arXiv subjects

Lele Zheng

Publications and source records attributed to Lele Zheng.

9 recordsLinked to original sources

arXiv2026

C$^2$T-OpenMax: A Novel Open-Set WiFi RF Fingerprinting Method via Center Constrained Learning and Confidence-Guided Tail Modeling

Radio frequency fingerprinting (RFF) enables device authentication from transmitter-specific hardware imperfections, but practical deployment requires cross-environment open-set recognition. Data augmentation improves environmental generalization, yet may yield dispersed, low-confidence known-class representations that distort the class statistics used by OpenMax. To address this problem, we propose C$^2$T-OpenMax, an enhanced OpenMax framework combining center-constrained learning with confidence-guided tail modeling. The former improves intra-class compactness, making class-wise representations more suitable for distance-based modeling. The latter retains only correctly classified, high-confidence logits for mean activation vector estimation and Weibull fitting, reducing bias from ambiguous boundary samples. Together, the two modules refine representation geometry and OpenMax construction while preserving augmentation benefits. Experiments on a public WiFi CSI dataset show that C$^2$T-OpenMax achieves the highest open-set accuracy in seven of eight location groups and outperforms all baselines in area under the receiver operating characteristic curve (AUROC) and open-set classification rate (OSCR) across every tested openness level. Under the largest-openness setting, it improves accuracy by 12.31%, AUROC by 0.0887, and OSCR by 0.0856 over the augmented OpenMax baseline.

cs.CR↗

arXiv2026

FedGSA: Geometry-Consistent Subspace Aggregation for Differentially Private Federated LoRA

Low-Rank Adaptation (LoRA) enables communication-efficient federated fine-tuning of pretrained language models. However, integrating differential privacy (DP) into federated LoRA remains challenging: independently perturbing and aggregating its two low-rank matrices can cause aggregation mismatch and the quadratic noise term. Existing methods mitigate these issues by freezing one low-rank matrix but still rely on Euclidean aggregation, which is basis-dependent and may distort the global update. To address this limitation, we propose FedGSA, a geometry-consistent aggregation framework for differentially private federated LoRA. FedGSA represents each privatized client update as a basis-invariant subspace on the Grassmann manifold. In each communication round, clients extract low-dimensional subspaces capturing dominant update directions and encode them as projection matrices. The server aggregates these representations to estimate a geometry-consistent global update subspace and reconstructs the global LoRA factors within it, reducing distortion caused by basis misalignment, privacy noise, and heterogeneous client updates. We prove that FedGSA incurs no additional privacy loss beyond client-side DP training and establish its convergence under standard assumptions. Experiments on four GLUE tasks and a language generation benchmark demonstrate consistent improvements across privacy budgets and degrees of data heterogeneity. In particular, FedGSA improves average accuracy over the strongest baseline by 2.17% and 2.27% under $\epsilon=6$ and $\epsilon=3$, respectively.

cs.CR↗

arXiv2026

Noise-Aware Shrinkage for Differentially Private Zeroth-Order Fine-Tuning of Large Language Models

Differentially private zeroth-order optimization (DP-ZO) enables memory-efficient private fine-tuning of large language models using only forward evaluations. Existing aggregation-based DP-ZO methods reconstruct model updates at a fixed scale, ignoring that the strength of useful signals varies throughout training. Consequently, noise-dominated updates may receive excessive weight and degrade model utility. To address this issue, we propose SAGE, a noise-aware shrinkage method that adaptively attenuates privatized estimates according to their estimated signal quality. SAGE subtracts the known Gaussian noise variance from the observed second moment to estimate the underlying signal energy, stabilizes this estimate through temporal tracking, and compares its current signal-to-noise level with a warm-up reference to derive a bounded shrinkage factor. As pure post-processing, SAGE requires neither additional privacy budget nor model queries and introduces only constant additional state. Our theoretical analysis shows that shrinkage reduces the quadratic update-risk term faster than the linear descent term, preserving useful descent while limiting the influence of noise-dominated updates. Experiments on RoBERTa-large, OPT-1.3B, and OPT-6.7B demonstrate that SAGE outperforms existing baselines in most settings under the same privacy budgets while preserving the forward-only memory efficiency of DP-ZO.

cs.LG↗

arXiv2026

When Convenience Becomes Risk: A Semantic View of Under-Specification in Host-Acting Agents

Host-acting agents promise a convenient interaction model in which users specify goals and the system determines how to realize them. We argue that this convenience introduces a distinct security problem: semantic under-specification in goal specification. User instructions are typically goal-oriented, yet they often leave process constraints, safety boundaries, persistence, and exposure insufficiently specified. As a result, the agent must complete missing execution semantics before acting, and this completion can produce risky host-side plans even when the user-stated goal is benign. In this paper, we develop a semantic threat model, present a taxonomy of semantic-induced risky completion patterns, and study the phenomenon through an OpenClaw-centered case study and execution-trace analysis. We further derive defense design principles for making execution boundaries explicit and constraining risky completion. These findings suggest that securing host-acting agents requires governing not only which actions are allowed at execution time, but also how goal-only instructions are translated into executable plans.

