AdaState：用于流式视频生成的自演化锚点

摘要

自回归视频扩散模型通过逐帧生成流式视频，并在每个片段中基于已生成的内容进行条件化处理。这类模型在结构上以首帧为锚点：其键值表示在注意力缓存中占据特殊地位，并作为整个生成过程中的主要场景参照。由于该锚点在缓存中是最清晰且无误差的位置，模型会过度关注它，从而抑制视频动态性，并使得场景构图锁定在初始视角，即使场景自然演变也无法改变。这导致生成的视频在时间维度上浅层化，其中的运动、镜头移动和场景发展均被弱化，以换取静态的一致性。为解决这一问题，我们将静态锚点替换为自适应状态——一个隐藏的隐变量，模型在每个片段中与内容一同去噪，但从不渲染它。模型不再参照固定的首帧，而是通过同时关注前一状态和当前内容，在每个步骤中生成自身的场景锚点，从而产生一个随生成内容演变的参照。与编码绝对时间概念的标准视频生成不同，我们的方法将时间视为相对的：每个生成步骤都看到相同的位置结构，无论生成进行到何种程度，且状态转换在每个片段中完全一致。这些特性共同在生成过程中引入了递归机制，其中去噪充当转换函数，键值缓存充当载体，无需外部模块。实验表明，自适应状态显著提升了视频动态性，使生成视频中包含更丰富的运动和自然的场景演变。

English

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.