Grounding Symbols in Multi-Modal Instructions

Yordan Hristov, Svetlin Penkov, Alex Lascarides, Subramanian Ramamoorthy


Abstract
As robots begin to cohabit with humans in semi-structured environments, the need arises to understand instructions involving rich variability—for instance, learning to ground symbols in the physical world. Realistically, this task must cope with small datasets consisting of a particular users’ contextual assignment of meaning to terms. We present a method for processing a raw stream of cross-modal input—i.e., linguistic instructions, visual perception of a scene and a concurrent trace of 3D eye tracking fixations—to produce the segmentation of objects with a correspondent association to high-level concepts. To test our framework we present experiments in a table-top object manipulation scenario. Our results show our model learns the user’s notion of colour and shape from a small number of physical demonstrations, generalising to identifying physical referents for novel combinations of the words.
Anthology ID:
W17-2807
Volume:
Proceedings of the First Workshop on Language Grounding for Robotics
Month:
August
Year:
2017
Address:
Vancouver, Canada
Editors:
Mohit Bansal, Cynthia Matuszek, Jacob Andreas, Yoav Artzi, Yonatan Bisk
Venue:
RoboNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
49–57
Language:
URL:
https://aclanthology.org/W17-2807
DOI:
10.18653/v1/W17-2807
Bibkey:
Cite (ACL):
Yordan Hristov, Svetlin Penkov, Alex Lascarides, and Subramanian Ramamoorthy. 2017. Grounding Symbols in Multi-Modal Instructions. In Proceedings of the First Workshop on Language Grounding for Robotics, pages 49–57, Vancouver, Canada. Association for Computational Linguistics.
Cite (Informal):
Grounding Symbols in Multi-Modal Instructions (Hristov et al., RoboNLP 2017)
Copy Citation:
PDF:
https://aclanthology.org/W17-2807.pdf