arXiv subjects

Zilong Chen

Publications and source records attributed to Zilong Chen.

At least 19 recordsLinked to original sources

arXiv2026

Scaling Properties of Text Conditioning in Visual Generation

We study empirical scaling properties for text conditioning in visual generation. Such properties have rarely been measured because diffusion loss does not scale with the number of tokens in natural-language prompts. Surprisingly, we find that the converged diffusion loss scales with the amount of structured language in the prompt. To quantify structured language, we adapt two complementary measures: a white-box likelihood metric (GPG) and a black-box attribute metric (ED). Across controlled training runs, the converged diffusion loss decreases approximately linearly with GPG and follows a power law with ED. Guided by these scaling properties, we improve \emph{diffusability} by constructing structured prompts with semantic and geometric annotations derived from images, and improve \emph{promptability} by training a prompter through supervised fine-tuning, cold-start, and verifier-gated on-policy distillation. The resulting system outperforms all evaluated open-weight models on nearly every compositional, reasoning, and world-knowledge benchmark, while matching or surpassing the strongest closed-weight models on most evaluations.

cs.CV↗

arXiv2026

ARM: An AutoRegressive Large Multimodal Model with Unified Discrete Representations

This paper introduces ARM, a discrete representation-based AutoRegressive Model that unifies image understanding, generation, and editing within a next-token prediction framework. ARM is built on three efforts: first, we train a discrete semantic visual tokenizer that maps images into compact token sequences. Our tokenizer is supervised with multiple objectives that jointly promote semantic discriminability, language alignment and faithful reconstruction, thereby supporting diverse tasks in a shared latent space. With this, we train a 7B autoregressive model over large-scale text and image token sequences, seamlessly developing vision-language perception and generation capabilities. Finally, to further improve preference-aligned behavior for text-to-image generation and instruction-guided editing, ARM applies reinforcement learning (RL) to optimize task-level objectives such as visual quality, instruction adherence, and edit consistency. Surprisingly, the results show that RL not only substantially improves performance on the target tasks (e.g., raising WISE overall from 0.50 to 0.56, GEdit-Bench-EN G_O from 5.75 to 6.68), but also induces cross-task synergy between text-to-image generation and editing. Collectively, these findings highlight autoregressive modeling, when paired with strong representations and preference optimization, as a scalable foundation for multimodal intelligence. Code: https://github.com/wdrink/ARM.

cs.CV↗

arXiv2026

Unified 4D World Action Modeling from Video Priors with Asynchronous Denoising

We propose X-WAM, a Unified 4D World Model that unifies real-time robotic action execution and high-fidelity 4D world synthesis (video + 3D reconstruction) in a single framework, addressing the critical limitations of prior unified world models (e.g., UWM) that only model 2D pixel-space and fail to balance action efficiency and world modeling quality. To leverage the strong visual priors of pretrained video diffusion models, X-WAM imagines the future world by predicting multi-view RGB-D videos, and obtains spatial information efficiently through a lightweight structural adaptation: replicating the final few blocks of the pretrained Diffusion Transformer into a dedicated depth prediction branch for the reconstruction of future spatial information. Moreover, we propose Asynchronous Noise Sampling (ANS) to jointly optimize generation quality and action decoding efficiency. ANS applies a specialized asynchronous denoising schedule during inference, which rapidly decodes actions with fewer steps to enable efficient real-time execution, while dedicating the full sequence of steps to generate high-fidelity video. Rather than entirely decoupling the timesteps during training, ANS samples from their joint distribution to align with the inference distribution. Pretrained on over 5,800 hours of robotic data, X-WAM achieves 79.2% and 90.7% average success rate on RoboCasa and RoboTwin 2.0 benchmarks, while producing high-fidelity 4D reconstruction and generation surpassing existing methods in both visual and geometric metrics.

cs.RO↗

arXiv2025

VQ-VA World: Towards High-Quality Visual Question-Visual Answering

This paper studies Visual Question-Visual Answering (VQ-VA): generating an image, rather than text, in response to a visual question -- an ability that has recently emerged in proprietary systems such as NanoBanana and GPT-Image. To also bring this capability to open-source models, we introduce VQ-VA World, a data-centric framework built around an agentic pipeline for large-scale, targeted data construction. Leveraging web-scale deployment, this pipeline crawls a massive amount of ~1.8M high-quality, interleaved image-text samples for model training. For evaluation, we further release IntelligentBench, a human-curated benchmark that systematically assesses VQ-VA along the aspects of world knowledge, design knowledge, and reasoning. Training with VQ-VA World data yields strong empirical gains: it helps LightFusion attain 53.06 on IntelligentBench, substantially surpassing the best prior open-source baselines (i.e., 7.78 from vanilla LightFusion; 1.94 from UniWorld-V1), and significantly narrowing the gap toward leading proprietary systems (e.g., 81.67 from NanoBanana; 82.64 from GPT-Image). By releasing the full suite of model weights, datasets, and pipelines, we hope to stimulate future research on VQ-VA.

