# Cross-lingual Sentence Classification on XNLI Language Transfer (test)

Source: [Cross-lingual Sentence Classification on XNLI Language Transfer (test)](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test>)

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

Data updated: 2026-02-26T06:09:07.228Z

Evaluation of cross-lingual sentence classification performance on the test split of the XNLI Language Transfer dataset, reporting per-language accuracy scores.

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 | Cross-lingual Sentence Classification |
| Dataset (including any stated split) | XNLI Language Transfer (test) |
| Metric | ar |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [ar](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=ar>): Higher is better; unit: PERCENT.

- [bg](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=bg>): Higher is better; unit: PERCENT.

- [de](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=de>): Higher is better; unit: PERCENT.

- [el](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=el>): Higher is better; unit: PERCENT.

- [en](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=en>): Higher is better; unit: PERCENT.

- [es](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=es>): Higher is better; unit: PERCENT.

- [fr](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=fr>): Higher is better; unit: PERCENT.

- [hi](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=hi>): Higher is better; unit: PERCENT.

- [ru](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=ru>): Higher is better; unit: PERCENT.

- [sw](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=sw>): Higher is better; unit: PERCENT.

- [th](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=th>): Higher is better; unit: PERCENT.

- [tr](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=tr>): Higher is better; unit: PERCENT.

- [ur](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=ur>): Higher is better; unit: PERCENT.

- [vi](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=vi>): Higher is better; unit: PERCENT.

- [zh](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=zh>): Higher is better; unit: PERCENT.

- [Average Accuracy](<https://www.sota2.com/research/sota/cross-lingual-sentence-classification-on-xnli-language-transfer-test.md?metric=Average%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | ar | Result date | Report paper |
| --- | --- | --- | --- |
| BEPA Bilingual | 71.7 | Not recorded | 695b455a53fdd5c361742e54 |
| BEPA Monolingual | 70.7 | Not recorded | 695b455a53fdd5c361742e54 |
| XLM-RoBERTa | 70.5 | Not recorded | 695b455a53fdd5c361742e54 |

### BEPA Bilingual

ar: 71.7. Result ID: 6996a8195698665ec4f1eed2.

[Reporting paper](<https://www.sota2.com/research/paper/695b455a53fdd5c361742e54.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695b455a53fdd5c361742e54.md>)

Recorded evaluation configuration: {"evaluation\_mode":"zero-shot","training\_set":"English","seeds":"5"}. This record does not assert independent reproduction.

### BEPA Monolingual

ar: 70.7. Result ID: 6996a8195698665ec4f1eed3.

[Reporting paper](<https://www.sota2.com/research/paper/695b455a53fdd5c361742e54.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695b455a53fdd5c361742e54.md>)

Recorded evaluation configuration: {"evaluation\_mode":"zero-shot","training\_set":"English","seeds":"5"}. This record does not assert independent reproduction.

### XLM-RoBERTa

ar: 70.5. Result ID: 6996a8195698665ec4f1eed1.

[Reporting paper](<https://www.sota2.com/research/paper/695b455a53fdd5c361742e54.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695b455a53fdd5c361742e54.md>)

Recorded evaluation configuration: {"evaluation\_mode":"zero-shot","training\_set":"English","seeds":"5"}. This record does not assert independent reproduction.

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