The paper introduces Relax Forcing, a training‑free memory mechanism for autoregressive video diffusion that structures temporal context into Sink, Tail, and History frames. By selecting History frames with a relaxation criterion, the method reduces error accumulation and attention overhead while preserving motion dynamics. Experiments on VBench‑Long demonstrate that this structured memory improves long‑video generation quality over existing baselines.
By Zengqun Zhao, Yanzuo Lu, Ziquan Liu, Jifei Song, Jiankang Deng, Ioannis Patras
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 during long autoregressive rollout.
LongVU‑TTT is a causal test‑time training method for long‑video multimodal large language models that inserts a convolutional resampler with fast‑weight updates between the vision encoder and the LLM. The fast weights adapt per video and contextualize frame features before compression, while a hybrid selector keeps explicit visual evidence for downstream reasoning. Experiments show that TTT‑Conv outperforms TTT‑MLP and bidirectional Mamba2 on MLVU, and beats attention‑ and fixed‑state recurrent resamplers on three benchmarks, achieving competitive results on five video‑understanding tasks after reducing 512 frames to 128 LLM frames.
By Mahmoud Ahmed, Sameh Abdulah, Olatunji Ruwase, Sam Ade Jacobs, Mathis Bode, Mohamed Elhoseiny
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:2602. 01801v2 Announce Type: replace-cross Abstract: Autoregressive video diffusion models enable streaming generation, opening the door to long-form synthesis, video world models, and interactive neural game engines.
By Dvir Samuel, Issar Tzachor, Matan Levy, Michael Green, Gal Chechik, Rami Ben-Ari
LayerRecall is a memory router for autoregressive video diffusion that selectively retrieves and injects historical key/value states into specific layers of the model, based on the current context. It addresses the problem that existing memory mechanisms expose nonlocal history but do not guarantee effective use, by recognizing that different layers prefer current, recent, or distant context. The method, combined with Cross‑Horizon Prediction Matching, achieves state‑of‑the‑art long‑range consistency on MemoBench and MovieBench while maintaining local continuity and incurring negligible inference overhead.
By Yixuan Ding, Jiahao Kong, Wei Huang, Ruijie Quan, Yi Yang
Long-video MLLMs must model temporal change before a limited visual-token budget removes most frame evidence. We introduce LongVU-TTT, which inserts a convolutional Test-Time Training (TTT) resampler...
RECAP-Forcing is a new method for long autoregressive video generation that addresses the memory challenge by organizing memory based on appearance novelty rather than recency. The approach retains key-value caches for newly appearing content—such as entering subjects, disoccluded regions, and new scenes—at the moment they first appear, ensuring consistent identities over time. It combines an attention sink for the initial scene with an optical-flow-based novelty bank for later frames, improving visual quality and semantic fidelity without adding learnable parameters.
By Haiyang Xu, Zheng Ding, Zhuowen Tu
arXiv:2608. 07408v1 Announce Type: cross Abstract: We study visual persistence in interactive video world models.
By Xindi Wu, Sven Elflein, James Lucas, Olga Russakovsky, Laura Leal-Taix\'e, Despoina Paschalidou, Jonathan Lorraine, Aljo\v{s}a O\v{s}ep
arXiv:2609.13012v1 Announce Type: new
Abstract: Vision-language models retain a substantial amount of pixel-decodable visual content in their visual key-value cache. We show, in our setting, that thi...
By Chenyu Zhou, Qiliang Jiang, Shuning Wu, Xu Zhou
arXiv:2607. 18759v1 Announce Type: new Abstract: Transformers with relative positional encodings often extrapolate to sequences longer than those seen during training, whereas transformers with learned absolute encodings typically do not.
By Subham Singh, Ashutosh Mishra, Subha Raut
arXiv:2606.06158v2 Announce Type: replace
Abstract: Adaptive video tokenisation seeks to dynamically allocate token budgets based on the underlying visual complexity of a sequence. Current continuous...
By Kevin Dave, Sai Aditya Patkuri, Chhaya Kumar Das, Gouranga Bala, Rajeshkumar SA, R. Venkatesh Babu