A Field Guide & Benchmark for Similarity Search in RAG
How to benchmark sparse, dense, hybrid and late-interaction retrieval on your own corpus, and how to read the numbers without fooling yourself.
Retrieval and RAG02
Subject · 2 posts
How to benchmark sparse, dense, hybrid and late-interaction retrieval on your own corpus, and how to read the numbers without fooling yourself.
Retrieval and RAG02
What BM25, Recall@k, answer-in-context rate and token F1 actually measure, which number to fix first, and how to compute them without fooling yourself.
Retrieval and RAG01