arXiv subjects

Tiziano Leidi

Publications and source records attributed to Tiziano Leidi.

11 recordsLinked to original sources

arXiv2026

TTM-Bench: A Framework for Text-to-Music System Performance Benchmarking

Text-to-music (TTM) systems are increasingly used to generate musical audio from natural-language descriptions. Robust evaluation is therefore essential, yet reliable performance comparison remains challenging. This difficulty stems from differences in system architecture, supported conditioning information, and access mode, as well as heterogeneous and fragmented metrics that cannot be applied uniformly across systems. To address these challenges, we introduce TTM-Bench, a framework that defines a common protocol for systematic, reproducible performance benchmarking of contemporary TTM systems. It evaluates performance along two dimensions: musical-content alignment, quantified by interpretable semantic, genre, and musical-descriptor agreement scores against a common musical specification and summarized by an aggregate score; and computational efficiency, characterized by generation latency and real-time factor, alongside resource use for local models and cost for hosted services. We demonstrate the framework through a preliminary comparative case study, illustrating the complementary evidence captured by these dimensions. The results show that higher musical-content alignment does not systematically coincide with lower computational demands, highlighting the importance of assessing TTM performance through distinct, interpretable measures rather than a reductive overall indicator.

cs.SD↗

arXiv2026

Explanation-Based Runtime Verification for Trustworthy ML-driven Optical Networks

Machine learning (ML) models are increasingly integrated into optical network automation frameworks to support tasks such as failure management, performance monitoring and resource allocation. In these environments, ML-driven predictions may be directly coupled with control-plane actions where incorrect decisions can immediately impact service quality, resource efficiency, and network stability. As automation levels increase, ensuring the reliability of individual decisions at deployment time becomes a critical requirement. Explainable artificial intelligence (XAI) techniques have emerged to improve transparency by highlighting the factors influencing ML predictions. In addition to identifying influential features, they provide insights into the underlying reasoning process of the model, revealing how different input variables contribute to the final outcome and how feature interactions shape the decision boundary. In this work, we introduce explanation-based runtime verification, an approach that exploits model explanations to assess the soundness of individual ML decisions before they are executed in the network control loop. The proposed approach evaluates explanation coherence and physics grounding consistency at runtime, enabling the system to defer or reject decisions flagged as uncertain. We demonstrate the effectiveness of our approach on a representative use case of lightpath quality of transmission classification. Experimental results show that explanation-based verification can intercept a significant fraction of erroneous decisions while preserving high automation rate.

cs.LG↗

arXiv2025

MuMeNet: A Network Simulator for Musical Metaverse Communications

The Metaverse, a shared and spatially organized digital continuum, is transforming various industries, with music emerging as a leading use case. Live concerts, collaborative composition, and interactive experiences are driving the Musical Metaverse (MM), but the requirements of the underlying network and service infrastructures hinder its growth. These challenges underscore the need for a novel modeling and simulation paradigm tailored to the unique characteristics of MM sessions, along with specialized service provisioning strategies capable of capturing their interactive, heterogeneous, and multicast-oriented nature. To this end, we make a first attempt to formally model and analyze the problem of service provisioning for MM sessions in 5G/6G networks. We first formalize service and network graph models for the MM, using "live audience interaction in a virtual concert" as a reference scenario. We then present MuMeNet, a novel discrete-event network simulator specifically tailored to the requirements and the traffic dynamics of the MM. We showcase the effectiveness of MuMeNet by running a linear programming based orchestration policy on the reference scenario and providing performance analysis under realistic MM workloads.

cs.NI↗

arXiv2025

BinomialHash: A Constant Time, Minimal Memory Consistent Hash Algorithm

Consistent hashing is a technique for distributing data across a network of nodes in a way that minimizes reorganization when nodes join or leave the network. It is extensively applied in modern distributed systems as a fundamental mechanism for routing and data placement. Similarly, distributed storage systems rely on consistent hashing for scalable and fault-tolerant data partitioning. This paper introduces BinomialHash, a consistent hashing algorithm that executes in constant time and requires minimal memory. We provide a detailed explanation of the algorithm, present a pseudo-code implementation, and formally establish its strong theoretical guarantees. Finally, we compare its performance against state-of-the-art constant-time consistent hashing algorithms, demonstrating that our solution is both highly competitive and effective, while also validating the theoretical boundaries.

