Searcharxiv⌕ Search

arXiv subjects

Jian Zheng

Publications and source records attributed to Jian Zheng.

At least 37 records · Page 2Linked to original sources

Robust Data-Driven Receding Horizon Control

This paper presents a data-driven receding horizon control framework for discrete-time linear systems that guarantees robust performance in the presence of bounded disturbances. Unlike the majority of existing data-driven predictive control methods, which rely on Willem's fundamental lemma, the proposed method enforces set-membership constraints for data-driven control and utilizes execution data to iteratively refine a set of compatible systems online. Numerical results demonstrate that the proposed receding horizon framework achieves better contractivity for the unknown system compared with regular data-driven control approaches.

math.OC↗

T2V-Turbo-v2: Enhancing Video Generation Model Post-Training through Data, Reward, and Conditional Guidance Design

In this paper, we focus on enhancing a diffusion-based text-to-video (T2V) model during the post-training phase by distilling a highly capable consistency model from a pretrained T2V model. Our proposed method, T2V-Turbo-v2, introduces a significant advancement by integrating various supervision signals, including high-quality training data, reward model feedback, and conditional guidance, into the consistency distillation process. Through comprehensive ablation studies, we highlight the crucial importance of tailoring datasets to specific learning objectives and the effectiveness of learning from diverse reward models for enhancing both the visual quality and text-video alignment. Additionally, we highlight the vast design space of conditional guidance strategies, which centers on designing an effective energy function to augment the teacher ODE solver. We demonstrate the potential of this approach by extracting motion guidance from the training datasets and incorporating it into the ODE solver, showcasing its effectiveness in improving the motion quality of the generated videos with the improved motion-related metrics from VBench and T2V-CompBench. Empirically, our T2V-Turbo-v2 establishes a new state-of-the-art result on VBench, with a Total score of 85.13, surpassing proprietary systems such as Gen-3 and Kling.

cs.CV↗

Spectral Turán problems for nondegenerate hypergraphs

Keevash, Lenz and Mubayi developed a general criterion for hypergraph spectral extremal problems in their seminal work (SIAM J. Discrete Math., 2014). Their framework shows that extremal results on the $α$-spectral radius (for $α> 1$) may be deduced from a corresponding hypergraph Turán problem exhibiting stability properties, provided its extremal construction satisfies certain continuity assumptions. In this paper, we establish a spectral stability result for nondegenerate hypergraphs, extending the Keevash--Lenz--Mubayi criterion. Applying this result, we derive two general spectral Turán theorems for hypergraphs with bipartite or multipartite pattern, thereby transforming spectral Turán problems into the corresponding purely combinatorial problems related to degree-stability in nondegenerate $k$-graph families. As applications, we determine the maximum $α$-spectral radius for several classes of hypergraph and characterize the corresponding extremal hypergraphs, such as the expansion of complete graphs, the generalized fans, the cancellative hypergraphs, the generalized triangles, and a special book hypergraph.

math.CO↗

A signless Laplacian spectral Erdös-Stone-Simonovits theorem

The celebrated Erdős--Stone--Simonovits theorem states that $\mathrm{ex}(n,F)= \big(1-\frac{1}{χ(F)-1}+o(1) \big)\frac{n^{2}}{2}$, where $χ(F)$ is the chromatic number of $F$. In 2009, Nikiforov proved a spectral extension of the Erdős--Stone--Simonovits theorem in terms of the adjacency spectral radius. In this paper, we shall establish a unified extension in terms of the signless Laplacian spectral radius. Let $q(G)$ be the signless Laplacian spectral radius of $G$ and we denote $\mathrm{ex}_{q}(n,F) =\max \{q(G):|G|=n ~\mbox{and}~F\nsubseteq G\}$. It is known that the Erdős--Stone--Simonovits type result for the signless Laplacian spectral radius does not hold for even cycles. We prove that if $F$ is a graph with $χ(F)\geq 3$, then $\mathrm{ex}_{q}(n,F)=\big(1-\frac{1}{χ(F)-1}+o(1) \big)2n$. This solves a problem proposed by Li, Liu and Feng (2022), which gives an entirely satisfactory answer to the problem of estimating $\mathrm{ex}_q(n,F)$. Furthermore, it extends the aforementioned result of Erdős, Stone and Simonovits as well as the spectral result of Nikiforov. Our result indicates that the Erdős--Stone--Simonovits type result regarding the signless Laplacian spectral radius is valid in general.

