arXiv subjects

Jiahao Wu

Publications and source records attributed to Jiahao Wu.

At least 37 records · Page 2Linked to original sources

arXiv2026

ClipGStream: Clip-Stream Gaussian Splatting for Any Length and Any Motion Multi-View Dynamic Scene Reconstruction

Dynamic 3D scene reconstruction is essential for immersive media such as VR, MR, and XR, yet remains challenging for long multi-view sequences with large-scale motion. Existing dynamic Gaussian approaches are either Frame-Stream, offering scalability but poor temporal stability, or Clip, achieving local consistency at the cost of high memory and limited sequence length. We propose ClipGStream, a hybrid reconstruction framework that performs stream optimization at the clip level rather than the frame level. The sequence is divided into short clips, where dynamic motion is modeled using clip-independent spatio-temporal fields and residual anchor compensation to capture local variations efficiently, while inter-clip inherited anchors and decoders maintain structural consistency across clips. This Clip-Stream design enables scalable, flicker-free reconstruction of long dynamic videos with high temporal coherence and reduced memory overhead. Extensive experiments demonstrate that ClipGStream achieves state-of-the-art reconstruction quality and efficiency. The project page is available at: https://liangjie1999.github.io/ClipGStreamWeb/

cs.CV↗

arXiv2026

LLM-Driven Instance-Specific Heuristic Generation and Selection

Combinatorial optimization problems are widely encountered in real-world applications. A critical research challenge lies in designing high-quality heuristic algorithms that efficiently approximate optimal solutions within a reasonable time. In recent years, many works have explored integrating Large Language Models (LLMs) with Evolutionary Algorithms to automate heuristic algorithm design through prompt engineering. However, these approaches generally adopt a problem-specific paradigm, applying a single algorithm across all problem instances, failing to account for the heterogeneity across instances. In this paper, we propose InstSpecHH, a novel framework that introduces the concept of instance-specific heuristic generation. InstSpecHH partitions the overall problem class into sub-classes based on instance features and performs differentiated, automated heuristic design for each problem subclass. By tailoring heuristics to the unique features of different sub-classes, InstSpecHH achieves better performance at the problem class level while avoiding redundant heuristic generation for similar instances, thus reducing computational overhead. This approach effectively balances the trade-off between the cost of automatic heuristic design and the quality of the obtained solutions. To evaluate the performance of InstSpecHH, we conduct comprehensive experiments on 4,500 subclasses of the Online Bin Packing Problem (OBPP) and 365 subclasses of the Capacitated Vehicle Routing Problem (CVRP). Experimental results show that InstSpecHH demonstrates strong intra-subclass and inter-subclass generalization capabilities. Compared to previous problem-specific methods, InstSpecHH reduces the average optimality gap by 6.06\% for OBPP and 0.66\% for CVRP. These results highlight the potential of instance-aware automatic heuristic design to further enhance solution quality.

cs.NE↗

arXiv2026

Intrinsic Geometry-Appearance Consistency Optimization for Sparse-View Gaussian Splatting

3D human reconstruction from a single image is a challenging problem and has been exclusively studied in the literature. Recently, some methods have resorted to diffusion models for guidance, optimizing a 3D representation via Score Distillation Sampling(SDS) or generating a back-view image for facilitating reconstruction. However, these methods tend to produce unsatisfactory artifacts (\textit{e.g.} flattened human structure or over-smoothing results caused by inconsistent priors from multiple views) and struggle with real-world generalization in the wild. In this work, we present \emph{MVD-HuGaS}, enabling free-view 3D human rendering from a single image via a multi-view human diffusion model. We first generate multi-view images from the single reference image with an enhanced multi-view diffusion model, which is well fine-tuned on high-quality 3D human datasets to incorporate 3D geometry priors and human structure priors. To infer accurate camera poses from the sparse generated multi-view images for reconstruction, an alignment module is introduced to facilitate joint optimization of 3D Gaussians and camera poses. Furthermore, we propose a depth-based Facial Distortion Mitigation module to refine the generated facial regions, thereby improving the overall fidelity of the reconstruction. Finally, leveraging the refined multi-view images, along with their accurate camera poses, MVD-HuGaS optimizes the 3D Gaussians of the target human for high-fidelity free-view renderings. Extensive experiments on Thuman2.0 and 2K2K datasets show that the proposed MVD-HuGaS achieves state-of-the-art performance on single-view 3D human rendering.

