arXiv subjects

Xiaojun Chen

Publications and source records attributed to Xiaojun Chen.

At least 37 records · Page 2Linked to original sources

arXiv2026

High-Throughput and Scalable Secure Inference Protocols for Deep Learning with Packed Secret Sharing

Most existing secure neural network inference protocols based on secure multi-party computation (MPC) typically support at most four participants, demonstrating severely limited scalability. Liu et al. (USENIX Security'24) presented the first relatively practical approach by utilizing Shamir secret sharing with Mersenne prime fields. However, when processing deeper neural networks such as VGG16, their protocols incur substantial communication overhead, resulting in particularly significant latency in wide-area network (WAN) environments. In this paper, we propose a high-throughput and scalable MPC protocol for neural network inference against semi-honest adversaries in the honest-majority setting. The core of our approach lies in leveraging packed Shamir secret sharing (PSS) to enable parallel computation and reduce communication complexity. The main contributions are three-fold: i) We present a communication-efficient protocol for vector-matrix multiplication, based on our newly defined notion of vector-matrix multiplication-friendly random share tuples. ii) We design the filter packing approach that enables parallel convolution. iii) We further extend all non-linear protocols based on Shamir secret sharing to the PSS-based protocols for achieving parallel non-linear operations. Extensive experiments across various datasets and neural networks demonstrate the superiority of our approach in WAN. Compared to Liu et al. (USENIX Security'24), our scheme reduces the communication upto 5.85x, 11.17x, and 6.83x in offline, online and total communication overhead, respectively. In addition, our scheme is upto 1.59x, 2.61x, and 1.75x faster in offline, online and total running time, respectively.

cs.CR↗

arXiv2026

An Adaptive Smoothing Algorithm for Non-Lipschitz Optimization on Manifolds with Complexity Guarantees

We study a class of optimization problems on Riemannian manifolds, where the objective function consists of a smooth term and quasi-norm type penalties with exponent $p \in (0, 1]$. The essential difficulty lies in the fact that the objective function may not be locally Lipschitz continuous, which places this type of problems beyond the reach of existing Riemannian techniques. To overcome this obstacle, this paper constructs a general smoothing framework and establishes fundamental properties for developing efficient algorithms. In particular, we propose a smoothing Riemannian gradient algorithm equipped with a smoothing-aware AdaGrad-type stepsize rule. Its global convergence is demonstrated together with an iteration complexity of $O (ε^{p - 4})$, which includes the best available iteration complexity of $O (ε^{- 3})$ for Lipschitz problems with $p = 1$ as a special case. To the best of our knowledge, this is the first complexity result for non-Lipschitz optimization on Riemannian manifolds. Preliminary numerical experiments corroborate the practical efficiency of the proposed approach in real-world applications arsing from machine learning and data science.

math.OC↗

arXiv2026

Bayesian distributionally robust variational inequalities: regularization and quantification

We propose a Bayesian distributionally robust variational inequality (DRVI) framework that models the data-generating distribution through a finite mixture family, which allows us to study the DRVI on a tractable finite-dimensional parametric ambiguity set. To address distributional uncertainty, we construct a data-driven ambiguity set with posterior coverage guarantees via Bayesian inference. We also employ a regularization approach to ensure numerical stability. We prove the existence of solutions to the Bayesian DRVI and the asymptotic convergence to a solution as sample size grows to infinity and the regularization parameter goes to zero. Moreover, we derive quantitative stability bounds and finite-sample guarantees under data scarcity and contamination. Numerical experiments on a distributionally robust multi-portfolio Nash equilibrium problem validate our theoretical results and demonstrate the robustness and reliability of Bayesian DRVI solutions in practice.

math.OC↗

arXiv2026

Nonconvex Nonsmooth Multicomposite Optimization and Its Applications to Recurrent Neural Networks

We consider a class of nonconvex nonsmooth multicomposite optimization problems where the objective function consists of a Tikhonov regularizer and a composition of multiple nonconvex nonsmooth component functions. Such optimization problems arise from tangible applications in machine learning and beyond. To define and compute its first-order and second-order d(irectional)-stationary points effectively, we first derive the closed-form expression of the tangent cone for the feasible region of its constrained reformulation. Building on this, we establish its equivalence with the corresponding constrained and $\ell_1$-penalty reformulations in terms of global optimality and d-stationarity. The equivalence offers indirect methods to attain the first-order and second-order d-stationary points of the original problem in certain cases. We apply our results to the training process of recurrent neural networks (RNNs).

