用双向自回归框架让开源视频世界模型更高效可控
BiWM: Advancing Open-Source Interactive Video World Models with Bidirectional Autoregression
- 提出双向自回归训练框架，仅两阶段即可完成
- 8块H200卡几百度数内收敛，支持多模型与实时控制
- 适合需要高效交互式视频生成的研究者和开发者
交互式视频世界模型通常通过控制微调、自回归训练、因果初始化和少量步骤蒸馏，将双向视频生成器转为因果自回归系统。该流程成本高，且冻结的因果历史会累积误差，影响长时序保真度与可控性。本文提出BiWM，首个开源的双向自回归视频世界模型全栈训练框架。它在每个生成片段中保留完整注意力，仅需两个阶段：相机/动作控制微调和少量步骤分布匹配蒸馏（DMD）。在8块H200 GPU上，两阶段均在几百步优化器迭代内收敛。框架支持Wan2.1-T2V-1.3B、Wan2.2-TI2V-5B、HunyuanVideo-1.5-TI2V-8B和LTX-2.3-22B，具备真实相机控制、可插拔长历史压缩及可选低比特部署能力。监督与前向KL锚点有效缓解DMD模式崩溃，保持场景动态。BiWM提供从预训练双向视频模型到可交互、可控制、高效的生成世界模型的紧凑、可复现路径。
原文摘要 · Abstract (English)
Interactive video world models commonly convert bidirectional video generators into causal autoregressive systems through control fine-tuning, autoregressive training, causal initialization, and few-step distillation. This pipeline is costly, while frozen causal histories accumulate errors that degrade long-horizon fidelity and controllability. We present BiWM, the first open-source full-stack training framework for bidirectional autoregressive video world models. BiWM retains full attention within each generated chunk and requires only two stages: camera/action-control fine-tuning and few-step Distribution Matching Distillation (DMD). Both stages converge within a few hundred optimizer steps on 8 H200 GPUs. The framework supports Wan2.1-T2V-1.3B, Wan2.2-TI2V-5B, HunyuanVideo-1.5-TI2V-8B, and LTX-2.3-22B, together with real-world camera control, pluggable long-history compression, and optional low-bit deployment. Supervised and forward-KL anchors mitigate DMD mode collapse and preserve scene dynamics. BiWM provides a compact, reproducible path from pretrained bidirectional video models to interactive, controllable, and efficient world models.
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。