In this paper, we point out that the objective of the retrieval algorithms is to align with the LLM, which is similar to the objective of knowledge distillation in LLMs. We analyze the similarity in information focus between the distilled language model(DLM) and the original LLM from the perspective of information theory, and thus propose a novel paradigm that leverages a DLM as the retrieval algorithm. Based on the insight, we present SpeContext, an algorithm and system co-design for long-context reasoning. (1) At the algorithm level, SpeContext proposes lightweight retrieval head based on the head-level attention weights of DLM, achieving > 90% parameters reduction by pruning the redundancy. (2) At the system level, SpeContext designs an asynchronous prefetch dataflow via the elastic loading strategy, effectively overlapping KV cache retrieval with the LLM computation. (3) At the compilation level, SpeContext constructs the theoretical memory model and implements an adaptive memory management system to achieve acceleration by maximizing GPU memory utilization. We deploy and evaluate SpeContext in two resourceconstrained environments, cloud and edge. Extensive experiments show that, compared with the Huggingface framework, SpeContext achieves up to 24.89x throughput improvement in cloud and 10.06x speedup in edge with negligible accuracy loss, pushing the Pareto frontier of accuracy and throughput.

Large language model (LLM) inference often suffers from high latency, particularly in resource-constrained environments such as on-device or edge deployments. To address this challenge, we present StorInfer, a novel storage-assisted LLM inference system that accelerates response time by precomputing and storing predictable query-response pairs offline. When a user query semantically matches a precomputed query, StorInfer bypasses expensive GPU inference and instantly returns the stored response, significantly reducing latency and compute costs. To maximize coverage and effectiveness, StorInfer employs an LLM-driven generator that adaptively produces diverse and deduplicated queries based on a given knowledge base. This is achieved via two techniques: adaptive query masking, which prevents regeneration of similar queries, and adaptive sampling, which dynamically tunes generation parameters to promote semantic diversity. The resulting query-response pairs are embedded and indexed using a disk-backed vector database to enable fast, similarity-based retrieval at runtime. Using this approach, we generated 150K unique precomputed pairs (taking up to 830 MB of storage space), achieving up to 17.3% latency reduction with no loss in response quality. Our evaluation across multiple QA datasets demonstrates the practicality and scalability of storage-assisted inference, especially in scenarios with predictable query distributions. StorInfer highlights a promising direction in leveraging storage as a primary enabler for efficient, low-latency LLM deployment.

Existing on-device AI architectures for resource-constrained environments face two critical limitations: they lack compactness, with parameter requirements scaling proportionally to task complexity, and they exhibit poor generalizability, performing effectively only on specific application domains (e.g., models designed for regression tasks cannot adapt to natural language processing (NLP) applications). In this paper, we propose CURA, an architecture inspired by analog audio signal processing circuits that provides a compact and lightweight solution for diverse machine learning tasks across multiple domains. Our architecture offers three key advantages over existing approaches: (1) Compactness: it requires significantly fewer parameters regardless of task complexity; (2) Generalizability: it adapts seamlessly across regression, classification, complex NLP, and computer vision tasks; and (3) Complex pattern recognition: it can capture intricate data patterns while maintaining extremely low model complexity. We evaluated CURA across diverse datasets and domains. For compactness, it achieved equivalent accuracy using up to 2,500 times fewer parameters compared to baseline models. For generalizability, it demonstrated consistent performance across four NLP benchmarks and one computer vision dataset, nearly matching specialized existing models (achieving F1-scores up to 90%). Lastly, it delivers superior forecasting accuracy for complex patterns, achieving 1.6 times lower mean absolute error and 2.1 times lower mean squared error than competing models.

This paper presents a comprehensive evaluation of lightweight deep learning models for image classification, emphasizing their suitability for deployment in resource-constrained environments such as low-memory devices. Five state-of-the-art architectures - MobileNetV3 Small, ResNet18, SqueezeNet, EfficientNetV2-S, and ShuffleNetV2 - are benchmarked across three diverse datasets: CIFAR-10, CIFAR-100, and Tiny ImageNet. The models are assessed using four key performance metrics: classification accuracy, inference time, floating-point operations (FLOPs), and model size. Additionally, we investigate the impact of hyperparameter tuning, data augmentation, and training paradigms by comparing pretrained models with scratch-trained counterparts, focusing on MobileNetV3 Small. Our findings reveal that transfer learning significantly enhances model accuracy and computational efficiency, particularly for complex datasets like Tiny ImageNet. EfficientNetV2 consistently achieves the highest accuracy, while MobileNetV3 offers the best balance between accuracy and efficiency, and SqueezeNet excels in inference speed and compactness. This study highlights critical trade-offs between accuracy and efficiency, offering actionable insights for deploying lightweight models in real-world applications where computational resources are limited. By addressing these challenges, this research contributes to optimizing deep learning systems for edge computing and mobile platforms.

Radiation Detection Systems (RDSs) play a vital role in ensuring public safety across various settings, from nuclear facilities to medical environments. However, these systems are increasingly vulnerable to cyber-attacks such as data injection, man-in-the-middle (MITM) attacks, ICMP floods, bot-net attacks, privilege escalation, and distributed denial-of-service (DDoS) attacks. Such threats could compromise the integrity and reliability of radiation measurements, posing significant public health and safety risks. This paper presents a new synthetic radiation dataset and an Intrusion Detection System (IDS) tailored for resource-constrained environments, bringing Machine Learning (ML) predictive capabilities closer to the sensing edge layer of critical infrastructure. Leveraging TinyML techniques, the proposed IDS employs an optimized XGBoost model enhanced with pruning, quantization, feature selection, and sampling. These TinyML techniques significantly reduce the size of the model and computational demands, enabling real-time intrusion detection on low-resource devices while maintaining a reasonable balance between efficiency and accuracy.

