arXiv subjects

Dario Rossi

Publications and source records attributed to Dario Rossi.

At least 19 recordsLinked to original sources

arXiv2026

Excitation factors and exceptional points of Kerr-de Sitter quasinormal modes

We compute the excitation factors of Kerr-de Sitter black hole quasinormal modes using two different techniques: a Heun function representation and an expansion of the solution in terms of hypergeometric functions. Exceptional points at which the complex frequencies of different overtones coincide are a generic feature of the Kerr-de Sitter quasinormal mode spectrum. Near exceptional points, the modes exhibit a "hysteresis phenomenon" that was previously found for Kerr black holes perturbed by massive fields and the absolute value of the excitation factors of the individual modes is significantly enhanced, while they acquire nearly opposite phases. We extrapolate the excitation factors to small values of the cosmological constant and find good agreement with previous calculations in the Kerr limit.

gr-qc↗

arXiv2026

Lightweight GenAI for Network Traffic Generation: Fidelity, Augmentation, and Classification

Network Traffic Classification (NTC) increasingly relies on data-driven models, yet its practical deployment is often constrained by limited labeled data, strict privacy requirements, and the cost of collecting representative traffic traces. While Network Traffic Generation (NTG) provides an effective means to mitigate data scarcity, conventional generative methods struggle to model the complex temporal dynamics of modern traffic and often incur high computational costs. In this article, we investigate lightweight Generative Artificial Intelligence (GenAI) architectures for practical NTG. Rather than generating raw packet bytes or relying on large foundation models, we synthesize compact flow-level traffic representations derived from early packet-header information, enabling transformer-based, state-space, and diffusion models with only a few million parameters. We present a modular GenAI pipeline for NTG and evaluate it along four complementary axes: (i) synthetic traffic fidelity, (ii) synthetic-only training for privacy-preserving NTC, (iii) data augmentation under low-data regimes, and (iv) computational efficiency. Experiments on two heterogeneous datasets show that lightweight transformer-based and state-space models preserve both static and temporal traffic characteristics, while providing useful synthetic data for downstream NTC. Among them, transformer-based models offer the best fidelity-efficiency trade-off, combining high-quality traffic generation with moderate computational overhead.

cs.NI↗

arXiv2026

Precision quantum simulation of magnon spectra and interactions

Quantum simulation promises to advance materials discovery by accurately simulating complex states of matter, their microscopic excitations, and macroscopic response functions. The central challenge in resolving the underlying interacting dynamics is to combine high-fidelity evolution with the sophisticated control necessary to manipulate individual quasi-particles in quantum many-body states. Here, we report on high-precision simulation of both linear and non-linear response functions in a 2D XY spin-1/2 magnet using an analog-digital superconducting processor of up to 97 qubits. By interleaving digital gates with analog evolution precisely characterized via Hamiltonian learning, we selectively excite magnons at tunable energy densities. Measuring first the linear magnon response -- a central probe in neutron-scattering experiments -- we extract temperature-dependent spectra and lifetimes. Our results reveal stark variations in magnon decay rates across the Brillouin zone, with enhancement near van Hove singularities and suppression for edge-localized modes. Next, we perform a suite of nonlinear measurements, including the study of self-scattering mechanisms, as well as pump-probe spectroscopy to directly characterize the magnon interactions. While matrix-product state simulations capture the dynamics well in either small systems or at low temperatures, their predictions become inaccurate away from these limits. This work demonstrates precise simulation of the interacting dynamics in quantum magnets, and provides key insights into quasi-particles and their microscopic scattering mechanisms.

