推理对齐：扩散Transformer在视频编辑中的隐式推理
Reasoning to Align: Implicit Reasoning in Diffusion Transformers for Video Editing
AI总结：
针对指令式视频编辑中条件信号未分化及交叉注意力监督不足的问题，提出RVEDiT框架，通过粒度路由令牌条件和参考锚定注意力对齐实现粗到细编辑与内部推理正则化。
摘要：
基于指令的视频编辑需要根据自然语言指令转换源视频，同时保留无关内容并保持时间连贯性。我们认为现有的扩散Transformer（DiT）编辑器由于两个结构原因难以完成此任务。首先，条件信号未分化地输入所有Transformer块，迫使单个令牌流同时编码全局编辑意图和细粒度视觉证据。其次，控制编辑的交叉注意力模式仅通过像素级重建间接监督，使得模型内部推理过程约束不足。为了解决这两个限制，我们提出了RVEDiT，一个隐式推理视频编辑DiT框架，围绕两个互补组件构建。第一个组件，粒度路由令牌条件，引入从多模态大语言模型蒸馏的可学习编辑令牌，并将其路由到浅层块，同时将原生视觉和文本令牌保留给深层块，从而在骨干网络内部诱导出从粗到细的编辑过程。第二个组件，参考锚定注意力对齐，在训练期间采用参数共享的参考分支，并最大化编辑分支和参考分支注意力特征之间的互信息，正则化模型的内部推理而不产生任何额外的推理成本。在标准基于指令的视频编辑基准上的实验表明，RVEDiT始终优于最先进的基线，特别是在局部和组合编辑方面取得了显著提升。
英文摘要：
Instruction-based video editing requires transforming a source video according to a natural-language instruction while preserving irrelevant content and remaining temporally coherent. We argue that existing Diffusion Transformer (DiT) editors struggle with this task for two structural reasons. First, conditioning signals are fed undifferentiated into all transformer blocks, forcing a single token stream to encode both global editing intent and fine-grained visual evidence. Second, the cross-attention patterns that govern the edit are supervised only indirectly through pixel-level reconstruction, leaving the model's internal reasoning process under-constrained. To address both limitations, we propose RVEDiT, an implicit Reasoning Video Editing DiT framework built around two complementary components. The first, Granularity-Routed Token Conditioning, introduces learnable editing tokens distilled from a multimodal LLM and routes them to shallow blocks, while reserving native visual and textual tokens for deeper blocks, thereby inducing a coarse-to-fine editing process inside the backbone. The second, Reference-Anchored Attention Alignment, employs a parameter-sharing reference branch during training and maximizes the mutual information between the attention features of the editing and reference branches, regularizing the model's internal reasoning without incurring any additional inference cost. Experiments on standard instruction-based video editing benchmarks show that RVEDiT consistently outperforms state-of-the-art baselines, with particularly strong gains on localized and compositional edits.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
