A Multilingual Study of Multi-Sentence Compression using Word Vertex-Labeled Graphs and Integer Linear Programming

Multi-Sentence Compression (MSC) aims to generate a short sentence with the\nkey information from a cluster of similar sentences. MSC enables summarization\nand question-answering systems to generate outputs combining fully formed\nsentences from one or several documents. This paper describes an Integer Linear\nProgramming method for MSC using a vertex-labeled graph to select different\nkeywords, with the goal of generating more informative sentences while\nmaintaining their grammaticality. Our system is of good quality and outperforms\nthe state of the art for evaluations led on news datasets in three languages:\nFrench, Portuguese and Spanish. We led both automatic and manual evaluations to\ndetermine the informativeness and the grammaticality of compressions for each\ndataset. In additional tests, which take advantage of the fact that the length\nof compressions can be modulated, we still improve ROUGE scores with shorter\noutput sentences.\n