StepGame: A New Benchmark for Robust Multi-Hop Spatial Reasoning in Texts

TL;DR

StepGame introduces a challenging new dataset for multi-hop spatial reasoning in texts, highlighting limitations of existing models and proposing a tensor-product memory network that outperforms baselines in robustness and generalization.

Contribution

The paper presents StepGame, a novel dataset for robust spatial reasoning, and a specialized tensor-product memory network that improves performance over existing models.

Findings

State-of-the-art models struggle with StepGame.

TP-MANN outperforms baselines in robustness.

Models show limited generalization on new spatial reasoning tasks.

Abstract

Inferring spatial relations in natural language is a crucial ability an intelligent system should possess. The bAbI dataset tries to capture tasks relevant to this domain (task 17 and 19). However, these tasks have several limitations. Most importantly, they are limited to fixed expressions, they are limited in the number of reasoning steps required to solve them, and they fail to test the robustness of models to input that contains irrelevant or redundant information. In this paper, we present a new Question-Answering dataset called StepGame for robust multi-hop spatial reasoning in texts. Our experiments demonstrate that state-of-the-art models on the bAbI dataset struggle on the StepGame dataset. Moreover, we propose a Tensor-Product based Memory-Augmented Neural Network (TP-MANN) specialized for spatial reasoning tasks. Experimental results on both datasets show that our model outperforms all the baselines with superior generalization and robustness performance.