(Apologies for cross-posting)

*** DeeLIO 2022 - Second Call for Papers ***

Deep Learning Inside Out (DeeLIO)

The Third Workshop on Knowledge Extraction and Integration for Deep Learning Architectures
Website: https://sites.google.com/view/deelio-ws/

Collocated with ACL 2022 in Dublin, Ireland or online


Two possible submission procedures and deadlines (see also below):

Direct submission: Feb 28 (deadline for submission)
Already reviewed by ARR: Mar 21 (deadline to commit to DeeLIO)


Workshop description

The third edition of the DeeLIO workshop will be collocated with ACL 2022. The previous editions were satellite workshops of the NAACL 2021 and EMNLP 2020 conferences.

The central topics of the workshop are knowledge extraction and integration for deep learning architectures. Following the success of deep learning methods in NLP, the focus of DeeLIO is to explore the linguistic and real-world knowledge that can be extracted from deep learning models, and to identify potentially useful sources of language-related information that can make the models “smarter” and better in natural language understanding tasks.

The Deep Learning Inside Out (DeeLIO) workshop aims precisely to bring together the knowledge interpretation, extraction and integration lines of research in deep learning, and cover the area in between. It will explore the linguistic and world knowledge neural networks encode, how this knowledge is exploited by the models for performing specific tasks, as well as ways to enhance the quality of this knowledge by leveraging external hand-crafted resources.

We especially encourage works on structurally diverse languages and low-data regimes. We hope that DeeLIO will inspire novel variation-aware transfer learning and multilingual solutions on how to use the knowledge from resource-rich languages to inform deep learning architectures in settings where external repositories are scarce or missing.

Topics of interest include but are not limited to:

Interpretation and explanation methodology (such as probing, adversarial techniques, counterfactual methodology);
Exploration of the types of linguistic and world knowledge neural models, architectures and representations encode;
Exploration of the knowledge neural models actually use for performing specific tasks;
Extraction of linguistic and world knowledge from deep learning models;
Analysis of the limitations of the knowledge acquired by current neural models;
Integration of external knowledge into deep learning models (under the form of semantic specialization of embeddings, retrofitting, joint modeling, or other);
Benefits of using external versus internally encoded knowledge, and their combination, for knowledge enhancement in neural networks;
Development and enrichment of lexico-semantic knowledge resources using deep learning models;
Using external knowledge in resource-lean languages through transfer techniques or joint multilingual modelling;
Analysis of the interplay between prompts and language models in few-shot learning.

We are delighted to announce our three keynote speakers:

Yejin Choi (Allen Institute for AI (AI2) and University of Washington)

Tal Linzen (New York University)

Allyson Ettinger (University of Chicago)

Important Dates

February 28, 2022: Submission Due Date
March 21, 2022: ARR Commitment deadline
March 26, 2022: Notification of Acceptance
April 10, 2022: Camera-ready papers due
May 26 or 27, 2022: Workshop co-located with ACL

Note: All deadlines are 11:59PM (UTC-12:00 / Anywhere on Earth).

Submission

We invite the submission of long and short papers on original and unpublished research in any topic related with knowledge interpretation and integration in deep neural networks. Long papers may consist of up to 8 pages of content + references. Short papers may consist of up to 4 pages of content + references. Upon acceptance, both types of papers will be given one additional page of content. Authors are encouraged to use this additional page for addressing reviewers’ comments in the final version.

Formatting: Papers should follow the paper formatting guidelines for *ACL conferences available here.

Submission: Authors can submit their papers on the DeeLIO website on OpenReview by February, 28. These papers will be reviewed by our reviewers. Papers that have been already reviewed through ARR will also be considered for acceptance. In this case, the authors need to commit to DeeLIO by March, 21.

Multiple submission policy: We follow the ARR Multiple Submission Policy. See the ARR CFP guidelines.

​Ethics Policy: Authors are required to honour the ethical code set out in the ACL Code of Ethics. See the ARR Author Checklist on Ethics.

​Reproducibility Criteria: See ARR Author Checklist on reproducibility.

Workshop Chairs

Eneko Agirre (University of the Basque Country)
Marianna Apidianaki (University of Pennsylvania)
Ivan Vulić (University of Cambridge)

Contact: deelio-ws at googlegroups dot com