Gamma-World：超越雙玩家的生成式多智能體世界建模

摘要

用於互動式影片生成的 world model 大多聚焦於單一智能體情境，亦即僅透過單一控制訊號生成未來觀測。然而，許多生成的環境需要多智能體互動：多位玩家、機器人或具身智能體在共享空間中同時行動。若要將 world model 擴展至此類場景，就必須建構具原則性的多智能體設計：智能體應維持獨立可控性、具置換對稱性，並支援高效推論，同時保持時間與視角之間的一致性。本文提出用於互動模擬的生成式多智能體 world model。其中引入單純形旋轉智能體編碼（Simplex Rotary Agent Encoding），這是 3D RoPE 的無參數擴展，將智能體表示為旋轉角度空間中正規單純形的頂點。此方法賦予每個智能體獨特的相位，同時使所有智能體具備置換等價性，無需學習每個 slot 的專屬標識或固定的智能體排序，即可實現可擴展的智能體身分識別。為避免智能體之間密集的全連接注意力，我們進一步提出稀疏中心注意力（Sparse Hub Attention），其中可學習的中心令牌（hub token）調節跨智能體的令牌互動，將跨智能體注意力成本從智能體數量的二次方降至線性。為了實現即時生成，我們將全上下文擴散教師模型蒸餾為因果學生模型，後者依序產生時間區塊並搭配 KV 快取，以 24 FPS 的速率生成可回應動作的影片。在多玩家虛擬環境的實驗顯示，我們的模型在影片真實度、動作可控性及智能體間一致性上，均優於基於 slot 與基於密集注意力的基線方法，且無需額外訓練即可從兩位玩家泛化至四位玩家。

English

World models for interactive video generation have largely focused on single-agent settings, where future observations are generated from a single control signal. However, many generated environments require multi-agent interaction: multiple players, robots, or embodied agents act simultaneously within a shared space. Scaling world models to such settings requires a principled multi-agent design: agents should remain independently controllable, permutation-symmetric, and support efficient inference while maintaining consistency across time and perspectives. In this paper, we present our generative multi-agent world model for interactive simulation. It introduces Simplex Rotary Agent Encoding, a parameter-free extension of 3D RoPE that represents agents as vertices of a regular simplex in rotary angle space. This gives each agent a distinct phase while making all agents permutation-equivalent, enabling scalable agent identity without learned per-slot identities or a fixed agent ordering. To avoid dense all-to-all attention across agents, we further propose Sparse Hub Attention, where learnable hub tokens mediate token interaction across agents, reducing cross-agent attention cost from quadratic to linear in the number of agents. For real-time rollout, we distill a full-context diffusion teacher into a causal student that generates temporal blocks sequentially with KV caching, enabling action-responsive generation at 24 FPS. Experiments in multiplayer virtual environments show that our model improves video fidelity, action controllability, and inter-agent consistency over slot-based and dense-attention baselines, while generalizing from two to four players without additional training.