arXiv subjects

Geoffrey Ye Li

Publications and source records attributed to Geoffrey Ye Li.

At least 19 recordsLinked to original sources

arXiv2026

SemISAC: Semantic Integrated Sensing and Communications

Conventional integrated sensing and communications (ISAC) systems primarily integrate communications and sensing through shared physical resources, without explicitly exploiting task-relevant semantic information. To move beyond such physical-level integration, we propose semantic ISAC (SemISAC), a general framework that unifies semantic communication (SemCom) and semantic sensing (SemS) to convey source meaning and acquire environmental meaning. Specifically, the transmitter combines source semantics and sensing task information with available side information to design the shared waveform and allocate radio resources, while the receiver-side communication and sensing task decoders recover the source meaning and infer the required environmental information, respectively. We also provide an information-theoretic interpretation to characterize the relationship between physical and task-relevant information and the resulting semantic trade-off in SemISAC. Building on this framework, we formulate the general SemISAC design problem and propose two realization methods, namely end-to-end (E2E) SemISAC optimization and modular SemISAC optimization. As a concrete realization, we apply modular SemISAC optimization to jointly design a learnable time-frequency (TF) precoder in an orthogonal frequency-division multiplexing (OFDM) system for representative SemCom and SemS tasks. Simulation results demonstrate that the proposed realization reduces sensing semantic distortion under a given communication requirement and achieves a more favorable communication-sensing trade-off than baseline designs.

eess.SP↗

arXiv2026

Token Encoding for Semantic Recovery

In generative semantic communication, semantic tokens guide receiver-side generative models to synthesize high-dimensional content. In challenging network environments, however, frequent token erasure distorts the conveyed semantics beyond what receiver-side recovery can restore. In this paper, we propose a token encoding framework (TokCode) for robust semantic recovery, achieving erasure resilience by restructuring redundancy in the semantic domain. TokCode uses a lightweight adapter to recast a general-purpose large language model (LLM) at the transmitter into a token encoder, exploiting the LLM's pretrained semantic prior to avoid introducing a dedicated deep model. To optimize the adapter efficiently and make it applicable across diverse channels, we develop a channel-quality-aware distillation approach for token encoder training~(CADET). Using a differentiable sentence-level semantic surrogate, CADET tunes T5 foundation models into experts for distinct erasure rates and distills them into a single reconfigurable low-rank adapter, enabling subsequent reinforcement learning (RL) to start above the plateau where direct RL stalls. Simulation results on token-based generative image transmission show that TokCode improves the image-level similarity over the best-performing receiver-side recovery benchmark by 14.1%--22.4%, closing 71.9%--76.5% of its gap to the erasure-aware oracle encoding, when only 20% to 50% of the tokens survive.

eess.SP↗

arXiv2026

Agentic UE-CoMIMO for 6G Terminals: From Virtual Antenna Augmentation to AI-Native Virtualization

End-user-centric collaborative MIMO (UE-CoMIMO) lets nearby devices form a virtual multi-antenna terminal to overcome the antenna limitations of individual user equipment. Extending such cooperation to communication, sensing, computing, and task-relevant information exchange requires a control layer that can interpret user intent, select cooperation mechanisms, and replan as conditions change. This article introduces Agentic UE-CoMIMO, in which device micro-agents, a smartphone or CPE hub agent, and edge/network agents coordinate device participation, relay modes, traffic splitting and duplication, compute placement, semantic-token exchange, and topology reconfiguration. Two system-level scenario studies on creator-centric live streaming and wearable-collaborative blind-spot sensing compare the proposed controller with capability-matched adaptive baselines. The results show that, by anticipating changes and preparing cooperation and fallback actions in advance, agentic control sustains high-quality streaming for longer and maintains blind-spot warnings through device outages. We also discuss the associated standardization, interoperability, trust, and validation challenges.

