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.
hub Baseline reference
Spatialvid: A large-scale video dataset with spatial annotations.arXiv preprint arXiv:2509.09676, 2025a
Baseline reference. 50% of citing Pith papers use this work as a benchmark or comparison.
22
Pith papers citing it
Baseline
50% of classified citations
hub tools
citation-role summary
background 3
dataset 3
citation-polarity summary
fields
cs.CV 22years
2026 22representative citing papers
MBench is a new benchmark that quantifies long-term memory in video world models via three hierarchical consistency dimensions evaluated on curated real videos.
4DThinker enables VLMs to perform dynamic spatial reasoning by thinking with 4D latent mental imagery using new fine-tuning and reinforcement learning methods.
M²-REPA decouples modality-specific features from diffusion intermediates and aligns them to complementary expert foundation models via a multi-modal alignment loss and modality-specific decoupling regularization for improved multimodal video generation.
DPPE decouples rotation and translation in camera positional encodings for multi-view transformers to resolve late-stage training stagnation and improve generalization in novel view synthesis.
MotionAtlas supplies a 2,073-question benchmark, a self-bootstrap pipeline yielding 159k captions, and fine-tuned Video-MLLMs that deliver 5.2-point gains over Qwen3-VL-4B on motion tasks.
CaR uses attention with viewpoint positional encoding and context compression for flexible memory retrieval in video world models, backed by a new SceneFly dataset, and reports SOTA results with open-domain generalization.
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.
RayDer is a unified transformer backbone for self-supervised static-scene novel view synthesis that absorbs dynamic content as a nuisance factor and shows power-law scaling with data and compute while matching supervised methods in zero-shot settings.
minWM supplies an end-to-end pipeline that fine-tunes bidirectional T2V/TI2V models with camera control then distills them via Causal Forcing into few-step autoregressive generators for low-latency rollout.
VLMs possess a latent 3D scene topology subspace corresponding to Laplacian eigenmaps that can be causally shaped via Dirichlet energy regularization to improve spatial task performance by up to 12.1%.
Relit-LiVE jointly predicts relit videos and viewpoint-aligned environment maps inside a single diffusion process to achieve physically consistent video relighting without camera pose input.
CHAI framework pairs AI pre-captions with expert human critiques to produce precise video descriptions, enabling open models to outperform closed ones like Gemini-3.1-Pro and improve fine-grained control in video generation models.
ArtifactWorld restores artifacts in 3D Gaussian Splatting by training a video diffusion backbone on 107.5K paired clips with an isomorphic predictor for artifact heatmaps and an Artifact-Aware Triplet Fusion mechanism to achieve better sparse-view novel synthesis.
SceneScribe-1M is a new dataset of 1 million videos with semantic text, camera parameters, dense depth, and consistent 3D point tracks to support monocular depth estimation, scene reconstruction, point tracking, and text-to-video synthesis.
Scaling data, model size, and compute for local feature matching produces large performance gains on challenging benchmarks and a new manually annotated HardMatch dataset.
VGGT-Ω improves feed-forward reconstruction accuracy and efficiency by architectural simplifications, register-based attention, and training on much larger supervised and unlabeled video data.
SANA-WM is a 2.6B-parameter efficient world model that synthesizes minute-scale 720p videos with 6-DoF camera control, trained on 213K public clips in 15 days on 64 H100s and runnable on single GPUs at 36x higher throughput than prior open baselines.
ST-Gen4D uses a world model that fuses global appearance and local dynamic graphs into a 4D cognition representation to guide consistent 4D Gaussian generation.
Learning rotation invariance in descriptors matches the performance of matcher-level invariance but allows earlier invariance, faster matchers, and no loss in upright performance when trained at scale.
Matrix-Game 3.0 delivers 720p real-time video generation at 40 FPS with minute-scale memory consistency by combining residual self-correction training, camera-aware memory injection, and DMD-based autoregressive distillation on a 5B model.
LingBot-World is presented as an open-source world model that delivers high-fidelity simulation, minute-level contextual consistency, and real-time interactivity under one second latency.
citing papers explorer
-
MemLearner: Learning to Query Context memory for Video World Models
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.
-
MBench: A Comprehensive Benchmark on Memory Capability for Video World Models
MBench is a new benchmark that quantifies long-term memory in video world models via three hierarchical consistency dimensions evaluated on curated real videos.
-
4DThinker: Thinking with 4D Imagery for Dynamic Spatial Understanding
4DThinker enables VLMs to perform dynamic spatial reasoning by thinking with 4D latent mental imagery using new fine-tuning and reinforcement learning methods.