quant-ph↗

arXiv2026

Autonomous LLM Agents & CTFs: A Second Look

Large Language Model (LLM) agents are increasingly proposed to automate offensive security tasks, with recent studies reporting near human-level success rates in Capture-the-Flag (CTF) challenges. We here revisit these results, providing a second look at these claims. We engineer different agent architectures of increasing complexity and modularity on 30 web-based CTFs challenges spanning 14 vulnerability classes. We instantiate these agents with multiple LLM backbones, and compare them with claude-code, a general-purpose agent that automatically determines its internal architecture. Our evaluation yields three main findings. First, claude-code achieves performance comparable to the engineered architectures (19/30 solved tasks), suggesting that general-purpose agents are strong baselines for offensive security tasks. Second, both our architectures and claude-code struggle in the same challenge categories, revealing persistent barriers that keep current agents below human-level capability. Third, by leveraging our manually designed architectures we can systematically measure the impact of additional components, finding that structured orchestration of specialized roles outperforms monolithic designs, improving run-to-run consistency, and reducing execution costs.

cs.CR↗

arXiv2026

Minimum mass, maximum charge and hyperbolicity in scalar Gauss-Bonnet gravity

We study the loss of hyperbolicity of perturbation equations for black hole solutions of scalar Gauss-Bonnet gravity. We consider a class of coupling functions allowing for static black hole solutions with arbitrary small masses. For masses below a minimum value, such solutions become unphysical, because the perturbation equations become elliptic; this arguably corresponds to the loss of validity of the effective field theory. We analyse the dependence of this minimum mass on the parameters of the theory, finding that with an appropriate choice of the coupling function, such mass can be chosen arbitrarily small. However, this does not correspond to larger deviations from general relativity, since observable quantities like the black hole scalar charge are bounded by above.

gr-qc↗

arXiv2026

Improving Generalization on Cybersecurity Tasks with Multi-Modal Contrastive Learning

The use of ML in cybersecurity has long been impaired by generalization issues: Models that work well in controlled scenarios fail to maintain performance in production. The root cause often lies in ML algorithms learning superficial patterns (shortcuts) rather than underlying cybersecurity concepts. We investigate contrastive multi-modal learning as a first step towards improving ML performance in cybersecurity tasks. We aim at transferring knowledge from data-rich modalities, such as text, to data-scarce modalities, such as payloads. We set up a case study on threat classification and propose a two-stage multi-modal contrastive learning framework that uses textual vulnerability descriptions to guide payload classification. First, we construct a semantically meaningful embedding space using contrastive learning on descriptions. Then, we align payloads to this space, transferring knowledge from text to payloads. We evaluate the approach on a large-scale private dataset and a synthetic benchmark built from public CVE descriptions and LLM-generated payloads. The methodology appears to reduce shortcut learning over baselines on both benchmarks. We release our synthetic benchmark and source code as open source.

cs.CR↗

arXiv2026

CyberSleuth: Autonomous Blue-Team LLM Agent for Web Attack Forensics

Post-mortem analysis of compromised systems is a key aspect of cyber forensics, today a mostly manual, slow, and error-prone task. Agentic AI, i.e., LLM-powered agents, is a promising avenue for automation. However, applying such agents to cybersecurity remains largely unexplored and difficult, as this domain demands long-term reasoning, contextual memory, and consistent evidence correlation - capabilities that current LLM agents struggle to master. In this paper, we present the first systematic study of LLM agents to automate post-mortem investigation. As a first scenario, we consider realistic attacks in which remote attackers try to abuse online services using well-known CVEs (30 controlled cases). The agent receives as input the network traces of the attack and extracts forensic evidence. We compare three AI agent architectures, six LLM backends, and assess their ability to i) identify compromised services, ii) map exploits to exact CVEs, and iii) prepare thorough reports. Our best-performing system, CyberSleuth, achieves 80% accuracy on 2025 incidents, producing complete, coherent, and practically useful reports (judged by a panel of 25 experts). We next illustrate how readily CyberSleuth adapts to face the analysis of infected machine traffic, showing that the effective AI agent design can transfer across forensic tasks. Our findings show that (i) multi-agent specialisation is key to sustained reasoning; (ii) simple orchestration outperforms nested hierarchical architectures; and (iii) the CyberSleuth design generalises across different forensic tasks.

cs.CR↗

arXiv2025

In Praise of Stubbornness: An Empirical Case for Cognitive-Dissonance Aware Continual Update of Knowledge in LLMs