cs.CV↗

arXiv2026

PEAR: Pixel-aligned Expressive humAn mesh Recovery

Reconstructing detailed 3D human meshes from a single in-the-wild image remains a fundamental challenge in computer vision. Existing SMPLX-based methods often suffer from slow inference, produce only coarse body poses, and exhibit misalignments or unnatural artifacts in fine-grained regions such as the face and hands. These issues make current approaches difficult to apply to downstream tasks. To address these challenges, we propose PEAR-a fast and robust framework for pixel-aligned expressive human mesh recovery. PEAR explicitly tackles three major limitations of existing methods: slow inference, inaccurate localization of fine-grained human pose details, and insufficient facial expression capture. Specifically, to enable real-time SMPLX parameter inference, we depart from prior designs that rely on high resolution inputs or multi-branch architectures. Instead, we adopt a clean and unified ViT-based model capable of recovering coarse 3D human geometry. To compensate for the loss of fine-grained details caused by this simplified architecture, we introduce pixel-level supervision to optimize the geometry, significantly improving the reconstruction accuracy of fine-grained human details. To make this approach practical, we further propose a modular data annotation strategy that enriches the training data and enhances the robustness of the model. Overall, PEAR is a preprocessing-free framework that can simultaneously infer EHM-s (SMPLX and scaled-FLAME) parameters at over 100 FPS. Extensive experiments on multiple benchmark datasets demonstrate that our method achieves substantial improvements in pose estimation accuracy compared to previous SMPLX-based approaches. Project page: https://wujh2001.github.io/PEAR

cs.CV↗

arXiv2026

MANGO:Natural Multi-speaker 3D Talking Head Generation via 2D-Lifted Enhancement

Current audio-driven 3D head generation methods mainly focus on single-speaker scenarios, lacking natural, bidirectional listen-and-speak interaction. Achieving seamless conversational behavior, where speaking and listening states transition fluidly remains a key challenge. Existing 3D conversational avatar approaches rely on error-prone pseudo-3D labels that fail to capture fine-grained facial dynamics. To address these limitations, we introduce a novel two-stage framework MANGO, which leveraging pure image-level supervision by alternately training to mitigate the noise introduced by pseudo-3D labels, thereby achieving better alignment with real-world conversational behaviors. Specifically, in the first stage, a diffusion-based transformer with a dual-audio interaction module models natural 3D motion from multi-speaker audio. In the second stage, we use a fast 3D Gaussian Renderer to generate high-fidelity images and provide 2D-level photometric supervision for the 3D motions through alternate training. Additionally, we introduce MANGO-Dialog, a high-quality dataset with over 50 hours of aligned 2D-3D conversational data across 500+ identities. Extensive experiments demonstrate that our method achieves exceptional accuracy and realism in modeling two-person 3D dialogue motion, significantly advancing the fidelity and controllability of audio-driven talking heads.

cs.CV↗

arXiv2025

MelTok: 2D Tokenization for Single-Codebook Audio Compression

Large Audio Language Models (LALMs) have emerged with strong performance across diverse audio understanding tasks and can be further enhanced by neural audio codecs. Transitioning from multi-layer residual vector quantizers to a single-layer quantizer has been shown to facilitate more efficient downstream language models decoding. However, the ability of a single codebook to capture fine-grained acoustic details remains limited, as the frequency-variant nature of 1D tokenizers leads to redundancy. To address this issue, we propose MelTok, a two-dimensional (2D) tokenizer that effectively compresses acoustic details of 44.1 KHz audio into a single codebook. The tokenizer encodes audio into a more compact representation than one-dimensional tokenizers. Furthermore, to recover audio from mel-spectrogram tokens, we propose a token-based vocoder. Both objective and subjective evaluations demonstrate that MelTok achieves quality comparable to multi-codebook codecs and outperforms existing state-of-the-art neural codecs with a single codebook on high-fidelity audio reconstruction. By preserving acoustic details, MelTok offers a strong representation for downstream understanding tasks.

cs.SD↗

arXiv2025

A Neuro-Symbolic Framework for Reasoning under Perceptual Uncertainty: Bridging Continuous Perception and Discrete Symbolic Planning