cs.CV↗

arXiv2025

Alias-free 4D Gaussian Splatting

Existing dynamic scene reconstruction methods based on Gaussian Splatting enable real-time rendering and generate realistic images. However, adjusting the camera's focal length or the distance between Gaussian primitives and the camera to modify rendering resolution often introduces strong artifacts, stemming from the frequency constraints of 4D Gaussians and Gaussian scale mismatch induced by the 2D dilated filter. To address this, we derive a maximum sampling frequency formulation for 4D Gaussian Splatting and introduce a 4D scale-adaptive filter and scale loss, which flexibly regulates the sampling frequency of 4D Gaussian Splatting. Our approach eliminates high-frequency artifacts under increased rendering frequencies while effectively reducing redundant Gaussians in multi-view video reconstruction. We validate the proposed method through monocular and multi-view video reconstruction experiments.Ours project page: https://4d-alias-free.github.io/4D-Alias-free/

cs.CV↗

arXiv2025

LightFusion: A Light-weighted, Double Fusion Framework for Unified Multimodal Understanding and Generation

Unified multimodal models have recently shown remarkable gains in both capability and versatility, yet most leading systems are still trained from scratch and require substantial computational resources. In this paper, we show that competitive performance can be obtained far more efficiently by strategically fusing publicly available models specialized for either generation or understanding. Our key design is to retain the original blocks while additionally interleaving multimodal self-attention blocks throughout the networks. This double fusion mechanism (1) effectively enables rich multi-modal fusion while largely preserving the original strengths of the base models, and (2) catalyzes synergistic fusion of high-level semantic representations from the understanding encoder with low-level spatial signals from the generation encoder. By training with only ~ 35B tokens, this approach achieves strong results across multiple benchmarks: 0.91 on GenEval for compositional text-to-image generation, 82.16 on DPG-Bench for complex text-to-image generation, 6.06 on GEditBench, and 3.77 on ImgEdit-Bench for image editing. By fully releasing the entire suite of code, model weights, and datasets, we hope to support future research on unified multimodal modeling.

cs.CV↗

arXiv2025

From Virtual Games to Real-World Play

We introduce RealPlay, a neural network-based real-world game engine that enables interactive video generation from user control signals. Unlike prior works focused on game-style visuals, RealPlay aims to produce photorealistic, temporally consistent video sequences that resemble real-world footage. It operates in an interactive loop: users observe a generated scene, issue a control command, and receive a short video chunk in response. To enable such realistic and responsive generation, we address key challenges including iterative chunk-wise prediction for low-latency feedback, temporal consistency across iterations, and accurate control response. RealPlay is trained on a combination of labeled game data and unlabeled real-world videos, without requiring real-world action annotations. Notably, we observe two forms of generalization: (1) control transfer-RealPlay effectively maps control signals from virtual to real-world scenarios; and (2) entity transfer-although training labels originate solely from a car racing game, RealPlay generalizes to control diverse real-world entities, including bicycles and pedestrians, beyond vehicles. Project page can be found: https://wenqsun.github.io/RealPlay/

cs.CV↗

arXiv2025

TEM^3-Learning: Time-Efficient Multimodal Multi-Task Learning for Advanced Assistive Driving

Multi-task learning (MTL) can advance assistive driving by exploring inter-task correlations through shared representations. However, existing methods face two critical limitations: single-modality constraints limiting comprehensive scene understanding and inefficient architectures impeding real-time deployment. This paper proposes TEM^3-Learning (Time-Efficient Multimodal Multi-task Learning), a novel framework that jointly optimizes driver emotion recognition, driver behavior recognition, traffic context recognition, and vehicle behavior recognition through a two-stage architecture. The first component, the mamba-based multi-view temporal-spatial feature extraction subnetwork (MTS-Mamba), introduces a forward-backward temporal scanning mechanism and global-local spatial attention to efficiently extract low-cost temporal-spatial features from multi-view sequential images. The second component, the MTL-based gated multimodal feature integrator (MGMI), employs task-specific multi-gating modules to adaptively highlight the most relevant modality features for each task, effectively alleviating the negative transfer problem in MTL. Evaluation on the AIDE dataset, our proposed model achieves state-of-the-art accuracy across all four tasks, maintaining a lightweight architecture with fewer than 6 million parameters and delivering an impressive 142.32 FPS inference speed. Rigorous ablation studies further validate the effectiveness of the proposed framework and the independent contributions of each module. The code is available on https://github.com/Wenzhuo-Liu/TEM3-Learning.

