#01Oct 15, 2026
cs.CV
GeoState: Learning View-Decoupled Geometry States for 3D World Modeling
Hongchang Liu, Jian Liu, Yujuan Ding and 5 more
Recent 3D world models increasingly use features from geometry foundation models (GFMs) as predictive world states, benefiting from their rich geometric priors. However, the view dependence of these features entangles scene dynamics with camera-induced variation, making the resulting state evolution harder to predict. We propose GeoState, a view-decoupled geometry representation for predictive 3D world modeling that separates scene-state evolution from camera-dependent observation. Specifically, a scene tokenizer aggregates multi-view GFM features into a view-decoupled scene state, which a geometry decoder queries using camera rays. An autoregressive world model then predicts future scene states, while camera motion is modeled separately. The predicted states can therefore be decoded under either predicted or externally specified cameras, while retaining a single context-derived scale throughout the rollout without per-frame scale alignment. Experiments show that GeoState outperforms prior 3D world models on Waymo, supports geometrically consistent cross-view querying, and generalizes zero-shot to KITTI and nuScenes. Project page: https://hongchang-liu.github.io/GeoState-project-page/
#02Oct 15, 2026
cs.CV
MJev-Spatial: System One Spatial Intelligence via Native Attention Matching
Luozheng Qin, Yuhan Zhu, Changhai Zhou and 9 more
Spatial intelligence lets an agent perceive and act in the physical world, and many of its judgments must be made at interaction rates: which spatial relation holds, how far an object is, or where to place it. Vision-language models (VLMs) produce these answers by free-form text generation, paying for autoregressive decoding and answer parsing even though the output space is compact: one of a few options, a scalar, or a 2D point. In light of this, we introduce MJev-Spatial, a System 1 multimodal decision model that tackles all three types of spatial understanding tasks within a unified classification framework. Specifically, MJev-Spatial treats each candidate as a class and scores it using the backbone's pretrained query and key projections, without adding parameters. The candidates are the given options for choice tasks, or value bins and image regions refined through coarse-to-fine classification for numerical and point prediction, respectively. At inference, each classification decision takes a single forward pass and generates no answer tokens. To train on heterogeneous spatial data efficiently, we supervise all coarse-to-fine stages jointly with teacher forcing in one forward pass, and combine this with length-aware global batching and cross-GPU load balancing, raising training throughput by 30.8% over the original batching strategy. Built on Qwen3.5-4B and trained on 12.3M question-answer pairs converted from existing spatial datasets, MJev-Spatial achieves the highest overall score (65.64%) among eight tested models, ranks first on seven of twelve benchmarks, and outperforms its unmodified autoregressive base model on all twelve, while running 2.5 times faster than the fastest autoregressive baseline and 5.6 times faster than the base model.
#03Oct 15, 2026
cs.CV
Speculative Forcing for Efficient Autoregressive Video Generation
Hyunwook Choi, Seungryong Kim, Dongyoon Han and 1 more
Few-step autoregressive video generators enable streaming generation, yet they run every token through the final denoising step, regardless of how much refinement it needs. We find this uniform computation is often unnecessary: many tokens already closely match their eventual full-step outputs in the initial steps, while others require refinement until the end. We introduce Speculative Forcing, a framework that treats intermediate clean estimates as drafts and predicts when individual tokens can be safely accepted early. Once accepted, a token is frozen as its final output and undergoes no further denoising, while its cached keys and values remain available as context for the remaining tokens. We explore two alternative verifier designs: a training-free one based on similarity to clean content and stability across steps, and a learned one that reads the frozen generator's intermediate features to estimate each token's remaining gap to its full-step output. Across short-video, long-video, and world generation, Speculative Forcing preserves the quality of full-step sampling while substantially reducing computation. On LongLive, it matches the four-step generator with ~1.7 denoising steps per token, running over 1.5x faster at 41.7 and 37.8 fps. On Lingbot-World 2.0, where camera motion continually changes the visible content, it maintains quality with ~2.5 denoising steps per token and runs 1.35x faster at 24.5 fps.
#04Oct 15, 2026
cs.CV
Skill-Decoupled On-Policy Expert Distillation for Unified Vision-Language Models
Yuqi Xiong, Chunyi Peng, Zhipeng Xu and 5 more
Reinforcement learning (RL) can elicit specialized visual capabilities in vision-language models (VLMs), yet jointly optimizing heterogeneous objectives often introduces cross-capability gradient conflicts. Training experts independently mitigates this interference, but direct reward optimization may entangle specialized knowledge with reward-hacking biases. This paper proposes Skill-Decoupled On-Policy Expert Distillation (SCOPED), which trains capability-specific experts and extracts explicit skills through reward-guided reflection. During on-policy distillation, these skills guide expert supervision while remaining unavailable to the unified student, encouraging it to internalize procedural knowledge without explicit skills at inference time. Across eight multimodal benchmarks, SCOPED consistently outperforms the backbone and mixed-capability RL, demonstrating the effectiveness of skill-decoupled on-policy expert distillation for integrating independently acquired capabilities. Further analyses reveal interference during joint capability optimization and suggest that skill augmentation promotes more consistent expert trajectories, greater attention to task-relevant visual evidence, and better preservation of specialized expertise in the unified student. All code and data are available at https://github.com/NEUIR/SCOPED.
#05Oct 15, 2026
cs.CV
Dense4D: Pixel-Level 4D Human Correspondence
Rıza Alp Güler
Reconstructing human motion in 4D requires identifying the same surface points through articulation, deformation and changes in visibility. We introduce Dense4D, to our knowledge the first feed-forward model dedicated to dense 4D correspondence across the body, clothing and hair from a set of monocular RGB images, without a body template. Trained specifically for human geometry, Dense4D jointly processes person-centred crops with known image-to-crop transforms to predict Human Point Maps (HPMs) and depth for each image. An HPM assigns each person pixel the 3D position of its observed surface point at a common reference time. These maps support amodal correspondence: points hidden in the reference image can be localised from observations in other frames. Matching HPM coordinates across crops establishes correspondence without predicting camera poses or large scene displacements; per-image depth then lifts the matches into 3D tracks. Experiments on real people demonstrate substantially higher tracking accuracy and closer agreement with scanned surfaces than the evaluated baselines. Points from different frames assemble into coherent human geometry in the reference pose, providing visible evidence of consistent dense correspondence through articulation and clothing deformation. Additional qualitative results, videos, and an interactive viewer are available at https://alpguler.com/dense4d.