arXiv subjects

Renjie Chen

Publications and source records attributed to Renjie Chen.

At least 19 recordsLinked to original sources

arXiv2026

A Survey on Self-Improving Test-Time Intelligence: Feedback-Driven Adapting, Learning, and Scaling at Inference

The ability of AI systems to improve their behavior during deployment is becoming increasingly important. As inference moves beyond the static execution of a fixed trained model, a growing body of work studies how models can refine their behavior on the fly by exploiting test-time information and additional computation. These developments have largely evolved along two directions: methods that modify the model's state using test-time signals, and methods that improve predictions through extra inference-time resources such as more sampling and tool use. However, these directions are often studied in separate communities with different terminology, making their connections harder to see. In this survey, we present feedback-driven Test-Time Intelligence (TTI) as a unified perspective for understanding such deployment-time improvement. We use this view to relate test-time adaptation, test-time learning, and test-time scaling, highlighting both their distinctions and their growing overlap in hybrid systems. This unified framework helps connect previously fragmented ideas and provides a clearer conceptual foundation for studying inference-time self-improvement. We review major methodological paradigms, representative applications, and open challenges across vision, language, multimodal learning, generative models, robotics, and healthcare. Our goal is to provide a coherent foundation and research roadmap for the study of self-improving AI systems at test time.

cs.LG↗

arXiv2026

Domain-Varying 2D Green' s Functions for Cage-based Deformation

In this work, we propose a novel theoretical view of cage-based deformation based on domain-varying Green' s functions and treat this domain as a new control space for the deformation effects. Harmonic Coordinates (HC) and Green Coordinates (GC) are classic methods in cage-based deformation and serve as the theoretical foundation for shape editing in a range of practical deformation tools. Our method revisits these two classical approaches. Specifically, we propose a framework based on Green' s functions across diverse domains (independent of the cage-enclosed domain) to unify these two techniques. To our knowledge, this represents the first such attempt in nearly two decades. Based on this perspective, we propose a novel cage-based deformation technique that introduces a new control space and utilizes domain-varying Green' s functions to yield varying deformation effects. Our method also establishes a continuous transition of effects from HC to GC as the Green' s function domain $\Theta$ expands from the cage region $\Omega$ to the entire $\mathbb{R}^2$. We call our method Domain-Varying Green Coordinates (DVGC). When $\Theta$ is a disk or a rectangle, the Green' s function possesses analytic or semi-analytic expressions, respectively, enabling the DVGC to be computed without finite element discretization. Furthermore, when $\Theta$ is a disk, the DVGC admit a closed-form expression for 2D simplicial cages, thereby eliminating the need for numerical integration. Experiments demonstrate that our method provides a novel control space ranging from more consistent with the cage to more shape-preserving, generating diverse deformation effects by varying the Green' s function domains.

cs.GR↗

arXiv2026

Bridging Video Understanding and Generation in a Unified Framework

Recently, unified image generation and understanding have been extensively explored. However, extending such unified modeling paradigms to the video domain remains largely underexplored. A central challenge is that video understanding favors compact, discriminative semantic representations, whereas video generation requires dense signals that preserve visual details and temporal coherence. Videos naturally capture both spatial semantics and temporal dynamics, making them a more suitable modality for unified multimodal modeling compared to static images. In this paper, we propose Vega, a unified framework that bridges video understanding and generation. Vega leverages a shared vocabulary to jointly model text and visual representations and employs a hybrid architecture combining autoregressive (AR) prediction with diffusion-based rendering. Specifically, the AR model focuses on predicting semantically meaningful visual tokens for keyframes, providing a structured representation that guides the diffusion module in rendering dense, high-resolution video frames. Extensive experiments demonstrate that Vega achieves strong performance on video generation benchmarks such as VBench and video understanding benchmarks like VideoMME.

