YSYanan Sun

Yanan Sun

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Yanan Sun has a total of 96 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • HEIRS-PS: Hybrid 3-D RRAM- and SRAM-CIM for Pipelined Sparse Multitask Transformer Acceleration

    by Liukai Xu, Shuai Yuan, Dengfeng Wang, Weifeng He, Naifeng Jing, Zhigang Ji, Jianfei Jiang, Guanghui He, Xueqing Li, Yanan Sun

    Multi-task transformers have garnered widespread attention for fast task adaptation at edge. However, in addition to matrix-vector multiplication with static and dynamic weights in the data-intensive transformer, task adaptation introduces new challenge of weight updating. Besides, while sparsity can help mitigate the rapid growth in parameter size and computational demand, it requires great hardware flexibility. Compute-in-memory (CIM) is widely explored to overcome the memory bottleneck in neural network acceleration, mainly with single type of memory devices, but struggles to balance the storage density with computing flexibility across multiple tasks. Meanwhile, the conventional pipeline parallel dataflow for high throughput suffers from the underutilization with bubbles caused by variable input sequence length and model sparsity. In this paper, HEIRS-PS, a hybrid three-dimension RRAM- and SRAM-CIM architecture, is proposed for the pipelined multi-task transformer acceleration with sparsity. With the 3D integrated RRAM-CIM and SRAM-CIM arrays, a novel hybrid- CIM paradigm is proposed with the collaborative addition to achieve both high storage density and computing flexibility for multi-task inference. Moreover, by fully leveraging the hybrid-CIM structure in HEIRS-PS, a new dynamic workload scheduling method is proposed with a layer regrouping method to balance the computing latency across pipeline stages. By eliminating the bubbles, the model inference latency is reduced with higher throughput. Experiment shows that the proposed HEIRS-PS architecture improves the energy efficiency by up to 13.3× and 8.68×, compared with state-of-the-art (SOTA) RRAM-CIM and SRAM-CIM works on Llama-1.3B. Moreover, HEIRS-PS also reduces the computing latency by up to 29.5% with the balanced latency across pipeline stages.

  • A Full-Digital Heterogeneous Computing-in-Memory With Dual-Bit Serialized Weight Stream and Ping-Pong Computing for Efficient Transformer Acceleration

    by Dengfeng Wang, Liukai Xu, Shuai Yuan, Weifeng He, Guanghui He, Fengbin Tu, Xueqing Li, Yanan Sun

    Vision Transformers (ViTs) have demonstrated remarkable performance across various computer vision tasks. However, the self-attention mechanism in transformers incurs high processing loads and costly off-chip data movement in conventional von Neumann architectures. This paper proposes a full-digital heterogeneous Computing-in-Memory (CIM) in-situ hybridizing differential RRAM and SRAM to enable efficient on-chip weight storage and flexible row-column dual-directional access for efficient transformer acceleration. A Ping-Pong computing dataflow is introduced to maximize the throughput density by allocating static and dynamic computations across CIM macros. A low-toggle weight tuning algorithm is proposed to minimize the dynamic power under the dual-bit-serial CIM scheme by introducing only small weight rounding distance. Besides, a patch-channel skip mechanism is used to reduce the redundant computing for dynamic matrix with high sparsity. Experimental results on different transformer models show that the proposed DBS-CIM architecture reduces layer-wise computing energy by up to 59.9% compared to conventional digital CIM. Compared to prior homogeneous CIM designs, it achieves $2.0\times $ to $3.3\times $ higher end-to-end energy efficiency, and outperforms existing heterogeneous CIM designs by $2.13\times $ in computational density.

  • IEEE WiCAS and YPCAS Host Joint Event IEEE at ISICAS 2025 in Qingdao to Explore Career Reshaping in the AI-Powered Era [CASS Conference Highlights]

    by Yanan Sun, Xiaozhe Wang

    The joint event hosted by IEEE Women in Circuits and Systems (WiCAS) and the Young Professionals of the Circuits and Systems Society (YPCAS) on 26 October 2025, in Qingdao brought together leading academics, industry experts, and emerging professionals to explore how artificial intelligence is reshaping careers in the circuits and systems field. Under the theme “Beyond AI: Reshaping Your Career in the AI-Powered Era,” the program featured keynote presentations covering advancements in integrated circuit reliability, biomedical microsystems, and AI-driven hardware design. The event concluded with an interdisciplinary roundtable discussion focused on the evolving role of AI in academia and industry, the critical competencies needed to remain competitive, and the high-value skills that complement AI technologies. This gathering fostered meaningful dialogue and provided strategic insights to help professionals navigate an increasingly AI-centric technological landscape.

