The meta-learning paradigm has demonstrated significant effectiveness in few-shot text classification. Currently, numerous efforts are grounded in metric-based learning, utilizing textual feature vectors for classification, with a common emphasis on enlarging inter-class distances to achieve improved classification effectiveness. However, many methods predominantly focus on enhancing the separation of prototypes without taking the semantic relationships between prototypes and class clusters into consideration. This oversight results in incomplete and inaccurate encoding of prototypes within the semantic space, affecting the generality of the learned metric space. In this paper, we propose the utilization of Semantically Enhanced Labels for calibrating class Prototypes (SELP), thereby obtaining prototypes that are more separated and semantically accurate. Additionally, we have devised a center loss to enhance intra-class compactness, coupled with the introduction of a simulated label distribution method to address the overfitting problem. Extensive experiments on eight few-shot text classification datasets show that the proposed method outperforms baselines significantly. Our code is available at https://github.com/tttyyyzzz-zty/SELP.git.

Reasoning over the Temporal Knowledge Graph (TKG) that predicts facts in the future has received much attention. Most previous works attempt to model temporal dynamics with knowledge graphs and graph convolution networks. However, these methods lack the consideration of high-order interactions between objects in TKG, which is an important factor to predict future facts. To address this problem, we introduce dynamic hypergraph embedding for temporal knowledge graph reasoning. Specifically, we obtain high-order interactions by constructing hypergraphs based on temporal knowledge graphs at different timestamps. Besides, we integrate the differences caused by time into the hypergraph representation in order to fit TKG. Then, we adapt dynamic meta-embedding for temporal hypergraph representation that allows our model to choose the appropriate high-order interactions for downstream reasoning. Experimental results on public TKG datasets show that our method outperforms the baselines. Furthermore, the analysis part demonstrates that the proposed method brings good interpretation for the predicted results.