arXiv subjects

Yingqi Liu

Publications and source records attributed to Yingqi Liu.

At least 19 recordsLinked to original sources

arXiv2026

On isotrivial cone structures of highest weight type

We study the projective-geometric criterion of Hwang-Li for isotrivial cone structures admitting characteristic conic connections. We give a complete classification of flag varieties satisfying this criterion, and in the process characterize flag varieties with an injective Gaussian map. As an application, we classify germs of minimal rational curves whose VMRT at a general point is isomorphic to that of a non-homogeneous smooth projective symmetric variety of Picard number one, apart from two exceptional cases.

math.AG↗

arXiv2025

Unveiling the Power of Multiple Gossip Steps: A Stability-Based Generalization Analysis in Decentralized Training

Decentralized training removes the centralized server, making it a communication-efficient approach that can significantly improve training efficiency, but it often suffers from degraded performance compared to centralized training. Multi-Gossip Steps (MGS) serve as a simple yet effective bridge between decentralized and centralized training, significantly reducing experiment performance gaps. However, the theoretical reasons for its effectiveness and whether this gap can be fully eliminated by MGS remain open questions. In this paper, we derive upper bounds on the generalization error and excess error of MGS using stability analysis, systematically answering these two key questions. 1). Optimization Error Reduction: MGS reduces the optimization error bound at an exponential rate, thereby exponentially tightening the generalization error bound and enabling convergence to better solutions. 2). Gap to Centralization: Even as MGS approaches infinity, a non-negligible gap in generalization error remains compared to centralized mini-batch SGD ($\mathcal{O}(T^{\frac{c\beta}{c\beta +1}}/{n m})$ in centralized and $\mathcal{O}(T^{\frac{2c\beta}{2c\beta +2}}/{n m^{\frac{1}{2c\beta +2}}})$ in decentralized). Furthermore, we provide the first unified analysis of how factors like learning rate, data heterogeneity, node count, per-node sample size, and communication topology impact the generalization of MGS under non-convex settings without the bounded gradients assumption, filling a critical theoretical gap in decentralized training. Finally, promising experiments on CIFAR datasets support our theoretical findings.

cs.LG↗

arXiv2025

Integrated Take-off Management and Trajectory Optimization for Merging Control in Urban Air Mobility Corridors

Urban Air Mobility (UAM) has the potential to revolutionize daily transportation, offering rapid and efficient aerial mobility services. Take-off and merging phases are critical for air corridor operations, requiring the coordination of take-off aircraft and corridor traffic while ensuring safety and seamless transition. This paper proposes an integrated take-off management and trajectory optimization for merging control in UAM corridors. We first introduce a novel take-off airspace design. To our knowledge, this paper is one of the first to propose a structured design for take-off airspace. Based on the take-off airspace design, we devise a hierarchical coordinated take-off and merging management (HCTMM) strategy. To be specific, the take-off airspace design can simplify aircraft dynamics and thus reduce the dimensionality of the trajectory optimization problem whilst mitigating obstacle avoidance complexities. The HCTMM strategy strictly ensures safety and improves the efficiency of take-off and merging operations. At the tactical level, a scheduling algorithm coordinates aircraft take-off times and selects dynamic merging points to reduce conflicts and ensure smooth take-off and merging processes. At the operational level, a trajectory optimization strategy ensures that each aircraft reaches the dynamic merging point efficiently while satisfying safety constraints. Simulation results show that, compared to representative strategies with fixed or dynamic merging points, the HCTMM strategy significantly improves operational efficiency and reduces computational burden, while ensuring safety under various corridor traffic conditions. Further results confirm the scalability of the HCTMM strategy and the computational efficiency enabled by the proposed take-off airspace design.

math.OC↗

arXiv2025

On linear sections of the spinor tenfold II

Following previous work by A. Kuznetsov, we study the Fano manifolds obtained as linear sections of the spinor tenfold in $\mathbb{P}^{15}$. Up to codimension three there are finitely many such sections, up to projective equivalence. In codimension four there are three moduli, and this family is particularly interesting because of its relationship with Kummer surfaces on the one hand, and a grading of the exceptional Lie algebra $\mathfrak{e}_8$ on the other hand. We show how the two approaches are intertwined, and we prove that codimension four sections of the spinor tenfolds and Kummer surfaces have the very same GIT moduli space. The Lie theoretic viewpoint provides a wealth of additional information. In particular we locate and study the unique section admitting an action of $SL_2\times SL_2$; similarly to the Mukai-Umemura variety in the family of prime Fano threefold of genus 12, it is a compactification of a quotient by a finite group.

