Text-to-image models show sharp performance drops on counterfactual image tasks, indicating reliance on real-world visual co-occurrence patterns rather than causal deduction.
2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Are Text-to-Image Models Inductivist Turkeys? A Counterfactual Benchmark for Causal Reasoning
Text-to-image models show sharp performance drops on counterfactual image tasks, indicating reliance on real-world visual co-occurrence patterns rather than causal deduction.