上一条：Conditional Variational Autoencoder for Sign Language Translation with Cross-Modal Alignment

下一条：Bridging the Domain Gaps in Context Representations for <i>k</i>-Nearest Neighbor Neural Machine Translation