# String-level response similarity on RA-QA Single-Verify, Discriminative tasks

Source: [String-level response similarity on RA-QA Single-Verify, Discriminative tasks](<https://www.sota2.com/research/sota/string-level-response-similarity-on-ra-qa-single-verify-discriminative-tasks>)

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

Data updated: 2026-03-09T09:15:10.852Z

Evaluation of model performance on the RA-QA Single-Verify subset focusing on string-level response similarity, measured using metrics like BERTScore and METEOR.

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 | String-level response similarity |
| Dataset (including any stated split) | RA-QA Single-Verify, Discriminative tasks |
| Metric | BERTScore |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [BERTScore](<https://www.sota2.com/research/sota/string-level-response-similarity-on-ra-qa-single-verify-discriminative-tasks.md?metric=BERTScore>): Higher is better; unit: PERCENT.

- [METEOR](<https://www.sota2.com/research/sota/string-level-response-similarity-on-ra-qa-single-verify-discriminative-tasks.md?metric=METEOR>): Higher is better; unit: PERCENT.

## Recorded results

| Method | BERTScore | Result date | Report paper |
| --- | --- | --- | --- |
| RAMoEA-QA | 94 | Not recorded | 69ae54062d603e6de3e42be2 |
| CareAQA-operaGT | 91 | Not recorded | 69ae54062d603e6de3e42be2 |
| CareAQA-operaCT | 87 | Not recorded | 69ae54062d603e6de3e42be2 |
| PENGI | 3 | Not recorded | 69ae54062d603e6de3e42be2 |

### RAMoEA-QA

BERTScore: 94. Result ID: 69ae8f896dbb7075b956ef60.

[Reporting paper](<https://www.sota2.com/research/paper/69ae54062d603e6de3e42be2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ae54062d603e6de3e42be2.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CareAQA-operaGT

BERTScore: 91. Result ID: 69ae8f896dbb7075b956ef5f.

[Reporting paper](<https://www.sota2.com/research/paper/69ae54062d603e6de3e42be2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa790b1e209131cea7bdb0.md>)

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

### CareAQA-operaCT

BERTScore: 87. Result ID: 69ae8f896dbb7075b956ef5e.

[Reporting paper](<https://www.sota2.com/research/paper/69ae54062d603e6de3e42be2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa790b1e209131cea7bdb0.md>)

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

### PENGI

BERTScore: 3. Result ID: 69ae8f896dbb7075b956ef5d.

[Reporting paper](<https://www.sota2.com/research/paper/69ae54062d603e6de3e42be2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71d71e209131cea534a3.md>)

[Implementation](<https://github.com/microsoft/Pengi>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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