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

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.

Figures (6)

• Figure 1: Architectural overview of the SAGE framework.
• Figure 2: The SAGE training and alignment pipeline. Stage 1 employs a GRPO-optimized RL agent to curate a subset $\mathcal{D}_{\text{cur}}$ from a noisy pool $\mathcal{D}_{\text{noisy}}$, guided by semantic proximity to a small expert reference $\mathcal{D}_{\text{exp}}$. Stage 2 utilizes LoRA to efficiently fine-tune the LLM on $\mathcal{D}_{\text{cur}}$, minimizing computational overhead while maximizing cultural alignment.
• Figure 3: Sensitivity analysis of performance relative to expert data size ($|\mathcal{D}_{\text{expert}}|$). The dashed black line represents the average BLEU-4 score across all seven languages.
• Figure 4: Efficiency evaluation of the SAGE framework.(a) Environmental Efficiency: SAGE reduces carbon emissions by over 95% compared to baseline fine-tuning by leveraging high-quality, culturally attuned data subsets. The hatched area represents the carbon savings. (b) Inference Throughput: Comparison of token generation speed.
• Figure 5: Environmental Impact: SAGE achieves comparable or superior performance while reducing training data usage by 97% and carbon footprint by over 95% compared to standard fine-tuning.