# Semantic Textual Similarity on STS Benchmark Slovak (val)

Source: [Semantic Textual Similarity on STS Benchmark Slovak (val)](<https://www.sota2.com/research/sota/semantic-textual-similarity-on-sts-benchmark-slovak-val>)

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

Data updated: 2026-02-26T05:57:40.796Z

Evaluation of Semantic Textual Similarity performance on the validation split of the Slovak version of the STS Benchmark dataset, typically measured using Pearson correlation.

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 | Semantic Textual Similarity |
| Dataset (including any stated split) | STS Benchmark Slovak (val) |
| Metric | Pearson Correlation |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 33 of 33 |
| 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](<https://www.sota2.com/research/sota/semantic-textual-similarity-on-sts-benchmark-slovak-val.md?metric=Pearson%20Correlation>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Pearson Correlation | Result date | Report paper |
| --- | --- | --- | --- |
| NLPCloud | 0.824 | Not recorded | 69842432a8b01e76cc8ee854 |
| GPT-4 | 0.78 | Not recorded | 69842432a8b01e76cc8ee854 |
| OpenAI full-text embeddings | 0.756 | Not recorded | 69842432a8b01e76cc8ee854 |
| SlovakBERT | 0.7537 | Not recorded | 69842432a8b01e76cc8ee854 |
| Gradient Boosting Regression | 0.685 | Not recorded | 69842432a8b01e76cc8ee854 |
| XGBoost | 0.678 | Not recorded | 69842432a8b01e76cc8ee854 |
| Bayesian Ridge Regression | 0.672 | Not recorded | 69842432a8b01e76cc8ee854 |
| Linear Regression | 0.671 | Not recorded | 69842432a8b01e76cc8ee854 |
| Random Forest Regression | 0.654 | Not recorded | 69842432a8b01e76cc8ee854 |
| Support Vector Regression | 0.648 | Not recorded | 69842432a8b01e76cc8ee854 |
| Decision Tree Regression | 0.635 | Not recorded | 69842432a8b01e76cc8ee854 |
| Ochiai | 0.58 | Not recorded | 69842432a8b01e76cc8ee854 |
| Jaccard | 0.568 | Not recorded | 69842432a8b01e76cc8ee854 |
| Sorensen-Dice | 0.562 | Not recorded | 69842432a8b01e76cc8ee854 |
| Cosine | 0.562 | Not recorded | 69842432a8b01e76cc8ee854 |
| Overlap | 0.56 | Not recorded | 69842432a8b01e76cc8ee854 |
| OpenAI word-level embeddings | 0.547 | Not recorded | 69842432a8b01e76cc8ee854 |
| LCSSeq | 0.532 | Not recorded | 69842432a8b01e76cc8ee854 |
| Jaro | 0.523 | Not recorded | 69842432a8b01e76cc8ee854 |
| Jaro-Winkler | 0.502 | Not recorded | 69842432a8b01e76cc8ee854 |
| Levenshtein | 0.492 | Not recorded | 69842432a8b01e76cc8ee854 |
| LCSStr | 0.491 | Not recorded | 69842432a8b01e76cc8ee854 |
| Damerau-Levenshtein | 0.485 | Not recorded | 69842432a8b01e76cc8ee854 |
| Smith-Waterman | 0.456 | Not recorded | 69842432a8b01e76cc8ee854 |
| FastText | 0.42 | Not recorded | 69842432a8b01e76cc8ee854 |
| Needleman-Wunsch | 0.397 | Not recorded | 69842432a8b01e76cc8ee854 |
| DISCO | 0.347 | Not recorded | 69842432a8b01e76cc8ee854 |
| ESA | 0.319 | Not recorded | 69842432a8b01e76cc8ee854 |
| Path | 0.265 | Not recorded | 69842432a8b01e76cc8ee854 |
| Hamming | 0.233 | Not recorded | 69842432a8b01e76cc8ee854 |
| Leacock-Chodorow | 0.231 | Not recorded | 69842432a8b01e76cc8ee854 |
| Wu-Palmer | 0.184 | Not recorded | 69842432a8b01e76cc8ee854 |
| HAL | 0.126 | Not recorded | 69842432a8b01e76cc8ee854 |

### NLPCloud

Pearson Correlation: 0.824. Result ID: 69964b4c5698665ec48f989b.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Third-party tool"}. This record does not assert independent reproduction.

### GPT-4

Pearson Correlation: 0.78. Result ID: 69964b4c5698665ec48f989a.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Third-party tool"}. This record does not assert independent reproduction.

### OpenAI full-text embeddings

Pearson Correlation: 0.756. Result ID: 69964b4c5698665ec48f9899.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Third-party tool"}. This record does not assert independent reproduction.

### SlovakBERT

Pearson Correlation: 0.7537. Result ID: 69964b4c5698665ec48f989c.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dc81e209131cea3c287.md>)

Recorded evaluation configuration: {"Algorithm Type":"Third-party tool","fine-tuned":"true"}. This record does not assert independent reproduction.

