arXiv subjects

Yong Zhang

Publications and source records attributed to Yong Zhang.

At least 163 records · Page 9Linked to original sources

arXiv2025

Gas-phase Molecules in Protoplanetary Nebulae with the 21 μm Emission Feature II. Carbon monosulfide

The carrier of the 21 $μ$m emission feature discovered in a handful of protoplanetary nebulae (PPNe) is one of the most intriguing enigmas in circumstellar chemistry. Investigating the gas-phase molecules in PPNe could yield important hints for understanding the 21 $μ$m feature. In this paper, we report observations of the CS $J = 5 \to 4$ line at 245 GHz and the CO $J = 1 \to 0$ line at 115 GHz toward seven PPNe exhibiting the 21 $μ$m feature. We find that CS is extremely scarce in these PPNe and the CS line is only detected in one source, IRAS Z02229+6208. Based on the assumption of local thermal equilibrium and negligible optical depth, we derive that the CS column densities and fractional abundances relative to H$_{2}$ are $N$(CS) < 9.1 ${\times}$ 10$^{13}$cm$^{-2}$ and $f$(CS) < 8.1 ${\times}$ 10$^{-7}$. A comparison of the CS abundances across different circumstellar envelopes reveals that the variations in CS abundance are complex, depending not only on the evolutionary stages but also on the properties of individual objects.

astro-ph.GA↗

arXiv2025

InfiniteTalk: Audio-driven Video Generation for Sparse-Frame Video Dubbing

Recent breakthroughs in video AIGC have ushered in a transformative era for audio-driven human animation. However, conventional video dubbing techniques remain constrained to mouth region editing, resulting in discordant facial expressions and body gestures that compromise viewer immersion. To overcome this limitation, we introduce sparse-frame video dubbing, a novel paradigm that strategically preserves reference keyframes to maintain identity, iconic gestures, and camera trajectories while enabling holistic, audio-synchronized full-body motion editing. Through critical analysis, we identify why naive image-to-video models fail in this task, particularly their inability to achieve adaptive conditioning. Addressing this, we propose InfiniteTalk, a streaming audio-driven generator designed for infinite-length long sequence dubbing. This architecture leverages temporal context frames for seamless inter-chunk transitions and incorporates a simple yet effective sampling strategy that optimizes control strength via fine-grained reference frame positioning. Comprehensive evaluations on HDTF, CelebV-HQ, and EMTD datasets demonstrate state-of-the-art performance. Quantitative metrics confirm superior visual realism, emotional coherence, and full-body motion synchronization.

cs.CV↗

arXiv2025

A compact quasi-zero stiffness metamaterial based on monolithic shells for vibration isolation

Quasi-zero stiffness (QZS) metamaterials are highly effective in isolating objects from low-frequency external vibrations, due to their high static stiffness but low dynamic stiffness characteristics. Traditionally, QZS metamaterials are designed by combining a negative-stiffness part with a positive-stiffness counterpart. Here, we present a novel QZS metamaterial design without relying on combining two components. The QZS characteristic is achieved solely through monolithic shell elements' unique geometry and nonlinear deformation. Using experimental and numerical approaches, we investigate the static and dynamic responses of the proposed metamaterials as a function of their geometric parameters. We then tune the structure's geometry to achieve ideal zero-stiffness behaviors and experimentally demonstrate an exceptional low-frequency vibration isolation mechanism. This concept can be further utilized as a building block for constructing metamaterials with multiple zero-stiffness features, enabling a broad range of applications.

physics.app-ph↗

arXiv2025

High-gain optical parametric amplification with a continuous-wave pump using a domain-engineered thin-film lithium niobate waveguide

While thin film lithium niobate (TFLN) is known for efficient signal generation, on-chip signal amplification remains challenging from fully integrated optical communication circuits. Here we demonstrate the continuous-wave-pump optical parametric amplification (OPA) using an x-cut domain-engineered TFLN waveguide, with high gain over the telecom band up to 13.9 dB, and test it for high signal-to-noise ratio signal amplification using a commercial optical communication module pair. Fabricated in wafer scale using common process as devices including modulators, this OPA device marks an important step in TFLN photonic integration.

