What is reciprocal rank fusion?

Reciprocal rank fusion, or RRF, is a method for combining several ranked lists into one final ranking. Each item scores the sum of 1 divided by (k plus its rank) across every list it appears in, where k is a small constant, often 60. Items ranked highly in multiple lists rise to the top. RRF's advantage is that it needs only ranks, not comparable scores, so you can fuse results from sources whose relevance scores are on different scales, like several social platforms. SocialCrawl's universal search fans a query out across platforms, then fuses the per-source rankings with RRF before an LLM rerank, which is why one endpoint returns a single sensible cross-platform list.