Verified65% confidenceFactExact time
RRF (Reciprocal Rank Fusion) uses only the ranking position information from each retrieval path (formula RRF(d)=Σ1/(k+rank_i(d)), where k is typically set to 60), is insensitive to score distribution, and bypasses the cross-modal score alignment problem
3
Sources
65%
Confidence
Long-term
Relevance
7/7/2026
First Seen
Sources
企业级RAG知识库落地的5大工程硬伤：从Demo到生产的真实门槛
bilibili赋范课堂7/4/2026
Related Claims
VerifiedRRF does not require normalization of raw scores from different retrievers and relies only on rank information, but when multiple retrievers systematically make the same type of error, RRF only amplifies the mistakes76% similarUnverifiednDCG@5 takes into account the relevance grades and position discounting of the top 5 results, and normalizes by dividing by the ideal ranking score.67% similarUnverifiedThe preference data that RLHF relies on has systematic flaws such as low inter-annotator agreement, position bias, and length bias.64% similarUnverifiedNeither Recall@k nor Reciprocal Rank@k measures whether the final answer is correct, safe, or grounded; both only answer questions at the retrieval level.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/180996API
curl https://kongchang.com/api/v1/knowledge/claims/180996MCP
get_claim(id=180996)