9 papers
OmniX: Any-view and Any-time 4D Reconstruction via Feed-forward Trajectory Fields
Yanqin Jiang, Tengfei Wang, Zhengwei Wang +6
Previous feed-forward 4D reconstruction methods either predict per-frame static point clouds, ignoring foreground motion, or estimate point cloud trajectories while being limited t…
WorldPlay: Towards Long-Term Geometric Consistency for Real-Time Interactive World Modeling
Wenqiang Sun, Haiyu Zhang, Haoyuan Wang +7
This paper presents WorldPlay, a streaming video diffusion model that enables real-time, interactive world modeling with long-term geometric consistency, resolving the trade-off be…
Pathwise Test-Time Correction for Autoregressive Long Video Generation
Xunzhi Xiang, Zixuan Duan, Guiyu Zhang +7
Distilled autoregressive diffusion models facilitate real-time short video synthesis but suffer from severe error accumulation during long-sequence generation. While existing Test-…
Elastic Diffusion Transformer
Jiangshan Wang, Zeqiang Lai, Jiarui Chen +5
Diffusion Transformers (DiT) have demonstrated remarkable generative capabilities but remain highly computationally expensive. Previous acceleration methods, such as pruning and di…
WorldCompass: Reinforcement Learning for Long-Horizon World Models
Zehan Wang, Tengfei Wang, Haiyu Zhang +9
This work presents WorldCompass, a novel Reinforcement Learning (RL) post-training framework for the long-horizon, interactive video-based world models, enabling them to explore th…
MoCA: Mixture-of-Components Attention for Scalable Compositional 3D Generation
Zhiqi Li, Wenhuan Li, Tengfei Wang +8
Compositionality is critical for 3D object and scene generation, but existing part-aware 3D generation methods suffer from poor scalability due to quadratic global attention costs…