Guiding Extractive Summarization with Question-Answering Rewards

Arumae, Kristjan; Liu, Fei

doi:10.18653/v1/n19-1264

Cited by 33 publications

(25 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Prior work proposed heuristic-based reward functions to train crossinput RL summarisers, in order to strengthen certain properties of the generated summaries. Arumae and Liu (2019) propose four reward functions to train an RL-based extractive summariser, including the question-answering competency rewards, which encourage the RL agent to generate summaries that can answer cloze-style questions. Such questions are automatically created by removing some words in the reference summaries.…”

Section: Related Workmentioning

confidence: 99%

Better Rewards Yield Better Summaries: Learning to Summarise Without References

Böhm¹,

Gao²,

Meyer³

et al. 2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

View full text Add to dashboard Cite

Reinforcement Learning (RL) based document summarisation systems yield state-of-the-art performance in terms of ROUGE scores, because they directly use ROUGE as the rewards during training. However, summaries with high ROUGE scores often receive low human judgement. To find a better reward function that can guide RL to generate human-appealing summaries, we learn a reward function from human ratings on 2,500 summaries. Our reward function only takes the document and system summary as input. Hence, once trained, it can be used to train RL-based summarisation systems without using any reference summaries. We show that our learned rewards have significantly higher correlation with human ratings than previous approaches. Human evaluation experiments show that, compared to the state-of-the-art supervised-learning systems and ROUGE-as-rewards RL summarisation systems, the RL systems using our learned rewards during training generate summaries with higher human ratings. The learned reward function and our source code are available at https://github.com/yg211/ summary-reward-no-reference.

show abstract

Section: Related Workmentioning

confidence: 99%

Better Rewards Yield Better Summaries: Learning to Summarise Without References

Böhm¹,

Gao²,

Meyer³

et al. 2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

View full text Add to dashboard Cite

show abstract

“…We proceed by evaluating our method against state-of-the-art extractive and abstractive summarization systems. Arumae and Liu (2019) present an approach to extract summary segments using question-answering as supervision signal, assuming a high quality summary can serve as document surrogate to answer questions. See et al 2017present pointer-generator networks, an abstractive summarization model and a reliable baseline for being both state-of-the-art, and also a vital tool for guiding our data creation.…”

Section: Resultsmentioning

confidence: 99%

“…The encoder hidden state dimension is set at 768, with 12 layers and 12 attention heads (BERT BASE uncased). We utilize dropout with p = 0.1, and λ is empirically set to 0.1. and an extractive summarizer (Arumae and Liu, 2019) whose CNN/DM results are macro-averaged. The bottom two sections showcase our models.…”

Section: Methodsmentioning

confidence: 99%

Proceedings of the 2nd Workshop on New Frontiers in Summarization

2019

View full text Add to dashboard Cite

In recent years, the task of Question Answering over passages, also pitched as a reading comprehension, has evolved into a very active research area. A reading comprehension system extracts a span of text, comprising of named entities, dates, small phrases, etc., which serve as the answer to a given question. However, these spans of text would result in an unnatural reading experience in a conversational system. Usually, dialogue systems solve this issue by using template-based language generation. These systems, though adequate for a domain specific task, are too restrictive and predefined for a domain independent system. In order to present the user with a more conversational experience, we propose a pointer generator based full-length answer generator which can be used with most QA systems. Our system generates a full-length answer given a question and the extracted factoid/span answer without relying on the passage from where the answer was extracted. We also present a dataset of 315,000 question, factoid answer and full-length answer triples. We have evaluated our system using ROUGE-1,2,L and BLEU and achieved 74.05 BLEU score and 86.25 Rogue-L score.

show abstract

“…Besides selective rationalization, the cooperative game has been studied in the latter direction above (Lee et al, 2018). It has also been applied to a relevant problem on summarization (Arumae and Liu, 2019), where the selected summary should be sufficient for answering questions related to the document. In this problem, the sum-mary is a special type of rationale of a document.…”

Section: Related Workmentioning

confidence: 99%

Rethinking Cooperative Rationalization: Introspective Extraction and Complement Control

Chang

Zhang

et al. 2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

View full text Add to dashboard Cite

Selective rationalization has become a common mechanism to ensure that predictive models reveal how they use any available features. The selection may be soft or hard, and identifies a subset of input features relevant for prediction. The setup can be viewed as a cooperate game between the selector (aka rationale generator) and the predictor making use of only the selected features. The co-operative setting may, however, be compromised for two reasons. First, the generator typically has no direct access to the outcome it aims to justify, resulting in poor performance. Second, there's typically no control exerted on the information left outside the selection. We revise the overall co-operative framework to address these challenges. We introduce an introspective model which explicitly predicts and incorporates the outcome into the selection process. Moreover, we explicitly control the rationale complement via an adversary so as not to leave any useful information out of the selection. We show that the two complementary mechanisms maintain both high predictive accuracy and lead to comprehensive rationales.

show abstract

Guiding Extractive Summarization with Question-Answering Rewards

Cited by 33 publications

References 43 publications

Better Rewards Yield Better Summaries: Learning to Summarise Without References

Better Rewards Yield Better Summaries: Learning to Summarise Without References

Proceedings of the 2nd Workshop on New Frontiers in Summarization

Rethinking Cooperative Rationalization: Introspective Extraction and Complement Control

Contact Info

Product

Resources

About