FineSightBench
Explore score history ↗FineSightBench probes fine-scale visual perception in VLMs across scales from 4 to 48 pixels, separating perception and reasoning tasks.
- Released
- 2026-06-05
- Institutions
- University of Luxembourg · Université Paris-Saclay
- Evaluates
- General AI, Multimodal, Reasoning
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- unreviewed
FineSightBench paper
No reported scores are on record for this benchmark yet.