The User-Centered Explanations in XAI (UCEX-XAI) Workshop aims to advance research on explainability in AI by focusing on human-centered approaches. While much of XAI research has emphasized algorithmic transparency and interpretability, there remains a gap in how AI-generated explanations align with the needs and cognitive processes of users. This workshop addresses key challenges in designing, evaluating, and applying user-centered explanations that support effective decision-making across different domains.
The workshop will facilitate discussions on best practices for designing adaptive, interactive, and multimodal explanations, as well as methodologies for evaluating their effectiveness. Topics will include user perception of AI explanations, cognitive biases, and ethical considerations in explanation design. By fostering interdisciplinary collaboration between AI, human-computer interaction (HCI), and cognitive science, the workshop aims to advance the development of more effective and accessible AI explanations.
The workshop will cover four key themes:
Designing Human-Centered XAI Systems – Approaches for creating explanations that are adaptable, interactive, and relevant to diverse users.
Evaluating XAI Explanations – Methods for assessing cognitive load, user trust, and effectiveness of AI explanations.
Dark Patterns and Ethical Concerns in XAI – Identifying misleading explanations and addressing cognitive biases in AI decision-making.
Applications Across Domains – Case studies and discussions on implementing user-centered XAI in different industries and AI models.
By providing a dedicated platform for researchers and practitioners, the UCEX-XAI workshop aims to contribute to the advancement of XAI systems that are not only technically sound but also meaningful and accessible to users.
📌 Paper Submission Deadline: June 2, 2025
📌 Notification of Acceptance: July 18, 2025
📌 Camera-Ready Submission: August 1, 2025
📌 Final workshop schedule published: 8 August 2024
📌 Workshop Date: October 25 or 26, 2025
All deadlines are at the end of the day specified, anywhere on Earth (UTC-12).