SPRING Lab IITM's submission to Low Resource Indic Language Translation Shared Task
This work addresses translation challenges for low-resource Indic languages, but it is incremental as it builds on existing models and techniques.
The researchers tackled low-resource translation for four Indic languages by developing a pipeline with data augmentation and fine-tuning, achieving improved performance over baselines for Assamese, Mizo, and Manipuri, and introducing special tokens for Khasi.
We develop a robust translation model for four low-resource Indic languages: Khasi, Mizo, Manipuri, and Assamese. Our approach includes a comprehensive pipeline from data collection and preprocessing to training and evaluation, leveraging data from WMT task datasets, BPCC, PMIndia, and OpenLanguageData. To address the scarcity of bilingual data, we use back-translation techniques on monolingual datasets for Mizo and Khasi, significantly expanding our training corpus. We fine-tune the pre-trained NLLB 3.3B model for Assamese, Mizo, and Manipuri, achieving improved performance over the baseline. For Khasi, which is not supported by the NLLB model, we introduce special tokens and train the model on our Khasi corpus. Our training involves masked language modelling, followed by fine-tuning for English-to-Indic and Indic-to-English translations.