Publications

Learning Event Schema Temporally and Transmodally (LESTAT)

Abstract

This report summarizes the University of Southern California (USC) Information Sciences Institute (ISI) Learning Event Schema Temporally and Transmodally (LESTAT) teams effort on DARPAs Knowledge-directed Artificial Intelligence Reasoning Over Schemas (KAIROS) program. The goal was twofold: (1) to develop a human-machine collaborative process for building a library of schemas through leveraging natural language processing (NLP) techniques to augment human background knowledge and (2) to create a repository of schemas. Each script includes a natural language description of the steps in the complex event with links to the Wikidata ontology.

Date
May 1, 2024
Authors
Marjorie Freedman, Alex Hedges, Yash Kankanampati, Jacob Lichtefeld, Jon May, Ralph Weischedel