Abstract--Large Language Models (LLMs) are emerging as powerful enablers for autonomous reasoning and natural-language coordination in unmanned aerial vehicle (UA V) swarms operating within Internet of Things (IoT) environments. However, existing LLM-driven UA V systems typically process sensor data, mission descriptions, and control outputs in plaintext, exposing sensitive operational information to privacy and security risks. This work introduces PrivLLMSwarm, a privacy-preserving framework that performs secure LLM inference for UA V swarm coordination through Secure Multi-Party Computation (MPC). The framework incorporates MPC-optimized transformer components, including efficient approximations of nonlinear activations and communication-aware attention mechanisms, enabling practical encrypted inference on resource-constrained aerial platforms. A fine-tuned GPT -based command generator, further enhanced through reinforcement learning in a realistic simulation environment, provides reliable natural-language instructions while maintaining end-to-end confidentiality. Experimental evaluation in an urban-scale simulation demonstrates that PrivLLMSwarm achieves high semantic accuracy, low encrypted inference latency, stable formation control, and robust obstacle-avoidance behavior under privacy constraints. Comparative analysis shows that PrivLLMSwarm offers a more favorable privacy-utility balance than differential privacy, federated learning, and plaintext baselines. PrivLLMSwarm establishes a practical foundation for secure, LLM-enabled UA V swarms in privacy-sensitive IoT applications including smart-city monitoring, emergency response, and critical infrastructure protection.

Agentic AI frameworks add a decision-making orchestrator embedded with external tools, including web search, Python interpreter, contextual database, and others, on top of monolithic LLMs, turning them from passive text oracles into autonomous problem-solvers that can plan, call tools, remember past steps, and adapt on the fly. This paper aims to characterize and understand the system bottlenecks introduced by agentic AI workloads from a largely overlooked CPU-centric perspective. We first systematically characterize Agentic AI on the basis of orchestrator/decision making component, inference path dynamics and repetitiveness of the agentic flow which directly influences the system-level performance. Thereafter, based on the characterization, we choose five representative agentic AI workloads- Haystack RAG, Toolformer, ChemCrow, Langchain and SWE-Agent to profile latency, throughput and energy metrics and demystify the significant impact of CPUs on these metrics relative to GPUs. We observe that - 1. Tool processing on CPUs can take up to 90.6% of the total latency; 2. Agentic throughput gets bottlenecked either by CPU factors - coherence, synchronization and over-subscription of cores or GPU factors - main memory capacity and bandwidth; \circled{3} CPU dynamic energy consumes up to 44% of the total dynamic energy at large batch sizes. Based on the profiling insights, we present two key optimizations- 1. CPU and GPU-Aware Micro-batching (CGAM) and 2. Mixed Agentic Workload Scheduling (MAWS) for homogeneous and heterogeneous agentic workloads respectively to demonstrate the potential to improve the performance, efficiency, and scalability of agentic AI. We achieve up to 2.1x and 1.41x P50 latency speedup compared to the multi-processing benchmark for homogeneous and heterogeneous agentic workloads respectively.

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.

The growing demand for on-device large language model (LLM) inference highlights the need for efficient mobile edge computing (MEC) solutions, especially in resource-constrained settings. Speculative decoding offers a promising solution by partitioning token generation between a lightweight draft model on mobile devices and a powerful target model on edge servers, but suffers from communication overhead and asynchronous delays. This paper is the first to propose a unified framework that jointly optimizes user association and resource allocation (UARA) to support efficient parallel speculative decoding. We solve the UARA problem using a multi-agent deep reinforcement learning algorithm. To evaluate our approach under realistic conditions, we conduct experiments using the Sionna simulator. Results show that our method achieves up to 28.0% and an average of 23.7% reduction in end-to-end latency without compromising inference accuracy, enabling scalable and low-latency LLM services in MEC systems.

--As the Middle East emerges as a strategic hub for artificial intelligence (AI) infrastructure, the feasibility of deploying sustainable datacenters in desert environments has become a topic of growing relevance. This paper presents an empirical study analyzing the energy consumption and carbon footprint of large language model (LLM) inference across four countries: the United Arab Emirates, Iceland, Germany, and the United States of America using DeepSeek Coder 1.3B and the HumanEval dataset on the task of code generation. We use the CodeCarbon library to track energy and carbon emissions and compare geographical trade-offs for climate-aware AI deployment. Our findings highlight both the challenges and potential of datacenters in desert regions and provide a balanced outlook on their role in global AI expansion. With the explosion of large-scale artificial intelligence workloads, the environmental footprint of datacenters has come under scrutiny. The AI compute coming online appears to be increasing by a factor of 10 every six months.