Bridging continuous perceptual signals and discrete symbolic reasoning is a fundamental challenge in AI systems that must operate under uncertainty. We present a neuro-symbolic framework that explicitly models and propagates uncertainty from perception to planning, providing a principled connection between these two abstraction levels. Our approach couples a transformer-based perceptual front-end with graph neural network (GNN) relational reasoning to extract probabilistic symbolic states from visual observations, and an uncertainty-aware symbolic planner that actively gathers information when confidence is low. We demonstrate the framework's effectiveness on tabletop robotic manipulation as a concrete application: the translator processes 10,047 PyBullet-generated scenes (3--10 objects) and outputs probabilistic predicates with calibrated confidences (overall F1=0.68). When embedded in the planner, the system achieves 94\%/90\%/88\% success on Simple Stack, Deep Stack, and Clear+Stack benchmarks (90.7\% average), exceeding the strongest POMDP baseline by 10--14 points while planning within 15\,ms. A probabilistic graphical-model analysis establishes a quantitative link between calibrated uncertainty and planning convergence, providing theoretical guarantees that are validated empirically. The framework is general-purpose and can be applied to any domain requiring uncertainty-aware reasoning from perceptual input to symbolic planning.

cs.AI↗

arXiv2025

Physics-informed machine learning for combustion: A review

Physics-informed machine learning (PIML) represents an emerging paradigm that integrates various forms of physical knowledge into machine learning (ML) components, thereby enhancing the physical consistency of ML models compared to purely data-driven paradigms. The field of combustion, characterized by a rich foundation of physical laws and abundant data, is undergoing a transformation due to PIML. This paper aims to provide a comprehensive overview of PIML for combustion, systematically outlining fundamental principles, significant contributions, key advancements, and available resources. The application of PIML in combustion is categorized into three domains: combustion chemical kinetics, combustion reacting flows, and other combustion-related scenarios. Additionally, current challenges, potential solutions, and practical guidelines for researchers and engineers will be discussed. A primary focus of this review is to demonstrate how combustion laws can be integrated into ML, either through soft or hard constraints, via loss functions or representation models, and within coordinate-to-variable or field-to-field paradigms. This paper shows that PIML offers a unified framework linking physics, model, and data in combustion--integrating physical knowledge in model-to-data simulation and reconstruction tasks, as well as data-to-model modeling tasks--resulting in enhanced data, improved physical models, and more reliable ML models. PIML for combustion presents significant opportunities for both the combustion and ML communities, encouraging greater collaboration and cross-disciplinary engagement.

physics.chem-ph↗

arXiv2025

NTIRE 2025 Challenge on Low Light Image Enhancement: Methods and Results

This paper presents a comprehensive review of the NTIRE 2025 Low-Light Image Enhancement (LLIE) Challenge, highlighting the proposed solutions and final outcomes. The objective of the challenge is to identify effective networks capable of producing brighter, clearer, and visually compelling images under diverse and challenging conditions. A remarkable total of 762 participants registered for the competition, with 28 teams ultimately submitting valid entries. This paper thoroughly evaluates the state-of-the-art advancements in LLIE, showcasing the significant progress.

cs.CV↗

arXiv2025

Jenga-Krotov algorithm: Efficient compilation of multi-qubit gates for exchange-only qubits

Exchange-only (EO) qubits, implemented in triple-quantum-dot systems, offer a compelling platform for scalable semiconductor-based quantum computing by enabling universal control through purely exchange interactions. While high-fidelity single- and two-qubit gates have been demonstrated, the synthesis of efficient multi-qubit operations-such as the Toffoli gate-remains a key bottleneck. Conventional gate decompositions into elementary operations lead to prohibitively long and error-prone pulse sequences, limiting practical deployment. In this work, we introduce a gradient-based optimization algorithm, Jenga-Krotov (JK), tailored to discover compact, high-fidelity EO gate sequences. Applying JK to the Toffoli gate, we reduce the number of required exchange unitaries from 216 (in direct decomposition) to 92, and compress the time steps required from 162 to 50, all while maintaining target fidelity. Under realistic noise, the accumulated gate error from our optimized sequence is an order of magnitude lower than that of conventional approaches. We have also applied the JK algorithm to other multi-qubit gates and algorithm. For the Fredkin gate, it reduces the number of time steps from 200 to 104 and the number of exchange unitaries from 276 to 172. For the quantum Fourier transform, it compresses the sequence from 180 to 80 time steps and from 237 to 202 exchange unitaries. These results demonstrate that the JK algorithm is a general and scalable strategy for multi-qubit gate synthesis in EO architectures, potentially facilitating realization of multi-qubit algorithms on semiconductor platforms.

quant-ph↗

arXiv2025

Quantification of Electrolyte Degradation in Lithium-ion Batteries with Neutron Imaging Techniques

