KG-BART: Knowledge Graph-Augmented BART for Generative Commonsense Reasoning

In progress paper talk and presentation Read more 27 Aug 2021 - less than 1 minute read

Fusing Context Into Knowledge Graph for Commonsense Question Answering

Method Knowledge Retrieval Suppose the question mentions an entity Eq ∈ V and the choice contains an entity Ec ∈ V. W We then employ the KCR method to select relation triples. If there is a direct edge r from Eq to Ec in G, we choose this triple (Eq , r, Ec). Otherwise, we retrieve all the N triples containing Ec. Each triple j is a... Read more 26 Aug 2021 - 2 minute read

A Semantic-based Method for Unsupervised Commonsense Question Answering

introduction Since existing methods can be easily distracted by irrelevant factors such as lexical perturbations, we argue that a commonsense question answering method should focus on the answers’ smantics and assign similar scores to synonymous choices. Method In this paper, we focus on unsupervised multiple-choice commonsense questio... Read more 26 Aug 2021 - 2 minute read

Connecting the Dots: A Knowledgeable Path Generator for Commonsense Question Answering

Paper contribution It is proposed to learn a multi-hop knowledge path generator dynamically to generate structured evidence based on the problem. The generator is based on a pre-trained language model, and uses a large amount of unstructured knowledge stored in the language model to supplement the incompleteness of the knowledge base. T... Read more 25 Aug 2021 - 8 minute read

KagNet: Knowledge-Aware Graph Networks for Commonsense Reasoning

Summary A proposed Knowledge-aware reasoning framework is , which mainly has the following two steps: schema graph grounding (see picture below) graph modeling for inference The core is the GCN-LSTM-HPA structure: Composed of GCN, LSTM, and hierarchical path-based attention mechanism Used for path... Read more 24 Aug 2021 - 4 minute read