arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2201.08860cs.CLcs.LG
收藏

GreaseLM:用于问答的图推理增强语言模型

GreaseLM: Graph REASoning Enhanced Language Models for Question Answering

Stanford University(斯坦福大学)

Xikun Zhang, Antoine Bosselut, Michihiro Yasunaga, Hongyu Ren, Percy Liang, Christopher D. Manning, Jure Leskovec

摘要:

对文本叙述的复杂问题进行推理需要对所陈述的上下文和作为其基础的世界知识进行推理,然而,作为大多数现代问答系统的基础的预训练语言模型(LM)不能鲁棒地表示概念之间的潜在关系,而这是推理所必需的。虽然知识图谱(KG)经常被用来增强LM与世界知识的结构化表示,它仍然是一个悬而未决的问题,如何有效地融合和推理的KG表示和语言的背景下,提供了情景约束和细微差别。在这项工作中,我们提出了GreaseLM,一个新的模型,融合编码表示从预先训练的LM和图神经网络在多个层次的模态交互操作。来自两种模态的信息传播到另一种模态,允许语言上下文表示以结构化的世界知识为基础,并允许语言的细微差别(例如,否定,模糊)的上下文中通知知识的图形表示。我们在常识推理的三个基准上的结果(即,CommonsenseQA、OpenbookQA)和医学问答(即,MedQA-USMLE)领域证明,GreaseLM可以更可靠地回答需要在情境约束和结构化知识上进行推理的问题,甚至优于8倍大的模型。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