Do Enterprise Systems Need Learned World Models? The Importance of Context to Infer Dynamics

TL;DR

This paper investigates the necessity of learned world models in enterprise systems, emphasizing the importance of runtime discovery of transition dynamics for robustness under deployment shifts.

Contribution

It introduces enterprise discovery agents that recover transition dynamics at runtime by reading configurations, enhancing robustness compared to offline-trained models.

Findings

Offline-trained models degrade under deployment shift.

Discovery-based agents maintain accuracy by grounding in current configurations.

Proposed CascadeBench benchmark evaluates enterprise cascade prediction.

Abstract

World models enable agents to anticipate the effects of their actions by internalizing environment dynamics. In enterprise systems, however, these dynamics are often defined by tenant-specific business logic that varies across deployments and evolves over time, making models trained on historical transitions brittle under deployment shift. We ask a question the world-models literature has not addressed: when the rules can be read at inference time, does an agent still need to learn them? We argue, and demonstrate empirically, that in settings where transition dynamics are configurable and readable, runtime discovery complements offline training by grounding predictions in the active system instance. We propose enterprise discovery agents, which recover relevant transition dynamics at runtime by reading the system's configuration rather than relying solely on internalized representations. We introduce CascadeBench, a reasoning-focused benchmark for enterprise cascade prediction that adopts the evaluation methodology of World of Workflows on diverse synthetic environments, and use it together with deployment-shift evaluation to show that offline-trained world models can perform well in-distribution but degrade as dynamics change, whereas discovery-based agents are more robust under shift by grounding their predictions in the current instance. Our findings suggest that, in configurable enterprise environments, agents should not rely solely on fixed internalized dynamics, but should incorporate mechanisms for discovering relevant transition logic at runtime.