← Benchmarks
PointGrounding
PointArena is a comprehensive platform for evaluating multimodal pointing across diverse reasoning scenarios. It includes Point-Bench, a curated dataset of ~1,000 pointing tasks across five categories: Spatial (positional references), Affordance (functional part identification), Counting (attribute-based grouping), Steerable (relative pointing), and Reasoning (open-ended visual inference). The benchmark evaluates language-guided pointing capabilities in vision-language models.
id pointgrounding · max 1 · 1 models reported
| # | Model | Score |
|---|
No scores for this benchmark yet.
