Omni-WorldBench: Towards a Comprehensive Interaction-Centric Evaluation for World Models

TL;DR

Omni-WorldBench introduces a comprehensive benchmark to evaluate the interactive response capabilities of 4D world models, addressing a critical gap in current evaluation methods by measuring how models respond to actions over space and time.

Contribution

The paper presents Omni-WorldBench, a novel benchmark with a prompt suite and agent-based metrics for systematically assessing interactive response in 4D world models.

Findings

Current models show limitations in interactive response capabilities.

Extensive evaluations reveal gaps in causal understanding of models.

Benchmark provides actionable insights for future improvements.

Abstract

Video--based world models have emerged along two dominant paradigms: video generation and 3D reconstruction. However, existing evaluation benchmarks either focus narrowly on visual fidelity and text--video alignment for generative models, or rely on static 3D reconstruction metrics that fundamentally neglect temporal dynamics. We argue that the future of world modeling lies in 4D generation, which jointly models spatial structure and temporal evolution. In this paradigm, the core capability is interactive response: the ability to faithfully reflect how interaction actions drive state transitions across space and time. Yet no existing benchmark systematically evaluates this critical dimension. To address this gap, we propose Omni--WorldBench, a comprehensive benchmark specifically designed to evaluate the interactive response capabilities of world models in 4D settings. Omni--WorldBench comprises two key components: Omni--WorldSuite, a systematic prompt suite spanning diverse interaction levels and scene types; and Omni--Metrics, an agent-based evaluation framework that quantifies world modeling capabilities by measuring the causal impact of interaction actions on both final outcomes and intermediate state evolution trajectories. We conduct extensive evaluations of 18 representative world models across multiple paradigms. Our analysis reveals critical limitations of current world models in interactive response, providing actionable insights for future research. Omni-WorldBench will be publicly released to foster progress in interactive 4D world modeling.