{"ID":5438597,"CreatedAt":"2026-07-01T01:17:58.482524686Z","UpdatedAt":"2026-07-03T01:40:09.565152011Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.31082","arxiv_id":"2606.31082","title":"Fleet: Few Shots Lead Effective AI-generated Image Detection","abstract":"AI-generated image (AIGI) detection is undergoing a critical transition from laboratory benchmarks to open-world adversarial defense. The prevalent paradigm focuses on finding static feature spaces, assuming that some invariant artifacts learned from historical data can achieve universal zero-shot generalization. While achieving saturation on several AIGI benchmarks, this static hypothesis suffers a severe performance drop against rapidly evolving generators (e.g., SD3, Nano Banana Pro). To address these limitations, we propose that the field should expand beyond \"static generalization\" to a new paradigm of \"dynamic adaptation\". We introduce Fleet, a framework that pioneers a dynamic paradigm of continuous few-shot evolution, enabling rapid alignment with emerging generative threats. Fleet improves few-shot adaptation by replacing unconstrained feature updates with constrained routing correction, where avoidance routing redirects novel AI samples away from Non-AI-dominated routes within decoupled subspaces. To validate this, we present Treasure, a benchmark spanning 64 models and 360k images, featuring diverse architectures and 20 closed-source commercial engines. Experiments reveal that while static SOTA methods fail catastrophically on modern generators, Fleet restores performance from 20.4% to 73.1% with only 10-shot adaptation on \"Doubao Seedream 4.0\". Code and data are available at https://github.com/ICTMCG/Fleet .","short_abstract":"AI-generated image (AIGI) detection is undergoing a critical transition from laboratory benchmarks to open-world adversarial defense. The prevalent paradigm focuses on finding static feature spaces, assuming that some invariant artifacts learned from historical data can achieve universal zero-shot generalization. While...","url_abs":"https://arxiv.org/abs/2606.31082","url_pdf":"https://arxiv.org/pdf/2606.31082v1","authors":"[\"Jiaan Wang\",\"Sirui Liu\",\"Yu Li\",\"Kaiyuan Yang\",\"Juan Cao\",\"Sheng Tang\"]","published":"2026-06-30T03:15:52Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false,"code_links":[{"ID":613761,"CreatedAt":"2026-07-01T01:17:58.482524686Z","UpdatedAt":"2026-07-01T01:17:58.482524686Z","DeletedAt":null,"paper_id":5438597,"paper_url":"https://arxiv.org/abs/2606.31082","paper_title":"Fleet: Few Shots Lead Effective AI-generated Image Detection","repo_url":"https://github.com/ICTMCG/Fleet","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}