The increasing adoption of Artificial Intelligence (AI) has led to larger, more complex models with numerous parameters that require substantial computing power -- resources often unavailable in many real-world application scenarios. Our paper addresses this challenge by introducing knowledge grafting, a novel mechanism that optimizes AI models for resource-constrained environments by transferring selected features (the scion) from a large donor model to a smaller rootstock model. The approach achieves an 88.54% reduction in model size (from 64.39 MB to 7.38 MB), while improving generalization capability of the model. Our new rootstock model achieves 89.97% validation accuracy (vs. donor's 87.47%), maintains lower validation loss (0.2976 vs. 0.5068), and performs exceptionally well on unseen test data with 90.45% accuracy. It addresses the typical size vs performance trade-off, and enables deployment of AI frameworks on resource-constrained devices with enhanced performance. We have tested our approach on an agricultural weed detection scenario, however, it can be extended across various edge computing scenarios, potentially accelerating AI adoption in areas with limited hardware/software support -- by mirroring in a similar manner the horticultural grafting enables productive cultivation in challenging agri-based environments.

Despite significant advances in foundation models like DeepSeek-R1 and ChatGPT, their deployment in medical settings faces critical challenges including computational requirements and professional knowledge barriers. This paper presents an efficient lightweight medical large language model architecture that systematically addresses these challenges through three-dimensional optimization: knowledge acquisition, model compression, and computational enhancement. We design a knowledge transfer pipeline from DeepSeek-R1-Distill-70B to DeepSeek-R1-Distill-7B using Low-Rank Adaptation (LoRA) for precise medical knowledge retention. Through 4-bit quantization and mixed-precision strategies, we achieve substantial model compression while preserving medical reasoning capabilities. The inference framework incorporates Flash Attention acceleration and continuous batching, complemented by specialized prompt templates for diverse medical queries. Experimental evaluation on medical benchmarks demonstrates that our approach maintains 92.1% accuracy on USMLE examinations while reducing memory consumption by 64.7% and inference latency by 12.4% compared to baseline models. This work provides a practical solution for deploying advanced language models in resource-constrained medical environments, enabling broader accessibility of AI-assisted healthcare.

Unsupervised Domain Adaptation (UDA) is a critical challenge in real-world vision systems, especially in resource-constrained environments like drones, where memory and computation are limited. Existing prompt-driven UDA methods typically rely on large vision-language models and require full access to source-domain data during adaptation, limiting their applicability. In this work, we propose Prmpt2Adpt, a lightweight and efficient zero-shot domain adaptation framework built around a teacher-student paradigm guided by prompt-based feature alignment. At the core of our method is a distilled and fine-tuned CLIP model, used as the frozen backbone of a Faster R-CNN teacher . A small set of low-level source features is aligned to the target domain semantics--specified only through a natural language prompt--via Prompt-driven Instance Normalization (PIN). These semantically steered features are used to briefly fine-tune the detection head of the teacher model. The adapted teacher then generates high-quality pseudo-labels, which guide the on-the-fly adaptation of a compact student model. Experiments on the MDS-A dataset demonstrate that Prmpt2Adpt achieves competitive detection performance compared to state-of-the-art methods, while delivering up to 7 faster adaptation and 5 faster inference speed using few source images--making it a practical and scalable solution for real-time adaptation in low-resource domains.

Mixture of Experts (MoE) LLMs, characterized by their sparse activation patterns, offer a promising approach to scaling language models while avoiding proportionally increasing the inference cost. However, their large parameter sizes present deployment challenges in resource-constrained environments with limited GPU memory capacity, as GPU memory is often insufficient to accommodate the full set of model weights. Consequently, typical deployments rely on CPU-GPU hybrid execution: the GPU handles compute-intensive GEMM operations, while the CPU processes the relatively lightweight attention mechanism. This setup introduces a key challenge: how to effectively optimize resource utilization across CPU and GPU? Prior work has designed system optimizations based on performance models with limited scope. Specifically, such models do not capture the complex interactions between hardware properties and system execution mechanisms. Therefore, previous approaches neither identify nor achieve the hardware limit. This paper presents MoE-Lens, a high-throughput MoE LLM inference system designed through holistic performance modeling for resource-constrained environments. Our performance model thoroughly analyzes various fundamental system components, including CPU memory capacity, GPU compute power, and workload characteristics, to understand the theoretical performance upper bound of MoE inference. Furthermore, it captures the system execution mechanisms to identify the key hardware bottlenecks and accurately predict the achievable throughput. Informed by our performance model, MoE-Lens introduces an inference system approaching hardware limits. Evaluated on diverse MoE models and datasets, MoE-Lens outperforms the state-of-the-art solution by 4.6x on average (up to 25.5x), with our theoretical model predicting performance with an average 94% accuracy.

Movements of maritime vessels are inherently complex and challenging to model due to the dynamic and often unpredictable nature of maritime operations. Even within structured maritime environments, such as shipping lanes and port approaches, where vessels adhere to navigational rules and predefined sea routes, uncovering underlying patterns is far from trivial. The necessity for accurate modeling of the mobility of maritime vessels arises from the numerous applications it serves, including risk assessment for collision avoidance, optimization of shipping routes, and efficient port management. This paper introduces FLP-XR, a model that leverages maritime mobility data to construct a robust framework that offers precise predictions while ensuring extremely fast training and inference capabilities. We demonstrate the efficiency of our approach through an extensive experimental study using three real-world AIS datasets. According to the experimental results, FLP-XR outperforms the current state-of-the-art in many cases, whereas it performs 2-3 orders of magnitude faster in terms of training and inference.