Pith. sign in

REVIEW 35 cited by

Video World Models with Long-term Spatial Memory

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2506.05284 v1 pith:BXELDHNB submitted 2025-06-05 cs.CV

Video World Models with Long-term Spatial Memory

classification cs.CV
keywords long-termmemorymodelsworldconsistencymechanismsspatialvideo
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Emerging world models autoregressively generate video frames in response to actions, such as camera movements and text prompts, among other control signals. Due to limited temporal context window sizes, these models often struggle to maintain scene consistency during revisits, leading to severe forgetting of previously generated environments. Inspired by the mechanisms of human memory, we introduce a novel framework to enhancing long-term consistency of video world models through a geometry-grounded long-term spatial memory. Our framework includes mechanisms to store and retrieve information from the long-term spatial memory and we curate custom datasets to train and evaluate world models with explicitly stored 3D memory mechanisms. Our evaluations show improved quality, consistency, and context length compared to relevant baselines, paving the way towards long-term consistent world generation.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 35 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. MemLearner: Learning to Query Context memory for Video World Models

    cs.CV 2026-06 unverdicted novelty 7.0

    MemLearner introduces a learning-based adaptive context query method using query tokens in video world models to improve long-term scene consistency over rule-based retrieval.

  2. What-If World: A Causal Benchmark for General World Models in Embodied Scenarios

    cs.CV 2026-05 unverdicted novelty 7.0

    What-If World is a new paired-prompt benchmark showing that nine state-of-the-art video generation models achieve at most 52% on causal intervention tests and cluster near 28% for open-source systems.

  3. Latent State Design for World Models under Sufficiency Constraints

    cs.AI 2026-05 unverdicted novelty 7.0

    World models succeed when their latent states are built to meet task-specific sufficiency constraints rather than preserving the maximum amount of information.

  4. MultiWorld: Scalable Multi-Agent Multi-View Video World Models

    cs.CV 2026-04 unverdicted novelty 7.0

    MultiWorld is a scalable framework for multi-agent multi-view video world models that improves controllability and consistency over single-agent baselines in game and robot tasks.

  5. World from Motion: Generative Dynamic Gaussian Reconstruction from Monocular Video

    cs.CV 2026-07 unverdicted novelty 6.0

    A generative video model conditioned on pixel-aligned 3D renderings produces consistent dynamic 3D Gaussian splats from monocular video and sets new SOTA in 4D reconstruction.

  6. Pano2World: End-to-End 3D Generation via Unified Multi-View Sequences

    cs.CV 2026-07 unverdicted novelty 6.0

    Pano2World generates an explorable 3D Gaussian scene directly from a single indoor panorama via coarse proxy rendering, view-aware joint denoising, and a latent feature adapter.

  7. Pano2World: End-to-End 3D Generation via Unified Multi-View Sequences

    cs.CV 2026-07 conditional novelty 6.0

    A single indoor panorama is converted end-to-end into an explorable 3D Gaussian scene via joint multi-view panoramic diffusion and a latent feature adapter that bypasses RGB re-encoding.

  8. PermaVid: Consistent Video Generation Across Edits via Disentangled Context Memory

    cs.CV 2026-06 unverdicted novelty 6.0

    PermaVid disentangles spatial context into semantic appearance and geometric structure via multi-modal memory banks and edit-aware updates to maintain long-term consistency in video generation after edits.

  9. Envision4D: Envisioning Visual Futures via Feed-forward 4D Gaussian Splatting for Autonomous Driving

    cs.CV 2026-06 unverdicted novelty 6.0

    Envision4D presents a feed-forward 4D Gaussian Splatting framework with future pose prediction, temporal attention, and conditioned motion lifting for pose-free extrapolation in autonomous driving scenes.

  10. Latent Spatial Memory for Video World Models

    cs.CV 2026-06 unverdicted novelty 6.0

    Mirage stores and queries 3D scene information in diffusion latent space via depth-guided lifting and warping, yielding 10.57× faster generation and 55× smaller memory than explicit RGB point-cloud baselines while rea...

