Generating Long Financial Report using Conditional Variational Autoencoders with Knowledge Distillation

Authors

  • Yunpeng Ren Harbin Institute of Technology Shenzhen Graduate School Qianhai Financial Holdings Co., Ltd.
  • Ziao Wang Harbin Institute of Technology Shenzhen Graduate School
  • Yiyuan Wang Harbin Institute of Technology Shenzhen Graduate School
  • Xiaofeng Zhang Harbin Institute of Technology Shenzhen Graduate School

DOI:

https://doi.org/10.1609/aaai.v35i18.17936

Keywords:

Deep Learning, Text Generation, Variational Autoencoders

Abstract

Automatically generating financial report from a piece of news is quite a challenging task. Apparently, the difficulty of this task lies in the lack of sufficient background knowledge to effectively generate long financial report. To address this issue, this paper proposes the conditional variational autoencoders (CVAE) based approach which distills external knowledge from a corpus of news-report data. Experimental results demonstrate that the proposed approach could achieve the SOTA performance.
AAAI-21 Proceedings Cover

Downloads

Published

2021-05-18

How to Cite

Ren, Y., Wang, Z., Wang, Y., & Zhang, X. (2021). Generating Long Financial Report using Conditional Variational Autoencoders with Knowledge Distillation. Proceedings of the AAAI Conference on Artificial Intelligence, 35(18), 15879-15880. https://doi.org/10.1609/aaai.v35i18.17936

Issue

Vol. 35 No. 18: AAAI-21 Student Papers and Demonstrations

Section

AAAI Student Abstract and Poster Program