Characterizing Datasets for Social Visual Question Answering, and the New TinySocial Dataset
Abstract
Modern social intelligence includes the ability to watch videos and answer questions about social and theory-of-mind-related content, e.g., for a scene in Harry Potter, "Is the father really upset about the boys flying the car?" Social visual question answering (social VQA) is emerging as a valuable methodology for studying social reasoning in both humans (e.g., children with autism) and AI agents. However, this problem space spans enormous variations in both videos and questions. We discuss methods for creating and characterizing social VQA datasets, including 1) crowdsourcing versus in-house authoring, including sample comparisons of two new datasets that we created (TinySocial-Crowd and TinySocial-InHouse) and the previously existing Social-IQ dataset; 2) a new rubric for characterizing the difficulty and content of a given video; and 3) a new rubric for characterizing question types. We close by describing how having well-characterized social VQA datasets will enhance the explainability of AI agents and can also inform assessments and educational interventions for people.
Keywords
Cite
@article{arxiv.2010.11997,
title = {Characterizing Datasets for Social Visual Question Answering, and the New TinySocial Dataset},
author = {Zhanwen Chen and Shiyao Li and Roxanne Rashedi and Xiaoman Zi and Morgan Elrod-Erickson and Bryan Hollis and Angela Maliakal and Xinyu Shen and Simeng Zhao and Maithilee Kunda},
journal= {arXiv preprint arXiv:2010.11997},
year = {2020}
}
Comments
To appear in the Joint IEEE International Conference on Development and Learning and on Epigenetic Robotics (ICDL), 2020