math.OC↗

arXiv2026

Optimization with Parametric Variational Inequality Constraints on a Moving Set

This paper focuses on optimization problems constrained by Parametric Variational Inequalities (PVI) defined on a moving set. Unlike most existing works on mathematical programs with equilibrium constraints, the equilibrium constraints have parameters not only in the function but also in the related set. We show that the solution function of the PVI is Lipschitz continuous with respect to the upper-level decision variables and the solution set of the optimization problem is nonempty and bounded. Moreover, we prove that the metric regularity of the constraints holds automatically, which allow us to characterize stationary points without any additional assumptions. A Smoothing Implicit Gradient Algorithm (SIGA) is proposed based on the smoothing approximation of the PVI. We prove the convergence of SIGA to a stationary point of the optimization problem and numerically validate the efficiency of SIGA by portfolio management problems with real data.

math.OC↗

arXiv2026

MLRecon: Robust Markerless Freehand 3D Ultrasound Reconstruction via Coarse-to-Fine Pose Estimation

Freehand 3D ultrasound (US) reconstruction promises volumetric imaging with the flexibility of standard 2D probes, yet existing tracking paradigms face a restrictive trilemma: marker-based systems demand prohibitive costs, inside-out methods require intrusive sensor attachment, and sensorless approaches suffer from severe cumulative drift. To overcome these limitations, we present MLRecon, a robust markerless 3D US reconstruction framework delivering drift-resilient 6D probe pose tracking using a single commodity RGB-D camera. Leveraging the generalization power of vision foundation models, our pipeline enables continuous markerless tracking of the probe, augmented by a vision-guided divergence detector that autonomously monitors tracking integrity and triggers failure recovery to ensure uninterrupted scanning. Crucially, we further propose a dual-stage pose refinement network that explicitly disentangles high-frequency jitter from low-frequency bias, effectively denoising the trajectory while maintaining the kinematic fidelity of operator maneuvers. Experiments demonstrate that MLRecon significantly outperforms competing sensorless and sensor-aided methods, achieving average position errors as low as 0.88 mm on complex trajectories and yielding high-quality 3D reconstructions with sub-millimeter mean surface accuracy. This establishes a new benchmark for low-cost, accessible volumetric US imaging in resource-limited clinical settings.

cs.CV↗

arXiv2026

Complexity of Projected Gradient Methods for Strongly Convex Optimization with Hölder Continuous Gradient Terms

This paper studies the complexity of projected gradient descent methods for a class of strongly convex constrained optimization problems where the objective function is expressed as a summation of $m$ component functions, each possessing a gradient that is Hölder continuous with an exponent $α_i \in (0, 1]$. Under this formulation, the gradient of the objective function may fail to be globally Hölder continuous, thereby rendering existing complexity results inapplicable to this class of problems. Our theoretical analysis reveals that, in this setting, the complexity of projected gradient methods is determined by $\hatα = \min_{i \in \{1, \dotsc, m\}} α_i$. We first prove that, with an appropriately fixed stepsize, the complexity bound for finding an approximate minimizer with a distance to the true minimizer less than $\varepsilon$ is $O (\log (\varepsilon^{-1}) \varepsilon^{2 (\hatα - 1) / (1 + \hatα)})$, which extends the well-known complexity result for $\hatα = 1$. Next we show that the complexity bound can be improved to $O (\log (\varepsilon^{-1}) \varepsilon^{2 (\hatα - 1) / (1 + 3 \hatα)})$ if the stepsize is updated by the universal scheme. We illustrate our complexity results by numerical examples arising from elliptic equations with a non-Lipschitz term.

math.OC↗

arXiv2026

PointSLAM++: Robust Dense Neural Gaussian Point Cloud-based SLAM

Real-time 3D reconstruction is crucial for robotics and augmented reality, yet current simultaneous localization and mapping(SLAM) approaches often struggle to maintain structural consistency and robust pose estimation in the presence of depth noise. This work introduces PointSLAM++, a novel RGB-D SLAM system that leverages a hierarchically constrained neural Gaussian representation to preserve structural relationships while generating Gaussian primitives for scene mapping. It also employs progressive pose optimization to mitigate depth sensor noise, significantly enhancing localization accuracy. Furthermore, it utilizes a dynamic neural representation graph that adjusts the distribution of Gaussian nodes based on local geometric complexity, enabling the map to adapt to intricate scene details in real time. This combination yields high-precision 3D mapping and photorealistic scene rendering. Experimental results show PointSLAM++ outperforms existing 3DGS-based SLAM methods in reconstruction accuracy and rendering quality, demonstrating its advantages for large-scale AR and robotics.

