VChain：视频生成推理中的视觉思维链

摘要

近期视频生成模型虽能制作出流畅且视觉吸引人的片段，但在合成具有连贯因果链的复杂动态场景时仍显不足。准确建模随时间推移的视觉结果与状态转换，仍是核心挑战。相比之下，大型语言与多模态模型（如GPT-4o）展现出强大的视觉状态推理与未来预测能力。为融合这些优势，我们提出了VChain，一种新颖的推理时视觉思维链框架，它将多模态模型的视觉推理信号注入视频生成过程。具体而言，VChain包含一个专用流程，利用大型多模态模型生成一组稀疏的关键帧作为快照，随后仅在这些关键时刻指导预训练视频生成器的稀疏推理时微调。我们的方法调优高效，引入额外开销极小，并避免了密集监督。在复杂多步骤场景上的大量实验表明，VChain显著提升了生成视频的质量。

English

Recent video generation models can produce smooth and visually appealing clips, but they often struggle to synthesize complex dynamics with a coherent chain of consequences. Accurately modeling visual outcomes and state transitions over time remains a core challenge. In contrast, large language and multimodal models (e.g., GPT-4o) exhibit strong visual state reasoning and future prediction capabilities. To bridge these strengths, we introduce VChain, a novel inference-time chain-of-visual-thought framework that injects visual reasoning signals from multimodal models into video generation. Specifically, VChain contains a dedicated pipeline that leverages large multimodal models to generate a sparse set of critical keyframes as snapshots, which are then used to guide the sparse inference-time tuning of a pre-trained video generator only at these key moments. Our approach is tuning-efficient, introduces minimal overhead and avoids dense supervision. Extensive experiments on complex, multi-step scenarios show that VChain significantly enhances the quality of generated videos.

VChain：视频生成推理中的视觉思维链

VChain: Chain-of-Visual-Thought for Reasoning in Video Generation

摘要

Support