Towards Universal Semantic Tagging

TL;DR

This paper introduces the concept of universal semantic tagging, a language-neutral approach to annotate words with semantically rich tags, enhancing multilingual semantic analysis and cross-lingual parsing capabilities.

Contribution

It presents the initial semantic tagset, demonstrates its semantic granularity and cross-lingual applicability, and provides baseline results for universal semantic tagging.

Findings

Tags offer fine-grained semantic information

Tags are effective for cross-lingual semantic parsing

Annotated corpus demonstrates practical application

Abstract

The paper proposes the task of universal semantic tagging---tagging word tokens with language-neutral, semantically informative tags. We argue that the task, with its independent nature, contributes to better semantic analysis for wide-coverage multilingual text. We present the initial version of the semantic tagset and show that (a) the tags provide semantically fine-grained information, and (b) they are suitable for cross-lingual semantic parsing. An application of the semantic tagging in the Parallel Meaning Bank supports both of these points as the tags contribute to formal lexical semantics and their cross-lingual projection. As a part of the application, we annotate a small corpus with the semantic tags and present new baseline result for universal semantic tagging.