Skip to content

ToRA is a series of Tool-integrated Reasoning LLM Agents designed to solve challenging mathematical reasoning problems by interacting with tools.

License

Notifications You must be signed in to change notification settings

Jaze-Developement/ToRA

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ToRA
ToRA: A Tool-Integrated Reasoning Agent


Project PagePaper

Repo for "ToRA: A Tool-Integrated Reasoning Agent for Mathematical Problem Solving"


Figure 1: Comparing ToRA with baselines on LLaMA-2 base models from 7B to 70B.

Tool-Integrated Reasoning Format


Figure 2: A basic example of single-round tool interaction, which interleaves rationales with program-based tool use.

ToRA Training Pipeline


Figure 3: Training ToRA contains ① Imitation Learning, and ② output space shaping.

Code & Models

🏝️ The code will be cleaned and uploaded within a few days, all ToRA models will be released.

☕️ Citation

If you find this repository helpful, please consider citing our paper:

@misc{gou2023tora,
      title={ToRA: A Tool-Integrated Reasoning Agent for Mathematical Problem Solving}, 
      author={Zhibin Gou and Zhihong Shao and Yeyun Gong and yelong shen and Yujiu Yang and Minlie Huang and Nan Duan and Weizhu Chen},
      year={2023},
      eprint={2309.17452},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

About

ToRA is a series of Tool-integrated Reasoning LLM Agents designed to solve challenging mathematical reasoning problems by interacting with tools.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published