arXiv subjects

Shuiguang Deng

Publications and source records attributed to Shuiguang Deng.

At least 37 records · Page 2Linked to original sources

arXiv2026

UniSpike: Accelerating Spiking Neural Networks on Neuromorphic Systems via Eliminating Address Redundancy

Many-core neuromorphic systems accelerate Spiking Neural Networks (SNNs), yet their packet-based spike communication can spend substantial traffic and energy repeatedly transmitting destination addresses. This overhead is amplified by the small payload of spike packets: in representative workloads, duplicate address transmissions account for up to 49% of the total traffic. This paper presents UniSpike, a hardware-software co-design that removes address redundancy by aggregating spikes destined for the same core into compact packets. UniSpike combines destination-centric spike scheduling, lightweight runtime packet assembly hardware, and destination-aware SNN partitioning. Across diverse SNN workloads, UniSpike reduces traffic by 1.93$\times$ on average, delivering 1.77$\times$ speedup and 1.50$\times$ energy efficiency improvement over state-of-the-art designs.

cs.NE↗

arXiv2026

DRAMA: Next-Gen Dynamic Orchestration for Resilient Multi-Agent Ecosystems in Flux

Multi-agent systems (MAS) have demonstrated significant effectiveness in addressing complex problems through coordinated collaboration among heterogeneous agents. However, real-world environments and task specifications are inherently dynamic, characterized by frequent changes, uncertainty, and variability. Despite this, most existing MAS frameworks rely on static architectures with fixed agent capabilities and rigid task allocation strategies, which greatly limits their adaptability to evolving conditions. This inflexibility poses substantial challenges for sustaining robust and efficient multi-agent cooperation in dynamic and unpredictable scenarios. To address these limitations, we propose DRAMA: a Dynamic and Robust Allocation-based Multi-Agent System designed to facilitate resilient collaboration in rapidly changing environments. DRAMA features a modular architecture with a clear separation between the control plane and the worker plane. Both agents and tasks are abstracted as resource objects with well-defined lifecycles, while task allocation is achieved via an affinity-based, loosely coupled mechanism. The control plane enables real-time monitoring and centralized planning, allowing flexible and efficient task reassignment as agents join, depart, or become unavailable, thereby ensuring continuous and robust task execution. The worker plane comprises a cluster of autonomous agents, each with local reasoning, task execution, the ability to collaborate, and the capability to take over unfinished tasks from other agents when needed.

cs.MA↗

arXiv2026

Frequency Matching in Spiking Neural Networks for mmWave Sensing

Millimeter-wave (mmWave) sensing enables privacy-preserving, always-on edge perception, but its measurements are often sparse, temporally irregular, and corrupted by high-frequency noise. Existing mmWave pipelines predominantly rely on artificial neural networks (ANNs), which achieve robustness through extensive preprocessing or deep architectures, thereby limiting their efficiency on edge devices. In this work, we study spiking neural networks (SNNs) for mmWave sensing from a mechanism-data alignment perspective. By leveraging the low-pass filtering behavior of leaky integrate-and-fire (LIF) dynamics, we analyze how their implicit temporal filtering interacts with the frequency structure of mmWave signals. Our analysis shows that when discriminative information resides in low-to-mid frequencies, LIF dynamics can inherently suppress high-frequency noise, clarifying when and why SNNs outperform ANNs. Based on this insight, we derive a principled criterion for configuring the membrane decay factor by matching the effective bandwidth of LIF dynamics to the data's discriminative spectral content. Experimental results across four widely used mmWave datasets validate the proposed frequency-matching hypothesis, yielding an average test-accuracy improvement of 6.22% and a 3.64$\times$ reduction in theoretical energy consumption relative to ANN baselines, under a unified evaluation protocol.

cs.NE↗

arXiv2026

SAFA-SNN: Sparsity-Aware On-Device Few-Shot Class-Incremental Learning with Fast-Adaptive Structure of Spiking Neural Network

