Neuro-symbolic Training for Reasoning over Spatial Language

TL;DR

This paper introduces a neuro-symbolic training approach that enhances language models' ability to perform complex spatial reasoning over text by incorporating logical rules as constraints, leading to better generalization.

Contribution

The paper proposes a novel neuro-symbolic training method that integrates spatial logical rules into language models to improve reasoning and question answering over spatial language.

Findings

Improved accuracy on spatial question-answering benchmarks

Enhanced ability to handle nested and multi-hop spatial reasoning

Better generalization across different spatial domains

Abstract

Spatial reasoning based on natural language expressions is essential for everyday human tasks. This reasoning ability is also crucial for machines to interact with their environment in a human-like manner. However, recent research shows that even state-of-the-art language models struggle with spatial reasoning over text, especially when facing nesting spatial expressions. This is attributed to not achieving the right level of abstraction required for generalizability. To alleviate this issue, we propose training language models with neuro-symbolic techniques that exploit the spatial logical rules as constraints, providing additional supervision to improve spatial reasoning and question answering. Training language models to adhere to spatial reasoning rules guides them in making more effective and general abstractions for transferring spatial knowledge to various domains. We evaluate our approach on existing spatial question-answering benchmarks. Our results indicate the effectiveness of our proposed technique in improving language models in complex multi-hop spatial reasoning over text.