English

Goal-Oriented Script Construction

Computation and Language 2021-09-01 v3

Abstract

The knowledge of scripts, common chains of events in stereotypical scenarios, is a valuable asset for task-oriented natural language understanding systems. We propose the Goal-Oriented Script Construction task, where a model produces a sequence of steps to accomplish a given goal. We pilot our task on the first multilingual script learning dataset supporting 18 languages collected from wikiHow, a website containing half a million how-to articles. For baselines, we consider both a generation-based approach using a language model and a retrieval-based approach by first retrieving the relevant steps from a large candidate pool and then ordering them. We show that our task is practical, feasible but challenging for state-of-the-art Transformer models, and that our methods can be readily deployed for various other datasets and domains with decent zero-shot performance.

Keywords

Cite

@article{arxiv.2107.13189,
  title  = {Goal-Oriented Script Construction},
  author = {Qing Lyu and Li Zhang and Chris Callison-Burch},
  journal= {arXiv preprint arXiv:2107.13189},
  year   = {2021}
}

Comments

INLG2021 (14th International Conference on Natural Language Generation)