Tiny Aya：跨尺度与多语言深度的桥梁
计算与语言
2026-03-13 v1
摘要
Tiny Aya 重新定义了小型多语言语言模型的可能。该模型在 70 种语言上训练，并通过区域感知后训练进行精炼，实现了翻译质量、多语言理解和高质量目标语言生成的状态-of-the-art 效果，仅需 3.35B 参数。发布包括预训练基础模型、全球均衡的指令微调变体，以及三个区域专精模型，分别针对非洲、南亚、欧洲、亚太和西亚地区的语言。本报告详细阐述了训练策略、数据构成及全面评估框架，呈现一种以效率为中心、跨语言性能均衡且具实际部署价值的多语言 AI 扩展路径。
引用
@article{arxiv.2603.11510,
title = {Tiny Aya: Bridging Scale and Multilingual Depth},
author = {Alejandro R. Salamanca and Diana Abagyan and Daniel D'souza and Ammar Khairi and David Mora and Saurabh Dash and Viraat Aryabumi and Sara Rajaee and Mehrnaz Mofakhami and Ananya Sahu and Thomas Euyang and Brittawnya Prince and Madeline Smith and Hangyu Lin and Acyr Locatelli and Sara Hooker and Tom Kocmi and Aidan Gomez and Ivan Zhang and Phil Blunsom and Nick Frosst and Joelle Pineau and Beyza Ermis and Ahmet Üstün and Julia Kreutzer and Marzieh Fadaee},
journal= {arXiv preprint arXiv:2603.11510},
year = {2026}
}