World Editing: Intervening on Executable Worlds at Increasing Depth

Max Ku, Nok-Kan Law, Yu-Chien Tang, Shih-Ying Yeh, Ping Nie, Andy Zheng, Tat Hei Lai, Fei-Yueh Chen, Nikko Yu, Wei-Chieh Sun, Suzy Huang, Chiao-Wei Hsu, Chih-Chuan Huang, Chak-Wing Mak, Ho Yin Sam Ng, Edisy Kin Wai Chan, Min-Hung Chen, Ho Kei Cheng

arXiv:2610.02331 · 2026-10-06 공개 · arXiv · PDF

coding-agents visual-consistency minecraft executable-worlds world-editing game-modding igmworld igmbench

Abstract

Interactive world models are increasingly capable of generating environments and acting within them, yet deliberately editing an existing executable world remains underexplored. We formulate world editing as intervening on an existing world while preserving properties that should remain unchanged, and introduce intervention depth as an axis describing how strongly an edit couples world entities, dynamics, and systems. We instantiate this capability through industry-grade game modding and introduce IGMWorld, together with IGMBench, a benchmark of 110 tasks and over 1.1K executable state and behavioral criteria across Minecraft and Terraria. The tasks span property, entity, dynamics, and system interventions and are evaluated through deterministic executability, behavioral, preservation, and visual checks. Frontier coding agents already exhibit substantial world-editing capability: the strongest configuration solves 78.2% of tasks under a strict task-level criterion, while criterion-level performance reaches 94.8%. Reliability generally decreases with intervention depth, and this pattern persists even among tasks with similar numbers of evaluation criteria. Most failed edits still build and load successfully, suggesting that the main difficulty is making the edited world behave as requested. Visual consistency remains a separate weakness, with all evaluated configurations below 50% joint visual pass rate. These results show that world editing is a distinct capability from world generation and interaction, and that executable games provide a practical testbed for studying it.

한국어 요약

한 줄 요약

실행 가능한 게임 세계에 대한 의도적 편집 능력을 평가하는 IGMWorld와 IGMBench를 제안하며, 편집 깊이가 성능에 미치는 영향을 분석한다.

핵심 기여도

핵심 아이디어

기존 연구는 주로 세계 생성(World Generation)과 상호작용(World Interaction)에 집중했으나, 본 연구는 **기존 실행 가능한 세계에 대한 의도적 편집**(World Editing)을 새로운 연구 영역으로 제시한다. 이는 단순히 시각적 수정을 넘어, **엔티티, 역학, 시스템 간의 결합**(Coupling)을 고려한 개입을 포함한다.

**Intervention Depth**는 편집이 세계 내 요소들을 얼마나 강하게 연결하는지를 기준으로 정의되며, 이는 편집의 복잡도와 신뢰도 사이의 관계를 분석하는 핵심 지표이다. 연구는 **게임 모드**(Game Modding)를 통해 이 개념을 구현하고, **실행 가능성**(Executability), **행동 정확도**(Behavioral Correctness), **시각 일관성**(Visual Consistency)을 평가하는 세 단계 평가 프레임워크를 제안한다.

기술적 접근법

주요 결과

의의 및 한계

실용적 활용