Deep Communicating Agents for Abstractive Summarization
نویسندگان
چکیده
We present deep communicating agents in an encoder-decoder architecture to address the challenges of representing a long document for abstractive summarization. With deep communicating agents, the task of encoding a long text is divided across multiple collaborating agents, each in charge of a subsection of the input text. These encoders are connected to a single decoder, trained end-to-end using reinforcement learning to generate a focused and coherent summary. Empirical results demonstrate that multiple communicating encoders lead to a higher quality summary compared to several strong baselines, including those based on a single encoder or multiple non-communicating encoders.
منابع مشابه
Deep Recurrent Generative Decoder for Abstractive Text Summarization
We propose a new framework for abstractive text summarization based on a sequence-to-sequence oriented encoderdecoder model equipped with a deep recurrent generative decoder (DRGN). Latent structure information implied in the target summaries is learned based on a recurrent latent random model for improving the summarization quality. Neural variational inference is employed to address the intra...
متن کاملNeural Abstractive Text Summarization
Abstractive text summarization is a complex task whose goal is to generate a concise version of a text without necessarily reusing the sentences from the original source, but still preserving the meaning and the key contents. We address this issue by modeling the problem as a sequence to sequence learning and exploiting Recurrent Neural Networks (RNNs). This work is a discussion about our ongoi...
متن کاملAutomatic Community Creation for Abstractive Spoken Conversations Summarization
Summarization of spoken conversations is a challenging task, since it requires deep understanding of dialogs. Abstractive summarization techniques rely on linking the summary sentences to sets of original conversation sentences, i.e. communities. Unfortunately, such linking information is rarely available or requires trained annotators. We propose and experiment automatic community creation usi...
متن کاملSmart Initialization Yields Better Convergence Properties in Deep Abstractive Summarization
Abstractive text summarization has been proposed as an alternative to the inherently limited extractive methods, but extant work is plagued with high training times. In this work, we introduce a set of extensions, including novel initialization techniques, that allow contemporary models to achieve 10x faster training time and comparable results. Our work also provides substantial evidence again...
متن کاملAbstractive Document Summarization with a Graph-Based Attentional Neural Model
Abstractive summarization is the ultimate goal of document summarization research, but previously it is less investigated due to the immaturity of text generation techniques. Recently impressive progress has been made to abstractive sentence summarization using neural models. Unfortunately, attempts on abstractive document summarization are still in a primitive stage, and the evaluation results...
متن کامل