Through systematic empirical investigation, we uncover a fundamental and concerning property of Large Language Models: while they can safely learn facts that don't contradict their knowledge, attempting to update facts with contradictory information triggers catastrophic corruption of unrelated knowledge. Unlike humans, who naturally resist contradictory information, these models indiscriminately accept contradictions, leading to devastating interference, destroying up to 80% of unrelated knowledge even when learning as few as 10-100 contradicting facts. To understand whether this interference could be mitigated through selective plasticity, we experiment with targeted network updates, distinguishing between previously used (stubborn) and rarely used (plastic) neurons. We uncover another asymmetry: while sparing frequently-used neurons significantly improves retention of existing knowledge for non-contradictory updates (98% vs 93% with standard updates), contradictory updates trigger catastrophic interference regardless of targeting strategy. This effect which persists across tested model scales (GPT-2 to GPT-J-6B), suggests a fundamental limitation in how neural networks handle contradictions. Finally, we demonstrate that contradictory information can be reliably detected (95%+ accuracy) using simple model features, offering a potential protective mechanism. These findings motivate new architectures that can, like humans, naturally resist contradictions rather than allowing destructive overwrites.

cs.CL↗

arXiv2025

Episodic Memories Generation and Evaluation Benchmark for Large Language Models

Episodic memory -- the ability to recall specific events grounded in time and space -- is a cornerstone of human cognition, enabling not only coherent storytelling, but also planning and decision-making. Despite their remarkable capabilities, Large Language Models (LLMs) lack a robust mechanism for episodic memory: we argue that integrating episodic memory capabilities into LLM is essential for advancing AI towards human-like cognition, increasing their potential to reason consistently and ground their output in real-world episodic events, hence avoiding confabulations. To address this challenge, we introduce a comprehensive framework to model and evaluate LLM episodic memory capabilities. Drawing inspiration from cognitive science, we develop a structured approach to represent episodic events, encapsulating temporal and spatial contexts, involved entities, and detailed descriptions. We synthesize a unique episodic memory benchmark, free from contamination, and release open source code and datasets to assess LLM performance across various recall and episodic reasoning tasks. Our evaluation of state-of-the-art models, including GPT-4 and Claude variants, Llama 3.1, and o1-mini, reveals that even the most advanced LLMs struggle with episodic memory tasks, particularly when dealing with multiple related events or complex spatio-temporal relationships -- even in contexts as short as 10k-100k tokens.

cs.CL↗

arXiv2024

Moire magnetism in CrBr3 multilayers emerging from differential strain

Interfaces between twisted 2D materials host a wealth of physical phenomena originating from the long-scale periodicity associated with the resulting moire structure. Besides twisting, an alternative route to create structures with comparably long or even longer periodicities is inducing a differential strain between adjacent layers in a van der Waals (vdW) material. Despite recent theoretical efforts analyzing its benefits, this route has not yet been implemented experimentally. Here we report evidence for the simultaneous presence of ferromagnetic and antiferromagnetic regions in CrBr3 _a hallmark of moire magnetism_ from the observation of an unexpected magnetoconductance in CrBr3 tunnel barriers with ferromagnetic Fe3GeTe2 and graphene electrodes. The observed magnetoconductance evolves with temperature and magnetic field as the magnetoconductance measured in small angle CrBr3 twisted junctions, in which moire magnetism occurs. Consistent with Raman measurements and theoretical modeling, we attribute the phenomenon to the presence of a differential strain in the CrBr3 multilayer, which locally modifies the stacking and the interlayer exchange between adjacent CrBr3 layers, resulting in spatially modulated spin textures. Our conclusions indicate that inducing differential strain in vdW multilayers is a viable strategy to create moire-like superlattices, which in the future may offer in-situ continuous tunability even at low temperatures.