physics.optics↗

arXiv2025

SMARTAPS: Tool-augmented LLMs for Operations Management

Large language models (LLMs) present intriguing opportunities to enhance user interaction with traditional algorithms and tools in real-world applications. An advanced planning system (APS) is a sophisticated software that leverages optimization to help operations planners create, interpret, and modify an operational plan. While highly beneficial, many customers are priced out of using an APS due to the ongoing costs of consultants responsible for customization and maintenance. To address the need for a more accessible APS expressed by supply chain planners, we present SmartAPS, a conversational system built on a tool-augmented LLM. Our system provides operations planners with an intuitive natural language chat interface, allowing them to query information, perform counterfactual reasoning, receive recommendations, and execute scenario analysis to better manage their operation. A short video demonstrating the system has been released: https://youtu.be/KtIrJjlDbyw

cs.AI↗

arXiv2025

R-Bot: An LLM-based Query Rewrite System

Query rewrite is essential for optimizing SQL queries to improve their execution efficiency without changing their results. Traditionally, this task has been tackled through heuristic and learning-based methods, each with its limitations in terms of inferior quality and low robustness. Recent advancements in LLMs offer a new paradigm by leveraging their superior natural language and code comprehension abilities. Despite their potential, directly applying LLMs like GPT-4 has faced challenges due to problems such as hallucinations, where the model might generate inaccurate or irrelevant results. To address this, we propose R-Bot, an LLM-based query rewrite system with a systematic approach. We first design a multi-source rewrite evidence preparation pipeline to generate query rewrite evidences for guiding LLMs to avoid hallucinations. We then propose a hybrid structure-semantics retrieval method that combines structural and semantic analysis to retrieve the most relevant rewrite evidences for effectively answering an online query. We next propose a step-by-step LLM rewrite method that iteratively leverages the retrieved evidences to select and arrange rewrite rules with self-reflection. We conduct comprehensive experiments on real-world datasets and widely used benchmarks, and demonstrate the superior performance of our system, R-Bot, surpassing state-of-the-art query rewrite methods. The R-Bot system has been deployed at Huawei and with real customers, and the results show that the proposed R-Bot system achieves lower query latency.

cs.DB↗

arXiv2025

Interval bifurcation theorems for Fredholm operator and its application to an elliptic overdetermined problem in bounded domains

We establish local interval bifurcation theorem and global interval bifurcation theorem for Fredholm operator with index $0$ via $0$-group. As one of applications, we investigate the existence of a family of nontrivial domains $Ω_ρ\subset \mathbb{R}^N$ ($N=2,3$ or $4$), bifurcating from a small ball, such that the problem \begin{equation} -Δu=u-\left(u^+\right)^3\,\, \text{in}\,\,Ω_ρ, \,\, u=0,\,\,\partial_νu=\text{const}\,\,\text{on}\,\,\partialΩ_ρ \nonumber \end{equation} has a sign-changing bounded solution. Compared with the recent result \cite[Theorem 2.1]{Ruiz}, here we obtain a family of domains $Ω_ρ$ instead of a sequence of domains.

math.AP↗

arXiv2025

The equilibrium distribution function for strongly nonlinear systems

The equilibrium distribution function determines macroscopic observables in statistical physics. While conventional methods correct equilibrium distributions in weakly nonlinear or near-integrable systems, they fail in strongly nonlinear regimes. We develop a framework to get the equilibrium distributions and dispersion relations in strongly nonlinear many-body systems, incorporating corrections beyond the random phase approximation and capturing intrinsic nonlinear effects. The theory is verified on the nonlinear Schrodinger equation, the Majda-McLaughlin-Tabak model, and the FPUT-beta model, demonstrating its accuracy across distinct types of nonlinear systems. Numerical results show substantial improvements over existing approaches, even in strong nonlinear regimes. This work establishes a theoretical foundation for equilibrium statistical properties in strongly nonlinear systems.

cond-mat.stat-mech↗

arXiv2025

Revisiting Multi-Wave Resonance in Classical Lattices: Quasi-Resonances, Not Exact Resonance, Govern Energy Redistribution

