cv
Basics
Name | Giordano Cicchetti |
giordano.cicchetti.gc@gmail.com | |
Summary | Ph.D. Candidate in Multimodal AI with expertise in machine learning and deep learning. Specialized in multimodal representation learning and alignment, and deep knowledge of multimodal generative AI. Proven track record in collaborative research environments and leadership in innovative projects. |
Language | Italian, English |
Education
-
2023 - 2026 Rome, Italy
Ph.D. in Information and Communication Technology (ICT)
Sapienza University of Rome
Information and Communication Technology (ICT)
-
2021 - 2023 Rome, Italy
-
2018 - 2021 Rome, Italy
Work
-
2024.09 - 2024.12 Teaching Assistant – MSc in Electronics Engineering
Sapienza University of Rome
Assisted in lectures, project preparation, creation of new teaching material, and evaluation for data science and machine learning basics.
-
2023.04 - 2023.10 AI Engineering Intern
Enel Global Services S.r.l.
Developed a Generative AI-based pipeline to restore original quality of dirty documents, improving OCR performance. Implemented AI solutions using AE, GAN, and Diffusion Models, with the best model being NAF-DPM.
- AWS services including Lambda, S3, and SageMaker
- Python, PyTorch
Publications
-
2025 Contrastive Learning Enables Low-Bandwidth Semantic Communication
AAAI 2025 AI4WCN Workshop
-
2025 Closing The Modality Gap Enables Novel Multimodal Learning Applications
ICLR 2025 Re-Align Workshop
-
2025 -
2024 NAF-DPM: A nonlinear activation-free diffusion probabilistic model for document enhancement
(Under review at IEEE Transactions on Image Processing)
Skills
Programming | |
Python | |
PyTorch | |
TensorFlow | |
Numpy | |
Java |
Development | |
Docker | |
Git | |
SQL | |
NoSQL | |
CI/CD | |
AWS | |
Linux |
Soft Skills | |
Leadership | |
Project Management | |
Scientific Writing | |
Public Speaking | |
Time Management |
Projects
-
-
-
Img2Img-SC
Implementation of Language-Oriented Semantic Latent Representation for Image Transmission.