Related Experiment Video
Updated: May 29, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
A survey of multilingual large language models
Libo Qin1, Qiguang Chen2, Yuhang Zhou2
1School of Computer Science and Engineering, Central South University, Changsha 410083, China.
This paper surveys multilingual large language models (MLLMs), offering a unified taxonomy and identifying emerging research frontiers. It provides resources to accelerate breakthroughs in MLLM development and application.
Area of Science:
- Natural Language Processing
- Artificial Intelligence
- Machine Learning
Background:
- Multilingual large language models (MLLMs) demonstrate significant success in processing queries across diverse languages.
- A comprehensive survey of existing MLLM approaches and recent advancements is currently lacking.
Purpose of the Study:
- To present a unified and thorough review of the field of multilingual large language models.
- To highlight recent progress, identify emerging trends, and discuss challenges in MLLM research.
Main Methods:
- Conducting an extensive survey of multilingual alignment techniques in MLLMs.
- Developing a unified taxonomy to categorize and summarize current MLLM progress.
- Identifying and discussing key emerging frontiers and associated challenges in the field.
Main Results:
- The paper provides the first comprehensive review of multilingual alignment in MLLMs.
- A novel, unified framework is introduced for summarizing MLLM research.
- Key emerging research frontiers and their challenges are identified.
Conclusions:
- This work offers a pioneering review of multilingual large language models, consolidating existing knowledge.
- The provided taxonomy and identified frontiers aim to guide future research and development in MLLMs.
- Abundant open-source resources are collected to facilitate community access and accelerate breakthroughs.
Related Concept Videos
Language and Cognition
Improving Translational Accuracy
Types of Surveys
Multi-input and Multi-variable systems
In the absence...
Typical Model Studies
One-Compartment Open Model: Wagner-Nelson and Loo Riegelman Method for ka Estimation
On...

