Tora Code 13b V1.0

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

ToRA is a tool - integrated reasoning agent designed to solve mathematical problems. It combines natural language reasoning with external tools, enhancing the efficiency of mathematical problem - solving.

🚀 Quick Start

For inference, evaluation, and training code, please refer to ToRA's GitHub repo.

✨ Features

• Tool - Integrated Reasoning: ToRA seamlessly integrates natural language reasoning with the use of external tools such as computation libraries and symbolic solvers.
• High - Performance on Math Datasets: ToRA models achieve high accuracy on various math datasets, like GSM8k and MATH. For example, ToRA - Code - 34B is the first and only open - source model to achieve over 50% accuracy (pass@1) on the MATH dataset.

📦 Installation

No installation steps are provided in the original document, so this section is skipped.

💻 Usage Examples

No code examples are provided in the original document, so this section is skipped.

📚 Documentation

🔥 News

• [2023/10/08] 🔥🔥🔥 All ToRA models released at [HuggingFace](https://huggingface.co/llm - agents)!!!
• [2023/09/29] ToRA paper, repo, and website released.

💡 Introduction

ToRA is a series of Tool - integrated Reasoning Agents designed to solve challenging mathematical reasoning problems by interacting with tools, e.g., computation libraries and symbolic solvers. ToRA series seamlessly integrate natural language reasoning with the utilization of external tools, thereby amalgamating the analytical prowess of language and the computational efficiency of external tools.

• ^*ToRA - Code - 34B is currently the first and only open - source model to achieve over 50% accuracy (pass@1) on the MATH dataset, which significantly outperforms GPT - 4’s CoT result (51.0 vs. 42.5), and is competitive with GPT - 4 solving problems with programs. By open - sourcing our codes and models, we hope more breakthroughs will come!

• ^†10 math tasks include GSM8k, MATH, GSM - Hard, SVAMP, TabMWP, ASDiv, SingleEQ, SingleOP, AddSub, and MultiArith.

⚡️ Training

The models are trained on ToRA - Corpus 16k, which contains tool - integrated reasoning trajectories of MATH and GSM8k from GPT - 4.

We use imitation learning (i.e., SFT) to fine - tune the models, and then apply our proposed output space shaping to improve tool - integrated reasoning behaviors. Please refer to the paper for more details.

🪁 Inference & Evaluation

Please refer to ToRA's GitHub repo for inference, evaluation, and training code.

🔧 Technical Details

The models are trained on ToRA - Corpus 16k, which contains tool - integrated reasoning trajectories of MATH and GSM8k from GPT - 4. Imitation learning (SFT) is used for fine - tuning, and output space shaping is applied to improve tool - integrated reasoning behaviors. For more details, please refer to the paper.

📄 License

The license used in this project is llama2.

☕️ 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}
}