Xuanchi Ren

Zijian Huang, Jay Zhangjie Wu, Zian Wang et al.

Natural language has become a powerful interface for image generation and editing, yet text-guided visual systems remain highly sensitive to prompt formulation. Semantically similar requests can produce different outputs depending on wording, specificity, and how explicitly visual constraints are stated, motivating prompt enhancement as a trainable component rather than a peripheral user choice. Existing strong enhancers often rely on large, proprietary LLMs such as ChatGPT or Gemini, adding cost, latency, and deployment dependence to the visual generation pipeline. We propose Agentic Prompt Enhancer (APE), a lightweight framework that post-trains small language models (SLMs) as prompt-enhancement agents. APE supports both single-agent rewriting and role-specialized multi-agent enhancement. Its single-agent instantiation, SAPE, rewrites the prompt in one pass, while its multi-agent instantiation, MAPE, decomposes enhancement into a router--rewriter--composer process for handling compositional constraints over objects, attributes, spatial relations, and edits. With task-aware rewards and post-training protocols, APE improves visual alignment and prompt following without modifying the downstream visual model. Experiments on challenging image generation and editing benchmarks demonstrate that post-trained small prompt enhancers reliably outperform their base counterparts, narrowing the gap to closed-source prompt enhancers; in addition, MAPE proves particularly strong on complex compositional tasks within these benchmarks.

Hassan Abu Alhaija, Jose Alvarez, Maciej Bala et al. · nvidia

We introduce Cosmos-Transfer, a conditional world generation model that can generate world simulations based on multiple spatial control inputs of various modalities such as segmentation, depth, and edge. In the design, the spatial conditional scheme is adaptive and customizable. It allows weighting different conditional inputs differently at different spatial locations. This enables highly controllable world generation and finds use in various world-to-world transfer use cases, including Sim2Real. We conduct extensive evaluations to analyze the proposed model and demonstrate its applications for Physical AI, including robotics Sim2Real and autonomous vehicle data enrichment. We further demonstrate an inference scaling strategy to achieve real-time world generation with an NVIDIA GB200 NVL72 rack. To help accelerate research development in the field, we open-source our models and code at https://github.com/nvidia-cosmos/cosmos-transfer1.

Aditi, Niket Agarwal, Arslan Ali et al.

We introduce Cosmos 3, a family of omnimodal world models designed to jointly process and generate language, image, video, audio, and action sequences within a unified mixture-of-transformers architecture. By supporting highly flexible input-output configurations, Cosmos 3 seamlessly unifies critical modalities for Physical AI -- effectively subsuming vision-language models, video generators, world simulators, and world-action models into a single framework. Our evaluation demonstrates that Cosmos 3 establishes a new state-of-the-art across a diverse suite of understanding and generation tasks, demonstrating omnimodal world models as scalable, general-purpose backbones for embodied agents. Our post-trained Cosmos 3 models were ranked as the best open-source Text-to-Image and Image-to-Video models by Artificial Analysis, and the best policy model by RoboArena at the time the technical report was written. To accelerate open research and deployment in Physical AI, we make our code, model checkpoints, curated synthetic datasets, and evaluation benchmark available under the Linux Foundation's OpenMDW-1.1 https://openmdw.ai/license/1-1/ License at https://github.com/nvidia/cosmos}{github.com/nvidia/cosmos and https://huggingface.co/collections/nvidia/cosmos3 . The project website is available at https://research.nvidia.com/labs/cosmos-lab/cosmos3 .

