Visual Question Answering (VQA) is a multimodal task that uses natural language to ask and answer questions based on image content. For multimodal tasks, obtaining accurate modality feature information is crucial. The existing researches on the visual question answering model mainly start from the perspective of attention mechanism and multimodal fusion, which will tend to ignore the impact of modal interaction learning and the introduction of noise information in the process of modal fusion on the overall performance of the model. This paper proposes a novel and efficient multimodal adaptive gated mechanism model, MAGM. The model adds an adaptive gate mechanism to the intra- and inter-modality learning and the modal fusion process. This model can effectively filter irrelevant noise information, obtain fine-grained modal features, and improve the ability of the model to adaptively control the contribution of the two modal features to the predicted answer. In intra- and inter-modality learning modules, the self-attention gated and self-guided-attention gated units are designed to filter text and image features’ noise information effectively. In modal fusion module, the adaptive gated modal feature fusion structure is designed to obtain fine-grained modal features and improve the accuracy of the model in answering questions. Quantitative and qualitative experiments on the two VQA task benchmark datasets, VQA 2.0 and GQA, proved that the method in this paper is superior to the existing methods. The MAGM model has an overall accuracy of 71.30% on the VQA 2.0 dataset and an overall accuracy of 57.57% on the GQA dataset.

Chinese short-text classification is an essential primary research direction in natural language processing. The main problem facing the current Chinese short text classification task is how to effectively extract critical semantic feature information and determine the text category within a short text length. Aiming at the problem of feature information extraction of Chinese short texts, this paper introduces a novel sparse attention mechanism. It proposes a Chinese short text classification model RGSC combined with capsule networks. After the model generates word vectors through the RoBERTa pre-training model, the Bi-GRU model initially extracts text feature information. The profound text feature extraction module SC, composed of sparse attention mechanism and capsule networks, is used to further extract critical semantic feature information. The results of ablation experiments on the TouTiao Chinese short text dataset and performance comparison experiments with various models show that the RGSC model can effectively extract the essential text feature information while effectively reducing the irrelevant noise information contained in the text features, and obtain fine-grained text feature information.