math.AG↗

arXiv2025

Atom-Molecule Superradiance and Entanglement with Cavity-Mediated Three-Body Interactions

Ultracold atoms coupled to optical cavities offer a powerful platform for studying strongly correlated many-body physics. Here, we propose an experimental scheme for creating biatomic molecules via cavity-enhanced photoassociation from an atomic condensate. This setup realizes long-range three-body interactions mediated by tripartite cavity-atom-molecule coupling. Beyond a critical pump strength, a self-organized square lattice phase for molecular condensate emerges, resulting in hybrid atom-molecule superradiance with spontaneous $U(1)$ symmetry breaking. Distinct from previously observed ultracold bosonic (fermionic) atomic superradiance, our findings demonstrate bosonic enhancement characterized by a cubic scaling of steady-state photon number with total atom number. Additionally, strong photon-matter entanglement is shown to effectively characterize superradiant quantum phase transition. Our findings deepen the understanding of quantum superchemistry and exotic many-body nonequilibrium dynamics in cavity-coupled quantum gases.

cond-mat.quant-gas↗

arXiv2024

TIPS: Threat Actor Informed Prioritization of Applications using SecEncoder

This paper introduces TIPS: Threat Actor Informed Prioritization using SecEncoder, a specialized language model for security. TIPS combines the strengths of both encoder and decoder language models to detect and prioritize compromised applications. By integrating threat actor intelligence, TIPS enhances the accuracy and relevance of its detections. Extensive experiments with a real-world benchmark dataset of applications demonstrate TIPS's high efficacy, achieving an F-1 score of 0.90 in identifying malicious applications. Additionally, in real-world scenarios, TIPS significantly reduces the backlog of investigations for security analysts by 87%, thereby streamlining the threat response process and improving overall security posture.

cs.CR↗

arXiv2024

SecEncoder: Logs are All You Need in Security

Large and Small Language Models (LMs) are typically pretrained using extensive volumes of text, which are sourced from publicly accessible platforms such as Wikipedia, Book Corpus, or through web scraping. These models, due to their exposure to a wide range of language data, exhibit impressive generalization capabilities and can perform a multitude of tasks simultaneously. However, they often fall short when it comes to domain-specific tasks due to their broad training data. This paper introduces SecEncoder, a specialized small language model that is pretrained using security logs. SecEncoder is designed to address the domain-specific limitations of general LMs by focusing on the unique language and patterns found in security logs. Experimental results indicate that SecEncoder outperforms other LMs, such as BERTlarge, DeBERTa-v3-large and OpenAI's Embedding (textembedding-ada-002) models, which are pretrained mainly on natural language, across various tasks. Furthermore, although SecEncoder is primarily pretrained on log data, it outperforms models pretrained on natural language for a range of tasks beyond log analysis, such as incident prioritization and threat intelligence document retrieval. This suggests that domain specific pretraining with logs can significantly enhance the performance of LMs in security. These findings pave the way for future research into security-specific LMs and their potential applications.

cs.CR↗

arXiv2024

Boosting the Performance of Decentralized Federated Learning via Catalyst Acceleration

Decentralized Federated Learning has emerged as an alternative to centralized architectures due to its faster training, privacy preservation, and reduced communication overhead. In decentralized communication, the server aggregation phase in Centralized Federated Learning shifts to the client side, which means that clients connect with each other in a peer-to-peer manner. However, compared to the centralized mode, data heterogeneity in Decentralized Federated Learning will cause larger variances between aggregated models, which leads to slow convergence in training and poor generalization performance in tests. To address these issues, we introduce Catalyst Acceleration and propose an acceleration Decentralized Federated Learning algorithm called DFedCata. It consists of two main components: the Moreau envelope function, which primarily addresses parameter inconsistencies among clients caused by data heterogeneity, and Nesterov's extrapolation step, which accelerates the aggregation phase. Theoretically, we prove the optimization error bound and generalization error bound of the algorithm, providing a further understanding of the nature of the algorithm and the theoretical perspectives on the hyperparameter choice. Empirically, we demonstrate the advantages of the proposed algorithm in both convergence speed and generalization performance on CIFAR10/100 with various non-iid data distributions. Furthermore, we also experimentally verify the theoretical properties of DFedCata.

