the.ai

Retrieval / Representation

verified

Sparse Retrieval

The oldest thing that works: score a document by how many of the query's rare words it contains. Common words count for little, rare ones for a lot, and a word repeated ten times counts barely more than one repeated five. It has no idea what anything means, and it is still hard to beat.

Viz primitive · update-spectrumsaturation = 0.2

8 values. The left group decays steeply; the right group is 36% of the way to flat, and reads flatter than the left.

Term-frequency contributions to a score, before and after BM25's saturation. Drag the saturation up to watch a repeated term stop paying for itself.

0.2

Reviewed by opendroid · 2026-08-04

  • arXiv:2104.08663 — BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Models