CompGuessWhat?!
An instance of the GROLLA framework
| CARVIEW |
Instance of a multi-task framework for evaluating the quality of learned neural representations, in particular concerning attribute grounding. Accepted as long paper to the Association for Computational Linguistics (ACL) 2020.
An instance of the GROLLA framework
Learning to guess an hidden object in a reference scene
Learning to predict the target object attributes
Learning to play games involving unknown objects and reference scenes