Peacock presents a comprehensive suite of Arabic MLLMs, Peacock models outperform multilingual baseline mBlip, Pretraining data is curated from publicly available English datasets translated into Arabic, Models trained in two stages - pretraining and instruction finetuning, Peacock models excel in VQA tasks on various datasets, Henna benchmark evaluates model capabilities related to Arabic culture.
Sang ngôn ngữ khác
từ nội dung nguồn
arxiv.org
Thông tin chi tiết chính được chắt lọc từ
by Fakhraddin A... lúc arxiv.org 03-05-2024
https://arxiv.org/pdf/2403.01031.pdfYêu cầu sâu hơn