Arslan Ali, Junjie Bai, Maciej Bala et al. · nvidia

We introduce [Cosmos-Predict2.5], the latest generation of the Cosmos World Foundation Models for Physical AI. Built on a flow-based architecture, [Cosmos-Predict2.5] unifies Text2World, Image2World, and Video2World generation in a single model and leverages [Cosmos-Reason1], a Physical AI vision-language model, to provide richer text grounding and finer control of world simulation. Trained on 200M curated video clips and refined with reinforcement learning-based post-training, [Cosmos-Predict2.5] achieves substantial improvements over [Cosmos-Predict1] in video quality and instruction alignment, with models released at 2B and 14B scales. These capabilities enable more reliable synthetic data generation, policy evaluation, and closed-loop simulation for robotics and autonomous systems. We further extend the family with [Cosmos-Transfer2.5], a control-net style framework for Sim2Real and Real2Real world translation. Despite being 3.5$\times$ smaller than [Cosmos-Transfer1], it delivers higher fidelity and robust long-horizon video generation. Together, these advances establish [Cosmos-Predict2.5] and [Cosmos-Transfer2.5] as versatile tools for scaling embodied intelligence. To accelerate research and deployment in Physical AI, we release source code, pretrained checkpoints, and curated benchmarks under the NVIDIA Open Model License at https://github.com/nvidia-cosmos/cosmos-predict2.5 and https://github.com/nvidia-cosmos/cosmos-transfer2.5. We hope these open resources lower the barrier to adoption and foster innovation in building the next generation of embodied intelligence.

Aarti Basant, Amlan Kar, Despoina Paschalidou et al. · nvidia

As autonomous vehicle capabilities advance, the safe evaluation of driving policies in long-tail scenarios remains a critical bottleneck. In closed-loop simulation, the driving policy model actively interacts with the environment, where its actions dynamically update the simulator state and directly influence the next set of generated sensor observations. While recent reconstruction-based neural simulators offer photorealism, they are fundamentally constrained by their initial captured data and struggle to generalize to highly dynamic or novel scenes. To overcome these limitations, we introduce OmniDreams, a foundation generative world model mid- and post-trained from the Cosmos diffusion model to autoregressively generate action-conditioned videos in real time. By leveraging the rich visual priors of Cosmos and mid- and post-training on 21k hours of driving scenarios, OmniDreams synthesizes complex, unobserved phenomena that are hard for traditional simulators to capture, such as extreme weather and unpredictable dynamic agent behaviors. Crucially, it autoregressively conditions its photorealistic sensor generation on past frames, the current simulator state, and immediate driving actions. Deployed in a closed-loop system with the Alpamayo 1 policy model and AlpaSim orchestrator, OmniDreams acts as a highly responsive, reactive environment, providing a scalable and comprehensive solution for training and evaluating next-generation autonomous driving policies. We additionally show preliminary results indicating that a world-action model (WAM) post-trained from OmniDreams achieves strong performance on the Physical AI Autonomous Vehicles NuRec dataset, surpassing the VLA-based Alpamayo 1.5 research policy model while using only 1/5 the total parameters. These results highlight the potential for a real-time world model like OmniDreams to also serve as a backbone for policy architectures.

Xuanchi Ren, Xiaolong Wang · ibm-research

Novel view synthesis from a single image has recently attracted a lot of attention, and it has been primarily advanced by 3D deep learning and rendering techniques. However, most work is still limited by synthesizing new views within relatively small camera motions. In this paper, we propose a novel approach to synthesize a consistent long-term video given a single scene image and a trajectory of large camera motions. Our approach utilizes an autoregressive Transformer to perform sequential modeling of multiple frames, which reasons the relations between multiple frames and the corresponding cameras to predict the next frame. To facilitate learning and ensure consistency among generated frames, we introduce a locality constraint based on the input cameras to guide self-attention among a large number of patches across space and time. Our method outperforms state-of-the-art view synthesis approaches by a large margin, especially when synthesizing long-term future in indoor 3D scenes. Project page at https://xrenaa.github.io/look-outside-room/.

Tianchang Shen, Sherwin Bahmani, Kai He et al. · nvidia, utoronto

Recent advances in video generation enable a new paradigm for 3D scene creation: generating camera-controlled videos that simulate scene walkthroughs, then lifting them to 3D via feed-forward reconstruction techniques. This generative reconstruction approach combines the visual fidelity and creative capacity of video models with 3D outputs ready for real-time rendering and simulation. Scaling to large, complex environments requires 3D-consistent video generation over long camera trajectories with large viewpoint changes and location revisits, a setting where current video models degrade quickly. Existing methods for long-horizon generation are fundamentally limited by two forms of degradation: spatial forgetting and temporal drifting. As exploration proceeds, previously observed regions fall outside the model's temporal context, forcing the model to hallucinate structures when revisited. Meanwhile, autoregressive generation accumulates small synthesis errors over time, gradually distorting scene appearance and geometry. We present Lyra 2.0, a framework for generating persistent, explorable 3D worlds at scale. To address spatial forgetting, we maintain per-frame 3D geometry and use it solely for information routing -- retrieving relevant past frames and establishing dense correspondences with the target viewpoints -- while relying on the generative prior for appearance synthesis. To address temporal drifting, we train with self-augmented histories that expose the model to its own degraded outputs, teaching it to correct drift rather than propagate it. Together, these enable substantially longer and 3D-consistent video trajectories, which we leverage to fine-tune feed-forward reconstruction models that reliably recover high-quality 3D scenes.

