# Correlation analysis with ground truth on Evaluation Dataset 2000 samples

Source: [Correlation analysis with ground truth on Evaluation Dataset 2000 samples](<https://www.sota2.com/research/sota/correlation-analysis-with-ground-truth-on-evaluation-dataset-2000-samples>)

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

Data updated: 2026-03-05T09:18:01.474Z

Evaluation of model performance using correlation metrics (Pearson and Spearman) against ground truth on the Evaluation Dataset 2000 samples.

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 | Correlation analysis with ground truth |
| Dataset (including any stated split) | Evaluation Dataset 2000 samples |
| Metric | Pearson Correlation Coefficient |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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

- [Pearson Correlation Coefficient](<https://www.sota2.com/research/sota/correlation-analysis-with-ground-truth-on-evaluation-dataset-2000-samples.md?metric=Pearson%20Correlation%20Coefficient>): Higher is better; unit: SCALAR.

- [Spearman Correlation Coefficient](<https://www.sota2.com/research/sota/correlation-analysis-with-ground-truth-on-evaluation-dataset-2000-samples.md?metric=Spearman%20Correlation%20Coefficient>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Pearson Correlation Coefficient | Result date | Report paper |
| --- | --- | --- | --- |
| sts\_paraphrase | 0.754 | Not recorded | 69a90e182d603e6de3e4091b |
| median over evaluators | 0.749 | Not recorded | 69a90e182d603e6de3e4091b |
| Semantic quality | 0.733 | Not recorded | 69a90e182d603e6de3e4091b |
| Composite | 0.513 | Not recorded | 69a90e182d603e6de3e4091b |
| Structure quality | 0.466 | Not recorded | 69a90e182d603e6de3e4091b |
| Agreement / uncertainty | -0.384 | Not recorded | 69a90e182d603e6de3e4091b |
| Query-output alignment | -0.437 | Not recorded | 69a90e182d603e6de3e4091b |

### sts\_paraphrase

Pearson Correlation Coefficient: 0.754. Result ID: 69a9494c6dbb7075b952fdec.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e182d603e6de3e4091b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66441e209131cea0f5ab.md>)

[Implementation](<https://github.com/UKPLab/sentence-transformers>)

Recorded evaluation configuration: {"category":"Best single evaluator"}. This record does not assert independent reproduction.

### median over evaluators

Pearson Correlation Coefficient: 0.749. Result ID: 69a9494c6dbb7075b952fded.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e182d603e6de3e4091b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69531e209131cea218fb.md>)

Recorded evaluation configuration: {"category":"Best consensus baseline"}. This record does not assert independent reproduction.

### Semantic quality

Pearson Correlation Coefficient: 0.733. Result ID: 69a9494c6dbb7075b952fdee.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e182d603e6de3e4091b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a90e182d603e6de3e4091b.md>)

Recorded evaluation configuration: {"type":"dimension"}. This record does not assert independent reproduction.

### Composite

Pearson Correlation Coefficient: 0.513. Result ID: 69a9494c6dbb7075b952fdeb.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e182d603e6de3e4091b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a90e182d603e6de3e4091b.md>)

Recorded evaluation configuration: {"weights":"default weights"}. This record does not assert independent reproduction.

### Structure quality

Pearson Correlation Coefficient: 0.466. Result ID: 69a9494c6dbb7075b952fdef.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e182d603e6de3e4091b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a90e182d603e6de3e4091b.md>)

Recorded evaluation configuration: {"type":"dimension"}. This record does not assert independent reproduction.

### Agreement / uncertainty

Pearson Correlation Coefficient: -0.384. Result ID: 69a9494c6dbb7075b952fdf1.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e182d603e6de3e4091b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a90e182d603e6de3e4091b.md>)

Recorded evaluation configuration: {"type":"dimension"}. This record does not assert independent reproduction.

### Query-output alignment

Pearson Correlation Coefficient: -0.437. Result ID: 69a9494c6dbb7075b952fdf0.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e182d603e6de3e4091b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a90e182d603e6de3e4091b.md>)

Recorded evaluation configuration: {"type":"dimension"}. This record does not assert independent reproduction.

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