cs.LG↗

arXiv2024

The synthetic gauge field and exotic vortex phase with spin-orbital-angular-momentum coupling

Ultracold atoms endowed with tunable spin-orbital-angular-momentum coupling (SOAMC) represent a promising avenue for delving into exotic quantum phenomena. Building on recent experimental advancements, we propose the generation of synthetic gauge fields ,and by including exotic vortex phases within spinor Bose-Einstein condensates, employing a combination of a running wave and Laguerre-Gaussian laser fields. We investigate the ground-state characteristics of the SOAMC condensate, revealing the emergence of exotic vortex states with controllable orbital angular momenta. It is shown that the interplay of the SOAMC and conventional spin-linear-momentum coupling induced by the running wave beam leads to the formation of a vortex state exhibiting a phase stripe hosting single multiply quantized singularity. The phase of the ground state will undergo the phase transition corresponding to the breaking of rotational symmetry while preserving the mirror symmetry. Importantly, the observed density distribution of the ground-state wavefunction, exhibiting broken rotational symmetry, can be well characterized by the synthetic magnetic field generated through light interaction with the dressed spin state. Our findings pave the way for further exploration into the rotational properties of stable exotic vortices with higher orbital angular momenta against splitting in the presence of synthetic gauge fields in ultracold quantum gases.

cond-mat.quant-gas↗

arXiv2024

Decentralized Directed Collaboration for Personalized Federated Learning

Personalized Federated Learning (PFL) is proposed to find the greatest personalized models for each client. To avoid the central failure and communication bottleneck in the server-based FL, we concentrate on the Decentralized Personalized Federated Learning (DPFL) that performs distributed model training in a Peer-to-Peer (P2P) manner. Most personalized works in DPFL are based on undirected and symmetric topologies, however, the data, computation and communication resources heterogeneity result in large variances in the personalized models, which lead the undirected aggregation to suboptimal personalized performance and unguaranteed convergence. To address these issues, we propose a directed collaboration DPFL framework by incorporating stochastic gradient push and partial model personalized, called \textbf{D}ecentralized \textbf{Fed}erated \textbf{P}artial \textbf{G}radient \textbf{P}ush (\textbf{DFedPGP}). It personalizes the linear classifier in the modern deep model to customize the local solution and learns a consensus representation in a fully decentralized manner. Clients only share gradients with a subset of neighbors based on the directed and asymmetric topologies, which guarantees flexible choices for resource efficiency and better convergence. Theoretically, we show that the proposed DFedPGP achieves a superior convergence rate of $\mathcal{O}(\frac{1}{\sqrt{T}})$ in the general non-convex setting, and prove the tighter connectivity among clients will speed up the convergence. The proposed method achieves state-of-the-art (SOTA) accuracy in both data and computation heterogeneity scenarios, demonstrating the efficiency of the directed collaboration and partial gradient push.

cs.LG↗

arXiv2024

LOTUS: Evasive and Resilient Backdoor Attacks through Sub-Partitioning

Backdoor attack poses a significant security threat to Deep Learning applications. Existing attacks are often not evasive to established backdoor detection techniques. This susceptibility primarily stems from the fact that these attacks typically leverage a universal trigger pattern or transformation function, such that the trigger can cause misclassification for any input. In response to this, recent papers have introduced attacks using sample-specific invisible triggers crafted through special transformation functions. While these approaches manage to evade detection to some extent, they reveal vulnerability to existing backdoor mitigation techniques. To address and enhance both evasiveness and resilience, we introduce a novel backdoor attack LOTUS. Specifically, it leverages a secret function to separate samples in the victim class into a set of partitions and applies unique triggers to different partitions. Furthermore, LOTUS incorporates an effective trigger focusing mechanism, ensuring only the trigger corresponding to the partition can induce the backdoor behavior. Extensive experimental results show that LOTUS can achieve high attack success rate across 4 datasets and 7 model structures, and effectively evading 13 backdoor detection and mitigation techniques. The code is available at https://github.com/Megum1/LOTUS.

