linagora/TunisianMMLU

Domaine:

natural language processing

Type de record:

dataset

Créateur:

Lin

Hôte:

TunisianMMLU is an evaluation benchmark designed to assess large language models' (LLM) performance in Tunisian Derja, a variety of Arabic. It consists of 22,027 multiple-choice questions, translated from selected subsets of the Massive Multitask Language Understanding (MMLU), ArabicMMLU and DarijaMMLU benchmarks to measure model performance on 44 subjects in Derja. Task Category: Multiple-choice question answering

linagora/Tunisian_Derja_Dataset

This is a collection of Tunisian dialect textual documents for Language Modeling. It was used for co

linagora/fineweb2_Tunisian_Arabic

This is the Tunisian Arabic Portion of The FineWeb2 Dataset. This dataset contains a rich collection

linagora-labs/ASR_train_kaldi_tunisian

# ASR_train_kaldi_tunisian ## Training Acoustic Models with Kaldi This script facilitates the trai

linagora/linto-dataset-audio-ar-tn

This is the first packaged version of the datasets used to train the Linto Tunisian dialect with cod

linagora/linto-dataset-text-ar-tn

This is a collection of Tunisian dialect textual documents for Language Modeling. It was used to tra

linagora/linto-dataset-audio-ar-tn-augmented

This is the augmented datasets used to train the Linto Tunisian dialect with code-switching STT lina

linagora/TunisianMMLU

Domaine:

Type de record:

Visit

Tasks

Languages