mlsanityData to 14 Oct 2026

WorldPlay: Towards Long-Term Geometric Consistency for Real-Time Interactive World Modeling

arXiv 2512.14614 ↗First posted 16 Dec 2025cs.CV cs.GRcited by 170

On the map: World models → Video-generation world models and simulators

also in 1 more
Abstract (arXiv)

This paper presents WorldPlay, a streaming video diffusion model that enables real-time, interactive world modeling with long-term geometric consistency, resolving the trade-off between speed and memory that limits current methods. WorldPlay draws power from three key ingredients. 1) We use a Dual Action Representation to enable robust action control in response to the user's keyboard and mouse inputs. 2) To enforce long-term consistency, our Reconstituted Context Memory dynamically rebuilds context from past frames and uses temporal reframing to keep geometrically important but long-past frames accessible, effectively alleviating memory attenuation. 3) We also propose Context Forcing, a novel distillation method designed for memory-aware model. Aligning memory context between the teacher and student preserves the student's capacity to use long-range information, enabling real-time speeds while preventing error drift. Taken together, WorldPlay generates long-horizon streaming 720p video at 24 FPS with superior consistency, comparing favorably with existing techniques and showing strong generalization across diverse scenes. Project page and online demo can be found: https://3d-models.hunyuan.tencent.com/world/ and https://3d.hunyuan.tencent.com/sceneTo3D.

Key facts. Cited by 170 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 87 of its 100 references that are in the map. First posted on arXiv 16 Dec 2025.

Builds on

87 of 100 references.

Closest ancestors · 17

Same direction, newest first

  1. 2025RELIC: Interactive Video World Model with Long-Horizon Memory3 Dec
  2. 2025Hunyuan-GameCraft-2: Instruction-following Interactive Game World Model28 Nov
  3. 2025Rolling Forcing: Autoregressive Long Video Diffusion in Real Time29 Sep
  4. 2025Matrix-game 2.0: An open-source real-time and streaming interactive world model18 Aug
  5. 2025Yume: An Interactive World Generation Model23 Jul
  6. 2025VRAG: Learning World Models for Interactive Video Generation28 May
  7. 2025WorldMem: Long-term Consistent World Simulation with Memory16 Apr
  8. 2025Wan: Open and Advanced Large-Scale Video Generative Models26 Mar
  9. 2025GameFactory: Creating New Games with Generative Interactive Videos14 Jan
  10. 2024GameGen-X: Interactive Open-world Game Video Generation1 Nov
  11. 2024Diffusion Models Are Real-Time Game Engines27 Aug
  12. 2024CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer12 Aug
5 more
  1. 2024StreamingT2V: Consistent, Dynamic, and Extendable Long Video Generation from Text21 Mar
  2. 2022Scalable Diffusion Models with Transformers19 Dec
  3. 2022Flow Matching for Generative Modeling6 Oct
  4. 2020Score-Based Generative Modeling through Stochastic Differential Equations26 Nov
  5. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 70