cs.CV↗

arXiv2024

Moduli of codimension two linear sections of subadjoint varieties

Let $G$ be a simple algebraic group of type $F_{4}$, $E_{6}$, $E_{7}$ or $E_{8}$, and let $\mathfrak{g}$ be its Lie algebra. The adjoint variety $X_{ad} \subseteq \mathbb{P} \mathfrak{g}$ is defined as the unique closed orbit of the adjoint action of $G$ on $\mathbb{P}\mathfrak{g}$. $X_{ad}$ is a Fano contact manifold covered by lines in $\mathbb{P} \mathfrak{g}$. The subadjoint variety $S \subseteq \mathbb{P} W$ is denoted by the variety of lines on $X_{ad}$ through a fixed point $x$, where $W \subseteq T_{x}X$ is taken as the contact hyperplane. It follows from a result in representation theory of Vinberg that the GIT quotient space of codimension two linear sections of $S$ is isomorphic to the weighted projective space $\mathbb{P}(1,3,4,6)$. In this note, we investigate the problem of finding a geometric interpretation of the above isomorphism. As a main result, for each $\mathfrak{g}$ of the above type, we construct a natural open embedding of the GIT quotient space of nonsingular codimension two linear sections of $S$ into $\mathbb{P}(1,3,4,6)$ whose complement is a fixed hypersurface of degree 12. The key ingredient of our construction is to apply a correspondence of Bahargava and Ho which relates the above moduli problem to a moduli problem on curves of genus one.

math.AG↗

arXiv2023

Make Landscape Flatter in Differentially Private Federated Learning

To defend the inference attacks and mitigate the sensitive information leakages in Federated Learning (FL), client-level Differentially Private FL (DPFL) is the de-facto standard for privacy protection by clipping local updates and adding random noise. However, existing DPFL methods tend to make a sharper loss landscape and have poorer weight perturbation robustness, resulting in severe performance degradation. To alleviate these issues, we propose a novel DPFL algorithm named DP-FedSAM, which leverages gradient perturbation to mitigate the negative impact of DP. Specifically, DP-FedSAM integrates Sharpness Aware Minimization (SAM) optimizer to generate local flatness models with better stability and weight perturbation robustness, which results in the small norm of local updates and robustness to DP noise, thereby improving the performance. From the theoretical perspective, we analyze in detail how DP-FedSAM mitigates the performance degradation induced by DP. Meanwhile, we give rigorous privacy guarantees with Rényi DP and present the sensitivity analysis of local updates. At last, we empirically confirm that our algorithm achieves state-of-the-art (SOTA) performance compared with existing SOTA baselines in DPFL. Code is available at https://github.com/YMJS-Irfan/DP-FedSAM

cs.LG↗

arXiv2023

Towards More Suitable Personalization in Federated Learning via Decentralized Partial Model Training

Personalized federated learning (PFL) aims to produce the greatest personalized model for each client to face an insurmountable problem--data heterogeneity in real FL systems. However, almost all existing works have to face large communication burdens and the risk of disruption if the central server fails. Only limited efforts have been used in a decentralized way but still suffers from inferior representation ability due to sharing the full model with its neighbors. Therefore, in this paper, we propose a personalized FL framework with a decentralized partial model training called DFedAlt. It personalizes the "right" components in the modern deep models by alternately updating the shared and personal parameters to train partially personalized models in a peer-to-peer manner. To further promote the shared parameters aggregation process, we propose DFedSalt integrating the local Sharpness Aware Minimization (SAM) optimizer to update the shared parameters. It adds proper perturbation in the direction of the gradient to overcome the shared model inconsistency across clients. Theoretically, we provide convergence analysis of both algorithms in the general non-convex setting for decentralized partial model training in PFL. Our experiments on several real-world data with various data partition settings demonstrate that (i) decentralized training is more suitable for partial personalization, which results in state-of-the-art (SOTA) accuracy compared with the SOTA PFL baselines; (ii) the shared parameters with proper perturbation make partial personalized FL more suitable for decentralized training, where DFedSalt achieves most competitive performance.

