Clustering is a fundamental tool that has garnered significant interest across a wide range of applications including text analysis. To improve clustering accuracy, many researchers have proposed incorporating background knowledge, typically in the form of must‑link and cannot‑link constraints, to guide the clustering process. With the recent advent of large language models (LLMs), there is growing interest in improving clustering quality through LLM-based automatic constraint generation. In this paper, we propose a novel constraint‑generation approach that reduces resource consumption by generating constraint sets rather than using traditional pairwise constraints. This improves both query efficiency and constraint accuracy compared to state‑of‑the‑art methods. We further introduce a constrained clustering algorithm tailored to the characteristics of LLM-generated constraints. Our method incorporates a confidence threshold and a penalty mechanism to address potentially inaccurate constraints. We evaluate our approach on five text datasets, considering both the cost of constraint generation and overall clustering performance. The results show that our method achieves clustering accuracy comparable to the state-of-the-art algorithms while reducing the number of LLM queries by more than 20 times.

Segment Routing (SR) provides an effective approach to path control with minimal control-plane signaling. This makes SR a strong candidate for routing in the Satellite-Terrestrial Integrated Network (STIN), the core infrastructure enabling ubiquitous Internet of Things (IoT) connectivity. However, directly applying existing SR solutions to satellite networks presents significant challenges, which include limited bandwidth and constrained onboard processing capabilities, hindering efficient IoT data transmission. To enable reliable and efficient IoT services, we propose Satellite-Tailored Segment Routing (STSR), a novel framework designed specifically for satellite networks in STIN. STSR is built as a lightweight and Segment Routing over IPv6 (SRv6)-compatible extension. It deploys a customized data plane that enables efficient source routing through bit-based encoding and streamlined processing, which is vital for resource-constrained satellites. Furthermore, we develop a quality of service (QoS)-aware route compression scheme designed to meet diverse IoT service demands. This scheme leverages the computational resources of terrestrial controllers to generate compact, STSR-encoded paths. By accounting for QoS requirements and satellite-specific dynamics, the embedded algorithm enhances routing performance for heterogeneous IoT flows within the satellite network. Simulation and evaluation results demonstrate that STSR outperforms existing SRv6-based approaches in path encoding efficiency, payload transmission efficiency, processing overhead, and traffic engineering performance.

Deploying agentic AI services, such as large language models (LLMs) training and inference, presents significant challenges due to their complex, interdependent design across multiple layers of strategy space(framework, system, transport, network). Addressing diverse user intents with limited cross disciplinary expertise further exacerbates this complexity. To overcome these hurdles, we introduce a novel simulation-guided closed-loop (SGCL) strategy orchestration and optimization framework that is inherently intent-aware. Our approach leverages high-fidelity simulators to evaluate candidate deployment strategies, employs a surrogate-based Bayesian optimization engine to guide the closed-loop process, and incorporates a layer-wise caching mechanism to minimize redundant simulations and reduce evaluation overhead. We demonstrate its efficacy in distributed LLM training. Compared to baselines, our method consistently achieves higher intent satisfaction ratio and significantly boosts orchestration efficiency, notably reducing time-to-target by up to 94.4%. These findings highlight SGCL as a practical and effective solution for reliable, cost- and time efficient deployment of agentic AI services.

Multicast traffic engineering (TE) over Segment Routing (SR) is crucial for large-scale Low Earth Orbit (LEO) satellite networks. However, introducing multicast Source Routing TE to the LEO satellite network faces challenges due to constrained computational resources and the need for service adaptability. To address these issues, we propose the SatelliteTailored Service-Adaptive Multicast TE protocol, abbreviated as ST-SAM, which emphasizes efficient TE through a lightweight design and adaptability. ST-SAM introduces a protocol that partitions the network into two domains to provide depth and breadth control of the multicast service tree. Furthermore, it incorporates a dynamic domain boundary adjustment mechanism that optimizes domain placement based on satellite node density and multicast service characteristics. Simulations show that our proposed ST-SAM effectively reduces network overhead by up to 53.68% compared to existing solutions and achieves optimal lightweight performance under over 75 % of evaluated conditions, including various network scales and multicast service demands.

The excessively long Segment Identifier (SID) list in SRv6 challenges the promotion of SRv6 network slicing in cloudedge collaborative network. In this paper, we studied the problem of packet overhead during SRv6 SID processing in SRv6 network slicing services. We first design a new mechanism called Behavior Expression Slicing (BES) mechanism to establish the feasibility of dynamically virtual address allocation as well as SID compression based on adjacency information. Then we propose an algorithm to optimize compression efficiency in proposed BES mechanism. The simulation results indicate that the compression rate of the proposed mechanism and algorithm can reach over 75%.

Mobile immersive streaming applications, such as virtual reality (VR) and augmented reality (AR), impose stringent requirements on network transmission to deliver a seamless user experience. However, existing transmission control methods often fall short in dynamic mobile network environments. Traditional transport-layer protocols typically prioritize network-wide Quality of Service (QoS), overlooking the strict deadlines and varying priorities intrinsic to immersive streaming content, thereby adversely impacting application-specific Quality of Experience (QoE). Furthermore, heuristic-based congestion control methods lack adaptability to rapidly changing network environments, whereas machine learning (ML)-driven approaches, despite their adaptability potential, frequently prove computationally intensive and impractical for deployment on resource-constrained mobile devices. To address these drawbacks, we propose an intelligent transmission control approach that integrates a deadline- and priority-aware scheduler with a learning enhanced congestion control mechanism to optimize immersive streaming content delivery. The scheduling module dynamically prioritizes data packets based on their urgency and importance, timely delivering critical data within strict deadlines as much as possible. Complementing this, the hybrid congestion control module strategically combines lightweight heuristics for baseline performance efficiency with a selectively invoked ML model designed to adaptively adjust the sending rate, efficiently responding to real-time fluctuations in network conditions. Experimental results highlight the effectiveness of our approach, achieving QoE improvements of 15% to 45% compared to existing methods across a range of streaming applications and mobile network scenarios.

