8th ECML PKDD International Workshop on eXplainable Knowledge Discovery in Data Mining
Focus Theme: Validation of Explanations in Explainable AI
The 8th edition of the International Workshop on eXplainable Knowledge Discovery in Data Mining (XKDD 2026) invites submissions from researchers and practitioners working on Explainable Artificial Intelligence (XAI), Data Mining and Machine Learning.
Co-located with ECML-PKDD 2026 in Naples, Italy, this year’s workshop focuses on one of the key open challenges in XAI: the validation of explanation methods.
Although a wide range of explanation techniques has been proposed in recent years, the field still lacks widely accepted methods to evaluate whether explanations are reliable, informative and aligned with ethical and societal expectations. XKDD 2026 aims to address this gap by bringing together contributions that investigate how explanation methods can be assessed and compared across models, tasks and data distributions.
The workshop welcomes work analysing the assumptions, strengths and limitations of existing evaluation metrics, as well as studies exploring their behaviour in different settings. Particular attention is given to properties such as faithfulness, stability, usefulness and alignment with human values.
Beyond quantitative evaluation, XKDD 2026 also encourages qualitative approaches, including expert assessment, participatory evaluation and real-world case studies, especially in high-stakes domains where numerical metrics alone may not sufficiently capture the quality and trustworthiness of explanations.
Topics of Interest
Topics include, but are not limited to:
- XAI and evaluation metrics
- Qualitative and quantitative validation of XAI methods
- XAI for Trustworthy AI and Social AI
- XAI aligned with human values
- Interpretable Machine Learning and Transparent Data Mining
- XAI for ethical, fair and transparent AI systems
- XAI for fairness checking, privacy preservation and regulatory compliance
- XAI for unlearning in machine learning models
- XAI for outlier and anomaly detection
- XAI methodologies for tabular data, images, text and time series
- XAI for Federated Learning and Graph-based approaches
- XAI for visualisation and multi-level explanations
- Accountability and liability from ethical and legal perspectives
- Real-world case studies and applications
Submission Information
The workshop welcomes:
- Full research and position papers of up to 16 pages, including references
- Short abstracts of 2–4 pages presenting emerging ideas or previously published work for discussion purposes
Submissions must be written in English and formatted according to the Springer Lecture Notes in Computer Science (LNCS) guidelines, following the ECML-PKDD 2026 workshop format.
Accepted full papers will be published in the LNCSI post-proceedings. Pre-proceedings will be made available online before the workshop.
Authors of accepted papers are expected to present their work at the workshop.
Important Dates
- Paper submission deadline: 5 June 2026
- Notification of acceptance: 5 July 2026
- Workshop date: 7 September 2026
Organising Committee
Programme Co-Chairs
- Francesca Naretto, University of Pisa, Italy
- Francesco Spinnato, University of Pisa, Italy
- Przemyslaw Biecek, Warsaw University of Technology, Poland
- Andreas Theissler, Justus Liebig University Giessen, Germany
Steering Committee
- Riccardo Guidotti, University of Pisa, Italy
- Anna Monreale, University of Pisa, Italy
- Salvatore Rinzivillo, ISTI-CNR, Italy
- Przemyslaw Biecek, Warsaw University of Technology, Poland
Further Information
- Workshop website: https://xkdd2026.isti.cnr.it/
- Submission portal: https://cmt3.research.microsoft.com/ECMLPKDDWT2026/Submission/Index
- ECML-PKDD 2026: https://ecmlpkdd.org/2026/
For further information, please contact: francesca.naretto@unipi.it or francesco.spinnato@di.unipi.it