Related papers: Adaptive Quantization for Key Generation in Low-Po…

Large language models require significant computational resources for deployment, making quantization essential for practical applications. However, the main obstacle to effective quantization lies in systematic outliers in activations and…

Efficient real-world deployments of large language models (LLMs) rely on Key-Value (KV) caching for processing and generating long outputs, reducing the need for repetitive computation. For large contexts, Key-Value caches can take up tens…

Low-Rank Adaptation (LoRA) is a crucial method for efficiently fine-tuning large language models (LLMs), with its effectiveness influenced by two key factors: rank selection and weight initialization. While numerous LoRA variants have been…

LoRaWAN has emerged as one of the promising low-power wide-area network technologies to enable long-range sensing and monitoring applications in Internet of Things. The LoRa physical layer used in LoRaWAN suffers from low data rates and…

The emergence of LLMs has ignited a fresh surge of breakthroughs in NLP applications, particularly in domains such as question-answering systems and text generation. As the need for longer context grows, a significant bottleneck in model…

LoRaWAN is a Low Power Wide Area Network technology featuring long transmission ranges and a simple MAC layer, which can support sensor data collection, control applications and reliable services thanks to the flexibility offered by a large…

Communication at high carrier frequencies such as millimeter wave (mmWave) and terahertz (THz) requires channel estimation for very large bandwidths at low SNR. Hence, allocating an orthogonal pilot tone for each coherence bandwidth leads…

We proposes an ultra low power wideband spectrum sensing architecture by utilizing a one-bit quantization at the cognitive radio (CR) receiver. The impact of this aggressive quantization is quantified and it is shown that the proposed…

Semantic communication, recognized as a promising technology for future intelligent applications, has received widespread research attention. Despite the potential of semantic communication to enhance transmission reliability, especially in…

Post training quantization is essential for deploying large language models (LLMs) on resource constrained hardware, yet state of the art methods enforce uniform bit widths across layers, yielding suboptimal accuracy efficiency trade offs.…

State estimation for discrete-time linear systems with quantized measurements is addressed. By exploiting the set-theoretic nature of the information provided by the quantizer, the problem is cast in the set membership estimation setting.…

Low power wide area network technologies (LPWANs) are attracting attention because they fulfill the need for long range low power communication for the Internet of Things. LoRa is one of the proprietary LPWAN physical layer (PHY)…

Guardrails have emerged as an alternative to safety alignment for content moderation of large language models (LLMs). Existing model-based guardrails have not been designed for resource-constrained computational portable devices, such as…

Reconfigurable Intelligent Surface (RIS) is a new paradigm that enables the reconfiguration of the wireless environment. Based on this feature, RIS can be employed to facilitate Physical-layer Key Generation (PKG). However, this technique…

Post Training Quantization (PTQ), a mainstream model compression technique, often leads to the paradoxical 'low error, high loss' phenomenon because it focuses solely on minimizing quantization error. The root cause lies in the Hessian…

Recent years have witnessed the proliferation of Low-power Wide Area Networks (LPWANs) in the unlicensed band for various Internet-of-Things (IoT) applications. Due to the ultra-low transmission power and long transmission duration, LPWAN…

This paper introduces a method for adapting LoRA adapters in smaller-sized language models to arbitrary downstream tasks. Unlike standard mixture-of-expert architectures, our method employs a gradient-free routing function to choose a…

Physical layer security (PLS) is superior to classical cryptography techniques due to its notion of perfect secrecy and independence to an eavesdropper's computational power. One form of PLS arises when Alice and Bob (the legitimate users)…

Large language models (LLMs) have not yet effectively leveraged the vast amounts of edge-device data, and federated learning (FL) offers a promising paradigm to collaboratively fine-tune LLMs without transferring private edge data to the…

The LoRA-finetuning quantization of LLMs has been extensively studied to obtain accurate yet compact LLMs for deployment on resource-constrained hardware. However, existing methods cause the quantized LLM to severely degrade and even fail…