NVIDIA & Tsinghua Propose Gamma-World: World Models Evolve from ‘Solo Play’ to ‘Multi-Agent Coexistence’
Gamma-World systematically solves architectural gaps in multi-agent world modeling via simplex agent encoding and sparse hub attention, achieving >40% average FVD reduction, zero-shot generalization from 2 to 4 agents, and 24 FPS real-time rollout.
入选理由:采用正单纯形(regular simplex)编码玩家身份,实现任意玩家间几何等距,支持零样本扩展至更多玩家且无需重训

