Update README.md
Browse files
README.md
CHANGED
|
@@ -29,22 +29,24 @@ base_model:
|
|
| 29 |
This is an instance of [xlm-roberta-large](https://huggingface.co/FacebookAI/xlm-roberta-large) that was fine-tuned on binary toxicity classification task based on our updated (2025) dataset [textdetox/multilingual_toxicity_dataset](https://huggingface.co/datasets/textdetox/multilingual_toxicity_dataset).
|
| 30 |
|
| 31 |
Now, the models covers 15 languages from various language families:
|
| 32 |
-
* English (en); F1: 0.9225
|
| 33 |
-
* Russian (ru); F1: 0.9525
|
| 34 |
-
* Ukrainian (uk); F1: 0.96
|
| 35 |
-
* German (de); F1: 0.7325
|
| 36 |
-
* Spanish (es); F1: 0.7125
|
| 37 |
-
* Arabic (ar); F1: 0.6625
|
| 38 |
-
* Amharic (am); F1: 0.5575
|
| 39 |
-
* Hindi (hi); F1: 0.9725
|
| 40 |
-
* Chinese (zh); F1: 0.9175
|
| 41 |
-
* Italian (it); F1: 0.5864
|
| 42 |
-
* French (fr); F1: 0.9235
|
| 43 |
-
* Hinglish (hin); F1: 0.61
|
| 44 |
-
* Hebrew (he); F1: 0.8775
|
| 45 |
-
* Japanese (ja); F1: 0.8773
|
| 46 |
-
* Tatar (tt); F1: 0.5744
|
| 47 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
|
| 49 |
## Citation
|
| 50 |
The model is prepared for [TextDetox 2025 Shared Task](https://pan.webis.de/clef25/pan25-web/text-detoxification.html) evaluation.
|
|
|
|
| 29 |
This is an instance of [xlm-roberta-large](https://huggingface.co/FacebookAI/xlm-roberta-large) that was fine-tuned on binary toxicity classification task based on our updated (2025) dataset [textdetox/multilingual_toxicity_dataset](https://huggingface.co/datasets/textdetox/multilingual_toxicity_dataset).
|
| 30 |
|
| 31 |
Now, the models covers 15 languages from various language families:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
|
| 33 |
+
| Language | Code | F1 Score |
|
| 34 |
+
|-----------|------|---------|
|
| 35 |
+
| English | en | 0.9225 |
|
| 36 |
+
| Russian | ru | 0.9525 |
|
| 37 |
+
| Ukrainian | uk | 0.96 |
|
| 38 |
+
| German | de | 0.7325 |
|
| 39 |
+
| Spanish | es | 0.7125 |
|
| 40 |
+
| Arabic | ar | 0.6625 |
|
| 41 |
+
| Amharic | am | 0.5575 |
|
| 42 |
+
| Hindi | hi | 0.9725 |
|
| 43 |
+
| Chinese | zh | 0.9175 |
|
| 44 |
+
| Italian | it | 0.5864 |
|
| 45 |
+
| French | fr | 0.9235 |
|
| 46 |
+
| Hinglish | hin | 0.61 |
|
| 47 |
+
| Hebrew | he | 0.8775 |
|
| 48 |
+
| Japanese | ja | 0.8773 |
|
| 49 |
+
| Tatar | tt | 0.5744 |
|
| 50 |
|
| 51 |
## Citation
|
| 52 |
The model is prepared for [TextDetox 2025 Shared Task](https://pan.webis.de/clef25/pan25-web/text-detoxification.html) evaluation.
|