GeoVAD-Bench
Explore score history ↗GeoVAD-Bench diagnoses visual chain-of-thought for geometry problem solving along five dimensions: perception, auxiliary quality, utilization, deductive reasoning, and final correctness, with controlled No-Aux, Auto-Aux, and GT-Aux intervention settings.
- Released
- 2026-09-11
- Institutions
- Southeast University · Ant Group (China)
- Evaluates
- Geometric reasoning, Language & Knowledge, Reasoning, Robotics & Embodied AI
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- ai-name-audit-deferred
GeoVAD-Bench paper
No reported scores are on record for this benchmark yet.