From the source
Summary by forck
Hugging Face introduces Matryoshka embedding models, which allow embeddings to be truncated to smaller dimensions while retaining performance, improving efficiency for downstream tasks.
🪆 Introduction to Matryoshka Embedding Models
Research paper
Original post · huggingface.co ↗
From the source
These Matryoshka embedding models are trained such that these small truncated embeddings would still be useful. In short, Matryoshka embedding models can produce useful embeddings of various dimensions.
huggingface.co