cs.CV↗

arXiv2025

Anisotropic Green Coordinates

We live in a world filled with anisotropy, a ubiquitous characteristic of both natural and engineered systems. In this study, we concentrate on space deformation and introduce Anisotropic Green Coordinates (AGC), which provide versatile effects for cage-based and variational deformations in both two and three dimensions. The AGC are derived from the anisotropic Laplace equation $\nabla\cdot(\mathbf{A}\nabla u)=0$, where $\mathbf{A}$ is a symmetric positive definite (SPD) matrix. Based on this equation, we establish the boundary integral formulation, which is subsequently discretized to derive the deformation coordinates defined on the vertices and normals of oriented simplicial cages. Our method satisfies basic properties such as linear reproduction and translation invariance, and possesses closed-form expressions for both 2D and 3D scenarios. We also give an intuitive geometric interpretation of the approach, demonstrating that our method can generate a quasi-conformal mapping. We demonstrate both theoretically and empirically that the deformation effect is more pronounced when the normal of the cage face aligns with the eigenvector corresponding to the larger eigenvalue of $\mathbf{A}$. This indicates that anisotropy amplifies the deformation sensitivity along this direction, enabling more targeted cage design and matrix selection. Furthermore, we derive the gradients and Hessians of the deformation coordinates and employ the local-global optimization framework to facilitate variational shape deformation, enabling flexible shape manipulation while achieving as-rigid-as-possible (ARAP) shape deformation. Experimental results demonstrate that AGC offer versatile and diverse deformation options, providing artists with enhanced flexibility and introducing a novel perspective on spatial deformation.

cs.GR↗

arXiv2025

NoPe-NeRF++: Local-to-Global Optimization of NeRF with No Pose Prior

In this paper, we introduce NoPe-NeRF++, a novel local-to-global optimization algorithm for training Neural Radiance Fields (NeRF) without requiring pose priors. Existing methods, particularly NoPe-NeRF, which focus solely on the local relationships within images, often struggle to recover accurate camera poses in complex scenarios. To overcome the challenges, our approach begins with a relative pose initialization with explicit feature matching, followed by a local joint optimization to enhance the pose estimation for training a more robust NeRF representation. This method significantly improves the quality of initial poses. Additionally, we introduce global optimization phase that incorporates geometric consistency constraints through bundle adjustment, which integrates feature trajectories to further refine poses and collectively boost the quality of NeRF. Notably, our method is the first work that seamlessly combines the local and global cues with NeRF, and outperforms state-of-the-art methods in both pose estimation accuracy and novel view synthesis. Extensive evaluations on benchmark datasets demonstrate our superior performance and robustness, even in challenging scenes, thus validating our design choices.

cs.CV↗

arXiv2025

SCOPE: Spectral Concentration by Distributionally Robust Joint Covariance-Precision Estimation

We propose a distributionally robust formulation for simultaneously estimating the covariance matrix and the precision matrix of a random vector.The proposed model minimizes the worst-case weighted sum of the Frobenius loss of the covariance estimator and Stein's loss of the precision matrix estimator against all distributions from an ambiguity set centered at the nominal distribution. The radius of the ambiguity set is measured via convex spectral divergence. We demonstrate that the proposed distributionally robust estimation model can be reduced to a convex optimization problem, thereby yielding quasi-analytical estimators. The joint estimators are shown to be nonlinear shrinkage estimators. The eigenvalues of the estimators are shrunk nonlinearly towards a positive scalar, where the scalar is determined by the weight coefficient of the loss terms. By tuning the coefficient carefully, the shrinkage corrects the spectral bias of the empirical covariance/precision matrix estimator. By this property, we call the proposed joint estimator the Spectral concentrated COvariance and Precision matrix Estimator (SCOPE). We demonstrate that the shrinkage effect improves the condition number of the estimator. We provide a parameter-tuning scheme that adjusts the shrinkage target and intensity that is asymptotically optimal. Numerical experiments on synthetic and real data show that our shrinkage estimators perform competitively against state-of-the-art estimators in practical applications.