Continuous learning of novel classes is crucial for edge devices to preserve data privacy and maintain reliable performance in dynamic environments. However, the scenario becomes particularly challenging when data samples are insufficient, requiring on-device few-shot class-incremental learning (FSCIL). Although existing work has explored parameter-efficient FSCIL frameworks based on artificial neural networks (ANNs), their deployment is still fundamentally constrained by limited device resources. Spiking neural networks (SNNs) process spatiotemporal information efficiently, offering lower energy consumption, greater biological plausibility, and compatibility with neuromorphic hardware than ANNs. In this work, we propose an SNN-based method containing Sparsity-Aware neuronal dynamics and Fast Adaptive structure (SAFA-SNN) for on-device FSCIL. By threshold regulation, most neurons exhibit stable spikes and others exhibit adaptive spikes. As a result, synaptic traces that encode base-class knowledge are naturally preserved, thereby alleviating catastrophic forgetting. To cope with spike non-differentiability in backpropagation, we employ a gradient-free technique, i.e., zeroth-order optimization. Moreover, class prototypes can limit overfitting on few-shot data but introduce bias. We enhance prototype discriminability by orthogonal subspace projection. Extensive experiments conducted on two standard benchmark datasets (CIFAR-100 and Mini-ImageNet) and three neuromorphic datasets (CIFAR10-DVS, DVS128 Gesture, and N-Caltech101) demonstrate that SAFA-SNN outperforms baselines, specifically achieving at least 4.01% improvement at the last incremental session on Mini-ImageNet and 20% lower energy cost on CIFAR-100 over baselines with practical implementation.

cs.LG↗

arXiv2026

Reliable Microservice Tail Latency Prediction via Decoupled Dual-Stream Learning and Gradient Modulation

Microservice architectures enable scalable cloud-native applications; however, the distributed nature of these systems complicates the maintenance of strict Service Level Objectives. Accurately predicting window-level P95 tail latency remains difficult due to the complex interactions between software workload propagation and infrastructure resource limits. Existing predictive models struggle to capture these dynamics because the lack of explicit separation between traffic metrics and resource metrics causes misaligned feature representations. Building on this suboptimal data treatment, the unified architectures of prior approaches fail to isolate cascading service dependencies from localized processing capacity. Due to this entanglement, joint training suffers from an optimization imbalance wherein resource features converge faster and dominate gradient updates, thereby preventing the learning of underlying software topologies. To address these challenges, we propose USRFNet, a dual-stream framework that separates the modeling of demand and capacity. The proposed framework utilizes a Graph Neural Network to model the spatial interactions of traffic workloads across software-level service dependencies, and a gating MLP to independently extract infrastructure-level resource dynamics. The model then integrates these representations through hierarchical tensor fusion. To resolve the training imbalance, we introduce a Reliability-Aware Gradient Modulation strategy that dynamically rescales gradients based on the generalization ratio of each data stream. Experiments on three large-scale real-world benchmarks demonstrate that USRFNet outperforms state-of-the-art methods in prediction accuracy. Specifically, compared to the best-performing baselines, the proposed framework achieves relative MAPE reductions ranging from 15.62% to 26.11% across the evaluated datasets.

cs.LG↗

arXiv2026

Toward Robust and Efficient ML-Based GPU Caching for Modern Inference

In modern GPU inference, cache efficiency remains a major bottleneck, and heuristic policies such as \textsc{LRU} can perform far worse than the offline optimum. Existing learning-based caching systems improve hit rates mainly through predictor design, but often follow learned predictions blindly, making performance unreliable when predictions are inaccurate. In contrast, emerging learning-augmented caching algorithms~\cite{pmlr-v80-lykouris18a,mitzenmacher2022algorithms} provide performance guarantees by carefully integrating predictions into caching policies, achieving both \emph{consistency} (near-optimality under perfect predictions) and \emph{robustness} (bounded worst-case performance under prediction errors). However, deployment remains challenging. A practical algorithm should satisfy strict time and space efficiency constraints, which some theoretical work overlooks, while also incurring low deployment overhead. We propose learning-augmented LRU, a deployment-oriented learning-augmented caching algorithm that guarantees \emph{1-consistency} and \emph{$O(k)$-robustness}, incurs low time and space overhead, and maintains strong compatibility. We further build a GPU cache, called \textsc{LCR}, on top of learning-augmented LRU to benefit from its theoretical guarantees and translate them into practical performance. In experiments, \textsc{LCR} reduces P99 time-to-first-token (TTFT) by up to 28.3\% on LLM workloads and increases throughput by up to 24.2\% on deep learning recommendation (DLRM) workloads. Even with poor predictions, performance degrades gracefully and remains close to \textsc{LRU}, demonstrating robustness with practical value.

