The growing presence of mobile and IoT devices has led to massive decentralized and evolving data, driving the rise of Federated Learning (FL) to enable collaborative training without data sharing. However, traditional FL assumes static data distributions, which is unrealistic for dynamic real-world environments. To address this challenge, Federated Class-Incremental Learning (FCIL) has emerged as a promising framework that enables flexible adaptation to newly introduced classes over time. Existing FCIL methods typically integrate old knowledge preservation into local client training. However, these methods cannot avoid spatial-temporal client drift caused by data heterogeneity and often incur significant computational and communication overhead, limiting practical deployment. To address these challenges simultaneously, we propose a novel approach, Spatial-Temporal Statistics Aggregation (STSA), which provides a unified framework to aggregate feature statistics both spatially (across clients) and temporally (across stages). The aggregated feature statistics are unaffected by data heterogeneity and can be used to update the classifier in closed form at each stage. Additionally, we introduce STSA-E, a communication-efficient variant that enables the server to approximate global second-order feature statistics using first-order statistics uploaded from clients. Theoretical analysis shows that it achieves similar performance to STSA with much lower communication overhead. Extensive experiments on three widely used FCIL datasets, with varying degrees of data heterogeneity, show that our method outperforms state-of-the-art FCIL methods in terms of performance, flexibility, and both communication and computation efficiency. The code is available at https://github.com/Yuqin-G/STSA.
论文检索
输入标题、作者或关键词，从 954 篇学术成果中精准定位
Federated learning is a distributed collaborative training framework that enables multiple clients to share model updates and jointly train deep neural networks without exchanging raw data. Although extensive research has explored data augmentation techniques in federated settings, the naturally non-IID data distributions among clients render blind augmentation prone to severe degradation of the learned model. To solve this problem, we suggest Responsive Augmentation and Approximate Update Method for Fed erated Learning with Non-IID Data (RAFed), aimed at alleviating feature shift in client samples. We leverage a Responsive Augmentation Method to accumulate shared data augmentation policy knowledge through local learning, guiding the policy gradient to consider the impact of data augmentation on unseen local data, and employ an Approximate Update Mechanism to reduce communication costs and achieve efficient policy search. To improve the adaptability of data augmentation policies to local data distributions, we introduce a Dynamic Adaptive Method for searching personalized augmentation policies tailored to heterogeneous clients. Experiments on four popular datasets show that RAFed achieves superior test accuracy and lower communication costs compared to related baselines while providing privacy advantages. The code is available via https://github.com/anonymously123-stcak/RAFed.
In the era of large language models (LLMs), weight-activation quantization helps fit models on edge device by reducing memory and compute bit-widths. However, three challenges persist for energy constrained hardware: (1) even after quantization, multiply-accumulate (MAC) operations remain unavoidable and continue to dominate energy consumption; (2) dequantization (or per-tensor/ channel rescaling) introduces extra arithmetic and data movement, increasing latency and energy; (3) uniform parameters bit widths clip salient values—while intra-channel mixed precision is generally impractical on current matrix hardware and memory. In contrast, brain-inspired Spiking Neural Networks (SNNs), owing to their binary spike-based information representation and the Integrate-and-Fire (IF) paradigm, naturally support mixed-precision storage and energy-efficient computation by replacing complex MACs with temporal Accumulate (ACCs). Motivated by this property, we propose SpikeQuant, which selectively applies mixed-precision quantization to activations with salient values and re-encodes them into binary spike counts, thereby enabling dynamic mixed storage of different bitwidths. Furthermore, by embedding the quantization scale into the threshold of the IF mechanism, our approach performs energy-efficient linear transformations on weights and activations while avoiding explicit dequantization. Experimental results demonstrate that SpikeQuant consistently achieves near-FP16 perplexity under W4A4 quantization while reducing energy cost by up to 4.6× compared to existing methods, highlighting its effectiveness for accurate and energy-efficient LLM deployment. Our code is open-sourced at https://github.com/wangchenyu929/SpikeQuant.git
