English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Transformer inference requires high compute accuracy; achieving this usinganalog CIMs has been difficult due to inherent computational errors. Toovercome this challenge, we propose a Capacitor-Reconfiguring CIM (CR-CIM) torealize high compute accuracy analog CIM with a 10-bit ADC attaininghigh-area/power efficiency. CR-CIM reconfigures its capacitor array to servedual purposes: for computation and ADC conversion, achieving significant areasavings. Furthermore, CR-CIMs eliminate signal attenuation by keeping thesignal charge stationary during operation, leading to a 4x improvement incomparator energy efficiency. We also propose a software-analog co-designtechnique integrating majority voting into the 10-bit ADC to dynamicallyoptimize the CIM noise performance based on the running layer to further saveinference power. Our CR-CIM achieves the highest compute-accuracy for analogCIMs, and the power efficiency of 818 TOPS/W is competitive with thestate-of-the-art. Furthermore, the FoM considering SQNR and CSNR is 2.3x and1.5x better than previous works, respectively. Vision Transformer (ViT)inference is achieved and realizes a highest CIFAR10 accuracy of 95.8% foranalog CIMs.

An 818-TOPS/W CSNR-31dB SQNR-45dB 10-bit Capacitor-Reconfiguring Computing-in-Memory Macro with Software-Analog Co-Design for Transformers