  • A Heterogeneous CIM Architecture With Splittable Nonvolatile Computing-in-SRAM Cell Pairs Enabling Efficient On-Chip Neural Network Inference

    by Dengfeng Wang, Liukai Xu, Yiming Chen, Weifeng He, Guanghui He, Xueqing Li, Yanan Sun

    Heterogeneous computing-in-memory (CIM) offers a promising solution for efficient neural network (NN) accelerations by leveraging the characteristics of different types of memories. However, the previous heterogeneous CIM either requires additional data transfer between different types of isolated memories or solely relies on in situ embedded single-level (SL) or three-level (TL) nonvolatile memories (NVMs), making it difficult to trade off between storage density and robustness benefits. In this article, a new heterogeneous CIM architecture (NVS-SPT) with enhanced storage density and inference robustness is proposed to enable full on-chip acceleration of practical-scale NNs while scalable to larger models. A splittable nonvolatile computing-in-static random access memory (SRAM) cell pairs (nvS2RAM-CIM) is proposed with hybrid in situ embedded SL and TL resistive random access memory (ReRAM) groups, allowing flexible configuration as split or linked state to enhance storage density and restore yield. A layerwise hybrid-coding search (LHCS) algorithm with bitwise and tritwise data-aware mapping (BTM) method is proposed to determine the optimal weight coding patterns with high array utilizations. In addition, a merged hybrid-coding block (MHCB) generation scheme is employed to enable high computing parallelism by merging the dense computing patterns. The proposed NVS-SPT demonstrates up to $4.2\times $ higher storage density compared with previous heterogeneous CIM with pure SL-ReRAMs and achieves up to 44.7% enhanced NN accuracy, compared with previous unified ternary coding. Furthermore, the proposed NVS-SPT exhibits up to $1.72\times $ and $1.44\times $ enhanced energy efficiency with $3.10\times $ and $1.42\times $ higher computing density, compared with previous heterogeneous CIM based on pure SL- or TL-ReRAMs, respectively.


  • DBP-CIM: Energy-Efficient 8T SRAM-Based Diagonal-Block Parallel Computing-in-Memory With Compact Data Layout for Arithmetic Operations

    by Dengfeng Wang, Chengjun Chang, Weifeng He, Guanghui He, Yanan Sun

    In this work, an energy-efficient bit-parallel static random-access memory (SRAM)-based computing-in-memory (SRAM-CIM) is proposed for general-purpose in-memory arithmetic operations to adapt diverse computing tasks. A compact diagonal-block parallel (DBP) mapping scheme and a novel arithmetic flow are proposed to address the hardware underutilization issue in conventional two-sided stationary bit-parallel CIM architectures. Specifically, the DBP mapping method is implemented to enhance the throughput by reorganizing the intermediate and final results into diagonal memory blocks, effectively reducing the vacant CIM cells caused by the dynamic bit width during computing. In addition, the proposed hardware-efficient arithmetic flows employ: 1) a pipelined ADD scheme to reduce the critical path latency in near-memory computing units; and 2) shift-based arithmetic operations that halve the hardware resources required for multiplication and division while reducing energy consumption. The post-layout simulations on 28-nm CMOS technology show that the proposed DBP-CIM achieves higher energy efficiency and throughput for general-purpose arithmetic operations, compared with state-of-the-art works. Furthermore, evaluations on the general-purpose benchmarks demonstrate that the DBP-CIM reduces energy consumption and computing cycles by up to 55.9% and 60.9%, compared to the conventional bit-parallel CIM.

  • Hop-CIM: An all-digital two-level approximate SRAM-CIM macro for high energy-efficient HNN acceleration with data-aware early exit and column-wise partial-sum reuse

    by Shunqin Cai, Liukai Xu, Wentao Liu, Dengfeng Wang, Keqing Ouyang, Jinyu Wang, Weizhong Wu, Qiang Huang, Zhi Li, Yanan Sun

