NVIDIA Kumo Tabular Sets a New Accuracy-Efficiency Frontier for Tabular Prediction
Hugging Face · 30 Sept 2026, 01:30 AEST
NVIDIA has introduced Kumo Tabular, a new foundation model designed for tabular classification and regression. This model leverages in-context learning to predict labels from tables efficiently.
Key facts
- Tabular data is the backbone of enterprise machine learning.
- For two decades, this work has been done with gradient-boosted trees, and it has worked well.
- Kumo Tabular is a Transformer built around the structure of a table, utilizing column, row and in-context attention.
- Given a table with labeled rows and the rows you want predictions for, Kumo Tabular returns class probabilities or numeric predictions in a single forward pass.
Read aloud
NVIDIA has introduced Kumo Tabular, a new foundation model designed for tabular classification and regression. This model leverages in-context learning to predict labels from tables efficiently. Tabular data is the backbone of enterprise machine learning. For two decades, this work has been done with gradient-boosted trees, and it has worked well. Kumo Tabular is a Transformer built around the structure of a table, utilizing column, row and in-context attention. Given a table with labeled rows and the rows you want predictions for, Kumo Tabular returns class probabilities or numeric predictions in a single forward pass.
Sources in this cluster
- NVIDIA Kumo Tabular Sets a New Accuracy-Efficiency Frontier for Tabular Prediction
Hugging Face · 30 Sept 2026, 01:30 AEST