Weiwei Shan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Weiwei Shan has a total of 181 co-authors. Here are the top co-authors by number of shared publications:
- Jun Yang
Southeast University
13 shared publications - Junyi Qian
School of Integrated Circuits, Southeast University, Nanjing, China
13 shared publications - Cai Li
School of Integrated Circuits, Southeast University, Nanjing, China
8 shared publications - Haochang Zhi
Southeast University
8 shared publications - Lishuo Deng
School of Integrated Circuits, Southeast University, Nanjing, China
7 shared publications - Jintao Li
Shenzhen Institute for Advanced Study, University of Electronic Science and Technology of China, Shenzhen, China
6 shared publications
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 73 (11) , 7214-7227
SNAP-FPCIM: A 28-nm Embedded-Serial-Alignment and Non-2's-Complement-MAC Floating-Point Compute-in-Memory Macro With Adjustable Computation Precision
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 73 (11) , 7214-7227
SNAP-FPCIM: A 28-nm Embedded-Serial-Alignment and Non-2's-Complement-MAC Floating-Point Compute-in-Memory Macro With Adjustable Computation Precision
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
NVLLM: A 3D NAND-Centric Architecture Enabling Edge on-Device LLM Inference
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessNVLLM: A 3D NAND-Centric Architecture Enabling Edge on-Device LLM Inference
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Beyond Fuzzer Islands: CPU Fuzzing via Smart Coordination
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessBeyond Fuzzer Islands: CPU Fuzzing via Smart Coordination
- 14 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. PP (00189200) , 1-12
A BEV Perception Transformer Accelerator With Saliency-Driven Image/Point Cloud Fusion and Phase-Linked Dataflow in 28 nm CMOS
Deploying advanced Transformer-based models for real-time, high-accuracy multimodal bird’s-eye-view (BEV) perception in autonomous driving imposes substantial hardware demands. To address this, we propose a low-cost, low-power image/point-cloud fusion Transformer accelerator that supports two modes: high-performance driving and ultra-low-power sentry operation. We first propose a cross-modal saliency evaluation mechanism that prunes redundant computation. We then propose a column-wise computing array with a three-level adaptive-precision strategy to enable saliency-driven fusion. Furthermore, we propose a phase-linked dataflow that optimizes on-chip data reuse and an on-chip data-forwarding method that reduces external memory access (EMA). Fabricated in 28nm CMOS, this chip achieves the best perception accuracy on the nuScenes dataset, surpassing prior state-of-the-art (SOTA) accelerators. This is the first BEV perception Transformer accelerator to support a dual-mode architecture of high-performance driving and low-power sentry mode, sustaining an active power of 26.6mW and a long-term average power of only 1.68mW in sparse-activity scenarios. In driving mode, it delivers a peak frame rate of 15.89FPS and a peak energy efficiency of 6.11TOPS/W. Overall, this chip outperforms SOTA accelerators in accuracy, frame rate, and energy efficiency.
- 14 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. PP (00189200) , 1-12
A BEV Perception Transformer Accelerator With Saliency-Driven Image/Point Cloud Fusion and Phase-Linked Dataflow in 28 nm CMOS
Deploying advanced Transformer-based models for real-time, high-accuracy multimodal bird’s-eye-view (BEV) perception in autonomous driving imposes substantial hardware demands. To address this, we propose a low-cost, low-power image/point-cloud fusion Transformer accelerator that supports two modes: high-performance driving and ultra-low-power sentry operation. We first propose a cross-modal saliency evaluation mechanism that prunes redundant computation. We then propose a column-wise computing array with a three-level adaptive-precision strategy to enable saliency-driven fusion. Furthermore, we propose a phase-linked dataflow that optimizes on-chip data reuse and an on-chip data-forwarding method that reduces external memory access (EMA). Fabricated in 28nm CMOS, this chip achieves the best perception accuracy on the nuScenes dataset, surpassing prior state-of-the-art (SOTA) accelerators. This is the first BEV perception Transformer accelerator to support a dual-mode architecture of high-performance driving and low-power sentry mode, sustaining an active power of 26.6mW and a long-term average power of only 1.68mW in sparse-activity scenarios. In driving mode, it delivers a peak frame rate of 15.89FPS and a peak energy efficiency of 6.11TOPS/W. Overall, this chip outperforms SOTA accelerators in accuracy, frame rate, and energy efficiency.
- 5 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 73 (6) , 648-652
Hybrid-Polarity Logic Design and an Automated Algorithm for Glitch-Optimized Arithmetic Units
Artificial Intelligence (AI) processors and Digital Signal Processing (DSP) chips require high-efficiency arithmetic units, while glitch-induced power has become a major obstacle to further efficiency improvement. To address this, a hybrid-polarity logic design method and an automated algorithm are proposed for low-power and glitch-optimized arithmetic units. The output inverters of full/half adders are removed to eliminate redundant switching and alleviate path imbalance. The logical duality of these inverted cells is exploited to enable both conventional logic (positive polarity) and its inverted form (negative polarity) to co-exist and hybrid-propagate in arithmetic units. Then, a glitch-analysis-based inverter re-insertion strategy is used to ensure logic correctness with minimal power cost. The automated algorithm merges partial products from complex computations and explores hybrid-polarity circuit netlists with optimized power efficiency. Simulated in a 28 nm CMOS process, the proposed unsigned and signed multipliers achieve up to 54.40% and 55.92% power gains, 13.67% and 11.43% delay reduction, respectively, compared with their corresponding Design Compiler-generated baselines. When applied to dot-product modules in an AI accelerator, our design achieves up to 57.87% power reduction. The results verify the effectiveness of our hybrid-polarity design in mitigating glitches and enhancing efficiency for large-scale digital systems.
