Transformers outperformed young children and matched their error profile on a geometry odd-one-out task, while vision-language models underperformed vision-only models.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Computer Vision Models Show Human-Like Sensitivity to Geometric and Topological Concepts
Transformers outperformed young children and matched their error profile on a geometry odd-one-out task, while vision-language models underperformed vision-only models.