Chenyang Lei, Xuanchi Ren, Zhaoxiang Zhang et al.

Many videos contain flickering artifacts. Common causes of flicker include video processing algorithms, video generation algorithms, and capturing videos under specific situations. Prior work usually requires specific guidance such as the flickering frequency, manual annotations, or extra consistent videos to remove the flicker. In this work, we propose a general flicker removal framework that only receives a single flickering video as input without additional guidance. Since it is blind to a specific flickering type or guidance, we name this "blind deflickering." The core of our approach is utilizing the neural atlas in cooperation with a neural filtering strategy. The neural atlas is a unified representation for all frames in a video that provides temporal consistency guidance but is flawed in many cases. To this end, a neural network is trained to mimic a filter to learn the consistent features (e.g., color, brightness) and avoid introducing the artifacts in the atlas. To validate our method, we construct a dataset that contains diverse real-world flickering videos. Extensive experiments show that our method achieves satisfying deflickering performance and even outperforms baselines that use extra guidance on a public benchmark.

Francis Williams, Jiahui Huang, Jonathan Swartz et al.

We present fVDB, a novel GPU-optimized framework for deep learning on large-scale 3D data. fVDB provides a complete set of differentiable primitives to build deep learning architectures for common tasks in 3D learning such as convolution, pooling, attention, ray-tracing, meshing, etc. fVDB simultaneously provides a much larger feature set (primitives and operators) than established frameworks with no loss in efficiency: our operators match or exceed the performance of other frameworks with narrower scope. Furthermore, fVDB can process datasets with much larger footprint and spatial resolution than prior works, while providing a competitive memory footprint on small inputs. To achieve this combination of versatility and performance, fVDB relies on a single novel VDB index grid acceleration structure paired with several key innovations including GPU accelerated sparse grid construction, convolution using tensorcores, fast ray tracing kernels using a Hierarchical Digital Differential Analyzer algorithm (HDDA), and jagged tensors. Our framework is fully integrated with PyTorch enabling interoperability with existing pipelines, and we demonstrate its effectiveness on a number of representative tasks such as large-scale point-cloud segmentation, high resolution 3D generative modeling, unbounded scale Neural Radiance Fields, and large-scale point cloud reconstruction.

Fangfu Liu, Kai He, Tianchang Shen et al.

World models for interactive video generation have largely focused on single-agent settings, where future observations are generated from a single control signal. However, many generated environments require multi-agent interaction: multiple players, robots, or embodied agents act simultaneously within a shared space. Scaling world models to such settings requires a principled multi-agent design: agents should remain independently controllable, permutation-symmetric, and support efficient inference while maintaining consistency across time and perspectives. In this paper, we present our generative multi-agent world model for interactive simulation. It introduces Simplex Rotary Agent Encoding, a parameter-free extension of 3D RoPE that represents agents as vertices of a regular simplex in rotary angle space. This gives each agent a distinct phase while making all agents permutation-equivalent, enabling scalable agent identity without learned per-slot identities or a fixed agent ordering. To avoid dense all-to-all attention across agents, we further propose Sparse Hub Attention, where learnable hub tokens mediate token interaction across agents, reducing cross-agent attention cost from quadratic to linear in the number of agents. For real-time rollout, we distill a full-context diffusion teacher into a causal student that generates temporal blocks sequentially with KV caching, enabling action-responsive generation at 24 FPS. Experiments in multiplayer virtual environments show that our model improves video fidelity, action controllability, and inter-agent consistency over slot-based and dense-attention baselines, while generalizing from two to four players without additional training.