cs.IT↗

arXiv2026

AI/ML Life Cycle Management for Interoperable AI Native RAN

Artificial intelligence (AI) and machine learning (ML) are rapidly becoming integral to the 5G Radio Access Network (RAN), enabling beam management, channel state information (CSI) feedback, positioning, and mobility prediction. However, without a standardized life-cycle management (LCM) framework, challenges such as model drift, vendor lock-in, and limited transparency hinder large-scale deployment. 3GPP Releases 17--20 have progressively introduced AI/ML management and air-interface support, covering model training, validation, deployment, inference, data collection, performance monitoring, applicability assessment, and feature-specific control. Release 20 further extends these capabilities to two-sided CSI compression and inter-vendor model operation. This article reviews the resulting five-block LCM architecture, KPI-driven monitoring mechanisms, and inter-vendor collaboration schemes. We further propose an enhanced LCM framework with detailed interactions across functional blocks and an integrated procedure for reference-model and vendor-model development in two-sided operation, and identify open challenges in resource-efficient monitoring, environment drift detection, intelligent decision-making, and flexible model training. These developments provide a foundation for AI-native transceivers in 6G.

cs.IT↗

arXiv2026

Decentralized Federated Learning by Partial Message Exchange

Decentralized federated learning (DFL) has emerged as a transformative server-free paradigm that enables collaborative learning over large-scale heterogeneous networks. However, it continues to face fundamental challenges, including data heterogeneity, restrictive assumptions for theoretical analysis, and degraded convergence when standard communication- or privacyenhancing techniques are applied. To overcome these drawbacks, this paper develops a novel algorithm, PaME (DFL by Partial Message Exchange). The central principle is to allow only randomly selected sparse coordinates to be exchanged between two neighbor nodes. Consequently, PaME achieves substantial reductions in communication costs while still preserving a high level of privacy, without sacrificing accuracy. Moreover, grounded in rigorous analysis, the algorithm is shown to converge at a linear rate under the gradient to be locally Lipschitz continuous and the communication matrix to be doubly stochastic. These two mild assumptions not only dispense with many restrictive conditions commonly imposed by existing DFL methods but also enables PaME to effectively address data heterogeneity. Furthermore, comprehensive numerical experiments demonstrate its superior performance compared with several representative decentralized learning algorithms.

cs.LG↗

arXiv2026

Semantic Sensing: Toward a Task-Oriented Paradigm

Sensing and communication are fundamental enablers of next-generation networks. While communication technologies have advanced significantly, sensing remains limited to conventional parameter estimation and is far from fully explored. Motivated by these limitations, we propose semantic sensing (SemS), a novel framework that shifts the design objective from reconstruction fidelity to semantic effective recognition. Specifically, we mathematically formulate the interaction between transmit waveforms and semantic entities, thereby establishing SemS as a semantics-oriented transceiver design. Within this architecture, we leverage the information bottleneck (IB) principle as a theoretical criterion to derive a unified objective, guiding the sensing pipeline to maximize task-relevant information extraction. To practically solve this optimization problem, we develop a deep learning (DL)-based framework that jointly designs transmit waveform parameters and receiver representations. The framework is implemented in an orthogonal frequency division multiplexing (OFDM) system, featuring a shared semantic encoder that employs a Gumbel-Softmax-based pilot selector to discretely mask task-irrelevant resources. At the receiver, we design distinct decoding architectures tailored to specific sensing objectives, comprising a 2D residual network (ResNet)-based classifier for target recognition and a correlation-driven 1D regression network for high-precision delay estimation. Numerical results demonstrate that the proposed semantic pilot design achieves superior classification accuracy and ranging precision compared to reconstruction-based baselines, particularly under constrained resource budgets.

eess.SP↗

arXiv2026

Ten Years of Deep Learning for Wireless Communications: From Learned Blocks to Deployable Wireless Intelligence

Over the past decade, deep learning has evolved from a tool for replacing isolated wireless blocks into a broader methodology for developing wireless intelligence. This article traces that trajectory through three shifts: learning wireless functional modules, redesigning and re-normalizing communication goals, and enabling generalization under practical physical constraints. Together, these shifts advance the broader pursuit of communication anytime and anywhere, through any appropriate means. Early studies showed that neural networks could approximate difficult physical-layer inference and network-optimization mappings, while subsequent research embedded domain-specific structure, shifted toward task-oriented semantics, and addressed the need for edge-efficient adaptation. Looking ahead, we argue that the next era of wireless artificial intelligence (AI) depends on more than scaling model capacity. Promising directions include physically grounded wireless world models, agentic reasoning and fulfillment, and standardization mechanisms that allow learned components to operate with clear boundaries, physical consistency, and system-level interoperability.

