Empowering Language Models with Knowledge Graph Reasoning for Question Answering

Ziniu Hu, Yichong Xu, Wenhao Yu, Shuohang Wang, Ziyi Yang, Chenguang Zhu, Kai-Wei Chang, Yizhou Sun

Keywords: Open-Domain Question Answering, Knowledge Graph Reasoning, Neural-Symbolic Model

TL;DR: We propose OREO-LM, a model that incorporates symbolic Knowledge Graph Reasoning with existing Language Models

Abstract:

Answering open-domain questions requires world knowledge about in-context entities. As pre-trained Language Models (LMs) lack the power to store required knowledge, external knowledge sources, such as knowledge graphs, are often used to augment LMs.
In this work, we propose knOwledge REasOning empowered Language Model (OREO-LM), which consists of a novel Knowledge Interaction Layer that can be flexibly plugged into existing Transformer-based LMs to interact with a differentiable Knowledge Graph Reasoning module collaboratively. In this way, LM guides KG to walk towards the desired answer, while the retrieved knowledge improves LM.
By adopting OREO-LM to RoBERTa and T5, we show significant performance gain, achieving state-of-art results in the Closed-Book setting. The performance enhancement is mainly from the KG reasoning's capacity to infer missing relational facts. In addition, OREO-LM provides reasoning paths as rationales to interpret the model's decision.