A hierarchical diagnostic benchmark for counting stacked 3D objects shows that MLLMs often reach correct totals without executing the human-like prerequisite sub-tasks, and that training on those sub-tasks markedly improves accuracy.
The kaufman assessment battery for children—second edition and kabc-ii normative update.Contemporary intellectual assessment: Theories, tests, and issues, pages 333–359, 2018
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Spatial-IQ: Deconstructing Spatial Intelligence via Hierarchical Capability Tests
A hierarchical diagnostic benchmark for counting stacked 3D objects shows that MLLMs often reach correct totals without executing the human-like prerequisite sub-tasks, and that training on those sub-tasks markedly improves accuracy.