The Impact of Answers in Referential Visual Dialog

Mauricio Mazuecos, Patrick Blackburn, Luciana Benotti


Abstract
In the visual dialog task GuessWhat?! two players maintain a dialog in order to identify a secret object in an image. Computationally, this is modeled using a question generation module and a guesser module for the questioner role and an answering model, the Oracle, to answer the generated questions. This raises a question: what’s the risk of having an imperfect oracle model?. Here we present work in progress in the study of the impact of different answering models in human generated questions in GuessWhat?!. We show that having access to better quality answers has a direct impact on the guessing task for human dialog and argue that better answers could help train better question generation models.
Anthology ID:
2021.reinact-1.2
Volume:
Proceedings of the Reasoning and Interaction Conference (ReInAct 2021)
Month:
October
Year:
2021
Address:
Gothenburg, Sweden
Editors:
Christine Howes, Simon Dobnik, Ellen Breitholtz, Stergios Chatzikyriakidis
Venue:
ReInAct
SIG:
SIGSEM
Publisher:
Association for Computational Linguistics
Note:
Pages:
8–13
Language:
URL:
https://aclanthology.org/2021.reinact-1.2
DOI:
Bibkey:
Cite (ACL):
Mauricio Mazuecos, Patrick Blackburn, and Luciana Benotti. 2021. The Impact of Answers in Referential Visual Dialog. In Proceedings of the Reasoning and Interaction Conference (ReInAct 2021), pages 8–13, Gothenburg, Sweden. Association for Computational Linguistics.
Cite (Informal):
The Impact of Answers in Referential Visual Dialog (Mazuecos et al., ReInAct 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.reinact-1.2.pdf
Data
GuessWhat?!MS COCO