Bingxin Ke, Qunjie Zhou, Jiahui Huang et al. · nvidia, utoronto

We introduce CAPA, a parameter-efficient test-time optimization framework that adapts pre-trained 3D foundation models (FMs) for depth completion, using sparse geometric cues. Unlike prior methods that train task-specific encoders for auxiliary inputs, which often overfit and generalize poorly, CAPA freezes the FM backbone. Instead, it updates only a minimal set of parameters using Parameter-Efficient Fine-Tuning (e.g. LoRA or VPT), guided by gradients calculated directly from the sparse observations available at inference time. This approach effectively grounds the foundation model's geometric prior in the scene-specific measurements, correcting distortions and misplaced structures. For videos, CAPA introduces sequence-level parameter sharing, jointly adapting all frames to exploit temporal correlations, improve robustness, and enforce multi-frame consistency. CAPA is model-agnostic, compatible with any ViT-based FM, and achieves state-of-the-art results across diverse condition patterns on both indoor and outdoor datasets. Project page: research.nvidia.com/labs/dvl/projects/capa.

Yifan Lu, Qi Wu, Jay Zhangjie Wu et al.

Most practical high-resolution text-to-image systems, including latent diffusion and autoregressive models, perform generation in a compact latent space, and a decoder maps the generated latents back to pixels. Yet the latent-to-pixel decoder is reconstruction-oriented, optimized to invert the encoder rather than synthesize more details, and becomes increasingly costly at megapixel scale. This drawback calls for a more expressive and efficient decoding paradigm. Motivated by recent progress in scalable pixel-space diffusion, we introduce PiD, a Pixel diffusion Decoder that reformulates latent decoding as conditional pixel diffusion, unifying decoding and upsampling into one generative module. By denoising directly in high-resolution pixel space, PiD synthesizes $4\times$ and even $8\times$ upscaled images with low latency. For latent conditioning, a lightweight sigma-aware adapter injects noise-corrupted latents into the pixel diffusion backbone, enabling PiD to decode partially denoised latents and terminate the latent diffusion process early. To further improve efficiency, we distill the model using DMD2, reducing inference to just 4 steps. PiD applies to both conventional VAE latents and semantic latents (e.g., SigLIP, DINOv2) used in recent RAE-based models. PiD decodes latents of $512 \times 512$ images into $2048 \times 2048$ pixels in under 1 second with 13 GB peak memory on a consumer RTX 5090, and as fast as 210 ms on a GB200 GPU, about $6\times$ faster than cascaded diffusion-based super-resolution pipelines with better visual fidelity.

Xuanchi Ren, Yifan Lu, Tianshi Cao et al. · nvidia, utoronto

Collecting and annotating real-world data for safety-critical physical AI systems, such as Autonomous Vehicle (AV), is time-consuming and costly. It is especially challenging to capture rare edge cases, which play a critical role in training and testing of an AV system. To address this challenge, we introduce the Cosmos-Drive-Dreams - a synthetic data generation (SDG) pipeline that aims to generate challenging scenarios to facilitate downstream tasks such as perception and driving policy training. Powering this pipeline is Cosmos-Drive, a suite of models specialized from NVIDIA Cosmos world foundation model for the driving domain and are capable of controllable, high-fidelity, multi-view, and spatiotemporally consistent driving video generation. We showcase the utility of these models by applying Cosmos-Drive-Dreams to scale the quantity and diversity of driving datasets with high-fidelity and challenging scenarios. Experimentally, we demonstrate that our generated data helps in mitigating long-tail distribution problems and enhances generalization in downstream tasks such as 3D lane detection, 3D object detection and driving policy learning. We open source our pipeline toolkit, dataset and model weights through the NVIDIA's Cosmos platform. Project page: https://research.nvidia.com/labs/toronto-ai/cosmos_drive_dreams

