[Hcitaly] 1st RAISE Workshop: Resilient AI Framework for Secure and Ethical Machine Learning Systems

1st RAISE Workshop: Resilient AI Framework for Secure and Ethical Machine Learning Systems co-located with the 16th Biannual Conference of the Italian SIGCHI Chapter - "Technologies and Methodologies of Human-Computer Interaction in the Third Millenium” (CHITALY 2025) - https://chitaly2025.di.unisa.it/ Salerno, Italy 6-10 October, 2025 Workshop website: https://raise-ws.github.io/chitaly2025 Call for papers The emerging field of Resilient AI recognizes that technological systems cannot be viewed as isolated, abstract entities. Instead, they need to be understood in their broader socio-technical context, where AI and algorithms function as complex systems shaped by collective and distributive practices. In order to ensure that AI systems are robust, fair and in line with human values, it is crucial to develop methods that enhance their resilience across multiple dimensions, such as reliability, accountability, fairness and ethical integrity. A number of challenges need to be addressed to guarantee resilient AI. These include the need to balance human control and automation to ensure that AI systems remain intuitive, transparent and trustworthy, while adapting to user preferences and needs. Robustness and reliability are critical concerns, requiring mechanisms to detect and mitigate functional and non-functional defects, such as errors, biases and vulnerabilities, to maintain essential properties such as safety and accuracy. Moreover, fairness must be addressed within ML pipelines, embedding techniques that promote inclusivity and unbiased decision-making to prevent discrimination and reinforce social responsibility. Another pressing issue is the verification, validation, and testing of AI models; while advances in AI and search-based software engineering have enabled automated test case generation and validation, empirical studies are still needed to investigate AI defects and their impact on system performance. Furthermore, ensuring high-quality data is a fundamental challenge since effective data profiling, imputation, and anomaly detection must be developed to handle missing values, outliers, and inconsistences, improving overall AI model performance and reliability. The RAISE workshop brings together researchers and practitioners to address challenges including data quality, privacy protection, human-centered AI design, and the validation of machine learning systems. RAISE promotes interdisciplinary discussion to explore innovative techniques and empirical studies for building AI systems that are secure, interpretable, and inclusive. Topics Topics include, but are not limited to: • Data profiling approaches for AI systems • Data quality approaches for AI systems • Tools and metrics to summarize data relationships • Privacy concerns of AI systems • Privacy-preserving machine learning • Ethics and trustworthy AI • Reliability and fairness of AI-based systems • Resilient AI models • Empirical evaluation of resilient AI solutions in real-world cases • Verification and validation methods for AI-based systems • Explainable AI • AI solutions to critical scenarios • Domain-specific AI solutions • AI-based systems for human interaction • AI-based systems for requirements analysis • AI-based adaptive systems • User-centered design solutions for AI-based systems • AR/VR user-centered interfaces for AI-based systems • Adaptation of AR/VR in learning contexts • Neuro-Symbolic AI Important Dates First submission round - Submission deadline: June 13th 2025 - Notification: July 11th 2025 - Camera-ready: July 31st 2025 Second submission round - Submission deadline: September 5th 2025 - Notification: September 12th 2025 - Camera-ready: September 22nd 2025 Submissions and Evaluation Criteria We invite submissions of original research and position papers. Submissions should be provided via EasyChair (https://easychair.org/conferences/?conf=raise2025). Accepted papers will be presented at the workshop and included in the CEUR Workshop Proceedings (http://ceur-ws.org/). The minimum length of a CEUR paper should be 5 pages, with approximately 2500 characters (= 380-400 words) per page. All papers will be reviewed by the organizers and by the program committee based on relevance and significance in order to provide constructive comments to the submitters. Reviewing will be single blind (i.e. author names and affiliations should be listed). If accepted, at least one of the authors must register and attend the workshop. Workshop organizers - Bernardo Breve – University of Naples “Federico II”, Italy (bernardo.breve(a)unina.it <mailto:bernardo.breve@unina.it>) - Loredana Caruccio – University of Salerno, Italy (lcaruccio(a)unisa.it <mailto:lcaruccio@unisa.it>) - Giuseppe Polese – University of Salerno, Italy (gpolese(a)unisa.it <mailto:gpolese@unisa.it>) -- Bernardo Breve, PhD Assistant Professor Department of Electrical Engineering and Information Technology University of Naples Federico II via Claudio, 21 80125 Napoli, NA - ITALY e-mail: bernardo.breve(a)unina.it