Important models and my work in CoQA task
Abstract
Firstly, I turned questions and paragraphs into an entity graph where each entity and sentence is a node. Secondly, I applied the graph recurrent network to extract global context information for embedding update. Thirdly, I predicted entities that are most relevant to the current question and built an answer based on these entities.
Date
May 8, 2021 1:00 PM — 3:00 PM
Event
presentation in lab for my CoQA project
Location
University at Buffalo
Presentation in lab to show the important models for CoQA task and explain my original multi-hop QA model
Barry Menglong Yao
PhD student in Computer Science
My research interests include multimodal entity linking and fake news detection.