Riccardo Rosati

Marianna Maranghi, Aris Anagnostopoulos, Irene Cannistraci et al. · eth-zurich

The Associazione Medici Diabetologi (AMD) collects and manages one of the largest worldwide-available collections of diabetic patient records, also known as the AMD database. This paper presents the initial results of an ongoing project whose focus is the application of Artificial Intelligence and Machine Learning techniques for conceptualizing, cleaning, and analyzing such an important and valuable dataset, with the goal of providing predictive insights to better support diabetologists in their diagnostic and therapeutic choices.

Francisco Bérchez-Moreno, Riccardo Rosati, Maria Chiara Fiorentino et al.

Background and objective. Conventional Deep Learning (DL) approaches for Knee Osteoarthritis (KOA) grading rely on one-hot labels, which fail to capture both the ordinal uncertainty of Kellgren--Lawrence (KL) and Calcium Pyrophosphate Deposition Disease (CPPD) severity scores and the asymmetric relationship between the two scales observed in clinical practice. Methods. We retrospectively collected 2172 knee X-ray images, including 968 radiographs jointly annotated for KL and CPPD severity. An ordinal DL framework based on soft-labelling was developed for both tasks, replacing one-hot targets with unimodal probability distributions centred on the annotated grade. Four formulations were investigated: binomial, beta, triangular, and exponential. Results. All soft-labelling strategies consistently outperformed the nominal baseline. For CPPD grading, the triangular formulation achieved the highest Quadratic Weighted Kappa (QWK) and the lowest Mean Absolute Error (MAE) (QWK = 0.796; MAE = 0.438), while the beta formulation yielded the most balanced class-wise performance considering Average MAE (AMAE) and Maximum MAE (MMAE) across classes (AMAE = 0.458; MMAE = 0.573). For KL grading, the beta-based approach provided the best overall performance, achieving the highest QWK together with the lowest MAE and class-wise errors (QWK = 0.777; MAE = 0.529; AMAE = 0.523; MMAE = 0.775). Statistical analysis demonstrated significant improvements over conventional one-hot supervision (p < 0.001).

Riccardo Rosati, Edoardo Colucci, Massimiliano Bolognini et al.

The rapid adoption of Large Language Models (LLMs) in interactive systems has enabled the creation of dynamic, open-ended Role-Playing Agents (RPAs). However, evaluating these agents remains a significant challenge, as standard NLP metrics fail to capture the nuances of role adherence, logical consistency, and long-term narrative stability. This paper introduces RPA-Check, a multi-stage automated evaluation framework designed to objectively assess the performance of LLM-based RPAs in complex, constraints-heavy environments. Our methodology is based on a four-step pipeline: (1) Dimension Definition, establishing high-level qualitative behavioral criteria; (2) Augmentation, where these requirements are expanded into granular boolean checklist indicators; (3) Semantic Filtering, to ensure indicator objectivity, no redundancy and agent isolation; and (4) LLM-as-a-Judge Evaluation, which employs chain-of-thought verification to score agent fidelity. We validate this framework by applying it to LLM Court, a serious game for forensic training involving several quantized local models. Experimental results across five distinct legal scenarios demonstrate the framework's ability to identify subtle trade-offs between model size, reasoning depth, and operational stability. Notably, the findings reveal an inverse relationship between parametric scale and procedural consistency, showing that smaller, adequately instruction-tuned models (8-9B) can outperform larger architectures prone to user-alignment bias or sycophancy. RPA-Check thus provides a standardized and reproducible metric for future research in generative agent evaluation within specialized domains.

Edoardo Conti, Riccardo Rosati, Lorenzo Federici et al.

