Adapting Neural Single-Document Summarization Model for Abstractive Multi-Document Summarization: A Pilot Study

Abstract

Till now, neural abstractive summarization methods have achieved great success for single document summarization (SDS). However, due to the lack of large scale multi-document summaries, such methods can be hardly applied to multi-document summarization (MDS). In this paper, we investigate neural abstractive methods for MDS by adapting a state-of-the-art neural abstractive summarization model for SDS. We propose an approach to extend the neural abstractive model trained on large scale SDS data to the MDS task. Our approach only makes use of a small number of multi-document summaries for fine tuning. Experimental results on two benchmark DUC datasets demonstrate that our approach can outperform a variety of baseline neural models.

Multi-News: A Large-Scale Multi-Document Summarization Dataset and Abstractive Hierarchical Model

“…The methods most related to this work are SDS adapted for MDS data. Zhang et al (2018a) adopts a hierarchical encoding framework trained on SDS data to MDS data by adding an additional document-level encoding. Baumel et al (2018) incorporates query relevance into standard sequence-to-sequence models.…”

Multi-document Summarization with Maximal Marginal Relevance-guided Reinforcement Learning

“…We take the MDS datasets from DUC and TAC competitions which are widely used in prior studies (Kulesza and Taskar, 2012;Lebanoff et al, 2018). Following convention Cao et al, 2017;Cho et al, 2019) Lebanoff et al, 2018;Zhang et al, 2018;Cho et al, 2019), we measure ROUGE-1/2/SU4 F1 scores (Lin, 2004). The evaluation parameters are set according to Hong et al (2014) with stemming and stopwords not removed.…”

“…Classical MDS explore both extractive (Erkan and Radev, 2004;Haghighi and Vanderwende, 2009) and abstractive methods (Barzilay et al, 1999;Ganesan et al, 2010). Many neural MDS methods (Yasunaga et al, 2017;Zhang et al, 2018) are merely comparable or even worse than classical methods due to the challenges of large search space and limited training data. Unlike DPP-Caps-Comb (Cho et al, 2019) that incorporates neural measures into classical MDS as features, RL-MMR opts for the opposite by endowing SDS methods with the capability to conduct MDS, enabling the potential of further improvement with advances in SDS.…”

“…Bridging SDS and MDS. Initial trials adapting SDS models to MDS (Lebanoff et al, 2018;Zhang et al, 2018) directly reuse SDS models (See et al, 2017;Tan et al, 2017). To deal with the large search space, a sentence ranker is used in the adapted models for candidate pruning.…”

“…There have been attempts to solve the aforementioned challenges in MDS. Regarding the large search space, prior studies (Lebanoff et al, 2018;Zhang et al, 2018) perform sentence filtering using a sentence ranker and only take top-ranked K sentences. However, such a hard cutoff of the search space makes these approaches insufficient in the exploration of the (already scarce) labeled data and limited by the ranker since most sentences are discarded, 2 albeit the discarded sentences are important and could have been favored.…”