# Many-to-one Sign Language Translation on SP-10 (test)

Source: [Many-to-one Sign Language Translation on SP-10 (test)](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test>)

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

Data updated: 2026-02-26T06:09:12.477Z

Evaluation of Many-to-one Sign Language Translation performance on the test split of the SP-10 dataset, utilizing detailed BLEU and ROUGE scores across various language pairs.

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 | Many-to-one Sign Language Translation |
| Dataset (including any stated split) | SP-10 (test) |
| Metric | csl BLEU |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [csl BLEU](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=csl%20BLEU>): Higher is better; unit: SCALAR.

- [ukl BLEU](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=ukl%20BLEU>): Higher is better; unit: SCALAR.

- [rsl BLEU](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=rsl%20BLEU>): Higher is better; unit: SCALAR.

- [bqn BLEU](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=bqn%20BLEU>): Higher is better; unit: SCALAR.

- [icl BLEU](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=icl%20BLEU>): Higher is better; unit: SCALAR.

- [gsg BLEU](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=gsg%20BLEU>): Higher is better; unit: SCALAR.

- [ise BLEU](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=ise%20BLEU>): Higher is better; unit: SCALAR.

- [swl BLEU](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=swl%20BLEU>): Higher is better; unit: SCALAR.

- [lls BLEU](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=lls%20BLEU>): Higher is better; unit: SCALAR.

- [bfi BLEU](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=bfi%20BLEU>): Higher is better; unit: SCALAR.

- [Mean BLEU](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=Mean%20BLEU>): Higher is better; unit: SCALAR.

- [csl ROUGE](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=csl%20ROUGE>): Higher is better; unit: SCALAR.

- [ukl ROUGE](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=ukl%20ROUGE>): Higher is better; unit: SCALAR.

- [rsl ROUGE](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=rsl%20ROUGE>): Higher is better; unit: SCALAR.

- [bqn ROUGE](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=bqn%20ROUGE>): Higher is better; unit: SCALAR.

- [icl ROUGE](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=icl%20ROUGE>): Higher is better; unit: SCALAR.

- [gsg ROUGE](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=gsg%20ROUGE>): Higher is better; unit: SCALAR.

- [ise ROUGE](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=ise%20ROUGE>): Higher is better; unit: SCALAR.

- [swl ROUGE](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=swl%20ROUGE>): Higher is better; unit: SCALAR.

- [lls ROUGE](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=lls%20ROUGE>): Higher is better; unit: SCALAR.

- [bfi ROUGE](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=bfi%20ROUGE>): Higher is better; unit: SCALAR.

- [Mean ROUGE](<https://www.sota2.com/research/sota/many-to-one-sign-language-translation-on-sp-10-test.md?metric=Mean%20ROUGE>): Higher is better; unit: SCALAR.

## Recorded results

| Method | csl BLEU | Result date | Report paper |
| --- | --- | --- | --- |
| Individual (10) | 6.24 | Not recorded | 68fa78e81e209131cea7b21d |
| Sign2(LID+Text) | 5.92 | Not recorded | 68fa78e81e209131cea7b21d |
| MLST | 5.19 | Not recorded | 68fa78e81e209131cea7b21d |
| Universal (1) | 2.72 | Not recorded | 68fa78e81e209131cea7b21d |
| Google multi | 2.28 | Not recorded | 68fa78e81e209131cea7b21d |

### Individual (10)

csl BLEU: 6.24. Result ID: 6995acb63badfaffe6ab6e95.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e81e209131cea7b21d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703d1e209131cea49db1.md>)

[Implementation](<https://github.com/tensorflow/tensor2tensor>)

Recorded evaluation configuration: {"Mode":"Individual","Number of Models":10}. This record does not assert independent reproduction.

### Sign2(LID+Text)

csl BLEU: 5.92. Result ID: 6995acb63badfaffe6ab6e99.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e81e209131cea7b21d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78e81e209131cea7b21d.md>)

[Implementation](<https://github.com/Claire874/Gloss-free-MLSLT>)

Recorded evaluation configuration: {"Mode":"Multilingual"}. This record does not assert independent reproduction.

### MLST

csl BLEU: 5.19. Result ID: 6995acb63badfaffe6ab6e98.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e81e209131cea7b21d.md>)

Recorded evaluation configuration: {"Mode":"Multilingual"}. This record does not assert independent reproduction.

### Universal (1)

csl BLEU: 2.72. Result ID: 6995acb63badfaffe6ab6e96.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e81e209131cea7b21d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703d1e209131cea49db1.md>)

[Implementation](<https://github.com/tensorflow/tensor2tensor>)

Recorded evaluation configuration: {"Mode":"Universal","Number of Models":1}. This record does not assert independent reproduction.

### Google multi

csl BLEU: 2.28. Result ID: 6995acb63badfaffe6ab6e97.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78e81e209131cea7b21d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa639f1e209131ce9ffc6d.md>)

Recorded evaluation configuration: {"Mode":"Multilingual"}. This record does not assert independent reproduction.

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