Few-Shot Learner Generalizes Across AI-Generated Image Detection

Shiyu Wu, Jing Liu, Jing Li, Yequan Wang
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:67449-67460, 2025.

Abstract

Current fake image detectors trained on large synthetic image datasets perform satisfactorily on limited studied generative models. However, these detectors suffer a notable performance decline over unseen models. Besides, collecting adequate training data from online generative models is often expensive or infeasible. To overcome these issues, we propose Few-Shot Detector (FSD), a novel AI-generated image detector which learns a specialized metric space for effectively distinguishing unseen fake images using very few samples. Experiments show that FSD achieves state-of-the-art performance by $+11.6%$ average accuracy on the GenImage dataset with only $10$ additional samples. More importantly, our method is better capable of capturing the intra-category commonality in unseen images without further training. Our code is available at https://github.com/teheperinko541/Few-Shot-AIGI-Detector.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-wu25r, title = {Few-Shot Learner Generalizes Across {AI}-Generated Image Detection}, author = {Wu, Shiyu and Liu, Jing and Li, Jing and Wang, Yequan}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {67449--67460}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/wu25r/wu25r.pdf}, url = {https://proceedings.mlr.press/v267/wu25r.html}, abstract = {Current fake image detectors trained on large synthetic image datasets perform satisfactorily on limited studied generative models. However, these detectors suffer a notable performance decline over unseen models. Besides, collecting adequate training data from online generative models is often expensive or infeasible. To overcome these issues, we propose Few-Shot Detector (FSD), a novel AI-generated image detector which learns a specialized metric space for effectively distinguishing unseen fake images using very few samples. Experiments show that FSD achieves state-of-the-art performance by $+11.6%$ average accuracy on the GenImage dataset with only $10$ additional samples. More importantly, our method is better capable of capturing the intra-category commonality in unseen images without further training. Our code is available at https://github.com/teheperinko541/Few-Shot-AIGI-Detector.} }
Endnote
%0 Conference Paper %T Few-Shot Learner Generalizes Across AI-Generated Image Detection %A Shiyu Wu %A Jing Liu %A Jing Li %A Yequan Wang %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-wu25r %I PMLR %P 67449--67460 %U https://proceedings.mlr.press/v267/wu25r.html %V 267 %X Current fake image detectors trained on large synthetic image datasets perform satisfactorily on limited studied generative models. However, these detectors suffer a notable performance decline over unseen models. Besides, collecting adequate training data from online generative models is often expensive or infeasible. To overcome these issues, we propose Few-Shot Detector (FSD), a novel AI-generated image detector which learns a specialized metric space for effectively distinguishing unseen fake images using very few samples. Experiments show that FSD achieves state-of-the-art performance by $+11.6%$ average accuracy on the GenImage dataset with only $10$ additional samples. More importantly, our method is better capable of capturing the intra-category commonality in unseen images without further training. Our code is available at https://github.com/teheperinko541/Few-Shot-AIGI-Detector.
APA
Wu, S., Liu, J., Li, J. & Wang, Y.. (2025). Few-Shot Learner Generalizes Across AI-Generated Image Detection. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:67449-67460 Available from https://proceedings.mlr.press/v267/wu25r.html.

Related Material