Multimodal sentiment analysis is an important research area that predicts speaker's sentiment tendency through features extracted from textual, visual and acoustic modalities. e central challenge is the fusion method of the multimodal information. A variety of fusion methods have been proposed, but few of them adopt endto-end translation models to mine the subtle correlation between modalities. Enlightened by recent success of Transformer in the area of machine translation, we propose a new fusion method, TransModality, to address the task of multimodal sentiment analysis. We assume that translation between modalities contributes to a be er joint representation of speaker's u erance. With Transformer, the learned features embody the information both from the source modality and the target modality. We validate our model on multiple multimodal datasets: CMU-MOSI, MELD, IEMOCAP. e experiments show that our proposed method achieves the state-ofthe-art performance. CCS CONCEPTS •Information systems → Data mining; •Computing methodologies → Natural language processing; Arti cial intelligence;

The incorporation of data augmentation method in grammatical error correction task has attracted much attention. However, existing data augmentation methods mainly apply noise to tokens, which leads to the lack of diversity of generated errors. In view of this, we propose a new data augmentation method that can apply noise to the latent representation of a sentence. By editing the latent representations of grammatical sentences, we can generate synthetic samples with various error types. Combining with some pre-defined rules, our method can greatly improve the performance and robustness of existing grammatical error correction models. We evaluate our method on public benchmarks of GEC task and it achieves the state-of-the-art performance on CoNLL-2014 and FCE benchmarks.

No abstract