arXiv subjects

Mohamed Seif

Publications and source records attributed to Mohamed Seif.

At least 19 recordsLinked to original sources

arXiv2026

Watermarkability: Measuring and Improving Language-Model Compatibility with Logit-Based Watermarks

Language-model watermarks can help identify model-generated text, encode provenance information, and detect post-generation edits. A broad class of logit-based watermarks increases the logits of a designated token set at each decoding step. Because a base model's native probability mass on this set varies across contexts, models, and watermark rules, the same logit bias can yield different adherence rates and increases in perplexity (PPL). We introduce watermarkability to characterize the compatibility between a model and the token sets requested by a watermark. At a target mass $τ$, we quantify this property through a watermarkability cost: the expected minimum KL divergence from the base-model distribution required to assign probability at least $τ$ to the requested set. A lower cost means that the model is more watermarkable for that request distribution. To reduce this cost, we propose a two-stage method. In the first stage, we train a target-set-conditioned low-rank adapter on unions drawn from a 16-way vocabulary partition to increase the requested set's probability mass. In the second stage, a closed-form controller applies the smallest remaining uniform logit bias needed to reach the target mass. We evaluate the method on six watermark schedules using OPT-125M and Qwen3-8B. Across the 12 model-schedule pairs, the pre-adaptation watermarkability cost correlates with the log-PPL increase under a near-hard fixed-bias baseline (pooled Pearson $r=0.929$). For every schedule, at least 99.80% of sampled tokens fall in the target set requested at their decoding step. At this reliability, the adapter and controller reduce average per-schedule prompt-conditional PPL by 46.3% on OPT-125M and 5.2% on Qwen3-8B relative to the fixed-bias baseline.

cs.CL↗

arXiv2026

NOEMA: Executable Contracts for Learned Wireless Comparisons

Learning-based wireless research increasingly combines simulation, external model training, and benchmarking. Rebuilding these stages for every experiment is time-consuming, while differences introduced between them can silently change the conditions of a comparison. We present NOEMA, an open-source toolkit that prepares model development and baseline evaluation from a shared, machine-checkable wireless scenario. From the same scenario, NOEMA can prepare benchmark execution, capture aligned training data, and export differentiable components for model development. Researchers can therefore focus on training and selecting their models, then return the selected component to the same controlled experiment for evaluation. NOEMA checks the returned model and verifies that declared comparison conditions, including available information, random conditions, and failure accounting, remain consistent through benchmarking and reporting. An included agentic-control example applies these contracts to a language-model supervisor that selects allocation policies and power budgets from historical link feedback. All 43 regression tests of the workflow checks produced their expected outcomes, including rejection of all 31 cases containing known inconsistencies. Separate fidelity tests showed that captured data and returned-model behavior were preserved. In a 20-replicate quadrature phase-shift keying phase-tracking study, the mean bit-error-rate difference between the learned receiver and five- pilot smoothing was -0.01526 (95% confidence interval: -0.01547 to -0.01506). NOEMA therefore provides both a reusable path for developing learned wireless components and a way to keep their final comparisons inspectable and checkable.

eess.SP↗

arXiv2026

Anchor-ECC: Local Integrity Checking for Watermarked LLM Outputs via Error-Correcting Codes

LLM watermarking has become an effective approach to distinguishing AI-generated text from human-written text by embedding detectable patterns during generation. However, a small post-generation edit may change the meaning of the text without removing its overall watermark signal, creating a risk that the modified content is still attributed to the original model. We propose Anchor-ECC, which incorporates the error-correcting code (ECC) constraints and explicit boundary anchors into the watermark structure and pairs them with a dynamic-programming decoder to detect and localize post-generation edits. Across Qwen3-8B, Mistral-7B-Instruct-v0.3, and OPT-125M, the approximate-hard setting achieves about 99.7% block-level true positive rate (TPR) with at most 7.6% false alarm rate (FAR) for edit detection under mixed insertions, deletions, and substitutions, while preserving the distinction between watermarked outputs and unwatermarked text. Additional quality experiments identify lower-perplexity configurations that retain strong edit-detection performance. Together, these results extend LLM watermarking from source identification to local integrity verification while supporting configurable trade-offs between detection reliability and generation quality.

cs.CR↗

arXiv2026

How Can Incentives and Cut Layer Selection Influence Data Contribution in Split Federated Learning?

