Autoregressive video diffusion models generate streaming video by producing frames sequentially, conditioning each chunk on previously generated content. These models are structurally anchored to the first frame: its key-value representation occupies a privileged position in the attention cache and serves as the primary scene reference throughout generation. As the cleanest and most error-free position in the cache, this anchor draws disproportionate attention, suppressing video dynamics, and locking scene composition to the initial viewpoint even as the scene naturally evolves. The result is a temporally shallow video in which motion, camera movement, and scene progression are dampened in favor of static consistency. To address this, we replace the static anchor with an adaptive state, a hidden latent that the model denoises alongside content at every chunk but never renders. Rather than referencing a frozen first frame, the model generates its own scene anchor at each step by attending to both the previous state and the current content, producing a reference that evolves with the generated content. Unlike standard video generation, which encodes an absolute notion of time, our formulation treats time as relative: every generation step sees the same positional structure regardless of how far generation has progressed, and the state transition is identical at every chunk. Together, these properties introduce a recurrence into the generation process, where denoising serves as the transition function, and the KV cache serves as the carrier, requiring no external module. Experiments demonstrate that the adaptive state substantially improves video dynamics, enabling richer motion and natural scene progression within generated videos.

翻译：自回归视频扩散模型通过逐块生成帧、并令每个块条件依赖于先前生成的内容，从而生成流式视频。这类模型在结构上锚定于第一帧：其键值表示在注意力缓存中占据特权位置，并在整个生成过程中充当主要场景参考。作为缓存中最干净、无误差的位置，该锚点吸引了不成比例的注意力，抑制了视频动态性，并将场景构图锁定在初始视角上，即使场景自然演变也无法改变。其结果是生成时间维度浅显的视频，其中运动、镜头移动和场景进展被抑制，转而倾向于静态一致性。为解决该问题，我们将静态锚点替换为自适应状态，即一个隐变量，该变量在每个块中与内容一同进行去噪，但从不渲染。模型不再参考冻结的第一帧，而是通过同时关注前一状态和当前内容，在每一步生成自身的场景锚点，从而产生一个随生成内容演变的参考。与编码绝对时间概念的标准视频生成不同，我们的公式将时间视为相对的：无论生成进展到多远，每个生成步骤都看到相同的定位结构，且每个块的状态转移都是相同的。这些特性共同在生成过程中引入了循环，其中去噪充当转移函数，KV缓存充当载体，无需外部模块。实验表明，自适应状态显著改善了视频动态性，使得生成视频中能够呈现更丰富的运动和自然的场景进展。