Ancient Korean Archive Translation: Comparison Analysis on Statistical phrase alignment, LLM in-context learning, and inter-methodological approach
This addresses the challenge of translating ancient archives for researchers, but it is incremental as it builds on existing methods.
This study tackled the problem of translating ancient Korean texts with sparse corpora by comparing three methods, and found that their proposed inter-methodological approach achieved a BLEU score of 36.71, outperforming SOLAR-10.7B in-context learning and the best existing Seq2Seq model.
This study aims to compare three methods for translating ancient texts with sparse corpora: (1) the traditional statistical translation method of phrase alignment, (2) in-context LLM learning, and (3) proposed inter methodological approach - statistical machine translation method using sentence piece tokens derived from unified set of source-target corpus. The performance of the proposed approach in this study is 36.71 in BLEU score, surpassing the scores of SOLAR-10.7B context learning and the best existing Seq2Seq model. Further analysis and discussion are presented.