cs.CV↗

arXiv2025

MeshGen: Generating PBR Textured Mesh with Render-Enhanced Auto-Encoder and Generative Data Augmentation

In this paper, we introduce MeshGen, an advanced image-to-3D pipeline that generates high-quality 3D meshes with detailed geometry and physically based rendering (PBR) textures. Addressing the challenges faced by existing 3D native diffusion models, such as suboptimal auto-encoder performance, limited controllability, poor generalization, and inconsistent image-based PBR texturing, MeshGen employs several key innovations to overcome these limitations. We pioneer a render-enhanced point-to-shape auto-encoder that compresses meshes into a compact latent space by designing perceptual optimization with ray-based regularization. This ensures that the 3D shapes are accurately represented and reconstructed to preserve geometric details within the latent space. To address data scarcity and image-shape misalignment, we further propose geometric augmentation and generative rendering augmentation techniques, which enhance the model's controllability and generalization ability, allowing it to perform well even with limited public datasets. For the texture generation, MeshGen employs a reference attention-based multi-view ControlNet for consistent appearance synthesis. This is further complemented by our multi-view PBR decomposer that estimates PBR components and a UV inpainter that fills invisible areas, ensuring a seamless and consistent texture across the 3D mesh. Our extensive experiments demonstrate that MeshGen largely outperforms previous methods in both shape and texture generation, setting a new standard for the quality of 3D meshes generated with PBR textures. See our code at https://github.com/heheyas/MeshGen, project page https://heheyas.github.io/MeshGen

cs.GR↗

arXiv2025

Video4DGen: Enhancing Video and 4D Generation through Mutual Optimization

The advancement of 4D (i.e., sequential 3D) generation opens up new possibilities for lifelike experiences in various applications, where users can explore dynamic objects or characters from any viewpoint. Meanwhile, video generative models are receiving particular attention given their ability to produce realistic and imaginative frames. These models are also observed to exhibit strong 3D consistency, indicating the potential to act as world simulators. In this work, we present Video4DGen, a novel framework that excels in generating 4D representations from single or multiple generated videos as well as generating 4D-guided videos. This framework is pivotal for creating high-fidelity virtual contents that maintain both spatial and temporal coherence. The 4D outputs generated by Video4DGen are represented using our proposed Dynamic Gaussian Surfels (DGS), which optimizes time-varying warping functions to transform Gaussian surfels (surface elements) from a static state to a dynamically warped state. We design warped-state geometric regularization and refinements on Gaussian surfels, to preserve the structural integrity and fine-grained appearance details. To perform 4D generation from multiple videos and capture representation across spatial, temporal, and pose dimensions, we design multi-video alignment, root pose optimization, and pose-guided frame sampling strategies. The leveraging of continuous warping fields also enables a precise depiction of pose, motion, and deformation over per-video frames. Further, to improve the overall fidelity from the observation of all camera poses, Video4DGen performs novel-view video generation guided by the 4D content, with the proposed confidence-filtered DGS to enhance the quality of generated sequences. With the ability of 4D and video generation, Video4DGen offers a powerful tool for applications in virtual reality, animation, and beyond.

cs.GR↗

arXiv2025

MMTL-UniAD: A Unified Framework for Multimodal and Multi-Task Learning in Assistive Driving Perception

Advanced driver assistance systems require a comprehensive understanding of the driver's mental/physical state and traffic context but existing works often neglect the potential benefits of joint learning between these tasks. This paper proposes MMTL-UniAD, a unified multi-modal multi-task learning framework that simultaneously recognizes driver behavior (e.g., looking around, talking), driver emotion (e.g., anxiety, happiness), vehicle behavior (e.g., parking, turning), and traffic context (e.g., traffic jam, traffic smooth). A key challenge is avoiding negative transfer between tasks, which can impair learning performance. To address this, we introduce two key components into the framework: one is the multi-axis region attention network to extract global context-sensitive features, and the other is the dual-branch multimodal embedding to learn multimodal embeddings from both task-shared and task-specific features. The former uses a multi-attention mechanism to extract task-relevant features, mitigating negative transfer caused by task-unrelated features. The latter employs a dual-branch structure to adaptively adjust task-shared and task-specific parameters, enhancing cross-task knowledge transfer while reducing task conflicts. We assess MMTL-UniAD on the AIDE dataset, using a series of ablation studies, and show that it outperforms state-of-the-art methods across all four tasks. The code is available on https://github.com/Wenzhuo-Liu/MMTL-UniAD.

