Learning to Plan and Generate Text with Citations

arXiv:2404.03381v1 Announce Type: new
Abstract: 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 …

abstract arxiv attribution capabilities citations cs.cl demand deployment evidence explore generate information llms paper queries responses systems text type