This paper presents a tree-structured neural topic model, which has a topic distribution over a tree with an infinite number of branches. Our model parameterizes an unbounded ancestral and fraternal topic distribution by applying doubly-recurrent neural networks. With the help of autoencoding ...
1. 论文阅读:End-to-End Neural Pipeline for Goal-Oriented Dialogue Systems using GPT-2, ACL2020.54(397) 2. 论文阅读:Tree-Structured Neural Topic Model,ACL2020.73(385) 3. 论文阅读:TransS-Driven Joint Learning Architecture for Implicit Discourse Relation Recognition(342) 4. 论文阅读:Evaluating...
Tree-Structured Neural Topic Model A code for "Tree-Structured Neural Topic Model" in ACL2020Corresponding paper: https://www.aclweb.org/anthology/2020.acl-main.73/Masaru Isonuma, Juncihiro Mori, Danushka Bollegala, and Ichiro Sakata (The University of Tokyo, University of Liverpool)...
First, compared to the baseline models, including sequence-based models, self-attention models, convolutional neural network models, and graph neural network models, the SDTGCN model outperforms the baseline models on the Rest14, Rest15, Rest16, and Twitter datasets. This effectively demonstrates th...