Behind Maya: Building a Multilingual Vision Language Model

Nahid Alam, Karthik Reddy Kanjula, Surya Guthikonda, Timothy Chung, Bala Krishna S Vegesna +14 more
5/13/2025
cs.CVcs.CL

Official code: nahidalam/maya (129 stars, Apache-2.0, last commit 2025-08-07).

Abstract

In recent times, we have seen a rapid development of large Vision-Language Models (VLMs). They have shown impressive results on academic benchmarks, primarily in widely spoken languages but lack performance on low-resource languages and varied cultural contexts. To address these limitations, we introduce Maya, an open-source Multilingual VLM. Our contributions are: 1) a multilingual image-text pretraining dataset in eight languages, based on the LLaVA pretraining dataset; and 2) a multilingual image-text model supporting these languages, enhancing cultural and linguistic comprehension in vision-language tasks. Code available at https://github.com/nahidalam/maya.

View on arXivView PDF

Code Implementations(1)

nahidalam/mayaOfficial100%
Apache-2.0

Maya: An Instruction Finetuned Multilingual Multimodal Model using Aya

12913CSS, HTMLNov 27, 20241 years agoApache-2.0

Similar Papers

Maya: An Instruction Finetuned Multilingual Multimodal Model

97% match1 repo1292024

TowerVision: Understanding and Improving Multilinguality in Vision-Language Models

86% match2025

Chitrarth: Bridging Vision and Language for a Billion People

86% match1 repo132025

Constructing Multilingual Visual-Text Datasets Revealing Visual Multilingual Ability of Vision Language Models

85% match2024

VLURes: Benchmarking VLM Visual and Linguistic Understanding in Low-Resource Languages

85% match2025

A Progressive Framework of Vision-language Knowledge Distillation and Alignment for Multilingual Scene

85% match2024

Cite this paper

@article{alam2025behind,
  title  = {Behind Maya: Building a Multilingual Vision Language Model},
  author = {Nahid Alam and Karthik Reddy Kanjula and Surya Guthikonda and Timothy Chung and Bala Krishna S Vegesna and Abhipsha Das and Anthony Susevski and Ryan Sze-Yin Chan and S M Iftekhar Uddin and Shayekh Bin Islam and Roshan Santhosh and Snegha A and Drishti Sharma and Chen Liu and Isha Chaturvedi and Genta Indra Winata and Ashvanth. S and Snehanshu Mukherjee and Alham Fikri Aji},
  year   = {2025},
  eprint = {2505.08910},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2505.08910v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2505.08910.svg)](https://codeofpaper.com/paper/2505.08910)

Discussion