cs.CR↗

arXiv2026

Differentially Private Subspace Fine-Tuning for Large Language Models

Fine-tuning large language models on downstream tasks is crucial for realizing their cross-domain potential but often relies on sensitive data, raising privacy concerns. Differential privacy (DP) offers rigorous privacy guarantees and has been widely adopted in fine-tuning; however, naively injecting noise across the high-dimensional parameter space creates perturbations with large norms, degrading performance and destabilizing training. To address this issue, we propose DP-SFT, a two-stage subspace fine-tuning method that substantially reduces noise magnitude while preserving formal DP guarantees. Our intuition is that, during fine-tuning, significant parameter updates lie within a low-dimensional, task-specific subspace, while other directions change minimally. Hence, we only inject DP noise into this subspace to protect privacy without perturbing irrelevant parameters. In phase one, we identify the subspace by analyzing principal gradient directions to capture task-specific update signals. In phase two, we project full gradients onto this subspace, add DP noise, and map the perturbed gradients back to the original parameter space for model updates, markedly lowering noise impact. Experiments on multiple datasets demonstrate that DP-SFT enhances accuracy and stability under rigorous DP constraints, accelerates convergence, and achieves substantial gains over DP fine-tuning baselines.

cs.LG↗

arXiv2025

Privacy in Fine-tuning Large Language Models: Attacks, Defenses, and Future Directions

Fine-tuning has emerged as a critical process in leveraging Large Language Models (LLMs) for specific downstream tasks, enabling these models to achieve state-of-the-art performance across various domains. However, the fine-tuning process often involves sensitive datasets, introducing privacy risks that exploit the unique characteristics of this stage. In this paper, we provide a comprehensive survey of privacy challenges associated with fine-tuning LLMs, highlighting vulnerabilities to various privacy attacks, including membership inference, data extraction, and backdoor attacks. We further review defense mechanisms designed to mitigate privacy risks in the fine-tuning phase, such as differential privacy, federated learning, and knowledge unlearning, discussing their effectiveness and limitations in addressing privacy risks and maintaining model utility. By identifying key gaps in existing research, we highlight challenges and propose directions to advance the development of privacy-preserving methods for fine-tuning LLMs, promoting their responsible use in diverse applications.

cs.AI↗

arXiv2024

Extracting Spatiotemporal Data from Gradients with Large Language Models

Recent works show that sensitive user data can be reconstructed from gradient updates, breaking the key privacy promise of federated learning. While success was demonstrated primarily on image data, these methods do not directly transfer to other domains, such as spatiotemporal data. To understand privacy risks in spatiotemporal federated learning, we first propose Spatiotemporal Gradient Inversion Attack (ST-GIA), a gradient attack algorithm tailored to spatiotemporal data that successfully reconstructs the original location from gradients. Furthermore, the absence of priors in attacks on spatiotemporal data has hindered the accurate reconstruction of real client data. To address this limitation, we propose ST-GIA+, which utilizes an auxiliary language model to guide the search for potential locations, thereby successfully reconstructing the original data from gradients. In addition, we design an adaptive defense strategy to mitigate gradient inversion attacks in spatiotemporal federated learning. By dynamically adjusting the perturbation levels, we can offer tailored protection for varying rounds of training data, thereby achieving a better trade-off between privacy and utility than current state-of-the-art methods. Through intensive experimental analysis on three real-world datasets, we reveal that the proposed defense strategy can well preserve the utility of spatiotemporal federated learning with effective security protection.

cs.LG↗

arXiv2024

Enhancing Privacy of Spatiotemporal Federated Learning against Gradient Inversion Attacks

Spatiotemporal federated learning has recently raised intensive studies due to its ability to train valuable models with only shared gradients in various location-based services. On the other hand, recent studies have shown that shared gradients may be subject to gradient inversion attacks (GIA) on images or texts. However, so far there has not been any systematic study of the gradient inversion attacks in spatiotemporal federated learning. In this paper, we explore the gradient attack problem in spatiotemporal federated learning from attack and defense perspectives. To understand privacy risks in spatiotemporal federated learning, we first propose Spatiotemporal Gradient Inversion Attack (ST-GIA), a gradient attack algorithm tailored to spatiotemporal data that successfully reconstructs the original location from gradients. Furthermore, we design an adaptive defense strategy to mitigate gradient inversion attacks in spatiotemporal federated learning. By dynamically adjusting the perturbation levels, we can offer tailored protection for varying rounds of training data, thereby achieving a better trade-off between privacy and utility than current state-of-the-art methods. Through intensive experimental analysis on three real-world datasets, we reveal that the proposed defense strategy can well preserve the utility of spatiotemporal federated learning with effective security protection.

cs.CR↗

arXiv2022

Graph Analysis in Decentralized Online Social Networks with Fine-Grained Privacy Protection

Graph analysts cannot directly obtain the global structure in decentralized social networks, and analyzing such a network requires collecting local views of the social graph from individual users. Since the edges between users may reveal sensitive social interactions in the local view, applying differential privacy in the data collection process is often desirable, which provides strong and rigorous privacy guarantees. In practical decentralized social graphs, different edges have different privacy requirements due to the distinct sensitivity levels. However, the existing differentially private analysis of social graphs provide the same protection for all edges. To address this issue, this work proposes a fine-grained privacy notion as well as novel algorithms for private graph analysis. We first design a fine-grained relationship differential privacy (FGR-DP) notion for social graph analysis, which enforces different protections for the edges with distinct privacy requirements. Then, we design algorithms for triangle counting and k-stars counting, respectively, which can accurately estimate subgraph counts given fine-grained protection for social edges. We also analyze upper bounds on the estimation error, including k-stars and triangle counts, and show their superior performance compared with the state-of-the-arts. Finally, we perform extensive experiments on two real social graph datasets and demonstrate that the proposed mechanisms satisfying FGR-DP have better utility than the state-of-the-art mechanisms due to the finer-grained protection.

cs.CR↗