# Natural Language Inference on XNLI Sw (dev)

Source: [Natural Language Inference on XNLI Sw (dev)](<https://www.sota2.com/research/sota/natural-language-inference-on-xnli-sw-dev>)

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

Data updated: 2026-02-26T05:29:53.460Z

Evaluation of cross-lingual Natural Language Inference performance on the Swahili (Sw) subset of the XNLI development split, typically measured by accuracy.

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 | Natural Language Inference |
| Dataset (including any stated split) | XNLI Sw (dev) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 18 of 18 |
| 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/natural-language-inference-on-xnli-sw-dev.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Tpl-div-k | 65.68 | Not recorded | 68fa74281e209131cea6014e |
| T-div-k | 64.44 | Not recorded | 68fa74281e209131cea6014e |
| Tpl-top-k | 64.38 | Not recorded | 68fa74281e209131cea6014e |
| Tpl-div-k | 63.88 | Not recorded | 68fa74281e209131cea6014e |
| Tpl-rand-k | 63.88 | Not recorded | 68fa74281e209131cea6014e |
| T-top-k | 63.82 | Not recorded | 68fa74281e209131cea6014e |
| T-top-k | 63.74 | Not recorded | 68fa74281e209131cea6014e |
| SRC+GEN | 63.52 | Not recorded | 68fa74281e209131cea6014e |
| SRC+GEN | 63.34 | Not recorded | 68fa74281e209131cea6014e |
| Tpl-rand-k | 63.24 | Not recorded | 68fa74281e209131cea6014e |
| T-div-k | 62.69 | Not recorded | 68fa74281e209131cea6014e |
| Tpl-top-k | 62.11 | Not recorded | 68fa74281e209131cea6014e |
| T-rand-k | 62.11 | Not recorded | 68fa74281e209131cea6014e |
| T-rand-k | 61.67 | Not recorded | 68fa74281e209131cea6014e |
| SRC | 59.16 | Not recorded | 68fa74281e209131cea6014e |
| SRC | 57.63 | Not recorded | 68fa74281e209131cea6014e |
| GEN | 56.71 | Not recorded | 68fa74281e209131cea6014e |
| GEN | 56.02 | Not recorded | 68fa74281e209131cea6014e |

### Tpl-div-k

Accuracy: 65.68. Result ID: 69933f7c3a6c31484914af9f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Implementation](<https://github.com/LLM-Based-Augmentations>)

Recorded evaluation configuration: {"Evaluation protocol":"translate-train","Supervision type":"hard prompt labels","Data selection strategy":"div-k"}. This record does not assert independent reproduction.

### T-div-k

Accuracy: 64.44. Result ID: 69933f7c3a6c31484914af9c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Implementation](<https://github.com/LLM-Based-Augmentations>)

Recorded evaluation configuration: {"Evaluation protocol":"translate-train","Supervision type":"teacher soft pseudolabels","Data selection strategy":"div-k"}. This record does not assert independent reproduction.

### Tpl-top-k

Accuracy: 64.38. Result ID: 69933f7c3a6c31484914af9d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Implementation](<https://github.com/LLM-Based-Augmentations>)

Recorded evaluation configuration: {"Evaluation protocol":"translate-train","Supervision type":"hard prompt labels","Data selection strategy":"top-k"}. This record does not assert independent reproduction.

### Tpl-div-k

Accuracy: 63.88. Result ID: 69933f7c3a6c31484914af45.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Implementation](<https://github.com/LLM-Based-Augmentations>)

Recorded evaluation configuration: {"selection strategy":"div-k","Zero-shot":"true"}. This record does not assert independent reproduction.

### Tpl-rand-k

Accuracy: 63.88. Result ID: 69933f7c3a6c31484914af9e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Implementation](<https://github.com/LLM-Based-Augmentations>)

Recorded evaluation configuration: {"Evaluation protocol":"translate-train","Supervision type":"hard prompt labels","Data selection strategy":"rand-k"}. This record does not assert independent reproduction.

### T-top-k

Accuracy: 63.82. Result ID: 69933f7c3a6c31484914af9a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Implementation](<https://github.com/LLM-Based-Augmentations>)

Recorded evaluation configuration: {"Evaluation protocol":"translate-train","Supervision type":"teacher soft pseudolabels","Data selection strategy":"top-k"}. This record does not assert independent reproduction.

### T-top-k

Accuracy: 63.74. Result ID: 69933f7c3a6c31484914af40.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Implementation](<https://github.com/LLM-Based-Augmentations>)

Recorded evaluation configuration: {"selection strategy":"top-k","Zero-shot":"true"}. This record does not assert independent reproduction.

### SRC+GEN

Accuracy: 63.52. Result ID: 69933f7c3a6c31484914af99.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

Recorded evaluation configuration: {"Evaluation protocol":"translate-train"}. This record does not assert independent reproduction.

### SRC+GEN

Accuracy: 63.34. Result ID: 69933f7c3a6c31484914af3f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

Recorded evaluation configuration: {"Zero-shot":"true"}. This record does not assert independent reproduction.

### Tpl-rand-k

Accuracy: 63.24. Result ID: 69933f7c3a6c31484914af44.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Implementation](<https://github.com/LLM-Based-Augmentations>)

Recorded evaluation configuration: {"selection strategy":"rand-k","Zero-shot":"true"}. This record does not assert independent reproduction.

### T-div-k

Accuracy: 62.69. Result ID: 69933f7c3a6c31484914af42.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Implementation](<https://github.com/LLM-Based-Augmentations>)

Recorded evaluation configuration: {"selection strategy":"div-k","Zero-shot":"true"}. This record does not assert independent reproduction.

### Tpl-top-k

Accuracy: 62.11. Result ID: 69933f7c3a6c31484914af43.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Implementation](<https://github.com/LLM-Based-Augmentations>)

Recorded evaluation configuration: {"selection strategy":"top-k","Zero-shot":"true"}. This record does not assert independent reproduction.

### T-rand-k

Accuracy: 62.11. Result ID: 69933f7c3a6c31484914af9b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Implementation](<https://github.com/LLM-Based-Augmentations>)

Recorded evaluation configuration: {"Evaluation protocol":"translate-train","Supervision type":"teacher soft pseudolabels","Data selection strategy":"rand-k"}. This record does not assert independent reproduction.

### T-rand-k

Accuracy: 61.67. Result ID: 69933f7c3a6c31484914af41.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

[Implementation](<https://github.com/LLM-Based-Augmentations>)

Recorded evaluation configuration: {"selection strategy":"rand-k","Zero-shot":"true"}. This record does not assert independent reproduction.

### SRC

Accuracy: 59.16. Result ID: 69933f7c3a6c31484914af98.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

Recorded evaluation configuration: {"Evaluation protocol":"translate-train"}. This record does not assert independent reproduction.

### SRC

Accuracy: 57.63. Result ID: 69933f7c3a6c31484914af3e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

Recorded evaluation configuration: {"Zero-shot":"true"}. This record does not assert independent reproduction.

### GEN

Accuracy: 56.71. Result ID: 69933f7c3a6c31484914af97.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

Recorded evaluation configuration: {"Evaluation protocol":"translate-train"}. This record does not assert independent reproduction.

### GEN

Accuracy: 56.02. Result ID: 69933f7c3a6c31484914af3d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74281e209131cea6014e.md>)

Recorded evaluation configuration: {"Zero-shot":"true"}. This record does not assert independent reproduction.

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