GPT Can Solve Mathematical Problems Without a Calculator
This addresses the problem of arithmetic limitations in language models for AI applications, though it is incremental as it builds on existing fine-tuning methods.
The paper challenges the assumption that large language models cannot perform arithmetic accurately without calculator tools, showing that a 2 billion-parameter model can achieve almost 100% accuracy in multi-digit operations, significantly outperforming GPT-4's 4.3% accuracy.
Previous studies have typically assumed that large language models are unable to accurately perform arithmetic operations, particularly multiplication of >8 digits, and operations involving decimals and fractions, without the use of calculator tools. This paper aims to challenge this misconception. With sufficient training data, a 2 billion-parameter language model can accurately perform multi-digit arithmetic operations with almost 100% accuracy without data leakage, significantly surpassing GPT-4 (whose multi-digit multiplication accuracy is only 4.3%). We also demonstrate that our MathGLM, fine-tuned from GLM-10B on a dataset with additional multi-step arithmetic operations and math problems described in text, achieves similar performance to GPT-4 on a 5,000-samples Chinese math problem test set. Our code and data are public at https://github.com/THUDM/MathGLM.