  1. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  2. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,226
  3. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,852
  4. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
  5. 2023Efficient Streaming Language Models with Attention Sinkscited by 2,186
  6. 2025InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Modelscited by 1,834
  7. 2023VBench: Comprehensive Benchmark Suite for Video Generative Modelscited by 1,822
  8. 2024HunyuanVideo: A Systematic Framework For Large Video Generative Modelscited by 1,660
  9. 2023DUSt3R: Geometric 3D Vision Made Easycited by 1,567
  10. 2023AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuningcited by 1,433
  11. 2023ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillationcited by 1,219
  12. 2023One-step Diffusion with Distribution Matching Distillationcited by 1,019
38 more
  1. 2023Adversarial Diffusion Distillationcited by 759
  2. 2024Improved Distribution Matching Distillation for Fast Image Synthesiscited by 684
  3. 2024Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusioncited by 681
  4. 2025Self Forcing: Bridging the Train-Test Gap in Autoregressive Video Diffusioncited by 625
  5. 2023MotionCtrl: A Unified and Flexible Motion Controller for Video Generationcited by 580
  6. 2024VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Modelscited by 579
  7. 2025Mean Flows for One-step Generative Modelingcited by 561
  8. 2023DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Visioncited by 537
  9. 2024CameraCtrl: Enabling Camera Control for Text-to-Video Generationcited by 460
  10. 2024From Slow Bidirectional to Fast Autoregressive Video Diffusion Modelscited by 433
  11. 2024ViewCrafter: Taming Video Diffusion Models for High-fidelity Novel View Synthesiscited by 403
  12. 2024One Step Diffusion via Shortcut Modelscited by 391
  13. 2025GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Controlcited by 341
  14. 2024Diffusion for World Modeling: Visual Details Matter in Ataricited by 300
  15. 2024Fast High-Resolution Image Synthesis with Latent Adversarial Diffusion Distillationcited by 279
  16. 2024Navigation World Modelscited by 272
  17. 2024SDXL-Lightning: Progressive Adversarial Diffusion Distillationcited by 266
  18. 2025LongLive: Real-time Interactive Long Video Generationcited by 265
  19. 2025Seedance 1.0: Exploring the Boundaries of Video Generation Modelscited by 259
  20. 2025Video models are zero-shot learners and reasonerscited by 220
  21. 2024WonderWorld: Interactive 3D Scene Generation from a Single Imagecited by 202
  22. 2025Self-Forcing++: Towards Minute-Scale High-Quality Video Generationcited by 192
  23. 2025Stable Virtual Camera: Generative View Synthesis with Diffusion Modelscited by 182
  24. 2025ViPE: Video Pose Engine for 3D Geometric Perceptioncited by 170
  25. 2025Context as Memory: Scene-Consistent Interactive Long Video Generation with Memory Retrievalcited by 163
  26. 2025Difix3D+: Improving 3D Reconstructions with Single-Step Diffusion Modelscited by 161
  27. 2024SageAttention: Accurate 8-Bit Attention for Plug-and-play Inference Accelerationcited by 160
  28. 2023WonderJourney: Going from Anywhere to Everywherecited by 158
  29. 2025WorldScore: A Unified Evaluation Benchmark for World Generationcited by 156
  30. 2024AC3D: Analyzing and Improving 3D Camera Control in Video Diffusion Transformerscited by 154
  31. 2025Diffusion Adversarial Post-Training for One-Step Video Generationcited by 151
  32. 2024ReconX: Reconstruct Any Scene from Sparse Views with Video Diffusion Modelcited by 143
  33. 2025Frame Context Packing and Drift Prevention in Next-Frame-Prediction Video Diffusion Modelscited by 143
  34. 2024DimensionX: Create Any 3D and 4D Scenes from a Single Image with Controllable Video Diffusioncited by 140
  35. 2024FIFO-Diffusion: Generating Infinite Videos from Text without Trainingcited by 137
  36. 2021Sequence Parallelism: Long Sequence Training from System Perspectivecited by 126
  37. 2025TrajectoryCrafter: Redirecting Camera Trajectory for Monocular Videos via Diffusion Modelscited by 125
  38. 2025VMem: Consistent Interactive Video Scene Generation with Surfel-Indexed View Memorycited by 109

50 most cited of 70 All 70 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 169.

170 papers.

Sort:

By area:

