Unicorn-CIM: Uncovering the Vulnerability and Improving the Resilience of High-Precision Compute-in-Memory
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915400007024640 |
|---|---|
| author | Li, Qiufeng Liang, Yiwen Cao, Weidong |
| author_facet | Li, Qiufeng Liang, Yiwen Cao, Weidong |
| contents | Compute-in-memory (CIM) architecture has been widely
explored to address the von Neumann bottleneck in accelerating deep
neural networks (DNNs). However, its reliability remains largely understudied, particularly in the emerging domain of floating-point (FP)
CIM, which is crucial for speeding up high-precision inference and on device training. This paper introduces Unicorn-CIM, a framework to
uncover the vulnerability and improve the resilience of high-precision
CIM, built on static random-access memory (SRAM)-based FP CIM
architecture. Through the development of fault injection and extensive
characterizations across multiple DNNs, Unicorn-CIM reveals how soft
errors manifest in FP operations and impact overall model performance.
Specifically, we find that high-precision DNNs are extremely sensitive
to errors in the exponent part of FP numbers. Building on this insight,
Unicorn-CIM develops an efficient algorithm-hardware co-design method
that optimizes model exponent distribution through fine-tuning and
incorporates a lightweight Error Correcting Code (ECC) scheme to
safeguard high-precision DNNs on FP CIM. Comprehensive experiments
show that our approach introduces just an 8.98% minimal logic overhead
on the exponent processing path while providing robust error protection
and maintaining model accuracy. This work paves the way for developing
more reliable and efficient CIM hardware. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_02311 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Unicorn-CIM: Uncovering the Vulnerability and Improving the Resilience of High-Precision Compute-in-Memory Li, Qiufeng Liang, Yiwen Cao, Weidong Hardware Architecture Compute-in-memory (CIM) architecture has been widely explored to address the von Neumann bottleneck in accelerating deep neural networks (DNNs). However, its reliability remains largely understudied, particularly in the emerging domain of floating-point (FP) CIM, which is crucial for speeding up high-precision inference and on device training. This paper introduces Unicorn-CIM, a framework to uncover the vulnerability and improve the resilience of high-precision CIM, built on static random-access memory (SRAM)-based FP CIM architecture. Through the development of fault injection and extensive characterizations across multiple DNNs, Unicorn-CIM reveals how soft errors manifest in FP operations and impact overall model performance. Specifically, we find that high-precision DNNs are extremely sensitive to errors in the exponent part of FP numbers. Building on this insight, Unicorn-CIM develops an efficient algorithm-hardware co-design method that optimizes model exponent distribution through fine-tuning and incorporates a lightweight Error Correcting Code (ECC) scheme to safeguard high-precision DNNs on FP CIM. Comprehensive experiments show that our approach introduces just an 8.98% minimal logic overhead on the exponent processing path while providing robust error protection and maintaining model accuracy. This work paves the way for developing more reliable and efficient CIM hardware. |
| title | Unicorn-CIM: Uncovering the Vulnerability and Improving the Resilience of High-Precision Compute-in-Memory |
| topic | Hardware Architecture |
| url | https://arxiv.org/abs/2506.02311 |