eess.SP↗

arXiv2026

Hierarchical Wireless Foundation Model for Multi-Task Optimization

The increasing complexity of next-generation wireless networks has driven the integration of artificial intelligence (AI) into wireless communications. However, most existing studies focus on developing task-specific deep learning techniques for single scenarios, which limits their ability to generalize across diverse tasks, channel conditions, and system configurations. To address this generalization bottleneck, we propose a hierarchical wireless foundation model (WFM) for multi-task optimization. The proposed WFM couples an upstream foundation channel encoder (FCE) with a downstream foundation optimization decoder (FOD) via geometry-aware cross-attention. Specifically, the FCE extracts task-agnostic channel representations via self-supervised masked reconstruction while the FOD generates multi-task optimization decisions through differentiable output heads. Moreover, a hybrid supervised-to-unsupervised training strategy is employed to overcome the performance ceiling of purely supervised learning, and the modular architecture of the WFM enables efficient adaptation to unseen communication tasks with minimal parameter overhead. Simulation results show that the proposed WFM learns high-fidelity channel representations and achieves competitive multi-task optimization performance while substantially reducing optimization inference latency relative to numerical baselines. Furthermore, it exhibits robust generalization to unseen propagation environments, varying constraint parameters, and heterogeneous system configurations.

eess.SP↗

arXiv2026

Preconditioned Inexact Stochastic ADMM for Deep Model

Deep learning models are usually trained with stochastic gradient descent-based algorithms, but these optimizers face inherent limitations, such as slow convergence and stringent assumptions for convergence. In particular, data heterogeneity arising from distributed settings poses significant challenges to their theoretical and numerical performance. This paper develops an algorithm, PISA (Preconditioned Inexact Stochastic Alternating Direction Method of Multipliers). Grounded in rigorous theoretical guarantees, the algorithm converges under the sole assumption of Lipschitz continuity of the gradient on a bounded region, thereby removing the need for other conditions commonly imposed by stochastic methods. This capability enables the proposed algorithm to tackle the challenge of data heterogeneity effectively. Moreover, the algorithmic architecture enables scalable parallel computing and supports various preconditions, such as second-order information, second moment, and orthogonalized momentum by Newton-Schulz iterations. Incorporating the latter two preconditions in PISA yields two computationally efficient variants: SISA and NSISA. Comprehensive experimental evaluations for training or fine-tuning diverse deep models, including vision models, large language models, reinforcement learning models, generative adversarial networks, and recurrent neural networks, demonstrate superior numerical performance of SISA and NSISA compared to various state-of-the-art optimizers.

cs.LG↗

arXiv2026

Twin-in-the-Loop Optimization and Fundamental Limits of Position--Velocity Estimation in Cell-Free ISAC Systems

Digital twin (DT) networks require tight integration with wireless sensing, yet the fundamental limits of such coupling in cell-free integrated sensing and communication (ISAC) systems remain largely unexplored, particularly in the presence of fluid intelligent metasurfaces (FIM). This paper establishes a joint position-velocity Cramer-Rao bound (CRB) framework, operationalized through a twin-in-the-loop architecture. By leveraging a scatter-matrix decomposition of the velocity Fisher information, we show that single-base-station systems are inherently rank-deficient for two-dimensional velocity estimation, whereas cell-free deployments with multiple access-point pairs achieve full observability. The resulting CRB reveals a spatio-temporal decoupling: FIM shape optimization significantly improves position accuracy but does not affect the velocity CRB under isotropic waveforms while Doppler coupling asymmetrically enhances position estimation accuracy. Building on this analysis, we develop a closed-loop DT framework, deriving the critical mismatch angle in closed form and showing that angular diversity in cell-free systems mitigates DT prediction errors. We further characterize the optimal synchronization period and propose a confidence-aware scheduling strategy that reduces the DT update rate. Numerical results demonstrate substantial performance gains over single-base-station systems, with improvements attributed to angular diversity, Doppler-position coupling, and FIM adaptation.