math.CO↗

Convex Data-Driven Contraction With Riemannian Metrics

The growing complexity of dynamical systems and advances in data collection necessitates robust data-driven control strategies without explicit system identification and robust synthesis. Data-driven stability has been explored in linear and nonlinear systems, often by turning the problem into a linear or positive semidefinite program. This paper focuses on a new emerging property called contractivity, which refers to the exponential convergence of all system trajectories toward each other under a specified metric. Data-driven closed loop contractivity has been studied for the case of the 2-norm and assuming nonlinearities are Lipschitz bounded in subsets of n dimensional euclidean space. We extend the analysis by considering Riemannian metrics for polynomial dynamics. The key to our derivation is to leverage the convex criteria for closed-loop contraction and duality results to efficiently check infinite dimensional membership constraints. Numerical examples demonstrate the effectiveness of the proposed method for both linear and nonlinear systems, highlighting its potential for robust data-driven contraction.

math.OC↗

The signless Laplacian spectral Turán problems for color-critical graphs

The well-known Turán theorem states that if $G$ is an $n$-vertex $K_{r+1}$-free graph, then $e(G)\le e(T_{n,r})$, with equality if and only if $G$ is the $r$-partite Turán graph $T_{n,r}$. A graph $F$ is called color-critical if it contains an edge whose deletion reduces its chromatic number. Extending the Turán theorem, Simonovits (1968) proved that for any color-critical graph $F$ with $χ(F)=r+1$ and sufficiently large $n$, the Turán graph $T_{n,r}$ is the unique graph with maximum number of edges among all $n$-vertex $F$-free graphs. Subsequently, Nikiforov [Electron. J. Combin., 16 (1) (2009)] proved a spectral version of the Simonovits theorem in terms of the adjacency spectral radius. In this paper, we show an extension of the Simonovits theorem for the signless Laplacian spectral radius. We prove that for any color-critical graph $F$ with $χ(F)=r+1\ge 4$ and sufficiently large $n$, if $G$ is an $F$-free graph on $n$ vertices, then $q(G)\le q(T_{n,r})$, with equality if and only if $G=T_{n,r}$. Our approach is to establish a signless Laplacian spectral version of the criterion of Keevash, Lenz and Mubayi [SIAM J. Discrete Math., 28 (4) (2014)]. Consequently, we can determine the signless Laplacian spectral extremal graphs for generalized books and even wheels. As an application, our result gives an upper bound on the degree power of an $F$-free graph. We show that if $n$ is sufficiently large and $G$ is an $F$-free graph on $n$ vertices with $m$ edges, then $\sum_{v\in V(G)} d^2(v) \le 2(1- \frac{1}{r})mn$, with equality if and only if $G$ is a regular Turán graph $T_{n,r}$. This extends a result of Nikiforov and Rousseau [J. Combin. Theory Ser B 92 (2004)].

math.CO↗

A one-dimensional mixing model for the impact of ablative Rayleigh-Taylor instability on compression dynamics

A one-dimensional mixing model, incorporating the effects of laser ablation and initial perturbations, is developed to study the influence of ablative Rayleigh-Taylor instability on compression dynamics. The length of the mixing region is determined with the buoyancy-drag model[arXiv:2411.12392v2 (2024)]. The mixing effect on laser ablation is mainly described with an additional heat source which depends on turbulent kinetic energy and initial perturbation level through a free multiplier. The model is integrated into a one-dimensional radiation hydrodynamics code and validated against two-dimensional planar simulations. The further application of our model to spherical implosion simulations reveals that the model can give reasonable predictions of implosion degradation due to mixing, such as lowered shell compression, reduced stagnation pressure, and decreased areal density, etc. It is found that the time interval between the convergence of the main shock and stagnation may offer an estimate of mixing level in single-shot experiments.

