Game model separates combat rules from video for better gameplay
Fight World Model: Customizable Combat through Visual Interaction Proxies
Computer Vision and Pattern Recognition
Summary
Video game characters fighting on screen need a system that keeps track of health and hit reactions properly. The authors introduce Fight World Model, which divides the game into two parts: one that strictly follows the rules and keeps track of fighting details, and another that creates the visuals. This lets players see consistent hit effects and health changes, and allows new fighting abilities without retraining the visual system. Their method works well with many characters and arenas and beats earlier models at keeping things accurate.
What this means in practice
- •For video game developers: Customize and extend fighting rules and character abilities in games without retraining animation or rendering systems.$Commercial implications: Enables selling more flexible and modifiable fighting game engines supporting new mechanics and character designs.
- •For interactive media creators: Create visual combat scenes with precise control over timing and effects, allowing new types of interactions beyond original game rules.
Authors
Jiahua Li, Zixu John, Tom Zhong, Fuping Wu, Tianhao Xu, Ruicheng Zhang, Jun Zhou, Zunnan Xu
Abstract
A playable, programmable game world must maintain persistent state, enforce executable rules, and support extensible gameplay mechanics. Existing video world models can generate plausible combat, yet struggle to keep hit reactions and health updates consistent with game rules. We introduce Fight World Model (FWM), which separates rule-governed combat execution from video rendering. An executable engine maintains state and resolves interactions. A deterministic compiler encodes these outcomes and attempted actions in a spatially and temporally aligned interaction proxy. The proxy aligns action and skill windows, brief hit reactions, and persistent health by fighter identity and timing. These cues can be controlled separately to render interactions under modified rules. Identity-bound references preserve character appearance across side switches. We train FWM on FightWorld Corpus, spanning 52 skill effects, 187 character-appearance variants, and 15 arenas. Without retraining the renderer, FWM supports damage scaling, limited-hit immunity, lifesteal, and multi-stage skill mechanisms beyond the native game. It also allows character and arena customization. Experiments demonstrate the effectiveness of FWM, which outperforms existing game world models in action fidelity, hit reactions, health-state consistency, and character appearance preservation. Our project is available at https://fight-world-model.github.io.