PAC Bench evaluates vision-language models on properties, affordances, and constraints for manipulation, finding near-zero performance on physical constraints.
PHYBench: Holistic evaluation of physical perception and reasoning in large language models, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
PAC Bench: Do Foundation Models Understand Prerequisites for Executing Manipulation Policies?
PAC Bench evaluates vision-language models on properties, affordances, and constraints for manipulation, finding near-zero performance on physical constraints.