Benchmark RadarAI BENCHMARK PROFILE
VLMsAreBlind
A vision-language benchmark that probes blind spots and brittle reasoning in multimodal models.
- Released
- Unknown
- Readiness
- Paper only
- Primary field
- General AI
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.