stat.ML↗

arXiv2025

Test-Time Model Adaptation for Quantized Neural Networks

Quantizing deep models prior to deployment is a widely adopted technique to speed up inference for various real-time applications, such as autonomous driving. However, quantized models often suffer from severe performance degradation in dynamic environments with potential domain shifts and this degradation is significantly more pronounced compared with their full-precision counterparts, as shown by our theoretical and empirical illustrations. To address the domain shift problem, test-time adaptation (TTA) has emerged as an effective solution by enabling models to learn adaptively from test data. Unfortunately, existing TTA methods are often impractical for quantized models as they typically rely on gradient backpropagation--an operation that is unsupported on quantized models due to vanishing gradients, as well as memory and latency constraints. In this paper, we focus on TTA for quantized models to improve their robustness and generalization ability efficiently. We propose a continual zeroth-order adaptation (ZOA) framework that enables efficient model adaptation using only two forward passes, eliminating the computational burden of existing methods. Moreover, we propose a domain knowledge management scheme to store and reuse different domain knowledge with negligible memory consumption, reducing the interference of different domain knowledge and fostering the knowledge accumulation during long-term adaptation. Experimental results on three classical architectures, including quantized transformer-based and CNN-based models, demonstrate the superiority of our methods for quantized model adaptation. On the quantized W6A6 ViT-B model, our ZOA is able to achieve a 5.0\% improvement over the state-of-the-art FOA on ImageNet-C dataset. The source code is available at https://github.com/DengZeshuai/ZOA.

cs.CV↗

arXiv2025

ContentV: Efficient Training of Video Generation Models with Limited Compute

Recent advances in video generation demand increasingly efficient training recipes to mitigate escalating computational costs. In this report, we present ContentV, an 8B-parameter text-to-video model that achieves state-of-the-art performance (85.14 on VBench) after training on 256 x 64GB Neural Processing Units (NPUs) for merely four weeks. ContentV generates diverse, high-quality videos across multiple resolutions and durations from text prompts, enabled by three key innovations: (1) A minimalist architecture that maximizes reuse of pre-trained image generation models for video generation; (2) A systematic multi-stage training strategy leveraging flow matching for enhanced efficiency; and (3) A cost-effective reinforcement learning with human feedback framework that improves generation quality without requiring additional human annotations. All the code and models are available at: https://contentv.github.io.

cs.CV↗

arXiv2025

Towards Self-Improvement of Diffusion Models via Group Preference Optimization

Aligning text-to-image (T2I) diffusion models with Direct Preference Optimization (DPO) has shown notable improvements in generation quality. However, applying DPO to T2I faces two challenges: the sensitivity of DPO to preference pairs and the labor-intensive process of collecting and annotating high-quality data. In this work, we demonstrate that preference pairs with marginal differences can degrade DPO performance. Since DPO relies exclusively on relative ranking while disregarding the absolute difference of pairs, it may misclassify losing samples as wins, or vice versa. We empirically show that extending the DPO from pairwise to groupwise and incorporating reward standardization for reweighting leads to performance gains without explicit data selection. Furthermore, we propose Group Preference Optimization (GPO), an effective self-improvement method that enhances performance by leveraging the model's own capabilities without requiring external data. Extensive experiments demonstrate that GPO is effective across various diffusion models and tasks. Specifically, combining with widely used computer vision models, such as YOLO and OCR, the GPO improves the accurate counting and text rendering capabilities of the Stable Diffusion 3.5 Medium by 20 percentage points. Notably, as a plug-and-play method, no extra overhead is introduced during inference.

cs.CV↗

arXiv2025

TrackGS: Optimizing COLMAP-Free 3D Gaussian Splatting with Global Track Constraints

We present TrackGS, a novel method to integrate global feature tracks with 3D Gaussian Splatting (3DGS) for COLMAP-free novel view synthesis. While 3DGS delivers impressive rendering quality, its reliance on accurate precomputed camera parameters remains a significant limitation. Existing COLMAP-free approaches depend on local constraints that fail in complex scenarios. Our key innovation lies in leveraging feature tracks to establish global geometric constraints, enabling simultaneous optimization of camera parameters and 3D Gaussians. Specifically, we: (1) introduce track-constrained Gaussians that serve as geometric anchors, (2) propose novel 2D and 3D track losses to enforce multi-view consistency, and (3) derive differentiable formulations for camera intrinsics optimization. Extensive experiments on challenging real-world and synthetic datasets demonstrate state-of-the-art performance, with much lower pose error than previous methods while maintaining superior rendering quality. Our approach eliminates the need for COLMAP preprocessing, making 3DGS more accessible for practical applications.

cs.CV↗

arXiv2025

EchoVideo: Identity-Preserving Human Video Generation by Multimodal Feature Fusion