To alleviate the training burden in federated learning while enhancing convergence speed, Split Federated Learning (SFL) has emerged as a promising approach by combining the advantages of federated and split learning. However, despite its advantages, existing SFL studies have largely overlooked the strategic interactions among self-interested participants during the SFL process. In this framework, the SFL model owner can choose the cut layer to balance the training load between the server and clients, ensuring the necessary level of privacy for the clients. Additionally, the SFL model owner sets incentives to encourage client participation in the SFL process. The optimization strategies employed by the SFL model owner influence clients' decisions regarding the amount of data they contribute, taking into account the shared incentives over clients and anticipated energy consumption from both computation and networking during SFL. To address this framework, we model the problem using a hierarchical decision-making approach, formulated as a single-leader multi-follower Stackelberg game. We demonstrate the existence and uniqueness of the Nash equilibrium among clients and analyze the Stackelberg equilibrium by examining the leader's game. Furthermore, we discuss privacy concerns related to differential privacy and the criteria for selecting the minimum required cut layer. Our findings show that the Stackelberg equilibrium solution maximizes the utility for both the clients and the SFL model owner while achieving a well-balanced trade-off between model accuracy and the associated computing and networking overhead during the SFL process.

cs.GT↗

arXiv2026

When Phase Doesn't Matter: Self-Coherent Over-the-Air Computation at Sub-THz

Over-the-air computation (OAC) enables efficient function aggregation in wireless networks by exploiting the superposition property of the multiple-access channel. However, practical deployment of OAC is severely challenged by the reliance on accurate carrier synchronization and coherent reception, which are costly and fragile, especially in short-range and low-complexity systems. In this work, we propose a \emph{self-coherent, synthesizer-free over-the-air computation framework} based on \emph{Kramers--Kronig (KK) reception}. By transmitting a biased aggregate waveform and employing direct detection followed by KK phase reconstruction at the receiver, the proposed scheme eliminates the need for explicit carrier recovery while preserving coherent-like signal aggregation. We develop a signal-domain system model for multi-user OAC under KK reception and provide a synchronization-relaxation analysis demonstrating that the proposed architecture fundamentally removes carrier-frequency offset (CFO) sensitivity between transmitters and receiver. By shifting synchronization complexity away from strict carrier-phase tracking and eliminating distributed phase alignment requirements, the framework reduces control overhead and improves scalability in multi-user aggregation. A detailed per-symbol mean-squared error (MSE) characterization isolates the impact of channel mismatch and KK reconstruction noise, showing that the proposed self-coherent architecture approaches the theoretical performance limits of baseband OAC under practical operating conditions. Finally, we demonstrate that the approach is particularly well suited for mmWave and sub-THz systems, where oscillator phase instability otherwise represents a fundamental bottleneck to scalable coherent OAC.

cs.IT↗

arXiv2026

Toward Autonomous O-RAN: A Multi-Scale Agentic AI Framework for Real-Time Network Control and Management

Open Radio Access Networks (O-RAN) promise flexible 6G network access through disaggregated, software-driven components and open interfaces, but this programmability also increases operational complexity. Multiple control loops coexist across the service management layer and RAN Intelligent Controller (RIC), while independently developed control applications can interact in unintended ways. In parallel, recent advances in generative Artificial Intelligence (AI) are enabling a shift from isolated AI models toward agentic AI systems that can interpret goals, coordinate multiple models and control functions, and adapt their behavior over time. This article proposes a multi-scale agentic AI framework for O-RAN that organizes RAN intelligence as a coordinated hierarchy across the Non-Real-Time (Non-RT), Near-Real-Time (Near-RT), and Real-Time (RT) control loops: (i) A Large Language Model (LLM) agent in the Non-RT RIC translates operator intent into policies and governs model lifecycles. (ii) Small Language Model (SLM) agents in the Near-RT RIC execute low-latency optimization and can activate, tune, or disable existing control applications; and (iii) Wireless Physical-layer Foundation Model (WPFM) agents near the distributed unit provide fast inference close to the air interface. We describe how these agents cooperate through standardized O-RAN interfaces and telemetry. Using a proof-of-concept implementation built on open-source models, software, and datasets, we demonstrate the proposed agentic approach in two representative scenarios: robust operation under non-stationary conditions and intent-driven slice resource control.

