Paper Notes

Home

❯

Papers

❯

World Model

❯

Long Horizon Generation

文件夹: Papers/World-Model/Long-Horizon-Generation

此文件夹下有17条笔记。

  • 2026年5月

    SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion Transformer

  • 2026年5月

    Video Generation with Predictive Latents

  • 2026年3月

    Out of Sight but Not Out of Mind: Hybrid Memory for Dynamic Video World Models

  • 2026年1月

    VideoMemory: Toward Consistent Video Generation via Memory Integration

  • 2025年12月

    PFP: Pretraining Frame Preservation in Autoregressive Video Memory Compression

  • 2025年12月

    Resampling Forcing: End-to-End Training for Autoregressive Video Diffusion via Self-Resampling

  • 2025年10月

    Self-Forcing++: Towards Minute-Scale High-Quality Video Generation

  • 2025年9月

    Rolling Forcing: Autoregressive Long Video Diffusion in Real Time

  • 2025年8月

    MMPL: Macro-from-Micro Planning for High-Quality and Parallelized Autoregressive Long Video Generation

  • 2025年6月

    DeepVerse: 4D Autoregressive Video Generation as a World Model

  • 2025年6月

    Self Forcing: Bridging the Train-Test Gap in Autoregressive Video Diffusion

  • 2025年5月

    MAGI-1: Autoregressive Video Generation at Scale

  • 2025年5月

    RLVR-World: Training World Models with Reinforcement Learning

  • 2025年4月

    FramePack: Frame Context Packing and Drift Prevention in Next-Frame-Prediction Video Diffusion Models

  • 2025年4月

    SkyReels-V2: Infinite-length Film Generative Model

  • 2025年3月

    FAR: Long-Context Autoregressive Video Modeling with Next-Frame Prediction

  • 2024年7月

    Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion


Created with Quartz v4.5.2 © 2026

  • Source