MathScale: Scaling Instruction Tuning for Mathematical Reasoning
- URL: http://arxiv.org/abs/2403.02884v1
- Date: Tue, 5 Mar 2024 11:42:59 GMT
- Title: MathScale: Scaling Instruction Tuning for Mathematical Reasoning
- Authors: Zhengyang Tang, Xingxing Zhang, Benyou Wan, Furu Wei
- Abstract summary: Large language models (LLMs) have demonstrated remarkable capabilities in problem-solving.
However, their proficiency in solving mathematical problems remains inadequate.
We propose MathScale, a simple and scalable method to create high-quality mathematical reasoning data.
- Score: 70.89605383298331
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Large language models (LLMs) have demonstrated remarkable capabilities in
problem-solving. However, their proficiency in solving mathematical problems
remains inadequate. We propose MathScale, a simple and scalable method to
create high-quality mathematical reasoning data using frontier LLMs (e.g., {\tt
GPT-3.5}). Inspired by the cognitive mechanism in human mathematical learning,
it first extracts topics and knowledge points from seed math questions and then
build a concept graph, which is subsequently used to generate new math
questions. MathScale exhibits effective scalability along the size axis of the
math dataset that we generate. As a result, we create a mathematical reasoning
dataset (MathScaleQA) containing two million math question-answer pairs. To
evaluate mathematical reasoning abilities of LLMs comprehensively, we construct
{\sc MwpBench}, a benchmark of Math Word Problems, which is a collection of ten
datasets (including GSM8K and MATH) covering K-12, college, and competition
level math problems. We apply MathScaleQA to fine-tune open-source LLMs (e.g.,
LLaMA-2 and Mistral), resulting in significantly improved capabilities in
mathematical reasoning. Evaluated on {\sc MwpBench}, MathScale-7B achieves
state-of-the-art performance across all datasets, surpassing its best peers of
equivalent size by 42.9\% in micro average accuracy and 43.7\% in macro average
accuracy, respectively.
Related papers
- MathBench: Evaluating the Theory and Application Proficiency of LLMs with a Hierarchical Mathematics Benchmark [82.64129627675123]
MathBench is a new benchmark that rigorously assesses the mathematical capabilities of large language models.
MathBench spans a wide range of mathematical disciplines, offering a detailed evaluation of both theoretical understanding and practical problem-solving skills.
arXiv Detail & Related papers (2024-05-20T17:52:29Z) - FineMath: A Fine-Grained Mathematical Evaluation Benchmark for Chinese Large Language Models [44.63505885248145]
FineMath is a fine-grained mathematical evaluation benchmark dataset for assessing Chinese Large Language Models (LLMs)
FineMath is created to cover the major key mathematical concepts taught in elementary school math, which are divided into 17 categories of math word problems.
All the 17 categories of math word problems are manually annotated with their difficulty levels according to the number of reasoning steps required to solve these problems.
arXiv Detail & Related papers (2024-03-12T15:32:39Z) - MATHSENSEI: A Tool-Augmented Large Language Model for Mathematical Reasoning [2.9104279358536647]
We present MathSensei, a tool-augmented large language model for mathematical reasoning.
We study the complementary benefits of the tools - knowledge retriever (Bing Web Search), program generator + executor (Python), and symbolic equation solver (Wolfram-Alpha API)
arXiv Detail & Related papers (2024-02-27T05:50:35Z) - MathCoder: Seamless Code Integration in LLMs for Enhanced Mathematical
Reasoning [52.97768001837269]
We present a method to fine-tune open-source language models, enabling them to use code for modeling and deriving math equations.
We propose a method of generating novel and high-quality datasets with math problems and their code-based solutions.
This approach yields the MathCoder models, a family of models capable of generating code-based solutions for solving challenging math problems.
arXiv Detail & Related papers (2023-10-05T17:52:09Z) - MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models [91.66694225955872]
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.
We release all the MetaMathQA dataset, the MetaMath models with different model sizes and the training code for public use.
arXiv Detail & Related papers (2023-09-21T17:45:42Z) - WizardMath: Empowering Mathematical Reasoning for Large Language Models
via Reinforced Evol-Instruct [128.89645483139236]
We present WizardMath, which enhances the mathematical reasoning abilities of Llama-2, by applying our proposed Reinforcement Learning from Evol-Instruct Feedback (RLEIF) method to the domain of math.
Our model even surpasses ChatGPT-3.5, Claude Instant-1, PaLM-2 and Minerva on GSM8k, simultaneously surpasses Text-davinci, PaLM-1 and GPT-3 on MATH.
arXiv Detail & Related papers (2023-08-18T14:23:21Z) - Measuring Mathematical Problem Solving With the MATH Dataset [55.4376028963537]
We introduce MATH, a dataset of 12,500 challenging competition mathematics problems.
Each problem has a full step-by-step solution which can be used to teach models to generate answer derivations and explanations.
We also contribute a large auxiliary pretraining dataset which helps teach models the fundamentals of mathematics.
arXiv Detail & Related papers (2021-03-05T18:59:39Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.