arXiv subjects

Tingting Tang

Publications and source records attributed to Tingting Tang.

At least 19 recordsLinked to original sources

arXiv2026

Spin-multiplexed terahertz image edge detection based on extrinsic chirality in all-dielectric metasurface

Extrinsically chiral metasurfaces can exhibit pronounced chiral responses in planar structures without intrinsic chirality, making them attractive for spin-selective light manipulation. Here, we propose a transmissive all-dielectric terahertz metasurface for image edge detection. Owing to the extrinsic chiral response, the metasurface enables edge-enhanced imaging under right-handed circularly polarized incidence near 0.69 THz, while preserving bright-field imaging under left-handed circularly polarized incidence. Spatial-frequency analysis shows that the effective numerical aperture of the device is approximately 0.4, indicating the terahertz image edge detection can be achieved over a broad spatial-frequency range. This work provides a new design route toward transmissive terahertz all-optical image processing with spin selectivity and a large numerical aperture.

physics.optics↗

arXiv2026

Differentially Private Retrieval-Augmented Generation

Retrieval-augmented generation (RAG) is a widely used framework for reducing hallucinations in large language models (LLMs) on domain-specific tasks by retrieving relevant documents from a database to support accurate responses. However, when the database contains sensitive corpora, such as medical records or legal documents, RAG poses serious privacy risks by potentially exposing private information through its outputs. Prior work has demonstrated that one can practically craft adversarial prompts that force an LLM to regurgitate the augmented contexts. A promising direction is to integrate differential privacy (DP), a privacy notion that offers strong formal guarantees, into RAG systems. However, naively applying DP mechanisms into existing systems often leads to significant utility degradation. Particularly for RAG systems, DP can reduce the usefulness of the augmented contexts leading to increase risk of hallucination from the LLMs. Motivated by these challenges, we present DP-KSA, a novel privacy-preserving RAG algorithm that integrates DP using the propose-test-release paradigm. DP-KSA follows from a key observation that most question-answering (QA) queries can be sufficiently answered with a few keywords. Hence, DP-KSA first obtains an ensemble of relevant contexts, each of which will be used to generate a response from an LLM. We utilize these responses to obtain the most frequent keywords in a differentially private manner. Lastly, the keywords are augmented into the prompt for the final output. This approach effectively compresses the semantic space while preserving both utility and privacy. We formally show that DP-KSA provides formal DP guarantees on the generated output with respect to the RAG database. We evaluate DP-KSA on two QA benchmarks using three instruction-tuned LLMs, and our empirical results demonstrate that DP-KSA achieves a strong privacy-utility tradeoff.

cs.CR↗

arXiv2026

LRD-MPC: Efficient MPC Inference through Low-rank Decomposition

Secure Multi-party Computation (MPC) enables untrusted parties to jointly compute a function without revealing their inputs. Its application to machine learning (ML) has gained significant attention, particularly for secure inference services deployed across multiple cloud virtual machines (VMs), where each VM acts as an MPC party. Model providers secret-share model weights, and users secret-share inputs, ensuring that each server operates only on random shares. While MPC provides strong cryptographic guarantees, it incurs substantial computational and communication overhead. Deep neural networks rely heavily on convolutional and fully connected layers, which require costly matrix multiplications in MPC. To reduce this cost, we propose leveraging low-rank decomposition (LRD) for linear layers, replacing one large matrix multiplication with two smaller ones. Each matrix multiplication in MPC incurs a round of communication, meaning decomposing one matrix multiplication into two leads to an additional communication round. Second, the added matrix multiplication requires an additional truncation step to maintain numerical precision. Since truncation itself requires communication and computation, these overheads can offset the gains from decomposition. To address this, we introduce two complementary optimizations: truncation skipping and efficient linear layer concatenation. Truncation skipping removes the extra truncation induced by LRD, while linear layer concatenation pipelines operations to hide the additional communication round. Together, these techniques mitigate the main overheads of LRD in MPC and improve overall efficiency. Our approach is broadly applicable across MPC protocols. Experiments show up to 25% speedup in n-PC and 33% in 3-PC protocols over full-rank baselines, along with up to 52% GPU energy savings and 88% reduction in offline-phase latency.

