MBZUAI-Paris/Egyptian-DPO-Mixture

Domaine:

natural language processing

Type de record:

dataset

Créateur:

MBZ

Hôte:

This dataset supports Direct Preference Optimization (DPO) fine-tuning using off-policy alignment signals to enhance stylistic control, code-switching behavior, and instruction adherence and on-policy alignment signals to improve instruction-following and safety behavior.

MBZUAI-Paris/Egyptian-SFT-Mixture

This dataset contains the Supervised Fine-Tuning samples for Nile-Chat in both Arabic and Latin scri

MBZUAI-Paris/Darija-SFT-Mixture

Note the ODC-BY license, indicating that different licenses apply to subsets of the data. This means

MBZUAI-Paris/DarijaMMLU

DarijaMMLU is an evaluation benchmark designed to assess large language models' (LLM) performance in

MBZUAI-Paris/DarijaBench

Note the ODC-BY license, indicating that different licenses apply to subsets of the data. This means

MBZUAI-Paris/DarijaAlpacaEval

Dataset Summary

MBZUAI-Paris/DarijaHellaSwag

DarijaHellaSwag is a challenging multiple-choice benchmark designed to evaluate machine reading comp