Jiahui Huang, Qunjie Zhou, Hesam Rabeti et al. · nvidia, utoronto

Accurate 3D geometric perception is an important prerequisite for a wide range of spatial AI systems. While state-of-the-art methods depend on large-scale training data, acquiring consistent and precise 3D annotations from in-the-wild videos remains a key challenge. In this work, we introduce ViPE, a handy and versatile video processing engine designed to bridge this gap. ViPE efficiently estimates camera intrinsics, camera motion, and dense, near-metric depth maps from unconstrained raw videos. It is robust to diverse scenarios, including dynamic selfie videos, cinematic shots, or dashcams, and supports various camera models such as pinhole, wide-angle, and 360° panoramas. We have benchmarked ViPE on multiple benchmarks. Notably, it outperforms existing uncalibrated pose estimation baselines by 18%/50% on TUM/KITTI sequences, and runs at 3-5FPS on a single GPU for standard input resolutions. We use ViPE to annotate a large-scale collection of videos. This collection includes around 100K real-world internet videos, 1M high-quality AI-generated videos, and 2K panoramic videos, totaling approximately 96M frames -- all annotated with accurate camera poses and dense depth maps. We open-source ViPE and the annotated dataset with the hope of accelerating the development of spatial AI systems.

Xuanchi Ren, Jiahui Huang, Xiaohui Zeng et al. · utoronto

We present XCube (abbreviated as $\mathcal{X}^3$), a novel generative model for high-resolution sparse 3D voxel grids with arbitrary attributes. Our model can generate millions of voxels with a finest effective resolution of up to $1024^3$ in a feed-forward fashion without time-consuming test-time optimization. To achieve this, we employ a hierarchical voxel latent diffusion model which generates progressively higher resolution grids in a coarse-to-fine manner using a custom framework built on the highly efficient VDB data structure. Apart from generating high-resolution objects, we demonstrate the effectiveness of XCube on large outdoor scenes at scales of 100m$\times$100m with a voxel size as small as 10cm. We observe clear qualitative and quantitative improvements over past approaches. In addition to unconditional generation, we show that our model can be used to solve a variety of tasks such as user-guided editing, scene completion from a single scan, and text-to-3D. The source code and more results can be found at https://research.nvidia.com/labs/toronto-ai/xcube/.

Xuanchi Ren, Tao Yang, Li Erran Li et al.

Motion prediction of vehicles is critical but challenging due to the uncertainties in complex environments and the limited visibility caused by occlusions and limited sensor ranges. In this paper, we study a new task, safety-aware motion prediction with unseen vehicles for autonomous driving. Unlike the existing trajectory prediction task for seen vehicles, we aim at predicting an occupancy map that indicates the earliest time when each location can be occupied by either seen and unseen vehicles. The ability to predict unseen vehicles is critical for safety in autonomous driving. To tackle this challenging task, we propose a safety-aware deep learning model with three new loss functions to predict the earliest occupancy map. Experiments on the large-scale autonomous driving nuScenes dataset show that our proposed model significantly outperforms the state-of-the-art baselines on the safety-aware motion prediction task. To the best of our knowledge, our approach is the first one that can predict the existence of unseen vehicles in most cases. Project page at {\url{https://github.com/xrenaa/Safety-Aware-Motion-Prediction}}.

Xuanchi Ren, Tao Yang, Yuwang Wang et al.

Content and style (C-S) disentanglement intends to decompose the underlying explanatory factors of objects into two independent subspaces. From the unsupervised disentanglement perspective, we rethink content and style and propose a formulation for unsupervised C-S disentanglement based on our assumption that different factors are of different importance and popularity for image reconstruction, which serves as a data bias. The corresponding model inductive bias is introduced by our proposed C-S disentanglement Module (C-S DisMo), which assigns different and independent roles to content and style when approximating the real data distributions. Specifically, each content embedding from the dataset, which encodes the most dominant factors for image reconstruction, is assumed to be sampled from a shared distribution across the dataset. The style embedding for a particular image, encoding the remaining factors, is used to customize the shared distribution through an affine transformation. The experiments on several popular datasets demonstrate that our method achieves the state-of-the-art unsupervised C-S disentanglement, which is comparable or even better than supervised methods. We verify the effectiveness of our method by downstream tasks: domain translation and single-view 3D reconstruction. Project page at https://github.com/xrenaa/CS-DisMo.

Xuanchi Ren, Tao Yang, Yuwang Wang et al.