cs.CR↗

arXiv2026

Optimal Harvesting in Stream Networks: Maximizing Biomass and Yield

In this study, we develop a metapopulation model framework to identify optimal harvesting strategies for a population in a stream network. We consider two distinct optimization objectives: maximization of total biomass and maximization of total yield, under the constraint of a fixed total harvesting effort. We examine in detail the special case of a two-patch network and fully characterize the optimal strategies for each objective. We show that when the population growth rate exceeds a critical threshold, a single harvesting strategy can simultaneously maximize both objectives. For general $n$-patch networks with homogeneous growth rates across patches, we focus on the regime of large growth rates and demonstrate that the optimal harvesting strategy selects patches according to their intraspecific competition rates and an effective net flow metric determined by network connectivity parameters.

math.DS↗

arXiv2025

Learning Polynomial Activation Functions for Deep Neural Networks

Activation functions are crucial for deep neural networks. This novel work frames the problem of training neural network with learnable polynomial activation functions as a polynomial optimization problem, which is solvable by the Moment-SOS hierarchy. This work represents a fundamental departure from the conventional paradigm of training deep neural networks, which relies on local optimization methods like backpropagation and gradient descent. Numerical experiments are presented to demonstrate the accuracy and robustness of optimum parameter recovery in presence of noises.

math.OC↗

arXiv2025

Impact of network connectivity on the dynamics of populations in stream environments

We consider the impact of network connectivity on the dynamics of a population in a stream environment. The population is modeled using a graph theoretical framework, with habitats represented by isolated patches. We introduce a change in connectivity into the model through the addition of a bi-directional or one-directional edge between two patches and examine the impact of this edge modification on the metapopulation growth rate and the network biomass. Our main results indicate that adding a bi-directional edge often decreases both measures, while the effect of adding an one-directional edge is more intricate and dependent on the model parameters. We establish complete analytical results for stream networks of three patches, and provide some generalizations and conjectures for more general stream networks of $n$ patches. These conjectures are supported with numerical simulations.

q-bio.PE↗

arXiv2024

Fastrack: Fast IO for Secure ML using GPU TEEs

As cloud-based ML expands, ensuring data security during training and inference is critical. GPU-based Trusted Execution Environments (TEEs) offer secure, high-performance solutions, with CPU TEEs managing data movement and GPU TEEs handling authentication and computation. However, CPU-to-GPU communication overheads significantly hinder performance, as data must be encrypted, authenticated, decrypted, and verified, increasing costs by 12.69 to 33.53 times. This results in GPU TEE inference becoming 54.12% to 903.9% slower and training 10% to 455% slower than non-TEE systems, undermining GPU TEE advantages in latency-sensitive applications. This paper analyzes Nvidia H100 TEE protocols and identifies three key overheads: 1) redundant CPU re-encryption, 2) limited authentication parallelism, and 3) unnecessary operation serialization. We propose Fastrack, optimizing with 1) direct GPU TEE communication, 2) parallelized authentication, and 3) overlapping decryption with PCI-e transmission. These optimizations cut communication costs and reduce inference/training runtime by up to 84.6%, with minimal overhead compared to non-TEE systems.

cs.CR↗

arXiv2024

Ethos: Rectifying Language Models in Orthogonal Parameter Space

Language models (LMs) have greatly propelled the research on natural language processing. However, LMs also raise concerns regarding the generation of biased or toxic content and the potential disclosure of private information from the training dataset. In this work, we present a new efficient approach, Ethos, that rectifies LMs to mitigate toxicity and bias in outputs and avoid privacy leakage. Ethos is built on task arithmetic. However, unlike current task arithmetic algorithms, Ethos distinguishes general beneficial and undesired knowledge when reconstructing task vectors. Specifically, Ethos first obtains a set of principal components from the pre-trained models using singular value decomposition. Then, by projecting the task vector onto principal components, Ethos identifies the principal components that encode general or undesired knowledge. Ethos performs negating using the task vector with undesired knowledge only, thereby minimizing collateral damage on general model utility. We demonstrate the efficacy of our approach on three different tasks: debiasing, detoxification, and memorization unlearning. Evaluations show Ethos is more effective in removing undesired knowledge and maintaining the overall model performance compared to current task arithmetic methods.