cs.NI↗

arXiv2026

Differentially Private Spectral Graph Clustering: Balancing Privacy, Accuracy, and Efficiency

We study spectral graph clustering under edge differential privacy. We propose a matrix shuffling mechanism that combines randomized edge flipping with a random permutation of the adjacency matrix. While edge flipping alone provides only a constant $\varepsilon$ guarantee as the graph grows, shuffling amplifies privacy so that the effective $\varepsilon$ tends to zero with the number of nodes. We develop a unified error analysis framework -- based on Davis--Kahan perturbation theory and a classification-margin bound -- that gives explicit misclassification rates for all the mechanisms considered as a function of the privacy budget, eigengap, and number of communities. Applying this framework, we show that the matrix shuffling mechanism achieves an error rate scaling of $\tilde{O}(1/n)$, a clear improvement over two canonical DP baselines from the private PCA literature: the Gaussian mechanism applied directly to the adjacency matrix (Analyze Gauss) and the noisy power method, both of which scale as $\tilde{O}(1)$ in $n$. We further propose a private spectral gap detection algorithm for estimating the number of communities. Experiments on synthetic and real-world networks validate our theoretical findings.

cs.IT↗

arXiv2026

Protecting Human Activity Signatures in Compressed IEEE 802.11 CSI Feedback

Explicit channel state information (CSI) feedback in IEEE~802.11 conveys \emph{transmit beamforming directions} by reporting quantized Givens rotation and phase angles that parametrize the right-singular subspace of the channel matrix. Because these angles encode fine-grained spatial signatures of the propagation environment, recent work have shown that plaintext CSI feedback can inadvertently reveal user activity, identity, and location to passive eavesdroppers. In this work, we introduce a standards-compatible \emph{differentially private (DP) quantization mechanism} that replaces deterministic angular quantization with an $\varepsilon$-DP stochastic quantizer applied directly to the Givens parameters of the transmit beamforming matrix. The mechanism preserves the 802.11 feedback structure, admits closed-form sensitivity bounds for the angular representation, and enables principled privacy calibration. Numerical simulations demonstrate strong privacy guarantees with minimal degradation in beamforming performance.

cs.IT↗

arXiv2026

Adversary-Aware Private Inference over Wireless Channels

AI-based sensing at wireless edge devices has the potential to significantly enhance Artificial Intelligence (AI) applications, particularly for vision and perception tasks such as in autonomous driving and environmental monitoring. AI systems rely both on efficient model learning and inference. In the inference phase, features extracted from sensing data are utilized for prediction tasks (e.g., classification or regression). In edge networks, sensors and model servers are often not co-located, which requires communication of features. As sensitive personal data can be reconstructed by an adversary, transformation of the features are required to reduce the risk of privacy violations. While differential privacy mechanisms provide a means of protecting finite datasets, protection of individual features has not been addressed. In this paper, we propose a novel framework for privacy-preserving AI-based sensing, where devices apply transformations of extracted features before transmission to a model server.

cs.IT↗

arXiv2025

Differentially Private Online Community Detection for Censored Block Models: Algorithms and Fundamental Limits

We study the private online change detection problem for dynamic communities, using a censored block model (CBM). We consider edge differential privacy (DP) in both local and central settings, and propose joint change detection and community estimation procedures for both scenarios. We seek to understand the fundamental tradeoffs between the privacy budget, detection delay, and exact community recovery of community labels. Further, we provide theoretical guarantees for the effectiveness of our proposed method by showing necessary and sufficient conditions for change detection and exact recovery under edge DP. Simulation and real data examples are provided to validate the proposed methods.

cs.SI↗

arXiv2025

Detecting Post-generation Edits to Watermarked LLM Outputs via Combinatorial Watermarking

