CVPR 2026 论文列表

We introduce a framework for learning latent representations of 4D objects which are descriptive, faithfully capturing object geometry and appearance; compressive, aiding in downstream efficiency; and accessible, requiring minimal input, i.e., an unstructured dynamic point cloud, to construct. Specifically, Velox trains an encoder to compress spatiotemporal color point clouds into a set of dynamic tokens. These tokens are supervised using two complementary decoders: a 4D surface decoder, which models the time-varying surface distribution capturing the geometry; and a Gaussian decoder, which maps the tokens to 3D Gaussians, helping learn appearance. To demonstrate the utility of our representation, we evaluate it across three downstream tasks: video-to-4D generation, 3D tracking, and cloth simulation via image-to-4D generation, and observe strong performances in all settings.

Latent diffusion has become the default paradigm for visual generation, yet we observe a persistent reconstruction-generation trade-off as latent dimensionality increases: higher-capacity autoencoders improve reconstruction fidelity but generation quality eventually declines. We trace this gap to the different behaviors in high-frequency tokenization and detokenization. Through controlled perturbations in both RGB and latent domains, we analyze encoder/decoder behaviors and find that decoders depend strongly on high-frequency latent components to recover details, whereas encoders under-represent high-frequency contents, yielding insufficient exposure and underfitting in high-frequency bands for diffusion model training. To address this issue, we introduce FreqWarm, a plug-and-play frequency warm-up curriculum that increases early-stage exposure to high-frequency latent signals during diffusion or flow-matching training -- without modifying or retraining the autoencoder. Applied across several high-dimensional tokenizers, FreqWarm consistently improves generation quality: decreasing gFID by 14.11 on Wan2.2-VAE, 6.14 on LTX-VAE, and 4.42 on DC-AE-f32, while remaining architecture-agnostic and compatible with diverse backbones. Our study shows that explicitly managing frequency exposure can successfully turn high-dimensional latent spaces into more diffusible targets.

Spatial intelligence is essential for multimodal large language models, yet current benchmarks largely assess it only from an understanding perspective. We ask whether modern generative or unified multimodal models also possess generative spatial intelligence (GSI)--the ability to respect and manipulate 3D spatial constraints during image generation--and whether such capability can be measured or improved. We introduce GSI-Bench, the first benchmark designed to quantify GSI through spatially grounded image editing. It consists of two complementary components: GSI-Real, a high-quality real-world dataset built via a 3D-prior-guided generation and filtering pipeline, and GSI-Syn, a large-scale synthetic benchmark with controllable spatial operations and fully automated labeling.Together with a unified evaluation protocol, GSI-Bench enables scalable, model-agnostic assessment of spatial compliance and editing fidelity. Experiments show that fine-tuning unified multimodal models on GSI-Syn yields substantial gains on both synthetic and real tasks and, strikingly, also improves downstream spatial understanding. This provides the first clear evidence that generative training can tangibly strengthen spatial reasoning--establishing a new pathway for advancing spatial intelligence in multimodal models.

Human-centric multi-view video has a clear semantic structure: a static background and dynamic human motion. We propose a generative compression framework that explicitly decouples these components. The background is modeled once with 3D Gaussian Splatting, while the human is represented by a personalized Gaussian avatar reconstructed from a sparse set of key views that are transmitted only once and driven by compact per-frame pose parameters from the Skinned Multi-Person Linear (SMPL) model. The encoder sends only three elements: the background, the key views, and the SMPL parameters, enabling high-fidelity multi-viewpoint synthesis at dramatically reduced bitrates. This shifts compression from low-level redundancy removal to semantics-aware generative modeling. Experiments across multiple human-centric datasets demonstrate superior rate-distortion performance, particularly for long and densely captured sequences, and naturally enable semantic editing.

Generative video models achieve high visual fidelity but often violate basic physical principles, limiting reliability in real-world settings. Prior attempts to inject physics rely on conditioning: frame-level signals are domain-specific and short-horizon, while global text prompts are coarse and noisy, missing fine-grained dynamics. We present PhysVid, a physics-aware local conditioning scheme that operates over temporally contiguous chunks of frames. Each chunk is annotated with physics-grounded descriptions of states, interactions, and constraints, which are fused with the global prompt via chunk-aware cross-attention during training. At inference, we introduce negative physics prompts (descriptions of locally relevant law violations) to steer generation away from implausible trajectories. On VideoPhy, PhysVid improves physical commonsense scores by ~ 33% over baseline video generators, and by up to ~ 8% on VideoPhy2. These results show that local, physics-aware guidance substantially increases physical plausibility in generative video and marks a step toward physics-grounded video models.

