#video generation
23 papers · 1 filter
FeatFix: Reuse What You Verify through Local Exact-Feature Correction for Faster Cached Diffusion Inference
Hanshuai Cui, Zhiqing Tang, Zhi Yao +3
FeatFix reuses exact intermediate features computed for verification to locally correct draft outputs in cached diffusion inference, speeding up image and video generation while pr…
Mitigating Compounding Error via Video Representation Regularization
Taiye Chen, Qi Zhang, Yisen Wang
The paper studies why autoregressive video generation models accumulate errors over time and introduces a lightweight regularization that stabilizes hidden representations, reducin…
ContactFlow: A video action conditioning that transfers across embodiments
Sami Azirar, Enrico Pallotta, Jan Nogga +3
The paper introduces Contact Flow, an embodiment‑agnostic representation that encodes manipulation as the trajectory of 3D contact points, enabling a video‑based world model traine…
CineWeaver: Training-Free Reference-Controllable Multi-Shot Long Video Generation for Cinematic Storytelling
Yuyang Huang, Yabo Chen, Wenrui Dai +6
CineWeaver introduces a training-free method that modifies pretrained video diffusion models to generate long, multi-shot cinematic videos with fine-grained reference control and c…
Wonder: Video World Model Done Better
Jiacong Xu, Hanwen Jiang, Zhixin Shu +3
Wonder is a video world model that lets users explore a generated scene in real time by moving a virtual camera, using a dense coordinate conditioning and a sparse attention memory…
Hierarchical Denoising For Multi-Step Visual Reasoning
Zezhong Qian, Xiaowei Chi, Chak-Wing Mak +9
The paper introduces HDR, a hierarchical denoising framework for causal video generation that enables multi-step visual reasoning with low-latency streaming, achieving higher succe…