docs: add "What is ordinal retrieval" framing to README

[Fieldnote-Echo]

What

Adds a ## What is ordinal retrieval section near the top of the README, above ## Benchmark at a glance.

The README previously jumped straight from what the crate is to the benchmark without ever stating the underlying idea. This supplies that frame, so the top now reads tagline → what the crate is → what the idea is → proof.

Content

Ordinal retrieval is a retrieval family where the index operates on order/sign quotients of embedding coordinates rather than metric magnitudes, with training-free compression, quotient-aware candidate generation, and asymmetric reranking against the original query signal.

Dense retrieval assumes magnitude geometry is the primitive. Ordinal retrieval asks what survives when embeddings are quotiented down to coordinate order, signs, and relative salience. Surprisingly, enough retrieval signal survives to match dense qrel quality at much smaller storage, and those quotients support quotient-aware candidate-generation rules that are training-free, append-friendly, and graph-optional.

Notes

• Adopts the "ordinal retrieval is a retrieval family" framing (contrasted with dense) as the public positioning.
• The "matches dense qrel quality at much smaller storage" line restates the existing in-repo BEIR result — it sits right above the hero blockquote that quantifies it (within BEIR qrel noise at 8–16× smaller) — not a paper number.
• Candidate-generation rules described as "quotient-aware" (not "new") to stay consistent with the README's existing FastScan / binary-quantization-lineage humility.

Test plan

• Prose-only change; the new section adds no code fences, so no doctests are introduced. CI markdown/link checks should pass.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

[qodo-code-review]

PR Summary by Qodo

Docs: add "What is ordinal retrieval" framing near top of README
📝 Documentation 🕐 Less than 5 minutes

Description

• Add a high-level "What is ordinal retrieval" section before the benchmark
• Explain ordinal vs dense retrieval in terms of order/sign quotients and reranking
• Clarify that the quality/storage claim restates the existing BEIR result

High-Level Assessment

Adding a short conceptual framing section immediately before the benchmark is the most effective approach for onboarding and avoids introducing new technical surfaces (e.g., diagrams or additional docs pages). Alternatives like a separate docs page or a longer tutorial would add navigation/maintenance overhead without improving the README’s above-the-fold clarity.

Files changed (1) +14 / -0

Documentation (1) +14 / -0

README.mdIntroduce an ordinal-retrieval conceptual framing section +14/-0

Introduce an ordinal-retrieval conceptual framing section

• Adds a new top-level section ("What is ordinal retrieval") above the benchmark to define ordinal retrieval, contrast it with dense retrieval, and summarize key properties (training-free compression, quotient-aware candidate generation, asymmetric reranking). Keeps the existing benchmark claim as the supporting evidence immediately below.

README.md

ⓘ You are approaching your monthly quota for Qodo. Upgrade your plan

[qodo-code-review]

Code Review by Qodo

🐞 Bugs (0) 📘 Rule violations (0) 📎 Requirement gaps (0)

Great, no issues found!

Qodo reviewed your code and found no material issues that require review

ⓘ You are approaching your monthly quota for Qodo. Upgrade your plan

[gemini-code-assist]

Code Review

This pull request updates the README.md to include a new section explaining the concept of ordinal retrieval. The review feedback suggests minor improvements to this section, specifically adding a question mark to the new heading for grammatical completeness and replacing 'order/sign' with 'ordinal and sign' to maintain consistency with the rest of the document.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.