Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2407.13343 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866910533410619392 |
|---|---|
| author | Liao, You-Cheng Yu, Chen-Jui Lin, Chi-Yi Yun, He-Feng Wang, Yen-Hsiang Li, Hsiao-Min Fan, Yao-Chung |
| author_facet | Liao, You-Cheng Yu, Chen-Jui Lin, Chi-Yi Yun, He-Feng Wang, Yen-Hsiang Li, Hsiao-Min Fan, Yao-Chung |
| contents | Using large language models, this paper presents techniques to improve extremely low-resourced indigenous language translations. Our approaches are grounded in the use of (1) the presence of a datastore consisting of a limited number of parallel translation examples, (2) the inherent capabilities of LLMs like GPT-3.5, and (3) a word-level translation dictionary. We harness the potential of LLMs and in-context learning techniques in such a setting for using LLMs as universal translators for extremely low-resourced languages. Our methodology hinges on utilizing LLMs as language compilers for selected language pairs, hypothesizing that they could internalize syntactic structures to facilitate accurate translation. We introduce three techniques: KNNPrompting with Retrieved Prompting Context, Chain-of-Thought Prompting and Learningfrom-Mistakes Prompting, with the last method addressing past errors. The evaluation results suggest that, even with limited corpora, LLMs can effectively translate extremely low-resource languages when paired with proper prompting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_13343 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Learning-From-Mistakes Prompting for Indigenous Language Translation Liao, You-Cheng Yu, Chen-Jui Lin, Chi-Yi Yun, He-Feng Wang, Yen-Hsiang Li, Hsiao-Min Fan, Yao-Chung Computation and Language Using large language models, this paper presents techniques to improve extremely low-resourced indigenous language translations. Our approaches are grounded in the use of (1) the presence of a datastore consisting of a limited number of parallel translation examples, (2) the inherent capabilities of LLMs like GPT-3.5, and (3) a word-level translation dictionary. We harness the potential of LLMs and in-context learning techniques in such a setting for using LLMs as universal translators for extremely low-resourced languages. Our methodology hinges on utilizing LLMs as language compilers for selected language pairs, hypothesizing that they could internalize syntactic structures to facilitate accurate translation. We introduce three techniques: KNNPrompting with Retrieved Prompting Context, Chain-of-Thought Prompting and Learningfrom-Mistakes Prompting, with the last method addressing past errors. The evaluation results suggest that, even with limited corpora, LLMs can effectively translate extremely low-resource languages when paired with proper prompting. |
| title | Learning-From-Mistakes Prompting for Indigenous Language Translation |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2407.13343 |