 – Introduction
01:53 – AI Virtual Cell project
07:39 – Computational social science and relational data
10:01 – Relational deep learning
18:31 – Use cases
20:44 – Limitations of single-table ML
23:52 – Multi-table relational benchmarks
27:25 – Kumo and RFM2
36:50 – Relational graph transformer and relational transformer papers
39:35 – Data requirements and in-context learning
42:53 – Benchmark results and real-world performance
44:59 – Cold start problems
45:51 – Kumo deployment use cases
49:38 – Explainability for relational foundation models
53:55 – Architectural bottlenecks and ideal use cases
59:02 – Spectrum on post-training and fine-tuning
01:01:25 – Future directions

🔗 LINKS & RESOURCES
===============================
Kumo.AI – https://kumo.ai/
Relational Graph Transformer – https://arxiv.org/abs/2505.10960
Relational Deep Learning: Challenges, Foundations and Next-Generation Architectures – https://arxiv.org/abs/2506.16654
KumoRFM-2: Scaling Foundation Models for Relational Learning – https://arxiv.org/abs/2604.12596
Transformers for Tabular Data at Capital One with Bayan Bruss – 591 – https://twimlai.com/podcast/twimlai/transformers-for-tabular-data-at-capital-one

📸 Camera: https://amzn.to/3TQ3zsg
🎙️Microphone: https://amzn.to/3t5zXeV
🚦Lights: https://amzn.to/3TQlX49
🎛️ Audio Interface: https://amzn.to/3TVFAIq
🎚️ Stream Deck: https://amzn.to/3zzm7F5