Adapt-NLP

Adapt-NLP: The Second Workshop on Domain Adaptation for NLP

The growth in computational power and the rise of Deep Neural Networks (DNNs) have revolutionized the field of Natural Language Processing (NLP). The ability to collect massive dataset with the capacity to train big models on powerful GPUs, has yielded NLP-based technology that was beyond imagination only a few years ago.

However, many NLP algorithms rely on the fundamental assumption that the training and the test sets follow the same underlying distribution. When these distributions do not match, a phenomenon known as domain shift, such models are likely to encounter performance drops. Despite the growing availability of heterogeneous data, many NLP domains still lack the amounts of labeled data required to feed data-hungry neural models, and in some domains and languages even unlabeled data is scarce. As a result, the problem of domain adaptation, training an algorithm on annotated data from one or more source domain and applying it to other target domains, is a fundamental NLP challenge.

Important Dates:

~~Jan 18, 2021 – Submission deadline.~~

~~Jan 24, 2021 – Submission deadline (Extension).~~

~~Feb 18, 2021 – Notification of acceptance.~~

~~Mar 1, 2021 – Camera-ready papers due.~~

~~April 20, 2021 – Workshop Dates.~~

Note: All deadlines are 11:59PM UTC-12:00.

Workshop Goals

We propose a workshop that aims to bring together researchers and ideas in the broad spectrum of NLP research in domain adaptation, zero-shot transfer learning and related setups where the data distribution differs between the train and test phases. In spite of the wide interest in the research community and its high relevance to industrial text-based applications, the last workshop that was devoted to domain adaptation was held in 2010. Considering the substantial progress in NLP since 2010, and the particular advancements related to DA, we believe there is a strong need in such a workshop, that will facilitate a discussion among interested researchers, and bring together researchers who today focus on an isolated instance of the more general problem.

Main Workshop Topics

The topics of the workshop include, but are not restricted to:

Novel DA algorithms, for existing as well as new setups.

Extending DA research to new domains and tasks through both novel datasets and algorithmic approaches.

Proposing novel zero-shot and few-shot algorithms and discussing their relevance for DA.

Exploring the similarities and differences between algorithmic approaches to DA, cross-lingual and cross-task learning.

A conceptual discussion of the definitions of fundamental concepts such as domain, transfer as well as zero-shot and few-shot learning.

Introducing and exploring novel or under-explored DA setups, aiming towards realistic and applicable ones (e.g., one-to-many DA, many-to-many DA and DA when the target domain is unknown when training on the source domain).

Invited Speakers