PulseAugur
实时 05:31:04

新AI模型可模拟复杂多人游戏,实现长时稳定预测

研究人员开发了一种新颖的多人世界模型,能够模拟具有复杂物理交互的高度动态环境。该模型是一个拥有50亿参数的潜在扩散模型,它以多个智能体的动作为条件,以正确归因场景变化并保持连贯性。该模型在10,000小时的Rocket League游戏数据上进行训练,可以在单个Nvidia B200 GPU上以每秒20帧的速度实时生成四人对战,预测可以保持长达五分钟甚至数小时的稳定。研究团队正在发布数据集、代码库和实时演示,以促进该领域的进一步研究。 AI

影响 能够更真实、更稳定地模拟复杂的多智能体环境,有望推动AI在游戏和机器人领域的进步。

排序理由 该集群描述了一篇研究论文,详细介绍了一种新的人工智能模型架构及其应用,包括代码和数据的发布。

在 Hugging Face Daily Papers 阅读 →

AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →

新AI模型可模拟复杂多人游戏,实现长时稳定预测

报道来源 [3]

  1. arXiv cs.AI TIER_1 English(EN) · Anthony Hu, V\'aclav Volhejn, Adrien Ramanana Rahary, Chris Mulder, Aditya Makkar, Am\'elie Royer, Manu Orsini, Alyx Liao, Adam Jelley, Eloi Alonso, Florian Laurent, Fredrik Nor\'en, James Swingos, Jan H\"unermann, Kent Rollins, Lucas Hosseini, Matthieu … ·

    Multiplayer Interactive World Models with Representation Autoencoders

    arXiv:2607.05352v1 Announce Type: cross Abstract: We introduce the first multiplayer world model for highly dynamic environments governed by complex physical interactions. Whereas single-player world models treat the other agents as part of the environment, ours conditions on the…

  2. arXiv cs.AI TIER_1 English(EN) · Patrick Pérez ·

    具有表示自编码器的多人交互世界模型

    We introduce the first multiplayer world model for highly dynamic environments governed by complex physical interactions. Whereas single-player world models treat the other agents as part of the environment, ours conditions on the action streams of multiple agents, learning to at…

  3. Hugging Face Daily Papers TIER_1 English(EN) ·

    Multiplayer Interactive World Models with Representation Autoencoders

    A large-scale multiplayer world model trained on extensive gameplay data demonstrates stable long-horizon rollouts in a complex physics-based environment while maintaining coherence across multiple agents' actions.