CALT: Computer ALgebra with Transformer

CALT (Computer ALgebra with Transformer) is a simple Python library for learning arithmetic and symbolic computation with Transformer models. It offers a basic Transformer model and training, and non-experts of deep learning (e.g., mathematicians) can focus on constructing datasets to train and evaluate the model. Particularly, users only need to implement an instance generator for their own task.

Installation

CALT can be installed via pip.

pip install calt-x

We highly recommend the users to use CALT codebase - a comprehensive template repository to build up your own projects using CALT. The quickstart guide can be found in CALT codebase documentation.

Citation

If you use this code in your research, please cite our paper:

@misc{kera2025calt,
  title={CALT: A Library for Computer Algebra with Transformer},
  author={Hiroshi Kera and Shun Arawaka and Yuta Sato},
  year={2025},
  archivePrefix={arXiv},
  eprint={2506.08600}
}

The following is a small list of such studies from our group.

• "Learning to Compute Gröbner Bases," Kera et al., 2024
• "Computational Algebra with Attention: Transformer Oracles for Border Basis Algorithms," Kera and Pelleriti et al., 2025
• "Geometric Generality of Transformer-Based Gröbner Basis Computation," Kambe et al., 2025

Refer to our paper "CALT: A Library for Computer Algebra with Transformer," Kera et al., 2025 for a comprehensive overview.