Leveraging Large Language Models for Building Interpretable Rule-Based Data-to-Text Systems
We introduce a simple approach that uses a large language model (LLM) to automatically implement a fully interpretable rule-based data-to-text system in pure Python. Experimental evaluation on the WebNLG dataset showed that such a constructed system produces text of better quality (according to the...
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Artículo |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2502.20609 http://arxiv.org/abs/2502.20609 |
| Etiquetas: |
Sin Etiquetas, Sea el primero en etiquetar este registro!
|