CLAISep 21, 2023

MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models

Cambridge
arXiv:2309.12284v4671 citationsh-index: 49Has Code
Originality Incremental advance
AI Analysis

This addresses the challenge of enhancing mathematical problem-solving in LLMs for researchers and developers, representing a strong incremental improvement over existing open-source models.

The paper tackles the problem of improving mathematical reasoning in open-source large language models by proposing MetaMath, a fine-tuned model that achieves state-of-the-art performance, with MetaMath-7B exceeding previous models by 11.5% on GSM8K and 8.7% on MATH, and MetaMath-70B slightly outperforming GPT-3.5-Turbo on GSM8K.

Large language models (LLMs) have pushed the limits of natural language understanding and exhibited excellent problem-solving ability. Despite the great success, most existing open-source LLMs (e.g., LLaMA-2) are still far away from satisfactory for solving mathematical problem due to the complex reasoning procedures. To bridge this gap, we propose MetaMath, a fine-tuned language model that specializes in mathematical reasoning. Specifically, we start by bootstrapping mathematical questions by rewriting the question from multiple perspectives without extra knowledge, which results in a new dataset called MetaMathQA. Then we fine-tune the LLaMA-2 models on MetaMathQA. Experimental results on two popular benchmarks (i.e., GSM8K and MATH) for mathematical reasoning demonstrate that MetaMath outperforms a suite of open-source LLMs by a significant margin. Our MetaMath-7B model achieves 66.4% on GSM8K and 19.4% on MATH, exceeding the state-of-the-art models of the same size by 11.5% and 8.7%. Particularly, MetaMath-70B achieves an accuracy of 82.3% on GSM8K, slightly better than GPT-3.5-Turbo. We release all the MetaMathQA dataset, the MetaMath models with different model sizes and the training code for public use.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes