arXiv subjects

Riccardo Rovatti

Publications and source records attributed to Riccardo Rovatti.

At least 19 recordsLinked to original sources

arXiv2026

SVD-Based Typicality Maps for Out-of-Distribution Detection in Vision Transformers

We present a method for analyzing the internal representations of Vision Transformers (ViTs) exploiting the geometry of their learned parameters. Each affine layer's weight matrix is factored via Singular Value Decomposition (SVD), and activations are projected onto the leading right singular vectors to obtain compact, layer-intrinsic representations. A class-conditional density model is then fitted at each layer, producing per-class \emph{typicality scores} that are stacked across depth into \emph{typicality maps}: two-dimensional summaries of how class-specific evidence evolves through the network. From these maps, we derive two post-hoc scores for Out-Of-Distribution (OOD) detection: a \emph{Prototype Alignment Score} (PAS), measuring agreement with class reference prototype patterns, and a \emph{Multi-Layer Soft Voting} (MLSV) score, capturing cross-layer consensus without stored prototypes. On ViT-B/16 fine-tuned on CIFAR-100, the proposed scores achieve competitive detection performance without retraining or OOD exposure.

cs.CV↗

arXiv2026

A Convolutional Layer Activation Dimensionality Reduction for Out-of-Distribution and Adversarial Attack Detection Methods

Despite the success of convolutional neural networks in image classification tasks and their general application in multi-modal models, their susceptibility to out-of-distribution and adversarial attack samples raises concerns regarding trustworthiness and safety. Among the approaches to tackle such issues, detection methods that analyze the model's intermediate activations to estimate a confidence score are a promising family that evaluates the decision process, relying on a dimensionality reduction step to enable efficient downstream processing of the high-dimensional activations. However, when considering convolutional layers, the dimensionality reduction methods in the literature either lack a mechanism to control the compression/information-loss trade-off or yield large representations. In this paper, we carefully analyze two state-of-the-art detection methods and their dimensionality reductions for convolutional layers and develop a novel reduction method with a controllable high-compression level. We extend these two state-of-the-art detection methods, enabling the usage of any dimensionality reduction, and evaluate their performance on out-of-distribution and adversarial attack detection. Results show that the detection methods with the proposed dimensionality reduction consistently perform better than, or comparable to, the strongest alternative. Furthermore, the proposed method is shown to reduce computation and memory footprints, given that it has the highest compression among the compared methods.

cs.CV↗

arXiv2026

What out-of-the-box LLMs can(t) do in law? A Turing test in Italian exams for lawyers, judges and notaries

The article reports on a blind Turing Test experiment, assessing the performance of out-of-the-box leading LLMs on three Italian legal professional exams: the Bar, Judges and Notary exams. Leading LLMs were asked to generate full written exam papers, which were made indistinguishable from human submissions and anonymously evaluated by expert examiners, using the same criteria applied in real examinations. Results reveal marked differences across both models and tasks. While some LLMs match or exceed top human performance in adversarial legal argumentation and doctrinal analysis, all models fail in the notary exam, which requires goal-directed legal planning under strict formal and substantive constraints. Beyond ranking models, the study identifies task-specific strengths, limitations and recurring legal failure patterns. Although limited to out-of-the-box systems, the findings provide qualitative evidence on the current scope and boundaries of the legal competence of LLMs across distinct professional tasks.

cs.CY↗

arXiv2026

Vanishing Contributions: A Unified Framework for Smooth and Iterative Model Compression

The increasing scale of Deep Neural Networks (DNNs) introduces the need for compression techniques such as pruning, quantization, and low-rank decomposition. While these methods are very effective at reducing memory, computation, and energy consumption, they may introduce severe accuracy degradation, which is often mitigated by using iterative, gradual compression. However, different compression techniques require distinct iterative approaches, and some result in unstable, discontinuous model fine-tuning. We introduce Vanishing Contributions (VCON), a unified framework for the smooth, iterative transition of DNNs into a compressed form. Rather than replacing the original network directly with its compressed version, VCON executes both in parallel during fine-tuning. The contribution of the original (uncompressed) model is progressively reduced, while that of the compressed model is gradually increased. This affine combination allows the network to slowly adapt, improving stability and mitigating accuracy degradation. We evaluate VCON on computer vision and natural language processing benchmarks, using multiple compression strategies. In most settings, our framework improves accuracy over post-shot and iterative baselines. Typical gains exceed 1%, while some configuration exhibits improvements above 15%. VCON is thus compatible with existing compression techniques and consistently improves performance across diverse tasks.

cs.LG↗

arXiv2026

On-board Telemetry Monitoring in Autonomous Satellites: Challenges and Opportunities

The increasing autonomy of spacecraft demands fault-detection systems that are both reliable and explainable. This work addresses eXplainable Artificial Intelligence for onboard Fault Detection, Isolation and Recovery within the Attitude and Orbit Control Subsystem by introducing a framework that enhances interpretability in neural anomaly detectors. We propose a method to derive low-dimensional, semantically annotated encodings from intermediate neural activations, called peepholes. Applied to a convolutional autoencoder, the framework produces interpretable indicators that enable the identification and localization of anomalies in reaction-wheel telemetry. Peepholes analysis further reveals bias detection and supports fault localization. The proposed framework enables the semantic characterization of detected anomalies while requiring only a marginal increase in computational resources, thus supporting its feasibility for on-board deployment.

cs.AI↗

arXiv2025

Multi-Layer Confidence Scoring for Detection of Out-of-Distribution Samples, Adversarial Attacks, and In-Distribution Misclassifications

The recent explosive growth in Deep Neural Networks applications raises concerns about the black-box usage of such models, with limited trasparency and trustworthiness in high-stakes domains, which have been crystallized as regulatory requirements such as the European Union Artificial Intelligence Act. While models with embedded confidence metrics have been proposed, such approaches cannot be applied to already existing models without retraining, limiting their broad application. On the other hand, post-hoc methods, which evaluate pre-trained models, focus on solving problems related to improving the confidence in the model's predictions, and detecting Out-Of-Distribution or Adversarial Attacks samples as independent applications. To tackle the limited applicability of already existing methods, we introduce Multi-Layer Analysis for Confidence Scoring (MACS), a unified post-hoc framework that analyzes intermediate activations to produce classification-maps. From the classification-maps, we derive a score applicable for confidence estimation, detecting distributional shifts and adversarial attacks, unifying the three problems in a common framework, and achieving performances that surpass the state-of-the-art approaches in our experiments with the VGG16 and ViTb16 models with a fraction of their computational overhead.

cs.LG↗

arXiv2025

RDD: Pareto Analysis of the Rate-Distortion-Distinguishability Trade-off

Extensive monitoring systems generate data that is usually compressed for network transmission. This compressed data might then be processed in the cloud for tasks such as anomaly detection. However, compression can potentially impair the detector's ability to distinguish between regular and irregular patterns due to information loss. Here we extend the information-theoretic framework introduced in [1] to simultaneously address the trade-off between the three features on which the effectiveness of the system depends: the effectiveness of compression, the amount of distortion it introduces, and the distinguishability between compressed normal signals and compressed anomalous signals. We leverage a Gaussian assumption to draw curves showing how moving on a Pareto surface helps administer such a trade-off better than simply relying on optimal rate-distortion compression and hoping that compressed signals can be distinguished from each other.

eess.SP↗

arXiv2025

Goal-Oriented Joint Source-Channel Coding: Distortion-Classification-Power Trade-off

Joint source-channel coding is a compelling paradigm when low-latency and low-complexity communication is required. This work proposes a theoretical framework that integrates classification and anomaly detection within the conventional signal reconstruction objective. Assuming a Gaussian scalar source and constraining the encoder to piecewise linear mappings, we derive tractable design rules and explicitly characterize the trade-offs between distortion, classification error, and transmission power.

eess.SP↗

arXiv2023

Legal Summarisation through LLMs: The PRODIGIT Project

We present some initial results of a large-scale Italian project called PRODIGIT which aims to support tax judges and lawyers through digital technology, focusing on AI. We have focused on generation of summaries of judicial decisions and on the extraction of related information, such as the identification of legal issues and decision-making criteria, and the specification of keywords. To this end, we have deployed and evaluated different tools and approaches to extractive and abstractive summarisation. We have applied LLMs, and particularly on GPT4, which has enabled us to obtain results that proved satisfactory, according to an evaluation by expert tax judges and lawyers. On this basis, a prototype application is being built which will be made publicly available.

cs.CL↗

arXiv2022

Anomaly Detection based on Compressed Data: an Information Theoretic Characterization

We analyze the effect of lossy compression in the processing of sensor signals that must be used to detect anomalous events in the system under observation. The intuitive relationship between the quality loss at higher compression and the possibility of telling anomalous behaviours from normal ones is formalized in terms of information-theoretic quantities. Some analytic derivations are made within the Gaussian framework and possibly in the asymptotic regime for what concerns the stretch of signals considered. Analytical conclusions are matched with the performance of practical detectors in a toy case allowing the assessment of different compression/detector configurations.

cs.IT↗

arXiv2015

On the security of a class of diffusion mechanisms for image encryption

The need for fast and strong image cryptosystems motivates researchers to develop new techniques to apply traditional cryptographic primitives in order to exploit the intrinsic features of digital images. One of the most popular and mature technique is the use of complex ynamic phenomena, including chaotic orbits and quantum walks, to generate the required key stream. In this paper, under the assumption of plaintext attacks we investigate the security of a classic diffusion mechanism (and of its variants) used as the core cryptographic rimitive in some image cryptosystems based on the aforementioned complex dynamic phenomena. We have theoretically found that regardless of the key schedule process, the data complexity for recovering each element of the equivalent secret key from these diffusion mechanisms is only O(1). The proposed analysis is validated by means of numerical examples. Some additional cryptographic applications of our work are also discussed.

cs.CR↗

arXiv2015

Interleaving techniques for high-throughput chaotic noise generation in CMOS

An interleaving technique is proposed to enhance the throughput of current-mode CMOS discrete time chaotic sources based on the iteration of unidimensional maps. A discussion of the reasons and the advantages offered by the approach is provided, together with analytical results about the conservation of some major statistical features. As an example, application to an FM-DCSK communication system is proposed. To conclude, a sample circuit capable of 20 Msample/s is presented.

nlin.CD↗

arXiv2015

A Tailed Tent Map Chaotic Circuit Exploiting S$^2$I Memory Elements

In the implementation of discrete time chaotic systems, designers have mostly focused on the choice and synthesis of suitable nonlinear blocks, while correct and fast operation cannot neglect analog memory elements. Herein, a realization exploiting the Tailed Tent Map coupled with S$^2$I sample and hold stages is proposed. Interfacing problems between the memory and the processing sub-circuits are tackled and the possibility to obtain a fivefold speed improvement with respect to previously reported results is demonstrated by means of SPICE and other computer aided simulations.

nlin.CD↗

arXiv2015

On Known-Plaintext Attacks to a Compressed Sensing-based Encryption: A Quantitative Analysis

Despite the linearity of its encoding, compressed sensing may be used to provide a limited form of data protection when random encoding matrices are used to produce sets of low-dimensional measurements (ciphertexts). In this paper we quantify by theoretical means the resistance of the least complex form of this kind of encoding against known-plaintext attacks. For both standard compressed sensing with antipodal random matrices and recent multiclass encryption schemes based on it, we show how the number of candidate encoding matrices that match a typical plaintext-ciphertext pair is so large that the search for the true encoding matrix inconclusive. Such results on the practical ineffectiveness of known-plaintext attacks underlie the fact that even closely-related signal recovery under encoding matrix uncertainty is doomed to fail. Practical attacks are then exemplified by applying compressed sensing with antipodal random matrices as a multiclass encryption scheme to signals such as images and electrocardiographic tracks, showing that the extracted information on the true encoding matrix from a plaintext-ciphertext pair leads to no significant signal recovery quality increase. This theoretical and empirical evidence clarifies that, although not perfectly secure, both standard compressed sensing and multiclass encryption schemes feature a noteworthy level of security against known-plaintext attacks, therefore increasing its appeal as a negligible-cost encryption method for resource-limited sensing applications.

cs.IT↗

arXiv2015

Low-complexity Multiclass Encryption by Compressed Sensing

The idea that compressed sensing may be used to encrypt information from unauthorised receivers has already been envisioned, but never explored in depth since its security may seem compromised by the linearity of its encoding process. In this paper we apply this simple encoding to define a general private-key encryption scheme in which a transmitter distributes the same encoded measurements to receivers of different classes, which are provided partially corrupted encoding matrices and are thus allowed to decode the acquired signal at provably different levels of recovery quality. The security properties of this scheme are thoroughly analysed: firstly, the properties of our multiclass encryption are theoretically investigated by deriving performance bounds on the recovery quality attained by lower-class receivers with respect to high-class ones. Then we perform a statistical analysis of the measurements to show that, although not perfectly secure, compressed sensing grants some level of security that comes at almost-zero cost and thus may benefit resource-limited applications. In addition to this we report some exemplary applications of multiclass encryption by compressed sensing of speech signals, electrocardiographic tracks and images, in which quality degradation is quantified as the impossibility of some feature extraction algorithms to obtain sensitive information from suitably degraded signal recoveries.

cs.IT↗

arXiv2013

Sample and Hold Errors in the Implementation of Chaotic Maps

Though considerable effort has recently been devoted to hardware realization of chaotic maps, the analysis generally neglects the influence of implementation inaccuracies. Here we investigate the consequences of S/H errors on Bernoulli shift, tent map and tailed tent map systems: an error model is proposed and implementations are characterized under its assumptions.

nlin.CD↗

arXiv2013

From Chirps to Random-FM Excitations in Pulse Compression Ultrasound Systems

Pulse compression is often practiced in ultrasound Non Destructive Testing (NDT) systems using chirps. However, chirps are inadequate for setups where multiple probes need to operate concurrently in Multiple Input Multiple Output (MIMO) arrangements. Conversely, many coded excitation systems designed for MIMO miss some chirp advantages (constant envelope excitation, easiness of bandwidth control, etc.) and may not be easily implemented on hardware originally conceived for chirp excitations. Here, we propose a system based on random-FM excitations, capable of enabling MIMO with minimal changes with respect to a chirp-based setup. Following recent results, we show that random-FM excitations retain many advantages of chirps and provide the ability to frequency-shape the excitations matching the transducers features.

physics.ins-det↗

arXiv2012

Rakeness in the design of Analog-to-Information Conversion of Sparse and Localized Signals

Design of Random Modulation Pre-Integration systems based on the restricted-isometry property may be suboptimal when the energy of the signals to be acquired is not evenly distributed, i.e. when they are both sparse and localized. To counter this, we introduce an additional design criterion, that we call rakeness, accounting for the amount of energy that the measurements capture from the signal to be acquired. Hence, for localized signals a proper system tuning increases the rakeness as well as the average SNR of the samples used in its reconstruction. Yet, maximizing average SNR may go against the need of capturing all the components that are potentially non-zero in a sparse signal, i.e., against the restricted isometry requirement ensuring reconstructability. What we propose is to administer the trade-off between rakeness and restricted isometry in a statistical way by laying down an optimization problem. The solution of such an optimization problem is the statistic of the process generating the random waveforms onto which the signal is projected to obtain the measurements. The formal definition of such a problems is given as well as its solution for signals that are either localized in frequency or in more generic domain. Sample applications, to ECG signals and small images of printed letters and numbers, show that rakeness-based design leads to non-negligible improvements in both cases.

cs.IT↗