Update README.md
Browse files
README.md
CHANGED
|
@@ -18,11 +18,14 @@ multilinguality:
|
|
| 18 |
license: cc-by-sa-4.0
|
| 19 |
inference: false
|
| 20 |
---
|
|
|
|
| 21 |
|
| 22 |
This learned metric is for evaluating models trained on the TaTA dataset. It was trained as per instructions in [TaTA: A Multilingual Table-to-Text Dataset for African Languages](https://aclanthology.org/2023.findings-emnlp.118/) (StATA-QE variant).
|
| 23 |
|
| 24 |
The original file can be found [here](https://github.com/google-research/url-nlp/tree/main/tata).
|
| 25 |
|
|
|
|
|
|
|
| 26 |
It achieves an RMSE loss of 0.36 on the dev split, and a Pearson correlation of 0.59 with human evaluations on the test split ("attributable" column) of [this dataset](https://huggingface.co/datasets/adenhaus/stata).
|
| 27 |
|
| 28 |
# Example use
|
|
|
|
| 18 |
license: cc-by-sa-4.0
|
| 19 |
inference: false
|
| 20 |
---
|
| 21 |
+
# Background
|
| 22 |
|
| 23 |
This learned metric is for evaluating models trained on the TaTA dataset. It was trained as per instructions in [TaTA: A Multilingual Table-to-Text Dataset for African Languages](https://aclanthology.org/2023.findings-emnlp.118/) (StATA-QE variant).
|
| 24 |
|
| 25 |
The original file can be found [here](https://github.com/google-research/url-nlp/tree/main/tata).
|
| 26 |
|
| 27 |
+
# Performance
|
| 28 |
+
|
| 29 |
It achieves an RMSE loss of 0.36 on the dev split, and a Pearson correlation of 0.59 with human evaluations on the test split ("attributable" column) of [this dataset](https://huggingface.co/datasets/adenhaus/stata).
|
| 30 |
|
| 31 |
# Example use
|