GreaseLM: Graph REASoning Enhanced Language Models for Question Answering
This addresses the challenge of robust reasoning in question answering for domains like commonsense and medical QA, though it is incremental in improving fusion methods.
The authors tackled the problem of answering complex questions requiring reasoning over both textual context and structured world knowledge by proposing GreaseLM, a model that fuses language model and graph neural network representations, achieving results that outperform models 8x larger on benchmarks like CommonsenseQA, OpenbookQA, and MedQA-USMLE.
Answering complex questions about textual narratives requires reasoning over both stated context and the world knowledge that underlies it. However, pretrained language models (LM), the foundation of most modern QA systems, do not robustly represent latent relationships between concepts, which is necessary for reasoning. While knowledge graphs (KG) are often used to augment LMs with structured representations of world knowledge, it remains an open question how to effectively fuse and reason over the KG representations and the language context, which provides situational constraints and nuances. In this work, we propose GreaseLM, a new model that fuses encoded representations from pretrained LMs and graph neural networks over multiple layers of modality interaction operations. Information from both modalities propagates to the other, allowing language context representations to be grounded by structured world knowledge, and allowing linguistic nuances (e.g., negation, hedging) in the context to inform the graph representations of knowledge. Our results on three benchmarks in the commonsense reasoning (i.e., CommonsenseQA, OpenbookQA) and medical question answering (i.e., MedQA-USMLE) domains demonstrate that GreaseLM can more reliably answer questions that require reasoning over both situational constraints and structured knowledge, even outperforming models 8x larger.