Visual Commonsense Causal Reasoning From a Still Image

• Published in: IEEE access Vol. 13; pp. 85084 - 85097
• Main Authors: Wu, Xiaojing, Guo, Rui, Li, Qin, Zhu, Ning
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2025; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Cause effect analysis; Commonsense reasoning; Correlation; Electronic mail; Estimation; Large language models; multimodal large language model; Natural languages; Question answering (information retrieval); Reasoning; Visual commonsense reasoning; Visual effects; visual event reasoning; Visualization

Even from a still image, humans exhibit the ability to ratiocinate diverse visual cause-and-effect relationships of events preceding, succeeding, and extending beyond the given image scope. Previous work on commonsense causal reasoning (CCR) aimed at understanding general causal dependencies among c...