Accurate multivariate time series forecasting (MTSF) is critical for intelligent web services in Web of Things. When confronted with unseen multivariate time series (MTS), the industry typically invests significant time and resources in training multiple models to identify the optimal model for deployment. This paper proposes a novel, efficient, and scalable MTSF model selection method that directly selects suitable MTSF methods based on data characteristics without extensive model training. Model selection is a core component of AutoML, which has made significant progress in recent years. However, existing methods incur high operational costs and cannot be directly applied to MTSF tasks. Moreover, there is a lack of a comprehensive and cohesive public time series library for MTSF model selection. To address these challenges, we compile the first large heterogeneous labeled MTSF model selection dataset, called the ModelPile, which covers 41 mainstream datasets across 11 domains. We then propose AutoMTSF, a large model-enabled model selection method that transforms the MTSF model selection problem into a time series classification problem and utilizes the ModelPile to unlock large-scale multi-dataset training. AutoMTSF first uses the pre-trained large model to encode raw MTS. Given the coarse-grained limitations of large model encoding, Recursive Temporal Pattern Feature (RTPF) is proposed to capture both fine-grained and global temporal feature evolution, thereby effectively mapping data characteristics to the MTSF method space. Experiments comparing AutoMTSF with 2 baselines, 17 MTSF methods, and 4 large time series models show that AutoMTSF outperforms state-of-the-art methods while maintaining comparable execution time. This work represents a critical step in validating the accuracy and efficiency of large model-enabled classification for MTSF.
In real-world web applications, especially those involving sensor networks and Internet of Things (IoT) devices, time series data are often incomplete due to network delays, device failures or logging constraints. Such missing data can severely affect downstream tasks including anomaly detection, recommendation, and A/B testing, making imputation a critical step for reliable web analytics. Diffusion models have recently achieved strong performance for time series imputation. As relevant research progresses, the spectral nature of time series has received increasing attention. However, most ''frequency-aware'' diffusion variants modify either the input or network architecture, but the variance schedule in the forward process remains unchanged, injecting noise with the same variance into every frequency bin. This limitation prevents diffusion from adapting to real data, where spectral energy varies irregularly across frequencies rather than following a simple high–low split. To address these issues, we propose Frequency-Shaped Diffusion (FSDI), which replaces the uniform variance schedule with a data-driven schedule in the frequency domain. Frequency bin variances are estimated from the spectral energy distribution of the data, allocated as inverses of that energy, and then Parseval-calibrated so the total noise energy exactly matches standard diffusion, preserving training stability and ensuring fair comparison. Experiments on real-world datasets demonstrate that FSDI achieves state-of-the-art performance. All code have been made publicly at https://github.com/decisionintelligence/FSDI.
Federated domain generalization (FDG) for time-series classification (TSC) poses a critical challenge for modern intelligent web services, which rely on edge-collected time-series signals from diverse mobile applications and web devices (e.g., wearables sensors) to support decision-making. The source heterogeneity and temporal dynamics give rise to out-of-distribution (OOD) patterns, which hinder the model's ability to generalize to previously unseen users and devices. In this work, we propose Federated Generalization via Diversity Generation (FedDiG), a diffusion-based FDG framework that captures intra-client distribution shifts from a frequency-domain perspective and employs cross-frequency sampling to synthesize time-series data with diverse spectral patterns. Specifically, FedDiG first performs frequency-proxy representation learning on clients to serve as diffusion conditions. The server then aggregates client-side frequency proxies to construct a global proxy pool and applies class-wise mixup to create novel frequency features. These features guide a global diffusion model to produce diverse data, enabling the simulation of previously unseen patterns and thereby enhancing model training. Extensive experiments on four cross-domain time-series benchmarks demonstrate that FedDiG significantly outperforms state-of-the-art federated learning and FDG baselines, particularly under small-data regimes and large-scale client scenarios, achieving robust generalization to unseen domains in federated settings. This work bridges distribution-diversity synthesis and FDG for time-series to support robust, scalable web applications fed by edge-collected signals, delivering web-scale generalization across heterogeneous web, mobile, and IoT clients.
