Jun Suzuki, Masaaki Nagata
This paper tackles the reduction of redundant repeating generation that is often observed in RNN-based encoder-decoder models. Our basic idea is to jointly estimate the upper-bound frequency of each target vocabulary in the encoder and control the output words based on the estimation in the decoder. Our method shows significant improvement over a strong RNN-based encoder-decoder baseline and achieved its best results on an abstractive summarization benchmark.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Text Summarization | DUC 2004 Task 1 | ROUGE-1 | 32.28 | EndDec+WFE |
| Text Summarization | DUC 2004 Task 1 | ROUGE-2 | 10.54 | EndDec+WFE |
| Text Summarization | DUC 2004 Task 1 | ROUGE-L | 27.8 | EndDec+WFE |
| Text Summarization | GigaWord | ROUGE-1 | 36.3 | EndDec+WFE |
| Text Summarization | GigaWord | ROUGE-2 | 17.31 | EndDec+WFE |
| Text Summarization | GigaWord | ROUGE-L | 33.88 | EndDec+WFE |