cs.LG↗

arXiv2026

SortingHat: Redefining Operating Systems Education with a Tailored Digital Teaching Assistant

Operating Systems (OS) courses are among the most challenging in computer science education due to the complexity of internal structures and the diversity of running environments. Traditional teaching methods often fail to address the diverse backgrounds, learning speeds, and practical needs of students. To tackle these challenges, we present SortingHat, a personalized digital teaching assistant tailored specifically for OS education. SortingHat integrates advanced AI technologies, including a retrieval augmented generation (RAG) framework and multi agent reinforcement learning (MARL), to deliver adaptive, scalable, and effective educational support. SortingHat features a 3D digital human interface powered by large language models (LLMs) to provide personalized, empathetic, and context aware guidance. It generates tailored exercises based on each student's learning history and academic performance, reinforcing weak areas and challenging advanced concepts. Additionally, the system incorporates a robust evaluation pipeline that ensures fair, consistent, and unbiased grading of student submissions while delivering personalized, actionable feedback for improvement. By combining personalized guidance, adaptive content creation, and automated assessment, SortingHat transforms OS education into an engaging, immersive, and scalable experience.

cs.HC↗

arXiv2026

A Heavy-Load-Enhanced and Changeable-Periodicity-Perceived Workload Prediction Network

Cloud providers can greatly benefit from accurate workload prediction. However, the workload of cloud servers is highly variable, with occasional workload bursts, which makes workload prediction challenging. The time series forecasting methods relying on periodicity information, often assume fixed and known periodicity length, which does not align with the periodicity-changeable nature of cloud service workloads. Although many state-of-the-art time-series forecasting methods do not rely on periodicity information and achieve high overall accuracy, they are vulnerable to data imbalance between heavy workloads and regular workloads. As a result, their prediction accuracy on rare heavy workloads is limited. Unfortunately, heavyload-prediction accuracy is more important than overall one, as errors in heavyload prediction are more likely to cause Service Level Agreement violations than errors in normal-load prediction. Thus, we propose a changeable-periodicity-perceived workload prediction network (PePNet) to fuse periodic information adaptively for periodicity-changeable time series and improve rare heavy workload prediction accuracy. It has two distinctive characteristics: (i) A Periodicity-Perceived Mechanism to detect the periodicity length automatically and fuses periodic information adaptively, which is suitable for periodicity-changeable time series, and (ii) An Achilles' Heel Loss Function that is used to iteratively optimize the most under-fitting part in predicting sequence for each step, thus evidently improving the prediction accuracy of heavy load. Extensive experiments conducted on real-world datasets demonstrate that PePNet improves accuracy for overall workload by 11.8% averagely, compared with state-of-the-art methods. Especially, PePNet improves accuracy for heavy workload by 21.0% averagely.

cs.DC↗

arXiv2026

SatFusion: A Unified Framework for Enhancing Remote Sensing Images via Multi-Frame and Multi-Source Images Fusion

