WorldRuler: Interactive World Modeling under Configurable Rules
Summary
WorldRuler 是一个基于可执行 3D 规则图的生成式世界模型,能在保存与加载时持久化用户自定义规则与世界状态,并通过闭环的规则执行、视频生成与感知反馈准确实现规则引发的视觉变化。作者还基于 GTA-V 和 Habitat 构建了 RuleVerse 数据集以支持规则驱动交互的研究。
View Cached Full Text
Cached at: 11/19/26, 07:27 AM
Paper page - WorldRuler: Interactive World Modeling under Configurable Rules
Source: https://huggingface.co/papers/2611.24869
Abstract
A rule-graph-based generative world model preserves user-defined rules and world states across saving and loading while accurately realizing rule-induced visual changes on designated objects and attributes.
Rules govern how player actions change interactive worlds. Different rules create distinct gameplay experiences within similar scenes and action spaces. For customizable gameplay,generative world modelsmust followuser-defined rulesbeyond real-world behaviors. However, realizinguser-defined rulespresents two challenges:Rule-State PersistenceandRule-to-Visual Realization. The former requires preserving rule definitions, dependencies, and world states across saving and loading, so that restored worlds follow the established rules. The latter requires accurately realizing rule-induced changes on designated objects and attributes, particularly when affected objects are unseen. In this paper, we propose WorldRuler, which organizes rule definitions, dependencies, and object states in anexecutable 3D rule graphand connects rule execution,video generation, andperceptual feedbackin a closed loop. The graph supports joint saving and restoration of rules and world states, whiledynamic object bindinggrounds rule-specified entities in visual instances and uses stored states to constrain generation from their first appearance. We further introduceobject attribute state replacementandcounterfactual reconstruction training, enabling accurateattribute controland compatibleappearance and motion changeswhile limiting interference with unrelated content. We develop an automated data collection and annotation pipeline using GTA-V and Habitat to construct RuleVerse, a dataset ofrule-driven interactionswith detailedstate and event annotations, including cases where rules take effect before affected objects become visible. Extensive experiments demonstrate the effectiveness of WorldRuler in preserving rule-state consistency across saving and loading and accurately realizing rule outcomes on designated objects and attributes.
View arXiv pageView PDFProject pageGitHub6Add to collection
Get this paper in your agent:
hf papers read 2611\.24869
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2611.24869 in a model README.md to link it from this page.
Datasets citing this paper1
#### AlayaLab/RuleVerse Preview• Updated32 minutes ago • 37
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2611.24869 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
WorldStream: Streaming Interactive 3D Worlds with Hierarchical Planning and Persistent State
WorldStream是一个开放式的3D世界模型,通过分层规划与流式编译实现地形无限扩展、处理开放语言指令并持久保存用户编辑。宏/微规划器在有类型的决策模型上生成结构化决策状态,并通过全局记录维护可复现的世界状态与状态化NPC。
Matrix-Game: Interactive World Foundation Model
Matrix-Game 是一个 17B 参数的交互式世界基础模型,通过两阶段训练(大规模无标注预训练 + 动作标注训练)实现可控的 Minecraft 游戏世界生成,在动作可控性与物理一致性上超越 Oasis 和 MineWorld 等开源模型,并将开源权重与 GameWorld Score 基准。
State World Models: Execute the Rules, Generate the Pixels
The paper introduces Stateful Harness, a framework that separates rule-based world state updates from frozen video rendering to improve consistency in interactive world models with hidden state, and evaluates it with the WorldState-Bench benchmark, where the full harness raises rule-compliance from 32% to 92%.
RAR: Rate-Adaptive Rollout for Time-Consistent Interactive World Models
RAR (Rate-Adaptive Rollout) 是一个让交互式世界模型以可控时间速率进行 rollout 的框架,通过 Stride-RoPE 将不同采样速率的潜在帧对齐到共享物理时间坐标,实现多速率生成、未见速率泛化与在线速率切换,同时用多步长视频采样兼顾细粒度动态与长时程演化。
WorldGuide: Goal-Directed Video World Model for Procedural Task Execution
提出了 WorldGuide,一种面向长程程序化任务的闭环视频世界模型,通过规划器与执行器耦合及分层视觉记忆,在自建的 WorldGuide-Bench 上以 33.33% 任务成功率超过 MiniMax-H3。