The Trinity of Consistency as a Defining Principle for General World Models

TL;DR

This paper proposes the Trinity of Consistency as a foundational principle for developing General World Models, emphasizing semantic, geometric, and causal consistencies, and introduces CoW-Bench for evaluating multimodal models in reasoning and generation tasks.

Contribution

It introduces a theoretical framework based on three types of consistency for world models and presents CoW-Bench, a new benchmark for multi-frame reasoning and generation evaluation.

Findings

Unified architectures enable emergent internal world simulators

CoW-Bench effectively evaluates multimodal models' reasoning and generation

The framework clarifies limitations and guides future model development

Abstract

The construction of World Models capable of learning, simulating, and reasoning about objective physical laws constitutes a foundational challenge in the pursuit of Artificial General Intelligence. Recent advancements represented by video generation models like Sora have demonstrated the potential of data-driven scaling laws to approximate physical dynamics, while the emerging Unified Multimodal Model (UMM) offers a promising architectural paradigm for integrating perception, language, and reasoning. Despite these advances, the field still lacks a principled theoretical framework that defines the essential properties requisite for a General World Model. In this paper, we propose that a World Model must be grounded in the Trinity of Consistency: Modal Consistency as the semantic interface, Spatial Consistency as the geometric basis, and Temporal Consistency as the causal engine. Through this tripartite lens, we systematically review the evolution of multimodal learning, revealing a trajectory from loosely coupled specialized modules toward unified architectures that enable the synergistic emergence of internal world simulators. To complement this conceptual framework, we introduce CoW-Bench, a benchmark centered on multi-frame reasoning and generation scenarios. CoW-Bench evaluates both video generation models and UMMs under a unified evaluation protocol. Our work establishes a principled pathway toward general world models, clarifying both the limitations of current systems and the architectural requirements for future progress.