-
Divide and Conquer: Decoupled Representation Alignment for Multimodal World Models
M²-REPA decouples modality-specific features from diffusion intermediates and aligns them to complementary expert foundation models via a multi-modal alignment loss and modality-specific decoupling regularization for improved multimodal video generation.
-
DPPE: Rethinking Camera-Based Positional Encoding for Scaling Multi-View Transformers
DPPE decouples rotation and translation in camera positional encodings for multi-view transformers to resolve late-stage training stagnation and improve generalization in novel view synthesis.
-
MotionAtlas: Detailed Region Captioning for Motion-Centric Videos
MotionAtlas supplies a 2,073-question benchmark, a self-bootstrap pipeline yielding 159k captions, and fine-tuned Video-MLLMs that deliver 5.2-point gains over Qwen3-VL-4B on motion tasks.
-
Compression and Retrieval: Implicit Memory Retrieval for Video World Models
CaR uses attention with viewpoint positional encoding and context compression for flexible memory retrieval in video world models, backed by a new SceneFly dataset, and reports SOTA results with open-domain generalization.
-
PermaVid: Consistent Video Generation Across Edits via Disentangled Context Memory
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.
-
RayDer: Scalable Self-Supervised Novel View Synthesis from Real-World Video
RayDer is a unified transformer backbone for self-supervised static-scene novel view synthesis that absorbs dynamic content as a nuisance factor and shows power-law scaling with data and compute while matching supervised methods in zero-shot settings.
-
minWM: A Full-Stack Open-Source Framework for Real-Time Interactive Video World Models
minWM supplies an end-to-end pipeline that fine-tunes bidirectional T2V/TI2V models with camera control then distills them via Causal Forcing into few-step autoregressive generators for low-latency rollout.
-
Uncovering and Shaping the Latent Representation of 3D Scene Topology in Vision-Language Models
VLMs possess a latent 3D scene topology subspace corresponding to Laplacian eigenmaps that can be causally shaped via Dirichlet energy regularization to improve spatial task performance by up to 12.1%.
-
Relit-LiVE: Relight Video by Jointly Learning Environment Video
Relit-LiVE jointly predicts relit videos and viewpoint-aligned environment maps inside a single diffusion process to achieve physically consistent video relighting without camera pose input.
-
Building a Precise Video Language with Human-AI Oversight
CHAI framework pairs AI pre-captions with expert human critiques to produce precise video descriptions, enabling open models to outperform closed ones like Gemini-3.1-Pro and improve fine-grained control in video generation models.
-
ArtifactWorld: Scaling 3D Gaussian Splatting Artifact Restoration via Video Generation Models
ArtifactWorld restores artifacts in 3D Gaussian Splatting by training a video diffusion backbone on 107.5K paired clips with an isomorphic predictor for artifact heatmaps and an Artifact-Aware Triplet Fusion mechanism to achieve better sparse-view novel synthesis.
-
SceneScribe-1M: A Large-Scale Video Dataset with Comprehensive Geometric and Semantic Annotations
SceneScribe-1M is a new dataset of 1 million videos with semantic text, camera parameters, dense depth, and consistent 3D point tracks to support monocular depth estimation, scene reconstruction, point tracking, and text-to-video synthesis.
-
LoMa: Local Feature Matching Revisited
Scaling data, model size, and compute for local feature matching produces large performance gains on challenging benchmarks and a new manually annotated HardMatch dataset.
-
VGGT-$\Omega$
VGGT-Ω improves feed-forward reconstruction accuracy and efficiency by architectural simplifications, register-based attention, and training on much larger supervised and unlabeled video data.
-
SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion Transformer
SANA-WM is a 2.6B-parameter efficient world model that synthesizes minute-scale 720p videos with 6-DoF camera control, trained on 213K public clips in 15 days on 64 H100s and runnable on single GPUs at 36x higher throughput than prior open baselines.
-
ST-Gen4D: Embedding 4D Spatiotemporal Cognition into World Model for 4D Generation
ST-Gen4D uses a world model that fuses global appearance and local dynamic graphs into a 4D cognition representation to guide consistent 4D Gaussian generation.
-
Who Handles Orientation? Investigating Invariance in Feature Matching
Learning rotation invariance in descriptors matches the performance of matcher-level invariance but allows earlier invariance, faster matchers, and no loss in upright performance when trained at scale.
-
Matrix-Game 3.0: Real-Time and Streaming Interactive World Model with Long-Horizon Memory
Matrix-Game 3.0 delivers 720p real-time video generation at 40 FPS with minute-scale memory consistency by combining residual self-correction training, camera-aware memory injection, and DMD-based autoregressive distillation on a 5B model.
-
Advancing Open-source World Models
LingBot-World is presented as an open-source world model that delivers high-fidelity simulation, minute-level contextual consistency, and real-time interactivity under one second latency.