CraftAssist Instruction Parsing: Semantic Parsing for a Voxel-World Assistant

Kavya Srinet, Yacine Jernite, Jonathan Gray, Arthur Szlam


Abstract
We propose a semantic parsing dataset focused on instruction-driven communication with an agent in the game Minecraft. The dataset consists of 7K human utterances and their corresponding parses. Given proper world state, the parses can be interpreted and executed in game. We report the performance of baseline models, and analyze their successes and failures.
Anthology ID:
2020.acl-main.427
Volume:
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2020
Address:
Online
Editors:
Dan Jurafsky, Joyce Chai, Natalie Schluter, Joel Tetreault
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4693–4714
Language:
URL:
https://aclanthology.org/2020.acl-main.427
DOI:
10.18653/v1/2020.acl-main.427
Bibkey:
Cite (ACL):
Kavya Srinet, Yacine Jernite, Jonathan Gray, and Arthur Szlam. 2020. CraftAssist Instruction Parsing: Semantic Parsing for a Voxel-World Assistant. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4693–4714, Online. Association for Computational Linguistics.
Cite (Informal):
CraftAssist Instruction Parsing: Semantic Parsing for a Voxel-World Assistant (Srinet et al., ACL 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.acl-main.427.pdf
Dataset:
 2020.acl-main.427.Dataset.zip
Video:
 http://slideslive.com/38929264