Textual Inference and Meaning Representation in Human Robot Interaction

Abstract

This paper provides a first investigation over existing textual inference paradigms in order to propose a generic framework able to capture major semantic aspects in Human Robot Interaction (HRI). We investigate the use of general semantic paradigms used in Natural Language Un-derstanding (NLU) tasks, such as Seman-tic Role Labeling, over typical robot com-mands. The semantic information ob-tained is then represented under the Ab-stract Meaning Representation. AMR is a general representation language useful to express different level of semantic infor-mation without a strong dependence to the syntactic structure of an underlying sen-tence. The final aim of this work is to find an effective synergy between HRI and NLU. 1