Benchmarking LLM-based Machine Translation on Cultural Awareness

arXiv:2305.14328v2 Announce Type: replace
Abstract: Translating cultural-specific content is crucial for effective cross-cultural communication. However, many MT systems still struggle to translate sentences containing cultural-specific entities accurately and understandably. Recent advancements in in-context learning utilize lightweight prompts to guide large language models (LLMs) in machine translation tasks. Nevertheless, the effectiveness of this approach in enhancing machine translation with cultural awareness remains uncertain. To address this gap, we introduce a new data curation pipeline to construct a culturally relevant parallel corpus, …

abstract arxiv benchmarking communication context cs.cl guide however in-context learning language language models large language large language models llm llms machine machine translation prompts struggle systems tasks translate translation type