- 5 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 73 (6) , 648-652
Hybrid-Polarity Logic Design and an Automated Algorithm for Glitch-Optimized Arithmetic Units
Artificial Intelligence (AI) processors and Digital Signal Processing (DSP) chips require high-efficiency arithmetic units, while glitch-induced power has become a major obstacle to further efficiency improvement. To address this, a hybrid-polarity logic design method and an automated algorithm are proposed for low-power and glitch-optimized arithmetic units. The output inverters of full/half adders are removed to eliminate redundant switching and alleviate path imbalance. The logical duality of these inverted cells is exploited to enable both conventional logic (positive polarity) and its inverted form (negative polarity) to co-exist and hybrid-propagate in arithmetic units. Then, a glitch-analysis-based inverter re-insertion strategy is used to ensure logic correctness with minimal power cost. The automated algorithm merges partial products from complex computations and explores hybrid-polarity circuit netlists with optimized power efficiency. Simulated in a 28 nm CMOS process, the proposed unsigned and signed multipliers achieve up to 54.40% and 55.92% power gains, 13.67% and 11.43% delay reduction, respectively, compared with their corresponding Design Compiler-generated baselines. When applied to dot-product modules in an AI accelerator, our design achieves up to 57.87% power reduction. The results verify the effectiveness of our hybrid-polarity design in mitigating glitches and enhancing efficiency for large-scale digital systems.
- 16 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. PP (15498328) , 1-13
A 2.2- μ W Keyword Spotting System With Filterbank Learning and Attention-Aware Soft-Threshold Denoising in 28-nm CMOS
Keyword spotting (KWS) is crucial for speech interactions in mobile devices, yet its widespread adoption is constrained by the trade-off between low power and high accuracy in diverse noisy environments. Here, we present an ultra-low-power and robust KWS chip to address this challenge through architecture and circuit co-optimization. First, a learnable Mel filter bank is proposed and co-designed with a group-based time-multiplexed architecture to perform adaptive spectral reshaping in a log-Mel feature extractor, reducing the downstream network computations by 25.8% while preserving feature representation capability. Second, a broadcasted residual network (BC-ResNet) is designed to support both KWS and speaker verification (SV) classification tasks, which reduces additions by 43.5% and parameters by 5% compared to a lightweight depthwise separable CNN (DSCNN). Third, an attention-aware soft-threshold denoising strategy is developed to dynamically filter noisy features, yielding up to a 5.8% increase in accuracy over the baseline without denoising at 0dB signal-to-noise ratio (SNR). Finally, a multiplier-less processing element (PE) array combined with stacked-transistor-based ultra-low-leakage custom memory is designed, reducing hardware area by 31.3% and leakage power by 20.4%. Fabricated in a 28nm CMOS process, the proposed 12-class KWS chip consumes only $2.2\mu $ W at 0.42V supply. It achieves 87.6%-93.5% accuracy on GSCD across 0dB to 20dB SNR levels, demonstrating high robustness for diverse scenarios.
- 16 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. PP (15498328) , 1-13
A 2.2- μ W Keyword Spotting System With Filterbank Learning and Attention-Aware Soft-Threshold Denoising in 28-nm CMOS
Keyword spotting (KWS) is crucial for speech interactions in mobile devices, yet its widespread adoption is constrained by the trade-off between low power and high accuracy in diverse noisy environments. Here, we present an ultra-low-power and robust KWS chip to address this challenge through architecture and circuit co-optimization. First, a learnable Mel filter bank is proposed and co-designed with a group-based time-multiplexed architecture to perform adaptive spectral reshaping in a log-Mel feature extractor, reducing the downstream network computations by 25.8% while preserving feature representation capability. Second, a broadcasted residual network (BC-ResNet) is designed to support both KWS and speaker verification (SV) classification tasks, which reduces additions by 43.5% and parameters by 5% compared to a lightweight depthwise separable CNN (DSCNN). Third, an attention-aware soft-threshold denoising strategy is developed to dynamically filter noisy features, yielding up to a 5.8% increase in accuracy over the baseline without denoising at 0dB signal-to-noise ratio (SNR). Finally, a multiplier-less processing element (PE) array combined with stacked-transistor-based ultra-low-leakage custom memory is designed, reducing hardware area by 31.3% and leakage power by 20.4%. Fabricated in a 28nm CMOS process, the proposed 12-class KWS chip consumes only $2.2\mu $ W at 0.42V supply. It achieves 87.6%-93.5% accuracy on GSCD across 0dB to 20dB SNR levels, demonstrating high robustness for diverse scenarios.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- p. 1-13
VITAMIN: A Variation-Tolerant Clock Duty-Cycle Control System to Mitigate Aging-Induced Clock Duty-Cycle Distortion and V min Degradation With Monitoring-Enabled Adjuster and Path Replica
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- p. 1-13
VITAMIN: A Variation-Tolerant Clock Duty-Cycle Control System to Mitigate Aging-Induced Clock Duty-Cycle Distortion and V min Degradation With Monitoring-Enabled Adjuster and Path Replica