MMTE: Corpus and Metrics for Evaluating Machine Translation Quality of Metaphorical Language
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916472395137024 |
|---|---|
| author | Wang, Shun Zhang, Ge Wu, Han Loakman, Tyler Huang, Wenhao Lin, Chenghua |
| author_facet | Wang, Shun Zhang, Ge Wu, Han Loakman, Tyler Huang, Wenhao Lin, Chenghua |
| contents | Machine Translation (MT) has developed rapidly since the release of Large Language Models and current MT evaluation is performed through comparison with reference human translations or by predicting quality scores from human-labeled data. However, these mainstream evaluation methods mainly focus on fluency and factual reliability, whilst paying little attention to figurative quality. In this paper, we investigate the figurative quality of MT and propose a set of human evaluation metrics focused on the translation of figurative language. We additionally present a multilingual parallel metaphor corpus generated by post-editing. Our evaluation protocol is designed to estimate four aspects of MT: Metaphorical Equivalence, Emotion, Authenticity, and Quality. In doing so, we observe that translations of figurative expressions display different traits from literal ones. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_13698 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MMTE: Corpus and Metrics for Evaluating Machine Translation Quality of Metaphorical Language Wang, Shun Zhang, Ge Wu, Han Loakman, Tyler Huang, Wenhao Lin, Chenghua Computation and Language Machine Translation (MT) has developed rapidly since the release of Large Language Models and current MT evaluation is performed through comparison with reference human translations or by predicting quality scores from human-labeled data. However, these mainstream evaluation methods mainly focus on fluency and factual reliability, whilst paying little attention to figurative quality. In this paper, we investigate the figurative quality of MT and propose a set of human evaluation metrics focused on the translation of figurative language. We additionally present a multilingual parallel metaphor corpus generated by post-editing. Our evaluation protocol is designed to estimate four aspects of MT: Metaphorical Equivalence, Emotion, Authenticity, and Quality. In doing so, we observe that translations of figurative expressions display different traits from literal ones. |
| title | MMTE: Corpus and Metrics for Evaluating Machine Translation Quality of Metaphorical Language |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2406.13698 |