Theory of sāmarthya in Indian grammatical tradition: The foundation of universal semantic representation

Abstract

Semantic Representations become useful resources for various multilingual NLP applications such as Machine Translation, Multilingual Generation, cross Lingual QA, to name a few. Universal Semantic Representation (USR) is a recently developed semantic representation system that is based on Indian Grammatical Tradition and Paninian Grammar. This paper critically examines the notion of sāmarthya from Indian Grammatical Tradition, proposes that the principles of sāmarthya can account for the well-formedness of integrated word-sentential-discourse structure, and finally explains how the theory of sāmarthya motivates the multi-layered information encoding strategy of USR.