Can LLMs Ask Good Questions?
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913899239964672 |
|---|---|
| author | Zhang, Yueheng Liu, Xiaoyuan Sun, Yiyou Alharbi, Atheer Alzahrani, Hend Shi, Tianneng Alomair, Basel Song, Dawn |
| author_facet | Zhang, Yueheng Liu, Xiaoyuan Sun, Yiyou Alharbi, Atheer Alzahrani, Hend Shi, Tianneng Alomair, Basel Song, Dawn |
| contents | We evaluate questions generated by large language models (LLMs) from context, comparing them to human-authored questions across six dimensions: question type, question length, context coverage, answerability, uncommonness, and required answer length. Our study spans two open-source and two proprietary state-of-the-art models. Results reveal that LLM-generated questions tend to demand longer descriptive answers and exhibit more evenly distributed context focus, in contrast to the positional bias often seen in QA tasks. These findings provide insights into the distinctive characteristics of LLM-generated questions and inform future work on question quality and downstream applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_03491 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Can LLMs Ask Good Questions? Zhang, Yueheng Liu, Xiaoyuan Sun, Yiyou Alharbi, Atheer Alzahrani, Hend Shi, Tianneng Alomair, Basel Song, Dawn Computation and Language Artificial Intelligence We evaluate questions generated by large language models (LLMs) from context, comparing them to human-authored questions across six dimensions: question type, question length, context coverage, answerability, uncommonness, and required answer length. Our study spans two open-source and two proprietary state-of-the-art models. Results reveal that LLM-generated questions tend to demand longer descriptive answers and exhibit more evenly distributed context focus, in contrast to the positional bias often seen in QA tasks. These findings provide insights into the distinctive characteristics of LLM-generated questions and inform future work on question quality and downstream applications. |
| title | Can LLMs Ask Good Questions? |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2501.03491 |