Teaching Neural Module Networks to Do Arithmetic

TL;DR

This paper enhances Neural Module Networks with numerical reasoning modules, significantly improving their ability to handle complex, multi-step questions involving arithmetic over raw text.

Contribution

The paper introduces addition and subtraction modules to NMNs, bridging the gap for numerical reasoning and achieving state-of-the-art results on DROP.

Findings

17.7% improvement in F1 score on DROP

Enhanced NMNs' numerical reasoning capabilities

Outperforms previous models significantly

Abstract

Answering complex questions that require multi-step multi-type reasoning over raw text is challenging, especially when conducting numerical reasoning. Neural Module Networks(NMNs), follow the programmer-interpreter framework and design trainable modules to learn different reasoning skills. However, NMNs only have limited reasoning abilities, and lack numerical reasoning capability. We up-grade NMNs by: (a) bridging the gap between its interpreter and the complex questions; (b) introducing addition and subtraction modules that perform numerical reasoning over numbers. On a subset of DROP, experimental results show that our proposed methods enhance NMNs' numerical reasoning skills by 17.7% improvement of F1 score and significantly outperform previous state-of-the-art models.