Images as Tables: In-Context Learning with TabPFN for Low-Data Detection of AI-Generated Images

2026-05-30

AI Summary

This work addresses the limited generalization of existing AI-generated image detection methods when confronted with novel generators and the scarcity of annotated samples. It reframes the problem as a structured prediction task under low-data conditions: image features extracted by a frozen DINOv2 encoder are reduced via PCA to construct structured tabular representations, which are then processed by TabPFN using in-context learning—eliminating the need for conventional fine-tuning. This approach enables rapid adaptation to new generators without retraining and outperforms the current state-of-the-art method, LATTE, by 8.2% in cross-generator and low-data settings, demonstrating the efficacy and potential of tabular foundation models in image forensics.

Abstract

AI-generated image detection is a moving-target problem: detectors trained on one generator often fail when a new generator appears, and only a few labeled examples are available. We study a simple image-to-table formulation for this regime, where each image is encoded by a frozen DINOv3 backbone, its CLS feature is reduced to a 500-dimensional structured row with PCA, and TabPFN performs real/fake classification by in-context tabular inference rather than task-specific classifier training. This turns fake-image detection into low-data structured prediction over learned visual features, making detector adaptation depend on the labeled context set instead of gradient-based fine-tuning. On GenImage, LATTE, a recent state-of-the-art detector, remains stronger when many labeled samples from all generators are available, by 7.4% in the largest pooled setting, but DINOv3-PCA-TabPFN is stronger in the practically important low-data regime, outperforming LATTE by up to 8.2%, and in transfer settings where the detector must generalize from one generator to another. These results position tabular foundation models as a strong complementary adaptation mechanism for image forensics, shifting adaptation from detector retraining to lightweight in-context updates with a small labeled set of examples. Code URL: https://github.com/jpwalter30/Towards-Generalizable-Detection-of-AI-Generated-Images

Problem

• AI-generated image detection
• low-data regime
• detector generalization
• in-context learning
• image forensics

Innovation

• in-context learning
• TabPFN
• low-data detection
• AI-generated image detection
• feature-to-table conversion