Purpose: This study provides the first comprehensive evaluation of foundation models in fetal ultrasound (US) imaging under low inter-class variability conditions. While recent vision foundation models such as DINOv3 have shown remarkable transferability across medical domains, their ability to discriminate anatomically similar structures has not been systematically investigated. We address this gap by focusing on fetal brain standard planes--transthalamic (TT), transventricular (TV), and transcerebellar (TC)--which exhibit highly overlapping anatomical features and pose a critical challenge for reliable biometric assessment. Methods: To ensure a fair and reproducible evaluation, all publicly available fetal ultrasound datasets were curated and aggregated into a unified multicenter benchmark, FetalUS-188K, comprising more than 188,000 annotated images from heterogeneous acquisition settings. DINOv3 was pretrained in a self-supervised manner to learn ultrasound-aware representations. The learned features were then evaluated through standardized adaptation protocols, including linear probing with frozen backbone and full fine-tuning, under two initialization schemes: (i) pretraining on FetalUS-188K and (ii) initialization from natural-image DINOv3 weights. Results: Models pretrained on fetal ultrasound data consistently outperformed those initialized on natural images, with weighted F1-score improvements of up to 20 percent. Domain-adaptive pretraining enabled the network to preserve subtle echogenic and structural cues crucial for distinguishing intermediate planes such as TV. Conclusion: Results demonstrate that generic foundation models fail to generalize under low inter-class variability, whereas domain-specific pretraining is essential to achieve robust and clinically reliable representations in fetal brain ultrasound imaging.

Gianluca Cima, Domenico Lembo, Lorenzo Marconi et al.

In this paper, we propose the use of epistemic dependencies to express data protection policies in Controlled Query Evaluation (CQE), which is a form of confidentiality-preserving query answering over ontologies and databases. The resulting policy language goes significantly beyond those proposed in the literature on CQE so far, allowing for very rich and practically interesting forms of data protection rules. We show the expressive abilities of our framework and study the data complexity of CQE for (unions of) conjunctive queries when ontologies are specified in the Description Logic DL-Lite_R. Interestingly, while we show that the problem is in general intractable, we prove tractability for the case of acyclic epistemic dependencies by providing a suitable query rewriting algorithm. The latter result paves the way towards the implementation and practical application of this new approach to CQE.

Oleksandr Kuznetsov, Emanuele Frontoni, Luca Romeo et al.

In the rapidly evolving landscape of digital security, biometric authentication systems, particularly facial recognition, have emerged as integral components of various security protocols. However, the reliability of these systems is compromised by sophisticated spoofing attacks, where imposters gain unauthorized access by falsifying biometric traits. Current literature reveals a concerning gap: existing liveness detection methodologies - designed to counteract these breaches - fall short against advanced spoofing tactics employing deepfakes and other artificial intelligence-driven manipulations. This study introduces a robust solution through novel deep learning models addressing the deficiencies in contemporary anti-spoofing techniques. By innovatively integrating texture analysis and reflective properties associated with genuine human traits, our models distinguish authentic presence from replicas with remarkable precision. Extensive evaluations were conducted across five diverse datasets, encompassing a wide range of attack vectors and environmental conditions. Results demonstrate substantial advancement over existing systems, with our best model (AttackNet V2.2) achieving 99.9% average accuracy when trained on combined data. Moreover, our research unveils critical insights into the behavioral patterns of impostor attacks, contributing to a more nuanced understanding of their evolving nature. The implications are profound: our models do not merely fortify the authentication processes but also instill confidence in biometric systems across various sectors reliant on secure access.

Lorenzo Marconi, Flavia Ricci, Riccardo Rosati

We investigate Controlled Query Evaluation (CQE) over ontologies, where information disclosure is regulated by epistemic dependencies (EDs), a family of logical rules recently proposed for the CQE framework. In particular, we combine EDs with the notion of optimal GA censors, i.e. maximal sets of ground atoms that are entailed by the ontology and can be safely revealed. We focus on answering Boolean unions of conjunctive queries (BUCQs) with respect to the intersection of all optimal GA censors - an approach that has been shown in other contexts to ensure strong security guarantees with favorable computational behavior. First, we characterize the security of this intersection-based approach and identify a class of EDs (namely, full EDs) for which it remains safe. Then, for a subclass of EDs and for DL-Lite_R ontologies, we show that answering BUCQs in the above CQE semantics is in AC^0 in data complexity by presenting a suitable, detailed first-order rewriting algorithm. Finally, we report on experiments conducted in two different evaluation scenarios, showing the practical feasibility of our rewriting function.

