Code-Based English Models Surprising Performance on Chinese QA Pair Extraction Task
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866929269748269056 |
|---|---|
| author | Zheng, Linghan Liu, Hui Lin, Xiaojun Dong, Jiayuan Sheng, Yue Shi, Gang Liu, Zhiwei Chen, Hongwei |
| author_facet | Zheng, Linghan Liu, Hui Lin, Xiaojun Dong, Jiayuan Sheng, Yue Shi, Gang Liu, Zhiwei Chen, Hongwei |
| contents | In previous studies, code-based models have consistently outperformed text-based models in reasoning-intensive scenarios. When generating our knowledge base for Retrieval-Augmented Generation (RAG), we observed that code-based models also perform exceptionally well in Chinese QA Pair Extraction task. Further, our experiments and the metrics we designed discovered that code-based models containing a certain amount of Chinese data achieve even better performance. Additionally, the capabilities of code-based English models in specified Chinese tasks offer a distinct perspective for discussion on the philosophical "Chinese Room" thought experiment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_10286 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Code-Based English Models Surprising Performance on Chinese QA Pair Extraction Task Zheng, Linghan Liu, Hui Lin, Xiaojun Dong, Jiayuan Sheng, Yue Shi, Gang Liu, Zhiwei Chen, Hongwei Computation and Language Artificial Intelligence In previous studies, code-based models have consistently outperformed text-based models in reasoning-intensive scenarios. When generating our knowledge base for Retrieval-Augmented Generation (RAG), we observed that code-based models also perform exceptionally well in Chinese QA Pair Extraction task. Further, our experiments and the metrics we designed discovered that code-based models containing a certain amount of Chinese data achieve even better performance. Additionally, the capabilities of code-based English models in specified Chinese tasks offer a distinct perspective for discussion on the philosophical "Chinese Room" thought experiment. |
| title | Code-Based English Models Surprising Performance on Chinese QA Pair Extraction Task |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2401.10286 |