[説明] このページは機械翻訳された日本語版です。内容に誤りがございましたら、報告していただけると助かります。

Rerankers

Addedv2.4.x

Reranking models for a sophisticated evaluation and refinement on retrieved results.

BGERerankFunction [READ MORE]

BGERerankFunction is a class in milvusmodel that takes a query and document as input and directly returns a similarity score instead of embeddings. This functionality uses the underlying BGE reranking model.

CohereRerankFunction [READ MORE]

CohereRerankFunction is a class in milvusmodel that takes a query and document as input and directly returns a similarity score instead of embeddings. This functionality uses the underlying Cohere reranking model.

CrossEncoderRerankFunction [READ MORE]

CrossEncoderRerankFunction is a class in milvusmodel that takes a query and document as input and directly returns a similarity score instead of embeddings. This functionality uses the underlying Cross-Encoder reranking model.

JinaRerankFunction [READ MORE]

JinaRerankFunction is a class in milvusmodel that takes a query and document as input and directly returns a similarity score instead of embeddings. This functionality uses the underlying Jina AI reranking model.

VoyageRerankFunction [READ MORE]

VoyageRerankFunction is a class in milvusmodel that takes a query and document as input and directly returns a similarity score instead of embeddings. This functionality uses the underlying Voyage reranking model.