Publications

Scientific papers, conference proceedings, and research outputs from the EXCALIBUR project.

Conference Paper

FAIRTOPIA: A Multi-agent Guardianship Framework for Disrupting Unfair AI Pipelines

Vakali, A., Dimitriadis, I., & Vei, S.

Big Data Analytics and Knowledge Discovery, Lecture Notes in Computer Science, Volume 16861, pp. 37–42, 2026.
DOI: 10.1007/978-3-032-34896-8_3

View publication

Conference Paper

Not All Harms Hurt Equally: Understanding AI Safety Through Ordinal Severity

Vei, S., Giudici, P., Sermpezis, P., Vakali, A., & Bernardelli, A. E.

Navigating Complexity: Statistical Methods, Data Analysis, and Machine Learning for Actionable Insights, Studies in Classification, Data Analysis, and Knowledge Organization, pp. 276–284, 2026.
DOI: 10.1007/978-3-032-32009-4_34

View publication

Journal Article

AI Harmonics: A human-centric and harms severity-adaptive AI risk assessment framework

Vei, S., Giudici, P., Sermpezis, P., Vakali, A., & Bernardelli, A. E.

Artificial Intelligence, Volume 358, Article 104587, 2026.
DOI: 10.1016/j.artint.2026.104587

View publication

Journal Article

KFS-TUNE: Kernel-based Feature Selection for efficiency and accuracy tuning in Time Series Classification

Vei, S., Tiakas, E., & Vakali, A.

Knowledge-Based Systems, Volume 349, Article 116523, 2026.

DOI: 10.1016/j.knosys.2026.116523

View publication