Active Learning for Massively Parallel Translation of Constrained Text into Low Resource Languages
This work addresses translation challenges for low-resource languages, offering incremental improvements in active learning strategies for constrained texts.
The study tackled the problem of translating a closed text into a low-resource language by comparing portion-based and random sampling approaches for active learning, finding that random sampling improved performance by +11.0 BLEU in English and +4.9 BLEU in Eastern Pokomchi when training on 1,000 lines and testing on 30,000 lines.
We translate a closed text that is known in advance and available in many languages into a new and severely low resource language. Most human translation efforts adopt a portion-based approach to translate consecutive pages/chapters in order, which may not suit machine translation. We compare the portion-based approach that optimizes coherence of the text locally with the random sampling approach that increases coverage of the text globally. Our results show that the random sampling approach performs better. When training on a seed corpus of ~1,000 lines from the Bible and testing on the rest of the Bible (~30,000 lines), random sampling gives a performance gain of +11.0 BLEU using English as a simulated low resource language, and +4.9 BLEU using Eastern Pokomchi, a Mayan language. Furthermore, we compare three ways of updating machine translation models with increasing amount of human post-edited data through iterations. We find that adding newly post-edited data to training after vocabulary update without self-supervision performs the best. We propose an algorithm for human and machine to work together seamlessly to translate a closed text into a severely low resource language.