21 citations · 34 across the 6 of their papers we have counts for
5 papers · 1 filter
BitROM: Weight Reload-Free CiROM Architecture Towards Billion-Parameter 1.58-bit LLM Inference
Wenlun Zhang, Xinyu Li, Shimpei Ando +1
Compute-in-Read-Only-Memory (CiROM) accelerators offer outstanding energy efficiency for CNNs by eliminating runtime weight updates. However, their scalability to Large Language Mo…
A Review of SRAM-based Compute-in-Memory Circuits
Kentaro Yoshioka, Shimpei Ando, Satomi Miyagi +2
This paper presents a tutorial and review of SRAM-based Compute-in-Memory (CIM) circuits, with a focus on both Digital CIM (DCIM) and Analog CIM (ACIM) implementations. We explore…
ASiM: Modeling and Analyzing Inference Accuracy of SRAM-Based Analog CiM Circuits
Wenlun Zhang, Shimpei Ando, Yung-Chin Chen +1
SRAM-based Analog Compute-in-Memory (ACiM) demonstrates promising energy efficiency for deep neural network (DNN) processing. Nevertheless, efforts to optimize efficiency frequentl…
PACiM: A Sparsity-Centric Hybrid Compute-in-Memory Architecture via Probabilistic Approximation
Wenlun Zhang, Shimpei Ando, Yung-Chin Chen +3
Approximate computing emerges as a promising approach to enhance the efficiency of compute-in-memory (CiM) systems in deep neural network processing. However, traditional approxima…
HALO-CAT: A Hidden Network Processor with Activation-Localized CIM Architecture and Layer-Penetrative Tiling
Yung-Chin Chen, Shimpei Ando, Daichi Fujiki +2
To address the 'memory wall' problem in NN hardware acceleration, we introduce HALO-CAT, a software-hardware co-design optimized for Hidden Neural Network (HNN) processing. HALO-CA…