Web service administrators must ensure the stability of multiple systems by promptly detecting anomalies in Key Performance Indicators (KPIs). Achieving the goal of ''train once, infer across scenarios'' remains a fundamental challenge for time series anomaly detection models. Beyond improving zero-shot generalization, such models must also flexibly handle sequences of varying lengths during inference—ranging from one hour to one week—without retraining. Conventional approaches rely on sliding-window encoding and self-supervised learning, which restrict inference to fixed length inputs. Large Language Models (LLMs) have demonstrated remarkable zero-shot capabilities across general domains. However, when applied to time series data, they face inherent limitations due to context length. To address this issue, we propose ViTs, a Vision-Language Model (VLM)-based framework that converts time series curves into visual representations. By rescaling time series images, temporal dependencies are preserved while maintaining a consistent input size, thereby enabling efficient processing of arbitrarily long sequences without context constraints. Training VLMs for this purpose introduces unique challenges, primarily due to the scarcity of aligned time series image–text data. To overcome this, we employ an evolutionary algorithm to automatically generate thousands of high-quality image–text pairs and design a three-stage training pipeline consisting of: (1) time series knowledge injection, (2) anomaly detection enhancement, and (3) anomaly reasoning refinement. Extensive experiments demonstrate that ViTs substantially enhance the ability of VLMs to understand and detect anomalies in time series data. All datasets and code will be publicly released at: https://github.com/Uni-WangZexin/ViTs.
Personalized Federated Learning (pFL) effectively addresses the challenge of statistical heterogeneity in traditional Federated Learning (FL), with feature alignment methods (e.g., FedProto) standing out due to their communication efficiency and model-agnostic design, making them practically viable in real-world non-IID scenarios. These methods directly align class-level features across clients without requiring model parameter transmission. However, they represent each class as a holistic prototype, which limits the diversity and expressiveness of shared features. This restriction hampers the model's ability to generalize across clients and impedes personalized adaptation, as clients lack sufficient semantic components to reconstruct discriminative features tailored to their local data distributions. To overcome these limitations, we propose Federated Visual Primitive Learning (FedVPL), a novel framework comprising two key components: (1) Visual Primitive Space Sharing, which decomposes class-level features into semantically meaningful and reusable visual primitives, enabling cross-client and cross-class sharing to enrich feature diversity and decouple communication cost from the number of classes, significantly improving efficiency; and (2) Text-Guided Semantic Alignment, a parameter-free personalization mechanism that leverages external language priors to align shared primitives with client-specific semantics, without requiring additional communication overhead. Extensive experiments across diverse non-IID benchmarks demonstrate that FedVPL substantially outperforms state-of-the-art baselines, achieving up to a 5.62% improvement in accuracy, reducing communication overhead by at least 12.5x, and effectively addressing generalization and personalization challenges in heterogeneous federated environments.
A reliable autonomous driving system requires a high-precision perception module. Collaborative perception is emerging as a web-scale information-sharing paradigm for autonomous driving, enabling multiple vehicles to collectively achieve a broader perception field than any single vehicle. However, existing approaches necessitate frequent one-to-many communication, which increases network load and leads to information redundancy. This paper presents Octopus, an innovative vehicle-to-road collaboration framework that leverages the computational capabilities of roadside units. Instead of frequent one-to-many communication, vehicles interact only with roadside units, which significantly reduces communication overhead and improves real-time processing efficiency. While this design alleviates communication burdens, vehicles may still struggle to achieve comprehensive situational awareness in highly dynamic environments. To further address this limitation, our framework incorporates global fusion results as prior knowledge, enabling closed-loop fusion to refine vehicle-side perception. Extensive experiments on OPV2V and V2V4Real datasets demonstrate that Octopus excels at collaborative perception, outperforming the state-of-the-art approach up to 11.58% on AP@0.7, 12.74% on R11@0.7 and 5514× reduction in communication volume.
