Multimodal Emotion Recognition in Conversations (MERC) is an important topic in human-computer interaction. In the MERC task, conversations exhibit dynamic emotional dependency, including inter-speaker and intra-speaker emotional dependency, both are vital in understanding the content. However, current research primarily integrates these two emotional dependencies into one unified module, limiting the accuracy of MERC. In this paper, we propose a dynamic emotion-dependent network with relational subgraph interaction named DEDNet. DEDNet introduces relational subgraphs to separately model two emotional dependencies, enabling structured learning paths for utterances based on distinct emotional dependency types. Specifically, nodes indicate the utterances at different moments in the conversation, while edges define the emotional dependency and temporal relationships between nodes. To explicitly capture the differences between these two emotional dependencies, distinct subgraphs are designed for comprehensive representations. Furthermore, we propose an incremental interactive strategy, sequentially leveraging two emotional dependencies to learn the changes in dependency relationships. We find that modeling inter-speaker emotional dependency can better identify negative emotions and modeling intra-speaker emotional dependency can better recognize positive emotions. Experimental results demonstrate that our model outperforms current state-of-the-art methods on three benchmark datasets, IEMOCAP, MELD and DailyDialog. Our code is available at https://github.com/NLPAlchemist/DEDNet .

Many studies have demonstrated that incorporating lexical information into characters can effectively improve the performance of Chinese Named Entity Recognition (CNER). However, we argue that previous studies have not extensively explored the interactive relationship between characters and lexical information, and have only used the lexical information to enhance character-level representation. To address this limitation, we propose an interactive fusion approach that integrates characters and lexical information for CNER. Specifically, we first design graph attention networks to initially fuse character and lexical information within an interactive graph structure. Additionally, by introducing methods such as feedforward neural networks, residual connections, and layer normalization, the fusion effect of the graph attention network is further enhanced. Finally, concatenating and reducing dimensionality of character feature vectors and lexical feature vectors to achieve secondary fusion, thereby obtaining a more comprehensive feature representation. Experimental results on multiple datasets demonstrate that our proposed model outperforms other models that fuse lexical information. Particularly, on the CCKS2020 and Ontonotes datasets, our model achieves higher F1 scores than previous state-of-the-art models. The code is available via the link: https://github.com/wangye0523/The-interactive-fusion-of-characters-and-lexical-information-for-Chinese-named-entity-recognition.

Counterfactual subgraphs explain Graph Neural Networks (GNNs) by answering the question: “How would the prediction change if a certain subgraph were absent in the input instance?” The differentiable proxy adjacency matrix is prevalent in current counterfactual subgraph discovery studies due to its ability to avoid exhaustive edge searching. However, a prediction gap exists when feeding the proxy matrix with continuous values and the thresholded discrete adjacency matrix to GNNs, compromising the optimization of the subgraph generator. Furthermore, the end-to-end learning schema adopted in the subgraph generator limits the diversity of counterfactual subgraphs. To this end, we propose CF-GFNExplainer, a flow-based approach for learning counterfactual subgraphs. CF-GFNExplainer employs a policy network with a discrete edge removal schema to construct counterfactual subgraph generation trajectories. Additionally, we introduce a loss function designed to guide CF-GFNExplainer’s optimization. The discrete adjacency matrix generated in each trajectory eliminates the prediction gap, enhancing the validity of the learned subgraphs. Furthermore, the multi-trajectories sampling strategy adopted in CF-GFNExplainer results in diverse counterfactual subgraphs. Extensive experiments conducted on synthetic and real-world datasets demonstrate the effectiveness of the proposed method in terms of validity and diversity. The data and code of CF-GFNExplainer are available ^{1 1}https://github.com/AmGracee/CF-GFNExplainer .