cs.CV↗

arXiv2025

A full breakthrough in vacuum ultraviolet nonlinear optical performance of NH4B4O6F

The lack of suitable vacuum ultraviolet (VUV) nonlinear optical (NLO) crystals has hindered the development of compact, high-power VUV sources via second harmonic generation (SHG). Here, we report on the development of the fluorooxoborate crystal NH4B4O6F (ABF) as a promising material for VUV light generation. For the first time, devices with specific phase-matching angles were constructed, achieving a record 158.9 nm VUV light through phase-matching SHG and a maximum nanosecond pulse energy of 4.8 mJ at 177.3 nm with a conversion efficiency of 5.9 %. The enhanced NLO performance is attributed to optimized arrangements of fluorine-based units creating asymmetric sublattices. This work marks a significant milestone in the field of NLO materials, facilitating the future applications of compact, high-power VUV lasers utilizing ABF.

cond-mat.mtrl-sci↗

arXiv2025

Nature Language Model: Deciphering the Language of Nature for Scientific Discovery

Foundation models have revolutionized natural language processing and artificial intelligence, significantly enhancing how machines comprehend and generate human languages. Inspired by the success of these foundation models, researchers have developed foundation models for individual scientific domains, including small molecules, materials, proteins, DNA, RNA and even cells. However, these models are typically trained in isolation, lacking the ability to integrate across different scientific domains. Recognizing that entities within these domains can all be represented as sequences, which together form the "language of nature", we introduce Nature Language Model (NatureLM), a sequence-based science foundation model designed for scientific discovery. Pre-trained with data from multiple scientific domains, NatureLM offers a unified, versatile model that enables various applications including: (i) generating and optimizing small molecules, proteins, RNA, and materials using text instructions; (ii) cross-domain generation/design, such as protein-to-molecule and protein-to-RNA generation; and (iii) top performance across different domains, matching or surpassing state-of-the-art specialist models. NatureLM offers a promising generalist approach for various scientific tasks, including drug discovery (hit generation/optimization, ADMET optimization, synthesis), novel material design, and the development of therapeutic proteins or nucleotides. We have developed NatureLM models in different sizes (1 billion, 8 billion, and 46.7 billion parameters) and observed a clear improvement in performance as the model size increases.

cs.AI↗

arXiv2024

MeshAnything V2: Artist-Created Mesh Generation With Adjacent Mesh Tokenization

Meshes are the de facto 3D representation in the industry but are labor-intensive to produce. Recently, a line of research has focused on autoregressively generating meshes. This approach processes meshes into a sequence composed of vertices and then generates them vertex by vertex, similar to how a language model generates text. These methods have achieved some success but still struggle to generate complex meshes. One primary reason for this limitation is their inefficient tokenization methods. To address this issue, we introduce MeshAnything V2, an advanced mesh generation model designed to create Artist-Created Meshes that align precisely with specified shapes. A key innovation behind MeshAnything V2 is our novel Adjacent Mesh Tokenization (AMT) method. Unlike traditional approaches that represent each face using three vertices, AMT optimizes this by employing a single vertex wherever feasible, effectively reducing the token sequence length by about half on average. This not only streamlines the tokenization process but also results in more compact and well-structured sequences, enhancing the efficiency of mesh generation. With these improvements, MeshAnything V2 effectively doubles the face limit compared to previous models, delivering superior performance without increasing computational costs. We will make our code and models publicly available. Project Page: https://buaacyw.github.io/meshanything-v2/

cs.CV↗

arXiv2024

DimensionX: Create Any 3D and 4D Scenes from a Single Image with Controllable Video Diffusion