cond-mat.mtrl-sci↗

arXiv2024

Thermalization and Criticality on an Analog-Digital Quantum Simulator

Understanding how interacting particles approach thermal equilibrium is a major challenge of quantum simulators. Unlocking the full potential of such systems toward this goal requires flexible initial state preparation, precise time evolution, and extensive probes for final state characterization. We present a quantum simulator comprising 69 superconducting qubits which supports both universal quantum gates and high-fidelity analog evolution, with performance beyond the reach of classical simulation in cross-entropy benchmarking experiments. Emulating a two-dimensional (2D) XY quantum magnet, we leverage a wide range of measurement techniques to study quantum states after ramps from an antiferromagnetic initial state. We observe signatures of the classical Kosterlitz-Thouless phase transition, as well as strong deviations from Kibble-Zurek scaling predictions attributed to the interplay between quantum and classical coarsening of the correlated domains. This interpretation is corroborated by injecting variable energy density into the initial state, which enables studying the effects of the eigenstate thermalization hypothesis (ETH) in targeted parts of the eigenspectrum. Finally, we digitally prepare the system in pairwise-entangled dimer states and image the transport of energy and vorticity during thermalization. These results establish the efficacy of superconducting analog-digital quantum processors for preparing states across many-body spectra and unveiling their thermalization dynamics.

quant-ph↗

arXiv2024

Generic Multi-modal Representation Learning for Network Traffic Analysis

Network traffic analysis is fundamental for network management, troubleshooting, and security. Tasks such as traffic classification, anomaly detection, and novelty discovery are fundamental for extracting operational information from network data and measurements. We witness the shift from deep packet inspection and basic machine learning to Deep Learning (DL) approaches where researchers define and test a custom DL architecture designed for each specific problem. We here advocate the need for a general DL architecture flexible enough to solve different traffic analysis tasks. We test this idea by proposing a DL architecture based on generic data adaptation modules, followed by an integration module that summarises the extracted information into a compact and rich intermediate representation (i.e. embeddings). The result is a flexible Multi-modal Autoencoder (MAE) pipeline that can solve different use cases. We demonstrate the architecture with traffic classification (TC) tasks since they allow us to quantitatively compare results with state-of-the-art solutions. However, we argue that the MAE architecture is generic and can be used to learn representations useful in multiple scenarios. On TC, the MAE performs on par or better than alternatives while avoiding cumbersome feature engineering, thus streamlining the adoption of DL solutions for traffic analysis.

cs.LG↗

arXiv2024

Data Augmentation for Traffic Classification

Data Augmentation (DA) -- enriching training data by adding synthetic samples -- is a technique widely adopted in Computer Vision (CV) and Natural Language Processing (NLP) tasks to improve models performance. Yet, DA has struggled to gain traction in networking contexts, particularly in Traffic Classification (TC) tasks. In this work, we fulfill this gap by benchmarking 18 augmentation functions applied to 3 TC datasets using packet time series as input representation and considering a variety of training conditions. Our results show that (i) DA can reap benefits previously unexplored, (ii) augmentations acting on time series sequence order and masking are better suited for TC than amplitude augmentations and (iii) basic models latent space analysis can help understanding the positive/negative effects of augmentations on classification performance.

cs.LG↗

arXiv2023

Toward Generative Data Augmentation for Traffic Classification

Data Augmentation (DA)-augmenting training data with synthetic samples-is wildly adopted in Computer Vision (CV) to improve models performance. Conversely, DA has not been yet popularized in networking use cases, including Traffic Classification (TC). In this work, we present a preliminary study of 14 hand-crafted DAs applied on the MIRAGE19 dataset. Our results (i) show that DA can reap benefits previously unexplored in TC and (ii) foster a research agenda on the use of generative models to automate DA design.

cs.LG↗

arXiv2023

Replication: Contrastive Learning and Data Augmentation in Traffic Classification Using a Flowpic Input Representation

