Quantity Tagger: A Latent-Variable Sequence Labeling Approach to Solving Addition-Subtraction Word Problems

TL;DR

Quantity Tagger is a novel sequence labeling method that automatically identifies the mathematical relations among quantities in word problems, significantly improving accuracy over previous approaches.

Contribution

It introduces a latent-variable model that tags quantities with signs indicating operations, capturing hidden relations for solving addition-subtraction problems.

Findings

Achieves 5 and 8 points accuracy improvements on two datasets.

Effectively discovers underlying mathematical relations.

Outperforms prior methods in accuracy.

Abstract

An arithmetic word problem typically includes a textual description containing several constant quantities. The key to solving the problem is to reveal the underlying mathematical relations (such as addition and subtraction) among quantities, and then generate equations to find solutions. This work presents a novel approach, Quantity Tagger, that automatically discovers such hidden relations by tagging each quantity with a sign corresponding to one type of mathematical operation. For each quantity, we assume there exists a latent, variable-sized quantity span surrounding the quantity token in the text, which conveys information useful for determining its sign. Empirical results show that our method achieves 5 and 8 points of accuracy gains on two datasets respectively, compared to prior approaches.