LongVie 2: Multimodal Controllable Ultra-Long Video World Model
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908713211658240 |
|---|---|
| author | Gao, Jianxiong Chen, Zhaoxi Liu, Xian Zhuang, Junhao Xu, Chengming Feng, Jianfeng Qiao, Yu Fu, Yanwei Si, Chenyang Liu, Ziwei |
| author_facet | Gao, Jianxiong Chen, Zhaoxi Liu, Xian Zhuang, Junhao Xu, Chengming Feng, Jianfeng Qiao, Yu Fu, Yanwei Si, Chenyang Liu, Ziwei |
| contents | Building video world models upon pretrained video generation systems represents an important yet challenging step toward general spatiotemporal intelligence. A world model should possess three essential properties: controllability, long-term visual quality, and temporal consistency. To this end, we take a progressive approach-first enhancing controllability and then extending toward long-term, high-quality generation. We present LongVie 2, an end-to-end autoregressive framework trained in three stages: (1) Multi-modal guidance, which integrates dense and sparse control signals to provide implicit world-level supervision and improve controllability; (2) Degradation-aware training on the input frame, bridging the gap between training and long-term inference to maintain high visual quality; and (3) History-context guidance, which aligns contextual information across adjacent clips to ensure temporal consistency. We further introduce LongVGenBench, a comprehensive benchmark comprising 100 high-resolution one-minute videos covering diverse real-world and synthetic environments. Extensive experiments demonstrate that LongVie 2 achieves state-of-the-art performance in long-range controllability, temporal coherence, and visual fidelity, and supports continuous video generation lasting up to five minutes, marking a significant step toward unified video world modeling. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_13604 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LongVie 2: Multimodal Controllable Ultra-Long Video World Model Gao, Jianxiong Chen, Zhaoxi Liu, Xian Zhuang, Junhao Xu, Chengming Feng, Jianfeng Qiao, Yu Fu, Yanwei Si, Chenyang Liu, Ziwei Computer Vision and Pattern Recognition Building video world models upon pretrained video generation systems represents an important yet challenging step toward general spatiotemporal intelligence. A world model should possess three essential properties: controllability, long-term visual quality, and temporal consistency. To this end, we take a progressive approach-first enhancing controllability and then extending toward long-term, high-quality generation. We present LongVie 2, an end-to-end autoregressive framework trained in three stages: (1) Multi-modal guidance, which integrates dense and sparse control signals to provide implicit world-level supervision and improve controllability; (2) Degradation-aware training on the input frame, bridging the gap between training and long-term inference to maintain high visual quality; and (3) History-context guidance, which aligns contextual information across adjacent clips to ensure temporal consistency. We further introduce LongVGenBench, a comprehensive benchmark comprising 100 high-resolution one-minute videos covering diverse real-world and synthetic environments. Extensive experiments demonstrate that LongVie 2 achieves state-of-the-art performance in long-range controllability, temporal coherence, and visual fidelity, and supports continuous video generation lasting up to five minutes, marking a significant step toward unified video world modeling. |
| title | LongVie 2: Multimodal Controllable Ultra-Long Video World Model |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.13604 |