Large language models (LLMs) are increasingly deployed under the Model-as-a-Service (MaaS) paradigm. To meet stringent quality-of-service (QoS) requirements, existing LLM serving systems disaggregate the prefill and decode phases of inference. However, decode instances often experience low GPU utilization due to their memory-bound nature and insufficient batching in dynamic workloads, leaving compute resources underutilized. We introduce Harli, a serving system that improves GPU utilization by co-locating parameter-efficient finetuning (PEFT) tasks with LLM decode instances. PEFT tasks are compute-bound and memory-efficient, making them ideal candidates for safe co-location. Specifically, Harli addresses key challenges--limited memory and unpredictable interference--using three components: a unified memory allocator for runtime memory reuse, a two-stage latency predictor for decode latency modeling, and a QoS-guaranteed throughput-maximizing scheduler for throughput maximization. Experimental results show that Harli improves the finetune throughput by 46.2% on average (up to 92.0%) over state-of-the-art serving systems, while maintaining strict QoS guarantees for inference decode.

IT environments typically have logging mechanisms to monitor system health and detect issues. However, the huge volume of generated logs makes manual inspection impractical, highlighting the importance of automated log analysis in IT Software Support. In this paper, we propose a log analytics tool that leverages Large Language Models (LLMs) for log data processing and issue diagnosis, enabling the generation of automated insights and summaries. We further present a novel approach for efficiently running LLMs on CPUs to process massive log volumes in minimal time without compromising output quality. We share the insights and lessons learned from deployment of the tool - in production since March 2024 - scaled across 70 software products, processing over 2000 tickets for issue diagnosis, achieving a time savings of 300+ man hours and an estimated $15, 444 per month in manpower costs compared to the traditional log analysis practices.

Recent advances in LLMs have outpaced the computational and memory capacities of edge platforms that primarily employ CPUs, thereby challenging efficient and scalable deployment. While ternary quantization enables significant resource savings, existing CPU solutions rely heavily on memory-based lookup tables (LUTs) which limit scalability, and FPGA or GPU accelerators remain impractical for edge use. This paper presents T-SAR, the first framework to achieve scalable ternary LLM inference on CPUs by repurposing the SIMD register file for dynamic, in-register LUT generation with minimal hardware modifications. T-SAR eliminates memory bottlenecks and maximizes data-level parallelism, delivering 5.6-24.5x and 1.1-86.2x improvements in GEMM latency and GEMV throughput, respectively, with only 3.2% power and 1.4% area overheads in SIMD units. T-SAR achieves up to 2.5-4.9x the energy efficiency of an NVIDIA Jetson AGX Orin, establishing a practical approach for efficient LLM inference on edge platforms.

The substantial memory bandwidth and computational demands of large language models (LLMs) present critical challenges for efficient inference. To tackle this, the literature has explored heterogeneous systems that combine neural processing units (NPUs) with DRAM-based processing-in-memory (PIM) for LLM acceleration. However, existing high-precision (e.g., FP16) PIM compute units incur significant area and power overhead in DRAM technology, limiting the effective computation throughput. In this paper, we introduce P3-LLM, a novel NPU-PIM integrated accelerator for LLM inference using hybrid numerical formats. Our approach is threefold: First, we propose a flexible mixed-precision quantization scheme, which leverages hybrid numerical formats to quantize different LLM operands with high compression efficiency and minimal accuracy loss. Second, we architect an efficient PIM accelerator for P3-LLM, featuring enhanced compute units to support hybrid numerical formats. Our careful choice of numerical formats allows to co-design low-precision PIM compute units that significantly boost the computation throughput under iso-area constraints. Third, we optimize the low-precision dataflow of different LLM modules by applying operator fusion to minimize the overhead of runtime dequantization. Evaluation on a diverse set of representative LLMs and tasks demonstrates that P3-LLM achieves state-of-the-art accuracy in terms of both KV-cache quantization and weight-activation quantization. Combining the proposed quantization scheme with PIM architecture co-design, P3-LLM yields an average of $4.9\times$, $2.0\times$, and $3.4\times$ speedups over the state-of-the-art LLM accelerators HBM-PIM, Ecco, and Pimba, respectively. Our quantization code is available at https://github.com/yc2367/P3-LLM.git

The growing demand for on-device large language model (LLM) inference is driving interest in deploying lightweight, cost-effective AI solutions on edge hardware. Single-board computers (SBCs) such as the Raspberry Pi and Orange Pi offer a promising platform for localized, privacy-preserving inference-but remain underexplored in the context of LLM workloads. In this work, we benchmark the performance of 25 quantized open-source LLMs across three SBCs-Raspberry Pi 4, Raspberry Pi 5, and Orange Pi 5 Pro-using two inference runtimes: Ollama and Llamafile. We evaluate generation throughput, memory usage, and power consumption under varying CPU configurations, using multiple prompt types to simulate realistic workloads. Our results show that SBCs can reliably support models up to 1.5B parameters, with Llamafile achieving up to 4x higher throughput and 30-40% lower power usage than Ollama. We identify architecture-specific bottlenecks, highlight runtime-level trade-offs, and provide practical deployment recommendations. This study offers the first broad evaluation of LLM inference on SBCs, bridging the gap between high-performance language models and affordable edge computing.