Existing single object trackers typically treat temporal modeling superficially by passing limited inter-frame information, such as propagated tokens or template updates, without intrinsic temporal supervision learning. To address this limitation, we propose TGTrack, a new unified tracking framework that incorporates a temporally generative supervision task to guide the model in learning temporal dynamics. The core of TGTrack is a temporally generative learning paradigm equipped with a transformer-based generative decoder, which consists of a gated fusion module and an autoregressive prediction mechanism. This joint design enables the model to infer future scenarios from preceding information, thereby improving its ability to model both visual appearance and temporal dynamics. Furthermore, we introduce a time token embedding to explicitly encode the temporal position of each frame. Experiments on 11 benchmarks spanning five modalities show that TGTrack achieves state-of-the-art performance in robust unified tracking. For instance, TGTrack-B384 achieves an AUC of 75.3% on LaSOT. Code is available at https://github.com/wtg1/TGTrack.

MeanFlow (MF) is a diffusion-motivated generative model that enables efficient few-step generation by learning long jumps directly from noise to data. In practice, it is often used as a latent MF by leveraging the pre-trained Stable Diffusion variational autoencoder (SD-VAE) for high-dimensional data modeling. However, MF training remains computationally demanding and is often unstable. During inference, the SD-VAE decoder dominates the generation cost, and MF depends on complex guidance hyperparameters for class-conditional generation. In this work, we develop an efficient training and sampling scheme for MF in the latent space of a Representation Autoencoder (RAE), where a pre-trained vision encoder (e.g., DINO) provides semantically rich latents paired with a lightweight decoder. We observe that naive MF training in the RAE latent space suffers from severe gradient explosion. To stabilize and accelerate training, we adopt Consistency Mid-Training for trajectory-aware initialization and use a two-stage scheme: distillation from a pre-trained flow matching teacher to speed convergence and reduce variance, followed by an optional bootstrapping stage with a one-point velocity estimator to further reduce deviation from the oracle mean flow. This design removes the need for guidance, simplifies training configurations, and reduces computation in both training and sampling. Empirically, our method achieves a 1-step FID of 2.03, outperforming vanilla MF's 3.43, while reducing sampling GFLOPS by 38% and total training cost by 83% on ImageNet 256. We further scale our approach to ImageNet 512, achieving a competitive one-step FID of 3.23 with the lowest GFLOPS among all baselines.

Anticipating diverse future states is a central challenge in video world modeling. Discriminative world models produce deterministic predictions that implicitly average over possible futures, while existing generative world models remain computationally expensive. Recent work demonstrates that predicting the future in the feature space of a vision foundation model (VFM), rather than a latent space optimized for pixel reconstruction, requires significantly fewer world model parameters. However, most such approaches remain discriminative. In this work, we introduce DeltaWorld, a generative VFM-based world model that efficiently generates diverse plausible futures. At the core of DeltaWorld is DeltaTok, a tokenizer that encodes the feature difference between consecutive frames into a single continuous "delta" token, reducing video from a three-dimensional spatio-temporal representation to a one-dimensional temporal sequence. For example, this yields a 1,024x token reduction with 512x512 frames. Delta tokens enable efficient and effective multi-hypothesis training, where many diverse futures are generated in parallel and only the best is supervised. At inference, this leads to diverse predictions in a single forward pass. Experiments on dense forecasting tasks demonstrate that DeltaWorld forecasts futures that more closely align with real-world outcomes, while having over 35x fewer parameters and using 2,000x fewer FLOPs than existing generative world models. Project page: https://deltatok.github.io.

We present Functional Mean Flow (FMF) as a one-step generative model defined in infinite-dimensional Hilbert space. FMF extends the one-step Mean Flow framework to functional domains by providing a theoretical formulation for Functional Flow Matching and a practical implementation for efficient training and sampling. We also introduce an x_1-prediction variant that improves stability over the original u-prediction form. The resulting framework is a practical one-step Flow Matching method applicable to a wide range of functional data generation tasks such as time series, images, PDEs, and 3D geometry.

