試験
通常
試験時間: 00:00問題時間: 00:00
問題一覧
問題 13 / 65
Which method is typically used to project images and text into the same embedding space in order to link product images and product descriptions for search and recommendation?
①Variational Autoencoder (VAE)
②Multimodal embeddings using contrastive learning (e.g., CLIP)
③Natural language generation (large-scale language models)
④K-means clustering
⑤A single convolutional neural network (CNN)