  11. Echo-Memory: A Controlled Study of Memory in Action World Models

    cs.CV 2026-06 unverdicted novelty 6.0

    A controlled study finds that block-wise state-space recurrence outperforms other memory designs for open-domain scene return in action-conditioned video models, and that standard replay metrics do not adequately meas...

  12. MetaWorld: Scaling Multi-Agent Video World Model from Single-view Video Data

    cs.CV 2026-06 unverdicted novelty 6.0

    MetaWorld scales multi-agent video world models from single-view videos using monocular decomposition into ego-motion and trajectories, subject-aware generation, and cross-attention alignment for consistency.

  13. Geometry-Aware Implicit Memory for Video World Models

    cs.CV 2026-06 unverdicted novelty 6.0

    GIM-World adds a camera-queryable geometry distillation head and pruning rule to implicit memory in video world models, claiming better long-horizon geometric consistency on the MIND benchmark than explicit and implic...

  14. Robust Dreamer: Deviation-Aware Latent Gaussian Memory for Action-Controlled AR Video Generation

    cs.CV 2026-05 unverdicted novelty 6.0

    Robust Dreamer uses Latent Gaussian Memory anchored to diffusion latents and Deviation Learning with a Dynamic Deviation Archive to reduce drift in long-horizon action-controlled image-to-video generation, reporting S...

  15. E$^3$C: Video Generation with 3D Environmental Memory and Ego-Exo Human Pose Control

    cs.CV 2026-05 unverdicted novelty 6.0

    E³C is a video diffusion model that disentangles persistent 3D scene structure via point-cloud memory from human dynamics via ego-exo pose controls for improved egocentric video generation on the Nymeria dataset.

  16. Remember to be Curious: Episodic Context and Persistent Worlds for 3D Exploration

    cs.LG 2026-05 unverdicted novelty 6.0

    A curiosity-based 3D exploration policy that pairs persistent online 3D reconstruction with episodic sequence modeling over RGB to outperform active-mapping baselines on HM3D and transfer zero-shot to Gibson and synth...

  17. GeoFlow: Enforcing Implicit Geometric Consistency in Video Generation

    cs.CV 2026-05 unverdicted novelty 6.0

    GeoFlow adds a geometry-consistency reward based on rigid camera flow and object appearance preservation, integrated via reinforcement fine-tuning to improve geometric coherence in video generation.

  18. Warp-as-History: Generalizable Camera-Controlled Video Generation from One Training Video

    cs.CV 2026-05 unverdicted novelty 6.0

    Warp-as-History enables zero-shot camera trajectory following in frozen video models by supplying camera-warped pseudo-history, with single-video LoRA fine-tuning improving generalization to unseen videos.

  19. SWIFT: Prompt-Adaptive Memory for Efficient Interactive Long Video Generation

    cs.CV 2026-05 unverdicted novelty 6.0

    SWIFT introduces a semantic injection cache with head-wise updates and an adaptive dynamic window plus segment anchors to achieve efficient multi-prompt long video generation at 22.6 FPS while preserving quality in ca...

  20. CityRAG: Stepping Into a City via Spatially-Grounded Video Generation

    cs.CV 2026-04 conditional novelty 6.0

    PlayCoder combines a repository-aware coding agent with a vision-based GUI testing agent and an automated program repair loop to detect and fix silent logic errors in LLM-generated interactive application code.

  21. CityRAG: Stepping Into a City via Spatially-Grounded Video Generation

    cs.CV 2026-04 unverdicted novelty 6.0

    CityRAG generates minutes-long 3D-consistent videos of real-world cities by grounding outputs in geo-registered data and using temporally unaligned training to disentangle fixed scenes from transient elements like weather.

  22. From Synchrony to Sequence: Exo-to-Ego Generation via Interpolation

    cs.CV 2026-04 unverdicted novelty 6.0

    Interpolating exo and ego videos into a single continuous sequence lets diffusion sequence models generate more coherent first-person videos than direct conditioning, even without pose interpolation.

