AIIC AI Intelligence Centre

SOURCE-LINKED INTELLIGENCE

Dual-Frontier: When Can an Agent Trust Its World Model?

arXiv · AI, language, vision and robotics · article · Sep 22, 2026 · UTC

Learned world models are becoming essential to general-purpose agents: by predicting action consequences, they support planning and decision-making while reducing reliance on costly trial and error. This reliance creates a fundamental ambiguity: when a world-model-guided decision fails, the trajectory alone may not reveal whether the agent's decision rule or the world model caused the loss. We formalize this failure-attribution problem as a counterfactual decomposition of return loss and prove that its components are not identifiable from passive interaction, even for finite-horizon planners.

Read original source ↗ Open in workspace

recordType
paper
region
Global

Evidence & attribution

First collected: 2026-09-23T04:11:12.117Z. This is not the publication date.