physics.flu-dyn↗

A buoyancy-drag model with a time-varying drag coefficient for evaluating bubble front penetration depth

To evaluate and control bubble front penetration depth ${{h}_{B}}$ induced by ablative Rayleigh-Taylor instability (ARTI) from a weakly nonlinear phase to a self-similar phase, we first propose an improved buoyancy-drag (BD) model with a time-varying drag coefficient. The coefficient incorporates the influence of multiple physical mechanisms, including non-steady ablation, preheating, and other mechanisms during this phase. The model is validated through simulations under various conditions, demonstrating improved accuracy compared to the classical BD model and the self-similar growth. Furthermore, the model suggests controlling ${{h}_{B}}$ by suppressing the "most dangerous mode", which is influenced by initial perturbations and ablative acceleration history, thus offering novel insights for target manufacturing and pulse optimization near the ignition threshold.

physics.flu-dyn↗

Switchable Chern insulator, isospin competitions and charge density waves in rhombohedral graphene moire superlattices

Graphene-based moire superlattices provide a versatile platform for exploring novel correlated and topological electronic states, driven by enhanced Coulomb interactions within flat bands. The intrinsic tunability of graphene s multiple degrees of freedom enables precise control over these complex quantum phases. In this study, we observe a range of competing phases and their transitions in rhombohedrally stacked hexalayer graphene on hexagonal boron nitride (r-6G/hBN) moire superlattices. When electrons are polarized away from the moire superlattice, we firstly identify a Chern insulator with reversible Chern numbers at v = 1 (one electron per moire cell), attributed to the competition between bulk and edge magnetizations.Then, we detect transitions between three distinct insulating states at v = 2, driven by vertical displacement field D and vertical magnetic field B. These insulating phases are distinguished as spin-antiferromagnetic, spin-polarized, and valley-polarized insulators, based on their responses to parallel and perpendicular magnetic fields. When electrons are polarized toward the moire superlattice, in a device with large twist angle, insulating states appear at v = 1/3 and 2/3 at zero magnetic field, and v = 1/2 in a magnetic field. Our findings reveal a rich interplay of charge, isospin, topology and magnetic field in rhombohedral graphene moire superlattices.

cond-mat.str-el↗

Generation of 10 kT Axial Magnetic Fields Using Multiple Conventional Laser Beams: A Sensitivity Study for kJ PW-Class Laser Facilities

Strong multi-kilotesla magnetic fields have various applications in high-energy density science and laboratory astrophysics, but they are not readily available. In our previous work [Y. Shi et al., Phys. Rev. Lett. 130, 155101 (2023)], we developed a novel approach for generating such fields using multiple conventional laser beams with a twist in the pointing direction. This method is particularly well-suited for multi-kilojoule petawatt-class laser systems like SG-II UP, which are designed with multiple linearly polarized beamlets. Utilizing three-dimensional kinetic particle-in-cell simulations, we examine critical factors for a proof-of-principle experiment, such as laser polarization, relative pulse delay, phase offset, pointing stability, and target configuration, and their impact on magnetic field generation. Our general conclusion is that the approach is very robust and can be realized under a wide range of laser parameters and plasma conditions. We also provide an in-depth analysis of the axial magnetic field configuration, azimuthal electron current, and electron and ion orbital angular momentum densities. Supported by a simple model, our analysis shows that the axial magnetic field decays due to the expansion of hot electrons.

physics.plasm-ph↗

Adaptive Critical Subgraph Mining for Cognitive Impairment Conversion Prediction with T1-MRI-based Brain Network