### Gradient Boosting Regression

Pearson Correlation: 0.685. Result ID: 69964b4c5698665ec48f9897.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Custom ML model"}. This record does not assert independent reproduction.

### XGBoost

Pearson Correlation: 0.678. Result ID: 69964b4c5698665ec48f9898.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Custom ML model"}. This record does not assert independent reproduction.

### Bayesian Ridge Regression

Pearson Correlation: 0.672. Result ID: 69964b4c5698665ec48f9893.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Custom ML model"}. This record does not assert independent reproduction.

### Linear Regression

Pearson Correlation: 0.671. Result ID: 69964b4c5698665ec48f9892.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Custom ML model"}. This record does not assert independent reproduction.

### Random Forest Regression

Pearson Correlation: 0.654. Result ID: 69964b4c5698665ec48f9896.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Custom ML model"}. This record does not assert independent reproduction.

### Support Vector Regression

Pearson Correlation: 0.648. Result ID: 69964b4c5698665ec48f9894.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Custom ML model"}. This record does not assert independent reproduction.

### Decision Tree Regression

Pearson Correlation: 0.635. Result ID: 69964b4c5698665ec48f9895.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Custom ML model"}. This record does not assert independent reproduction.

### Ochiai

Pearson Correlation: 0.58. Result ID: 69964b4c5698665ec48f9889.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Term-based"}. This record does not assert independent reproduction.

### Jaccard

Pearson Correlation: 0.568. Result ID: 69964b4c5698665ec48f9885.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Term-based"}. This record does not assert independent reproduction.

### Sorensen-Dice

Pearson Correlation: 0.562. Result ID: 69964b4c5698665ec48f9886.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Term-based"}. This record does not assert independent reproduction.

### Cosine

Pearson Correlation: 0.562. Result ID: 69964b4c5698665ec48f9888.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Term-based"}. This record does not assert independent reproduction.

### Overlap

Pearson Correlation: 0.56. Result ID: 69964b4c5698665ec48f9887.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Term-based"}. This record does not assert independent reproduction.

### OpenAI word-level embeddings

Pearson Correlation: 0.547. Result ID: 69964b4c5698665ec48f988e.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Statistical"}. This record does not assert independent reproduction.

### LCSSeq

Pearson Correlation: 0.532. Result ID: 69964b4c5698665ec48f9883.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Character-based"}. This record does not assert independent reproduction.

### Jaro

Pearson Correlation: 0.523. Result ID: 69964b4c5698665ec48f987f.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Character-based"}. This record does not assert independent reproduction.

### Jaro-Winkler

Pearson Correlation: 0.502. Result ID: 69964b4c5698665ec48f9880.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Character-based"}. This record does not assert independent reproduction.

### Levenshtein

Pearson Correlation: 0.492. Result ID: 69964b4c5698665ec48f987d.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Character-based"}. This record does not assert independent reproduction.

### LCSStr

Pearson Correlation: 0.491. Result ID: 69964b4c5698665ec48f9884.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Character-based"}. This record does not assert independent reproduction.

### Damerau-Levenshtein

Pearson Correlation: 0.485. Result ID: 69964b4c5698665ec48f987e.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Character-based"}. This record does not assert independent reproduction.

### Smith-Waterman

Pearson Correlation: 0.456. Result ID: 69964b4c5698665ec48f9882.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Character-based"}. This record does not assert independent reproduction.

### FastText

Pearson Correlation: 0.42. Result ID: 69964b4c5698665ec48f988d.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa642f1e209131cea034d3.md>)

[Implementation](<https://fasttext.cc/>)

Recorded evaluation configuration: {"Algorithm Type":"Statistical"}. This record does not assert independent reproduction.

### Needleman-Wunsch

Pearson Correlation: 0.397. Result ID: 69964b4c5698665ec48f9881.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Character-based"}. This record does not assert independent reproduction.

### DISCO

Pearson Correlation: 0.347. Result ID: 69964b4c5698665ec48f988c.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Statistical"}. This record does not assert independent reproduction.

### ESA

Pearson Correlation: 0.319. Result ID: 69964b4c5698665ec48f988b.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Statistical"}. This record does not assert independent reproduction.

### Path

Pearson Correlation: 0.265. Result ID: 69964b4c5698665ec48f9890.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Knowledge-based"}. This record does not assert independent reproduction.

### Hamming

Pearson Correlation: 0.233. Result ID: 69964b4c5698665ec48f987c.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Character-based"}. This record does not assert independent reproduction.

### Leacock-Chodorow

Pearson Correlation: 0.231. Result ID: 69964b4c5698665ec48f9891.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Knowledge-based"}. This record does not assert independent reproduction.

### Wu-Palmer

Pearson Correlation: 0.184. Result ID: 69964b4c5698665ec48f988f.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Knowledge-based"}. This record does not assert independent reproduction.

### HAL

Pearson Correlation: 0.126. Result ID: 69964b4c5698665ec48f988a.

[Reporting paper](<https://www.sota2.com/research/paper/69842432a8b01e76cc8ee854.md>)

Recorded evaluation configuration: {"Algorithm Type":"Statistical"}. This record does not assert independent reproduction.

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