A Hierarchical Model for Data-to-Text Generation - Sorbonne Université
Conference Papers Year : 2020

A Hierarchical Model for Data-to-Text Generation

Abstract

Transcribing structured data into natural language descriptions has emerged as a challenging task, referred to as “data-to-text”. These structures generally regroup multiple elements, as well as their attributes. Most attempts rely on translation encoder-decoder methods which linearize elements into a sequence. This however loses most of the structure contained in the data. In this work, we propose to overpass this limitation with a hierarchical model that encodes the data-structure at the element-level and the structure level. Evaluations on RotoWire show the effectiveness of our model w.r.t. qualitative and quantitative metrics.

Dates and versions

hal-02774325 , version 1 (04-06-2020)

Identifiers

Cite

Clément Rebuffel, Laure Soulier, Geoffrey Scoutheeten, Patrick Gallinari. A Hierarchical Model for Data-to-Text Generation. 42nd European Conference on IR Research, ECIR 2020, Apr 2020, Lisbon, Portugal. pp.65-80, ⟨10.1007/978-3-030-45439-5_5⟩. ⟨hal-02774325⟩
198 View
6 Download

Altmetric

Share

More