Beyond task success: A closer look at jointly learning to see, ask, and GuessWhat

Ravi Shekhar, Aashish Venkatesh, Tim Baumgärtner, Elia Bruni, Barbara Plank, Raffaella Bernardi, Raquel Fernández


Abstract
We propose a grounded dialogue state encoder which addresses a foundational issue on how to integrate visual grounding with dialogue system components. As a test-bed, we focus on the GuessWhat?! game, a two-player game where the goal is to identify an object in a complex visual scene by asking a sequence of yes/no questions. Our visually-grounded encoder leverages synergies between guessing and asking questions, as it is trained jointly using multi-task learning. We further enrich our model via a cooperative learning regime. We show that the introduction of both the joint architecture and cooperative learning lead to accuracy improvements over the baseline system. We compare our approach to an alternative system which extends the baseline with reinforcement learning. Our in-depth analysis shows that the linguistic skills of the two models differ dramatically, despite approaching comparable performance levels. This points at the importance of analyzing the linguistic output of competing systems beyond numeric comparison solely based on task success.
Anthology ID:
N19-1265
Volume:
Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)
Month:
June
Year:
2019
Address:
Minneapolis, Minnesota
Editors:
Jill Burstein, Christy Doran, Thamar Solorio
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2578–2587
Language:
URL:
https://aclanthology.org/N19-1265
DOI:
10.18653/v1/N19-1265
Bibkey:
Cite (ACL):
Ravi Shekhar, Aashish Venkatesh, Tim Baumgärtner, Elia Bruni, Barbara Plank, Raffaella Bernardi, and Raquel Fernández. 2019. Beyond task success: A closer look at jointly learning to see, ask, and GuessWhat. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 2578–2587, Minneapolis, Minnesota. Association for Computational Linguistics.
Cite (Informal):
Beyond task success: A closer look at jointly learning to see, ask, and GuessWhat (Shekhar et al., NAACL 2019)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-5/N19-1265.pdf
Code
 additional community code
Data
GuessWhat?!MS COCOVisDial