arXiv:2607.09024v1 Announce Type: cross Abstract: Driven by next-token prediction, NLP shifted from task-specific models into powerful generalist foundation models. What, then, is the equivalent catalyst needed to achieve a general-purpose model in computer vision? In this paper,…
arXiv:2607.07763v1 Announce Type: new Abstract: World models are typically trained to predict discrete-time physical dynamics with a fixed step size baked into the model weights, preventing prediction at variable temporal resolutions. This matters for hierarchical planning, sim-t…
Driven by next-token prediction, NLP shifted from task-specific models into powerful generalist foundation models. What, then, is the equivalent catalyst needed to achieve a general-purpose model in computer vision? In this paper, we contend that large-scale text-to-video generat…
We propose OPSD-V, an on-policy self-distillation paradigm for post-training few-step autoregressive (AR) video diffusion models. Existing few-step AR video generators can produce long videos with low latency, but still suffer from error accumulation and weakened motion dynamics …
arXiv:2607.06631v1 Announce Type: cross Abstract: Video Diffusion Models (VDMs) have demonstrated superior generation quality but suffer from prohibitive computational costs. While recent few-step distillation techniques significantly accelerate inference, they typically enforce …
arXiv cs.LG
TIER_1Deutsch(DE)·Michael King, Aravindh Mahendran, Matthew Koichi Grimes, Fedor Kitashov, Adham Elarabawy, Pedro Velez, Maks Ovsjanikov, Viorica P\u{a}tr\u{a}ucean·
arXiv:2607.06856v1 Announce Type: cross Abstract: Prior work suggests that diffusion representations capture low-level geometry but struggle with high-level semantics. We demonstrate that state-of-the-art video diffusion models overcome this limitation. By systematically probing …
OPSD-V enhances few-step autoregressive video diffusion models by using real long-video data for temporal context during training, providing dense trajectory-level supervision that improves visual quality and motion dynamics without altering inference mechanisms.
arXiv:2606.23327v2 Announce Type: replace-cross Abstract: Video editing has become essential in digital media creation, yet existing automated systems are restricted to short segment processing and domain-specific tasks. They face two critical limitations: i) inability to handle …
CineMobile enables efficient image-to-video generation on mobile devices through distillation-guided pruning, diffusion distillation, and hybrid quantization techniques while maintaining visual quality and achieving significant speedup.
Vidu S1 is a real-time interactive video generation model that supports voice-controlled digital character animation with infinite-length output and high frame rate on consumer hardware.
arXiv cs.CV
TIER_1English(EN)·Thanh-Nhan Vo, Trong-Thuan Nguyen, Trung-Hoang Le, Tam V. Nguyen, Minh-Triet Tran·
arXiv:2607.08020v1 Announce Type: new Abstract: Autoregressive video diffusion enables efficient streaming and long-horizon video generation, but repeatedly reusing generated latents as causal context can amplify temporal errors, resulting in flickering, motion jitter, and struct…
arXiv cs.CV
TIER_1English(EN)·Utkarsh A. Mishra, Yongxin Chen, Danfei Xu, Yang Liu, Xi Chen, Jiayuan Mao·
arXiv:2607.08127v1 Announce Type: new Abstract: Generative video foundation models exhibit strong compositional priors, yet world-action models (WAMs) and video-action models (VAMs) often lose these priors after finetuning on robotic action data. We refer to this discrepancy as t…
arXiv:2607.08766v1 Announce Type: new Abstract: We propose OPSD-V, an on-policy self-distillation paradigm for post-training few-step autoregressive (AR) video diffusion models. Existing few-step AR video generators can produce long videos with low latency, but still suffer from …
arXiv:2512.01803v3 Announce Type: replace Abstract: Despite rapid advances in video generative models, robust metrics for evaluating visual and temporal correctness of complex human actions remain elusive. Critically, existing pure-vision encoders and Multimodal Large Language Mo…
arXiv cs.CV
TIER_1English(EN)·Zixin Guo, Yehonathan Litman, Yifeng He, John Miller, Chuhan Chen, Deva Ramanan·
arXiv:2607.08016v1 Announce Type: new Abstract: Video relighting requires balancing long-form temporal consistency with a physically grounded understanding of light transport, which depends on accurate estimation of intrinsic scene properties such as materials, geometry, and illu…
Driven by next-token prediction, NLP shifted from task-specific models into powerful generalist foundation models. What, then, is the equivalent catalyst needed to achieve a general-purpose model in computer vision? In this paper, we contend that large-scale text-to-video generat…
We propose OPSD-V, an on-policy self-distillation paradigm for post-training few-step autoregressive (AR) video diffusion models. Existing few-step AR video generators can produce long videos with low latency, but still suffer from error accumulation and weakened motion dynamics …
Generative video foundation models exhibit strong compositional priors, yet world-action models (WAMs) and video-action models (VAMs) often lose these priors after finetuning on robotic action data. We refer to this discrepancy as the video-action generalization gap. In this pape…
arXiv:2506.19348v2 Announce Type: replace Abstract: Training-free motion customization imposes motion patterns from reference videos onto video generators through test-time computation. Most existing methods target full diffusion models, requiring many denoising steps and high co…
Autoregressive video diffusion enables efficient streaming and long-horizon video generation, but repeatedly reusing generated latents as causal context can amplify temporal errors, resulting in flickering, motion jitter, and structural drift. In this paper, we investigate this f…
Video relighting requires balancing long-form temporal consistency with a physically grounded understanding of light transport, which depends on accurate estimation of intrinsic scene properties such as materials, geometry, and illumination. Existing methods follow two paradigms:…
Prior work suggests that diffusion representations capture low-level geometry but struggle with high-level semantics. We demonstrate that state-of-the-art video diffusion models overcome this limitation. By systematically probing their intermediate activations using recent mutual…
arXiv cs.CV
TIER_1English(EN)·Cong Wei, Quande Liu, Zixuan Ye, Qiulin Wang, Xintao Wang, Pengfei Wan, Kun Gai, Wenhu Chen·
arXiv:2510.08377v4 Announce Type: replace Abstract: Unified multimodal models have shown promising results in multimodal content generation and editing but remain largely limited to the image domain. In this work, we present UniVideo, a versatile framework that extends unified mo…
