Adaptive Multidocument Summarization Via Graph Representation Learning

The goal of multidocument summarization (MDS) is to generate a comprehensive and concise summary from multiple documents, which should not only be grammatically correct but also semantically contains the refined content of the overall texts. Existing summarizers based on sequential pretrained large language models often cognize documents as linear sequences, which overlook the hierarchical structure correlations of sentences and paragraphs within or between documents. Additionally, those models also have limitations in handling long text input. To alleviate these two problems, a multidocument summarization model is proposed, with a heterogeneous graph of sentences, paragraphs and documents, called HeterMDS, to uncover deep semantic meanings and local–global context within documents. By integrating large language model and graph encoder with bootstrapped graph latents, the proposed HeterMDS can learn a semantically rich document representation and generate a coherent, concise and fact-consistent summary. It can be flexibly applied to current pretrained language models, effectively improving their performance in MDS. Extensive experiment results can verify the effectiveness of the proposed HeterMDS and its contained modules, and demonstrate its competitiveness against the state-of-the-art models.