Causal Event Graph-Guided Language-based Spatiotemporal Question Answering


  • Kaushik Roy University of South Carolina
  • Alessandro Oltramari Bosch Center for Artificial Intelligence
  • Yuxin Zi University of South Carolina
  • Chathurangi Shyalika University of South Carolina
  • Vignesh Narayanan University of South Carolina
  • Amit Sheth University of South Carolina



Causal Knowledge, Language Understanding, Question Answering, Spatiotemporal


Large Language Models have excelled at encoding and leveraging language patterns in large text-based corpora for various tasks, including spatiotemporal event-based question answering (QA). However, due to encoding a text-based projection of the world, they have also been shown to lack a full bodied understanding of such events, e.g., a sense of intuitive physics, and cause-and-effect relationships among events. In this work, we propose using causal event graphs (CEGs) to enhance language understanding of spatiotemporal events in language models, using a novel approach that also provides proofs for the model’s capture of the CEGs. A CEG consists of events denoted by nodes, and edges that denote cause and effect relationships among the events. We perform experimentation and evaluation of our approach for benchmark spatiotemporal QA tasks and show effective performance, both quantitative and qualitative, over state-of-the-art baseline methods.






Empowering Machine Learning and Large Language Models with Domain and Commonsense Knowledge