Libo Qin, Qiguang Chen, Yuhang Zhou, Zhi Chen, Yinghui Li, Lizi Liao, Min Li, Wanxiang Che et al.
A comprehensive survey paper that summarizes approaches, recent advances, and challenges of multilingual large language models (MLLMs).
Although LLMs with multilingual capabilities have gained attention, there has been a lack of a unified survey that integrates existing research. Therefore, a systematic understanding of the overall status and development direction of MLLMs is needed.
This paper organizes MLLM research into a unified taxonomy and analyzes recent trends from the perspective of multilingual alignment. It also identifies key frontiers and challenges, and collects open-source resources such as relevant papers, data corpora, and leaderboards.
As the first comprehensive survey in the MLLM field, it provides a unified taxonomy and abundant resources, enhancing accessibility for researchers. It is expected to serve as a benchmark for future multilingual LLM research.