让视频生成像导演一样实时切换镜头，保持剧情连贯。
CausalCine: Real-Time Autoregressive Generation for Multi-Shot Video Narratives
- 用因果模型跨镜头生成，动态记忆检索避免内容错乱。
- 支持实时修改提示词，旧镜头不重算，生成更高效。
- 适合需要即时交互的影视创作或动画生成场景。
自回归视频生成旨在实现实时、无边界合成。然而，电影叙事不仅涉及单一场景的持续延伸，还需经历事件演变、视角切换和离散镜头边界。现有自回归模型在此场景下表现不佳：主要针对短时续接训练，将长序列视为单一镜头延续，导致长期生成中出现运动停滞与语义漂移。为此，我们提出CausalCine，一种交互式自回归框架，将多镜头视频生成转化为在线导播过程。CausalCine在镜头切换间进行因果生成，支持实时动态提示输入，并通过复用历史上下文避免重复生成。首先，我们在原生多镜头序列上训练因果基础模型，学习复杂镜头转换先验；其次，提出内容感知记忆路由（CAMR），依据注意力相关性而非时间邻近度动态检索历史键值对，保障有限活跃内存下的跨镜头一致性；最后，将因果基础模型蒸馏为少步生成器，实现实时交互生成。大量实验表明，CausalCine显著优于自回归基线，接近双向模型性能，同时实现了因果生成的流式交互能力。演示地址：https://yihao-meng.github.io/CausalCine/
原文摘要 · Abstract (English)
Autoregressive video generation aims at real-time, open-ended synthesis. Yet, cinematic storytelling is not merely the endless extension of a single scene; it requires progressing through evolving events, viewpoint shifts, and discrete shot boundaries. Existing autoregressive models often struggle in this setting. Trained primarily for short-horizon continuation, they treat long sequences as extended single shots, inevitably suffering from motion stagnation and semantic drift during long rollouts. To bridge this gap, we introduce CausalCine, an interactive autoregressive framework that transforms multi-shot video generation into an online directing process. CausalCine generates causally across shot changes, accepts dynamic prompts on the fly, and reuses context without regenerating previous shots. To achieve this, we first train a causal base model on native multi-shot sequences to learn complex shot transitions prior to acceleration. We then propose Content-Aware Memory Routing (CAMR), which dynamically retrieves historical KV entries according to attention-based relevance scores rather than temporal proximity, preserving cross-shot coherence under bounded active memory. Finally, we distill the causal base model into a few-step generator for real-time interactive generation. Extensive experiments demonstrate that CausalCine significantly outperforms autoregressive baselines and approaches the capability of bidirectional models while unlocking the streaming interactivity of causal generation. Demo available at https://yihao-meng.github.io/CausalCine/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。