cs.IT↗

arXiv2026

Foundation Models for Wireless Communications: From PHY Intelligence to Network Autonomy

6G networks will introduce unprecedented complexity, which calls for a paradigm shift in network optimization and management. Artificial intelligence (AI)-based solutions, especially those enabled by the recently developed foundation models, have been recognized as promising candidates. Foundation models are large-scale AI models with general-purpose feature extraction capabilities, and once trained on massive amounts of data, they can be adapted to solve a wide range of downstream tasks, either in a zero-shot manner or with few-shot fine-tuning. This article provides a comprehensive overview of how foundation models are reshaping physical-layer processing and wireless resource management across three progressive paradigms. First, we examine the adaptation of off-the-shelf pre-trained foundation models to various wireless tasks. Second, we explore wireless-native foundation models, built from scratch on wireless data to bridge cross-domain modality gaps and capture universal wireless-domain physical characteristics. Third, we highlight agentic foundation models, which elevate static data processing into autonomous, reasoning-driven network orchestration. Furthermore, we discuss the impact of applying foundation models to emerging 6G frontiers, including integrated sensing and communications (ISAC), new multiple-input multiple-output (MIMO) architectures, semantic communications, and system-level network autonomy. Finally, we identify critical open challenges and opportunities, charting a promising path toward fully intelligent and adaptive wireless networks.

eess.SP↗

arXiv2026

Massive MIMO CSI Feedback with Spiking Neural Networks

Deep learning-based channel state information (CSI) feedback has achieved empirical success in massive multiple-input multiple-output (MIMO) systems. However, existing approaches largely rely on dense artificial neural networks (ANNs), whose computational overhead limits their practical applications. In this article, we exploit bio-inspired spiking neural networks (SNNs) for massive MIMO CSI feedback, referred to as SpikingCSINet, where both the feedback and the main network computations are implemented through spikes. To overcome the information bottleneck of binary spikes in high-dimensional reconstruction, we develop a progressive residual (PR) architecture that exploits the natural temporal dimension of SNNs, encoding successive residuals across time steps to enhance information compactness. Experiments on the COST 2100 benchmark show that SpikingCSINet attains a more favorable performance-efficiency tradeoff than lightweight convolutional baselines. Moreover, it achieves performance competitive with Transformer-based feedback while reducing energy consumption by over $93\%$.

eess.SP↗

arXiv2026

Enabling High Error Tolerance in Satellite Video Transmissions by Generative Semantic Communication

Low Earth orbit (LEO) satellite relays will significantly extend the coverage of mobile networks, enabling users in remote areas to transmit data of real-time events. Nevertheless, the limited power of user devices and the long distance to satellites lead to low signal-to-noise ratio (SNR), which results in high error rates and frequent retransmissions, severely hindering the transmissions of high-dimensional data such as videos. In this paper, we propose a novel method to achieve high error tolerance in satellite-relay video transmissions using generative semantic communications (GSC). For the transmitter, we design and optimize a semantic encoder integrating a pre-trained video encoder with a low-density parity-check (LDPC) encoder, efficiently achieving generalizability and enabling forward error correction. For the receiver, we fine-tune a generative video model using an efficient in-context adaptation algorithm, enabling it to reconstruct videos from error-corrupted semantic information. Simulation results show that our method achieves 2.5 dB higher video peak SNR than conventional semantic communications at an error rate of 45%, and remains robust when the error rate exceeds 80%.

eess.SP↗

arXiv2026

0/1 Constrained Optimization Solving Sample Average Approximation for Chance Constrained Programming

