arXiv AI

Compress to Remember: Learning Compact Memory via On-Policy Distillation for Long Video Generation

The paper introduces Prediction‑Aligned Context Compaction (PACC), a method that learns a compact memory representation for long‑video generation by distilling a frozen video generator. PACC trains a compressor to aggregate past frames into memory tokens, using the generator as both teacher and student during on‑policy distillation. Experiments on MBench and VBench‑Long show that PACC improves memory‑event coverage and consistency, achieving better scores than strong baselines and producing competitive minute‑long videos.

arXiv Computer Vision
Aug 31

Relax Forcing: Relaxed KV-Memory for Consistent Long Video Generation

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
arXiv Computer Vision
Sep 24

The Past Frames the Future: Memory for Autoregressive Video Generation

arXiv:2609.28466v1 Announce Type: new Abstract: Advances in generative models have improved video fidelity, enabling long-horizon generation, interactive world modeling, and evolving visual environme...

By Harold Haodong Chen, Rongjin Guo, Disen Lan, Wen-Jie Shu, Hongfei Zhang, Hanzhe Hu, Shengtao Yao, Zixin Zhang, Guibin Zhang, Zhefan Rao, Jinxiu Liu, Yexin Liu, Rui Peng, Yuhao Liu, Bin Ren, Shuai Yang, Yukang Chen, Salman Khan, Ying-Cong Chen, Ser-Nam Lim, Rynson W. H. Lau, Nicu Sebe, Yu Cheng, Ming-Hsuan Yang, Qifeng Chen
arXiv Computer Vision
Aug 28

Ring Forcing: Towards Precise Long-Term Memory for Autoregressive Video Diffusion

Ring Forcing is an autoregressive video diffusion framework that enhances long‑term memory by enforcing retrieval from distant history through a ring‑structured training strategy. It introduces a compression and timestep composition method to extend effective historical span to minutes, and a sparse RoPE mechanism for scalable memory adaptation. Experiments show that Ring Forcing outperforms state‑of‑the‑art models in minutes‑long coherence and object permanence.

By Bowen Xue, Brandon Y. Feng, Chenguo Lin, Yuchen Lin, Yujia Zeng, Lvmin Zhang, Maneesh Agrawala, Honglei Yan, Panwang Pan
arXiv Computer Vision
6d ago

Video-Mirai: Autoregressive Video Diffusion Models Need Foresight

Video-Mirai is a training-only method that improves autoregressive video diffusion models by adding foresight. It uses a frozen foresight encoder to read completed rollouts and a lightweight predictor to distill future-informed targets into causal states, without altering the generator’s inference process. The approach raises the 5‑second VBench Total Score from 83.8 to 84.6 and improves subject and background consistency on 30‑second rollouts.

By Yonghao Yu, Lang Huang, Runyi Li, Zerun Wang, Toshihiko Yamasaki
arXiv Machine Learning
Sep 10

SimpleMemVLA: A Simple but Effective Native-Video Memory for Vision-Language-Action Models

arXiv:2609.05533v1 Announce Type: cross Abstract: Long-horizon manipulation is partially observable: the information needed to choose the next action may appear only in observations from minutes earl...

By Cheng Yin, Wang Xu, Junpeng Yang, Sikyuen Tam, Hanyu Liu, Yuan Yao, Xiangrui Zeng, Junbo Cui, Yequan Wang, Zhouping Yin, Yankai Lin
arXiv Computer Vision
Aug 31

LayerRecall: A State-Conditioned Memory Router for Long-Horizon Consistency in Video Generation

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