Federico Cocchi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Federico Cocchi has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Marcella Cornia
University of Modena and Reggio Emilia
6 shared publications - Rita Cucchiara
University of Modena and Reggio Emilia
6 shared publications - Lorenzo Baraldi
University of Modena and Reggio Emilia
5 shared publications - Alessandro Nicolosi
Leonardo Labs
2 shared publications - Samuele Poppi
University of Pisa
2 shared publications - Anna Mambelli
University of Modena and Reggio Emilia
1 shared publication
Latest Publications
- 15 September 2025
- Published by Springer Nature
- p. 36-52
Generating Synthetic Data with Large Language Models for Low-Resource Sentence Retrieval
- 15 September 2025
- Published by Springer Nature
- p. 36-52
Generating Synthetic Data with Large Language Models for Low-Resource Sentence Retrieval
- 4 December 2024
- Published by Springer Nature
- p. 111-126
Adapt to Scarcity: Few-Shot Deepfake Detection via Low-Rank Adaptation
- 4 December 2024
- Published by Springer Nature
- p. 111-126
Adapt to Scarcity: Few-Shot Deepfake Detection via Low-Rank Adaptation
- 1 December 2024
- Published by Springer Nature
- p. 340-356
Safe-CLIP: Removing NSFW Concepts from Vision-and-Language Models
- 1 December 2024
- Published by Springer Nature
- p. 340-356
Safe-CLIP: Removing NSFW Concepts from Vision-and-Language Models
- 21 November 2024
- Published by Springer Nature
- p. 199-216
Contrasting Deepfakes Diffusion via Contrastive Learning and Global-Local Similarities
- 21 November 2024
- Published by Springer Nature
- p. 199-216
Contrasting Deepfakes Diffusion via Contrastive Learning and Global-Local Similarities
- Open Access
Contrasting Deepfakes Diffusion via Contrastive Learning and Global-Local Similarities
Discerning between authentic content and that generated by advanced AI methods has become increasingly challenging. While previous research primarily addresses the detection of fake faces, the identification of generated natural images has only recently surfaced. This prompted the recent exploration of solutions that employ foundation vision-and-language models, like CLIP. However, the CLIP embedding space is optimized for global image-to-text alignment and is not inherently designed for deepfake detection, neglecting the potential benefits of tailored training and local image features. In this study, we propose CoDE (Contrastive Deepfake Embeddings), a novel embedding space specifically designed for deepfake detection. CoDE is trained via contrastive learning by additionally enforcing global-local similarities. To sustain the training of our model, we generate a comprehensive dataset that focuses on images generated by diffusion models and encompasses a collection of 9.2 million images produced by using four different generators. Experimental results demonstrate that CoDE achieves state-of-the-art accuracy on the newly collected dataset, while also showing excellent generalization capabilities to unseen image generators. Our source code, trained models, and collected dataset are publicly available at: https://github.com/aimagelab/CoDE.
Open AccessContrasting Deepfakes Diffusion via Contrastive Learning and Global-Local Similarities
Discerning between authentic content and that generated by advanced AI methods has become increasingly challenging. While previous research primarily addresses the detection of fake faces, the identification of generated natural images has only recently surfaced. This prompted the recent exploration of solutions that employ foundation vision-and-language models, like CLIP. However, the CLIP embedding space is optimized for global image-to-text alignment and is not inherently designed for deepfake detection, neglecting the potential benefits of tailored training and local image features. In this study, we propose CoDE (Contrastive Deepfake Embeddings), a novel embedding space specifically designed for deepfake detection. CoDE is trained via contrastive learning by additionally enforcing global-local similarities. To sustain the training of our model, we generate a comprehensive dataset that focuses on images generated by diffusion models and encompasses a collection of 9.2 million images produced by using four different generators. Experimental results demonstrate that CoDE achieves state-of-the-art accuracy on the newly collected dataset, while also showing excellent generalization capabilities to unseen image generators. Our source code, trained models, and collected dataset are publicly available at: https://github.com/aimagelab/CoDE.
- 5 September 2023
- Published by Springer Nature
- p. 345-356
Unveiling the Impact of Image Transformations on Deepfake Detection: An Experimental Analysis
- 5 September 2023
- Published by Springer Nature
- p. 345-356
Unveiling the Impact of Image Transformations on Deepfake Detection: An Experimental Analysis