Sample average approximation (SAA) is a tractable approach for dealing with chance constrained programming, a challenging stochastic optimization problem. The constraint of SAA is characterized by the $0/1$ loss function which results in considerable complexities in devising numerical algorithms. Most existing methods have been devised based on reformulations of SAA, such as binary integer programming or relaxed problems. However, the development of viable methods to directly tackle SAA remains elusive, let alone providing theoretical guarantees. In this paper, we investigate a general $0/1$ constrained optimization, providing a new way to address SAA rather than its reformulations. Specifically, starting with deriving the Bouligand tangent and Fr$\acute{e}$chet normal cones of the $0/1$ constraint, we establish several optimality conditions. One of them can be equivalently expressed by a system of equations, enabling the development of a semismooth Newton-type algorithm. The algorithm demonstrates a locally superlinear or quadratic convergence rate under standard assumptions, along with nice numerical performance compared to several leading solvers.

math.OC↗

arXiv2026

Beam Prediction Based on Multimodal Large Language Models

Accurate beam prediction is a key enabler for next-generation wireless communication systems. In this paper, we propose a multimodal large language model (LLM)-based beam prediction framework that effectively utilizes contextual information, provided by sensory data including RGB camera images and LiDAR point clouds. To effectively fuse heterogeneous modalities, we design specialized modality encoders together with a beam-guided attention masking mechanism and a high-frequency temporal alignment strategy, enabling robust cross-modal feature integration under dynamic environments. Furthermore, we construct a large-scale multimodal dataset for communication, named Multimodal-Wireless, which covers diverse weather and traffic conditions with high-fidelity ray-tracing labels. Extensive simulation results demonstrate that the proposed approach significantly reduces the reliance on oracle angle-of-departure knowledge and consistently outperforms state-of-the-art multimodal LLM-based beam prediction methods in terms of beam accuracy and communication performance, improving the average Top-1 accuracy to 80.8% and the average normalized gain to 89.1%.

eess.SP↗

arXiv2026

Large Language Models for Wireless Communications: From Adaptation to Autonomy

The emergence of large language models (LLMs) has revolutionized artificial intelligence, offering unprecedented capabilities in reasoning, generalization, and zero-shot learning. These strengths open new frontiers in wireless communications, where increasing complexity and dynamics demand intelligent and adaptive solutions. This article explores the role of LLMs in transforming wireless systems across three key directions: adapting pretrained LLMs for communication tasks, developing wireless-specific foundation models to balance versatility and efficiency, and enabling agentic LLMs with autonomous reasoning and coordination capabilities. We highlight recent advances, practical case studies, and the unique benefits of LLM-based approaches over traditional methods. Finally, we outline open challenges and research opportunities, including multimodal fusion, collaboration with lightweight models, and self-improving capabilities, charting a path toward intelligent, adaptive, and autonomous wireless networks.

cs.AI↗

arXiv2026

Event-Triggered Gossip for Distributed Learning

While distributed learning offers a new learning paradigm for distributed network with no central coordination, it is constrained by communication bottleneck between nodes. We develop a new event-triggered gossip framework for distributed learning to reduce inter-node communication overhead. The framework introduces an adaptive communication control mechanism that enables each node to autonomously decide in a fully decentralized fashion when to exchange model information with its neighbors based on local model deviations. We analyze the ergodic convergence of the proposed framework under noconvex objectives and interpret the convergence guarantees under different triggering conditions. Simulation results show that the proposed framework achieves substantially lower communication overhead than the state-of-the-art distributed learning methods, reducing cumulative point-to-point transmissions by \textbf{71.61\%} with only a marginal performance loss, compared with the conventional full-communication baseline.

eess.SP↗

arXiv2026

Multimodal-Wireless: A Large-Scale Dataset for Sensing and Communication

This paper presents Multimodal-Wireless, a large-scale open-source dataset for multimodal sensing and communication research. The dataset is generated through an integrated and customizable data pipeline built upon the CARLA simulator and Sionna framework, and features high-resolution communication channel state information (CSI) fully synchronized with five other sensor modalities, namely LiDAR, RGB and depth camera, inertial measurement unit (IMU) and radar, all sampled at 100 Hz. It contains approximately 160,000 frames collected across four virtual towns, sixteen communication scenarios, and three weather conditions. This paper provides a comprehensive overview of the dataset, outlining its key features, overall framework, and technical implementation details. In addition, it explores potential research applications concerning communication and collaborative perception, exemplified by beam prediction using a multimodal large language model. The dataset is open in https://le-liang.github.io/mmw/.

eess.SP↗