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

arXiv: https://arxiv.org/abs/2603.25716 · HTML: https://mlsanity.com/paper/2603.25716/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- Placed in Video-generation world models and simulators, one of 11 directions in World models.
- Builds on 48 of its 53 references that are in the map.
- First posted on arXiv 26 Mar 2026.

## Placed in

- World models / [Video-generation world models and simulators](https://mlsanity.com/wm/video-generation-world-models-and-simulators/)

## Builds on (most cited first, 30 of 48)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer](https://mlsanity.com/paper/2408.06072/) (2024)
- [VBench: Comprehensive Benchmark Suite for Video Generative Models](https://mlsanity.com/paper/2311.17982/) (2023)
- [HunyuanVideo: A Systematic Framework For Large Video Generative Models](https://mlsanity.com/paper/2412.03603/) (2024)
- [MiniCPM-V: A GPT-4V Level MLLM on Your Phone](https://mlsanity.com/paper/2408.01800/) (2024)
- [Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion](https://mlsanity.com/paper/2407.01392/) (2024)
- [Open-Sora: Democratizing Efficient Video Production for All](https://mlsanity.com/paper/2412.20404/) (2024)
- [Self Forcing: Bridging the Train-Test Gap in Autoregressive Video Diffusion](https://mlsanity.com/paper/2506.08009/) (2025)
- [CameraCtrl: Enabling Camera Control for Text-to-Video Generation](https://mlsanity.com/paper/2404.02101/) (2024)
- [YOLOv11: An Overview of the Key Architectural Enhancements](https://mlsanity.com/paper/2410.17725/) (2024)
- [Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllability](https://mlsanity.com/paper/2405.17398/) (2024)
- [ReCamMaster: Camera-Controlled Generative Rendering from A Single Video](https://mlsanity.com/paper/2503.11647/) (2025)
- [Navigation World Models](https://mlsanity.com/paper/2412.03572/) (2024)
- [Matrix-game 2.0: An open-source real-time and streaming interactive world model](https://mlsanity.com/paper/2508.13009/) (2025)
- [Stable Virtual Camera: Generative View Synthesis with Diffusion Models](https://mlsanity.com/paper/2503.14489/) (2025)
- [History-Guided Video Diffusion](https://mlsanity.com/paper/2502.06764/) (2025)
- [WorldPlay: Towards Long-Term Geometric Consistency for Real-Time Interactive World Modeling](https://mlsanity.com/paper/2512.14614/) (2025)
- [WorldMem: Long-term Consistent World Simulation with Memory](https://mlsanity.com/paper/2504.12369/) (2025)
- [Context as Memory: Scene-Consistent Interactive Long Video Generation with Memory Retrieval](https://mlsanity.com/paper/2506.03141/) (2025)
- [VD3D: Taming Large Video Diffusion Transformers for 3D Camera Control](https://mlsanity.com/paper/2407.12781/) (2024)
- [WorldScore: A Unified Evaluation Benchmark for World Generation](https://mlsanity.com/paper/2504.00983/) (2025)
- [Frame Context Packing and Drift Prevention in Next-Frame-Prediction Video Diffusion Models](https://mlsanity.com/paper/2504.12626/) (2025)
- [Emu3.5: Native Multimodal Models are World Learners](https://mlsanity.com/paper/2510.26583/) (2025)
- [GameFactory: Creating New Games with Generative Interactive Videos](https://mlsanity.com/paper/2501.08325/) (2025)
- [GameGen-X: Interactive Open-world Game Video Generation](https://mlsanity.com/paper/2411.00769/) (2024)