In this paper, we introduce \textbf{DimensionX}, a framework designed to generate photorealistic 3D and 4D scenes from just a single image with video diffusion. Our approach begins with the insight that both the spatial structure of a 3D scene and the temporal evolution of a 4D scene can be effectively represented through sequences of video frames. While recent video diffusion models have shown remarkable success in producing vivid visuals, they face limitations in directly recovering 3D/4D scenes due to limited spatial and temporal controllability during generation. To overcome this, we propose ST-Director, which decouples spatial and temporal factors in video diffusion by learning dimension-aware LoRAs from dimension-variant data. This controllable video diffusion approach enables precise manipulation of spatial structure and temporal dynamics, allowing us to reconstruct both 3D and 4D representations from sequential frames with the combination of spatial and temporal dimensions. Additionally, to bridge the gap between generated videos and real-world scenes, we introduce a trajectory-aware mechanism for 3D generation and an identity-preserving denoising strategy for 4D generation. Extensive experiments on various real-world and synthetic datasets demonstrate that DimensionX achieves superior results in controllable video generation, as well as in 3D and 4D scene generation, compared with previous methods.

cs.CV↗

arXiv2024

Freeplane: Unlocking Free Lunch in Triplane-Based Sparse-View Reconstruction Models

Creating 3D assets from single-view images is a complex task that demands a deep understanding of the world. Recently, feed-forward 3D generative models have made significant progress by training large reconstruction models on extensive 3D datasets, with triplanes being the preferred 3D geometry representation. However, effectively utilizing the geometric priors of triplanes, while minimizing artifacts caused by generated inconsistent multi-view images, remains a challenge. In this work, we present \textbf{Fre}quency modulat\textbf{e}d tri\textbf{plane} (\textbf{Freeplane}), a simple yet effective method to improve the generation quality of feed-forward models without additional training. We first analyze the role of triplanes in feed-forward methods and find that the inconsistent multi-view images introduce high-frequency artifacts on triplanes, leading to low-quality 3D meshes. Based on this observation, we propose strategically filtering triplane features and combining triplanes before and after filtering to produce high-quality textured meshes. These techniques incur no additional cost and can be seamlessly integrated into pre-trained feed-forward models to enhance their robustness against the inconsistency of generated multi-view images. Both qualitative and quantitative results demonstrate that our method improves the performance of feed-forward models by simply modulating triplanes. All you need is to modulate the triplanes during inference.

cs.CV↗

arXiv2024

Vidu4D: Single Generated Video to High-Fidelity 4D Reconstruction with Dynamic Gaussian Surfels

Video generative models are receiving particular attention given their ability to generate realistic and imaginative frames. Besides, these models are also observed to exhibit strong 3D consistency, significantly enhancing their potential to act as world simulators. In this work, we present Vidu4D, a novel reconstruction model that excels in accurately reconstructing 4D (i.e., sequential 3D) representations from single generated videos, addressing challenges associated with non-rigidity and frame distortion. This capability is pivotal for creating high-fidelity virtual contents that maintain both spatial and temporal coherence. At the core of Vidu4D is our proposed Dynamic Gaussian Surfels (DGS) technique. DGS optimizes time-varying warping functions to transform Gaussian surfels (surface elements) from a static state to a dynamically warped state. This transformation enables a precise depiction of motion and deformation over time. To preserve the structural integrity of surface-aligned Gaussian surfels, we design the warped-state geometric regularization based on continuous warping fields for estimating normals. Additionally, we learn refinements on rotation and scaling parameters of Gaussian surfels, which greatly alleviates texture flickering during the warping process and enhances the capture of fine-grained appearance details. Vidu4D also contains a novel initialization state that provides a proper start for the warping fields in DGS. Equipping Vidu4D with an existing video generative model, the overall framework demonstrates high-fidelity text-to-4D generation in both appearance and geometry.

cs.CV↗

arXiv2024

Point Resampling and Ray Transformation Aid to Editable NeRF Models

In NeRF-aided editing tasks, object movement presents difficulties in supervision generation due to the introduction of variability in object positions. Moreover, the removal operations of certain scene objects often lead to empty regions, presenting challenges for NeRF models in inpainting them effectively. We propose an implicit ray transformation strategy, allowing for direct manipulation of the 3D object's pose by operating on the neural-point in NeRF rays. To address the challenge of inpainting potential empty regions, we present a plug-and-play inpainting module, dubbed differentiable neural-point resampling (DNR), which interpolates those regions in 3D space at the original ray locations within the implicit space, thereby facilitating object removal & scene inpainting tasks. Importantly, employing DNR effectively narrows the gap between ground truth and predicted implicit features, potentially increasing the mutual information (MI) of the features across rays. Then, we leverage DNR and ray transformation to construct a point-based editable NeRF pipeline PR^2T-NeRF. Results primarily evaluated on 3D object removal & inpainting tasks indicate that our pipeline achieves state-of-the-art performance. In addition, our pipeline supports high-quality rendering visualization for diverse editing operations without necessitating extra supervision.

cs.CV↗