Abstractive Text Summarization Using LSTMs with Rich Features

Quoc, Viet Nguyen; Thanh, Huong Le; Minh, Tuan Luu

doi:10.1007/978-981-15-6168-9_3

Cited by 5 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most studies to construct abstractive summaries in Vietnamese use an encoder-decoder framework or a pre-trained model. Quoc et al (2019) integrate sentence positions and term frequencies into a pointer generator network with a coverage mechanism to perform the abstractive summarization for Vietnamese documents. Lam et al ( 2022) construct abstractive summaries for online newspapers using RNN with attention, BiLSTM with copy generator, standard Transformer, BERT, and sequence-to-sequence abstractive models using bottom-up approach.…”

Section: Related Workmentioning

confidence: 99%

Abstractive Text Summarization Using the BRIO Training Paradigm

Lam¹,

Doan²,

Pham³

et al. 2023

Findings of the Association for Computational Linguistics: ACL 2023

View full text Add to dashboard Cite

sentences produced by abstractive summarization models may be coherent and comprehensive, but they lack control and rely heavily on reference summaries. The BRIO training paradigm assumes a non-deterministic distribution to reduce the model's dependence on reference summaries, and improve model performance during inference. This paper presents a straightforward but effective technique to improve abstractive summaries by finetuning pre-trained language models, and training them with the BRIO paradigm. We build a text summarization dataset for Vietnamese, called VieSum. We perform experiments with abstractive summarization models trained with the BRIO paradigm on the CNNDM and the VieSum datasets. The results show that the models, trained on basic hardware, outperform all existing abstractive summarization models, especially for Vietnamese.

show abstract

Section: Related Workmentioning

confidence: 99%