Watermarking has become a key technique for proprietary language models, enabling the distinction between AI-generated and human-written text. However, in many real-world scenarios, LLM-generated content may undergo post-generation edits, such as human revisions or even spoofing attacks, making it critical to detect and localize such modifications. In this work, we introduce a new task: detecting post-generation edits locally made to watermarked LLM outputs. To this end, we propose a combinatorial pattern-based watermarking framework, which partitions the vocabulary into disjoint subsets and embeds the watermark by enforcing a deterministic combinatorial pattern over these subsets during generation. We accompany the combinatorial watermark with a global statistic that can be used to detect the watermark. Furthermore, we design lightweight local statistics to flag and localize potential edits. We introduce two task-specific evaluation metrics, Type-I error rate and detection accuracy, and evaluate our method on open-source LLMs across a variety of editing scenarios, demonstrating strong empirical performance in edit localization.

cs.LG↗

arXiv2025

Federated Learning-Enabled Hybrid Language Models for Communication-Efficient Token Transmission

Hybrid Language Models (HLMs) combine the low-latency efficiency of Small Language Models (SLMs) on edge devices with the high accuracy of Large Language Models (LLMs) on centralized servers. Unlike traditional end-to-end LLM inference, HLMs reduce latency and communication by invoking LLMs only when local SLM predictions are uncertain, i.e., when token-level confidence is low or entropy is high. However, ambiguous or low-confidence predictions still require frequent offloading to the LLM, leading to significant communication overhead in bandwidth-constrained settings. To address this, we propose FedHLM, a communication-efficient HLM framework that integrates uncertainty-aware inference with Federated Learning (FL). FedHLM's key innovation lies in collaboratively learning token-level uncertainty thresholds that govern when LLM assistance is needed. Rather than using static or manually tuned thresholds, FedHLM employs FL to optimize these thresholds in a privacy-preserving, distributed manner. Additionally, it leverages embedding-based token representations for Peer-to-Peer (P2P) resolution, enabling clients to reuse tokens inferred by semantically similar peers without engaging the LLM. We further introduce hierarchical model aggregation: edge servers refine local routing policies through client updates, while cross-cluster coordination aligns global decision boundaries. This layered design captures recurring uncertainty patterns, reducing redundant LLM queries. Experiments on large-scale news classification tasks show that FedHLM reduces LLM transmissions by over 95 percent with negligible accuracy loss, making it well-suited for scalable and efficient edge-AI applications.

cs.LG↗

arXiv2025

On the Price of Differential Privacy for Spectral Clustering over Stochastic Block Models

We investigate privacy-preserving spectral clustering for community detection within stochastic block models (SBMs). Specifically, we focus on edge differential privacy (DP) and propose private algorithms for community recovery. Our work explores the fundamental trade-offs between the privacy budget and the accurate recovery of community labels. Furthermore, we establish information-theoretic conditions that guarantee the accuracy of our methods, providing theoretical assurances for successful community recovery under edge DP.

cs.SI↗

arXiv2024

Deep Learning-Based Image Compression for Wireless Communications: Impacts on Reliability,Throughput, and Latency

In wireless communications, efficient image transmission must balance reliability, throughput, and latency, especially under dynamic channel conditions. This paper presents an adaptive and progressive pipeline for learned image compression (LIC)-based architectures tailored to such environments. We investigate two state-of-the-art learning-based models: the hyperprior model and Vector Quantized Generative Adversarial Network (VQGAN). The hyperprior model achieves superior compression performance through lossless compression in the bottleneck but is susceptible to bit errors, necessitating the use of error correction or retransmission mechanisms. In contrast, the VQGAN decoder demonstrates robust image reconstruction capabilities even in the absence of channel coding, enhancing reliability in challenging transmission scenarios. We propose progressive versions of both models, enabling partial image transmission and decoding under imperfect channel conditions. This progressive approach not only maintains image integrity under poor channel conditions but also significantly reduces latency by allowing immediate partial image availability. We evaluate our pipeline using the Kodak high-resolution image dataset under a Rayleigh fading wireless channel model simulating dynamic conditions. The results indicate that the progressive transmission framework enhances reliability and latency while maintaining or improving throughput compared to non-progressive counterparts across various Signal-to-Noise Ratio (SNR) levels. Specifically, the progressive-hyperprior model consistently outperforms others in latency metrics, particularly in the 99.9th percentile waiting time-a measure indicating the maximum waiting time experienced by 99.9% of transmission instances-across all SNRs, and achieves higher throughput in low SNR scenarios. where Adaptive WebP fails.

