---
markdown_generated_at: "2026-10-14"
title: "Geometric Algebra Transformers"
url: "https://www.qualcomm.com/research/artificial-intelligence/papers-with-code/paper-13"
description: "Improve robot perception with Qualcomm's Geometric Algebra Transformer. See Paper 13 details and more from our geometric ai company."
canonicalUrl: "https://www.qualcomm.com/research/artificial-intelligence/papers-with-code/paper-13"
---

PAPER 13

# Geometric Algebra Transformers

Johann Brehmer (Qualcomm AI Research)
Pim de Haan (Qualcomm AI Research)
Sönke Behrends (Qualcomm AI Research)
Taco Cohen (formerly Qualcomm AI Research)

NeurIPS 2023

![](https://s7d1.scene7.com/is/image/dmqualcommprod/paper-13_Geometric%20Algebra%20Transformers?$QC_Responsive$&fmt=png-alpha)

## Summary

Geometric algebra transformers (GATr) is a novel data-efficient architecture model to improve robots’ perception of their environment. GATr considers geometric structures of the physical environment through geometric algebra representations and equivariance. It has the scalability and expressivity of transformers. By embedding various kinds of geometric data into a single geometric algebra, GATr can process more geometric data types, making it suitable for a wide range of applications without requiring modifications to the network architecture.

[![](https://s7d1.scene7.com/is/image/dmqualcommprod/document-icon-v2_2?$QC_Responsive$&fmt=png-alpha)

#### Paper](https://arxiv.org/pdf/2305.18415.pdf)

[![](https://s7d1.scene7.com/is/image/dmqualcommprod/code-icon_2?$QC_Responsive$&fmt=png-alpha)

#### Code](https://github.com/qualcomm-ai-research/geometric-algebra-transformer)

## Citation

*@inproceedings{brehmer2023geometric, title = {Geometric Algebra Transformer}, author = {Brehmer, Johann and de Haan, Pim and Behrends, S{\"o}nke and Cohen, Taco}, booktitle = {Advances in Neural Information Processing Systems}, year = {2023}, volume = {37}, eprint = {2305.18415}, url = {https://arxiv.org/abs/2305.18415}, }*

## Results

We tested our method on several tasks, including robotic block stacking. In the graph above, our method outperforms all previous methods with 1% of the training data. As we scale the number of items, our method continues to outperform. GATr scales to tens of thousands of tokens, outperforming the geometric deep learning baselines.

## Looking for more papers with code?

[View all papers](https://www.qualcomm.com/research/artificial-intelligence/papers-with-code)

\* Qualcomm AI Research is an initiative of Qualcomm Technologies, Inc.