SAGE: Sustainable Agent-Guided Expert-tuning for Culturally Attuned Translation in Low-Resource Southeast Asia

2026-03-20

AI Summary

This work addresses the challenges of low-resource machine translation for Southeast Asian languages, which are hindered by the scarcity of high-quality, culturally relevant data and the high energy costs of training large models. To overcome these limitations, the authors propose SAGE, a novel framework that pioneers an energy-efficient paradigm centered on “precise data” rather than “big data.” SAGE employs a reinforcement learning agent based on Group Relative Policy Optimization to autonomously select a small set of high-value, culturally aligned dialogue samples, which are then used to fine-tune open-source large language models via LoRA. Evaluated across seven low-resource Southeast Asian languages, SAGE achieves state-of-the-art performance while using 97.1% less data and consuming 95.2% less energy than full-data baselines, effectively integrating cultural alignment and environmental sustainability into the translation pipeline.

Technology Category

• Natural Language Processing: (Large) Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

The vision of an inclusive World Wide Web is impeded by a severe linguistic divide, particularly for communities in low-resource regions of Southeast Asia. While large language models (LLMs) offer a potential solution for translation, their deployment in data-poor contexts faces a dual challenge: the scarcity of high-quality, culturally relevant data and the prohibitive energy costs of training on massive, noisy web corpora. To resolve the tension between digital inclusion and environmental sustainability, we introduce Sustainable Agent-Guided Expert-tuning (SAGE). This framework pioneers an energy-aware paradigm that prioritizes the "right data" over "big data". Instead of carbon-intensive training on unfiltered datasets, SAGE employs a reinforcement learning (RL) agent, optimized via Group Relative Policy Optimization (GRPO), to autonomously curate a compact training set. The agent utilizes a semantic reward signal derived from a small, expert-constructed set of community dialogues to filter out noise and cultural misalignment. We then efficiently fine-tune open-source LLMs on this curated data using Low-Rank Adaptation (LoRA). We applied SAGE to translation tasks between English and seven low-resource languages (LRLs) in Southeast Asia. Our approach establishes new state-of-the-art performance on BLEU-4 and COMET-22 metrics, effectively capturing local linguistic nuances. Crucially, SAGE surpasses baselines trained on full datasets while reducing data usage by 97.1% and training energy consumption by 95.2%. By delivering high-performance models with a minimal environmental footprint, SAGE offers a scalable and responsible pathway to bridge the digital divide in the Global South.

Problem

• low-resource languages
• culturally attuned translation
• sustainable AI
• digital divide
• energy-efficient training

Innovation

• Sustainable AI
• Agent-Guided Data Curation
• Low-Resource Translation
• Reinforcement Learning for NLP
• Energy-Efficient Fine-Tuning