CrossIn: An Efficient Instruction Tuning Approach for Cross-Lingual Knowledge Alignment
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_ | 1866913635406708736 |
|---|---|
| author | Lin, Geyu Wang, Bin Liu, Zhengyuan Chen, Nancy F. |
| author_facet | Lin, Geyu Wang, Bin Liu, Zhengyuan Chen, Nancy F. |
| contents | Multilingual proficiency presents a significant challenge for large language models (LLMs). English-centric models are usually suboptimal in other languages, particularly those that are linguistically distant from English. This performance discrepancy mainly stems from the imbalanced distribution of training data across languages during pre-training and instruction tuning stages. To address this problem, we propose a novel approach called CrossIn, which utilizes a mixed composition of cross-lingual instruction tuning data. Our method leverages the compressed representation shared by various languages to efficiently enhance the model's task-solving capabilities and multilingual proficiency within a single process. In addition, we introduce a multi-task and multi-faceted benchmark to evaluate the effectiveness of CrossIn. Experimental results demonstrate that our method substantially improves performance across tasks and languages, and we provide extensive insights into the impact of cross-lingual data volume and the integration of translation data on enhancing multilingual consistency and accuracy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_11932 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | CrossIn: An Efficient Instruction Tuning Approach for Cross-Lingual Knowledge Alignment Lin, Geyu Wang, Bin Liu, Zhengyuan Chen, Nancy F. Computation and Language Artificial Intelligence Multilingual proficiency presents a significant challenge for large language models (LLMs). English-centric models are usually suboptimal in other languages, particularly those that are linguistically distant from English. This performance discrepancy mainly stems from the imbalanced distribution of training data across languages during pre-training and instruction tuning stages. To address this problem, we propose a novel approach called CrossIn, which utilizes a mixed composition of cross-lingual instruction tuning data. Our method leverages the compressed representation shared by various languages to efficiently enhance the model's task-solving capabilities and multilingual proficiency within a single process. In addition, we introduce a multi-task and multi-faceted benchmark to evaluate the effectiveness of CrossIn. Experimental results demonstrate that our method substantially improves performance across tasks and languages, and we provide extensive insights into the impact of cross-lingual data volume and the integration of translation data on enhancing multilingual consistency and accuracy. |
| title | CrossIn: An Efficient Instruction Tuning Approach for Cross-Lingual Knowledge Alignment |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2404.11932 |