Abstract
The current encode-attend-decode paradigm suffers from noisy encoder problem. We implemented a seq2seq model consisting of a novel selective mechanism for query focused abstractive document summarization using neural networks to solve this problem. Selective mechanism was used for the better representation of input (passage) sequence. We conducted experiments on Debatepedia dataset and have demonstrated that our model outperforms the state-of-the-art model in all ROUGE scores.
Cite
CITATION STYLE
Aryal, C., & Chali, Y. (2020). Selection Driven Query Focused Abstractive Document Summarization. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 12109 LNAI, pp. 118–124). Springer. https://doi.org/10.1007/978-3-030-47358-7_11
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
