[CFP] ExUM Workshop @UMAP 2026 - Call For Papers - *** DEADLINE APRIL, 9, 2026 ***

[Marco Polignano]

*** Apologies for cross postings ***

*ExUM Workshop @UMAP 2026 - Call For Papers - *** DEADLINE APRIL, 9, 2026 ***
----------------------------------------------------------------------
Workshop on Explainable User Models and Personalised Systems (ExUM@UMAP 2026)
co-located with UMAP 2026 (https://www.um.org/umap2026/) - 34th ACM Conference on User Modeling, Adaptation and Personalization, June 8-11, 2026 | Gothenburg, Sweden.

Twitter: https://x.com/ExUM_Workshop
Web: https://exum-umap.github.io

For any information: marco.p...@uniba.it, catald...@uniba.it

=================
IMPORTANT DATES
=================

* Submission Deadline: April 9, 2026
* Notification: April 28, 2026
* Workshop Papers Camera-ready Submission (CEUR proceedings): May 7, 2026

Please note: All deadlines refer to 11:59 pm AoE (Anywhere on Earth) time.

=========
ABSTRACT
=========

Adaptive and personalized systems, including Large Language Models (LLMs) have rapidly emerged as transformative technologies, deeply integrated into various aspects of modern life. From conversational agents that provide human-like interactions to recommendation algorithms that curate personalized content such as music, movies, or products, these systems are reshaping how individuals interact with digital platforms. As their influence grows in supporting decision-making, content delivery, and user engagement, it becomes increasingly important to address key issues such as transparency, fairness, and user trust. Frameworks like the EU General Data Protection Regulation (GDPR) and EU AI-Act have highlighted the 'right to explanation,' underscoring the need for users to understand the mechanisms driving these intelligent systems. Despite that, a significant portion of research in these fields has been geared toward maximizing performance, i.e., improving the relevance of the results of personalized systems, often at the expense of explainability. This trade-off risks eroding user trust and poses problems of compliance with ethical and regulatory standards. This initiative aims to create a forum for discussing the pressing challenges, innovative methodologies, and future directions in exploring how transparency, explainability, and user-centric design can be incorporated into these technologies to make them not only effective but also trustworthy, ethical, and aligned with the diverse needs and expectations of their users.

======
TOPICS
======
Topics of interest include but are not limited to:

- TRANSPARENT AND EXPLAINABLE PERSONALIZATION STRATEGIES
    Scrutable User Models
    Transparent User Profiling and Personal Data Extraction
    Explainable Personalization and Adaptation Methodologies
    Novel strategies (e.g., conversational recommender systems) for building transparent algorithms
    Transparent Personalization and Adaptation to Groups of users

- TRANSPARENT PERSONALIZATION BASED ON LARGE LANGUAGE MODELS

- DESIGNING EXPLANATION ALGORITHMS
    Explanation algorithms based on item description and item properties
    Explanation algorithms based on user-generated content (e.g., reviews)
    Explanation algorithms based on collaborative information
    Building explanation algorithms for opaque personalization techniques (e.g., neural networks, matrix factorization, deep learning approaches)
    Explanation algorithms based on methods to build group models

- DESIGNING TRANSPARENT AND EXPLAINABLE USER INTERFACES
    Transparent User Interfaces
    Designing Transparent Interaction methodologies
    Novel paradigms (e.g. chatbots, LLMs) for building transparent models

- EVALUATING TRANSPARENCY AND EXPLAINABILITY
    Evaluating Transparency in interaction or personalization
    Evaluating Explainability of the algorithms
    Designing User Studies for evaluating transparency and explainability
    Novel metrics and experimental protocols

- OPEN ISSUES IN TRANSPARENT AND EXPLAINABLE USER MODELS AND PERSONALIZED SYSTEMS
    Ethical issues (fairness and biases) in user / group models and personalized systems
    Privacy management of personal and social data
    Discussing Recent Regulations (GDPR) and future directions


============
SUBMISSIONS
============
We encourage the submission of contributions investigating novel methodologies to exploit heterogeneous personal data and approaches to build transparent and scrutable user models. In particular, we accept three kinds of submissions:

(A) Regular papers (10 or more standard pages, including references (CEUR format));
(B) Short papers (5–9 standard pages, including references (CEUR format));
(C) Ongoing projects, Demo, Position and Perspective Papers (less than 5 standard pages, including references (CEUR format));

Submission site: https://easychair.org/my2/conference?conf=exum2026

All submitted papers will be evaluated by at least two members of the program committee, based on originality, significance, relevance, and technical quality. Note that the references do not count toward page limits. Submissions should be single-blinded, i.e. authors’ names should be included in the submissions. Papers must be formatted according to the workflow for CEUR publications. All accepted papers will be published by CEUR as a joint volume of Workshop UMAP 2026 Proceedings. At least one author of each accepted paper must register for the particular workshop and present the paper there.

CEUR Templates and Formatting:

All papers must use the CEUR-WS template in one-column format (LaTeX is strongly preferred).
    Offline version: http://ceur-ws.org/Vol-XXX/CEURART.zip
    Overleaf version: https://www.overleaf.com/latex/templates/template-for-submissions-to-ceur-workshop-proceedings-ceur-ws-dot-org/wqyfdgftmcfw

If LaTeX is not used, authors must strictly follow the ODT template instructions (provided within the CEURART package).

    Microsoft Word must not be used for the ODT template.
    The Libertinus font family is mandatory; installation instructions are included in the template.

Papers that do not comply with these requirements will not be suitable for publication in CEUR-WS.

Declaration of Generative AI:

Each paper must include a mandatory Declaration of Generative AI, in accordance with the CEUR-WS Generative AI Policy.
https://ceur-ws.org/GenAI/Policy.html


=============
ORGANIZATION
=============
Marco Polignano - Bari, Italy
Amra Delic - University of Sarajevo, Bosnia-Herzegovina
Cataldo Musto - University of Bari, Italy
Amon Rapp – University of Torino, Italy
Giovanni Semeraro - University of Bari, Italy
Juergen Ziegler - University of Duisburg Essen, Germany