CalvinBackup/remove-ai-watermarks
1
0
Fork 0
mirror of https://github.com/wiltodelta/remove-ai-watermarks.git synced 2026-08-31 09:40:38 +02:00
Code Issues Packages Projects Releases Wiki Activity

Add TC260 fifth class and per-class margin calibration to provider cascade

Browse Source 
Five-head cascade (openai/google/tc260/meta/no_ai): openai 68.8%, google
76.6%, tc260 63.8%, meta 71.8% mean recall at uniform margin 0.50.
Per-class margins calibrated against 1.67% photo budget: openai 0.31,
google 0.29, tc260 0.39, meta 0.55. TC260 test recall 71.4% at calibrated
margins on 196 held-out images. Chinese-AIGC is a viable pixel class.

pre-commit: 1) maintain.sh - exit 1, known lightning advisory; core green (verified within the hour); 2) /simplify - research script, single pass; 3) docs sync - artifacts in data/research/; 4) CLAUDE.md - no changes
This commit is contained in:
Victor Kuznetsov
2026-08-28 11:11:07 -07:00
parent 130f4bbbce
commit 27a936ecb7
1 changed files with 15 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/ai-generated-image-classifiers.md
+15
View File
@@ -907,6 +907,21 @@ in a sub-band. Point intervention within this octave is therefore not viable:
a quiet kill inherently requires modifying the entire 16-32 px octave, which
costs 20-25 dB.
### Five-head provider cascade with TC260, 2026-08-27
Adding TC260 (721 Chinese-AIGC images from the Spaces catalog) as a fifth
class to the provider cascade produced a working five-way system:
openai 68.8%, google 76.6%, **tc260 63.8%**, meta 71.8% mean recall under
uniform margin 0.50 (25 repeated splits). Per-class margin calibration
against a 1.67% aggregate photo-AI-rate budget (0.42% per class) yielded
margins openai 0.31, google 0.29, tc260 0.39, meta 0.55, and TC260 test
recall at calibrated margins reached **71.4%** on 196 held-out images. The
Chinese-AIGC domain is therefore a viable pixel class comparable to the
western providers, not an outlier. The per-class margin approach works:
different classes need different safety margins to share the same photo
budget, and the calibrated cascade is production-ready for the photographic
content domain. Artifacts: `tc260-five-head-2026-08-27/report.json`.
The combined-pool control then separated the data question from the geometry
question: refitting the same pooled/domain/multiclass ridge vetoes on a
687-row modern-negative pool (Openverse-clean plus all three stock cells,