cs.CV↗

arXiv2025

LiNeXt: Revisiting LiDAR Completion with Efficient Non-Diffusion Architectures

3D LiDAR scene completion from point clouds is a fundamental component of perception systems in autonomous vehicles. Previous methods have predominantly employed diffusion models for high-fidelity reconstruction. However, their multi-step iterative sampling incurs significant computational overhead, limiting its real-time applicability. To address this, we propose LiNeXt-a lightweight, non-diffusion network optimized for rapid and accurate point cloud completion. Specifically, LiNeXt first applies the Noise-to-Coarse (N2C) Module to denoise the input noisy point cloud in a single pass, thereby obviating the multi-step iterative sampling of diffusion-based methods. The Refine Module then takes the coarse point cloud and its intermediate features from the N2C Module to perform more precise refinement, further enhancing structural completeness. Furthermore, we observe that LiDAR point clouds exhibit a distance-dependent spatial distribution, being densely sampled at proximal ranges and sparsely sampled at distal ranges. Accordingly, we propose the Distance-aware Selected Repeat strategy to generate a more uniformly distributed noisy point cloud. On the SemanticKITTI dataset, LiNeXt achieves a 199.8x speedup in inference, reduces Chamfer Distance by 50.7%, and uses only 6.1% of the parameters compared with LiDiff. These results demonstrate the superior efficiency and effectiveness of LiNeXt for real-time scene completion.

cs.CV↗

arXiv2025

RWKV-PCSSC: Exploring RWKV Model for Point Cloud Semantic Scene Completion

Semantic Scene Completion (SSC) aims to generate a complete semantic scene from an incomplete input. Existing approaches often employ dense network architectures with a high parameter count, leading to increased model complexity and resource demands. To address these limitations, we propose RWKV-PCSSC, a lightweight point cloud semantic scene completion network inspired by the Receptance Weighted Key Value (RWKV) mechanism. Specifically, we introduce a RWKV Seed Generator (RWKV-SG) module that can aggregate features from a partial point cloud to produce a coarse point cloud with coarse features. Subsequently, the point-wise feature of the point cloud is progressively restored through multiple stages of the RWKV Point Deconvolution (RWKV-PD) modules. By leveraging a compact and efficient design, our method achieves a lightweight model representation. Experimental results demonstrate that RWKV-PCSSC reduces the parameter count by 4.18$\times$ and improves memory efficiency by 1.37$\times$ compared to state-of-the-art methods PointSSC. Furthermore, our network achieves state-of-the-art performance on established indoor (SSC-PC, NYUCAD-PC) and outdoor (PointSSC) scene dataset, as well as on our proposed datasets (NYUCAD-PC-V2, 3D-FRONT-PC).

cs.CV↗

arXiv2025

DeepTracer: Tracing Stolen Model via Deep Coupled Watermarks

Model watermarking techniques can embed watermark information into the protected model for ownership declaration by constructing specific input-output pairs. However, existing watermarks are easily removed when facing model stealing attacks, and make it difficult for model owners to effectively verify the copyright of stolen models. In this paper, we analyze the root cause of the failure of current watermarking methods under model stealing scenarios and then explore potential solutions. Specifically, we introduce a robust watermarking framework, DeepTracer, which leverages a novel watermark samples construction method and a same-class coupling loss constraint. DeepTracer can incur a high-coupling model between watermark task and primary task that makes adversaries inevitably learn the hidden watermark task when stealing the primary task functionality. Furthermore, we propose an effective watermark samples filtering mechanism that elaborately select watermark key samples used in model ownership verification to enhance the reliability of watermarks. Extensive experiments across multiple datasets and models demonstrate that our method surpasses existing approaches in defending against various model stealing attacks, as well as watermark attacks, and achieves new state-of-the-art effectiveness and robustness.

cs.CR↗

arXiv2025

Value-Aligned Prompt Moderation via Zero-Shot Agentic Rewriting for Safe Image Generation

