ChartDialogs: Plotting from Natural Language Instructions

Yutong Shao, Ndapa Nakashole


Abstract
This paper presents the problem of conversational plotting agents that carry out plotting actions from natural language instructions. To facilitate the development of such agents, we introduce ChartDialogs, a new multi-turn dialog dataset, covering a popular plotting library, matplotlib. The dataset contains over 15,000 dialog turns from 3,200 dialogs covering the majority of matplotlib plot types. Extensive experiments show the best-performing method achieving 61% plotting accuracy, demonstrating that the dataset presents a non-trivial challenge for future research on this task.
Anthology ID:
2020.acl-main.328
Volume:
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2020
Address:
Online
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3559–3574
Language:
URL:
https://aclanthology.org/2020.acl-main.328
DOI:
10.18653/v1/2020.acl-main.328
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/2020.acl-main.328.pdf
Video:
 http://slideslive.com/38928845