A Multi-Task Learning Framework for Abstractive Text Summarization

Lu, Yao; Liu, Linqing; Jiang, Zhile; Yang, Min; Goebel, Randy

doi:10.1609/aaai.v33i01.33019987

“…In RL experiments, we train using BART from scratch, as opposed to using a model already fine-tuned on answer summarization, as we found that this model better learned to follow the given rewards. Following similar ratios in Lu et al (2019), we set (γ rl ,γ ml ) = (0.9, 0.1). Hyperparameters are tuned on the validation set.…”

Section: Rewardsmentioning

confidence: 99%

AnswerSumm: A Manually-Curated Dataset and Pipeline for Answer Summarization

Fabbri

¹

,

Wu

²

,

Iyer

³

et al. 2022

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua

View full text Add to dashboard Cite

Community Question Answering (CQA) fora such as Stack Overflow and Yahoo! Answers contain a rich resource of answers to a wide range of community-based questions. Each question thread can receive a large number of answers with different perspectives. One goal of answer summarization is to produce a summary that reflects the range of answer perspectives. A major obstacle for this task is the absence of a dataset to provide supervision for producing such summaries. Recent works propose heuristics to create such data, but these are often noisy and do not cover all answer perspectives present. This work introduces a novel dataset of 4,631 CQA threads for answer summarization curated by professional linguists. Our pipeline gathers annotations for all subtasks of answer summarization, including relevant answer sentence selection, grouping these sentences based on perspectives, summarizing each perspective, and producing an overall summary. We analyze and benchmark state-of-the-art models on these subtasks and introduce a novel unsupervised approach for multi-perspective data augmentation that boosts summarization performance according to automatic evaluation. Finally, we propose reinforcement learning rewards to improve factual consistency and answer coverage and analyze areas for improvement.

show abstract

“…Abstractive summarization has been enhanced in multitask learning frameworks with one additional task, by integrating it with text entailment generation (Pasunuru et al, 2017), extractive summarization (Chen et al, 2019;Hsu et al, 2018), and sentiment classification (Chan et al, 2020;Ma et al, 2018). While other research has combined multiple tasks, Lu et al (2019) integrated only predictive tasks, while Guo et al (2018) used only generative tasks. Recently, Dou and Neubig (2021) proposed using different tasks as guiding signals.…”

Section: Related Workmentioning

confidence: 99%

Exploring Multitask Learning for Low-Resource Abstractive Summarization

Magooda¹,

Litman²,

Elaraby³

2021

Findings of the Association for Computational Linguistics: EMNLP 2021

5

0

View full text Add to dashboard Cite

This paper explores the effect of using multitask learning for abstractive summarization in the context of small training corpora. In particular, we incorporate four different tasks (extractive summarization, language modeling, concept detection, and paraphrase detection) both individually and in combination, with the goal of enhancing the target task of abstractive summarization via multitask learning. We show that for many task combinations, a model trained in a multitask setting outperforms a model trained only for abstractive summarization, with no additional summarization data introduced. Additionally, we do a comprehensive search and find that certain tasks (e.g. paraphrase detection) consistently benefit abstractive summarization, not only when combined with other tasks but also when using different architectures and training corpora.

show abstract

“…Abstractive summarization has been enhanced in multitask learning frameworks with one additional task, by integrating it with text entailment generation (Pasunuru et al, 2017), extractive summarization (Chen et al, 2019;Hsu et al, 2018), and sentiment classification (Chan et al, 2020;Ma et al, 2018). While other research has combined multiple tasks, Lu et al (2019) integrated only predictive tasks, while Guo et al (2018) used only generative tasks. Recently, Dou and Neubig (2021) proposed using different tasks as guiding signals.…”

Section: Related Workmentioning

confidence: 99%

Exploring Multitask Learning for Low-Resource AbstractiveSummarization

Magooda¹,

Elaraby²,

Litman³

2021

Preprint

0

View full text Add to dashboard Cite

This paper explores the effect of using multitask learning for abstractive summarization in the context of small training corpora. In particular, we incorporate four different tasks (extractive summarization, language modeling, concept detection, and paraphrase detection) both individually and in combination, with the goal of enhancing the target task of abstractive summarization via multitask learning. We show that for many task combinations, a model trained in a multitask setting outperforms a model trained only for abstractive summarization, with no additional summarization data introduced. Additionally, we do a comprehensive search and find that certain tasks (e.g. paraphrase detection) consistently benefit abstractive summarization, not only when combined with other tasks but also when using different architectures and training corpora.

show abstract

A Multi-Task Learning Framework for Abstractive Text Summarization

Cited by 11 publications

References 2 publications

AnswerSumm: A Manually-Curated Dataset and Pipeline for Answer Summarization

AnswerSumm: A Manually-Curated Dataset and Pipeline for Answer Summarization

Exploring Multitask Learning for Low-Resource Abstractive Summarization

Exploring Multitask Learning for Low-Resource AbstractiveSummarization

Contact Info

Product

Resources

About