Does Structure Matter? Leveraging Data-to-Text Generation for Answering Complex Information Needs - Sorbonne Université
Conference Papers Year : 2022

Does Structure Matter? Leveraging Data-to-Text Generation for Answering Complex Information Needs

Abstract

In this work, our aim is to provide a structured answer in natural language to a complex information need. Particularly, we envision using generative models from the perspective of data-to-text generation. We propose the use of a content selection and planning pipeline which aims at structuring the answer by generating intermediate plans. The experimental evaluation is performed using the TREC Complex Answer Retrieval (CAR) dataset. We evaluate both the generated answer and its corresponding structure and show the effectiveness of planning-based models in comparison to a text-to-text model.
Fichier principal
Vignette du fichier
ECIR2022_ComplexAnswerGeneration (3).pdf (287.78 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03563311 , version 1 (09-02-2022)

Licence

Identifiers

Cite

Hanane Djeddal, Thomas Gérald, Laure Soulier, Karen Pinel-Sauvagnat, Lynda Tamine. Does Structure Matter? Leveraging Data-to-Text Generation for Answering Complex Information Needs. 44th European Conference on Information Retrieval (ECIR 2022), Apr 2022, Stavanger, Norway. ⟨10.48550/arXiv.2112.04344⟩. ⟨hal-03563311⟩
106 View
83 Download

Altmetric

Share

More