Feature-aware conditional GAN for category text generation
Category text generation receives considerable attentions since it is beneficial for various natural language processing tasks. Recently, the generative adversarial network (GAN) has attained promising performance in text generation, attributed to its adversarial training process. However, there...
Saved in:
Main Authors: | , , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
2023
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/171459 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-171459 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1714592023-10-26T08:43:36Z Feature-aware conditional GAN for category text generation Li, Xinze Mao, Kezhi Lin, Fanfan Feng, Zijian School of Electrical and Electronic Engineering Interdisciplinary Graduate School (IGS) Engineering::Electrical and electronic engineering Category Text Generation Text Processing Category text generation receives considerable attentions since it is beneficial for various natural language processing tasks. Recently, the generative adversarial network (GAN) has attained promising performance in text generation, attributed to its adversarial training process. However, there are several issues in text GANs, including discreteness, training instability, mode collapse, lack of diversity and controllability etc. To address these issues, this paper proposes a novel GAN framework, the feature-aware conditional GAN (FA-GAN), for controllable category text generation. In FA-GAN, the generator has a sequence-to-sequence structure for improving sentence diversity, which consists of three encoders including a special feature-aware encoder and a category-aware encoder, and one relational-memory-core-based decoder with the Gumbel SoftMax activation function. The discriminator has an additional category classification head. To generate sentences with specified categories, the multi-class classification loss is supplemented in the adversarial training. Comprehensive experiments have been conducted, and the results show that FA-GAN consistently outperforms 10 state-of-the-art text generation approaches on 6 text classification datasets. The case study demonstrates that the synthetic sentences generated by FA-GAN can match the required categories and are aware of the features of conditioned sentences, with good readability, fluency, and text authenticity. 2023-10-26T08:43:36Z 2023-10-26T08:43:36Z 2023 Journal Article Li, X., Mao, K., Lin, F. & Feng, Z. (2023). Feature-aware conditional GAN for category text generation. Neurocomputing, 547, 126352-. https://dx.doi.org/10.1016/j.neucom.2023.126352 0925-2312 https://hdl.handle.net/10356/171459 10.1016/j.neucom.2023.126352 2-s2.0-85163434694 547 126352 en Neurocomputing © 2023 Elsevier B.V. All rights reserved. |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Electrical and electronic engineering Category Text Generation Text Processing |
spellingShingle |
Engineering::Electrical and electronic engineering Category Text Generation Text Processing Li, Xinze Mao, Kezhi Lin, Fanfan Feng, Zijian Feature-aware conditional GAN for category text generation |
description |
Category text generation receives considerable attentions since it is
beneficial for various natural language processing tasks. Recently, the
generative adversarial network (GAN) has attained promising performance in text
generation, attributed to its adversarial training process. However, there are
several issues in text GANs, including discreteness, training instability, mode
collapse, lack of diversity and controllability etc. To address these issues,
this paper proposes a novel GAN framework, the feature-aware conditional GAN
(FA-GAN), for controllable category text generation. In FA-GAN, the generator
has a sequence-to-sequence structure for improving sentence diversity, which
consists of three encoders including a special feature-aware encoder and a
category-aware encoder, and one relational-memory-core-based decoder with the
Gumbel SoftMax activation function. The discriminator has an additional
category classification head. To generate sentences with specified categories,
the multi-class classification loss is supplemented in the adversarial
training. Comprehensive experiments have been conducted, and the results show
that FA-GAN consistently outperforms 10 state-of-the-art text generation
approaches on 6 text classification datasets. The case study demonstrates that
the synthetic sentences generated by FA-GAN can match the required categories
and are aware of the features of conditioned sentences, with good readability,
fluency, and text authenticity. |
author2 |
School of Electrical and Electronic Engineering |
author_facet |
School of Electrical and Electronic Engineering Li, Xinze Mao, Kezhi Lin, Fanfan Feng, Zijian |
format |
Article |
author |
Li, Xinze Mao, Kezhi Lin, Fanfan Feng, Zijian |
author_sort |
Li, Xinze |
title |
Feature-aware conditional GAN for category text generation |
title_short |
Feature-aware conditional GAN for category text generation |
title_full |
Feature-aware conditional GAN for category text generation |
title_fullStr |
Feature-aware conditional GAN for category text generation |
title_full_unstemmed |
Feature-aware conditional GAN for category text generation |
title_sort |
feature-aware conditional gan for category text generation |
publishDate |
2023 |
url |
https://hdl.handle.net/10356/171459 |
_version_ |
1781793766850428928 |