Saved in:
Bibliographic Details
Main Authors: Guo, Zhixin, Yan, Minyxuan, Qi, Jiexing, Zhou, Jianping, He, Ziwei, Zheng, Guanjie, Wang, Xinbing
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2302.12468
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909151452463104
author Guo, Zhixin
Yan, Minyxuan
Qi, Jiexing
Zhou, Jianping
He, Ziwei
Zheng, Guanjie
Wang, Xinbing
author_facet Guo, Zhixin
Yan, Minyxuan
Qi, Jiexing
Zhou, Jianping
He, Ziwei
Zheng, Guanjie
Wang, Xinbing
contents Pretrained language models (PLMs) have made remarkable progress in table-to-text generation tasks. However, the lack of domain-specific knowledge makes it challenging to bridge the topological gap between tabular data and text, especially in real-world applications with limited resources. To mitigate the limitation of insufficient labeled data, we propose a novel framework: Adapt-Knowledge-to-Generate (AKG). The core insight of AKG is to adapt unlabeled domain-specific knowledge into the model, which brings at least three benefits: (1) it injects representation of normal table-related descriptions to bridge the topological gap between tabular data and texts; (2) it enables us to use large amounts of unlabeled domain-specific knowledge fully, which can alleviate the PLMs' inherent shortcomings of lacking domain knowledge; (3) it allows us to design various tasks to employ the domain-specific knowledge. Extensive experiments and analyses are conducted on three open-domain, few-shot natural language generation (NLG) data sets: Humans, Songs, and Books. Compared to previous state-of-the-art approaches, our model achieves superior performance in terms of both fluency and accuracy as judged by human and automatic evaluations.
format Preprint
id arxiv_https___arxiv_org_abs_2302_12468
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Adapting Knowledge for Few-shot Table-to-Text Generation
Guo, Zhixin
Yan, Minyxuan
Qi, Jiexing
Zhou, Jianping
He, Ziwei
Zheng, Guanjie
Wang, Xinbing
Computation and Language
Pretrained language models (PLMs) have made remarkable progress in table-to-text generation tasks. However, the lack of domain-specific knowledge makes it challenging to bridge the topological gap between tabular data and text, especially in real-world applications with limited resources. To mitigate the limitation of insufficient labeled data, we propose a novel framework: Adapt-Knowledge-to-Generate (AKG). The core insight of AKG is to adapt unlabeled domain-specific knowledge into the model, which brings at least three benefits: (1) it injects representation of normal table-related descriptions to bridge the topological gap between tabular data and texts; (2) it enables us to use large amounts of unlabeled domain-specific knowledge fully, which can alleviate the PLMs' inherent shortcomings of lacking domain knowledge; (3) it allows us to design various tasks to employ the domain-specific knowledge. Extensive experiments and analyses are conducted on three open-domain, few-shot natural language generation (NLG) data sets: Humans, Songs, and Books. Compared to previous state-of-the-art approaches, our model achieves superior performance in terms of both fluency and accuracy as judged by human and automatic evaluations.
title Adapting Knowledge for Few-shot Table-to-Text Generation
topic Computation and Language
url https://arxiv.org/abs/2302.12468