This leaderboard evaluates models and datasets focused on the Tunisian dialect of Arabic.It highlights performance on key resources such as TSAC (fbougares/tsac) and the Tunisian Dialect Corpus (arbml/Tunisian_Dialect_Corpus).

How it works

We evaluate models on:

• TSAC (fbougares/tsac): Sentiment analysis in Tunisian dialect.
• Tunisian Dialect Corpus (arbml/Tunisian_Dialect_Corpus): Coverage and language understanding.

Reproducibility

To reproduce our results, use the following commands (replace with your model):

from transformers import AutoConfig, AutoModel, AutoTokenizer
config = AutoConfig.from_pretrained("your model name", revision=revision)
model = AutoModel.from_pretrained("your model name", revision=revision)
tokenizer = AutoTokenizer.from_pretrained("your model name", revision=revision)

Some good practices before submitting a model

1) Make sure your model is trained or evaluated on Tunisian dialect data (e.g., TSAC, Tunisian Dialect Corpus).

2) Make sure you can load your model and tokenizer using AutoClasses:

from transformers import AutoConfig, AutoModel, AutoTokenizer
config = AutoConfig.from_pretrained("your model name", revision=revision)
model = AutoModel.from_pretrained("your model name", revision=revision)
tokenizer = AutoTokenizer.from_pretrained("your model name", revision=revision)

If this step fails, follow the error messages to debug your model before submitting it. It's likely your model has been improperly uploaded.

Note: make sure your model is public! Note: if your model needs use_remote_code=True, we do not support this option yet but we are working on adding it, stay posted!

3) Convert your model weights to safetensors

It's a new format for storing weights which is safer and faster to load and use. It will also allow us to add the number of parameters of your model to the Extended Viewer!

4) Make sure your model has an open license!

This is a leaderboard for Open LLMs, and we'd love for as many people as possible to know they can use your model 🤗