52 citations · 122 across the 21 of their papers we have counts for
25 papers
DANTINOX: A Unified Framework for Multi-Paradigm Language Modeling
Marco Simoni, Aleksandar Fontana, Giulio Rossolini +1
Language generation research increasingly spans three paradigms: autoregressive decoding, discrete masked diffusion, and continuous flow-matching. Comparing them is difficult becau…
Position Matters: Feature Inversion Attacks in ViT Split Inference with Token Reduction and Shuffling
Stefano Leggio, Giulio Rossolini, Alessandro Biondi
Vision Transformers (ViTs) are increasingly used in split-inference systems, where edge devices transmit intermediate token representations to a remote cloud. In this setting, toke…
Adversarial Decoys: Misdirecting Attention-Based Defenses in ViT
Giulia Marchiori Pietrosanti, Giulio Rossolini, Giorgio Buttazzo
Vision Transformers (ViTs) remain vulnerable to localized adversarial attacks, e.g., adversarial patches, while recent test-time defenses mitigate them by suppressing image tokens…
PairedGTA: Generating Driving Datasets for Controlled Photometric Shift Analysis
Andrea Chianese, Giulio Rossolini, Alessandro Biondi +2
Evaluating the performance of visual perception systems for autonomous driving is essential to ensure reliable operation across diverse environmental scenarios. Ideally, a balanced…
Learning Robustness at Test-Time from a Non-Robust Teacher
Stefano Bianchettin, Giulio Rossolini, Giorgio Buttazzo
Nowadays, pretrained models are increasingly used as general-purpose backbones and adapted at test-time to downstream environments where target data are scarce and unlabeled. While…
How Worst-Case Are Adversarial Attacks? Linking Adversarial and Perturbation Robustness
Giulio Rossolini
Adversarial attacks are widely used to identify model vulnerabilities; however, their validity as proxies for robustness to random perturbations remains debated. We ask whether an…
On the Hidden Objective Biases of Group-based Reinforcement Learning
Aleksandar Fontana, Marco Simoni, Giulio Rossolini +2
Group-based reinforcement learning methods, like Group Relative Policy Optimization (GRPO), are widely used nowadays to post-train large language models. Despite their empirical su…
KGQuest: Template-Driven QA Generation from Knowledge Graphs with LLM-Based Refinement
Sania Nayab, Marco Simoni, Giulio Rossolini +1
The generation of questions and answers (QA) from knowledge graphs (KG) plays a crucial role in the development and testing of educational platforms, dissemination tools, and large…
GTPO: Stabilizing Group Relative Policy Optimization via Gradient and Entropy Control
Marco Simoni, Aleksandar Fontana, Giulio Rossolini +2
Group Relative Policy Optimization (GRPO) is a promising policy-based approach for Large Language Model alignment, yet its performance is often limited by training instability and…
Exploiting Edge Features for Transferable Adversarial Attacks in Distributed Machine Learning
Giulio Rossolini, Fabio Brau, Alessandro Biondi +2
As machine learning models become increasingly deployed across the edge of internet of things environments, a partitioned deep learning paradigm in which models are split across mu…
Improving LLM Reasoning for Vulnerability Detection via Group Relative Policy Optimization
Marco Simoni, Aleksandar Fontana, Giulio Rossolini +1
Improving and understanding the training dynamics and reasoning of Large Language Models (LLMs) has become essential for their deployment in AI-based security tools, such as softwa…
Leveraging Knowledge Graphs and LLMs for Structured Generation of Misinformation
Sania Nayab, Marco Simoni, Giulio Rossolini
The rapid spread of misinformation, further amplified by recent advances in generative AI, poses significant threats to society, impacting public opinion, democratic stability, and…
Benchmarking the Spatial Robustness of DNNs via Natural and Adversarial Localized Corruptions
Giulia Marchiori Pietrosanti, Giulio Rossolini, Alessandro Biondi +1
The robustness of deep neural networks is a crucial factor in safety-critical applications, particularly in complex and dynamic environments (e.g., medical or driving scenarios) wh…
Video Deblurring by Sharpness Prior Detection and Edge Information
Yang Tian, Fabio Brau, Giulio Rossolini +2
Video deblurring is essential task for autonomous driving, facial recognition, and security surveillance. Traditional methods directly estimate motion blur kernels, often introduci…
Edge-Only Universal Adversarial Attacks in Distributed Learning
Giulio Rossolini, Tommaso Baldi, Alessandro Biondi +1
Distributed learning frameworks, which partition neural network models across multiple computing nodes, enhance efficiency in collaborative edge-cloud systems, but may also introdu…
Concise Thoughts: Impact of Output Length on LLM Reasoning and Cost
Sania Nayab, Giulio Rossolini, Marco Simoni +4
Today's large language models (LLMs) can solve challenging question-answering tasks, and prompt engineering techniques, such as chain-of-thought (CoT), have gained attention for en…
Attention-Based Real-Time Defenses for Physical Adversarial Attacks in Vision Applications
Giulio Rossolini, Alessandro Biondi, Giorgio Buttazzo
Deep neural networks exhibit excellent performance in computer vision tasks, but their vulnerability to real-world adversarial attacks, achieved through physical objects that can c…
TrainSim: A Railway Simulation Framework for LiDAR and Camera Dataset Generation
Gianluca D'Amico, Mauro Marinoni, Federico Nesti +4
The railway industry is searching for new ways to automate a number of complex train functions, such as object detection, track discrimination, and accurate train positioning, whic…
Robust-by-Design Classification via Unitary-Gradient Neural Networks
Fabio Brau, Giulio Rossolini, Alessandro Biondi +1
The use of neural networks in safety-critical systems requires safe and robust models, due to the existence of adversarial attacks. Knowing the minimal adversarial perturbation of…
CARLA-GeAR: a Dataset Generator for a Systematic Evaluation of Adversarial Robustness of Vision Models
Federico Nesti, Giulio Rossolini, Gianluca D'Amico +2
Adversarial examples represent a serious threat for deep neural networks in several application domains and a huge amount of work has been produced to investigate them and mitigate…
Defending From Physically-Realizable Adversarial Attacks Through Internal Over-Activation Analysis
Giulio Rossolini, Federico Nesti, Fabio Brau +2
This work presents Z-Mask, a robust and effective strategy to improve the adversarial robustness of convolutional networks against physically-realizable adversarial attacks. The pr…
On the Real-World Adversarial Robustness of Real-Time Semantic Segmentation Models for Autonomous Driving
Giulio Rossolini, Federico Nesti, Gianluca D'Amico +3
The existence of real-world adversarial examples (commonly in the form of patches) poses a serious threat for the use of deep learning models in safety-critical computer vision tas…
On the Minimal Adversarial Perturbation for Deep Neural Networks with Provable Estimation Error
Fabio Brau, Giulio Rossolini, Alessandro Biondi +1
Although Deep Neural Networks (DNNs) have shown incredible performance in perceptive and control tasks, several trustworthy issues are still open. One of the most discussed topics…
Evaluating the Robustness of Semantic Segmentation for Autonomous Driving against Real-World Adversarial Patch Attacks
Federico Nesti, Giulio Rossolini, Saasha Nair +2
Deep learning and convolutional neural networks allow achieving impressive performance in computer vision tasks, such as object detection and semantic segmentation (SS). However, r…