aiqbench analysis
This arXiv preprint proposes GILA, a method that pulls features from several layers of a frozen pretrained vision encoder and combines them with inter-layer attention, rather than relying only on the final layer, to detect AI-generated images. The reported gains in detection and cross-dataset generalization are measured on specific backbones, generators, and external datasets, so they need independent replication before being treated as robust. For readers, the practical point is that intermediate-layer representations may carry forensic signal that last-layer probes miss, which could matter for keeping detectors useful as generators change.
Beyond the Last Layer: Global Inter-Layer Aggregation for AI-Generated Image Detection
arXiv:2611.00017v1 Announce Type: cross Abstract: The rapid development of generative models has made AI-generated images increasingly close to real images. It has also brought several challenges to detectors, including rapid generator updates, large changes in data distribution, and limited cross-dataset generalization. Many existing methods based on pretrained vision encoders primarily use the global representation from the last layer, potentially leaving useful information from intermediate layers underexplored. To address this issue, we propose Global Inter-Layer Aggregation (GILA). GILA extracts global representations from different depths of a frozen encoder, models interactions among layer representations through inter-layer self-attention, and adaptively aggregates the resulting contextualized representations. The method is not tied to a single pretraining objective and does not require fine-tuning the pretrained encoder. We evaluated GILA on three vision backbones with different pretraining objectives, ten seen generators, and three external datasets. The experimental results show that GILA makes better use of multi-layer information in pretrained encoders and improves detection performance and cross-dataset generalization.
Abstract:The rapid development of generative models has made AI-generated images increasingly close to real images. It has also brought several challenges to detectors, including rapid generator updates, large changes in data distribution, and limited cross-dataset generalization. Many existing methods based on pretrained vision encoders primarily use the global representation from the last layer, potentially leaving useful information from intermediate layers underexplored. To address this issue, we propose Global Inter-Layer Aggregation (GILA). GILA extracts global representations from different depths of a frozen encoder, models interactions among layer representations through inter-layer self-attention, and adaptively aggregates the resulting contextualized representations. The method is not tied to a single pretraining objective and does not require fine-tuning the pretrained encoder. We evaluated GILA on three vision backbones with different pretraining objectives, ten seen generators, and three external datasets. The experimental results show that GILA makes better use of multi-layer information in pretrained encoders and improves detection performance and cross-dataset generalization.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2611.00017 [cs.CV]
(or arXiv:2611.00017v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2611.00017 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 253 words