Disaggregated Memory Systems (DMS) hold substantial potential for cloud datacenters but face critical deployment barriers in multi-tenant RDMA environments. Existing DMS designs rely on idealized assumptions-overlooking interference from co-located RDMA applications, oversimplifying fabric topology considerations, and lacking elastic service-level objectives (SLOs) guarantees-resulting in performance degradation and resource inefficiency. To bridge these gaps, we introduce Wiseswap, an elastic, datacenter network-aware DMS that delivers robust memory disaggregation for multi-tenant clouds through three key innovations: (1) Preemption-enabled isolation: A low-overhead kernel mechanism utilizes WAIT/ENABLE RDMA primitives to prioritize latency-critical swap operations over user-space RDMA flows, maintaining tenant fairness; (2) Adaptive fabric path selection: In-kernel telemetry dynamically probes latency and routes memory traffic through uncongested paths, mitigating interference from elephant flows; (3) Feedback-directed autoscaling: Fine-grained optimization of DMS-specific parameters-dynamically optimizes resource allocation under fluctuating workloads, guaranteeing stringent SLOs while minimizing resource overhead. Evaluations demonstrate that Wiseswap improves throughput by 1.3-2.4x and reduces tail latency by 57-73% under contention compared to state-of-the-art solutions, while consistently meeting strict SLO targets.
Large Language Models (LLMs) increasingly underpin intelligent web applications, from chatbots to search and recommendation, where efficient specialization is essential. Low-Rank Adaptation (LoRA) enables such adaptation with minimal overhead, while federated LoRA allows web service providers to fine-tune shared models without data sharing. However, in privacy-sensitive deployments, clients inject varying levels of differential privacy (DP) noise, creating privacy heterogeneity that misaligns individual incentives and global performance. In this paper, we propose WinFLoRA, a privacy-heterogeneous federated LoRA that utilizes aggregation weights as incentives with noise awareness. Specifically, the noises from clients are estimated based on the uploaded LoRA adapters. A larger weight indicates greater influence on the global model and better downstream task performance, rewarding lower-noise contributions. By up-weighting low-noise updates, WinFLoRA improves global accuracy while accommodating clients' heterogeneous privacy requirements. Consequently, WinFLoRA aligns heterogeneous client utility in terms of privacy and downstream performance with global model objectives without third-party involvement. Extensive evaluations demonstrate that across multiple LLMs and datasets, WinFLoRA achieves up to 52.58% higher global accuracy and up to 2.56× client utility than state-of-the-art benchmarks. Source code is publicly available at https://github.com/koums24/WinFLoRA.git.
Collaborative learning between edge servers (e.g., base stations) and end devices (e.g., drones) enables simultaneous model training in web applications through knowledge sharing. The resulting models effectively reduce service latency. However, existing approaches either assume isomorphic models on edge servers and end devices or incur substantial transmission overhead when training. Moreover, edge servers are often unable to access data from end devices on time due to long-distance constraints or strict data privacy regulations. This paper proposes a Prototype Augmentation-based Edge-end Collaborative Learning method (PAECL). It simultaneously trains heterogeneous edge and end models in the absence of data on edge servers by transmitting only augmented class-wise feature vectors (prototypes), significantly reducing communication overhead compared to sharing models, data, or logits. Specifically, on end devices, prototype-implied latent knowledge is augmented via local prototype contrast and global prototype alignment. On edge servers, prototypes are further augmented to produce bounded virtual vectors by mixing them with random noise, and the augmented prototypes are then delivered to generative models to provide data during edge model training. Through simulations and field experiments, PAECL achieves the highest accuracy for edge and end models under limited training resources and reduces the transmission burden by at least 297 times compared to existing edge-end heterogeneous learning methods.
Mobile real-time video streaming (RTVS) demands ultra-low latency to preserve content timeliness. Packet loss in mobile networks significantly inflates frame latency and thus degrades the quality of experience (QoE). As a promising solution, Forward Error Correction (FEC) encoding has been widely deployed in RTVS systems to recover from packet loss by introducing redundancy. However, existing schemes focus on per-frame FEC protection, failing to optimize QoE because they cannot precisely allocate redundancy to handle burst loss events. These events typically occur at the single-frame level, but can be smoothed out at the multi-frame level. We propose Breath, an adaptive FEC scheme that dynamically adjusts the protection boundary based on network and video dynamics. We have implemented Breath in a RTVS system and evaluated it in emulated mobile networks using network traces collected from the production system. Results show that, compared to state-of-the-art FEC schemes, Breath reduces deadline missing rate by 17.2%-22.5% while improving the average video bitrate by 10.6%-14.2%.
