Learning to Plan and Generate Text with Citations
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866929431376822272 |
|---|---|
| author | Fierro, Constanza Amplayo, Reinald Kim Huot, Fantine De Cao, Nicola Maynez, Joshua Narayan, Shashi Lapata, Mirella |
| author_facet | Fierro, Constanza Amplayo, Reinald Kim Huot, Fantine De Cao, Nicola Maynez, Joshua Narayan, Shashi Lapata, Mirella |
| contents | The increasing demand for the deployment of LLMs in information-seeking scenarios has spurred efforts in creating verifiable systems, which generate responses to queries along with supporting evidence. In this paper, we explore the attribution capabilities of plan-based models which have been recently shown to improve the faithfulness, grounding, and controllability of generated text. We conceptualize plans as a sequence of questions which serve as blueprints of the generated content and its organization. We propose two attribution models that utilize different variants of blueprints, an abstractive model where questions are generated from scratch, and an extractive model where questions are copied from the input. Experiments on long-form question-answering show that planning consistently improves attribution quality. Moreover, the citations generated by blueprint models are more accurate compared to those obtained from LLM-based pipelines lacking a planning component. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_03381 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Learning to Plan and Generate Text with Citations Fierro, Constanza Amplayo, Reinald Kim Huot, Fantine De Cao, Nicola Maynez, Joshua Narayan, Shashi Lapata, Mirella Computation and Language The increasing demand for the deployment of LLMs in information-seeking scenarios has spurred efforts in creating verifiable systems, which generate responses to queries along with supporting evidence. In this paper, we explore the attribution capabilities of plan-based models which have been recently shown to improve the faithfulness, grounding, and controllability of generated text. We conceptualize plans as a sequence of questions which serve as blueprints of the generated content and its organization. We propose two attribution models that utilize different variants of blueprints, an abstractive model where questions are generated from scratch, and an extractive model where questions are copied from the input. Experiments on long-form question-answering show that planning consistently improves attribution quality. Moreover, the citations generated by blueprint models are more accurate compared to those obtained from LLM-based pipelines lacking a planning component. |
| title | Learning to Plan and Generate Text with Citations |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2404.03381 |