Chair: Grégoire Montavon
(Freie Universität Berlin)

Following the success of eXplainable AI at generating faithful and understandable explanations of complex Machine Learning (ML) models, researchers have started to ask how to integrate explainability more tightly into machine learning, data science, or decision-making pipelines. This includes how to use explanations for supporting high-stakes human decisions, how to systematically improve an ML model from one or multiple explanations, how to use explanations to make a model comply with specific norms, or how to generate scientific hypotheses from a trained ML model augmented with explanations. The topics of interest include but are not limited to:

Rich, structured explanations (e.g. higher-order, concepts-based, hierarchical, counterfactual, disentangled).
XAI-based methods for data/model visualization and decision support.
Integration of user-in-the-loop mechanisms in XAI.
XAI techniques to lower a model’s reliance on spuriously correlated features.
XAI techniques to enrich predictions with predictive uncertainty.
XAI techniques for generating new scientific hypotheses.
Evaluation metrics and benchmarks for XAI actionability.
Use cases for actionable XAI.

Submitted manuscripts must be novel and not substantially duplicate existing work. Manuscripts must be written using Springer’s Lecture Notes in Computer Science (LNCS) in the format provided here. Latex and word files are admitted: however, the former is preferred (word template, latex template, latex in overleaf). All submissions and reviews will be handled electronically. The conference has a no dual submission policy, so submitted manuscripts should not be currently under review at another publication venue.

Articles must be submitted using the easy-chair platform here.

The contact author must provide the following information: paper title, all author names, affiliations, postal address, e-mail address, and at least three keywords.

The conference will not require a strict page number, as we believe authors have different writing styles and would like to produce scientific material differently. However, the following types of articles are admitted:

full articlesbetween 12 and 24 pages (including references)
short articlesbetween 8 and 12 pages (including references)
extended abstractsbetween 4 and 8 pages (including references)

Full articles should report on original and substantial contributions of lasting value, and the work should concern the theory and/or practice of Explainable Artificial Intelligence (xAI). Moreover, manuscripts showcasing the innovative use of xAI methods, techniques, and approaches and exploring the benefits and challenges of applying xAI-based technology in real-life applications and contexts are welcome. Evaluations of proposed solutions and applications should be commensurate with the claims made in the article. Full articles should reflect more complex innovations or studies and have a more thorough discussion of related work. Research procedures and technical methods should be presented sufficiently to ensure scrutiny and reproducibility. We recognise that user data may be proprietary or confidential, therefore we encourage sharing (anonymized, cleaned) data sets, data collection procedures, and code. Results and findings should be communicated clearly, and implications of the contributions for xAI as a field and beyond should be explicitly discussed.
Shorter articles should generally report on advances that can be described, set into context, and evaluated concisely. These articles are not ‘work-in-progress’ reports but complete studies focused on smaller but complete research work, simple to describe. For these articles, the discussion of related work and contextualisation in the wider body of knowledge can be smaller than that of full articles.
Extended abstracts should contain the definition of a problem and the presentation of a solution, comparisons to related work, and other details expected in a research manuscript but not in an abstract. They are not simply long abstracts or ‘work-in-progress’. An extended abstract is a research article whose ideas and significance can be understood in less than an hour. Producing an extended abstract can be more demanding than producing a full or short research article. Some things that can be omitted from an extended abstract, such as future work, details of proofs or implementation that should seem plausible to reviewers, and ramifications not relevant to the key ideas of the abstract. It should also contain enough bibliographic references to follow the main argument of the proposed research.

Special track articles

The article submitted to the special tracks follows the normal procedure and must be submitted via easy-chair, as mentioned above. The types of articles admitted are full articles, shorter articles and extended abstracts, as described above. The author of an article for a special track must select the name of such special track in the list of topics in easy-chair, along with other relevant topics.

Anonymity for review

The submitted article (a .pdf) must be anonymous because the conference uses a double-blind review process. Therefore, authors must omit their names and affiliations in the submitted .pdf file and avoid obvious identifying statements. For instance, citations to the author’s prior work should be made in the third person. Failure to anonymize your submission could result in desk rejection.