Recent advancements in video generation have significantly impacted various downstream applications, particularly in identity-preserving video generation (IPT2V). However, existing methods struggle with "copy-paste" artifacts and low similarity issues, primarily due to their reliance on low-level facial image information. This dependence can result in rigid facial appearances and artifacts reflecting irrelevant details. To address these challenges, we propose EchoVideo, which employs two key strategies: (1) an Identity Image-Text Fusion Module (IITF) that integrates high-level semantic features from text, capturing clean facial identity representations while discarding occlusions, poses, and lighting variations to avoid the introduction of artifacts; (2) a two-stage training strategy, incorporating a stochastic method in the second phase to randomly utilize shallow facial information. The objective is to balance the enhancements in fidelity provided by shallow features while mitigating excessive reliance on them. This strategy encourages the model to utilize high-level features during training, ultimately fostering a more robust representation of facial identities. EchoVideo effectively preserves facial identities and maintains full-body integrity. Extensive experiments demonstrate that it achieves excellent results in generating high-quality, controllability and fidelity videos.

cs.CV↗

arXiv2025

Flexible 3D Cage-based Deformation via Green Coordinates on B\'{e}zier Patches

Cage-based deformation is a fundamental problem in geometry processing, where a cage, a user-specified boundary of a region, is used to deform the ambient space of a given mesh. Traditional 3D cages are typically composed of triangles and quads. While quads can represent non-planar regions when their four corners are not coplanar, they form ruled surfaces with straight isoparametric curves, which limits their ability to handle curved and high-curvature deformations. In this work, we extend the cage for curved boundaries using B\'{e}zier patches, enabling flexible and high-curvature deformations with only a few control points. The higher-order structure of the B\'{e}zier patch also allows for the creation of a more compact and precise curved cage for the input model. Based on Green's third identity, we derive the Green coordinates for the B\'{e}zier cage, achieving shape-preserving deformation with smooth surface boundaries. These coordinates are defined based on the vertex positions and normals of the B\'{e}zier control net. Given that the coordinates are approximately calculated through the Riemann summation, we propose a global projection technique to ensure that the coordinates accurately conform to the linear reproduction property. Experimental results show that our method achieves high performance in handling curved and high-curvature deformations.

cs.GR↗

arXiv2024

HybridGS: Decoupling Transients and Statics with 2D and 3D Gaussian Splatting

Generating high-quality novel view renderings of 3D Gaussian Splatting (3DGS) in scenes featuring transient objects is challenging. We propose a novel hybrid representation, termed as HybridGS, using 2D Gaussians for transient objects per image and maintaining traditional 3D Gaussians for the whole static scenes. Note that, the 3DGS itself is better suited for modeling static scenes that assume multi-view consistency, but the transient objects appear occasionally and do not adhere to the assumption, thus we model them as planar objects from a single view, represented with 2D Gaussians. Our novel representation decomposes the scene from the perspective of fundamental viewpoint consistency, making it more reasonable. Additionally, we present a novel multi-view regulated supervision method for 3DGS that leverages information from co-visible regions, further enhancing the distinctions between the transients and statics. Then, we propose a straightforward yet effective multi-stage training strategy to ensure robust training and high-quality view synthesis across various settings. Experiments on benchmark datasets show our state-of-the-art performance of novel view synthesis in both indoor and outdoor scenes, even in the presence of distracting elements.

cs.CV↗

arXiv2024

Distributional stability of sparse inverse covariance matrix estimators

Finding an approximation of the inverse of the covariance matrix, also known as precision matrix, of a random vector with empirical data is widely discussed in finance and engineering. In data-driven problems, empirical data may be ``contaminated''. This raises the question as to whether the approximate precision matrix is reliable from a statistical point of view. In this paper, we concentrate on a much-noticed sparse estimator of the precision matrix and investigate the issue from the perspective of distributional stability. Specifically, we derive an explicit local Lipschitz bound for the distance between the distributions of the sparse estimator under two different distributions (regarded as the true data distribution and the distribution of ``contaminated'' data). The distance is measured by the Kantorovich metric on the set of all probability measures on a matrix space. We also present analogous results for the standard estimators of the covariance matrix and its eigenvalues. Furthermore, we discuss several applications and conduct some numerical experiments.