eess.SP↗

arXiv2024

Collaborative Inference over Wireless Channels with Feature Differential Privacy

Collaborative inference among multiple wireless edge devices has the potential to significantly enhance Artificial Intelligence (AI) applications, particularly for sensing and computer vision. This approach typically involves a three-stage process: a) data acquisition through sensing, b) feature extraction, and c) feature encoding for transmission. However, transmitting the extracted features poses a significant privacy risk, as sensitive personal data can be exposed during the process. To address this challenge, we propose a novel privacy-preserving collaborative inference mechanism, wherein each edge device in the network secures the privacy of extracted features before transmitting them to a central server for inference. Our approach is designed to achieve two primary objectives: 1) reducing communication overhead and 2) ensuring strict privacy guarantees during feature transmission, while maintaining effective inference performance. Additionally, we introduce an over-the-air pooling scheme specifically designed for classification tasks, which provides formal guarantees on the privacy of transmitted features and establishes a lower bound on classification accuracy.

cs.CR↗

arXiv2024

Privacy Preserving Semi-Decentralized Mean Estimation over Intermittently-Connected Networks

We consider the problem of privately estimating the mean of vectors distributed across different nodes of an unreliable wireless network, where communications between nodes can fail intermittently. We adopt a semi-decentralized setup, wherein to mitigate the impact of intermittently connected links, nodes can collaborate with their neighbors to compute a local consensus, which they relay to a central server. In such a setting, the communications between any pair of nodes must ensure that the privacy of the nodes is rigorously maintained to prevent unauthorized information leakage. We study the tradeoff between collaborative relaying and privacy leakage due to the data sharing among nodes and, subsequently, propose PriCER: Private Collaborative Estimation via Relaying -- a differentially private collaborative algorithm for mean estimation to optimize this tradeoff. The privacy guarantees of PriCER arise (i) implicitly, by exploiting the inherent stochasticity of the flaky network connections, and (ii) explicitly, by adding Gaussian perturbations to the estimates exchanged by the nodes. Local and central privacy guarantees are provided against eavesdroppers who can observe different signals, such as the communications amongst nodes during local consensus and (possibly multiple) transmissions from the relays to the central server. We substantiate our theoretical findings with numerical simulations. Our implementation is available at https://github.com/rajarshisaha95/private-collaborative-relaying.

eess.SP↗

arXiv2024

Over-the-Air Collaborative Inference with Feature Differential Privacy

Collaborative inference in next-generation networks can enhance Artificial Intelligence (AI) applications, including autonomous driving, personal identification, and activity classification. This method involves a three-stage process: a) data acquisition through sensing, b) feature extraction, and c) feature encoding for transmission. Transmission of the extracted features entails the potential risk of exposing sensitive personal data. To address this issue, in this work a new privacy-protecting collaborative inference mechanism is developed. Under this mechanism, each edge device in the network protects the privacy of extracted features before transmitting them to a central server for inference. This mechanism aims to achieve two main objectives while ensuring effective inference performance: 1) reducing communication overhead, and 2) maintaining strict privacy guarantees during features transmission.

cs.IT↗

arXiv2024

Clustering Mixtures of Discrete Distributions: A Note on Mitra's Algorithm

In this note, we provide a refined analysis of Mitra's algorithm \cite{mitra2008clustering} for classifying general discrete mixture distribution models. Built upon spectral clustering \cite{mcsherry2001spectral}, this algorithm offers compelling conditions for probability distributions. We enhance this analysis by tailoring the model to bipartite stochastic block models, resulting in more refined conditions. Compared to those derived in \cite{mitra2008clustering}, our improved separation conditions are obtained.

cs.LG↗