From Pixels to States: Rethinking Interactive World Models as Game Engines

Zhen Li, Zian Meng, Shuwei Shi, Mingliang Zhai, Jiaming Tan, Chuanhao Li, Kaipeng Zhang

arXiv:2607.14076 · 2026-07-17 공개 · arXiv · PDF

real-time-generation state-dynamics data-engine interactive-world-models black-myth-wukong game-state-persistence player-action-control game-engines

Abstract

Building interactive worlds that respond coherently to player actions has long been a shared goal of computer graphics, games, and artificial intelligence. Recent video generative models provide a data-driven route toward this goal by predicting future observations conditioned on user actions, and are increasingly regarded as potential next-generation game engines. Realizing a genuinely interactive game world, however, requires interaction outcomes that follow rules over evolving game conditions, consequences that persist over long horizons, and a generation loop that operates in real time. Conventional game engines realize these properties through a recurrent action-state-observation loop, in which player actions update an explicit game state according to predefined rules and observations are rendered from the resulting state. Taking this loop as an organizing lens, this paper examines interactive game world modeling along four dimensions: player action control, game state dynamics, state-observation persistence, and real-time interactive generation. For each dimension, we start from the capabilities required by an interactive game world, group existing approaches into representative families, and discuss the strengths and trade-offs of each family. Complementing this analysis, we present a scalable data engine for Black Myth: Wukong that collects over 90 hours of gameplay with frame-aligned player actions, ground-truth game states, and visual observations, together with structured and semantic annotations, as a resource for state-aware game world modeling. We hope this paper offers a clear picture of where the field stands and fosters progress toward interactive game worlds.

한국어 요약

한 줄 요약

게임 엔진의 action-state-observation 루프를 기반으로 한 대화형 게임 세계 모델링을 체계적으로 분석하고, 90시간 분량의 Black Myth: Wukong 데이터셋을 제시한다.

핵심 기여도

핵심 아이디어

기존 게임 엔진은 플레이어 액션 → 게임 상태 업데이트 → 관찰 렌더링의 반복 루프를 통해 대화형 세계를 구축한다. 이 논문은 이 루프를 기반으로, 데이터 기반 생성 모델이 대화형 세계를 구현하기 위해 갖춰야 할 4가지 핵심 차원을 정의한다. 이는 단순히 미래 관찰을 예측하는 것을 넘어, 게임 규칙에 따른 상태 변화와 장기적 결과의 일관성, 실시간 생성을 포함한다. 특히, 대부분의 생성 모델이 게임 상태를 암묵적으로 유지하는 반면, 이 논문은 명시적 상태 표현의 중요성을 강조하며, 이를 위한 데이터셋 구축을 제안한다.

기술적 접근법

주요 결과

의의 및 한계

실용적 활용