arXiv subjects

Jinane Bazzi

Publications and source records attributed to Jinane Bazzi.

4 recordsLinked to original sources

arXiv2026

EdgeCIM: A Hardware-Software Co-Design for CIM-Based Acceleration of Small Language Models

The growing demand for deploying Small Language Models (SLMs) on edge devices, including laptops, smartphones, and embedded platforms, has exposed fundamental inefficiencies in existing accelerators. While GPUs handle prefill workloads efficiently, the autoregressive decoding phase is dominated by GEMV operations that are inherently memory-bound, resulting in poor utilization and prohibitive energy costs at the edge. In this work, we present EdgeCIM, a hardware-software co-design framework that rethinks accelerator design for end-to-end decoder-only inference. At its core is a CIM macro, implemented in 65nm, coupled with a tile-based mapping strategy that balances pipeline stages, maximizing parallelism while alleviating DRAM bandwidth bottlenecks. Our simulator enables design space exploration of SLMs up to 4B parameters, identifying Pareto-optimal configurations in terms of latency and energy. Compared to an NVIDIA Orin Nano, EdgeCIM achieves up to 7.3x higher throughput and 49.59x better energy efficiency on LLaMA3.2-1B, and delivers 9.95x higher throughput than Qualcomm SA8255P on LLaMA3.2-3B. Extensive benchmarks on TinyLLaMA-1.1B, LLaMA3.2 (1B, 3B), Phi-3.5-mini-3.8B, Qwen2.5 (0.5B, 1.5B, 3B), SmolLM2-1.7B, SmolLM3-3B, and Qwen3 (0.6B, 1.7B, 4B) reveal that our accelerator, under INT4 precision, achieves on average 336.42 tokens/s and 173.02 tokens/J. These results establish EdgeCIM as a compelling solution towards real-time, energy-efficient edge-scale SLM inference.

cs.AR↗

arXiv2025

Mixed-Precision Quantization for Language Models: Techniques and Prospects

The rapid scaling of language models (LMs) has resulted in unprecedented computational, memory, and energy requirements, making their training and deployment increasingly unsustainable. Quantization has emerged as an essential compression technique to reduce model size, alleviate memory bottlenecks, and accelerate inference. However, while uniform low-bit quantization (e.g., INT8, INT4) provides significant efficiency gains, it can degrade accuracy in sensitive components of transformer-based LMs. Mixed-precision quantization offers a promising alternative by selectively allocating precision across layers or within tensors to balance efficiency and accuracy. This survey provides a comprehensive overview of Mixed-Precision quantization frameworks for LMs (MXPLMs). We first review quantization fundamentals, including uniform and non-uniform quantizers, quantization granularity, and methods widely used in post-training quantization. We then categorize and compare recent MXPLM frameworks according to their bit allocation strategies and precision configurations across weights, activations, and key-value caches. A comparative analysis highlights differences in perplexity, zero-shot task performance, and deployment trade-offs. Furthermore, we contrast MXPLMs with earlier mixed-precision quantization methods for deep neural networks, identifying strategies that transfer and those that face challenges in the LM setting. Finally, we summarize open issues and future directions, including hardware-aware design, activation quantization, and scalable optimization methods for billion-parameter models. By consolidating recent advances, this work serves as a reference for understanding the current landscape and research prospects of mixed-precision quantization for large-scale language models.

cs.LG↗

arXiv2022

DNA Pattern Matching Acceleration with Analog Resistive CAM

DNA pattern matching is essential for many widely used bioinformatics applications. Disease diagnosis is one of these applications, since analyzing changes in DNA sequences can increase our understanding of possible genetic diseases. The remarkable growth in the size of DNA datasets has resulted in challenges in discovering DNA patterns efficiently in terms of run time and power consumption. In this paper, we propose an efficient hardware and software codesign that determines the chance of the occurrence of repeat-expansion diseases using DNA pattern matching. The proposed design parallelizes the DNA pattern matching task using associative memory realized with analog content-addressable memory and implements an algorithm that returns the maximum number of consecutive occurrences of a specific pattern within a DNA sequence. We fully implement all the required hardware circuits with PTM 45-nm technology, and we evaluate the proposed architecture on a practical human DNA dataset. The results show that our design is energy-efficient and significantly accelerates the DNA pattern matching task compared to previous approaches described in the literature.

cs.AR↗

arXiv2022

Efficient Analog CAM Design

Content Addressable Memories (CAMs) are considered a key-enabler for in-memory computing (IMC). IMC shows order of magnitude improvement in energy efficiency and throughput compared to traditional computing techniques. Recently, analog CAMs (aCAMs) were proposed as a means to improve storage density and energy efficiency. In this work, we propose two new aCAM cells to improve data encoding and robustness as compared to existing aCAM cells. We propose a methodology to choose the margin and interval width for data encoding. In addition, we perform a comprehensive comparison against prior work in terms of the number of intervals, noise sensitivity, dynamic range, energy, latency, area, and probability of failure.

cs.AR↗