# Semantic Textual Similarity on STS-B (Accuracy/Bias evaluation)

Source: [Semantic Textual Similarity on STS-B (Accuracy/Bias evaluation)](<https://www.sota2.com/research/sota/semantic-textual-similarity-on-sts-b-accuracy-bias-evaluation>)

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

Data updated: 2026-02-26T05:29:47.979Z

Evaluation of Semantic Textual Similarity performance on the STS-B dataset, measured using Accuracy and Bias metrics.

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-B |
| Metric | Accuracy |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/semantic-textual-similarity-on-sts-b-accuracy-bias-evaluation.md?metric=Accuracy>): Higher is better; unit: SCALAR.

- [Bias](<https://www.sota2.com/research/sota/semantic-textual-similarity-on-sts-b-accuracy-bias-evaluation.md?metric=Bias>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| EAR | 0.595 | Not recorded | 68fa738e1e209131cea5d2e0 |
| MABEL | 0.591 | Not recorded | 68fa738e1e209131cea5d2e0 |
| Vanilla-tuning | 0.578 | Not recorded | 68fa738e1e209131cea5d2e0 |
| MABEL | 0.57 | Not recorded | 68fa738e1e209131cea5d2e0 |
| Debiased-tuning | 0.518 | Not recorded | 68fa738e1e209131cea5d2e0 |
| EAR | 0.509 | Not recorded | 68fa738e1e209131cea5d2e0 |
| Vanilla-tuning | 0.507 | Not recorded | 68fa738e1e209131cea5d2e0 |
| ProSocialTuning | 0.494 | Not recorded | 68fa738e1e209131cea5d2e0 |
| Debiased-tuning | 0.473 | Not recorded | 68fa738e1e209131cea5d2e0 |
| ProSocialTuning | 0.46 | Not recorded | 68fa738e1e209131cea5d2e0 |

### EAR

Accuracy: 0.595. Result ID: 6992ff533a6c31484912f0a1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d2e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c101e209131cea31d71.md>)

Recorded evaluation configuration: {"Backbone":"RoBERTa-base"}. This record does not assert independent reproduction.

### MABEL

Accuracy: 0.591. Result ID: 6992ff533a6c31484912f0a2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d2e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dc51e209131cea3c093.md>)

Recorded evaluation configuration: {"Backbone":"RoBERTa-base"}. This record does not assert independent reproduction.

### Vanilla-tuning

Accuracy: 0.578. Result ID: 6992ff533a6c31484912f09f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d2e0.md>)

Recorded evaluation configuration: {"Backbone":"RoBERTa-base"}. This record does not assert independent reproduction.

### MABEL

Accuracy: 0.57. Result ID: 6992ff533a6c31484912f09d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d2e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dc51e209131cea3c093.md>)

Recorded evaluation configuration: {"Backbone":"BERT-base"}. This record does not assert independent reproduction.

### Debiased-tuning

Accuracy: 0.518. Result ID: 6992ff533a6c31484912f0a0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d2e0.md>)

Recorded evaluation configuration: {"Backbone":"RoBERTa-base"}. This record does not assert independent reproduction.

### EAR

Accuracy: 0.509. Result ID: 6992ff533a6c31484912f09c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d2e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c101e209131cea31d71.md>)

Recorded evaluation configuration: {"Backbone":"BERT-base"}. This record does not assert independent reproduction.

### Vanilla-tuning

Accuracy: 0.507. Result ID: 6992ff533a6c31484912f09a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d2e0.md>)

Recorded evaluation configuration: {"Backbone":"BERT-base"}. This record does not assert independent reproduction.

### ProSocialTuning

Accuracy: 0.494. Result ID: 6992ff533a6c31484912f0a3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d2e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d2e0.md>)

[Implementation](<https://github.com/MSU-NLP-CSS/ProSocialTuning>)

Recorded evaluation configuration: {"Backbone":"RoBERTa-base"}. This record does not assert independent reproduction.

### Debiased-tuning

Accuracy: 0.473. Result ID: 6992ff533a6c31484912f09b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d2e0.md>)

Recorded evaluation configuration: {"Backbone":"BERT-base"}. This record does not assert independent reproduction.

### ProSocialTuning

Accuracy: 0.46. Result ID: 6992ff533a6c31484912f09e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d2e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d2e0.md>)

[Implementation](<https://github.com/MSU-NLP-CSS/ProSocialTuning>)

Recorded evaluation configuration: {"Backbone":"BERT-base"}. This record does not assert independent reproduction.

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