  23. From Synchrony to Sequence: Exo-to-Ego Generation via Interpolation

    cs.CV 2026-04 conditional novelty 6.0

    Interpolating only the video frames between synchronized exo and ego clips already turns discontinuous cross-view generation into continuous sequence modeling and measurably improves diffusion-based Exo2Ego synthesis.

  24. Lyra 2.0: Explorable Generative 3D Worlds

    cs.CV 2026-04 unverdicted novelty 6.0

    Lyra 2.0 produces persistent 3D-consistent video sequences for large explorable worlds by using per-frame geometry for information routing and self-augmented training to correct temporal drift.

  25. Rein3D: Reinforced 3D Indoor Scene Generation with Panoramic Video Diffusion Models

    cs.CV 2026-04 unverdicted novelty 6.0

    Rein3D generates photorealistic, globally consistent 3D indoor scenes by using a restore-and-refine process where radial panoramic videos are restored via diffusion models and then used to update a 3D Gaussian field.

  26. Geometry Forcing: Marrying Video Diffusion and 3D Representation for Consistent World Modeling

    cs.CV 2025-07 unverdicted novelty 6.0

    Geometry Forcing aligns video diffusion representations with geometric foundation model features via angular cosine and scale regression objectives to improve 3D consistency in generated videos.

  27. Directing the World: Fast Autoregressive Video Generation with Compositional Human-Camera Control

    cs.CV 2026-06 unverdicted novelty 5.0

    A decoupled-control autoregressive video model using Fast-Slow Memory training, dynamic projection, and staged camera control to produce stable long-horizon outputs with human and viewpoint guidance.

  28. WorldOlympiad: Can Your World Model Survive a Triathlon?

    cs.CV 2026-06 unverdicted novelty 5.0

    WorldOlympiad is a new benchmark decomposing world-model evaluation into physical, geometry, and interaction tracks using segmentation, MLLM judges, Gaussian splatting, and action prompts on diverse scenarios.

  29. DecMem: Towards Minute-Long Consistent World Generation with Decoupled Memory

    cs.CV 2026-05 unverdicted novelty 5.0

    DecMem proposes a decoupled memory system using sparse global and anchored local components to enable consistent minute-long controllable video generation in world models.

  30. World-R1: Reinforcing 3D Constraints for Text-to-Video Generation

    cs.CV 2026-04 unverdicted novelty 5.0

    World-R1 applies RL via Flow-GRPO on a new text dataset for world simulation to enforce 3D constraints in video generation while preserving visual quality.

  31. World-R1: Reinforcing 3D Constraints for Text-to-Video Generation

    cs.CV 2026-04 unverdicted novelty 5.0

    World-R1 uses Flow-GRPO reinforcement learning and a new text dataset to enforce 3D consistency in text-to-video generation while keeping the original model's visual quality.

  32. InSpatio-WorldFM: An Open-Source Real-Time Generative Frame Model

    cs.CV 2026-03 unverdicted novelty 5.0

    InSpatio-WorldFM is a frame-independent generative model that uses explicit 3D anchors and spatial memory to deliver real-time multi-view consistent spatial intelligence via a three-stage training pipeline from pretra...

  33. World-R1: Reinforcing 3D Constraints for Text-to-Video Generation

    cs.CV 2026-04 unverdicted novelty 4.0

    World-R1 applies reinforcement learning via Flow-GRPO and a text dataset to align text-to-video models with 3D constraints from pre-trained foundation models, improving consistency while keeping original visual quality.

  34. World-R1: Reinforcing 3D Constraints for Text-to-Video Generation

    cs.CV 2026-04 unverdicted novelty 4.0

    World-R1 uses RL with 3D model feedback and a new text dataset to improve geometric consistency in text-to-video generation while keeping the base model unchanged.

  35. Evolution of Video Generative Foundations

    cs.CV 2026-04 unverdicted novelty 2.0

    This survey traces video generation technology from GANs to diffusion models and then to autoregressive and multimodal approaches while analyzing principles, strengths, and future trends.