Researchers have published a paper on arXiv titled "From Pixels to States: Rethinking Interactive World Models as Game Engines." The paper explores how recent video generative models can be used as next-generation game engines by predicting future observations based on user actions. It analyzes interactive game world modeling across four dimensions: player action control, game state dynamics, state-observation persistence, and real-time interactive generation. To support this research, the authors developed a data engine for the game Black Myth: Wukong, collecting over 90 hours of gameplay data including actions, game states, and observations. AI
IMPACT This research could advance the development of more coherent and responsive interactive worlds in games and AI applications.
RANK_REASON Academic paper published on arXiv
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
