English-Sinhala-Tamil Machine Translation with Large Language Models

Principal Investigator: Nisansa de Silva

The proposed research is expected to support the completion of ongoing English-Sinhala-Tamil machine translation research through the development of human-annotated evaluation datasets and cloud computing resources; benchmarking and fine-tuning multilingual large language models using transfer learning and multi-domain data mixing techniques, with the aim of releasing an evaluation dataset and publishing the findings in a Q1 journal.

Objectives:

• Completion of ongoing English-Sinhala-Tamil machine translation research through the development of human-annotated evaluation datasets and cloud computing resources