Preserving Cultural Identity with Context-Aware Translation Through Multi-Agent AI Systems

• Authors: Mahfuz Ahmed Anik, Abdur Rahman, Azmine Toushik Wasi, Md Manjurul Ahsan
• Accepted in NAACL 2025 Workshop on Language Models for Underserved Communities
• Read in arXiv

Abstract: Language is a cornerstone of cultural identity, yet globalization and the dominance of major languages have placed nearly 3,000 languages at risk of extinction. Existing AI-driven translation models prioritize efficiency but often fail to capture cultural nuances, idiomatic expressions, and historical significance, leading to translations that marginalize linguistic diversity. To address these challenges, we propose a multi-agent AI framework designed for culturally adaptive translation in underserved language communities. Our approach leverages specialized agents for translation, interpretation, content synthesis, and bias evaluation, ensuring that linguistic accuracy and cultural relevance are preserved. Using CrewAI and LangChain, our system enhances contextual fidelity while mitigating biases through external validation. Comparative analysis shows that our framework outperforms GPT-4o, producing contextually rich and culturally embedded translations—a critical advancement for Indigenous, regional, and low-resource languages. This research underscores the potential of multi-agent AI in fostering equitable, sustainable, and culturally sensitive NLP technologies, aligning with the AI Governance, Cultural NLP, and Sustainable NLP pillars of Language Models for Underserved Communities.

---

Instructions

1. Start LLM Server for LLM Inference

a. Install Ollama:

curl -fsSL https://ollama.com/install.sh | sh

b. Start the Ollama Server and Download the Llama3 Model:

ollama serve & ollama pull aya-expanse:8b

c. Install LiteLLM with Proxy Support:

pip install 'litellm[proxy]'

d. Start the LiteLLM Proxy Server with the Ollama Llama3 Model:

litellm --model ollama/aya-expanse:8b

2. Prepare Environment

Install required packages.

pip install colab-xterm duckduckgo-search
pip install crewai==0.28.8 crewai_tools==0.1.6 langchain_community==0.0.29

3. Run The Agent

Change the English text in line 81, in main.py. Then, run the script.

python main.py

---

Cite as:

@inproceedings{
anonymous2025preserving,
title={Preserving Cultural Identity with Context-Aware Translation Through Multi-Agent {AI} Systems},
author={Mahfuz Ahmed Anik and Abdur Rahman and Azmine Toushik Wasi and Md Manjurul Ahsan},
booktitle={NAACL 2025 Workshop on Language Models for Underserved Communities},
year={2025},
url={https://openreview.net/forum?id=RiCfefEHII}
}

or,

@misc{anik2025preservingculturalidentitycontextaware,
      title={Preserving Cultural Identity with Context-Aware Translation Through Multi-Agent AI Systems}, 
      author={Mahfuz Ahmed Anik and Abdur Rahman and Azmine Toushik Wasi and Md Manjurul Ahsan},
      year={2025},
      eprint={2503.04827},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2503.04827}, 
}