cs.DC↗

arXiv2024

Liquid Neural Network-based Adaptive Learning vs. Incremental Learning for Link Load Prediction amid Concept Drift due to Network Failures

Adapting to concept drift is a challenging task in machine learning, which is usually tackled using incremental learning techniques that periodically re-fit a learning model leveraging newly available data. A primary limitation of these techniques is their reliance on substantial amounts of data for retraining. The necessity of acquiring fresh data introduces temporal delays prior to retraining, potentially rendering the models inaccurate if a sudden concept drift occurs in-between two consecutive retrainings. In communication networks, such issue emerges when performing traffic forecasting following a~failure event: post-failure re-routing may induce a drastic shift in distribution and pattern of traffic data, thus requiring a timely model adaptation. In this work, we address this challenge for the problem of traffic forecasting and propose an approach that exploits adaptive learning algorithms, namely, liquid neural networks, which are capable of self-adaptation to abrupt changes in data patterns without requiring any retraining. Through extensive simulations of failure scenarios, we compare the predictive performance of our proposed approach to that of a reference method based on incremental learning. Experimental results show that our proposed approach outperforms incremental learning-based methods in situations where the shifts in traffic patterns are drastic.

cs.NI↗

arXiv2024

MementoHash: A Stateful, Minimal Memory, Best Performing Consistent Hash Algorithm

Consistent hashing is used in distributed systems and networking applications to spread data evenly and efficiently across a cluster of nodes. In this paper, we present MementoHash, a novel consistent hashing algorithm that eliminates known limitations of state-of-the-art algorithms while keeping optimal performance and minimal memory usage. We describe the algorithm in detail, provide a pseudo-code implementation, and formally establish its solid theoretical guarantees. To measure the efficacy of MementoHash, we compare its performance, in terms of memory usage and lookup time, to that of state-of-the-art algorithms, namely, AnchorHash, DxHash, and JumpHash. Unlike JumpHash, MementoHash can handle random failures. Moreover, MementoHash does not require fixing the overall capacity of the cluster (as AnchorHash and DxHash do), allowing it to scale indefinitely. The number of removed nodes affects the performance of all the considered algorithms. Therefore, we conduct experiments considering three different scenarios: stable (no removed nodes), one-shot removals (90% of the nodes removed at once), and incremental removals. We report experimental results that averaged a varying number of nodes from ten to one million. Results indicate that our algorithm shows optimal lookup performance and minimal memory usage in its best-case scenario. It behaves better than AnchorHash and DxHash in its average-case scenario and at least as well as those two algorithms in its worst-case scenario. However, the worst-case scenario for MementoHash occurs when more than 70% of the nodes fail, which describes a unlikely scenario. Therefore, MementoHash shows the best performance during the regular life cycle of a cluster.

cs.DC↗

arXiv2021

The Virtual Emotion Loop: Towards Emotion-Driven Services via Virtual Reality

The importance of emotions in service and in product design is well known. However, it is still not very well understood how users' emotions can be incorporated in a product or service lifecycle. We argue that this gap is due to a lack of a methodological framework for an effective investigation of the emotional response of persons when using products and services. Indeed, the emotional response of users is generally investigated by means of methods (e.g., surveys) that are not effective for this purpose. In our view, Virtual Reality (VR) technologies represent the perfect medium to evoke and recognize users' emotional response, as well as to prototype products and services (and, for the latter, even deliver them). In this paper, we first provide our definition of emotion-driven services, and then we propose a novel methodological framework, referred to as the Virtual-Reality-Based Emotion-Elicitation-and-Recognition loop (VEE-loop), that can be exploited to realize it. Specifically, the VEE-loop consists in a continuous monitoring of users' emotions, which are then provided to service designers as an implicit users' feedback. This information is used to dynamically change the content of the VR environment, until the desired affective state is solicited. Finally, we discuss issues and opportunities of this VEE-loop, and we also present potential applications of the VEE-loop in research and in various application areas.

cs.HC↗

arXiv2016

Simulation of capillary infiltration into packing structures by the Lattice-Boltzmann method for the optimization of ceramic materials

