Update README.md
#1
by prince4332 - opened
README.md
CHANGED
|
@@ -86,10 +86,3 @@ MargEmbed-Akan is trained by **cross-lingual listwise score distillation** from
|
|
| 86 |
- Optimized for Twi questions over English passages; for English-only search, the base model or proprietary models are equal or better.
|
| 87 |
- The AfriQA evaluation corpus has 5,357 passages; scores on a full Wikipedia-scale index will be lower.
|
| 88 |
|
| 89 |
-
## License
|
| 90 |
-
|
| 91 |
-
The training data includes NLLB-200 translations (CC BY-NC 4.0), so this model is released for **research and non-commercial use**.
|
| 92 |
-
|
| 93 |
-
## Citation
|
| 94 |
-
|
| 95 |
-
Paper forthcoming. Please cite this model page until then.
|
|
|
|
| 86 |
- Optimized for Twi questions over English passages; for English-only search, the base model or proprietary models are equal or better.
|
| 87 |
- The AfriQA evaluation corpus has 5,357 passages; scores on a full Wikipedia-scale index will be lower.
|
| 88 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|