作者位次 #2/6ACL 2026Findings· Findings of ACL
N eo A ra BERT : A Modern Foundation Model for A rabic Embeddings with Diacritics-Aware Tokenization and POS -Targeted Masking
Chadi Abou Chakra · Hadi Khaled Hamoud · Osama Rakan Al Mraikhat · Qusai Abu Obaida · Mohamad Ballout · Fadi Zaraket
摘要
We present NeoAraBERT, a state-of-the-art open-source Arabic text-embedding model built on the NeoBERT architecture. We pre-train NeoAraBERT on diverse open-source and internal datasets covering modern standard, classical, and dialectal Arabic. We guided our design choices with Arabic tailored ab...