Acegpt, Localizing Large Language Models In Arabic

Huang Huang, Yu Fei, Zhu Jianqing, Sun Xuening, Cheng Hao, Song Dingjie, Chen Zhihong, Alharthi Abdulmohsen, An Bang, He Juncai, Liu Ziche, Zhang Zhiyi, Chen Junying, Li Jianquan, Wang Benyou, Zhang Lian, Sun Ruoyu, Wan Xiang, Li Haizhou, Xu Jinchao. Arxiv 2023

[Paper] [Code]    
Agentic Fine Tuning GPT Has Code Model Architecture Pretraining Methods Reinforcement Learning Training Techniques

This paper is devoted to the development of a localized Large Language Model (LLM) specifically for Arabic, a language imbued with unique cultural characteristics inadequately addressed by current mainstream models. Significant concerns emerge when addressing cultural sensitivity and local values. To address this, the paper proposes a comprehensive solution that includes further pre-training with Arabic texts, Supervised Fine-Tuning (SFT) utilizing native Arabic instructions, and GPT-4 responses in Arabic, alongside Reinforcement Learning with AI Feedback (RLAIF) employing a reward model attuned to local culture and values. The goal is to cultivate culturally cognizant and value-aligned Arabic LLMs capable of accommodating the diverse, application-specific needs of Arabic-speaking communities. Comprehensive evaluations reveal that the resulting model, dubbed `AceGPT’, sets the state-of-the-art standard for open Arabic LLMs across various benchmarks. Codes, data, and models are in https://github.com/FreedomIntelligence/AceGPT.

Similar Work