A Generative Approach for Multi-Document Summarization using Semantic-Discursive information

Multi-document summarization is the automatic produ ction of a unique summary from a collection of texts. In this paper, we propose a statistical generative approach for multi-document summarization that combines simple information such as sentence positi on in the text and semantic-discursive information from CST (Cross-Doc ument Structure Theory). In particular, we formulate the multi-docu ment summarization task using a Noisy-Channel model.