  1. 2026Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generationcited by 184
  2. 2026Advancing Open-source World Modelscited by 154
  3. 2026DreamDojo: A Generalist Robot World Model from Large-Scale Human Videoscited by 135
  4. 2026Matrix-Game 3.0: Real-Time and Streaming Interactive World Model with Long-Horizon Memorycited by 100
  5. 2026Being-H0.7: A Latent World-Action Model from Egocentric Videoscited by 75
  6. 2026Helios: Real Real-Time Long Video Generation Modelcited by 65
  7. 2026Context Forcing: Consistent Autoregressive Video Generation with Long Contextcited by 59
  8. 2026Causal Forcing++: Scalable Few-Step Autoregressive Diffusion Distillation for Real-Time Interactive Video Generationcited by 56
  9. 2026SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion Transformercited by 55
  10. 2026Infinite Worlds with Versatile Interactionscited by 48
  11. 2026HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worldscited by 44
  12. 2026DreamX-World 1.0: A General-Purpose Interactive World Modelcited by 40
38 more
  1. 2026MIND: Benchmarking Memory Consistency and Action Control in World Modelscited by 35
  2. 2026LiveWorld: Simulating Out-of-Sight Dynamics in Generative Video World Modelscited by 28
  3. 2026MemRoPE: Training-Free Infinite Video Generation via Evolving Memory Tokenscited by 27
  4. 2026MosaicMem: Hybrid Spatial Memory for Controllable Video World Modelscited by 27
  5. 2026Out of Sight but Not Out of Mind: Hybrid Memory for Dynamic Video World Modelscited by 26
  6. 2026WorldMark: A Unified Benchmark Suite for Interactive Video World Modelscited by 26
  7. 2026WorldCompass: Reinforcement Learning for Long-Horizon World Modelscited by 24
  8. 2026Quant VideoGen: Auto-Regressive Long Video Generation via 2-Bit KV-Cache Quantizationcited by 23
  9. 2026RealWonder: Real-Time Physical Action-Conditioned Video Generationcited by 22
  10. 2026INSPATIO-WORLD: A Real-Time 4D World Simulator via Spatiotemporal Autoregressive Modelingcited by 21
  11. 2026minWM: A Full-Stack Open-Source Framework for Real-Time Interactive Video World Modelscited by 19
  12. 2026Omni-WorldBench: Towards a Comprehensive Interaction-Centric Evaluation for World Modelscited by 18
  13. 2026LongLive-RAG: A General Retrieval-Augmented Framework for Long Video Generationcited by 16
  14. 2026Pathwise Test-Time Correction for Autoregressive Long Video Generationcited by 15
  15. 2026AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memoriescited by 15
  16. 2026HiAR: Efficient Autoregressive Long Video Generation via Hierarchical Denoisingcited by 15
  17. 2026WorldKV: Efficient World Memory with World Retrieval and Compressioncited by 15
  18. 2026One-Forcing: Towards Stable One-Step Autoregressive Video Generationcited by 15
  19. 2026Gamma-World: Generative Multi-Agent World Modeling Beyond Two Playerscited by 15
  20. 2026Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligencecited by 15
  21. 2026Addressable Memory for Video World Modelscited by 15
  22. 2026Agentic World Modeling: Foundations, Capabilities, Laws, and Beyondcited by 14
  23. 2026iWorld-Bench: A Benchmark for Interactive World Models with a Unified Action Generation Frameworkcited by 14
  24. 2026ActWorld: From Explorable to Interactive World Model via Action-Aware Memorycited by 14
  25. 2026Alaya-EVOKE: From Linear-Scaling Supervision to Endless Worldcited by 14
  26. 2026Code World Model: Coding Agent as World Braincited by 14
  27. 2026Out of Sight, Out of Mind? Evaluating State Evolution in Video World Modelscited by 13
  28. 2026SCOPE: Simulating Cross-game Operations in Playable Environments for FPS World Modelscited by 13
  29. 2026Geometry-Aware Implicit Memory for Video World Modelscited by 13
  30. 2026EchoWM: Open and Enterable Omnimodal World Modelscited by 13
  31. 2026Current World Models Lack a Persistent State Corecited by 12
  32. 2026SolarWM: Open Data and Scalable Training for Long-Horizon Video World Modelscited by 12
  33. 2026GigaWorld-1: A Roadmap to Build World Models for Robot Policy Evaluationcited by 11
  34. 2026Olaf-World: Orienting Latent Actions for Video World Modelingcited by 10
  35. 2026ShareVerse: Multi-Agent Consistent Video Generation for Shared World Modelingcited by 10
  36. 2026SVG-EAR: Parameter-Free Linear Compensation for Sparse Video Generation via Error-aware Routingcited by 10
  37. 2026ACWM-Phys: Investigating Generalized Physical Interaction in Action-Conditioned Video World Modelscited by 10
  38. 2026Incantation: Natural Language as the Action Interface for Multi-Entity Video World Modelscited by 10

50 most cited of 170 All 170 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Video World Models with Long-term Spatial Memory
  2. 2026Beyond Pixel Histories: World Models with Persistent 3D State
  3. 2025MagicWorld: Towards Long-Horizon Stability for Interactive Video World Exploration
  4. 2025WorldPack: Dynamic Frame Compression for Long-context Video World Modeling
  5. 2026LIVE: Long-horizon Interactive Video World Modeling
  6. 2026Infinite-World: Scaling Interactive World Models to 1000-Frame Horizons via Pose-Free Hierarchical Memory
  7. 2026AlayaWorld: Interactive Long-Horizon World Modeling -- Full Technical Report
  8. 2026UCM: Unified Modeling of Camera Control and Memory with Time-aware Positional Encoding Warping for World Models