Non-destructive characterization of lithium-ion batteries provides critical insights for optimizing performance and lifespan while preserving structural integrity. Optimizing electrolyte design in commercial LIBs requires consideration of composition, electrolyte-to-capacity ratio, spatial distribution, and associated degradation pathways. However, existing non-destructive methods for studying electrolyte infiltration, distribution, and degradation in LIBs lack the spatiotemporal resolution required for precise observation and quantification of the electrolyte. In this study, we employ neutron imaging with sufficient spatial resolution ~150 um and large field of view 20x20 cm2 to quantitatively resolve the electrolyte inventory and distribution within LiFePO4/graphite pouch cells under high-temperature accelerated aging. Quantitative standard curves based on neutron transmission attenuation reveal a clear electrolyte dry-out threshold at 3.18 g Ah-1 and the two stages evolutions of EI during cell aging were quantified. By integrating non-destructive electrochemical diagnostics, accelerated graphite material loss and liquid phase Li+ diffusion degradation is observed during pore-drying. Further analysis, including operando cyclic aging, reveals that the neutron transmission below the saturation reference is due to the enrichment of hydrogen nuclei within the solid-electrolyte interphase. Assumed pore-drying does not occur, the SEI signal of the electrodes can be quantitatively decoupled during ageing. Combined analyses with NI, TOF-SIMS, and SEM reveal that high EI cells exhibit uniform SEI growth and reduced degradation, while low EI cells show uneven SEI formation, accelerating capacity loss. This study unveils a dynamic electrolyte infiltration-consumption-dry-out process in LIBs, offering non-destructive and quantitative insights to guide sustainable and durable battery development.

physics.app-ph↗

arXiv2025

OneForecast: A Universal Framework for Global and Regional Weather Forecasting

Accurate weather forecasts are important for disaster prevention, agricultural planning, etc. Traditional numerical weather prediction (NWP) methods offer physically interpretable high-accuracy predictions but are computationally expensive and fail to fully leverage rapidly growing historical data. In recent years, deep learning models have made significant progress in weather forecasting, but challenges remain, such as balancing global and regional high-resolution forecasts, excessive smoothing in extreme event predictions, and insufficient dynamic system modeling. To address these issues, this paper proposes a global-regional nested weather forecasting framework (OneForecast) based on graph neural networks. By combining a dynamic system perspective with multi-grid theory, we construct a multi-scale graph structure and densify the target region to capture local high-frequency features. We introduce an adaptive messaging mechanism, using dynamic gating units to deeply integrate node and edge features for more accurate extreme event forecasting. For high-resolution regional forecasts, we propose a neural nested grid method to mitigate boundary information loss. Experimental results show that OneForecast performs excellently across global to regional scales and short-term to long-term forecasts, especially in extreme event predictions. Codes link https://github.com/YuanGao-YG/OneForecast.

cs.LG↗

arXiv2025

Trapping of Single Atoms in Metasurface Optical Tweezer Arrays

Optical tweezer arrays have emerged as a key experimental platform for quantum computation, quantum simulation, and quantum metrology, enabling unprecedented levels of control over single atoms and molecules. However, existing tweezer platforms have fundamental limitations in array geometry, size, and scalability. Here we demonstrate the trapping of single strontium atoms in optical tweezer arrays generated via holographic metasurfaces. We realize two dimensional arrays with more than 1000 trapped atoms, arranged in arbitrary geometries with trap spacings as small as 1.5 um. The arrays have a high uniformity in terms of trap depth, trap frequency, and positional accuracy, rivaling or surpassing existing approaches. This is enabled by highly efficient holographic metasurfaces fabricated from high-refractive index materials, silicon-rich silicon nitride and titanium dioxide. Leveraging sub-micrometer pixel sizes and high pixel densities, our platform allows scaling far beyond current capabilities. As a demonstration, we realize an optical tweezer array with 360,000 traps. These advances will facilitate tweezer-array based quantum applications that require large system sizes.

physics.atom-ph↗

arXiv2025

Robot Metabolism: Towards machines that can grow by consuming other machines

Biological lifeforms can heal, grow, adapt, and reproduce -- abilities essential for sustained survival and development. In contrast, robots today are primarily monolithic machines with limited ability to self-repair, physically develop, or incorporate material from their environments. While robot minds rapidly evolve new behaviors through AI, their bodies remain closed systems, unable to systematically integrate material to grow or heal. We argue that open-ended physical adaptation is only possible when robots are designed using a small repertoire of simple modules. This allows machines to mechanically adapt by consuming parts from other machines or their surroundings and shed broken components. We demonstrate this principle on a truss modular robot platform. We show how robots can grow bigger, faster, and more capable by consuming materials from their environment and other robots. We suggest that machine metabolic processes like those demonstrated here will be an essential part of any sustained future robot ecology.

