Multi-Document Summarization with Determinantal Point Processes and Contextualized Representations
This work addresses extractive summarization for NLP researchers, offering an incremental improvement by blending deep learning with traditional methods.
The paper tackled the problem of improving multi-document summarization by integrating deep contextualized representations with determinantal point processes, finding that combining these representations with surface indicators is necessary for effective sentence selection, achieving competitive results on benchmark datasets.
Emerged as one of the best performing techniques for extractive summarization, determinantal point processes select the most probable set of sentences to form a summary according to a probability measure defined by modeling sentence prominence and pairwise repulsion. Traditionally, these aspects are modelled using shallow and linguistically informed features, but the rise of deep contextualized representations raises an interesting question of whether, and to what extent, contextualized representations can be used to improve DPP modeling. Our findings suggest that, despite the success of deep representations, it remains necessary to combine them with surface indicators for effective identification of summary sentences.