Lorenzo Marconi, Riccardo Rosati

Consistent Query Answering (CQA) is an inconsistency-tolerant approach to data access in knowledge bases and databases. The goal of CQA is to provide meaningful (consistent) answers to queries even in the presence of inconsistent information, e.g. a database whose data conflict with meta-data (typically the database integrity constraints). The semantics of CQA is based on the notion of repair, that is, a consistent version of the initial, inconsistent database that is obtained through minimal modifications. We study CQA in databases with data dependencies expressed by existential rules. More specifically, we focus on the broad class of disjunctive embedded dependencies with inequalities (DEDs), which extend both tuple-generating dependencies and equality-generated dependencies. We first focus on the case when the database predicates are closed, i.e. the database is assumed to have complete knowledge about such predicates, thus no tuple addition is possible to repair the database. In such a scenario, we provide a detailed analysis of the data complexity of CQA and associated tasks (repair checking) under different semantics (AR and IAR) and for different classes of existential rules. In particular, we consider the classes of acyclic, linear, full, sticky and guarded DEDs, and their combinations.

Víctor Manuel Vargas, Riccardo Rosati, César Hervás-Martínez et al.

Predictive Maintenance (PdM) methods aim to facilitate the scheduling of maintenance work before equipment failure. In this context, detecting early faults in automated teller machines (ATMs) has become increasingly important since these machines are susceptible to various types of unpredictable failures. ATMs track execution status by generating massive event-log data that collect system messages unrelated to the failure event. Predicting machine failure based on event logs poses additional challenges, mainly in extracting features that might represent sequences of events indicating impending failures. Accordingly, feature learning approaches are currently being used in PdM, where informative features are learned automatically from minimally processed sensor data. However, a gap remains to be seen on how these approaches can be exploited for deriving relevant features from event-log-based data. To fill this gap, we present a predictive model based on a convolutional kernel (MiniROCKET and HYDRA) to extract features from the original event-log data and a linear classifier to classify the sample based on the learned features. The proposed methodology is applied to a significant real-world collected dataset. Experimental results demonstrated how one of the proposed convolutional kernels (i.e. HYDRA) exhibited the best classification performance (accuracy of 0.759 and AUC of 0.693). In addition, statistical analysis revealed that the HYDRA and MiniROCKET models significantly overcome one of the established state-of-the-art approaches in time series classification (InceptionTime), and three non-temporal ML methods from the literature. The predictive model was integrated into a container-based decision support system to support operators in the timely maintenance of ATMs.

Piero Bonatti, Gianluca Cima, Domenico Lembo et al.

Controlled Query Evaluation (CQE) has been recently studied in the context of Semantic Web ontologies. The goal of CQE is concealing some query answers so as to prevent external users from inferring confidential information. In general, there exist multiple, mutually incomparable ways of concealing answers, and previous CQE approaches choose in advance which answers are visible and which are not. In this paper, instead, we study a dynamic CQE method, namely, we propose to alter the answer to the current query based on the evaluation of previous ones. We aim at a system that, besides being able to protect confidential data, is maximally cooperative, which intuitively means that it answers affirmatively to as many queries as possible; it achieves this goal by delaying answer modifications as much as possible. We also show that the behavior we get cannot be intensionally simulated through a static approach, independent of query history. Interestingly, for OWL 2 QL ontologies and policy expressed through denials, query evaluation under our semantics is first-order rewritable, and thus in AC0 in data complexity. This paves the way for the development of practical algorithms, which we also preliminarily discuss in the paper.

Gianluca Cima, Domenico Lembo, Riccardo Rosati et al.

We study privacy-preserving query answering in Description Logics (DLs). Specifically, we consider the approach of controlled query evaluation (CQE) based on the notion of instance indistinguishability. We derive data complexity results for query answering over DL-Lite$_{\mathcal{R}}$ ontologies, through a comparison with an alternative, existing confidentiality-preserving approach to CQE. Finally, we identify a semantically well-founded notion of approximated query answering for CQE, and prove that, for DL-Lite$_{\mathcal{R}}$ ontologies, this form of CQE is tractable with respect to data complexity and is first-order rewritable, i.e., it is always reducible to the evaluation of a first-order query over the data instance.