The rapid growth of web content has spurred the widespread adoption of on-device AI assistants powered by large language models (LLMs). However, deploying and personalizing these assistants in real-world environments remains challenging due to limited annotation budgets and scarce on-device fine-tuning resources. Existing edge–cloud collaboration frameworks typically rely on costly cloud-based supervision or perform full-layer finetuning, leading to inefficiencies in both computation and adaptation. To address these limitations, we propose EcoTune, a budget-constrained framework for efficient edge–cloud collaborative adaptation. EcoTune jointly optimizes representative data selection for cloud annotation and selective on-device model adaptation within a unified closed-loop process. Specifically, it employs a multi-armed bandit–based strategy to identify highvalue user interactions for cloud supervision and a layer importance–driven adaptation mechanism to update only critical components of the small language model (SLM). This coordinated optimization enables dynamic, resource-efficient personalization under stringent annotation and tuning budgets. Experiments on real-world testbeds demonstrate that EcoTune achieves up to 20%-60% reduction in annotation costs and significantly lowers fine-tuning memory consumption compared to state-of-the-art baselines, providing a practical and scalable solution for personalized on-device LLMs.
Mobile agents have made progress toward reliable smartphone automation, yet performance in complex applications remains limited by incomplete knowledge and weak generalization to unseen environments. We introduce a curiosity driven knowledge retrieval framework that formalizes uncertainty during execution as a curiosity score. When this score exceeds a threshold, the system retrieves external information from documentation, code repositories, and historical trajectories. Retrieved content is organized into structured AppCards, which encode functional semantics, parameter conventions, interface mappings, and interaction patterns. During execution, an enhanced agent selectively integrates relevant AppCards into its reasoning process, thereby compensating for knowledge blind spots and improving planning reliability. Evaluation on the AndroidWorld benchmark shows consistent improvements across backbones, with an average gain of six percentage points and a new state of the art success rate of 88.8% when combined with GPT-5. Analysis indicates that AppCards are particularly effective for multi step and cross application tasks, while improvements depend on the backbone model. Case studies further confirm that AppCards reduce ambiguity, shorten exploration, and support stable execution trajectories. Task trajectories are publicly available at https://lisalsj.github.io/Droidrun-appcard/.
Modern Large Language Model (LLM) training clusters increasingly mix heterogeneous GPUs, diverse intra-node fabrics, and inter-node interconnects, combined with varied parallelism strategies. Exploring this massive design space, further amplified by heterogeneity, through real deployments is prohibitively slow and costly. Existing simulators, which are primarily designed and tuned for homogeneous clusters, either trade fidelity for speed or require heavyweight workflows with non-negligible overhead. We propose HeteroSim, a high-fidelity simulation framework for heterogeneous LLM training systems. It introduces: (i) a LLM training workload compiler that captures realistic training graphs, microbatching schedules, and compute-communication overlap; (ii) a heterogeneity-aware computation planner using roofline-style scaling across GPU generations; (iii) a collective communication planner that reproduces NCCL-like behaviors with per-link models, message channelization, and configurable routing. Across a wide range of heterogeneity levels, experimental results show that HeteroSim achieves near-real simulation accuracy while keeping low overhead.
Cross-view geo-localization (CVGL) establishes correspondences between ground-level and satellite images of the same geographic location, serving as a fundamental technology for smart city applications, including autonomous navigation, urban planning, and location-based services. Current CVGL approaches fall into two categories: feature-based methods achieve superior performance through 2D representation learning but lack interpretability. Spatial-based methods provide geometric understanding and interpretable matching but suffer from limited spatial modeling and weak cross-view alignment, leading to lower performance. We reformulate CVGL from a spatial perspective and propose an auxiliary task-enhanced network. The network captures spatial semantics and provides explicit alignment processes with visualizable results. We introduce an auxiliary spatial semantic alignment (SSA) task that learns spatial structure via vision foundation models (VFM) and BEV transformation to enhance the primary CVGL task. The primary task captures visual semantics, including texture and appearance. Within this unified framework with shared encoders, the primary task enriches the learned embeddings by fusing spatial structure with visual semantics, yielding spatially complete representations. Extensive experiments on three standard CVGL benchmarks demonstrate that our method significantly surpasses previous spatial-based approaches while maintaining competitive performance with state-of-the-art (SOTA) feature-based methods, achieving 98.48% R@1 on CVUSA and 71.05% R@1 on CVACT\_test. We provide comprehensive analyses through pixel-level activation maps and feature-space UMAP visualizations to validate both effectiveness and interpretability.