From the intuitive notion of disentanglement, the image variations corresponding to different factors should be distinct from each other, and the disentangled representation should reflect those variations with separate dimensions. To discover the factors and learn disentangled representation, previous methods typically leverage an extra regularization term when learning to generate realistic images. However, the term usually results in a trade-off between disentanglement and generation quality. For the generative models pretrained without any disentanglement term, the generated images show semantically meaningful variations when traversing along different directions in the latent space. Based on this observation, we argue that it is possible to mitigate the trade-off by $(i)$ leveraging the pretrained generative models with high generation quality, $(ii)$ focusing on discovering the traversal directions as factors for disentangled representation learning. To achieve this, we propose Disentaglement via Contrast (DisCo) as a framework to model the variations based on the target disentangled representations, and contrast the variations to jointly discover disentangled directions and learn disentangled representations. DisCo achieves the state-of-the-art disentangled representation learning and distinct direction discovering, given pretrained non-disentangled generative models including GAN, VAE, and Flow. Source code is at https://github.com/xrenaa/DisCo.

Xuanchi Ren, Zian Qian, Qifeng Chen

Motion blur in videos captured by autonomous vehicles and robots can degrade their perception capability. In this work, we present a novel approach to video deblurring by fitting a deep network to the test video. Our key observation is that some frames in a video with motion blur are much sharper than others, and thus we can transfer the texture information in those sharp frames to blurry frames. Our approach heuristically selects sharp frames from a video and then trains a convolutional neural network on these sharp frames. The trained network often absorbs enough details in the scene to perform deblurring on all the video frames. As an internal learning method, our approach has no domain gap between training and test data, which is a problematic issue for existing video deblurring approaches. The conducted experiments on real-world video data show that our model can reconstruct clearer and sharper videos than state-of-the-art video deblurring approaches. Code and data are available at https://github.com/xrenaa/Deblur-by-Fitting.

Xuanchi Ren, Tianchang Shen, Jiahui Huang et al. · nvidia, utoronto

We present GEN3C, a generative video model with precise Camera Control and temporal 3D Consistency. Prior video models already generate realistic videos, but they tend to leverage little 3D information, leading to inconsistencies, such as objects popping in and out of existence. Camera control, if implemented at all, is imprecise, because camera parameters are mere inputs to the neural network which must then infer how the video depends on the camera. In contrast, GEN3C is guided by a 3D cache: point clouds obtained by predicting the pixel-wise depth of seed images or previously generated frames. When generating the next frames, GEN3C is conditioned on the 2D renderings of the 3D cache with the new camera trajectory provided by the user. Crucially, this means that GEN3C neither has to remember what it previously generated nor does it have to infer the image structure from the camera pose. The model, instead, can focus all its generative power on previously unobserved regions, as well as advancing the scene state to the next frame. Our results demonstrate more precise camera control than prior work, as well as state-of-the-art results in sparse-view novel view synthesis, even in challenging settings such as driving scenes and monocular dynamic video. Results are best viewed in videos. Check out our webpage! https://research.nvidia.com/labs/toronto-ai/GEN3C/

Jay Zhangjie Wu, Yuxuan Zhang, Haithem Turki et al.

Neural Radiance Fields and 3D Gaussian Splatting have revolutionized 3D reconstruction and novel-view synthesis task. However, achieving photorealistic rendering from extreme novel viewpoints remains challenging, as artifacts persist across representations. In this work, we introduce Difix3D+, a novel pipeline designed to enhance 3D reconstruction and novel-view synthesis through single-step diffusion models. At the core of our approach is Difix, a single-step image diffusion model trained to enhance and remove artifacts in rendered novel views caused by underconstrained regions of the 3D representation. Difix serves two critical roles in our pipeline. First, it is used during the reconstruction phase to clean up pseudo-training views that are rendered from the reconstruction and then distilled back into 3D. This greatly enhances underconstrained regions and improves the overall 3D representation quality. More importantly, Difix also acts as a neural enhancer during inference, effectively removing residual artifacts arising from imperfect 3D supervision and the limited capacity of current reconstruction models. Difix3D+ is a general solution, a single model compatible with both NeRF and 3DGS representations, and it achieves an average 2$\times$ improvement in FID score over baselines while maintaining 3D consistency.

