Preference ≠ Tattooability: Auditing Reward Models for AI Tattoo Generation
Description
AI tattoo generation is usually ranked with the same reward models that rank general text-to-image outputs. Those models predict which image a crowd annotator prefers. A tattoo that will be worn also has to survive a needle, a gap that heals shut, and ink that spreads. We audited that gap with deterministic, physically scaled probes and with CLIPScore, PickScore, HPSv2.1 and ImageReward, on tattoo-subject pairs from HPDv2 and ImageRewardDB, on controlled defects, and on a print-scale ink-spread stress test. On HPDv2 tattoo-subject pairs, a-priori probes agree with human preference below chance, and that contrast remains after a Holm correction across scorers. Nothing here ranks a commercial AI tattoo generator.
Competing interest: the author operates tatany.app; that product was not evaluated or ranked.
Files
Preference_Tattooability_v1.0.pdf
Additional details
Related works
- Is documented by
- Other: https://tatany.app (URL)
- Is supplement to
- Software: https://github.com/Stark-Will/tattoo-probes (URL)
Software
- Repository URL
- https://github.com/Stark-Will/tattoo-probes