Alessio Borgi - About
🚀 I’m always open to collaborate, exchange ideas or just talk about anything!
I’m eager to work with anyone who has great ideas, wants to learn more and more and also share their experience to others. Don’t hesitate to write me if you’d like to propose your help or ask for mine on a project, research, paper-idea, or a moonshot you’re cooking up.
Hi, I’m Alessio 👋
I’m a PhD student in Graph Neural Networks and Generative AI, under the supervision of Prof. Pietro Liò (University of Cambridge) and co-supervised by Prof. Fabrizio Silvestri (Sapienza University of Rome). I have obtained my Master of Engineering in Artificial Intelligence & Robotics and my Bachelor of Engineering in Applied Computer Science and Artificial Intelligence at Sapienza University of Rome, both with the highest marks. My research sits at the intersection of Graph Neural Networks, Geometric Deep Learning, Topological Deep Learning and Diffusion Models, with applications to Robotics, Vision, and Biomedical AI.
Places I’ve Been
Latest Publications
Polynomial Neural Sheaf Diffusion: A Spectral Filtering Approach on Cellular Sheaves
Published in arXiv preprint arXiv:2512.00242, 2025
ArXiv preprint on Polynomial Neural Sheaf Diffusion: A Spectral Filtering Approach on Cellular Sheaves.
Z-SASLM: Zero-Shot Style-Aligned SLI Blending for Latent Manipulation
Published in CVPR (Computer Vision and Pattern Recognition) 2025 Workshops (Nashville, USA 🇺🇸), 2025
Z-SASLM introduces a zero-shot, fine-tuning-free approach to style alignment in diffusion models by blending multiple reference styles directly in latent space using spherical linear interpolation (SLI) with learned, context-aware weights. The method avoids model retraining, preserves content semantics, and yields consistent style transfer across prompts and seeds.
Latest Blog Post
Top Languages 
Contribution Heatmap
Popular Repositories
AMR_CleaningRobot
Simulation of a Cleaning Robot with the capability of performing SLAM of the environment the robot is navigating, Planning Trajectories to calculate optimal paths considering static and dynamic obstacles, and Dynamic Obstacle Avoidance to detect and navigate around obstacles.
BioHeat-PINNs
Improving hyperthermia treatment by controlling temperature distribution in both 1D and 2D domains and thermal energy applied to cutaneous and subcutaneous tissues, through Bio-Heat equation with Physics-Informed Neural Networks (PINNs).
RoboMAT
A comprehensive MATLAB library for solving a wide range of robotics tasks, providing tools and functions for robotic simulations, control systems, kinematics, and path planning.
RealTime-VLM
RealTime-VLM brings real-time VLM inference to the browser. It continuously captures webcam frames, sends image+text to an OpenAI-compatible API, and displays responses with sub-second latency. Works with local or hosted VLMs.
Z-SAMB_StyleAligned_MultiReference-MultiModal
Novel framework for Zero-Shot Style Alignment in Text-to-Image generation, incorporating Multi-Modal Context-Awareness and Multi-Reference Style Alignment, using minimal attention sharing, ensuring consistent style transfer without fine-tuning.
AdaViT
Adaptive Vision Transformer for efficient image classification, implementing dynamic token sparsification to reduce computational costs while maintaining accuracy.
Z-SASLM
[CVPR 2025] Z-SASLM is a zero-shot framework for multi-style image synthesis leveraging Spherical Linear Interpolation (SLI) to achieve smooth, coherent blending—without any fine-tuning.
MoonBot-Navigation
Autonomous robot for lunar navigation and object interaction, developed during TESP '25 at the Space Robotics Lab (Tohoku University). Features custom robot design, Dijkstra-based path planning, object detection with vision, and gripper control.
XGNN-GraphGenRL
Explainable Graph Neural Networks (XGNNs) and Reinforcement Learning (RL) techniques for graph generation and optimization tasks. The project aims to explore and validate the use of explainable AI for creating interpretable graph structures.
SkinMe
A deep learning-based application for skin disease detection and classification.
ALPR-Automatic-License-Plate-Recognition
A comprehensive system for detecting and recognizing license plates in real-time, combining image processing, object detection, and optical character recognition (OCR) to provide an accurate and efficient solution for automatic license plate recognition, with also double GUI, both for the Security Manager and for the Car User.
Clustering-Deepening
An in-depth exploration of clustering algorithms and techniques in machine learning, with applications focus on Object Tracking and Image Segmentation.