Generate & Rank: A Multi-task Framework for Math Word Problems

TL;DR

This paper introduces Generate & Rank, a multi-task framework combining generation and ranking to improve math word problem solving, effectively reducing mistakes and outperforming existing methods on benchmark datasets.

Contribution

The paper proposes a novel multi-task framework that jointly trains generation and ranking models, incorporating tree-based disturbances and online updates to enhance accuracy in math word problems.

Findings

Outperforms baselines on multiple datasets

Achieves 7% higher accuracy on Math23k

Effectively reduces generation mistakes

Abstract

Math word problem (MWP) is a challenging and critical task in natural language processing. Many recent studies formalize MWP as a generation task and have adopted sequence-to-sequence models to transform problem descriptions to mathematical expressions. However, mathematical expressions are prone to minor mistakes while the generation objective does not explicitly handle such mistakes. To address this limitation, we devise a new ranking task for MWP and propose Generate & Rank, a multi-task framework based on a generative pre-trained language model. By joint training with generation and ranking, the model learns from its own mistakes and is able to distinguish between correct and incorrect expressions. Meanwhile, we perform tree-based disturbance specially designed for MWP and an online update to boost the ranker. We demonstrate the effectiveness of our proposed method on the benchmark and the results show that our method consistently outperforms baselines in all datasets. Particularly, in the classical Math23k, our method is 7% (78.4% $\rightarrow$ 85.4%) higher than the state-of-the-art.