研究人员推出了MaineCoon，一个拥有220亿参数的音视频自回归模型，专为实时社交互动而设计。该模型在单个GPU上可实现高达47.5 FPS的帧率，并支持带有代理推理框架的长时序生成。MaineCoon采用了新颖的训练技术，如自重采样和强化在线策略蒸馏，旨在为针对AI原生社交平台的低延迟、高质量音视频生成设定新标杆。 AI
影响 为实时音视频生成设定了新标杆，有望赋能下一代AI原生社交平台。
排序理由 该集群描述了一篇在arXiv上发布的新研究论文，详细介绍了一个新颖的音视频模型。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
MaineCoon represents the first real-time audio-visual autoregressive model for social worlds, achieving high frame rates and long-horizon generation through novel training techniques and inference frameworks.
arXiv:2606.17800v1 Announce Type: new Abstract: As an increasing majority of global video content is consumed on social platforms for interactive social purposes, video generation models built for social worlds are important but largely overlooked by previous studies. In this wor…
As an increasing majority of global video content is consumed on social platforms for interactive social purposes, video generation models built for social worlds are important but largely overlooked by previous studies. In this work, we define the position of social world models…