A new arXiv preprint, Matrix-Game 4.0, argues that moving from controllable videos to genuinely playable worlds requires three things: persistent world state, executable game rules, and low-latency interaction. The authors position these as the essential barriers to real-time playability.
The paper outlines a roadmap for the next generation of Matrix-Game, though the abstract provides only a partial view of the approach. The text begins to describe a separation of game-related components, but the sentence is cut off in the available source, leaving the specific architecture unspecified.
As the sole source, there are no conflicting accounts to compare. The abstract does not yet detail implementation or evaluation plans, so the roadmap's feasibility remains an open question.