High-quality remote sensing (RS) image acquisition is fundamentally constrained by physical limitations. While Multi-Frame Super-Resolution (MFSR) and Pansharpening address this by exploiting complementary information, they are typically studied in isolation: MFSR lacks high-resolution (HR) structural priors for fine-grained texture recovery, whereas Pansharpening relies on upsampled low-resolution (LR) inputs and is sensitive to noise and misalignment. In this paper, we propose SatFusion, a novel and unified framework that seamlessly bridges multi-frame and multi-source RS image fusion. SatFusion extracts HR semantic features by aggregating complementary information from multiple LR multispectral frames via a Multi-Frame Image Fusion (MFIF) module, and integrates fine-grained structural details from an HR panchromatic image through a Multi-Source Image Fusion (MSIF) module with implicit pixel-level alignment. To further alleviate the lack of structural priors during multi-frame fusion, we introduce an advanced variant, SatFusion*, which integrates a panchromatic-guided mechanism into the MFIF stage. Through structure-aware feature embedding and transformer-based adaptive aggregation, SatFusion* enables spatially adaptive feature selection, strengthening the coupling between multi-frame and multi-source representations. Extensive experiments on four benchmark datasets validate our core insight: synergistically coupling multi-frame and multi-source priors effectively resolves the fragility of existing paradigms, delivering superior reconstruction fidelity, robustness, and generalizability.

eess.IV↗

arXiv2026

EpiDroid: Dependency-Guided Recomposition for Deep State Discovery in Mobile GUI Testing

The increasing scale and complexity of mobile applications make automated GUI exploration essential for software quality assurance. However, existing methods often neglect state dependencies between test fragments, which leads to redundant exploration and prevents access to deep application states. We introduce EpiDroid, a black-box, pluggable framework that augments existing explorers through semantic state dependency awareness. EpiDroid distills raw traces into stable test fragments to extract underlying dependencies. It then employs a Recomposition-Replay paradigm to perform impact reasoning via LLM and deterministic replay on high-value mutable state elements. Through iterative feedback, EpiDroid refines the state-dependency graph to systematically reach deep application states. We integrated EpiDroid into both industrial and state-of-the-art research tools and evaluated it on 20 real-world apps. The results show that EpiDroid consistently improves the performance of all baselines, increasing average code coverage by 10--28\% and delivering 3--4$\times$ more coverage gain compared to continuing the baselines alone from the same starting point. This demonstrates that dependency-guided recomposition unlocks deep states that forward exploration cannot access, irrespective of additional budget.

cs.SE↗

arXiv2026

GRAB-ANNS: High-Throughput Indexing and Hybrid Search via GPU-Native Bucketing

Hybrid search, which jointly optimizes vector similarity and structured predicate filtering, has become a fundamental building block for modern AI-driven systems. While recent predicate-aware ANN indices improve filtering efficiency on CPUs, their performance is increasingly constrained by limited memory bandwidth and parallelism. Although GPUs offer massive parallelism and superior memory bandwidth, directly porting CPU-centric hybrid search algorithms to GPUs leads to severe performance degradation due to architectural mismatches, including irregular memory access, branch divergence, and excessive CPU-GPU synchronization. In this paper, we present GRAB-ANNS, a high-throughput, GPU-native graph index for dynamic hybrid search. Our key insight is to rethink hybrid indexing from a hardware-first perspective. We introduce a bucket-based memory layout that transforms range predicates into lightweight bucket selection, enabling coalesced memory accesses and efficient SIMT execution. To preserve global navigability under arbitrary filters, we design a hybrid graph topology that combines dense intra-bucket local edges with sparse inter-bucket remote edges. We further develop an append-only update pipeline that supports efficient batched insertions and parallel graph maintenance on GPUs. Extensive experiments on large-scale datasets show that GRAB-ANNS achieves up to 240.1 times higher query throughput and 12.6 times faster index construction than state-of-the-art CPU-based systems, and up to 10 times higher throughput compared to optimized GPU-native reimplementations, while maintaining high recall.

cs.DB↗

arXiv2026

Synergizing Large Language Models and Task-specific Models for Time Series Anomaly Detection

In anomaly detection, methods based on large language models (LLMs) can incorporate expert knowledge by reading professional document, while task-specific small models excel at extracting normal data patterns and detecting value fluctuations from training data of target applications. Inspired by the human nervous system, where the brain stores expert knowledge and the peripheral nervous system and spinal cord handle specific tasks like withdrawal and knee-jerk reflexes, we propose CoLLaTe, a framework designed to facilitate collaboration between LLMs and task-specific models, leveraging the strengths of both models for anomaly detection. In particular, we first formulate the collaboration process and identify two key challenges in the collaboration: (1) the misalignment between the expression domains of the LLMs and task-specific small models, and (2) error accumulation arising from the predictions of both models. To address these challenges, we then introduce two key components in CoLLaTe: a model alignment module and a collaborative loss function. Through theoretical analysis and experimental validation, we demonstrate that these components effectively mitigate the identified challenges and achieve better performance than both LLM-based and task-specific models.