cs.RO↗

arXiv2025

LocalDyGS: Multi-view Global Dynamic Scene Modeling via Adaptive Local Implicit Feature Decoupling

Due to the complex and highly dynamic motions in the real world, synthesizing dynamic videos from multi-view inputs for arbitrary viewpoints is challenging. Previous works based on neural radiance field or 3D Gaussian splatting are limited to modeling fine-scale motion, greatly restricting their application. In this paper, we introduce LocalDyGS, which consists of two parts to adapt our method to both large-scale and fine-scale motion scenes: 1) We decompose a complex dynamic scene into streamlined local spaces defined by seeds, enabling global modeling by capturing motion within each local space. 2) We decouple static and dynamic features for local space motion modeling. A static feature shared across time steps captures static information, while a dynamic residual field provides time-specific features. These are combined and decoded to generate Temporal Gaussians, modeling motion within each local space. As a result, we propose a novel dynamic scene reconstruction framework to model highly dynamic real-world scenes more realistically. Our method not only demonstrates competitive performance on various fine-scale datasets compared to state-of-the-art (SOTA) methods, but also represents the first attempt to model larger and more complex highly dynamic scenes. Project page: https://wujh2001.github.io/LocalDyGS/.

cs.CV↗

arXiv2025

Enhancement of quantum coherence in solid-state qubits via interface engineering

Shallow nitrogen-vacancy (NV) centers in diamond are promising quantum sensors but suffer from noise-induced short coherence times due to bulk and surface impurities. We present interfacial engineering via oxygen termination and graphene patching, extending shallow NV coherence to over 1 ms, approaching the T1 limit. Raman spectroscopy and density-functional theory reveal surface termination-driven graphene charge transfer reduces spin noise by pairing surface electrons, supported by double electron-electron resonance spectroscopy showing fewer unpaired spins. Enhanced sensitivity enables detection of single weakly coupled 13C nuclear spins and external 11B spins from a hexagonal boron nitride (h-BN) layer, achieving nanoscale nuclear magnetic resonance. A protective h-BN top layer stabilizes the platform, ensuring robustness against harsh treatments and compatibility with target materials. This integrated approach advances practical quantum sensing by combining extended coherence, improved sensitivity, and device durability.

cond-mat.mes-hall↗

arXiv2025

The Second Order 2D Behaviors of a 3D Bose Gases in the Gross-Pitaevskii Regime

We consider a system of $N$ bosons interacting in a three-dimensional box endowed with periodic boundary condition that is strongly confined in one direction such that the normalized thickness of the box $d\ll1$. We assume particles to interact through a repulsive, radially symmetric and short-range interaction potential with scattering length scale $a\ll d$. We present a comprehensive study of such system in the Gross-Pitaevskii regime, up to the second order ground state energy, starting from proving optimal Bose-Einstein condensation results which were not previously available. The fine interplay between the parameters $N$, $a$ and $d$ generates three regions. Our result in one region on the one hand, is compatible with the classical three-dimensional Lee-Huang-Yang formula. On the other hand, it reveals a new mechanism exhibiting how the second order correction compensates and modifies the first order energy, which was previously thought of as containing a jump, and thus explains how a three-dimensional Bose gas system smoothly transits into two-dimensional system. Moreover, delving into the analysis of this new mechanism exclusive to the second order, we discover a dimensional coupling correlation effect, deeply buried away from the expected 3D and quasi-2D renormalizations, and calculate a new second order correction to the ground state energy.

math-ph↗

arXiv2025

The second order Huang-Yang approximation to the Fermi thermodynamic pressure

We consider a dilute Fermi gas in the thermodynamic limit with interaction potential scattering length $\mathfrak{a}_0$ at temperature $T>0$. We prove the 2nd order Huang-Yang approximation for the Fermi pressure of the system, in which there is a 2nd order term carrying the positive temperature efffect.Our formula is valid up to the temperature $T<ρ^{\frac{2}{3}+\frac{1}{6}}$, which is, by scaling, also necessary for the Huang-Yang formula to hold. Here, $T_F\simρ^{\frac{2}{3}}$ is the Fermi temperature. We also establish during the course of the proof, a conjecture regarding the second order approximation of density $ρ$ by R. Seiringer \cite{FermithermoTpositive}. Our proof uses frequency localization techniques from the analysis of nonlinear PDEs and does not involve spatial localization or Bosonization. In particular, our method covers the classical Huang-Yang formula at zero temperature.

math-ph↗