Syntactically Look-Ahead Attention Network for Sentence Compression
Syntactically Look-Ahead Attention Network for Sentence Compression
复制标题
DOI：
10.1609/aaai.v34i05.6315
复制
发表时间：
2020-02
期刊：
影响因子：
--
通讯作者：
Hidetaka Kamigaito;M. Okumura
中科院分区：
文献类型：
--
作者：
Hidetaka Kamigaito;M. Okumura
句子压缩是通过删除冗余词将长句子压缩成短句子的任务。在基于序列到序列（Seq2Seq）的模型中，解码器单向决定保留或删除单词。因此，它通常不能显式地捕获解码单词和未见单词之间的关系，这些单词将在未来的时间步骤中被解码。因此，为了避免产生不合语法的句子，解码器有时会在压缩句子时省略重要的单词。为了解决这个问题，我们提出了一种新的Seq2Seq模型，即句法前瞻性注意网络（SLAHAN），该模型可以通过在解码过程中明确跟踪依赖的父词和子词，并捕获未来将被解码的重要词，从而生成信息丰富的摘要。在谷歌句子压缩数据集上的自动评价结果表明，SLAHAN的基于保留标记的f1、ROUGE-1、ROUGE-2和ROUGE-L的得分分别为85.5、79.3、71.3和79.1。SLAHAN还提高了长句子的摘要性能。此外，在人类评价中，SLAHAN在不失去可读性的情况下提高了信息量。
Sentence compression is the task of compressing a long sentence into a short one by deleting redundant words. In sequence-to-sequence (Seq2Seq) based models, the decoder unidirectionally decides to retain or delete words. Thus, it cannot usually explicitly capture the relationships between decoded words and unseen words that will be decoded in the future time steps. Therefore, to avoid generating ungrammatical sentences, the decoder sometimes drops important words in compressing sentences. To solve this problem, we propose a novel Seq2Seq model, syntactically look-ahead attention network (SLAHAN), that can generate informative summaries by explicitly tracking both dependency parent and child words during decoding and capturing important words that will be decoded in the future. The results of the automatic evaluation on the Google sentence compression dataset showed that SLAHAN achieved the best kept-token-based-F1, ROUGE-1, ROUGE-2 and ROUGE-L scores of 85.5, 79.3, 71.3 and 79.1, respectively. SLAHAN also improved the summarization performance on longer sentences. Furthermore, in the human evaluation, SLAHAN improved informativeness without losing readability.