CoDraw: Collaborative Drawing as a Testbed for Grounded Goal-driven Communication
This work provides a testbed for studying language, perception, and action in AI, addressing the problem of grounded communication for researchers in human-AI interaction, though it is incremental as it builds on existing collaborative task frameworks.
The authors introduced CoDraw, a collaborative drawing game between two agents to study grounded goal-driven communication, and collected a dataset of ~10K dialogs with ~138K messages to evaluate models through automated and human-in-the-loop methods.
In this work, we propose a goal-driven collaborative task that combines language, perception, and action. Specifically, we develop a Collaborative image-Drawing game between two agents, called CoDraw. Our game is grounded in a virtual world that contains movable clip art objects. The game involves two players: a Teller and a Drawer. The Teller sees an abstract scene containing multiple clip art pieces in a semantically meaningful configuration, while the Drawer tries to reconstruct the scene on an empty canvas using available clip art pieces. The two players communicate with each other using natural language. We collect the CoDraw dataset of ~10K dialogs consisting of ~138K messages exchanged between human players. We define protocols and metrics to evaluate learned agents in this testbed, highlighting the need for a novel "crosstalk" evaluation condition which pairs agents trained independently on disjoint subsets of the training data. We present models for our task and benchmark them using both fully automated evaluation and by having them play the game live with humans.