Prediction the conversion to early-stage dementia is critical for mitigating its progression but remains challenging due to subtle cognitive impairments and structural brain changes. Traditional T1-weighted magnetic resonance imaging (T1-MRI) research focus on identifying brain atrophy regions but often fails to address the intricate connectivity between them. This limitation underscores the necessity of focuing on inter-regional connectivity for a comprehensive understand of the brain's complex network. Moreover, there is a pressing demand for methods that adaptively preserve and extract critical information, particularly specialized subgraph mining techniques for brain networks. These are essential for developing high-quality feature representations that reveal critical spatial impacts of structural brain changes and its topology. In this paper, we propose Brain-SubGNN, a novel graph representation network to mine and enhance critical subgraphs based on T1-MRI. This network provides a subgraph-level interpretation, enhancing interpretability and insights for graph analysis. The process begins by extracting node features and a correlation matrix between nodes to construct a task-oriented brain network. Brain-SubGNN then adaptively identifies and enhances critical subgraphs, capturing both loop and neighbor subgraphs. This method reflects the loop topology and local changes, indicative of long-range connections, and maintains local and global brain attributes. Extensive experiments validate the effectiveness and advantages of Brain-SubGNN, demonstrating its potential as a powerful tool for understanding and diagnosing early-stage dementia. Source code is available at https://github.com/Leng-10/Brain-SubGNN.

cs.CV↗

Phase Diagram of growth modes in Graphene Growth on Cooper by Vapor Deposition

Understanding the atomistic mechanism in graphene growth is crucial for controlling the number of layers or domain sizes to meet practical needs. In this work, focusing on the growth of graphene by chemical vapor deposition on copper substrates, the surface kinetics in the growth are systematically investigated by first-principles calculations. The phase diagram, predicting whether the growth mode is monolayer graphene or bilayer graphene under various experimental conditions, is constructed based on classical nucleation theory. Our phase diagram well illustrates the effect of high hydrogen pressure on bilayer graphene growth and clarifies the mechanism of the most widely used experimental growth approaches. The phase diagram can provide guidance and predictions for experiments and inspires the study of other two-dimensional materials with graphene-like growth mechanisms.

cond-mat.mtrl-sci↗

A novel nuclear recoil calibration for liquid noble gas detectors

According to many dark matter models, a potential signal registered in a detector would feature a single-scattering nuclear recoil (NR). So, it is crucial to calibrate the detector's response to NR events. The conventional calibrations implement $\sim$ keV to MeV neutrons, which can be produced by an accelerator, a neutron generator, or a radioactive source. Although the calibrating methods have been widely employed, they could be improved in several ways: (a) the incident neutron energy should be more monoenergetic, (b) the calibrating NR energy should line up with the region of interest (ROI) of the experiment, and (c) the intensity of the beam should be appropriate. In the paper, we introduce a novel NR calibration method for liquid helium detectors, in which a helium beam ($α$ particles) will be implemented to calibrate the detectors. The helium beam can (i) be tuned precisely to have a jitter of $\lesssim $ 4\% (the $α$ beam's kinetic energy is equivalent to the recoil energy in the conventional calibrations with fast neutrons); (ii) have an energy between $\sim$ 100 eV and tens of keV; and (iii) provide a tunable flux from nA to 100 $μ$A, which presents convenience in beam pipe configuration to obtain a $\sim$ 100 Hz events rate so that the events pileup would be ignorable.

physics.ins-det↗

FlexEControl: Flexible and Efficient Multimodal Control for Text-to-Image Generation

Controllable text-to-image (T2I) diffusion models generate images conditioned on both text prompts and semantic inputs of other modalities like edge maps. Nevertheless, current controllable T2I methods commonly face challenges related to efficiency and faithfulness, especially when conditioning on multiple inputs from either the same or diverse modalities. In this paper, we propose a novel Flexible and Efficient method, FlexEControl, for controllable T2I generation. At the core of FlexEControl is a unique weight decomposition strategy, which allows for streamlined integration of various input types. This approach not only enhances the faithfulness of the generated image to the control, but also significantly reduces the computational overhead typically associated with multimodal conditioning. Our approach achieves a reduction of 41% in trainable parameters and 30% in memory usage compared with Uni-ControlNet. Moreover, it doubles data efficiency and can flexibly generate images under the guidance of multiple input conditions of various modalities.

cs.CV↗

Self-generated magnetic field in three-dimensional ablative Rayleigh-Taylor instability