Generative vision-language models like Stable Diffusion demonstrate remarkable capabilities in creative media synthesis, but they also pose substantial risks of producing unsafe, offensive, or culturally inappropriate content when prompted adversarially. Current defenses struggle to align outputs with human values without sacrificing generation quality or incurring high costs. To address these challenges, we introduce VALOR (Value-Aligned LLM-Overseen Rewriter), a modular, zero-shot agentic framework for safer and more helpful text-to-image generation. VALOR integrates layered prompt analysis with human-aligned value reasoning: a multi-level NSFW detector filters lexical and semantic risks; a cultural value alignment module identifies violations of social norms, legality, and representational ethics; and an intention disambiguator detects subtle or indirect unsafe implications. When unsafe content is detected, prompts are selectively rewritten by a large language model under dynamic, role-specific instructions designed to preserve user intent while enforcing alignment. If the generated image still fails a safety check, VALOR optionally performs a stylistic regeneration to steer the output toward a safer visual domain without altering core semantics. Experiments across adversarial, ambiguous, and value-sensitive prompts show that VALOR significantly reduces unsafe outputs by up to 100.00% while preserving prompt usefulness and creativity. These results highlight VALOR as a scalable and effective approach for deploying safe, aligned, and helpful image generation systems in open-world settings.

cs.AI↗

arXiv2025

Solutions of Two-stage Stochastic Minimax Problems

This paper introduces a class of two-stage stochastic minimax problems where the first-stage objective function is nonconvex-concave while the second-stage objective function is strongly convex-concave. We establish properties of the second-stage minimax value function and solution functions, and characterize the existence and relationships among saddle points, minimax points, and KKT points. We apply the sample average approximation (SAA) to the class of two-stage stochastic minimax problems and prove the convergence of the KKT points as the sample size tends to infinity. An inexact parallel proximal gradient descent ascent algorithm is proposed to solve this class of problems with the SAA. Numerical experiments demonstrate the effectiveness of the proposed algorithm and validate the convergence properties of the SAA approach.

math.OC↗

arXiv2025

Singularity categories and singular loci of certain quotient singularities

Let $V$ be a finite dimensional $k$-vector space, where $k$ is an algebraic closed field of characteristic zero. Let $G \subseteq \mathrm{SL}(V)$ be a finite abelian group, and denote by $S$ the $G$-invariant subring of the polynomial ring $k[V]$. It is shown that the singularity category $D_{sg}(S)$ recovers the reduced singular locus of $\mathrm{Spec}(S)$.

math.AG↗

arXiv2025

Who Speaks for the Trigger? Dynamic Expert Routing in Backdoored Mixture-of-Experts Transformers

Large language models (LLMs) with Mixture-of-Experts (MoE) architectures achieve impressive performance and efficiency by dynamically routing inputs to specialized subnetworks, known as experts. However, this sparse routing mechanism inherently exhibits task preferences due to expert specialization, introducing a new and underexplored vulnerability to backdoor attacks. In this work, we investigate the feasibility and effectiveness of injecting backdoors into MoE-based LLMs by exploiting their inherent expert routing preferences. We thus propose BadSwitch, a novel backdoor framework that integrates task-coupled dynamic trigger optimization with a sensitivity-guided Top-S expert tracing mechanism. Our approach jointly optimizes trigger embeddings during pretraining while identifying S most sensitive experts, subsequently constraining the Top-K gating mechanism to these targeted experts. Unlike traditional backdoor attacks that rely on superficial data poisoning or model editing, BadSwitch primarily embeds malicious triggers into expert routing paths with strong task affinity, enabling precise and stealthy model manipulation. Through comprehensive evaluations across three prominent MoE architectures (Switch Transformer, QwenMoE, and DeepSeekMoE), we demonstrate that BadSwitch can efficiently hijack pre-trained models with up to 100% success rate (ASR) while maintaining the highest clean accuracy (ACC) among all baselines. Furthermore, BadSwitch exhibits strong resilience against both text-level and model-level defense mechanisms, achieving 94.07% ASR and 87.18% ACC on the AGNews dataset. Our analysis of expert activation patterns reveals fundamental insights into MoE vulnerabilities. We anticipate this work will expose security risks in MoE systems and contribute to advancing AI safety.