Low Earth Orbit (LEO) satellite ISPs promise universal Internet connectivity, yet their interaction with content delivery remains poorly understood. We present the first comprehensive measurement study decomposing Starlink's web content delivery performance decomposed across Point of Presence (PoP), DNS, and CDN layers. To quantify how satellite architecture disrupts terrestrial CDN assumptions, we conduct a measurement study spanning two years. We identify three distinct performance regimes based on infrastructure density. Regions with local content-rich PoPs achieve near-terrestrial latencies with the satellite segment dominating 80–90% of RTT. Infrastructure-sparse regions suffer cascading penalties: remote PoPs force distant resolver selection, which triggers CDN mislocalization, pushing latencies beyond 200 ms. Dense-infrastructure regions show minimal sensitivity to PoP changes. Leveraging Starlink's infrastructure expansion in early 2025 as a natural experiment, we demonstrate that relocating PoPs closer to user location reduces median page-fetch times by 60%. Our findings reveal that infrastructure proximity, not satellite coverage, influences web performance, requiring fundamental changes to CDN mapping and DNS resolution for satellite ISPs.
The sparse mixture-of-experts (MoE) architecture is a promising backbone of foundation models for a wide range of applications in edge. However, deploying them locally presents a significant challenge to memory-constrained GPUs. Previous techniques utilize CPUs for expert offloading, which suffer from inaccurate expert prefetching and on-demand loading latency. To address these challenges, we propose self-speculative MoE ( SS-MoE ), an algorithm-system co-design framework that facilitates inference under limited GPU memory. Our insight is that only a subset of routed experts, i.e., draft model, can still tackle easy tasks and generate draft tokens. Second, we deem GPU memory as the experts cache, and on-demand update it to mitigate IO overhead. Draft tokens from fewer routed experts are generated quickly, and these experts are then routed for verification. Additionally, we design a confidence-based policy to adaptively accept or verify draft tokens, which selectively decreases or increases the number of verification tokens of speculative decoding and achieves acceleration. Notably, under conservative verification, our approach preserves model accuracy and surpasses the decoding speed of the 4-bit quantized counterpart model. Under adaptive verification, our method significantly enhances decoding speed by 3.72x over state-of-the-art methods while maintaining nearly lossless accuracy.
WebAssembly (Wasm) is a web standard that defines a binary instruction format serving as a portable compilation target for high-level programming languages. As a cornerstone design feature, code porting empowers developers to migrate code across platforms without extensive redevelopment. However, we observe that certain native code exhibits functional discrepancies compared to its Wasm counterpart. These discrepancies impede migration by causing functional alterations, generating misleading messages, and forcing platform-specific workarounds. In this paper, we present an in-depth analysis of these Native-to-Wasm discrepancies, demonstrating that they often stem from complex underlying causes, leading to unexpected outcomes in real-world applications. We characterize the root causes and practical implications of these discrepancies, providing critical insights into WebAssembly code porting challenges. To facilitate this large-scale analysis, we develop WasmDiff, an automated differential fuzzing framework that detects Native-to-Wasm discrepancies by comparing the execution of identical code across platforms to identify input-triggered behavioral differences. We identify 14,053 discrepancies across 64,001 test samples from the basic benchmark and 3,102 discrepancies within eight real-world projects. Our findings reveal that discrepancies are prevalent, input-dependent and rooted in diverse execution stacks, underscoring the need for greater attention to code porting issues in the WebAssembly ecosystem.