Study measures value leakage in multilingual text-to-image generation
A new arXiv paper introduces a method to measure 'visual value leakage' in text-to-image systems across languages, noting prior cultural-representation evaluations focused mainly on recognizability.
Read the full story at arXiv cs.CV ↗
Timeline · 1 report
- 2026-11-05 05:00 · arXiv cs.CV
Whose Values Are Generated? Measuring Visual Value Leakage in Multilingual Text-to-Image Generation