Yifan Lu, Xuanchi Ren, Jiawei Yang et al. · nvidia, utoronto

We present InfiniCube, a scalable method for generating unbounded dynamic 3D driving scenes with high fidelity and controllability. Previous methods for scene generation either suffer from limited scales or lack geometric and appearance consistency along generated sequences. In contrast, we leverage the recent advancements in scalable 3D representation and video models to achieve large dynamic scene generation that allows flexible controls through HD maps, vehicle bounding boxes, and text descriptions. First, we construct a map-conditioned sparse-voxel-based 3D generative model to unleash its power for unbounded voxel world generation. Then, we re-purpose a video model and ground it on the voxel world through a set of carefully designed pixel-aligned guidance buffers, synthesizing a consistent appearance. Finally, we propose a fast feed-forward approach that employs both voxel and pixel branches to lift the dynamic videos to dynamic 3D Gaussians with controllable objects. Our method can generate controllable and realistic 3D driving scenes, and extensive experiments validate the effectiveness and superiority of our model.

Xuanchi Ren, Yifan Lu, Hanxue Liang et al.

We present SCube, a novel method for reconstructing large-scale 3D scenes (geometry, appearance, and semantics) from a sparse set of posed images. Our method encodes reconstructed scenes using a novel representation VoxSplat, which is a set of 3D Gaussians supported on a high-resolution sparse-voxel scaffold. To reconstruct a VoxSplat from images, we employ a hierarchical voxel latent diffusion model conditioned on the input images followed by a feedforward appearance prediction model. The diffusion model generates high-resolution grids progressively in a coarse-to-fine manner, and the appearance network predicts a set of Gaussians within each voxel. From as few as 3 non-overlapping input images, SCube can generate millions of Gaussians with a 1024^3 voxel grid spanning hundreds of meters in 20 seconds. Past works tackling scene reconstruction from images either rely on per-scene optimization and fail to reconstruct the scene away from input views (thus requiring dense view coverage as input) or leverage geometric priors based on low-resolution models, which produce blurry results. In contrast, SCube leverages high-resolution sparse networks and produces sharp outputs from few views. We show the superiority of SCube compared to prior art using the Waymo self-driving dataset on 3D reconstruction and demonstrate its applications, such as LiDAR simulation and text-to-scene generation.

Sherwin Bahmani, Tianchang Shen, Jiawei Ren et al. · nvidia, utoronto

The ability to generate virtual environments is crucial for applications ranging from gaming to physical AI domains such as robotics, autonomous driving, and industrial AI. Current learning-based 3D reconstruction methods rely on the availability of captured real-world multi-view data, which is not always readily available. Recent advancements in video diffusion models have shown remarkable imagination capabilities, yet their 2D nature limits the applications to simulation where a robot needs to navigate and interact with the environment. In this paper, we propose a self-distillation framework that aims to distill the implicit 3D knowledge in the video diffusion models into an explicit 3D Gaussian Splatting (3DGS) representation, eliminating the need for multi-view training data. Specifically, we augment the typical RGB decoder with a 3DGS decoder, which is supervised by the output of the RGB decoder. In this approach, the 3DGS decoder can be purely trained with synthetic data generated by video diffusion models. At inference time, our model can synthesize 3D scenes from either a text prompt or a single image for real-time rendering. Our framework further extends to dynamic 3D scene generation from a monocular input video. Experimental results show that our framework achieves state-of-the-art performance in static and dynamic 3D scene generation.

Shaowei Liu, Xuanchi Ren, Tianchang Shen et al.

Generating motion-controlled videos--where user-specified actions drive physically plausible scene dynamics under freely chosen viewpoints--demands two capabilities: (1) disentangled motion control, allowing users to separately control the object motion and adjust camera viewpoint; and (2) motion causality, ensuring that user-driven actions trigger coherent reactions from other objects rather than merely displacing pixels. Existing methods fall short on both fronts: they entangle camera and object motion into a single tracking signal and treat motion as kinematic displacement without modeling causal relationships between object motion. We introduce MoRight, a unified framework that addresses both limitations through disentangled motion modeling. Object motion is specified in a canonical static-view and transferred to an arbitrary target camera viewpoint via temporal cross-view attention, enabling disentangled camera and object control. We further decompose motion into active (user-driven) and passive (consequence) components, training the model to learn motion causality from data. At inference, users can either supply active motion and MoRight predicts consequences (forward reasoning), or specify desired passive outcomes and MoRight recovers plausible driving actions (inverse reasoning), all while freely adjusting the camera viewpoint. Experiments on three benchmarks demonstrate state-of-the-art performance in generation quality, motion controllability, and interaction awareness.

