Gender-specific Machine Translation with Large Language Models

arXiv:2309.03175v2 Announce Type: replace
Abstract: While machine translation (MT) systems have seen significant improvements, it is still common for translations to reflect societal biases, such as gender bias. Decoder-only Large Language Models (LLMs) have demonstrated potential in MT, albeit with performance slightly lagging behind traditional encoder-decoder Neural Machine Translation (NMT) systems. However, LLMs offer a unique advantage: the ability to control the properties of the output through prompts. In this study, we leverage this flexibility to explore LLaMa's capability to …

abstract arxiv bias biases cs.cl decoder encoder encoder-decoder gender gender bias however improvements language language models large language large language models llms machine machine translation neural machine translation performance systems translation translations type