BDIQA: A New Dataset for Video Question Answering to Explore Cognitive Reasoning through Theory of Mind

Authors

  • Yuanyuan Mao Shanghai Key Laboratory of Multidimensional Information Processing, ECNU, Shanghai, China Department of Computer Science and Technology, East China Normal University
  • Xin Lin Shanghai Key Laboratory of Multidimensional Information Processing, ECNU, Shanghai, China Department of Computer Science and Technology, East China Normal University
  • Qin Ni Key Laboratory of Multilingual Education with AI, Shanghai International Studies University
  • Liang He Shanghai Key Laboratory of Multidimensional Information Processing, ECNU, Shanghai, China Department of Computer Science and Technology, East China Normal University

DOI:

https://doi.org/10.1609/aaai.v38i1.27814

Keywords:

CMS: Conceptual Inference and Reasoning, CMS: Simulating Human Behavior, CV: Video Understanding & Activity Analysis

Abstract

As a foundational component of cognitive intelligence, theory of mind (ToM) can make AI more closely resemble human thought processes, thereby enhancing their interaction and collaboration with human. In particular, it can significantly improve a model's comprehension of videos in complex scenes. However, current video question answer (VideoQA) datasets focus on studying causal reasoning within events, few of them genuinely incorporating human ToM. Consequently, there is a lack of development in ToM reasoning tasks within the area of VideoQA. This paper presents BDIQA, the first benchmark to explore the cognitive reasoning capabilities of VideoQA models in the context of ToM. BDIQA is inspired by the cognitive development of children's ToM and addresses the current deficiencies in machine ToM within datasets and tasks. Specifically, it offers tasks at two difficulty levels, assessing Belief, Desire and Intention (BDI) reasoning in both simple and complex scenarios. We conduct evaluations on several mainstream methods of VideoQA and diagnose their capabilities with zero-shot, few-shot and supervised learning. We find that the performance of pre-trained models on cognitive reasoning tasks remains unsatisfactory. To counter this challenge, we undertake thorough analysis and experimentation, ultimately presenting two guidelines to enhance cognitive reasoning derived from ablation analysis.

Published

2024-03-25

How to Cite

Mao, Y., Lin, X., Ni, Q., & He, L. (2024). BDIQA: A New Dataset for Video Question Answering to Explore Cognitive Reasoning through Theory of Mind. Proceedings of the AAAI Conference on Artificial Intelligence, 38(1), 583-591. https://doi.org/10.1609/aaai.v38i1.27814

Issue

Section

AAAI Technical Track on Cognitive Modeling & Cognitive Systems