Multi-label intent detection aims to assign multiple labels to utterances and attracts increasing attention as a practical task in task-oriented dialogue systems. As dialogue domains change rapidly and new intents emerge fast, the lack of annotated data motivates multi-label few-shot intent detection. However, previous studies are confused by the identical representation of the utterance with multiple labels and overlook the intrinsic intra-class and inter-class interactions. To address these two limitations, we propose a novel dual class knowledge propagation network in this paper. In order to learn well-separated representations for utterances with multiple intents, we first introduce a label-semantic augmentation module incorporating class name information. For better consideration of the inherent intra-class and inter-class relations, an instance-level and a class-level graph neural network are constructed, which not only propagate label information but also propagate feature structure. And we use a simple yet effective method to predict the intent count of each utterance. Extensive experimental results on two multi-label intent datasets have demonstrated that our proposed method outperforms strong baselines by a large margin.

Aspect sentiment triplet extraction (ASTE) is an essential task, which aims to extract triplets(aspect, opinion, sentiment). However, overlapped triplets, especially multi-overlap triplets,make ASTE a challenge. Most existing methods suffer from multi-overlap triplets becausethey focus on the single interactions between an aspect and an opinion. To solve the aboveissues, we propose a novel multi-overlap triplet extraction method, which decodes the complexrelations between multiple aspects and opinions by learning their cooperative interactions. Overall, the method is based on an encoder-decoder architecture. During decoding, we design ajoint decoding mechanism, which employs a multi-channel strategy to generate aspects andopinions through the cooperative interactions between them jointly. Furthermore, we constructa correlation-enhanced network to reinforce the interactions between related aspectsand opinions for sentiment prediction. Besides, a relation-wise calibration scheme is adoptedto further improve performance. Experiments show that our method outperforms baselines,especially multi-overlap triplets.

This paper describes our participation in SemEval 2018 Task 2: Multilingual Emoji Prediction, in which participants are asked to predict a tweet’s most associated emoji from 20 emojis. Instead of regarding it as a 20-class classification problem we regard it as a text similarity problem. We propose a vector similarity based approach for this task. First the distributed representation (tweet vector) for each tweet is generated, then the similarity between this tweet vector and each emoji’s embedding is evaluated. The most similar emoji is chosen as the predicted label. Experimental results show that our approach performs comparably with the classification approach and shows its advantage in classifying emojis with similar semantic meaning.