With the rapid development of the Internet, the massive amount of web textual data has grown exponentially, which has brought considerable challenges to downstream tasks, such as document management, text classification, and information retrieval. Automatic text summarization (ATS) is becoming an extremely important means to solve this problem. The  core of ATS is to mine the gist of the original text and automatically generate a concise and readable summary. Recently, to better balance and develop these two aspects, deep learning (DL)-based abstractive summarization models have been developed. At present, for ATS tasks, almost all state-of-the-art (SOTA) models are based on DL architecture. However, a comprehensive literature survey is still lacking in the field of DL-based abstractive text summarization. To fill this gap, this paper provides researchers with a comprehensive survey of DL-based abstractive summarization. We first give an overview of abstractive summarization and DL. Then, we summarize several typical frameworks of abstractive summarization. After that, we also give a comparison of several popular datasets that are commonly used for training, validation, and testing. We further analyze the performance of several typical abstractive summarization systems on common datasets. Finally, we highlight some open challenges in the abstractive summarization task and outline some future research trends. We hope that these explorations will provide researchers with new insights into DL-based abstractive summarization.

No abstract

No abstract

The remarkable success of deep learning technologies has provided new ideas for solving complex tracking problems. It is difficult for traditional algorithms to directly estimate the trajectory vector and target class from the received signal due to the limitation of modelling ability, which causes inevitable information loss. Moreover, existing algorithms suffer severe performance degradation when dealing with problems that are difficult to mathematically model in advance, such as highly nonlinear observations and manoeuvring scenarios. To address these issues, we propose a deep learning algorithm for joint direct tracking and classification (DeepDTC), which is a novel direct tracking framework. Specifically, we construct a convolutional neural network (CNN)‐based signal processing component to capture observation features, and a Transformer‐based trajectory tracking component to capture the features of target state and identity. Meanwhile, in signal processing component, we design an attribute network to learn auxiliary knowledge features. Finally, we construct a multi‐task learning network to connect these components and estimate trajectory vectors and target classes simultaneously. Our algorithm takes the transformer with attention mechanism as the core network, which is highly scalable and suitable for introducing various auxiliary knowledge. The comparison experiments with traditional methods demonstrate the effectiveness and advancement of the proposed algorithm. The comparative experiments with the single‐task model show that DeepDTC can further improve the tracking accuracy by utilising the learnt classification information.

No abstract