visheratin
commited on
Commit
·
ace3b37
1
Parent(s):
d706dc3
Update README.md
Browse files
README.md
CHANGED
@@ -279,4 +279,8 @@ To get the image, use the following format:
|
|
279 |
|
280 |
```
|
281 |
https://nllb-data.com/{id}.jpg
|
282 |
-
```
|
|
|
|
|
|
|
|
|
|
279 |
|
280 |
```
|
281 |
https://nllb-data.com/{id}.jpg
|
282 |
+
```
|
283 |
+
|
284 |
+
## Paper
|
285 |
+
|
286 |
+
The dataset was used to train the models in the paper: "[NLLB-CLIP - train performant multilingual image retrieval model on a budget](https://arxiv.org/abs/2309.01859)".
|