Zero-Shot Cross-Lingual Abstractive Sentence Summarization through Teaching Generation and Attention

Duan, Xiangyu; Yin, Mingming; Zhang, Min; Chen, Boxing; Luo, Weihua

doi:10.18653/v1/p19-1305

Cited by 56 publications

(68 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Shen et al (2018) propose zero-shot cross-lingual headline generation to generate Chinese headlines for English articles, via a teacher-student framework, using two teacher models. Duan et al (2019) propose a similar approach for cross-lingual abstractive sentence summarization. We note that our approach is much simpler and also focuses on a different kind of summarization task.…”

Section: Related Workmentioning

confidence: 99%

WikiLingua: A New Benchmark Dataset for Cross-Lingual Abstractive Summarization

Ladhak¹,

Durmus²,

Cardie³

et al. 2020

Findings of the Association for Computational Linguistics: EMNLP 2020

104

View full text Add to dashboard Cite

We introduce WikiLingua, a large-scale, multilingual dataset for the evaluation of crosslingual abstractive summarization systems. We extract article and summary pairs in 18 languages from WikiHow 12 , a high quality, collaborative resource of how-to guides on a diverse set of topics written by human authors. We create gold-standard articlesummary alignments across languages by aligning the images that are used to describe each how-to step in an article. As a set of baselines for further studies, we evaluate the performance of existing cross-lingual abstractive summarization methods on our dataset. We further propose a method for direct crosslingual summarization (i.e., without requiring translation at inference time) by leveraging synthetic data and Neural Machine Translation as a pre-training step. Our method significantly outperforms the baseline approaches, while being more cost efficient during inference.

show abstract

Section: Related Workmentioning

confidence: 99%

WikiLingua: A New Benchmark Dataset for Cross-Lingual Abstractive Summarization

Ladhak¹,

Durmus²,

Cardie³

et al. 2020

Findings of the Association for Computational Linguistics: EMNLP 2020

104

View full text Add to dashboard Cite

show abstract

“…While the MultiLing benchmark covers 40 languages, it provides relatively few examples (10k in the 2019 release). Most proposed approaches, so far, have used an extractive approach given the lack of a multilingual corpus to train abstractive models (Duan et al, 2019).…”

Section: Multilingual Text Summarizationmentioning

confidence: 99%

MLSUM: The Multilingual Summarization Corpus

Scialom¹,

Dray²,

Lamprier³

et al. 2020

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

We present MLSUM, the first large-scale Mul-tiLingual SUMmarization dataset. Obtained from online newspapers, it contains 1.5M+ article/summary pairs in five different languages -namely, French, German, Spanish, Russian, Turkish. Together with English news articles from the popular CNN/Daily mail dataset, the collected data form a large scale multilingual dataset which can enable new research directions for the text summarization community. We report cross-lingual comparative analyses based on state-of-the-art systems. These highlight existing biases which motivate the use of a multi-lingual dataset.

show abstract

“…We circumvent this by replacing tags with task-specific transformer encoder layers which are added on top of the base encoder. This proposed architecture allows us to transfer supervision signals across languages and is potentially useful for other generation tasks, including question generation and sentence compression (Shen et al, 2018;Duan et al, 2019).…”

Section: Related Workmentioning

confidence: 99%

Zero-Shot Crosslingual Sentence Simplification

Mallinson¹,

Sennrich²,

Lapata³

2020

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

Sentence simplification aims to make sentences easier to read and understand. Recent approaches have shown promising results with encoder-decoder models trained on large amounts of parallel data which often only exists in English. We propose a zero-shot modeling framework which transfers simplification knowledge from English to another language (for which no parallel simplification corpus exists) while generalizing across languages and tasks. A shared transformer encoder constructs language-agnostic representations, with a combination of task-specific encoder layers added on top (e.g., for translation and simplification). Empirical results using both human and automatic metrics show that our approach produces better simplifications than unsupervised and pivot-based methods.

show abstract

Zero-Shot Cross-Lingual Abstractive Sentence Summarization through Teaching Generation and Attention

Cited by 56 publications

References 26 publications

WikiLingua: A New Benchmark Dataset for Cross-Lingual Abstractive Summarization

WikiLingua: A New Benchmark Dataset for Cross-Lingual Abstractive Summarization

MLSUM: The Multilingual Summarization Corpus

Zero-Shot Crosslingual Sentence Simplification

Contact Info

Product

Resources

About