# Context Delivery on Context delivery 47-query (test)

Source: [Context Delivery on Context delivery 47-query (test)](<https://www.sota2.com/research/sota/context-delivery-on-context-delivery-47-query-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-14T12:22:47.795Z

Evaluation of model performance on context delivery capabilities using the 47-query test set, focusing on content leakage rates, noise ratio, and token waste.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Context Delivery |
| Dataset (including any stated split) | Context delivery 47-query (test) |
| Metric | Content Leak Rate (Cross-Domain) |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 2 of 2 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Content Leak Rate (Cross-Domain)](<https://www.sota2.com/research/sota/context-delivery-on-context-delivery-47-query-test.md?metric=Content%20Leak%20Rate%20(Cross-Domain)>): Lower is better; unit: PERCENT.

- [Noise Ratio (Irrelevant Units Delivered)](<https://www.sota2.com/research/sota/context-delivery-on-context-delivery-47-query-test.md?metric=Noise%20Ratio%20(Irrelevant%20Units%20Delivered)>): Lower is better; unit: PERCENT.

- [Total Content Leaks](<https://www.sota2.com/research/sota/context-delivery-on-context-delivery-47-query-test.md?metric=Total%20Content%20Leaks>): Lower is better; unit: SCALAR.

- [Token Waste](<https://www.sota2.com/research/sota/context-delivery-on-context-delivery-47-query-test.md?metric=Token%20Waste>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Content Leak Rate (Cross-Domain) | Result date | Report paper |
| --- | --- | --- | --- |
| Context Kubernetes | 21.6 | Not recorded | 69ddca7d7d0172140ec79f96 |
| Ungoverned context delivery | 26.5 | Not recorded | 69ddca7d7d0172140ec79f96 |

### Context Kubernetes

Content Leak Rate (Cross-Domain): 21.6. Result ID: 69de318547f34ce924c6fea4.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f96.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f96.md>)

Recorded evaluation configuration: {"Governance mode":"Governed"}. This record does not assert independent reproduction.

### Ungoverned context delivery

Content Leak Rate (Cross-Domain): 26.5. Result ID: 69de318547f34ce924c6fea3.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca7d7d0172140ec79f96.md>)

Recorded evaluation configuration: {"Governance mode":"Ungoverned"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