Over the last years we witnessed a renewed interest toward Traffic Classification (TC) captivated by the rise of Deep Learning (DL). Yet, the vast majority of TC literature lacks code artifacts, performance assessments across datasets and reference comparisons against Machine Learning (ML) methods. Among those works, a recent study from IMC22 [16] is worth of attention since it adopts recent DL methodologies (namely, few-shot learning, self-supervision via contrastive learning and data augmentation) appealing for networking as they enable to learn from a few samples and transfer across datasets. The main result of [16] on the UCDAVIS19, ISCX-VPN and ISCX-Tor datasets is that, with such DL methodologies, 100 input samples are enough to achieve very high accuracy using an input representation called "flowpic" (i.e., a per-flow 2d histograms of the packets size evolution over time). In this paper (i) we reproduce [16] on the same datasets and (ii) we replicate its most salient aspect (the importance of data augmentation) on three additional public datasets (MIRAGE19, MIRAGE22 and UTMOBILENET21). While we confirm most of the original results, we also found a 20% accuracy drop on some of the investigated scenarios due to a data shift in the original dataset that we uncovered. Additionally, our study validates that the data augmentation strategies studied in [16] perform well on other datasets too. In the spirit of reproducibility and replicability we make all artifacts (code and data) available to the research community at https://tcbenchstack.github.io/tcbench/

cs.LG↗

arXiv2023

Kagome chiral spin liquid in transition metal dichalcogenide moiré bilayers

At $n=3/4$ filling of the moiré flat band, transition metal dichalcogenide moiré bilayers will develop kagome charge order. We derive an effective spin model for the resulting localized spins and find that its further neighbor spin interactions can be much less suppressed than the corresponding electron hopping strength. Using density matrix renormalization group simulations, we study its phase diagram and, for realistic model parameters relevant for WSe$_2$/WS$_2$, we show that this material can realize the exotic chiral spin liquid phase and the highly debated kagome spin liquid. Our work thus demonstrates that the frustration and strong interactions present in TMD heterobilayers provide an exciting platform to study spin liquid physics.

cond-mat.str-el↗

arXiv2023

A Lightweight, Efficient and Explainable-by-Design Convolutional Neural Network for Internet Traffic Classification

Traffic classification, i.e. the identification of the type of applications flowing in a network, is a strategic task for numerous activities (e.g., intrusion detection, routing). This task faces some critical challenges that current deep learning approaches do not address. The design of current approaches do not take into consideration the fact that networking hardware (e.g., routers) often runs with limited computational resources. Further, they do not meet the need for faithful explainability highlighted by regulatory bodies. Finally, these traffic classifiers are evaluated on small datasets which fail to reflect the diversity of applications in real-world settings. Therefore, this paper introduces a new Lightweight, Efficient and eXplainable-by-design convolutional neural network (LEXNet) for Internet traffic classification, which relies on a new residual block (for lightweight and efficiency purposes) and prototype layer (for explainability). Based on a commercial-grade dataset, our evaluation shows that LEXNet succeeds to maintain the same accuracy as the best performing state-of-the-art neural network, while providing the additional features previously mentioned. Moreover, we illustrate the explainability feature of our approach, which stems from the communication of detected application prototypes to the end-user, and we highlight the faithfulness of LEXNet explanations through a comparison with post hoc methods.

cs.LG↗

arXiv2023

Many or Few Samples? Comparing Transfer, Contrastive and Meta-Learning in Encrypted Traffic Classification

The popularity of Deep Learning (DL), coupled with network traffic visibility reduction due to the increased adoption of HTTPS, QUIC and DNS-SEC, re-ignited interest towards Traffic Classification (TC). However, to tame the dependency from task-specific large labeled datasets we need to find better ways to learn representations that are valid across tasks. In this work we investigate this problem comparing transfer learning, meta-learning and contrastive learning against reference Machine Learning (ML) tree-based and monolithic DL models (16 methods total). Using two publicly available datasets, namely MIRAGE19 (40 classes) and AppClassNet (500 classes), we show that (i) using large datasets we can obtain more general representations, (ii) contrastive learning is the best methodology and (iii) meta-learning the worst one, and (iv) while ML tree-based cannot handle large tasks but fits well small tasks, by means of reusing learned representations, DL methods are reaching tree-based models performance also for small tasks.

cs.LG↗