Brought a Gun to a Knife fight: Modern VFM Baselines Outgun Specialized Detectors on In-the-Wild AI Image Detection

Yue Zhou, Xinan He, Kaiqing Lin, Bing Fan, Feng Ding, Jinhua Zeng, Bin Li \sthanks Corresponding author · 2025

Read Brought a Gun to a Knife fight: Modern VFM Baselines Outgun Specialized Detectors on In-the-Wild AI Image Detection by Yue Zhou, Xinan He, Kaiqing Lin, Bing Fan, Feng Ding, Jinhua Zeng, Bin Li \sthanks Corresponding author (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-12995.txt.

Description

Yue Zhou, Xinan He and five co-authors show in 2025 that a plain head on a fresh vision base beats craft detectors for wild fake images. Craft tools shine on neat tests but miss many fakes in the wild, with high miss rates.

Trained on the same data, the plain head lifts wild scores by more than twenty points. Probes show fresh vision and tongue models link fake images to forgery words. Older models lack that link. Scores crash on fresh scraped data past the base cut-off. True tests thus need data free of all past training touch.

Catalog record

CREATORYue Zhou, Xinan He, Kaiqing Lin, Bing Fan, Feng Ding, Jinhua Zeng, Bin Li \sthanks Corresponding author
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE21 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-12995

Browse the full library · About the Tower of Babel