In this work we want to simulate with the Lattice-Boltzmann method in 2D the capillary infiltration into porous structures obtained from the packing of particles. The experimental problem motivating our work is the densification of carbon preforms by reactive melt infiltration. The aim is to determine optimization principles for the manufacturing of high-performance ceramics. Simulations are performed for packings with varying structural properties. Our analysis suggests that the observed slow infiltrations can be ascribed to interface dynamics. Pinning represents the primary factor retarding fluid penetration. The mechanism responsible for this phenomenon is analyzed in detail. When surface growth is allowed, it is found that the phenomenon of pinning becomes stronger. Systems trying to reproduce typical experimental conditions are also investigated. It turns out that the standard for accurate simulations is challenging. The primary obstacle to overcome for enhanced accuracy seems to be the over-occurrence of pinning.

cond-mat.mtrl-sci↗

arXiv2015

Lattice Boltzmann simulations on the role of channel structure for reactive capillary infiltration

It is widely recognized that the structure of porous media is of relevance for a variety of mechanical and physical phenomena. The focus of the present work is on capillarity, a pore-scale process occurring at the micron scale. We attempt to characterize the influence of pore shape for capillary infiltration by means of Lattice Boltzmann simulations in 2D with reactive boundaries leading to surface growth and ultimately to pore closure. The systems under investigation consist of single channels with different simplified morphologies: namely, periodic profiles with sinusoidal, step-shaped and zig-zag walls, as well as constrictions and expansions with rectangular, convex and concave steps. This is a useful way to decompose the complexity of typical porous media into basic structures. The simulations show that the minimum radius alone fails to characterize properly the infiltration dynamics. The structure of the channels emerge as the dominant property controlling the process. A factor responsible for this behavior is identified as being the occurrence of pinning of the contact line. It turns out that the optimal configuration for the pore structure arises from the packing of large particles with round shapes. In this case, the probability to have flow paths wide and straight is higher. Faceted surfaces presenting sharp edges should be avoided because of the phenomenon of pinning near narrow-to-wide parts. This study is motivated by the infiltration of molten metals into carbon preforms. This is a manufacturing technique for ceramic components devised to advanced applications. Guidelines for experimental work are discussed.

cond-mat.soft↗

arXiv2014

Surface growth effects on reactive capillary-driven flow: Lattice Boltzmann investigation

The Washburn law has always played a critical role for ceramics. In the microscale, surface forces take over volume forces and the phenomenon of spontaneous infiltration in narrow interstices becomes of particular relevance. The Lattice Boltzmann method is applied in order to ascertain the role of surface reaction and subsequent deformation of a single capillary in 2D for the linear Washburn behavior. The proposed investigation is motivated by the problem of reactive infiltration of molten silicon into carbon preforms. This is a complex phenomenon arising from the interplay between fluid flow, the transition to wetting, surface growth and heat transfer. Furthermore, it is characterized by slow infiltration velocities in narrow interstices resulting in small Reynolds numbers that are difficult to reproduce with a single capillary. In our simulations, several geometric characteristics for the capillaries are considered, as well as different infiltration and reaction conditions. The main result of our work is that the phenomenon of pore closure can be regarded as independent of the infiltration velocity, and in turn a number of other parameters. The instrumental conclusion drawn from our simulations is that short pores with wide openings and a round-shaped morphology near the throats represent the optimal configuration for the underlying structure of the porous preform in order to achieve faster infiltration. The role of the approximations is discussed in detail and the robustness of our findings is assessed.

cond-mat.soft↗

arXiv2013

Lattice Boltzmann simulation of the surface growth effects for the infiltration of molten Si in carbon preforms

The infiltration of molten silicon into carbon preforms is a widespread technique employed in the industry in order to enhance the thermal and mechanical properties of the final ceramic products. A proper understanding of this phenomenon is quite challenging since it stems from the reciprocal action and reaction between fluid flow, the transition to wetting, mass transport, precipitation, surface growth as well as heat transfer. As a result, the exhaustive modeling of such problem is an involved task. Lattice Boltzmann simulations in 2D for capillary infiltration are carried out in the isothermal regime taking into account surface reaction and subsequent surface growth. Precisely, for a single capillary in the linear Washburn regime, special attention is paid to the retardation for the infiltration process induced by the thickening of the surface behind the contact line of the invading front. Interestingly, it turns out that the process of surface growth leading to pore closure marginally depends on the infiltration velocity. We conclude that porous matrices with straight and wide pathways represent the optimal case for impregnation. Our analysis includes also a comparison between the radii characterizing the infiltration process (i.e., minimum, hydraulic, average and effective radii).

cond-mat.soft↗