cs.CL↗

arXiv2024

Edge Private Graph Neural Networks with Singular Value Perturbation

Graph neural networks (GNNs) play a key role in learning representations from graph-structured data and are demonstrated to be useful in many applications. However, the GNN training pipeline has been shown to be vulnerable to node feature leakage and edge extraction attacks. This paper investigates a scenario where an attacker aims to recover private edge information from a trained GNN model. Previous studies have employed differential privacy (DP) to add noise directly to the adjacency matrix or a compact graph representation. The added perturbations cause the graph structure to be substantially morphed, reducing the model utility. We propose a new privacy-preserving GNN training algorithm, Eclipse, that maintains good model utility while providing strong privacy protection on edges. Eclipse is based on two key observations. First, adjacency matrices in graph structures exhibit low-rank behavior. Thus, Eclipse trains GNNs with a low-rank format of the graph via singular values decomposition (SVD), rather than the original graph. Using the low-rank format, Eclipse preserves the primary graph topology and removes the remaining residual edges. Eclipse adds noise to the low-rank singular values instead of the entire graph, thereby preserving the graph privacy while still maintaining enough of the graph structure to maintain model utility. We theoretically show Eclipse provide formal DP guarantee on edges. Experiments on benchmark graph datasets show that Eclipse achieves significantly better privacy-utility tradeoff compared to existing privacy-preserving GNN training methods. In particular, under strong privacy constraints ($ε$ < 4), Eclipse shows significant gains in the model utility by up to 46%. We further demonstrate that Eclipse also has better resilience against common edge attacks (e.g., LPA), lowering the attack AUC by up to 5% compared to other state-of-the-art baselines.

cs.LG↗

arXiv2024

Comparative Analysis of Practical Identifiability Methods for an SEIR Model

Identifiability of a mathematical model plays a crucial role in parameterization of the model. In this study, we establish the structural identifiability of a Susceptible-Exposed-Infected-Recovered (SEIR) model given different combinations of input data and investigate practical identifiability with respect to different observable data, data frequency, and noise distributions. The practical identifiability is explored by both Monte Carlo simulations and a Correlation Matrix approach. Our results show that practical identifiability benefits from higher data frequency and data from the peak of an outbreak. The incidence data gives the best practical identifiability results compared to prevalence and cumulative data. In addition, we compare and distinguish the practical identifiability by Monte Carlo simulations and a Correlation Matrix approach, providing insights for when to use which method for other applications.

stat.ME↗

arXiv2024

Polarization splitter rotator on thin film lithium niobate based on multimode interference

Polarization splitter-rotators (PSRs) are the key elements to realize on-chip polarization manipulation. Current PSRs on thin film lithium niobate (TFLN) rely on sub-micron gaps to realize modes separation, which increase the difficulties of lithography and etching. In this paper, a polarization splitter-rotator on TFLN based on multimode interference (MMI) is demonstrated. Mode division is achieved by an MMI-based mode demultiplexer. The feature size of the PSR is 1.5 μm, which can be fabricated with low priced i-line contact aligners. Experimental results show a polarization extinction ratio (PER) > 20 dB and insertion loss (IL) <1.5 dB are achieved in a wavelength range of 1542-1600 nm for TE-polarized light. And a PER > 9.5 dB and an IL <3.0 dB are achieved in a wavelength range of 1561-1600 nm for TM-polarized light. This PSR could find application in the low-cost fabrication of dual-polarization TFLN integrated photonic devices.

physics.optics↗

arXiv2024

Thermal Stress Analysis of the LNG Corrugated Cryogenic Hose During Gas Pre-Cooling Process

In this study, thermal-fluid-solid coupled simulations on the gas-phase pre-cooling operation of the corrugated cryogenic hoses were performed. Attention was focused on the temporal evolution and spatial distribution of transient thermal stress in the hose structure caused by convective heat transfer of the cooling medium, Liquefied Natural Gas Boil-Off Gas (BOG). The effects of different corrugated hose parameters, i.e., boundary conditions, hose lengths, BOG inlet flow rates, and corrugation shapes (C-type and U-type), on the transient thermal stress behavior were thoroughly assessed. The thermal stress developed at different locations of the corrugated hoses with these parameters is found to be governed by two major factors: the boundary constraint and local temperature gradient. The objective of this study is to offer practical insights for the structural strength design of corrugated cryogenic hoses and effective pre-cooling strategies, aiming to mitigate structural safety risks caused by excessive thermal stress.

physics.flu-dyn↗

arXiv2023

A numerical method for solving elliptic equations on real closed algebraic curves and surfaces