A world model is an internal model that simulates how the world evolves. Given past observations and actions, it predicts the future physical state of both the embodied agent and its environment. Accurate world models are essential for enabling agents to think, plan, and reason effectively in complex, dynamic settings. However, existing world models often focus on random generation of open worlds, but neglect the need for high-fidelity modeling of deterministic scenarios (such as fixed-map mazes and static space robot navigation). In this work, we take a step toward building a truly accurate world model by addressing a fundamental yet open problem: constructing a model that can fully clone a deterministic world. 1) Through diagnostic experiment, we quantitatively demonstrate that high-fidelity cloning is feasible and the primary bottleneck for long-horizon fidelity is the geometric structure of the latent representation, not the dynamics model itself. 2) Building on this insight, we show that applying temporal contrastive learning principle as a geometric regularization can effectively curate a latent space that better reflects the underlying physical state manifold, demonstrating that contrastive constraints can serve as a powerful inductive bias for stable world modeling; we call this approach Geometrically-Regularized World Models (GRWM). At its core is a lightweight geometric regularization module that can be seamlessly integrated into standard autoencoders, reshaping their latent space to provide a stable foundation for effective dynamics modeling. By focusing on representation quality, GRWM offers a simple yet powerful pipeline for improving world model fidelity. Code is available at https://github.com/XiaFire/Clone_Deterministic_Environment.

Fully decentralized deep learning removes global servers and ensures local data privacy. However, Euclidean consensus, averaging weights, gradients or momentum, may degrade under the situations with non-i.i.d. (non-independent and identically distributed) data and client size imbalance. We propose a geometry-aware approach based on natural gradient variational inference. Clients communicate in the expectation parameter space of an exponential family, where simple linear mixing yields a forward Kullback--Leibler (KL) barycenter consensus. The aggregate is the model closest to all client distributions, aligning updates across heterogeneous sites and mitigating distribution shift. We further provide a lightweight decentralized Adam implementation where each client maintains a diagonal Gaussian posterior and operates entirely in the expectation space for both optimization and consensus. We prove convergence for convex losses on connected graphs. On CIFAR-100 and a medical image segmentation benchmark, our method substantially outperforms Euclidean-space consensus baselines under severe non-i.i.d. and client-imbalance cases, achieving around 20% accuracy gain on CIFAR-100, while matching the communication budget and improving training stability. The code is available at https://github.com/x-lu/Beyond-euclidean-gossip.

Aligning video generative models with human preferences remains challenging: current approaches rely on Vision-Language Models (VLMs) for reward modeling, but these models struggle to capture subtle temporal dynamics. We propose a fundamentally different approach: repurposing video generative models, which are inherently designed to model temporal structure, as reward models. We present the Generative-Transformer-based Self-Supervised Video Judge (\modelname), a novel evaluation model that transforms state-of-the-art video generation models into powerful temporally-aware reward models. Our key insight is that generative models can be reformulated as energy-based models (EBMs) that assign low energy to high-quality videos and high energy to degraded ones, enabling them to discriminate video quality with remarkable precision when trained via contrastive objectives. To prevent the model from exploiting superficial differences between real and generated videos, we design challenging synthetic negative videos through controlled latent-space perturbations: temporal slicing, feature swapping, and frame shuffling, which simulate realistic but subtle visual degradations. This forces the model to learn meaningful spatiotemporal features rather than trivial artifacts. \modelname achieves state-of-the-art performance on GenAI-Bench and MonteBench using only 30K human-annotations: 6xto 65xfewer than existing VLM-based approaches.

Dynamic objects in our physical 4D (3D + time) world are constantly evolving, deforming, and interacting with other objects, leading to diverse 4D scene dynamics. In this paper, we present a universal generative pipeline, CHORD, for CHOReographing Dynamic objects and scenes and synthesizing this type of phenomena. Traditional rule-based graphics pipelines to create these dynamics are based on category-specific heuristics, yet are labor-intensive and not scalable. Recent learning-based methods typically demand large-scale datasets, which may not cover all object categories in interest. Our approach instead inherits the universality from the video generative models by proposing a distillation-based pipeline to extract the rich Lagrangian motion information hidden in the Eulerian representations of 2D videos. Our method is universal, versatile, and category-agnostic. We demonstrate its effectiveness by conducting experiments to generate a diverse range of multi-body 4D dynamics, show its advantage compared to existing methods, and demonstrate its applicability in generating robotics manipulation policies.