Abstract

This demo paper describes a web-based editor of Abstract Meaning Representations (AMRs) to build semantic representations of English. It supports annotators in building a corpus of gold AMRs, which can be used in applications such as semantic parsing and semantics-based machine translation. The AMR Editor currently has about 50 active users worldwide, including at the Linguistic Data Consortium (LDC) and SDL Language Weaver, and has been used to write more than 10,000 AMRs in 2012, for sentences from newswire, web and other genres. It is publicly available at http://www. isi. edu/ulf/amr/AMR-editor. html with a guest access option for users without an account.

Date

March 19, 2026

Authors

Ulf Hermjakob

Journal

Marina del Rey, CA. USC Information Sciences Institute