A task-specific text embedding model that generates customized embeddings based on natural language instructions. INSTRUCTOR achieves state-of-the-art performance on 70 diverse embedding tasks by allowing users to specify the task objective and domain.
더 불러오는 중......
이 페이지에서
INSTRUCTOR
INSTRUCTOR is an instruction-finetuned text embedding model that generates domain-specific and task-aware embeddings without any fine-tuning, simply by providing task instructions.
Unlike traditional embedding models with fixed representations, INSTRUCTOR computes embeddings based on instructions explaining the use case (e.g., "Represent the Science question for retrieving supporting documents").
State-of-the-art on 70 diverse embedding tasks (MTEB leaderboard)
Hugging Face, LangChain, Haystack support
이와 관련된 다른 항목을 둘러보세요