The self-generated magnetic field in three-dimensional (3D) single-mode ablative Rayleigh-Taylor instabilities (ARTI) relevant to the acceleration phase of a direct-drive inertial confinement fusion (ICF) implosion is investigated. It is found that stronger magnetic fields up to a few thousands of T can be generated by 3D ARTI than by its two-dimensional (2D) counterpart. The Nernst effects significantly alter the magnetic fields convection and amplify the magnetic fields. The scaling law for the magnetic flux obtained in the 2D simulations performs reasonably well in the 3D cases. While the magnetic field significantly accelerates the bubble growth in the short-wavelength 2D modes through modifying the heat fluxes, the magnetic field mostly accelerates the spike growth but has little influence on the bubble growth in 3D ARTI.

physics.plasm-ph↗

Bootstrapping Chest CT Image Understanding by Distilling Knowledge from X-ray Expert Models

Radiologists highly desire fully automated versatile AI for medical imaging interpretation. However, the lack of extensively annotated large-scale multi-disease datasets has hindered the achievement of this goal. In this paper, we explore the feasibility of leveraging language as a naturally high-quality supervision for chest CT imaging. In light of the limited availability of image-report pairs, we bootstrap the understanding of 3D chest CT images by distilling chest-related diagnostic knowledge from an extensively pre-trained 2D X-ray expert model. Specifically, we propose a language-guided retrieval method to match each 3D CT image with its semantically closest 2D X-ray image, and perform pair-wise and semantic relation knowledge distillation. Subsequently, we use contrastive learning to align images and reports within the same patient while distinguishing them from the other patients. However, the challenge arises when patients have similar semantic diagnoses, such as healthy patients, potentially confusing if treated as negatives. We introduce a robust contrastive learning that identifies and corrects these false negatives. We train our model with over 12,000 pairs of chest CT images and radiology reports. Extensive experiments across multiple scenarios, including zero-shot learning, report generation, and fine-tuning processes, demonstrate the model's feasibility in interpreting chest CT images.

cs.CV↗

Fragmented Moments, Balanced Choices: How Do People Make Use of Their Waiting Time?

Everyone spends some time waiting every day. HCI research has developed tools for boosting productivity while waiting. However, little is known about how people naturally spend their waiting time. We conducted an experience sampling study with 21 working adults who used a mobile app to report their daily waiting time activities over two weeks. The aim of this study is to understand the activities people do while waiting and the effect of situational factors. We found that participants spent about 60% of their waiting time on leisure activities, 20% on productive activities, and 20% on maintenance activities. These choices are sensitive to situational factors, including accessible device, location, and certain routines of the day. Our study complements previous ones by demonstrating that people purpose waiting time for various goals beyond productivity and to maintain work-life balance. Our findings shed light on future empirical research and system design for time management.

cs.HC↗

Text-to-image Editing by Image Information Removal

Diffusion models have demonstrated impressive performance in text-guided image generation. Current methods that leverage the knowledge of these models for image editing either fine-tune them using the input image (e.g., Imagic) or incorporate structure information as additional constraints (e.g., ControlNet). However, fine-tuning large-scale diffusion models on a single image can lead to severe overfitting issues and lengthy inference time. Information leakage from pretrained models also make it challenging to preserve image content not related to the text input. Additionally, methods that incorporate structural guidance (e.g., edge maps, semantic maps, keypoints) find retaining attributes like colors and textures difficult. Using the input image as a control could mitigate these issues, but since these models are trained via reconstruction, a model can simply hide information about the original image when encoding it to perfectly reconstruct the image without learning the editing task. To address these challenges, we propose a text-to-image editing model with an Image Information Removal module (IIR) that selectively erases color-related and texture-related information from the original image, allowing us to better preserve the text-irrelevant content and avoid issues arising from information hiding. Our experiments on CUB, Outdoor Scenes, and COCO reports our approach achieves the best editability-fidelity trade-off results. In addition, a user study on COCO shows that our edited images are preferred 35% more often than prior work.

cs.CV↗