ITU
ITU

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

Research output: Conference Article in Proceeding or Book/Report chapterArticle in proceedingsResearchpeer-review

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

View graph of relations

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 lever- ages 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.
Original languageEnglish
Title of host publicationNAACL (North American Association for Computational Linguistics)
Place of PublicationMinneapolis
PublisherAssociation for Computational Linguistics
Publication dateJun 2019
ISBN (Electronic)978-1-950737-13-0
Publication statusPublished - Jun 2019

ID: 83863134