cs.AI↗

arXiv2026

SegQuant: A Semantics-Aware and Generalizable Quantization Framework for Diffusion Models

Diffusion models have demonstrated exceptional generative capabilities but are computationally intensive, posing significant challenges for deployment in resource-constrained or latency-sensitive environments. Quantization offers an effective means to reduce model size and computational cost, with post-training quantization (PTQ) being particularly appealing due to its compatibility with pre-trained models without requiring retraining or training data. However, existing PTQ methods for diffusion models often rely on architecture-specific heuristics that limit their generalizability and hinder integration with industrial deployment pipelines. To address these limitations, we propose SegQuant, a unified quantization framework that adaptively combines complementary techniques to enhance cross-model versatility. SegQuant consists of a segment-aware, graph-based quantization strategy (SegLinear) that captures structural semantics and spatial heterogeneity, along with a dual-scale quantization scheme (DualScale) that preserves polarity-asymmetric activations, which is crucial for maintaining visual fidelity in generated outputs. SegQuant is broadly applicable beyond Transformer-based diffusion models, achieving strong performance while ensuring seamless compatibility with mainstream deployment tools.

cs.CV↗

arXiv2026

CodeGlance: Understanding Code Reasoning Challenges in LLMs through Multi-Dimensional Feature Analysis

In modern software development, developers frequently need to understand code behavior at a glance -- whether reviewing pull requests, debugging issues, or navigating unfamiliar codebases. This ability to reason about dynamic program behavior is fundamental to effective software engineering and increasingly supported by Large Language Models (LLMs). However, existing studies on code reasoning focus primarily on isolated code snippets, overlooking the complexity of real-world scenarios involving external API interactions and unfamiliar functions. This gap hinders our understanding of what truly makes code reasoning challenging for LLMs across diverse programming contexts. We present CodeGlance, a multi-dimensional benchmark investigating code reasoning challenges across three realistic scenarios: intrinsic logic reasoning, API interaction reasoning, and unseen function reasoning. Through systematic evaluation of 7 state-of-the-art LLMs, we reveal that unseen function reasoning poses significant challenges especially for smaller models, with Qwen2.5-3b achieving only 6.0\% accuracy on unseen functions compared to 37.5\% on familiar APIs. We identify critical code complexity features -- including execution trace length, API invocation count, and control flow complexity -- that significantly impact code reasoning difficulty across scenarios. We further investigate how common augmentation strategies, including CoT, document retrieval, and code search, can improve reasoning performance, finding that their effectiveness varies substantially depending on whether challenges stem from logical complexity or knowledge gaps. These findings provide actionable guidance for developing more capable code reasoning systems and deploying LLM-based programming assistants in real-world software development.

cs.SE↗

arXiv2026

TADS: Task-Aware Data Selection for Multi-Task Multimodal Pre-Training

Large-scale multimodal pre-trained models like CLIP rely heavily on high-quality training data, yet raw web-crawled datasets are often noisy, misaligned, and redundant, leading to inefficient training and suboptimal generalization. Existing data selection methods are either heuristic-based, suffering from bias and limited diversity, or data-driven but task-agnostic, failing to optimize for multi-task scenarios. To address these gaps, we introduce TADS (Task-Aware Data Selection), a novel framework for multi-task multimodal pre-training that integrates Intrinsic Quality, Task Relevance, and Distributional Diversity into a learnable value function. TADS employs a comprehensive quality assessment system with unimodal and cross-modal operators, quantifies task relevance via interpretable similarity vectors, and optimizes diversity through cluster-based weighting. A feedback-driven meta-learning mechanism adaptively refines the selection strategy based on proxy model performance across multiple downstream tasks. Experiments on CC12M demonstrate that TADS achieves superior zero-shot performance on benchmarks like ImageNet, CIFAR-100, MS-COCO, and Flickr30K, using only 36% of the data while outperforming baselines by an average of 1.0%. This highlights that TADS significantly enhances data efficiency by curating a high-utility subset that yields a much higher performance ceiling within the same computational constraints.

