SummaRuNNer: A Recurrent Neural Network based Sequence Model for Extractive Summarization of Documents
2016-11-14Code Available1· sign in to hype
Ramesh Nallapati, FeiFei Zhai, Bo-Wen Zhou
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/dennlinger/summariesnone★ 11
- github.com/amagooda/SummaRuNNer_coattentionpytorch★ 1
- github.com/no-execution/Summa_labelpytorch★ 0
- github.com/austnbell/TextSummarizationFederatedLearningnone★ 0
- github.com/kotyukov/huaweiNLPnone★ 0
- github.com/kedz/nnsumpytorch★ 0
- github.com/Snoopy666/Extractive-User-Review-Summaruzation-using-LSTMtf★ 0
Abstract
We present SummaRuNNer, a Recurrent Neural Network (RNN) based sequence model for extractive summarization of documents and show that it achieves performance better than or comparable to state-of-the-art. Our model has the additional advantage of being very interpretable, since it allows visualization of its predictions broken up by abstract features such as information content, salience and novelty. Another novel contribution of our work is abstractive training of our extractive model that can train on human generated reference summaries alone, eliminating the need for sentence-level extractive labels.
Tasks
Benchmark Results
| Dataset | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| CNN / Daily Mail (Anonymized) | SummaRuNNer | ROUGE-1 | 39.6 | — | Unverified |
| CNN / Daily Mail (Anonymized) | Lead-3 baseline | ROUGE-1 | 39.2 | — | Unverified |