无尽世界：实时 3D 感知长视频生成
Endless World: Real-Time 3D-Aware Long Video Generation
条件自回归训练策略把新生成内容与已有帧对齐，配合全局 3D-aware 注意力注入几何约束，可在单卡实时生成无限长且 3D 一致的视频；这套「对齐已有帧 + 3D 引导」的思路可直接迁移到流式视频生成与长序列扩散采样。
原文依据 a conditional autoregressive training strategy that aligns newly generated content with existing video frames
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论生成具有稳定三维结构的长时连贯视频序列仍是一项重大挑战，在流式场景中尤为如此。受此驱动，我们提出 Endless World，一个用于无限、三维一致视频生成的实时框架。为支持无限视频生成，我们引入条件自回归训练策略，使新生成的内容与已有视频帧对齐。该设计在保持长程依赖的同时维持计算高效，从而无需额外训练开销即可在单块 GPU 上实现实时推理。此外，我们的 Endless World 集成全局三维感知注意力，以在时间维度上提供连续的几何引导。我们的三维注入机制在长序列中强制物理合理性与几何一致性，解决了长时程与动态场景合成中的关键挑战。大量实验表明，Endless World 能够生成长时间、稳定且视觉连贯的视频，在视觉保真度与空间一致性两方面均取得与现有方法相当或更优的性能。我们的代码将在评审结束后发布。
英文原文摘要 1144 字符
Producing long, coherent video sequences with stable 3D structure remains a major challenge, particularly in streaming scenarios. Motivated by this, we introduce Endless World, a real-time framework for infinite, 3D-consistent video generation. To support infinite video generation, we introduce a conditional autoregressive training strategy that aligns newly generated content with existing video frames. This design preserves long-range dependencies while remaining computationally efficient, enabling real-time inference on a single GPU without additional training overhead. Moreover, our Endless World integrates global 3D-aware attention to provide continuous geometric guidance across time. Our 3D injection mechanism enforces physical plausibility and geometric consistency throughout extended sequences, addressing key challenges in long-horizon and dynamic scene synthesis. Extensive experiments demonstrate that Endless World produces long, stable, and visually coherent videos, achieving competitive or superior performance to existing methods in both visual fidelity and spatial consistency. Our code will be released after review.