  • Robust Monolithic 3D Carbon-Based Computing-in-SRAM With Variation-Aware Bit-Wise Data-Mapping for High-Performance and Integration Density

    by Dengfeng Wang, Weifeng He, Qin Wang, Hailong Jiao, Yanan Sun

    Bit-serial computing-in memory with SRAM cells (SRAM-CIM) enables a full set of integer and floating-point arithmetic operations and various data-intensive computations. Carbon nanotube field-effect transistors (CN-MOSFETs) with high scalability, energy-efficiency, and low process thermal budget are attractive to realize high-dense monolithic three-dimensional (M3D) SRAM-CIM. However, CN-MOSFETs possess unique process variations with asymmetric spatial correlations which can significantly influence the performance and reliability of carbon-based SRAM-CIM. In this paper, new M3D-4N4P SRAM-CIM cells with CN-MOSFETs are proposed with optimized profiles for achieving ultra-high integration density while preserving robustness of data-access and computation. Furthermore, the variation-aware bit-wise data-mapping method is proposed for enhancing the performance of carbon-based SRAM-CIM by leveraging the spatial correlations of CN-MOSFETs. By minimizing the area skew of vertically-stacked layers, the areas of proposed M3D-4N4P SRAM-CIM cells are reduced by up to 50.32% compared to the previous 6N2P SRAM-CIM cells assuming carbon nanotube transistor technology. The proposed M3D-4N4P SRAM-CIM array also achieves by up to 2.17 $\times$ higher throughput on arithmetic operations and 18.34% lower computing latency with 25.36% reduced energy consumptions on MAC-based benchmarks, respectively, compared to the previous 2D-6N2P SRAM-CIM array.


  • HyCTor: A Hybrid CNN-Transformer Network Accelerator With Flexible Weight/Output Stationary Dataflow and Multicore Extension

    by Shuai Yuan, Weifeng He, Zhenhua Zhu, Fangxin Liu, Zhuoran Song, Guohao Dai, Guanghui He, Yanan Sun

    Hybrid convolutional neural network (CNN) and Transformer networks are emerging in computer vision, combining convolutional, linear, and attention layers to achieve high accuracies with moderate model sizes. Developing the accelerators for hybrid networks is pivotal to simultaneously optimize the static matrix multiplication (MM) in convolutional and linear layers, as well as dynamic MM in attention layers. However, the existing accelerators are primarily designed for either CNNs or Transformers, resulting in increased data movement to support dynamic MM and potential under-utilization of hardware for static MM. To enhance computational performance and energy efficiency for hybrid networks, we propose HyCTor, an accelerator featuring flexible output-stationary (OS) and weight-stationary (WS) dataflows, along with a multi-core extension for higher throughput. The parrallel array of HyCTor supports interlayer slicing and intra-layer splicing to improve the utilization for static MM, and enables seamless switching between OS and WS dataflow to minimize the data movement in dynamic MM. By leveraging structured sparsity in OS dataflow and unstructured sparsity in WS dataflow, the computational efficiency is further boosted for each layer through flexible dataflow selection based on the sparsity ratio. Besides, a novel QuadLoop-Mesh topology is proposed to address the complex data dependencies in hybrid networks and minimize data transmission distances in the multi-core HyCTor. Experimental results on ResNet-18, ViTB, and TransIAR-AF show that the proposed single-core HyCTor achieves 1.83×, 1.65×, and 2.41× speedup than state-of-the-art (SOTA) accelerators with 100% utilization rate in most layers, and 3.82× to 38.5× speedup than RTX4090 GPU. The energy efficiency of HyCTor is improved by 1.81× to 8.77× compared with SOTA accelerators. Moreover, the 4-core HyCTor achieves speedups of 3.32×, 2.58×, and 2.91×, while the 16-core HyCTor achieves speedups of 7.05×, 4.05×, and 9.64× compared to 1-core HyCTor on three networks.


  • Open Access

    Descriptor: MOSFET Electrical Simulation Dataset (MESD)

    by Yuhang Zhang, Kain Lu Low, Longfan Li, Leilai Shao, Yanan Sun, Jian Zhao, Yabin Sun, Yanling Shi, Yongfu Li

    This paper introduces the Metal-Oxide-Semiconductor Field Effect Transistor (MOSFET) Electrical Simulation Dataset, MESD, an extensive collection of I-V and C-V characteristics data simulated across different foundries’ Berkeley Short-channel IGFET Models (BSIMs). The MESD dataset covers a range of bias voltages, temperatures, and MOSFET physical dimensions across several process technologies from 3 to 350 nm. We have detailed the design methodology used in the generation of the dataset and have outlined the structure and accessibility of the data. MESD is designed to serve as a standardized benchmark for evaluating and comparing MOSFET models, aiding semiconductor researchers and model developers in assessing the accuracy of their simulations against a robust set of controlled data.