## Built on by (most cited first, 26 of 26)

- [Next Forcing: Causal World Modeling with Multi-Chunk Prediction](https://mlsanity.com/paper/2606.11187/) (2026)
- [Geometry-Aware Implicit Memory for Video World Models](https://mlsanity.com/paper/2606.02436/) (2026)
- [Current World Models Lack a Persistent State Core](https://mlsanity.com/paper/2606.20545/) (2026)
- [Prisma-World: Camera-Controllable Multi-Agent Video World Model](https://mlsanity.com/paper/2606.09507/) (2026)
- [Towards Interactive Video World Modeling: Frontiers, Challenges, Benchmarks, and Future Trends](https://mlsanity.com/paper/2606.01164/) (2026)
- [Echo-Memory: A Controlled Study of Memory in Action World Models](https://mlsanity.com/paper/2606.09803/) (2026)
- [Compression and Retrieval: Implicit Memory Retrieval for Video World Models](https://mlsanity.com/paper/2606.23105/) (2026)
- [WorldDirector: Building Controllable World Simulators with Persistent Dynamic Memory](https://mlsanity.com/paper/2607.02517/) (2026)
- [PlayWorld: Benchmarking World Models with Agent Players over Long-Horizon Objectives](https://mlsanity.com/paper/2608.13552/) (2026)
- [From Pixels to States: Rethinking Interactive World Models as Game Engines](https://mlsanity.com/paper/2607.14076/) (2026)
- [RECAP-Forcing: Retaining Content Appearances for Long Video Generation](https://mlsanity.com/paper/2608.26671/) (2026)
- [DecMem: Towards Minute-Long Consistent World Generation with Decoupled Memory](https://mlsanity.com/paper/2605.31336/) (2026)
- [Closing the Loop: Training-Free Revisit Consistency for Autoregressive Generative Rendering](https://mlsanity.com/paper/2607.21848/) (2026)
- [WorldCrafter: Consistent Video World Model with Implicit 3D-aware Memory](https://mlsanity.com/paper/2609.24984/) (2026)
- [Can 4D Foundation Models Remember?](https://mlsanity.com/paper/2609.20819/) (2026)
- [Code Plans, Diffusion Renders: Open-Ended Generative World Modeling](https://mlsanity.com/paper/2609.26458/) (2026)
- [DynaTokens: Teaching Dynamics to Camera-Controlled Video Models at Test Time](https://mlsanity.com/paper/2609.35704/) (2026)
- [Honeycomb: Constant-Size Scene Memory Representation for Video World Models](https://mlsanity.com/paper/2609.37690/) (2026)
- [LOCI: Spatial Linear Memory for Streaming World Models](https://mlsanity.com/paper/2609.40222/) (2026)
- [4Director: Controlling Video World Models with Rigid 3D Geometry](https://mlsanity.com/paper/2610.02160/) (2026)
- [World Observer: Joint Actor-Observer Generation for Persistent World Modeling](https://mlsanity.com/paper/2610.02162/) (2026)
- [Kepler4D: Controllable Future Video Generation via 4D Scene State Evolution](https://mlsanity.com/paper/2610.04152/) (2026)
- [Artemis: Geometry-Grounded Multi-Agent Driving World Models with Shared 3D State and Progressive Memory Update](https://mlsanity.com/paper/2610.07031/) (2026)
- [4LC: 4D Latent Cache for Efficient Conditioning in Dynamic World Modeling](https://mlsanity.com/paper/2610.17625/) (2026)
- [Grid or Nodes? Probing Object Memory in World Models Under Partial Observability](https://mlsanity.com/paper/2610.18637/) (2026)
- [StateCraft: Augmenting Video World Models with Abstract-State Reasoning](https://mlsanity.com/paper/2610.19795/) (2026)

## Abstract (arXiv)

Video world models have shown immense potential in simulating the physical world, yet existing memory mechanisms primarily treat environments as static canvases. When dynamic subjects hide out of sight and later re-emerge, current methods often struggle, leading to frozen, distorted, or vanishing subjects. To address this, we introduce Hybrid Memory, a novel paradigm requiring models to simultaneously act as precise archivists for static backgrounds and vigilant trackers for dynamic subjects, ensuring motion continuity during out-of-view intervals. To facilitate research in this direction, we construct HM-World, the first large-scale video dataset dedicated to hybrid memory. It features 59K high-fidelity clips with decoupled camera and subject trajectories, encompassing 17 diverse scenes, 49 distinct subjects, and meticulously designed exit-entry events to rigorously evaluate hybrid coherence. Furthermore, we propose HyDRA, a specialized memory architecture that compresses memory into tokens and utilizes a spatiotemporal relevance-driven retrieval mechanism. By selectively attending to relevant motion cues, HyDRA effectively preserves the identity and motion of hidden subjects. Extensive experiments on HM-World demonstrate that our method significantly outperforms state-of-the-art approaches in both dynamic subject consistency and overall generation quality. Code is publicly available at https://github.com/H-EmbodVis/HyDRA.