cs.LG↗

arXiv2023

Towards the Flatter Landscape and Better Generalization in Federated Learning under Client-level Differential Privacy

To defend the inference attacks and mitigate the sensitive information leakages in Federated Learning (FL), client-level Differentially Private FL (DPFL) is the de-facto standard for privacy protection by clipping local updates and adding random noise. However, existing DPFL methods tend to make a sharp loss landscape and have poor weight perturbation robustness, resulting in severe performance degradation. To alleviate these issues, we propose a novel DPFL algorithm named DP-FedSAM, which leverages gradient perturbation to mitigate the negative impact of DP. Specifically, DP-FedSAM integrates Sharpness Aware Minimization (SAM) optimizer to generate local flatness models with improved stability and weight perturbation robustness, which results in the small norm of local updates and robustness to DP noise, thereby improving the performance. To further reduce the magnitude of random noise while achieving better performance, we propose DP-FedSAM-$top_k$ by adopting the local update sparsification technique. From the theoretical perspective, we present the convergence analysis to investigate how our algorithms mitigate the performance degradation induced by DP. Meanwhile, we give rigorous privacy guarantees with Rényi DP, the sensitivity analysis of local updates, and generalization analysis. At last, we empirically confirm that our algorithms achieve state-of-the-art (SOTA) performance compared with existing SOTA baselines in DPFL.

cs.LG↗

arXiv2023

Detecting Backdoors in Pre-trained Encoders

Self-supervised learning in computer vision trains on unlabeled data, such as images or (image, text) pairs, to obtain an image encoder that learns high-quality embeddings for input data. Emerging backdoor attacks towards encoders expose crucial vulnerabilities of self-supervised learning, since downstream classifiers (even further trained on clean data) may inherit backdoor behaviors from encoders. Existing backdoor detection methods mainly focus on supervised learning settings and cannot handle pre-trained encoders especially when input labels are not available. In this paper, we propose DECREE, the first backdoor detection approach for pre-trained encoders, requiring neither classifier headers nor input labels. We evaluate DECREE on over 400 encoders trojaned under 3 paradigms. We show the effectiveness of our method on image encoders pre-trained on ImageNet and OpenAI's CLIP 400 million image-text pairs. Our method consistently has a high detection accuracy even if we have only limited or no access to the pre-training dataset.

cs.CV↗

arXiv2023

FLIP: A Provable Defense Framework for Backdoor Mitigation in Federated Learning

Federated Learning (FL) is a distributed learning paradigm that enables different parties to train a model together for high quality and strong privacy protection. In this scenario, individual participants may get compromised and perform backdoor attacks by poisoning the data (or gradients). Existing work on robust aggregation and certified FL robustness does not study how hardening benign clients can affect the global model (and the malicious clients). In this work, we theoretically analyze the connection among cross-entropy loss, attack success rate, and clean accuracy in this setting. Moreover, we propose a trigger reverse engineering based defense and show that our method can achieve robustness improvement with guarantee (i.e., reducing the attack success rate) without affecting benign accuracy. We conduct comprehensive experiments across different datasets and attack settings. Our results on eight competing SOTA defense methods show the empirical superiority of our method on both single-shot and continuous FL backdoor attacks. Code is available at https://github.com/KaiyuanZh/FLIP.

cs.CR↗

arXiv2023

A characterization of irreducible Hermitian symmetric spaces of tube type by $\mathbb{C}^{*}$-actions

A $\mathbb{C}^{*}$-action on a projective variety $X$ is said to be of Euler type at a nonsingular fixed point $x$ if the isotropy action of $\mathbb{C}^{*}$ on $T_{x}X$ is by scalar multiplication. In this paper, it's proven that a smooth projective variety of Picard number one $X$ is isomorphic to an irreducible Hermitian symmetric space of tube type if and only if for a general pair of points $x,y$ on $X$, there exists a $\mathbb{C}^{*}$-action on $X$ which is of Euler type at $x$ and its inverse action is of Euler type at $y$.

math.AG↗