cs.CR↗

arXiv2025

AILoRA: Function-Aware Asymmetric Initialization for Low-Rank Adaptation of Large Language Models

Parameter-efficient finetuning (PEFT) aims to mitigate the substantial computational and memory overhead involved in adapting large-scale pretrained models to diverse downstream tasks. Among numerous PEFT strategies, Low-Rank Adaptation (LoRA) has emerged as one of the most widely adopted approaches due to its robust empirical performance and low implementation complexity. In practical deployment, LoRA is typically applied to the $W^Q$ and $W^V$ projection matrices of self-attention modules, enabling an effective trade-off between model performance and parameter efficiency. While LoRA has achieved considerable empirical success, it still encounters challenges such as suboptimal performance and slow convergence. To address these limitations, we introduce \textbf{AILoRA}, a novel parameter-efficient method that incorporates function-aware asymmetric low-rank priors. Our empirical analysis reveals that the projection matrices $W^Q$ and $W^V$ in the self-attention mechanism exhibit distinct parameter characteristics, stemming from their functional differences. Specifically, $W^Q$ captures task-specific semantic space knowledge essential for attention distributions computation, making its parameters highly sensitive to downstream task variations. In contrast, $W^V$ encodes token-level feature representations that tend to remain stable across tasks and layers. Leveraging these insights, AILoRA performs a function-aware initialization by injecting the principal components of $W^Q$ to retain task-adaptive capacity, and the minor components of $W^V$ to preserve generalizable feature representations. This asymmetric initialization strategy enables LoRA modules to better capture the specialized roles of attention parameters, thereby enhancing both finetuning performance and convergence efficiency.

cs.AI↗

arXiv2025

On the graded singularity category of Abelian quotient singularities, I. Smooth categorical compactification

Given a singularity which is the quotient of an affine space $V$ by a finite Abelian group $G \subseteq \mathrm{SL}(V)$, we study the DG enhancement $\mathcal{D}^{b}(\mathrm{tails}(k[V]^G))$ of the bounded derived category of the non-commutative projective space $\mathrm{tails}(k[V]^G)$ and the DG enhancement $\mathcal{D}_{sg}^{\mathbb{Z}}(k[V]^G)$ of its graded singularity category. In this paper, we construct smooth categorical compactifications, in the sense of Efimov, of $\mathcal{D}^b(\mathrm{tails}(k[V]^G))$ and $\mathcal{D}_{sg}^{\mathbb{Z}}(k[V]^G)$ respectively, via the non-commutative crepant resolution of $k[V]^G$. We give explicit constructions of canonical classical generators in the kernels of such compactifications.

math.AG↗

arXiv2025

Orthogonal Constrained Minimization with Tensor $\ell_{2,p}$ Regularization for HSI Denoising and Destriping

Hyperspectral images~(HSIs) are often contaminated by a mixture of noise such as Gaussian noise, dead lines, stripes, and so on. In this paper, we propose a multi-scale low-rank tensor regularized $\ell_{2,p}$ (MLTL2p) approach for HSI denoising and destriping, which consists of an orthogonal constrained minimization model and an iterative algorithm with convergence guarantees. The model of the proposed MLTL2p approach is built based on a new sparsity-enhanced Multi-scale Low-rank Tensor regularization and a tensor $\ell_{2,p}$ norm with $p\in (0,1)$. The multi-scale low-rank regularization for HSI denoising utilizes the global and local spectral correlation as well as the spatial nonlocal self-similarity priors of HSIs. The corresponding low-rank constraints are formulated based on independent higher-order singular value decomposition with sparsity enhancement on its core tensor to prompt more low-rankness. The tensor $\ell_{2,p}$ norm for HSI destriping is extended from the matrix $\ell_{2,p}$ norm. A proximal block coordinate descent algorithm is proposed in the MLTL2p approach to solve the resulting nonconvex nonsmooth minimization with orthogonal constraints. We show any accumulation point of the sequence generated by the proposed algorithm converges to a first-order stationary point, which is defined using three equalities of substationarity, symmetry, and feasibility for orthogonal constraints. In the numerical experiments, we compare the proposed method with state-of-the-art methods including a deep learning based method, and test the methods on both simulated and real HSI datasets. Our proposed MLTL2p method demonstrates outperformance in terms of metrics such as mean peak signal-to-noise ratio as well as visual quality.

math.OC↗