arXiv:2603. 13042v2 Announce Type: replace Abstract: Digital Compute-in-Memory (DCiM) accelerates neural networks by reducing data movement.
arXiv:2505. 14303v3 Announce Type: replace-cross Abstract: Using Resistive Random Access Memory (RRAM) crossbars in Computing-in-Memory (CIM) architectures offers a promising solution to overcome the von Neumann bottleneck.
HBQ: Hierarchical Scaling Block Quantization with Hardware‑Efficiency‑Aware Design for Accurate LLM Inference proposes a new block‑quantization scheme that uses large blocks and low‑overhead significand scaling to balance hardware efficiency and accuracy. The authors demonstrate that larger blocks improve efficiency by amortizing dequantization and accumulation costs, while their SIG scaling compensates for the resulting accuracy loss. Experiments on a 28 nm ASIC accelerator show that HBQ achieves up to 4.6× higher area/energy efficiency than state‑of‑the‑art weight‑only quantization, with 1.5–3.0× speedup and 1.6–3.3× system energy reduction over existing BQ methods.
arXiv:2609.36584v1 Announce Type: new Abstract: Analog in-memory computing (AIMC) offers an alternative for model training by executing matrix operations directly where weights are stored. However, s...
arXiv:2606. 02781v1 Announce Type: cross Abstract: Deep neural networks (DNNs) have achieved state-of-the-art performance across diverse domains.
arXiv:2512.04705v3 Announce Type: replace-cross Abstract: The deployment of Early-Exiting Neural Networks (EENNs) on edge accelerators requires optimizing not only the network architecture but also i...
arXiv:2602. 19330v2 Announce Type: replace Abstract: Graph Neural Networks (GNNs) are increasingly explored for physical design analysis in Electronic Design Automation, particularly for modeling Clock Tree Synthesis behavior such as clock skew and buffering complexity.
arXiv:2511. 15503v3 Announce Type: replace-cross Abstract: High-performance Host processors can integrate Processing-In-Memory (PIM) devices, which can accelerate memory-intensive kernels of Machine Learning (ML) models, including Large Language Models (LLMs), by leveraging the large memory bandwidth available at PIM cores.
arXiv:2607. 05240v1 Announce Type: cross Abstract: Computing-in-Memory (CIM) accelerators execute Matrix-Vector Multiplications (MVMs) in memory, making them a compelling solution for Machine Learning (ML) workloads.
arXiv:2607. 28418v1 Announce Type: cross Abstract: Pruning is a promising approach for improving the efficiency of LLMs.
arXiv:2607. 15123v1 Announce Type: cross Abstract: Recent FPGAs have improved deep learning (DL) inference efficiency through dedicated tensor blocks and in-BRAM computation.
The paper introduces a holistic framework that jointly optimizes architecture, token, and bit-width for Vision Transformers (ViTs). It uses Neural Architecture Search (AutoFormer) to find compact backbones, token merging (ToMe) to reduce information processing, and fp16 mixed‑precision inference to accelerate operations. Experiments on ImageNet‑1K and a real‑world 3D X‑ray semiconductor defect dataset show over 10× improvements in throughput and reductions in parameters, FLOPs, and energy while preserving accuracy.
arXiv:2605. 24391v2 Announce Type: replace-cross Abstract: As the demand for deep learning grows, cost reduction through quantization has become essential for both training and inference.