Benchmark RadarAI BENCHMARK PROFILE
INCLUDE-BENCH
Evaluates disability-related bias in text-to-image models using 119K generated images across bias dimensions and contexts, with the Stereotype Content Model Score.
- Released
- 2026-07-09
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Addresses the underexplored area of disability stereotypes in T2I models, providing a large-scale evaluation for representational harms.
Motivation
Text-to-image (T2I) models have been shown to exhibit social biases.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.