Kezhi 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
Kezhi Li has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Jiaying Zhu
The Chinese University of Hong Kong, Hong Kong, Hong Kong, Hong Kong
2 shared publications - Mengxia Tao
National Center of Technology Innovation for EDA, Nanjing, Jiangsu, China
2 shared publications - Min Li
Southeast University, Nanjing, Jiangsu, China
2 shared publications - Qiang Xu
The Chinese University of Hong Kong
2 shared publications - Baoqi Zhang
National Center of Technology Innovation for EDA, Nanjing, China
1 shared publication - Chunxiang Wang
Shanghai Jiao Tong University
1 shared publication
Latest Publications
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Miter-Aware LUT Mapping: Aligning Structure and Solvability for Efficient Logic Equivalence Checking
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessMiter-Aware LUT Mapping: Aligning Structure and Solvability for Efficient Logic Equivalence Checking
- Open Access
- 5 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 102-106
EquivFusion: Unifying Hardware Equivalence Checking from Algorithms to Netlists via MLIR
Ensuring functional consistency between high-level algorithmic models and low-level hardware implementations is a critical challenge, particularly as modern design flows increasingly span heterogeneous abstractions—from deep learning frameworks to hardware netlists. In this paper, we present EquivFusion, an end-to-end equivalence checking tool tailored for multi-modal circuit designs. Unlike traditional flows that rely on siloed tools or ad-hoc translation, EquivFusion leverages a verification-oriented MLIR lowering pipeline to unify diverse entry points—including PyTorch, C/C++, Chisel, Verilog, and gate-level netlists—into a common intermediate representation. This architecture enables automated, pairwise equivalence checking across diverse abstraction levels by rigorously translating designs into standard formal verification formats, i.e., SMT-LIB, BTOR2, AIGER. We demonstrate EquivFusion's feasibility to bridge the semantic gap between software specifications and hardware realizations, showcasing its effectiveness in facilitating "shift-left" formal verification for datapath-intensive hardware designs. EquivFusion is available online1.
- 5 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 102-106
Open AccessEquivFusion: Unifying Hardware Equivalence Checking from Algorithms to Netlists via MLIR
Ensuring functional consistency between high-level algorithmic models and low-level hardware implementations is a critical challenge, particularly as modern design flows increasingly span heterogeneous abstractions—from deep learning frameworks to hardware netlists. In this paper, we present EquivFusion, an end-to-end equivalence checking tool tailored for multi-modal circuit designs. Unlike traditional flows that rely on siloed tools or ad-hoc translation, EquivFusion leverages a verification-oriented MLIR lowering pipeline to unify diverse entry points—including PyTorch, C/C++, Chisel, Verilog, and gate-level netlists—into a common intermediate representation. This architecture enables automated, pairwise equivalence checking across diverse abstraction levels by rigorously translating designs into standard formal verification formats, i.e., SMT-LIB, BTOR2, AIGER. We demonstrate EquivFusion's feasibility to bridge the semantic gap between software specifications and hardware realizations, showcasing its effectiveness in facilitating "shift-left" formal verification for datapath-intensive hardware designs. EquivFusion is available online1.
- 25 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Vehicles
- Vol. 9 (1) , 2564-2575
Forehead or Headlights – At Which Height Should LiDARs be Mounted on the Vehicle?
An increasing number of production vehicles are opting to integrate LiDAR systems to achieve superior perception capabilities. However, there are significant differences in the installation positions, particularly the heights, of LiDAR systems across different car models from various companies. For instance, some are installed on the forehead of the vehicle, while others are mounted near the headlights. The installation height of LiDAR has a significant impact on its perception capability, including the LiDAR detection ability and the performance of 3D object detection algorithms. To investigate the influence of LiDAR installation height, we introduced the Height3 dataset, which is collected simultaneously by LiDAR sensors with three heights in the CARLA simulation environment. Using this dataset, we first conducted a comprehensive statistical analysis of the LiDAR detection ability and compared the number of detected vehicles and the detection rate of LiDARs at three heights. Furthermore, we investigated the impact of point clouds captured from different heights on the performance of multiple 3D vehicle detection algorithms. Based on the studies on the simulated Height3 dataset, we have an overall understanding of the performance differences of commercial vehicle-mounted LiDARs installed at different heights, providing suggestions for the installation positions of LiDAR sensors on production vehicles. The dataset is available on http://www.kaggle.com/datasets/kezhili/height3 .
- 25 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Vehicles
- Vol. 9 (1) , 2564-2575
Forehead or Headlights – At Which Height Should LiDARs be Mounted on the Vehicle?
An increasing number of production vehicles are opting to integrate LiDAR systems to achieve superior perception capabilities. However, there are significant differences in the installation positions, particularly the heights, of LiDAR systems across different car models from various companies. For instance, some are installed on the forehead of the vehicle, while others are mounted near the headlights. The installation height of LiDAR has a significant impact on its perception capability, including the LiDAR detection ability and the performance of 3D object detection algorithms. To investigate the influence of LiDAR installation height, we introduced the Height3 dataset, which is collected simultaneously by LiDAR sensors with three heights in the CARLA simulation environment. Using this dataset, we first conducted a comprehensive statistical analysis of the LiDAR detection ability and compared the number of detected vehicles and the detection rate of LiDARs at three heights. Furthermore, we investigated the impact of point clouds captured from different heights on the performance of multiple 3D vehicle detection algorithms. Based on the studies on the simulated Height3 dataset, we have an overall understanding of the performance differences of commercial vehicle-mounted LiDARs installed at different heights, providing suggestions for the installation positions of LiDAR sensors on production vehicles. The dataset is available on http://www.kaggle.com/datasets/kezhili/height3 .