Teaching Neural Module Networks to Do Arithmetic
This work addresses the limitation of NMNs in handling numerical reasoning for complex question-answering tasks, representing an incremental advancement in the field.
The paper tackled the problem of enabling Neural Module Networks (NMNs) to perform numerical reasoning by introducing addition and subtraction modules and bridging the interpreter-question gap, resulting in a 17.7% F1 score improvement on a DROP subset and outperforming previous state-of-the-art models.
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.