pith. machine review for the scientific record. sign in

arxiv: 2604.16405 · v1 · submitted 2026-03-31 · 💻 cs.RO · cs.AI· cs.CV· cs.LG

ICAT: Incident-Case-Grounded Adaptive Testing for Physical-Risk Prediction in Embodied World Models

classification 💻 cs.RO cs.AIcs.CVcs.LG
keywords modelsembodiedriskworldicatplanningsevereseverity
0
0 comments X
read the original abstract

Video-generative world models are increasingly used as neural simulators for embodied planning and policy learning, yet their ability to predict physical risk and severe consequences is rarely evaluated.We find that these models often downplay or omit key danger cues and severe outcomes for hazardous actions, which can induce unsafe preferences during planning and training on imagined rollouts. We propose ICAT, which grounds testing in real incident reports and safety manuals by building structured risk memories and retrieving/composing them to constrain the generation of risk cases with causal chains and severity labels. Experiments on an ICAT-based benchmark show that mainstream world models frequently miss mechanisms and triggering conditions and miscalibrate severity, falling short of the reliability required for safety-critical embodied deployment.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.