The multi-wave exact resonance condition is a fundamental principle for understanding energy transfer in condensed matter systems, yet the dynamical evolution of waves satisfying this condition remains unexplored. Here, we reveal that the multi-wave resonant kinetic equations possess distinctive symmetry properties that preferentially induce energy equalization between counter-propagating waves of identical frequency. This initial equalization disrupts the exact resonance condition, rendering it dynamically invalid. We further demonstrate that nonlinearity-mediated multi-wave quasi-resonances--not exact resonances--overn energy transfer and drive the system toward thermalization. Crucially, the strength of exact resonances decays with increasing system size, while quasi-resonance strength grows. Moreover, exact resonance strength remains independent of nonlinearity, whereas quasi-resonance strength diminishes with reduced nonlinearity. These observations provide additional evidence supporting the aforementioned conclusion while elucidating the size-dependent thermalization characteristics in lattice systems.

cond-mat.stat-mech↗

arXiv2025

Surface Reduction Boosts Free Electron Concentration in MXene for Enhanced Photothermal Performance

The photothermal properties of MXenes originate from their high free electron concentration, which drives localized surface plasmon resonance (LSPR). However, their intrinsic electron concentration is limited by suboptimal d-orbital occupancy, while electronegative surface terminations actively deplete free electrons through orbital-selective withdrawal. Herein, we report a sodium (Na)-mediated surface reduction strategy in molten salts to transform Ti3C2Clx into electronically tunable Ti3C2. Specifically, Na atoms remove -Cl terminations to eliminate electron withdrawal and simultaneously inject electrons into the MXene lattice via a reduction reaction. This dual effect saturates Ti 3d-orbital vacancies while reducing surface coordination sites, achieving an increase in carrier concentration to 4.92-fold, an increase in mobility to 2.63-fold, and an enhancement in conductivity to 12.96-fold. Consequently, the optimized reduced MXene achieves a record photothermal conversion efficiency of 92.36% under 808 nm laser irradiation. As a proof-of-concept, a photothermal antibacterial woundplast with ultralow MXene content demonstrates a 91.39% bacterial kill rate. This work not only shows an effective way to tune the photothermal properties of MXenes but also inspires applications that require high electron concentration, such as energy storage, sensors, and electromagnetic interference shielding.

cond-mat.mtrl-sci↗

arXiv2025

FedPall: Prototype-based Adversarial and Collaborative Learning for Federated Learning with Feature Drift

Federated learning (FL) enables collaborative training of a global model in the centralized server with data from multiple parties while preserving privacy. However, data heterogeneity can significantly degrade the performance of the global model when each party uses datasets from different sources to train a local model, thereby affecting personalized local models. Among various cases of data heterogeneity, feature drift, feature space difference among parties, is prevalent in real-life data but remains largely unexplored. Feature drift can distract feature extraction learning in clients and thus lead to poor feature extraction and classification performance. To tackle the problem of feature drift in FL, we propose FedPall, an FL framework that utilizes prototype-based adversarial learning to unify feature spaces and collaborative learning to reinforce class information within the features. Moreover, FedPall leverages mixed features generated from global prototypes and local features to enhance the global classifier with classification-relevant information from a global perspective. Evaluation results on three representative feature-drifted datasets demonstrate FedPall's consistently superior performance in classification with feature-drifted data in the FL scenario.

cs.LG↗

arXiv2025

FAST search for circumstellar atomic hydrogen. III. Planetary nebulae at Galactic latitudes |b|>10degree

The detection of circumstellar atomic hydrogen (\ion{H}{1}) via the 21\,cm line remains a persistent challenge in planetary nebula (PN) studies, primarily due to contamination from ubiquitous interstellar \ion{H}{1} 21\,cm emission. In this paper, we report the results of a \ion{H}{1} survey of 12 high surface brightness PNe located at Galactic latitudes $|b|\geq10\arcdeg$, performed with the Five-hundred-meter Aperture Spherical radio Telescope (FAST), which is currently the most sensitive telescope in the $L$ band. Although the contamination from interstellar emission is still severe, we detect or tentatively detect circumstellar \ion{H}{1} 21\,cm absorption associated with two PNe: NGC\,6905 and NGC\,7662. \ion{H}{1} exhibits a comparatively high detection frequency in bipolar PNe. We estimate the \ion{H}{1} masses of the two PNe to range from 0.01 to 0.14 $M_{\sun}$, resulting in atomic-to-ionized hydrogen ratios of 0.02--0.3. The \ion{H}{1} shells have dynamical ages of 2100--2400 years. Our measurements confirm previous findings that the optical depth of \ion{H}{1} decreases with increasing linear radius of the nebula. The mass loss rates traced by the \ion{H}{1} absorption are larger than 10$^{-5}$\,M$_\sun$ yr$^{-1}$, indicating that they originate from the superwind phase at the tip of the asymptotic giant branch.