cs.LG↗

arXiv2026

Morphis: SLO-Aware Resource Scheduling for Microservices with Time-Varying Call Graphs

Modern microservice systems exhibit continuous structural evolution in their runtime call graphs due to workload fluctuations, fault responses, and deployment activities. Despite this complexity, our analysis of over 500,000 production traces from ByteDance reveals a latent regularity: execution paths concentrate around a small set of recurring invocation patterns. However, existing resource management approaches fail to exploit this structure. Industrial autoscalers like Kubernetes HPA ignore inter-service dependencies, while recent academic methods often assume static topologies, rendering them ineffective under dynamic execution contexts. In this work, we propose Morphis, a dependency-aware provisioning framework that unifies pattern-aware trace analysis with global optimization. It introduces structural fingerprinting that decomposes traces into a stable execution backbone and interpretable deviation subgraphs. Then, resource allocation is formulated as a constrained optimization problem over predicted pattern distributions, jointly minimizing aggregate CPU usage while satisfying end-to-end tail-latency SLOs. Our extensive evaluations on the TrainTicket benchmark demonstrate that Morphis reduces CPU consumption by 35-38% compared to state-of-the-art baselines while maintaining 98.8% SLO compliance.

cs.SE↗

arXiv2026

Reinforcement Fine-Tuning for History-Aware Dense Retriever in RAG

Retrieval-augmented generation (RAG) enables large language models (LLMs) to produce evidence-based responses, and its performance hinges on the matching between the retriever and LLMs. Retriever optimization has emerged as an efficient alternative to fine-tuning LLMs. However, existing solutions suffer from objective mismatch between retriever optimization and the goal of RAG pipeline. Reinforcement learning (RL) provides a promising solution to address this limitation, yet applying RL to retriever optimization introduces two fundamental challenges: 1) the deterministic retrieval is incompatible with RL formulations, and 2) state aliasing arises from query-only retrieval in multi-hop reasoning. To address these challenges, we replace deterministic retrieval with stochastic sampling and formulate RAG as a Markov decision process, making retriever optimizable by RL. Further, we incorporate retrieval history into the state at each retrieval step to mitigate state aliasing. Extensive experiments across diverse RAG pipelines, datasets, and retriever scales demonstrate consistent improvements of our approach in RAG performance.

cs.LG↗

arXiv2026

Can Vision-Language Models Handle Long-Context Code? An Empirical Study on Visual Compression

Large Language Models (LLMs) struggle with long-context code due to window limitations. Existing textual code compression methods mitigate this via selective filtering but often disrupt dependency closure, causing semantic fragmentation. To address this, we introduce LongCodeOCR, a visual compression framework that renders code into compressed two-dimensional image sequences for Vision-Language Models (VLMs). By preserving a global view, this approach avoids the dependency breakage inherent in filtering. We systematically evaluate LongCodeOCR against the state-of-the-art LongCodeZip across four benchmarks spanning code summarization, code question answering, and code completion. Our results demonstrate that visual code compression serves as a viable alternative for tasks requiring global understanding. At comparable compression ratios ($\sim$1.7$\times$), LongCodeOCR improves CompScore on Long Module Summarization by 36.85 points over LongCodeZip. At a 1M-token context length with Glyph (a specialized 9B VLM), LongCodeOCR maintains higher accuracy than LongCodeZip while operating at about 4$\times$ higher compression. Moreover, compared with LongCodeZip, LongCodeOCR drastically reduces compression-stage overhead (reducing latency from $\sim$4.3 hours to $\sim$1 minute at 1M tokens). Finally, our results characterize a fundamental coverage--fidelity trade-off: visual code compression retains broader context coverage to support global dependencies, yet faces fidelity bottlenecks on exactness-critical tasks; by contrast, textual code compression preserves symbol-level precision while sacrificing structural coverage.

cs.SE↗