Workshop Description

Linguistic annotation of natural language corpora is the backbone of supervised methods of statistical natural language processing and has contributed to breakthroughs in practical natural language applications, most notably in Machine Translation, Machine Reading, Question Answering, and Recognizing Textual Entailment. Nonetheless, challenges remain in developing systems that can actually ``understand'' human language, including the ability to track entities in a text, understand the relations between these entities, track events and their participants described in a text, understand how events unfold in time, and distinguish events that have actually happened from events that are planned or intended, are uncertain, or did not happen at all. The fifteenth LAW and third DMR workshops are combined for the first time to provide a forum for presentation and discussion of innovative research on all aspects of linguistic annotation, with a special focus on the design and annotation of representations of the various elements of meaning contributing to understanding.

This workshop intends to bring together researchers who are producers and consumers of linguistic annotations and meaning representations. Through their interaction, we will gain a deeper understanding of both the kinds of annotations generally, and the key elements of meaning representations specifically, that are the most valuable to the NLP community. The workshop will also provide an opportunity for researchers to critically examine existing annotation frameworks and to explore opportunities and identify challenges in the design and use of linguistic annotations in multilingual settings.

Given that this is the first joint LAW and DMR workshop, a final goal of this workshop will be to explore multi-layered annotations and how different types of annotations can be brought together to enhance progress towards true understanding. This may include annotation schemas representing multiple facets of meaning (for example, both semantic roles and coreference), or leveraging one type of annotation to bootstrap obtaining another (for example, leveraging automatic SRL for a sentential meaning representation). This may also include annotations from multiple modalities (for example, transcribed text and accompanying images), as understanding can often hinge upon linguistic meaning within a particular situated context.

The workshop will include long (8+1 pages) and short (4+1 pages) papers, posters, and demonstrations relating to the following topics:

• Creation/evaluation of annotation schemas and meaning representations;
• Cross-framework comparison of annotations and meaning representations;
• Methods for manual and automatic annotation, including automatic parsing of meaning representations;
• Use and evaluation of annotation software and frameworks;
• Semi-supervised "human in the loop" methods of annotation and crowd-sourcing approaches;
• Using annotations and meaning representations in real-world applications;
• Issues in applying annotations and meaning representations to multilingual settings;
• Multi-layered and multi-modal annotations;
• Any other topics that address the design, processing, and use of linguistic annotations and/or meaning representations.

Background

Linguistic annotation of natural language corpora is the backbone of supervised methods of statistical natural language processing. The Linguistic Annotation Workshop (LAW) is the annual workshop of the ACL Special Interest Group on Annotation (SIGANN), and it provides a forum for the presentation and discussion of innovative research on all aspects of linguistic annotation, including the creation and evaluation of annotation schemes, methods for automatic and manual annotation, use and evaluation of annotation software and frameworks, representation of linguistic data and annotations, semi-supervised “human in the loop” methods of annotation, crowd-sourcing approaches, and more. As in the past, the LAW will provide a forum for annotation researchers to work towards standardization, best practices, and interoperability of annotation information and software.

Anti-Harassment Policy

ACL Anti-Harassment Policy.

Author Responsibilities

Papers must be of original, previously-unpublished work. Papers must be anonymized to support double-blind reviewing. If the paper is available as a preprint, this must be indicated on the submission form but not in the paper itself. In addition, EMNLP 2021 will follow the same policy as ACL'2018 establishing an anonymity period (from submission to author notification) during which non-anonymous posting of preprints is not allowed. Also included in that policy are instructions to reviewers to not rate papers down for not citing recent preprints. Authors are asked to cite published versions of papers instead of preprint versions when possible.