EdTech Tracker

← Coding lens

relevance score

technique judged 65% likely a named technique or architecture

First seen
2024-06-11
Independent sources
6
Articles
13
Last 4 weeks vs prior
0 vs 0

See how it has spread across media tiers

Mentions over time

Articles per week (by publish date) using the term, in any spelling.
0242024-W242026-W242026-W36
Table view
WeekArticlesSources
2024-W2411
2026-W0711
2026-W0921
2026-W1141
2026-W2411
2026-W3211
2026-W3511
2026-W3622

In their words

Then a cross-encoder reranker reads the query and each candidate together, producing a much sharper relevance score by attending across both texts at once.

Under the hood: the anatomy of a remembered turn Retrieval — four sources, searched in parallel Before the model runs, retrieveMemories() fans out across four memory sources concurrently: How it works: the anatomy of a remembered turn In provider and middleware mode, every model call passes through the same lifecycle: The results are deduplicated by content, ranked by relevance score when the back

At query time, your query is also encoded as a set, and the relevance score is the sum of best matches between query tokens and document tokens (a quantity called MaxSim).

In our baseline run, the context-relevance score sat at 0.92 out of 1.00 while four answers in ten came back wrong, because duplicate chunks and one outdated record were filling the five results the agent could read.

A cross-encoder sees both the query and document simultaneously, computing a joint relevance score.

If retrieved chunks have high relevance scores and the query is a direct factual lookup, route to a cheap model.

When to use semantic chunking: Narrative documents (blog posts, research papers, support articles, PDFs), when your embedding-based evaluation shows low context relevance scores, or when document topics shift frequently within a single page.

This is fast but approximate — the similarity scores are directional, not precise relevance scores.