Jay Zhangjie Wu, Xuanchi Ren, Tianchang Shen et al. · nvidia, utoronto

Recent advances in large generative models have greatly enhanced both image editing and in-context image generation, yet a critical gap remains in ensuring physical consistency, where edited objects must remain coherent. This capability is especially vital for world simulation related tasks. In this paper, we present ChronoEdit, a framework that reframes image editing as a video generation problem. First, ChronoEdit treats the input and edited images as the first and last frames of a video, allowing it to leverage large pretrained video generative models that capture not only object appearance but also the implicit physics of motion and interaction through learned temporal consistency. Second, ChronoEdit introduces a temporal reasoning stage that explicitly performs editing at inference time. Under this setting, target frame is jointly denoised with reasoning tokens to imagine a plausible editing trajectory that constrains the solution space to physically viable transformations. The reasoning tokens are then dropped after a few steps to avoid the high computational cost of rendering a full video. To validate ChronoEdit, we introduce PBench-Edit, a new benchmark of image-prompt pairs for contexts that require physical consistency, and demonstrate that ChronoEdit surpasses state-of-the-art baselines in both visual fidelity and physical plausibility. Project page for code and models: https://research.nvidia.com/labs/toronto-ai/chronoedit

Dacheng Yin, Xuanchi Ren, Chong Luo et al.

This paper addresses the unsupervised learning of content-style decomposed representation. We first give a definition of style and then model the content-style representation as a token-level bipartite graph. An unsupervised framework, named Retriever, is proposed to learn such representations. First, a cross-attention module is employed to retrieve permutation invariant (P.I.) information, defined as style, from the input data. Second, a vector quantization (VQ) module is used, together with man-induced constraints, to produce interpretable content tokens. Last, an innovative link attention module serves as the decoder to reconstruct data from the decomposed content and style, with the help of the linking keys. Being modal-agnostic, the proposed Retriever is evaluated in both speech and image domains. The state-of-the-art zero-shot voice conversion performance confirms the disentangling ability of our framework. Top performance is also achieved in the part discovery task for images, verifying the interpretability of our representation. In addition, the vivid part-based style transfer quality demonstrates the potential of Retriever to support various fascinating generative tasks. Project page at https://ydcustc.github.io/retriever-demo/.

Tao Yang, Xuanchi Ren, Yuwang Wang et al.

Disentangled representation learning is one of the major goals of deep learning, and is a key step for achieving explainable and generalizable models. A well-defined theoretical guarantee still lacks for the VAE-based unsupervised methods, which are a set of popular methods to achieve unsupervised disentanglement. The Group Theory based definition of representation disentanglement mathematically connects the data transformations to the representations using the formalism of group. In this paper, built on the group-based definition and inspired by the n-th dihedral group, we first propose a theoretical framework towards achieving unsupervised representation disentanglement. We then propose a model, based on existing VAE-based methods, to tackle the unsupervised learning problem of the framework. In the theoretical framework, we prove three sufficient conditions on model, group structure, and data respectively in an effort to achieve, in an unsupervised way, disentangled representation per group-based definition. With the first two of the conditions satisfied and a necessary condition derived for the third one, we offer additional constraints, from the perspective of the group-based definition, for the existing VAE-based models. Experimentally, we train 1800 models covering the most prominent VAE-based methods on five datasets to verify the effectiveness of our theoretical framework. Compared to the original VAE-based methods, these Groupified VAEs consistently achieve better mean performance with smaller variances.

Xuanchi Ren, Haoran Li, Zijian Huang et al.

We present a learning-based approach with pose perceptual loss for automatic music video generation. Our method can produce a realistic dance video that conforms to the beats and rhymes of almost any given music. To achieve this, we firstly generate a human skeleton sequence from music and then apply the learned pose-to-appearance mapping to generate the final video. In the stage of generating skeleton sequences, we utilize two discriminators to capture different aspects of the sequence and propose a novel pose perceptual loss to produce natural dances. Besides, we also provide a new cross-modal evaluation to evaluate the dance quality, which is able to estimate the similarity between two modalities of music and dance. Finally, a user study is conducted to demonstrate that dance video synthesized by the presented approach produces surprisingly realistic results. The results are shown in the supplementary video at https://youtu.be/0rMuFMZa_K4