Learning to Retrieve: Internalizing Memory Retrieval for Video World Models

arXiv:2610.11444 · cs.CV · Submitted 2026-10-08 · Read on arXiv

cs.CV

Submitted: 2026-10-08

Updated: 2026-10-08

Project page: https://jkhu29.github.io/l2r

Terminology

• Consistency — 33× in this paper · explained in 3 episodes
• World Model — 29× in this paper · explained in 7 episodes
• Benchmark — 14× in this paper · explained in 2 episodes
• Transformer — 7× in this paper · explained in 2 episodes
• Ablation — 6× in this paper · explained in 1 episode
• Alignment — 6× in this paper · explained in 1 episode
• Diffusion Model — 4× in this paper · explained in 8 episodes
• Fidelity — 4× in this paper · explained in 2 episodes
• Accuracy — 3× in this paper · explained in 1 episode
• Diffusion Models — 3× in this paper · explained in 6 episodes
• Ablation Studies — 2× in this paper · explained in 1 episode
• Core — 2× in this paper · explained in 2 episodes
• Gated Linear Attention — 2× in this paper
• Generalization — 2× in this paper · explained in 13 episodes
• 3D Gaussian Splatting — 1× in this paper · explained in 2 episodes
• 3D Gaussian Splatting (3DGS) — 1× in this paper · explained in 4 episodes
• Adaptation — 1× in this paper · explained in 1 episode
• Attention Mechanism — 1× in this paper · explained in 3 episodes
• Diffusion Process — 1× in this paper · explained in 2 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Gaussian Splatting — 1× in this paper · explained in 4 episodes
• Generative Model — 1× in this paper · explained in 2 episodes
• Generative Models — 1× in this paper · explained in 3 episodes
• In International Conference on Machine Learning — 1× in this paper
• Latent Space — 1× in this paper · explained in 7 episodes
• Learned Perceptual Image Patch Similarity — 1× in this paper
• Point Clouds — 1× in this paper · explained in 2 episodes
• Reconstruction Error — 1× in this paper · explained in 2 episodes
• Sparse Attention — 1× in this paper · explained in 5 episodes
• Structural Similarity Index Measure — 1× in this paper
• Supervised Fine-Tuning (SFT) — 1× in this paper · explained in 8 episodes
• Tokenization — 1× in this paper · explained in 2 episodes

Sources

• Context Forcing: Consistent Autoregressive Video Generation with Long Context
• Echo-Memory: A Controlled Study of Memory in Action World Models
• Attend Locally, Remember Linearly: Linear Attention as Cross-Frame Memory for Autoregressive Video Diffusion
• Scaling Linear Attention with Sparse State Expansion
• Compression and Retrieval: Implicit Memory Retrieval for Video World Models
• HYDRA: Unifying Multi-modal Generation and Understanding via Representation-Harmonized Tokenization
• Wan: Open and Advanced Large-Scale Video Generative Models
• Why Do DiT Editors Drift? Plug-and-Play Low Frequency Alignment in VAE Latent Space
• Geometry-Aware Implicit Memory for Video World Models
• Wonder: Video World Model Done Better
• LongLive: Real-time Interactive Long Video Generation
• NeoVerse: Enhancing 4D World Model with in-the-wild Monocular Videos
• Deep Forcing: Training-Free Long Video Generation with Deep Sink and Participative Compression
• Context as Memory: Scene-Consistent Interactive Long Video Generation with Memory Retrieval
• MemLearner: Learning to Query Context memory for Video World Models
• MosaicMem: Hybrid Spatial Memory for Controllable Video World Models
• VideoSSM: Autoregressive Long Video Generation with Hybrid State-Space Memory
• HYDRA-X: Native Unified Multimodal Models with Holistic Visual Tokenizers
• RIFLEx: A Free Lunch for Length Extrapolation in Video Diffusion Transformers
• RealCam-Vid: High-resolution Video Dataset with Dynamic Scenes and Metric-scale Camera Movements

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models