Wenyong Zhou

Dr. Wenyong Zhou

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 April 2026 in ACM Transactions on Design Automation of Electronic Systems
Verified
Journal Article
Open access
TWTaiqiang Wu
Taiqiang Wu
CDChenchen Ding
Chenchen Ding
WZWenyong Zhou
Wenyong Zhou
YCYuxin Cheng
Yuxin Cheng
XFXincheng Feng
Xincheng Feng
NWNgai Wong
Ngai Wong
Low-rank adaptation (LoRA) is a predominant parameter-efficient finetuning method for adapting large language models (LLMs) to downstream tasks. Meanwhile, Compute-in-Memory (CIM) architectures demonstrate superior energy efficiency due to their array-level parallel in-memory computing designs. In this article, we propose deploying the LoRA-finetuned LLMs on the hybrid CIM architecture (i.e., pretrained weights onto energy-efficient Resistive Random-Access Memory (RRAM) and LoRA branches onto noise-free Static Random-Access Memory (SRAM)), reducing the energy cost to about 3% compared with the Nvidia A100 GPU. However, the inherent noise of RRAM on the saved weights leads to performance degradation, simultaneously. To address this issue, we design a novel Hardware-aware Low-rank Adaptation (HaLoRA) method. The key insight is to train a LoRA branch that is robust toward such noise and then deploy it on noise-free SRAM, while the extra cost is negligible since the parameters of LoRAs are much fewer than pretrained weights (e.g., 0.15% for LLaMA-3.2 1B model). To improve the robustness towards the noise, we theoretically analyze the gap between the optimization trajectories of the LoRA branch under both ideal and noisy conditions and further design an extra loss to minimize the upper bound of this gap. Therefore, we can enjoy both energy efficiency and accuracy during inference. Experiments finetuning the Qwen and LLaMA series demonstrate the effectiveness of HaLoRA across multiple reasoning tasks, achieving up to 22.7 improvement in average score while maintaining robustness at various noise types and noise levels.

4 August 2025 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
WZWenyong Zhou
Wenyong Zhou
Zhengwu LiuZhengwu Liu
Zhengwu Liu
YRYuan Ren
Yuan Ren
NWNgai Wong
Ngai Wong
compute-in-memory (CIM) accelerators have emerged as a promising way for enhancing the energy efficiency of convolutional neural networks (CNNs). Deploying CNNs on CIM platforms generally requires quantization of network weights and activations to meet hardware constraints. However, existing approaches either prioritize hardware efficiency with binary weight and activation quantization at the cost of accuracy, or utilize multibit weights and activations for greater accuracy but limited efficiency. In this article, we introduce a novel binary weight multibit activation (BWMA) method for CNNs on CIM-based accelerators. Our contributions include: deriving closed-form solutions for weight quantization in each layer, significantly improving the representational capabilities of binarized weights; and developing a differentiable function for activation quantization, approximating the ideal multibit function while bypassing the extensive search for optimal settings. Through comprehensive experiments on CIFAR-10 and ImageNet datasets, we show that BWMA achieves notable accuracy improvements over existing methods, registering gains of 1.44%–5.46% and 0.35%–5.37% on respective datasets. Moreover, hardware simulation results indicate that 4-bit activation quantization strikes the optimal balance between hardware cost and model performance.

27 October 2025
Verified
Conference Paper
Open access
YCYuxin Cheng
Yuxin Cheng
BHBinxiao Huang
Binxiao Huang
WZWenyong Zhou
Wenyong Zhou
TWTaiqiang Wu
Taiqiang Wu
Zhengwu LiuZhengwu Liu
Zhengwu Liu
NWNgai Wong
Ngai Wong

10 June 2025 in IEEE Transactions on Circuits and Systems II: Express Briefs
Verified
Journal Article
Closed access
WZWenyong Zhou
Wenyong Zhou
BLBoyu Li
Boyu Li
JRJiachen Ren
Jiachen Ren
TWTaiqiang Wu
Taiqiang Wu
ZAZhilin Ai
Zhilin Ai
NWNgai Wong
Ngai Wong
Implicit Neural Representations (INRs) encode discrete signals continuously while addressing spectral bias through activation functions (AFs). Previous approaches mitigate this bias by employing complex AFs, which often incur significant hardware overhead. To tackle this challenge, we introduce QuadINR, a hardware-efficient INR that utilizes piecewise quadratic AFs to achieve superior performance with dramatic reductions in hardware consumption. The quadratic functions encompass rich harmonic content in their Fourier series, delivering enhanced expressivity for high-frequency signals, as verified through Neural Tangent Kernel (NTK) analysis. We develop a unified N-stage pipeline framework that facilitates efficient hardware implementation of various AFs in INRs. We demonstrate FPGA implementations on the VCU128 platform and an ASIC implementation in a 28nm process. Experiments across images and videos show that QuadINR achieves up to 2.06dB PSNR improvement over prior work, with an area of only 1914μm and a dynamic power of 6.14mW, reducing resource and power consumption by up to 97% and improving latency by up to 93% vs existing baselines.

26 July 2026
Verified
Conference Paper
Open access
XFXincheng Feng
Xincheng Feng
WZWenyong Zhou
Wenyong Zhou
TWTaiqiang Wu
Taiqiang Wu
Zhengwu LiuZhengwu Liu
Zhengwu Liu
MLMeng Li
Meng Li
NWNgai Wong
Ngai Wong

1 July 2026 in IEEE Transactions on Computers
Verified
Journal Article
Closed access
XFXincheng Feng
Xincheng Feng
WZWenyong Zhou
Wenyong Zhou
TWTaiqiang Wu
Taiqiang Wu
Zhengwu LiuZhengwu Liu
Zhengwu Liu
MLMeng Li
Meng Li
NWNgai Wong
Ngai Wong