Reciprocal Rank Fusion (RRF)
Parameter-Free Rank Merging for Lexical BM25 & Dense Semantic Search
#Search#Hybrid Search#RRF#Information Retrieval#RAG#Ranking
Choose Presentation Mode:
STAGE 1 / 7— Anti-Pattern
Section 1: The LLM Anti-Pattern vs Right-Sized Model
The Naive Generative LLM Approach:
Asking an LLM agent to merge two different search result lists and decide which document belongs at rank 1.
Why It Fails in Production:
LLMs suffer from severe ordering biases, slow down search responses by 2,000ms, and cost unnecessary money for a simple ranking formula.
Targeted Algorithm (Reciprocal Rank Fusion (RRF))
Latency:0.01ms
Cost / 1M Ops:$0.00
Determinism:100% Deterministic Fusion
Generative LLM Alternative
Latency:2,000ms
Cost / 1M Ops:$6,000
Determinism:Inconsistent Sorting