What-If World is a new paired-prompt benchmark showing that nine state-of-the-art video generation models achieve at most 52% on causal intervention tests and cluster near 28% for open-source systems.
The essential role of causality in foundation world models for embodied ai.arXiv preprint arXiv:2402.06665, 2024
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
LLMs fail causal discovery due to a kernel obstruction in observational learning, but interventional agents using frozen LLMs in Bayesian loops succeed without training on causal graph benchmarks.
citing papers explorer
-
What-If World: A Causal Benchmark for General World Models in Embodied Scenarios
What-If World is a new paired-prompt benchmark showing that nine state-of-the-art video generation models achieve at most 52% on causal intervention tests and cluster near 28% for open-source systems.
-
Why LLMs Fail at Causal Discovery and How Interventional Agents Escape
LLMs fail causal discovery due to a kernel obstruction in observational learning, but interventional agents using frozen LLMs in Bayesian loops succeed without training on causal graph benchmarks.