Back to Learn
Deep Dive

Cross-Encoder

A neural network architecture used in RAG reranking where a query and a document are fed into the model simultaneously to compute a high-precision relevance score. Unlike Bi-Encoders, it allows for full self-attention across both inputs, resulting in superior accuracy at the cost of high computational latency and the inability to pre-compute embeddings.

Definition

A neural network architecture used in RAG reranking where a query and a document are fed into the model simultaneously to compute a high-precision relevance score. Unlike Bi-Encoders, it allows for full self-attention across both inputs, resulting in superior accuracy at the cost of high computational latency and the inability to pre-compute embeddings.

Disambiguation

Used for re-scoring a small candidate set, not for initial vector database retrieval.

Visual Metaphor

"A magnifying glass examining a key and a lock side-by-side to ensure every groove matches perfectly, rather than just checking their general shape."

Conceptual Overview

A neural network architecture used in RAG reranking where a query and a document are fed into the model simultaneously to compute a high-precision relevance score. Unlike Bi-Encoders, it allows for full self-attention across both inputs, resulting in superior accuracy at the cost of high computational latency and the inability to pre-compute embeddings.

Disambiguation

Used for re-scoring a small candidate set, not for initial vector database retrieval.

Visual Analog

A magnifying glass examining a key and a lock side-by-side to ensure every groove matches perfectly, rather than just checking their general shape.

Related Articles