English

Learning Action-Effect Dynamics for Hypothetical Vision-Language Reasoning Task

Computer Vision and Pattern Recognition 2022-12-09 v1

Abstract

'Actions' play a vital role in how humans interact with the world. Thus, autonomous agents that would assist us in everyday tasks also require the capability to perform 'Reasoning about Actions & Change' (RAC). This has been an important research direction in Artificial Intelligence (AI) in general, but the study of RAC with visual and linguistic inputs is relatively recent. The CLEVR_HYP (Sampat et. al., 2021) is one such testbed for hypothetical vision-language reasoning with actions as the key focus. In this work, we propose a novel learning strategy that can improve reasoning about the effects of actions. We implement an encoder-decoder architecture to learn the representation of actions as vectors. We combine the aforementioned encoder-decoder architecture with existing modality parsers and a scene graph question answering model to evaluate our proposed system on the CLEVR_HYP dataset. We conduct thorough experiments to demonstrate the effectiveness of our proposed approach and discuss its advantages over previous baselines in terms of performance, data efficiency, and generalization capability.

Keywords

Cite

@article{arxiv.2212.03866,
  title  = {Learning Action-Effect Dynamics for Hypothetical Vision-Language Reasoning Task},
  author = {Shailaja Keyur Sampat and Pratyay Banerjee and Yezhou Yang and Chitta Baral},
  journal= {arXiv preprint arXiv:2212.03866},
  year   = {2022}
}

Comments

11 pages, 9 figures; Accepted at Findings of EMNLP 2022. arXiv admin note: substantial text overlap with arXiv:2212.03433

R2 v1 2026-06-28T07:25:07.627Z