Improving Cross-Modal Alignment in Vision Language Navigation via Syntactic Information

Jialu Li, Hao Tan, Mohit Bansal


Abstract
Vision language navigation is the task that requires an agent to navigate through a 3D environment based on natural language instructions. One key challenge in this task is to ground instructions with the current visual information that the agent perceives. Most of the existing work employs soft attention over individual words to locate the instruction required for the next action. However, different words have different functions in a sentence (e.g., modifiers convey attributes, verbs convey actions). Syntax information like dependencies and phrase structures can aid the agent to locate important parts of the instruction. Hence, in this paper, we propose a navigation agent that utilizes syntax information derived from a dependency tree to enhance alignment between the instruction and the current visual scenes. Empirically, our agent outperforms the baseline model that does not use syntax information on the Room-to-Room dataset, especially in the unseen environment. Besides, our agent achieves the new state-of-the-art on Room-Across-Room dataset, which contains instructions in 3 languages (English, Hindi, and Telugu). We also show that our agent is better at aligning instructions with the current visual information via qualitative visualizations.
Anthology ID:
2021.naacl-main.82
Volume:
Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
Month:
June
Year:
2021
Address:
Online
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1041–1050
Language:
URL:
https://aclanthology.org/2021.naacl-main.82
DOI:
10.18653/v1/2021.naacl-main.82
Bibkey:
Cite (ACL):
Jialu Li, Hao Tan, and Mohit Bansal. 2021. Improving Cross-Modal Alignment in Vision Language Navigation via Syntactic Information. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 1041–1050, Online. Association for Computational Linguistics.
Cite (Informal):
Improving Cross-Modal Alignment in Vision Language Navigation via Syntactic Information (Li et al., NAACL 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.naacl-main.82.pdf
Video:
 https://aclanthology.org/2021.naacl-main.82.mp4
Code
 jialuli-luka/SyntaxVLN
Data
RxR