Benchmark RadarAI BENCHMARK PROFILE
NumBench
Text-to-image (T2I) models often generate the wrong number of objects, yet existing benchmarks are too small or weakly controlled to explain why.
- Released
- 2026-08-28
- Readiness
- Paper only
- Primary field
- General AI
Motivation
Text-to-image (T2I) models often generate the wrong number of objects, yet existing benchmarks are too small or weakly controlled to explain why.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.