math.ST↗

arXiv2024

Infrared resonance-lattice device technology

We present subwavelength resonant lattices fashioned as nano- and microstructured films as a basis for a host of device concepts. Whereas the canonical physical properties are fully embodied in a one-dimensional periodic lattice, the final device constructs are often patterned in two-dimensionally-modulated films in which case we may refer to them as photonic crystal slabs, metamaterials, or metasurfaces. These surfaces can support lateral modes and localized field signatures with propagative and evanescent diffraction channels critically controlling the response. The governing principle of guided-mode, or lattice, resonance enables diverse spectral expressions such that a single-layer component can behave as a sensor, reflector, filter, or polarizer. This structural sparsity contrasts strongly with the venerable field of multi-layer thin-film optics that is basis for most optical components on the market today. The lattice resonance effect can be exploited in all major spectral regions with appropriate low-loss materials and fabrication resources. In this paper, we highlight resonant device technology and present our work on design, fabrication, and characterization of optical elements operating in the near-IR, mid-IR, and long-wave IR spectral regions. Examples of fabricated and tested devices include biological sensors, high-contrast-ratio polarizers, narrow-band notch filters, and wideband high reflectors.

physics.optics↗

arXiv2024

Learning Neural Volumetric Pose Features for Camera Localization

We introduce a novel neural volumetric pose feature, termed PoseMap, designed to enhance camera localization by encapsulating the information between images and the associated camera poses. Our framework leverages an Absolute Pose Regression (APR) architecture, together with an augmented NeRF module. This integration not only facilitates the generation of novel views to enrich the training dataset but also enables the learning of effective pose features. Additionally, we extend our architecture for self-supervised online alignment, allowing our method to be used and fine-tuned for unlabelled images within a unified framework. Experiments demonstrate that our method achieves 14.28% and 20.51% performance gain on average in indoor and outdoor benchmark scenes, outperforming existing APR methods with state-of-the-art accuracy.

cs.CV↗

arXiv2023

Topological Magnetoresistance of Magnetic Skyrmionic Bubbles

Magnetic skyrmions offer promising prospects for constructing future energy-efficient and high-density information technology, leading to extensive explorations of new skyrmionic materials recently. The topological Hall effect has been widely adopted as a distinctive marker of skyrmion emergence. Alternately, here we propose a novel signature of skyrmion state by quantitatively investigating the magnetoresistance (MR) induced by skyrmionic bubbles in CeMn2Ge2. An intriguing finding was revealed: the anomalous MR measured at different temperatures can be normalized into a single curve, regardless of sample thickness. This behavior can be accurately reproduced by the recent chiral spin textures MR model. Further analysis of the MR anomaly allowed us to quantitatively examine the effective magnetic fields of various scattering channels. Remarkably, the analyses, combined with the Lorentz transmission electronic microscopy results, indicate that the in-plane scattering channel with triplet exchange interactions predominantly governs the magnetotransport in the Bloch-type skyrmionic bubble state. Our results not only provide insights into the quantum correction on MR induced by skyrmionic bubble phase, but also present an electrical probing method for studying chiral spin texture formation, evolution and their topological properties, which opens up exciting possibilities for identifying new skyrmionic materials and advancing the methodology for studying chiral spin textures.

cond-mat.mtrl-sci↗

arXiv2022

PCDNF: Revisiting Learning-based Point Cloud Denoising via Joint Normal Filtering

Recovering high quality surfaces from noisy point clouds, known as point cloud denoising, is a fundamental yet challenging problem in geometry processing. Most of the existing methods either directly denoise the noisy input or filter raw normals followed by updating point positions. Motivated by the essential interplay between point cloud denoising and normal filtering, we revisit point cloud denoising from a multitask perspective, and propose an end-to-end network, named PCDNF, to denoise point clouds via joint normal filtering. In particular, we introduce an auxiliary normal filtering task to help the overall network remove noise more effectively while preserving geometric features more accurately. In addition to the overall architecture, our network has two novel modules. On one hand, to improve noise removal performance, we design a shape-aware selector to construct the latent tangent space representation of the specific point by comprehensively considering the learned point and normal features and geometry priors. On the other hand, point features are more suitable for describing geometric details, and normal features are more conducive for representing geometric structures (e.g., sharp edges and corners). Combining point and normal features allows us to overcome their weaknesses. Thus, we design a feature refinement module to fuse point and normal features for better recovering geometric information. Extensive evaluations, comparisons, and ablation studies demonstrate that the proposed method outperforms state-of-the-arts for both point cloud denoising and normal filtering.

cs.CV↗