Multiverse Mechanica: A Testbed for Learning Game Mechanics via Counterfactual Worlds
Abstract
We study how generative world models trained on video games can go beyond mere reproduction of gameplay visuals to learning game mechanics-the modular rules that causally govern gameplay. We introduce a formalization of the concept of game mechanics that operationalizes mechanic-learning as a causal counterfactual inference task and uses the causal consistency principle to address the challenge of generating gameplay with world models that do not violate game rules. We present Multiverse Mechanica, a playable video game testbed that implements a set of ground truth game mechanics based on our causal formalism. The game natively emits training data, where each training example is paired with a set of causal DAGs that encode causality, consistency, and counterfactual dependence specific to the mechanic that is in play-these provide additional artifacts that could be leveraged in mechanic-learning experiments. We provide a proofof-concept that demonstrates fine-tuning a pre-trained model that targets mechanic learning. Multiverse Mechanica is a testbed that provides a reproducible, low-cost path for studying and comparing methods that aim to learn game mechanics-not just pixels. See Appendix D for author contribution details.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- GameGen-X: Interactive Open-world Game Video Generation cites Genie: Generative Interactive Environments
- CLIPScore: A Reference-free Evaluation Metric for Image Captioning cites Learning Transferable Visual Models From Natural Language Supervision
Builds on7
- Learning Transferable Visual Models From Natural Language SupervisionAlec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh et al.ICML 2021 · 47,906 citations
- Elucidating the Design Space of Diffusion-Based Generative ModelsTero Karras, Miika Aittala, Timo Aila, Samuli LaineNeurIPS 2022 · 3,959 citations
- CLIPScore: A Reference-free Evaluation Metric for Image CaptioningJack Hessel, Ari Holtzman, Maxwell Forbes, Ronan Le Bras et al.EMNLP 2021 · 937 citations
- Genie: Generative Interactive EnvironmentsJake Bruce, Michael D. Dennis, Ashley Edwards, Jack Parker-Holder et al.ICML 2024 · 513 citations
- GameGen-X: Interactive Open-world Game Video GenerationHaoxuan Che, Xuanhua He, Quande Liu, Cheng Jin et al.ICLR 2025 · 2 citations
Related papers
- CoPhy: Counterfactual Learning of Physical DynamicsFabien Baradel, Natalia Neverova, Julien Mille, Greg Mori et al.ICLR 2020 · 105 citations
- Disentangled Counterfactual Learning for Physical Audiovisual Commonsense ReasoningChangsheng Lv, Shuai Zhang, Yapeng Tian, Mengshi Qi et al.NeurIPS 2023 · 26 citations
- Causally Consistent Normalizing FlowQingyang Zhou, Kangjie Lu, Meng XuAAAI 2025
- Distilling Counterfactual Reasoning from Language to Vision: Causal Graph-Guided Post-Training for Video UnderstandingYuefei Chen, Jiang Liu, Xiaodong Lin, Ruixiang TangCVPR Findings 2026
- MECD: Unlocking Multi-Event Causal Discovery in Video ReasoningTieyuan Chen, Huabin Liu, Tianyao He, Yihang Chen et al.NeurIPS 2024 · 36 citations
