Update README.md
Browse files
README.md
CHANGED
|
@@ -263,7 +263,7 @@ model-index:
|
|
| 263 |
|
| 264 |
**[Tucano](https://huggingface.co/TucanoBR)** is a series of decoder-transformers natively pretrained in Portuguese. All Tucano models were trained on **[GigaVerbo](https://huggingface.co/datasets/TucanoBR/GigaVerbo)**, a concatenation of deduplicated Portuguese text corpora amounting to 200 billion tokens.
|
| 265 |
|
| 266 |
-
Read our preprint [here](https://arxiv.org/abs/
|
| 267 |
|
| 268 |
## Details
|
| 269 |
|
|
@@ -370,7 +370,7 @@ Hence, even though our models are released with a permissive license, we urge us
|
|
| 370 |
|
| 371 |
## Evaluations
|
| 372 |
|
| 373 |
-
The table below compares our models against several Portuguese and multilingual language models on the evaluation harness used in our study. More information on it can be found [here](https://github.com/Nkluge-correa/Tucano/tree/main/evaluations/README.md). To learn more about our evaluation harness selection, [read our preprint](https://arxiv.org/abs/
|
| 374 |
|
| 375 |
| | Average | Calame-PT | Lambada-PT | ARC-PT | HellaSwag-PT |
|
| 376 |
|-----------------|---------|-----------|------------|--------|--------------|
|
|
@@ -397,11 +397,14 @@ The table below compares our models against several Portuguese and multilingual
|
|
| 397 |
## Cite as 🤗
|
| 398 |
|
| 399 |
```latex
|
| 400 |
-
@misc{
|
| 401 |
-
|
| 402 |
-
|
| 403 |
-
|
| 404 |
-
|
|
|
|
|
|
|
|
|
|
| 405 |
}
|
| 406 |
```
|
| 407 |
|
|
|
|
| 263 |
|
| 264 |
**[Tucano](https://huggingface.co/TucanoBR)** is a series of decoder-transformers natively pretrained in Portuguese. All Tucano models were trained on **[GigaVerbo](https://huggingface.co/datasets/TucanoBR/GigaVerbo)**, a concatenation of deduplicated Portuguese text corpora amounting to 200 billion tokens.
|
| 265 |
|
| 266 |
+
Read our preprint [here](https://arxiv.org/abs/2411.07854).
|
| 267 |
|
| 268 |
## Details
|
| 269 |
|
|
|
|
| 370 |
|
| 371 |
## Evaluations
|
| 372 |
|
| 373 |
+
The table below compares our models against several Portuguese and multilingual language models on the evaluation harness used in our study. More information on it can be found [here](https://github.com/Nkluge-correa/Tucano/tree/main/evaluations/README.md). To learn more about our evaluation harness selection, [read our preprint](https://arxiv.org/abs/2411.07854).
|
| 374 |
|
| 375 |
| | Average | Calame-PT | Lambada-PT | ARC-PT | HellaSwag-PT |
|
| 376 |
|-----------------|---------|-----------|------------|--------|--------------|
|
|
|
|
| 397 |
## Cite as 🤗
|
| 398 |
|
| 399 |
```latex
|
| 400 |
+
@misc{correa2024tucanoadvancingneuraltext,
|
| 401 |
+
title={{Tucano: Advancing Neural Text Generation for Portuguese}},
|
| 402 |
+
author={Corr{\^e}a, Nicholas Kluge and Sen, Aniket and Falk, Sophia and Fatimah, Shiza},
|
| 403 |
+
year={2024},
|
| 404 |
+
eprint={2411.07854},
|
| 405 |
+
archivePrefix={arXiv},
|
| 406 |
+
primaryClass={cs.CL},
|
| 407 |
+
url={https://arxiv.org/abs/2411.07854},
|
| 408 |
}
|
| 409 |
```
|
| 410 |
|