Xin Ju

Orcid: 0009-0006-0875-196X

Affiliations:
  • National University of Defense Technology, Changsha, China


According to our database1, Xin Ju authored at least 9 papers between 2024 and 2026.

Collaborative distances:
  • Dijkstra number2 of four.
  • Erdős number3 of four.

Timeline

Legend:

Book  In proceedings  Article  PhD thesis  Dataset  Other 

Links

Online presence:

On csauthors.net:

Bibliography

2026
C-CIM: A Multi-Mode Convolution-Capable SRAM-CIM.
ACM Trans. Design Autom. Electr. Syst., January, 2026

Precision boundary modeling for area-efficient Block Floating Point accumulation.
J. Syst. Archit., 2026

2025
A quantized network processor for mixed-precision deep learning models based on enhanced Microscaling format.
J. Syst. Archit., 2025

CAMO: A High-Performance CIM-based Lightweight CNN Accelerator for Mobile Devices.
Proceedings of the IEEE International Symposium on Circuits and Systems, 2025

Super Microscaling: Enhancing Precision and Hardware Efficiency in Deep Learning Quantization.
Proceedings of the IEEE International Symposium on Circuits and Systems, 2025

SmartBlock: Adaptive Block Floating Point Quantization for Efficient DNN Acceleration.
Proceedings of the 54th International Conference on Parallel Processing, 2025

WinAcc: Window-based Acceleration of Neural Networks Using Block Floating Point.
Proceedings of the Design, Automation & Test in Europe Conference, 2025

2024
ABS: Accumulation Bit-Width Scaling Method for Designing Low-Precision Tensor Core.
IEEE Trans. Very Large Scale Integr. Syst., September, 2024

Enhancing the PE Utilization for Multi-Precision Systolic Array via Optimizing Computation Latency.
Proceedings of the IEEE International Symposium on Circuits and Systems, 2024


  Loading...