Incorporating Commonsense Knowledge Graph in Pretrained Models for Social Commonsense Tasks

2021/05/12 by Ting-Yun Chang, Chang, Ting-Yun, Yang Liu +10
Computer Science · #Advanced Graph Neural Networks #Natural Language Processing Techniques #Topic Modeling #cs.CL

paper · pdf · doi:10.48550/arxiv.2105.05457

EMNLP2020 Workshop

arxiv created 2021/05/12 · arxiv updated 2021/05/13

Abstract

Pretrained language models have excelled at many NLP tasks recently; however, their social intelligence is still unsatisfactory. To enable this, machines need to have a more general understanding of our complicated world and develop the ability to perform commonsense reasoning besides fitting the specific downstream tasks. External commonsense knowledge graphs (KGs), such as ConceptNet, provide rich information about words and their relationships. Thus, towards general commonsense learning, we propose two approaches to implicitly and explicitly infuse such KGs into pretrained language models. We demonstrate our proposed methods perform well on SocialIQA, a social commonsense reasoning task, in both limited and full training data regimes.

Citations

• Bidirectional Attention Flow for Machine Comprehension

Related

• Commonsense Knowledge Transfer for Pre-trained Language Models
• Go Beyond Plain Fine-tuning: Improving Pretrained Models for Social Commonsense
• Do Language Models Perform Generalizable Commonsense Inference?
• Great Truths are Always Simple: A Rather Simple Knowledge Encoder for Enhancing the Commonsense Reasoning Capacity of Pre-Trained Models
• Symbolic Knowledge Distillation: from General Language Models to Commonsense Models