Individual skill

reranking

Rerank retrieval candidates with cross-encoder models — two-stage pipelines, model selection, latency trade-offs, and evaluation. Use when first-stage retrieval is broad but imprecise.

skills/reranking/SKILL.md

Read pinned source

Source commit faf37c0859ebf627bfcb5225e8057d088691e0ae

What was checked

Matches the observed repository HEAD: Yes
Scanner rules are current: No
Repository scan complete: No
Skill scan complete: Yes
Observed behavior: Not checked yet
Last checked: Nov 8, 2026

These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.

Install this skill

Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.

An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.

Repository context

Repository stars (not skill ratings): 54
Repository last updated: Oct 10, 2026
License: MIT

Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.

agent-skills Repository topic
ai-agents Repository topic
ai-agents-automation Repository topic
ai-skills Repository topic
awesome Repository topic
awesome-list Repository topic
chatgpt Repository topic
claude-skills Repository topic
meta-ai Repository topic
muse-ai Repository topic
engineering Derived category

Files and excerpt

File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.

Browse files

SKILL.mdcore · 3,766 bytes

Read excerpt


# Reranking

First-stage retrieval is built for speed over a huge corpus; it's approximate by design. Reranking 
applies a slower, more precise model to the top candidates — a cross-encoder that reads the query 
and document together — and reorders them. It's the highest-ROI step in most retrieval pipelines.

## Overview

The two-stage pattern: retrieve broadly (top 50–200 from bi-encoder/hybrid search), then rerank 
to the top 5–10 with a cross-encoder. Cross-encoders score query-document pairs jointly — far 
more accurate than separate embeddings, far too slow for the full corpus. The trade is latency for 
precision; the art is sizing the candidate set and choosing the reranker for your quality/latency 
budget.