Papers › A Deep Reinforced Model for Abstractive Summarization

A Deep Reinforced Model for Abstractive Summarization

11 May 2017ICLR 2018 1arXiv:1705.04304archive 2025-07-28

Romain Paulus, Caiming Xiong, Richard Socher

Attentional, RNN-based encoder-decoder models for abstractive summarization have achieved good performance on short input and output sequences. For longer documents and summaries however these models often include repetitive and incoherent phrases. We introduce a neural network model with a novel intra-attention that attends over the input and continuously generated output separately, and a new training method that combines standard supervised word prediction and reinforcement learning (RL). Models trained only with supervised learning often exhibit "exposure bias" - they assume ground truth is provided at each step during training. However, when standard word prediction is combined with the global sequence prediction training of RL the resulting summaries become more readable. We evaluate this model on the CNN/Daily Mail and New York Times datasets. Our model obtains a 41.16 ROUGE-1 score on the CNN/Daily Mail dataset, an improvement over previous state-of-the-art models. Human evaluation also shows that our model produces higher quality summaries.

PaperPDFConference PDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

AndreyKolomiets/News_Headline_Generation mentioned on GitHubtfSyntology: not harvested report
Hmkhalla/NLP-summarizer mentioned on GitHubpytorchSyntology: not harvested report
JRC1995/Abstractive-Summarization mentioned on GitHubtfSyntology: not harvested report
Kumar-Tarun/summarization mentioned on GitHubpytorchSyntology: not harvested report
VD44/Rouge-L-Tensorflow mentioned on GitHubtfSyntology: not harvested report
cosmoquester/2021-dialogue-summary-competition mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation,... (arXiv:1910.13461; that paper's own run record: 40 ran (of which 7 constructed an object rather than computing a result; 30 with no instrument failure: 1 honoured, 2 violated, 27 with no contract checked; 10 where Syntology's instrument failed) · 13 unverified) and R-Drop: Regularized Dropout for Neural Networks (arXiv:2106.14448; that paper's own run record: 4 ran (of which 2 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
manshri/tesum mentioned on GitHubpytorchSyntology: not harvested report
oceanypt/A-DEEP-REINFORCED-MODEL-FOR-ABSTRACTIVE-SUMMARIZATION mentioned on GitHubpytorchSyntology: not harvested report
rohithreddy024/Text-Summarizer-Pytorch mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Get To The Point: Summarization with Pointer-Generator Networks (arXiv:1704.04368; that paper's own run record: 30 ran (of which 11 constructed an object rather than computing a result; 25 with no instrument failure: 4 honoured, 2 violated, 19 with no contract checked; 5 where Syntology's instrument failed) · 34 unverified); nothing checks that they implement this paper's method. report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Abstractive Text SummarizationDecoderPredictionReinforcement LearningReinforcement Learning (RL)Text Summarizationmodel

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Document Summarization CNN / Daily Mail ML + RL (Paulus et al., 2017) ROUGE-1 39.87 #20 of 26 Archive leaderboard report
Document Summarization CNN / Daily Mail ML + RL (Paulus et al., 2017) ROUGE-2 15.82 #20 of 26 Archive leaderboard report
Document Summarization CNN / Daily Mail ML + RL (Paulus et al., 2017) ROUGE-L 36.90 #20 of 26 Archive leaderboard report
Document Summarization CNN / Daily Mail ML + Intra-Attention (Paulus et al., 2017) ROUGE-1 38.30 #24 of 26 Archive leaderboard report
Document Summarization CNN / Daily Mail ML + Intra-Attention (Paulus et al., 2017) ROUGE-2 14.81 #24 of 26 Archive leaderboard report
Document Summarization CNN / Daily Mail ML + Intra-Attention (Paulus et al., 2017) ROUGE-L 35.49 #24 of 26 Archive leaderboard report
Text Summarization CNN / Daily Mail (Anonymized) ML+RL, with intra-attention ROUGE-1 39.87 #6 of 13 Archive leaderboard report
Text Summarization CNN / Daily Mail (Anonymized) ML+RL, with intra-attention ROUGE-2 15.82 #6 of 13 Archive leaderboard report
Text Summarization CNN / Daily Mail (Anonymized) ML+RL, with intra-attention ROUGE-L 36.90 #6 of 13 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections