Yueting Li
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
Yueting Li has a total of 41 co-authors. Here are the top co-authors by number of shared publications:
- Weisheng Zhao
National Key Laboratory of Spintronics, Hangzhou International Innovation Institute, Beihang University 1 , Hangzhou 311115
8 shared publications - Ngai Wong
Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong SAR, China
5 shared publications - Terry Tao Ye
School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, China
3 shared publications - Wendong Xu
The University of Hong Kong Department of Electrical and Electronic Engineering
3 shared publications - Daoqian Zhu
Fert Beijing Institute, School of Integrated Circuit Science and Engineering, Beihang University, Beijing, China
2 shared publications - Keni Qiu
Capital Normal University, Beijing, China
2 shared publications
Latest Publications
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
PEACE: Algorithm-Hardware Co-Design for PE-Granular Energy-Aware Compressed GEMM Acceleration
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessPEACE: Algorithm-Hardware Co-Design for PE-Granular Energy-Aware Compressed GEMM Acceleration
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-2
DScNMP: A Dataflow Scheduling for the Precision-Scalable Near-memory Processing Microarchitecture
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-2
Open AccessDScNMP: A Dataflow Scheduling for the Precision-Scalable Near-memory Processing Microarchitecture
- 3 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 75 (5) , 1820-1834
ReNN-RV: Run-Time PE Re configuration for D NN Inference Acceleration With Custom R ISC- V ISA
Deep neural network (DNN) accelerators integrated with RISC-V Instruction Set Architecture (ISA) extensions have enabled efficient computing on resource-constrained platforms. However, their specialization in regular compute patterns limits effectiveness on irregular workloads, making it challenging to achieve high throughput and energy efficiency. To tackle these challenges, we present ReNN-RV, which integrates a computation-aware RISC-V ISA extension with an instruction-driven processing pipeline to efficiently accelerate run-time reconfigurable processing elements (RePEs). The computation-aware ISA employs configurable opcodes and custom encodings to support fine-grained task scheduling, while an instruction-driven pipeline implements it with minimal control complexity. Moreover, the RePE accelerator provides seamless switching between multiply-accumulate (MAC) and non-MAC operations by configuring a path multiplexer to realize multiple operators at run time. Experimental results demonstrate that ReNN-RV achieves average reductions of$14.6\times$ in cycle count and$15.3\times$ in execution time across representative DNN workloads compared with the baseline RISC-V design. On average, ReNN-RV outperforms state-of-the-art designs by$10.1\times$ for energy efficiency and$10.3\times$ for computational throughput.- 3 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 75 (5) , 1820-1834
ReNN-RV: Run-Time PE Re configuration for D NN Inference Acceleration With Custom R ISC- V ISA
Deep neural network (DNN) accelerators integrated with RISC-V Instruction Set Architecture (ISA) extensions have enabled efficient computing on resource-constrained platforms. However, their specialization in regular compute patterns limits effectiveness on irregular workloads, making it challenging to achieve high throughput and energy efficiency. To tackle these challenges, we present ReNN-RV, which integrates a computation-aware RISC-V ISA extension with an instruction-driven processing pipeline to efficiently accelerate run-time reconfigurable processing elements (RePEs). The computation-aware ISA employs configurable opcodes and custom encodings to support fine-grained task scheduling, while an instruction-driven pipeline implements it with minimal control complexity. Moreover, the RePE accelerator provides seamless switching between multiply-accumulate (MAC) and non-MAC operations by configuring a path multiplexer to realize multiple operators at run time. Experimental results demonstrate that ReNN-RV achieves average reductions of$14.6\times$ in cycle count and$15.3\times$ in execution time across representative DNN workloads compared with the baseline RISC-V design. On average, ReNN-RV outperforms state-of-the-art designs by$10.1\times$ for energy efficiency and$10.3\times$ for computational throughput.- Open Access
- 24 February 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Parallel Computing
S-TRAC: An Algorithm–Hardware Co-design of Sparsity-aware Threshold Adjustment for Accelerator-based RISC-V ISA Extensions
Deep neural networks (DNNs) have become foundational to modern applications, yet their substantial computational and memory demands pose major obstacles to energy-efficient inference. Moreover, the rapidly expanding parameter footprint and structural diversity further amplify data movement, leading to substantial energy consumption and latency overheads. To address these issues, we propose a novel accelerator, S-TRAC , that dynamically adjusts sparsity t h r esholds through a lgorithm-hardware c o-design to enable efficient DNN inference. At the algorithm level, we employ a static sparse-dense storage format and a dynamic bit-processing scheme to skip non-contributing bits without sacrificing weight precision. At the hardware level, we introduce a column-wise processing-element array with LUT-based shift-accumulate multiplication and a global partial-sum accumulator to sustain energy-efficient execution. To support the proposed algorithm-hardware co-design, we propose a RISC-V extension that coordinates the read, arrangement, multiplication, accumulation, and write stages to support end-to-end accelerator execution. Experimental results show that S-TRAC increases effective sparsity by an average factor of 8.4 × across the evaluated DNN models, enabling substantial memory savings. S-TRAC design achieves 11.16 × energy efficiency and 37.03 × hardware efficiency improvements over state-of-the-art solutions.
- 24 February 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Parallel Computing
Open AccessS-TRAC: An Algorithm–Hardware Co-design of Sparsity-aware Threshold Adjustment for Accelerator-based RISC-V ISA Extensions
Deep neural networks (DNNs) have become foundational to modern applications, yet their substantial computational and memory demands pose major obstacles to energy-efficient inference. Moreover, the rapidly expanding parameter footprint and structural diversity further amplify data movement, leading to substantial energy consumption and latency overheads. To address these issues, we propose a novel accelerator, S-TRAC , that dynamically adjusts sparsity t h r esholds through a lgorithm-hardware c o-design to enable efficient DNN inference. At the algorithm level, we employ a static sparse-dense storage format and a dynamic bit-processing scheme to skip non-contributing bits without sacrificing weight precision. At the hardware level, we introduce a column-wise processing-element array with LUT-based shift-accumulate multiplication and a global partial-sum accumulator to sustain energy-efficient execution. To support the proposed algorithm-hardware co-design, we propose a RISC-V extension that coordinates the read, arrangement, multiplication, accumulation, and write stages to support end-to-end accelerator execution. Experimental results show that S-TRAC increases effective sparsity by an average factor of 8.4 × across the evaluated DNN models, enabling substantial memory savings. S-TRAC design achieves 11.16 × energy efficiency and 37.03 × hardware efficiency improvements over state-of-the-art solutions.
- Open Access
- 17 December 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
- Vol. 31 (2) , 1-24
PPD: A Portable and Highly Parallel Dispatching System for Deep Learning
The acceleration of inference process for deep learning models is closely tied with the parallelization capability of computational graph operators and the parallel scheduling strategies. Most existing deep learning compilers focus on optimizing intra-operator parallelism, while neglecting inter-operator parallelism. Furthermore, most industrial inference engines, such as PyTorch and TensorFlow, utilize a dataflow-based model to describe tasks and schedule operators. They are computationally expensive and operate in a topological order and are parallelized to run within a single CUDA stream. However, they fail to fully exploit the parallelism capabilities of multiple CUDA streams. In this article, we propose PPD, a portable, highly parallel dispatching system. It boosts the inference performance by dividing the computational graph into multiple taskflow-based subgraphs. Additionally, PPD entails a dispatching algorithm on a single GPU with multiple CUDA streams to enhance the parallelism and performance of model inference. PPD offers users a lightweight model definition and an inference C++ interface, allowing for seamless integration into any context. We also verify the feasibility of PPD on AMD and other graphics cards. We validate our approach on widely adopted neural network models with varying degrees of parallelism, and compare it with industrial inference engines. Experiments demonstrate that PPD outperforms SOTA methods by up to 2.28× .
- 17 December 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
- Vol. 31 (2) , 1-24
Open AccessPPD: A Portable and Highly Parallel Dispatching System for Deep Learning
The acceleration of inference process for deep learning models is closely tied with the parallelization capability of computational graph operators and the parallel scheduling strategies. Most existing deep learning compilers focus on optimizing intra-operator parallelism, while neglecting inter-operator parallelism. Furthermore, most industrial inference engines, such as PyTorch and TensorFlow, utilize a dataflow-based model to describe tasks and schedule operators. They are computationally expensive and operate in a topological order and are parallelized to run within a single CUDA stream. However, they fail to fully exploit the parallelism capabilities of multiple CUDA streams. In this article, we propose PPD, a portable, highly parallel dispatching system. It boosts the inference performance by dividing the computational graph into multiple taskflow-based subgraphs. Additionally, PPD entails a dispatching algorithm on a single GPU with multiple CUDA streams to enhance the parallelism and performance of model inference. PPD offers users a lightweight model definition and an inference C++ interface, allowing for seamless integration into any context. We also verify the feasibility of PPD on AMD and other graphics cards. We validate our approach on widely adopted neural network models with varying degrees of parallelism, and compare it with industrial inference engines. Experiments demonstrate that PPD outperforms SOTA methods by up to 2.28× .
- Open Access
- 28 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 212-215
A Custom RISC-V ISA with Scalable Processing Units for Efficient Neural Network Inference
- 28 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 212-215
Open AccessA Custom RISC-V ISA with Scalable Processing Units for Efficient Neural Network Inference
- Open Access
- 25 April 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 23 (3) , 1-24
Toward Energy-efficient STT-MRAM-based Near Memory Computing Architecture for Embedded Systems
Convolutional Neural Networks (CNNs) have significantly impacted embedded system applications across various domains. However, this exacerbates the real-time processing and hardware resource-constrained challenges of embedded systems. To tackle these issues, we propose spin-transfer torque magnetic random-access memory (STT-MRAM)-based near memory computing (NMC) design for embedded systems. We optimize this design from three aspects: Fast-pipelined STT-MRAM readout scheme provides higher memory bandwidth for NMC design, enhancing real-time processing capability with a non-trivial area overhead. Direct index compression format in conjunction with digital sparse matrix-vector multiplication (SpMV) accelerator supports various matrices of practical applications that alleviate computing resource requirements. Custom NMC instructions and stream converter for NMC systems dynamically adjust available hardware resources for better utilization. Experimental results demonstrate that the memory bandwidth of STT-MRAM achieves 26.7 GB/s. Energy consumption and latency improvement of digital SpMV accelerator are up to 64× and 1,120× across sparsity matrices spanning from 10% to 99.8%. Single-precision and double-precision elements transmission increased up to 8× and 9.6×, respectively. Furthermore, our design achieves a throughput of up to 15.9× over state-of-the-art designs.
- 25 April 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 23 (3) , 1-24
Open AccessToward Energy-efficient STT-MRAM-based Near Memory Computing Architecture for Embedded Systems
Convolutional Neural Networks (CNNs) have significantly impacted embedded system applications across various domains. However, this exacerbates the real-time processing and hardware resource-constrained challenges of embedded systems. To tackle these issues, we propose spin-transfer torque magnetic random-access memory (STT-MRAM)-based near memory computing (NMC) design for embedded systems. We optimize this design from three aspects: Fast-pipelined STT-MRAM readout scheme provides higher memory bandwidth for NMC design, enhancing real-time processing capability with a non-trivial area overhead. Direct index compression format in conjunction with digital sparse matrix-vector multiplication (SpMV) accelerator supports various matrices of practical applications that alleviate computing resource requirements. Custom NMC instructions and stream converter for NMC systems dynamically adjust available hardware resources for better utilization. Experimental results demonstrate that the memory bandwidth of STT-MRAM achieves 26.7 GB/s. Energy consumption and latency improvement of digital SpMV accelerator are up to 64× and 1,120× across sparsity matrices spanning from 10% to 99.8%. Single-precision and double-precision elements transmission increased up to 8× and 9.6×, respectively. Furthermore, our design achieves a throughput of up to 15.9× over state-of-the-art designs.
- 1 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 43 (8) , 2405-2410
APIM: An Antiferromagnetic MRAM-Based Processing-In-Memory System for Efficient Bit-Level Operations of Quantized Convolutional Neural Networks
Quantized Convolutional Neural Network (QCNN) is an attractive approach that reduces hardware overheads, especially for energy-constrained systems. However, existing QCNNs still require non-trivial hardware resources and memory capacity in order not to compromise model accuracy. To address this issue, we propose an antiferromagnetic magnetic random-access memory (ARAM)-based processing-in-memory (PIM) system, leveraging bit-level sparsity. Three optimization techniques are proposed to optimize hardware resource utilization while preserving CNN accuracy. Firstly, the ARAM-based memory subsystem allows dynamic adaptation of variable bit-width across CNN layers. Secondly, the bit-level accelerator employs the bit-fusion format engineered for processing data from the ARAM subsystem. Thirdly, a customized data path within the RISC-V core guarantees efficient instruction processing to the ARAM-based memory subsystem and bit-level accelerator, enabling optimal bit-level data transmission and computation. Experimental results demonstrate that this design remarkably reduces data movement by 50%-83% across existing CNNs. Compared to state-of-the-art designs, it enhances throughput and latency by an average of 5x and 10x, respectively. In addition, this design achieves speedups between 1.63x and 2.96x, outstripping other designs in AlexNet, VGG16, and ResNet18 benchmarks.
- 1 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 43 (8) , 2405-2410
APIM: An Antiferromagnetic MRAM-Based Processing-In-Memory System for Efficient Bit-Level Operations of Quantized Convolutional Neural Networks
Quantized Convolutional Neural Network (QCNN) is an attractive approach that reduces hardware overheads, especially for energy-constrained systems. However, existing QCNNs still require non-trivial hardware resources and memory capacity in order not to compromise model accuracy. To address this issue, we propose an antiferromagnetic magnetic random-access memory (ARAM)-based processing-in-memory (PIM) system, leveraging bit-level sparsity. Three optimization techniques are proposed to optimize hardware resource utilization while preserving CNN accuracy. Firstly, the ARAM-based memory subsystem allows dynamic adaptation of variable bit-width across CNN layers. Secondly, the bit-level accelerator employs the bit-fusion format engineered for processing data from the ARAM subsystem. Thirdly, a customized data path within the RISC-V core guarantees efficient instruction processing to the ARAM-based memory subsystem and bit-level accelerator, enabling optimal bit-level data transmission and computation. Experimental results demonstrate that this design remarkably reduces data movement by 50%-83% across existing CNNs. Compared to state-of-the-art designs, it enhances throughput and latency by an average of 5x and 10x, respectively. In addition, this design achieves speedups between 1.63x and 2.96x, outstripping other designs in AlexNet, VGG16, and ResNet18 benchmarks.
- 20 December 2023
- Published by Elsevier in Sustainable Cities and Society
- Vol. 101, 105145
Study on the impact of smart energy on carbon emissions in smart cities from single and holistic perspectives – Empirical evidence from China
- 20 December 2023
- Published by Elsevier in Sustainable Cities and Society
- Vol. 101, 105145
Study on the impact of smart energy on carbon emissions in smart cities from single and holistic perspectives – Empirical evidence from China
- 15 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-2
Toward Low Soft Error Sensitivity and Latency Near Memory Computing with EB MRAM
Magnetic random-access memory (MRAM) suffers from reliability issues subject to the harsh industrial environment and applications with limited hardware resources. Thus, we propose the new spin-orbit torque MRAM (SOT-MRAM) scheme called exchange bias (EB) MRAM for error-tolerant application. EB MRAM enhances reliability from its immunity to external magnetic and higher endurance. In addition, this paper presents near-memory computing (NMC) with EB MRAM to further achieve low soft error sensitivity and latency implementation. In the proposed scheme, the NMC controller exploits the accelerated design to reduce data movement, and EB MRAM as the computation memory is performed with subword-level compression. The experimental results show that the access time and energy of EB MRAM is only 0.9ns and 0.113pJ/bit while the data compressed rate is up to 75% with 2.05% loss over the state-of-the-art designs.
- 15 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-2
Toward Low Soft Error Sensitivity and Latency Near Memory Computing with EB MRAM
Magnetic random-access memory (MRAM) suffers from reliability issues subject to the harsh industrial environment and applications with limited hardware resources. Thus, we propose the new spin-orbit torque MRAM (SOT-MRAM) scheme called exchange bias (EB) MRAM for error-tolerant application. EB MRAM enhances reliability from its immunity to external magnetic and higher endurance. In addition, this paper presents near-memory computing (NMC) with EB MRAM to further achieve low soft error sensitivity and latency implementation. In the proposed scheme, the NMC controller exploits the accelerated design to reduce data movement, and EB MRAM as the computation memory is performed with subword-level compression. The experimental results show that the access time and energy of EB MRAM is only 0.9ns and 0.113pJ/bit while the data compressed rate is up to 75% with 2.05% loss over the state-of-the-art designs.