There are many numerical methods for solving partial different equations (PDEs) on manifolds such as classical implicit, finite difference, finite element, and isogeometric analysis methods which aim at improving the interoperability between finite element method and computer aided design (CAD) software. However, these approaches have difficulty when the domain has singularities since the solution at the singularity may be multivalued. This paper develops a novel numerical approach to solve elliptic PDEs on real, closed, connected, orientable, and almost smooth algebraic curves and surfaces. Our method integrates numerical algebraic geometry, differential geometry, and a finite difference scheme which is demonstrated on several examples.

math.NA↗

arXiv2023

Impact of resource distributions on the competition of species in stream environment

Our earlier work in \cite{nguyen2022population} shows that concentrating the resources on the upstream end tends to maximize the total biomass in a metapopulation model for a stream species. In this paper, we continue our research direction by further considering a Lotka-Voletrra competition patch model for two stream species. We show that the species whose resource allocations maximize the total biomass has competitive advantage.

q-bio.PE↗

arXiv2023

Maximizing Metapopulation Growth Rate and Biomass in Stream Networks

We consider the logistic metapopulation model over a stream network and use the metapopulation growth rate and the total biomass (of the positive equilibrium) as metrics for different aspects of population persistence. Our objective is to find distributions of resources that maximize these persistence measures. We begin our study by considering stream networks consisting of three nodes and prove that the strategy to maximize the total biomass is to concentrate all the resources in the most upstream locations. In contrast, when the diffusion rates are sufficiently small, the metapopulation growth rate is maximized when all resources are concentrated in one of the most downstream locations. These two main results are generalized to stream networks with any number of patches.

math.DS↗

arXiv2022

Numerical Study of Viscoelastic Upstream Instability

In this work, we report numerical results on the flow instability and bifurcation of a viscoelastic fluid in the upstream region of a confined cylinder in a narrow channel. Two-dimensional direct numerical simulations based on the FENE-P model (the finite-extensible nonlinear elastic model with the Peterlin closure) are conducted with numerical stabilization techniques. Our results show that the macroscopic viscoelastic constitutive relation can capture the viscoelastic upstream instability reported in previous experiments for low-Reynolds-number flows. The numerical simulations reveal that the non-dimensional recirculation length ($L_D$) is affected by the cylinder blocking rate ($BR$), the Weissenberg number ($Wi$), the viscosity ratio ($β$), and the maximum polymer extension ($L$). Close to the onset of upstream recirculation, depending on the values of $L$ and $β$, the bifurcation can be supercritical or subcritical. $β$ has a nonlinear influence on the upstream recirculation length. This work contributes to our theoretical understanding of this new instability mechanism in viscoelastic wake flows.

physics.flu-dyn↗

arXiv2022

Computing real powers of monomial ideals

This paper concerns the exponentiation of monomial ideals. While it is customary for the exponentiation operation on ideals to consider natural powers, we extend this notion to powers where the exponent is a positive real number. Real powers of a monomial ideal generalize the integral closure operation and highlight many interesting connections to the theory of convex polytopes. We provide multiple algorithms for computing the real powers of a monomial ideal. An important result is that given any monomial ideal $I$, the function taking real numbers to the corresponding real power of $I$ is a step function whose jumping points are rational. This reduces the problem of determining real powers to rational exponents.

math.AC↗

arXiv2022

A machine learning based method to generate random packed isotropic porous media with desired porosity and permeability

Porous materials are used in many fields, including energy industry, agriculture, medical industry, etc. The generation of digital porous media facilitates the fabrication of real porous media and the analysis of their properties. The past random digital porous media generation methods are unable to generate a porous medium with a specific permeability. A new method is proposed in the present study, which can generate the random packed isotropic porous media with specific porosity and permeability. Firstly, the process of generating the random packed isotropic porous media is detailed. Secondly, the permeability of the generated porous media is calculated with the multi-relaxation time (MRT) lattice Boltzmann method (LBM), which is prepared for the training of convolutional neural network (CNN). Thirdly, 3000 samples on the microstructure of porous media and their permeabilities are used to train the CNN model. The trained model is very effective in predicting the permeability of a porous medium. Finally, our method is elaborated and the choice of target permeability in this method is discussed. With the support of a powerful computer, a porous medium that satisfies the error condition of porosity and permeability can be generated in a short time.

physics.flu-dyn↗