Ethical & Human Subjects Considerations

The conference organisers expect authors to discuss the ethical considerations and the impact of the presented work and/or its intended application, where appropriate. Additionally, all authors must comply with ethical standards and regulatory guidelines associated with human subjects research, including using personally identifiable data and research involving human participants. Manuscripts reporting on human subjects research must include a statement identifying any regulatory review the research is subject to (and identifying the form of approval provided) or explaining the lack of required review.

Further style instructions

We ask the authors to start the reference section on a new page. Appendices count toward the page limit. Supplementary material, if any, should be linked to an external source using an anonymized URL.

*All dates are Anywhere on Heart time (AoE)

Article (main track & special tracks)

Abstracts registration deadline* (easy-chair):April 15, 2023
Article submission deadline* (easy-chair):April 20, 2023
Notification of acceptance:May 12, 2023
Registration & camera ready:May 19, 2023
Publication (Springer CCIS series)August 2023
*full, short and extended abstract articles

Doctoral consortium (DC) proposal

DC Proposal registration deadline (easy-chair):April 16th 2023
DC Proposal submission deadline (easy-chair):April 30, 2023
Notification of acceptance:May 7, 2023
Registration:May 19, 2023
Publication (planned with CEUR-WS.org*)August 2023
*Proceedings shall be submitted to CEUR-WS.org for online publication

Late-breaking work & demos

Late-breaking work & demo registration (easy-chair):May 21, 2023
Late-breaking work & demo submission (easy-chair):May 28, 2023
Notification of acceptance:June 06, 2023
Publication (planned with CEUR-WS.org*)August 2023
*Proceedings shall be submitted to CEUR-WS.org for online publication

Special track proposal

Proposal submission (contact):Anytime before February 08 February 15 2023

Panel discussion

Panel Discussion proposals:May 21, 2023
Notification of acceptance:May 28, 2023
Registration of Panel Discussions facilitators:June 06, 2023

Conference

The World Conference on eXplainable AI26-28 July 2023

The Peer-Review process

All articles submitted within the deadlines and according to the guidelines will be subjected to a double-blind review. Papers that are out of scope, incomplete, or lack sufficient evidence to support the basic claims may be rejected without full review. Furthermore, reviewers will be asked to comment on whether the length is appropriate for the contribution. Each of the submitted articles will be reviewed by at least three members of the Scientific Committee.

After completion of the review process, the authors will be informed about the acceptance or rejection of the submitted work. The reviewers’ comments will be available to the authors in both cases. In case of acceptance, authors must meet the recommendations for improvement and prepare and submit the definitive version of the work up to the camera-ready paper submission deadline. In case of failure to consider the recommendations made by the reviewers, the organizing committee and the editors reserve the right not to include these works in the conference proceedings.

The article’s final version must follow the appropriate style guide and contain the authors’ data (names, institutions and emails) and the ORCID details. Submitted articles will be evaluated according to their originality, technical soundness, significance of findings, contribution to knowledge, and clarity of exposition and organisation.

According to the quality of the accepted article and its rank among all the other accepted manuscripts, it can be accepted for a full or short presentation or as a poster.

Code of Ethics

Inspired by the code of ethics put forward by the Association of Computing Machinery, the programme committee, supervised by the general conference chairs and organisers, have the right to desk-reject manuscripts that perpetuate harmful stereotypes, employ unethical research practices, or uncritically present outcomes or implications that disadvantage minoritized communities. Further, reviewers of the scientific committee will be explicitly asked to consider whether the research was conducted in compliance with professional, ethical standards and applicable regulatory guidelines. Failure to do so could lead to a desk-rejection

Proceedings publication

Each accepted and presented full, short and extended abstract manuscript, either as an oral presentation or as a poster, will be included in the conference proceedings by Springer in Communications in Computer and Information Science, edited by the general chair.  At least one author must register for the conference by the early registration deadline. The official publication date is when the publisher makes the proceedings available online. This date will be after the conference and can take a number of weeks.

Indexing