astro-ph.GA↗

arXiv2025

A 3mm molecular line survey toward IRC+10216

IRC+10216 is the brightest infrared source in the northern sky, known for its rich chemical composition. It is often used as a standard reference for studying the circumstellar envelope (CSE) of carbon-rich stars. While pioneering 3\,mm spectral surveys have laid foundational datasets, their system temperature limitations rendered spectral line detection thresholds inadequate for probing the source's complex organic molecule inventory at this band, which made superseding observations necessary. We aim to gain an unbiased view regarding circumstellar chemistry and investigate whether IRC+10216 is typical or anomalous in terms of its chemical composition. We carried out an in-depth spectral line survey of the circumstellar envelope of IRC+10216 utilizing the Arizona Radio Observatory 12\,m telescope. We achieved complete spectral sampling across the 90--116 GHz atmospheric window ($λ=2.6$--3.3\,mm). A total of 214 emission lines belonging to 43 molecular species are identified in the CSE of IRC+10216, among which 28 lines are newly detected in this object and four emission lines remain unidentified. The excitation temperatures and column densities of 16 molecules are determined through rotation diagrams. We estimate the isotopic ratios of carbon, oxygen, and silicon elements. For the majority of the molecular species, the line intensity ratios between IRC+10216 and CIT\,6 are inversely proportional to the square of their distance, which suggests that the chemical processes occurring within them are similar. Nevertheless, there is evidence suggesting that the emission of C$_{4}$H and C$_{3}$N in IRC+10216 is unusually strong. These observations stand as the most sensitive and unbiased line survey of IRC+10216 within the $λ=3$ window carried out by a single-dish telescope. They offer a valuable reference for the astronomical community. (Abridged)

astro-ph.GA↗

arXiv2025

DAM-VSR: Disentanglement of Appearance and Motion for Video Super-Resolution

Real-world video super-resolution (VSR) presents significant challenges due to complex and unpredictable degradations. Although some recent methods utilize image diffusion models for VSR and have shown improved detail generation capabilities, they still struggle to produce temporally consistent frames. We attempt to use Stable Video Diffusion (SVD) combined with ControlNet to address this issue. However, due to the intrinsic image-animation characteristics of SVD, it is challenging to generate fine details using only low-quality videos. To tackle this problem, we propose DAM-VSR, an appearance and motion disentanglement framework for VSR. This framework disentangles VSR into appearance enhancement and motion control problems. Specifically, appearance enhancement is achieved through reference image super-resolution, while motion control is achieved through video ControlNet. This disentanglement fully leverages the generative prior of video diffusion models and the detail generation capabilities of image super-resolution models. Furthermore, equipped with the proposed motion-aligned bidirectional sampling strategy, DAM-VSR can conduct VSR on longer input videos. DAM-VSR achieves state-of-the-art performance on real-world data and AIGC data, demonstrating its powerful detail generation capabilities.

cs.CV↗

arXiv2025

Efficiently Serving Large Multimodal Models Using EPD Disaggregation

Large Multimodal Models (LMMs) extend Large Language Models (LLMs) by handling diverse inputs such as images, audio, and video, but at the cost of adding a multimodal encoding stage that increases both computational and memory overhead. This step negatively affects key Service Level Objectives (SLOs), such as time to first token (TTFT) and time per output token (TPOT). We introduce Encode-Prefill-Decode (EPD) Disaggregation, a novel framework that separates the encoding, prefill, and decode stages onto dedicated resources. Unlike current systems, which bundle encoding and prefill together, our approach decouples these steps, unlocking new opportunities and optimizations. These include a mechanism to cache multimedia tokens for efficient transfer, a novel way to parallelize the encoding load within a request, a module for optimal resource allocation for disaggregated serving, and a novel role-switching method to handle changing workload characteristics. Experimental evaluations with popular LMMs show substantial gains in memory efficiency (up to 15x lower peak memory utilization), batch sizes (up to 22x larger), 10x more images per request, and 2.2x larger KV caches. Furthermore, it leads to significant improvements in SLO attainment (up to 90-100% improvement) and TTFT (up to 71% reduction), compared to systems that do not disaggregate. The code is available at https://github.com/vbdi/epdserve.