In recent years, with the rapid development of the Internet of Vehicles (IoV) and the widespread application of integrated sensing and communication (ISAC) in the IoV, the integrated sensing and computation offloading in vehicular edge networks has attracted widespread attention from academia and industry. Due to the generation of a large number of latency-sensitive tasks from vehicles’ real-time sensing of road conditions, coupled with the rise of other computing-intensive vehicle applications, the current computing capabilities of the onboard devices cannot meet the diverse demands of vehicular users. Therefore, it is necessary to combine the computing resources around the vehicle to complete computing tasks. With the goal of maximizing the difference between gain and consumption, this article studies the integrated sensing and computation offloading involving roadside units (RSUs) and vehicle platoon in vehicular edge networks. Specifically, we first consider that mobile vehicles can simultaneously offload computing tasks to the RSU and the vehicle platoon via nonorthogonal multiple access (NOMA) technology, and construct a multiobjective optimization problem with the goal of maximizing the utility of the three parties. Then, we construct a game model among the ISAC vehicle, the RSU, and the vehicle platoon based on the Stackelberg game. By seeking the equilibrium of the game, the optimal offloading and pricing strategy are derived, while the utility of the three parties is maximized. Finally, the simulation results show that the proposed scheme is superior to other traditional schemes, and each party in the game obtains its optimal strategy.

Segment Routing (SR) is of great significance in the evolving landscape of Space-Air-Ground Integrated Networks (SAGIN). However, owing to hardware limitations and bandwidth constraints, directly applying existing SR-related solutions of terrestrial networks to satellite networks faces resource and performance challenges. Thus, in this paper, we introduce a novel framework called Satellite-Tailored Segment Routing (STSR) for satellite networks in SAGIN. We first propose the STSR data-plane protocol to provide lightweight source routing. Then, a routing policy integration scheme is proposed to support routing consistency in SAGIN. Simulation results reveal that compared to SRv6, in the case of 2000 satellites, STSR reduces routing path information in the header by 43.51 % and increases load efficiency by 43%-92%.

In situ flow information telemetry (IFIT) is a state-of-the-art in-band telemetry framework for operator networks and can serve as the information foundation for network intelligence in the emerging sixth-generation regime. However, the performance advantage of IFIT comes at the cost of excessive telemetry data notification overhead, which makes it challenging to promote IFIT extensively. Therefore, we propose an approach named APN-integrated IFIT information notification (AIIN) to provide data notification overhead adaptability to IFIT. AIIN introduces a requirement-aware capability and reactive differentiated treatment into IFIT. In AIIN, we first enhance application-aware networking (APN) and integrate it into IFIT notification to support the explicit expression of data notification requirements. Then, oriented toward the different timeliness (telemetry data lag time) and accuracy (telemetry data retention rate) requirements expressed in APN, we design different behavioral treatment models to define reactive functions and procedures to make network devices explicitly process these requirements without decisions. The AIIN prototype is implemented on P4 switches. We also deploy the prototype on the China Environment for Network Innovation (CENI) network. Emulation results show that AIIN can achieve nanosecond line speed performance with differentiated and reactive data notification overhead reduction and, in the best case, can reduce bandwidth occupation by approximately 84%.

The emergency backup problem of optical inter-datacenter (inter-DC) networks is widely studied to avoid data loss caused by disasters. However, when unpredictable disasters such as earthquakes and large-scale power outages attack the optical inter-DC network, it is impossible to predict the network damage under a discrete time domain through the early warning system (EWS). Designing a reliable emergency backup scheme is becoming a key challenge in the optical inter-DC network in unpredictable disasters. Previous work only optimized emergency backup in the optical inter-DC networks in predictable disasters, which would lead to the unreliability of emergency backup of the optical inter-DC networks in unpredictable disasters. Therefore, this work proposes a reliable emergency backup method for the optical inter-DC networks in unpredictable disasters. Firstly, the disaster probability propagation model of the optical inter-DC network in unpredictable disasters is established based on the Markov process to quantify the network damage under a discrete time domain and install the emergency backup reliability problem. Next, constructing a time-sensitive variable time extension network (TS-VTEN) transforms the dynamic emergency backup reliability problem into a static flow problem. Finally, the problem is expressed as integer linear programming (ILP). The proposed ILP method outperforms state-of-the-art emergency backup methods by simultaneously achieving high reliability and time efficiency.

In this paper, we study the problem of compression efficiency of SRH (Segment Routing Header) in G-SRv6 (Generalized SRv6). We propose the Function-decoupled Segment Routing Mechanism (FDSRM) to optimize the SID list in SRH while ensuring that the routing policy decision would not be affected. FDSRM logically decouples the functions of SID/G-SID according to the function in routing decision and instruction indication. Based on FDSRM, we propose a mathematical optimization framework that leverages the LSTM neural network to optimize the SID allocation to adapt to future traffic. Simulation results indicate that FDSRM can improve the probability of SRH compression by 50.86%, and compress more than 24.50 bytes when the hop count is greater than 20.