The paper introduces GraftOOD, a method that improves video reward optimization by grafting out‑of‑distribution video exemplars into advantage‑weighted rectification. By replacing parts of online rollouts with exemplar supervision, GraftOOD addresses motion collapse and achieves significant gains in dynamic degree and other metrics on the Wan2.1-1.3B benchmark. The authors also release GraftSet‑Video, a dataset of 9,588 exemplars for 2,397 prompts, and provide code for reproducibility.
By Zheming Fu, Ruizhe He, Wenbin Ning, Xiaoxiao Ma, Xuan Yang, Lianrui Mu, Haoyang Huang, Chang Liu, Lei Wang, Siming Fu
arXiv:2608.29904v1 Announce Type: new
Abstract: Modern video generators routinely fail at physical dynamics: objects float, trajectories violate gravity, contacts vanish. Standard denoising and flow-...
By Hai Nguyen-Truong, Tuan-Anh Vu, Dang Huynh
arXiv:2606. 30376v1 Announce Type: new Abstract: Aligning generative flow models on continuous spaces via online reinforcement learning is constrained by intractable trajectory likelihoods.
By Zheming Fu, Ruizhe He, Wei Shang, Xiaoxiao Ma, Lei Wang, Chang Liu, Siming Fu
arXiv:2609. 03241v1 Announce Type: cross Abstract: A reasoning model can improve from its own on-policy experience, but this inner loop is fragile: terminal verifiers provide reliable yet sparse supervision, while dense same-model guidance can reinforce false confidence or overconcentrate learning on a narrow solution mode.
By Zixun Huang, Kishan Panaganti, Haitao Mi, Leowei Liang
arXiv:2610.03691v1 Announce Type: new
Abstract: We present FlowHMR, a framework for recovering physically plausible global 3D human motion from monocular video. Previous learning-based methods typica...
By Zhanke Wang, Chengfeng Zhao, Qing Shuai, Jingzhong Lin, Heng Li, Zeyu Ling, Yuxin Wen, Jing Li, Di Kang, Chunchao Guo, Linchao Bao
arXiv:2609.39563v1 Announce Type: new
Abstract: Existing video language models encode sampled RGB frames independently, so a long video must either exhaust the token budget or drop the changes betwee...
By Can Zhang, Xiaotian Han, Junyuan Shang, Yuchen Ding, Zhenyu Zhang, Shuohuan Wang, Dianhai Yu, Ruirui Li
The paper introduces Recency Forcing, a technique that addresses the long‑horizon degradation in autoregressive video generation caused by KV eviction mismatch. By applying a timestep‑dependent bias—Temporal Response Bias—derived from a positional response measure, the method reduces the influence of distant frames during inference without altering context length or training objectives. An exact reformulation, Biased Attention Reparameterization, enables this bias to be applied as a standard FlashAttention call with zero overhead, achieving state‑of‑the‑art long‑horizon generation quality on VBench datasets.
By Tri Cao, Hung Nguyen, Phong Nguyen, Khoi Nguyen
We present FlowHMR, a framework for recovering physically plausible global 3D human motion from monocular video. Previous learning-based methods typically regress human motion directly from video and...
LinMorph proposes a learnable softmax‑to‑linear weight transformation for Video Diffusion Transformers, replacing traditional random initialization or manual weight reuse. By learning a shared morphism that maps each target Gated DeltaNet parameter to a combination of pretrained query, key, value, and output projections, it generates layer‑specific initializations that preserve pretrained knowledge. The method enables faster recovery and higher generation quality across pretrained Video DiTs, especially when applied to larger video scales after training on short, low‑resolution clips.
By Lingyi Long, Congyue Zhao, Xiaoou Li, Yanmin Qian, Tao Huang
DiffusionART introduces a new reinforcement learning framework for diffusion models that aggregates prompt rewards into a shared direction, eliminating consistency forces inherent in prior methods. It employs a trust‑region‑masked second update for drift‑adaptive restoration, removing the need for EMA tuning and differentiable reward models. Experiments on image and video generation show that DiffusionART trains more efficiently and stably, achieving higher mean rewards with significantly fewer GPU hours compared to existing approaches.
By Zhenglin Wan, Jiayu Chen, Zhengfeiyang Zhang, Pengfei Zhou, Wangbo Zhao, Fangyuan Kong, Pengfei Wan, Kun Gai, Yang You
arXiv:2606. 27771v1 Announce Type: new Abstract: Reinforcement learning (RL) post-training improves the reward alignment of flow-based generators, but often degrades perceptual quality in ways that are not captured by the reward proxy.
By Tianlin Pan, Lianyu Pang, Cheng Da, Huan Yang, Changqian Yu, Kun Gai, Wenhan Luo
arXiv:2606. 04775v1 Announce Type: cross Abstract: Text-to-video (T2V) models trained on large-scale web data can generate undesired content, motivating interventions that reduce harmful outputs without sacrificing visual quality.
By Jihoon Hong, Alice Chan, Qiyue Dai, Julian Skifstad, Glen Chou