cs.DC↗

arXiv2025

AnchorCrafter: Animate Cyber-Anchors Selling Your Products via Human-Object Interacting Video Generation

The generation of anchor-style product promotion videos presents promising opportunities in e-commerce, advertising, and consumer engagement. Despite advancements in pose-guided human video generation, creating product promotion videos remains challenging. In addressing this challenge, we identify the integration of human-object interactions (HOI) into pose-guided human video generation as a core issue. To this end, we introduce AnchorCrafter, a novel diffusion-based system designed to generate 2D videos featuring a target human and a customized object, achieving high visual fidelity and controllable interactions. Specifically, we propose two key innovations: the HOI-appearance perception, which enhances object appearance recognition from arbitrary multi-view perspectives and disentangles object and human appearance, and the HOI-motion injection, which enables complex human-object interactions by overcoming challenges in object trajectory conditioning and inter-occlusion management. Extensive experiments show that our system improves object appearance preservation by 7.5\% and doubles the object localization accuracy compared to existing state-of-the-art approaches. It also outperforms existing approaches in maintaining human motion consistency and high-quality video generation. Project page including data, code, and Huggingface demo: https://github.com/cangcz/AnchorCrafter.

cs.CV↗

arXiv2025

FPT Constant Approximation Algorithms for Colorful Sum of Radii

We study the colorful sum of radii problem, where the input is a point set $P$ partitioned into classes $P_1, P_2, \dots, P_ω$, along with per-class outlier bounds $m_1, m_2, \dots, m_ω$, summing to $m$. The goal is to select a subset $\mathcal{C} \subseteq P$ of $k$ centers and assign points to centers in $\mathcal{C}$, allowing up to $m_i$ unassigned points (outliers) from each class $P_i$, while minimizing the sum of cluster radii. The radius of a cluster is defined as the maximum distance from any point in the cluster to its center. The classical (non-colorful) version of the sum of radii problem is known to be NP-hard, even on weighted planar graphs. The colorful sum of radii is introduced by Chekuri et al. (2022), who provide an $O(\log ω)$-approximation algorithm. In this paper, we present the first constant-factor approximation algorithms for the colorful sum of radii running in FPT (fixed-parameter tractable) time. Our contributions are twofold: We design an iterative covering algorithm that achieves a $(2+\varepsilon)$-approximation with running time exponential in both $k$ and $m$; We further develop a $(7+\varepsilon)$-approximation algorithm by leveraging a colorful $k$-center subroutine, improving the running time by removing the exponential dependency on $m$.

cs.CG↗

arXiv2025

Computing the Bogoliubov-de Gennes excitations of two-component Bose-Einstein condensates

In this paper, we present an efficient and spectrally accurate numerical method to compute elementary/collective excitations in two-component Bose-Einstein condensates (BEC), around their mean-field ground state, by solving the associated Bogoliubov-de Gennes (BdG) equation. The BdG equation is essentially an eigenvalue problem for a non-Hermitian differential operator with an eigenfunction normalization constraint. Firstly, we investigate its analytical properties, including the exact eigenpairs, generalized nullspace structure and bi-orthogonality of eigenspaces. Subsequently, by combining the Fourier spectral method for spatial discretization and a stable modified Gram-Schmidt bi-orthogonal algorithm, we propose a structure-preserving iterative method for the resulting large-scale dense non-Hermitian discrete eigenvalue problem. Our method is matrix-free, and the matrix-vector multiplication (or the operator-function evaluation) is implemented with a near-optimal complexity ${\mathcal O}(N_{\rm t}\log(N_{\rm t}))$, where $N_{\rm t}$ is the total number of grid points, thanks to the utilization of the discrete Fast Fourier Transform (FFT). Therefore, it is memory-friendly, spectrally accurate, and highly efficient. Finally, we carry out a comprehensive numerical investigation to showcase its superiority in terms of accuracy and efficiency, alongside some applications to compute the excitation spectrum and Bogoliubov amplitudes in one, two, and three-dimensional problems.

math.NA↗