CounterScene endows generative world models with explicit counterfactual reasoning for safety-critical driving evaluation.
March 24, 2026
Original Paper
CounterScene: Counterfactual Causal Reasoning in Generative World Models for Safety-Critical Closed-Loop Evaluation
arXiv · 2603.21104
The Takeaway
Instead of using heuristic adversarial perturbations, this framework uses causal interaction graphs to identify and modify critical agents. This significantly improves the realism and effectiveness of safety testing for autonomous vehicles in closed-loop simulations.
From the abstract
Generating safety-critical driving scenarios requires understanding why dangerous interactions arise, rather than merely forcing collisions. However, existing methods rely on heuristic adversarial agent selection and unstructured perturbations, lacking explicit modeling of interaction dependencies and thus exhibiting a realism--adversarial trade-off. We present CounterScene, a framework that endows closed-loop generative BEV world models with structured counterfactual reasoning for safety-critic