In Conclusion Not Repetition: Comprehensive Abstractive Summarization With Diversified Attention Based On Determinantal Point Processes
This work addresses the issue of repetitive and low-abstraction summaries in abstractive summarization, which is important for applications like news aggregation and document analysis, though it appears incremental as it builds on existing Seq2Seq architectures.
The paper tackled the problem of generating comprehensive abstractive summaries by addressing degenerated attention distribution in Seq2Seq models, introducing a Diverse Convolutional Seq2Seq Model with Determinantal Point Processes that achieves higher comprehensiveness compared to vanilla models and strong baselines.
Various Seq2Seq learning models designed for machine translation were applied for abstractive summarization task recently. Despite these models provide high ROUGE scores, they are limited to generate comprehensive summaries with a high level of abstraction due to its degenerated attention distribution. We introduce Diverse Convolutional Seq2Seq Model(DivCNN Seq2Seq) using Determinantal Point Processes methods(Micro DPPs and Macro DPPs) to produce attention distribution considering both quality and diversity. Without breaking the end to end architecture, DivCNN Seq2Seq achieves a higher level of comprehensiveness compared to vanilla models and strong baselines. All the reproducible codes and datasets are available online.