modal matching tasks. In this work, we study the emergence of hubs in Visual Semantic Embeddings (VSE) with applica- tion to text-image matching.
確定! 回上一頁