Call for Papers

Overview

Natural Language Processing is being revolutionized by deep learning. However, deep learning requires large amounts of annotated data, and its advantage over traditional statistical methods typically diminishes when such data is not available. Large amounts of annotated data simply do not exist for many low-resource languages. Even for high-resource languages it can be difficult to find linguistically annotated data of sufficient size and quality to allow neural methods to excel; this remains true even as few-shot learning approaches have gained popularity in recent years.

This workshop aims to bring together researchers from the NLP and ML communities who work on learning with neural methods when there is not enough data for those methods to succeed out-of-the-box. Specifically, it will provide attendees with an overview of new and existing approaches from various disciplines, and enable them to distill principles that can be more generally applicable. We will also discuss the main challenges arising in this setting, and outline potential directions for future progress.

Topics of interest include but are not limited to:

• Few- and zero-shot learning

• Prompting

• Transfer learning

• Multi-task learning

• Learning-to-Learn and Meta-Learning

• Semi-supervised learning

• Self-supervised learning

• Unsupervised learning

• Bandit/Reinforcement learning to learn from weak/sparse supervision

• Domain adaptation

• Universal or multilingual representations

• Active learning

Submission Guidelines

Please submit your paper using START: https://www.softconf.com/naacl2022/DeepLo2022/

We solicit three categories of papers: regular workshop papers, extended abstracts and cross-submissions. All papers may be long (maximum 8 pages plus references) or short (maximum 4 pages plus references), both with unlimited references, following the ARR formatting requirements (see https://aclrollingreview.org/cfp). Reviewing will be double-blind, and thus no author information should be included in the papers; self-reference that identifies the authors should be avoided or anonymized. Camera-ready versions of papers will be given one additional page of content, and no limit to additional pages for references.

Regular Workshop Papers:

The reported research should be substantially original. Accepted papers will appear in the workshop proceedings. Each submission can be accompanied by a single PDF appendix (supplementary material). The paper submissions need to remain fully self-contained, as these supplementary materials are completely optional, and reviewers are not even asked to review or download them. Supplementary materials need to be fully anonymized to preserve the double-blind reviewing policy.

Only regular workshop papers will be included in the proceedings as archival publications, and only regular workshop papers will be eligible for best paper prizes.

Extended Abstracts:

Preliminary but interesting ideas or results that have not been published before may be submitted as extended abstracts. Accepted extended abstracts will be presented as posters, and included in the workshop program and handbook, but will not be included in the workshop proceedings. Extended abstract submissions are therefore ideal for preliminary work which would benefit from exposure but is not ready for publication.

Cross-Submissions:

In addition to previously unpublished work, we also solicit recent papers on relevant topics that have appeared in a non-NLP venue (e.g., workshop or conference papers at NeurIPS or ICML). Accepted cross-submissions will be presented as posters, with an indication of original venue, but will not be included in the workshop program or proceedings, and are not eligible for best paper prizes. Cross- submissions are ideal for related work which would benefit from exposure to the DeepLo audience. Submission length and format are determined by the original venue. Interested authors should submit their papers in PDF format with a note on the original venue.