T
traeai
Sign in

概念

Sparse Hub Attention

Gamma-World提出的稀疏注意力拓扑,通过枢纽token实现线性复杂度跨智能体通信。

已跟踪 2 条高相关材料

TraeAI 观察

相关材料

已收录 2 条与 Sparse Hub Attention 相关的内容,按评分排序。

英伟达清华团队提出Gamma-World:世界模型从「一个人玩」到「多人共处」

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

FeaturedArticle#World Model#Multi-Agent#Transformer#NVIDIA#Tsinghua中文
英伟达清华团队提出Gamma-World:世界模型从「一个人玩」到「多人共处」

Gamma-World systematically solves multi-agent world modeling via simplex agent encoding and sparse hub attention, enabling zero-shot generalization from 2-player training to 4-player inference and 24 FPS real-time rollout, with average FVD reduction >40%.

入选理由:采用正单纯形(regular simplex)编码实现玩家身份等距、无参数、可扩展,支持训练时2人→推理时4人零样本泛化

FeaturedArticle#World Model#Multi-Agent#Transformer#NVIDIA#Tsinghua中文

跨材料问答 · Sparse Hub Attention

回答基于:Sparse Hub Attention 相关 2 条材料
    0 / 500

    AI may generate inaccurate information. Please verify important content.