News

One paper on “Trinity: A General Purpose FHE Accelerator” is accepted by DATE'2024 as a regular paper. This is a joint work w/ Prof. Mingzhe Zhang from IIE CAS and Ant Group.

Dr. Meng Li received CCF Integrated Circuit Early Career Awards during CCF Chip 2024 for his contribution in the area of efficient and secure AI acceleration circuit and system.

Seven papers, including 2 papers on privacy-preserving AI and 5 papers on efficient AI, are accepted by ICCAD'2024 as regular papers.

Dr. Meng Li received CCF-ANT Award Funding, which is used to drive the research on efficient private AI inference based on low-precision quantization.

Three papers on efficient and privacy-preserving deep learning are accepted by DAC'2024 as regular papers, includin “Alchemist: A Unified Accelerator Architecture for Cross-Scheme Fully Homomorphic Encryption”, “FastQuery: Communication-efficient Embedding Table Query for Private LLMs inference”, and “MoteNN: Memory Optimization via Fine-grained Scheduling for Deep Neural Networks on Tiny Devices”.

One paper on “MixCIM: A Hybrid-Cell-Based Computing-in-Memory Macro with Less-Data-Movement and Activation-Memory-Reuse for Depthwise Separable Neural Networks” is accepted by IEEE CICC'2024.

One paper on “A 16.38TOPS and 4.55POPS/W SRAM Computing-in-Memory Macro for Signed Operands Computation and Batch Normalization Implementation” is accepted by IEEE TCASI'2024.

One paper on “ASCEND: Accurate yet Efficient End-to-End Stochastic Computing Acceleration of Vision Transformer” is accepted by DATE'2024 as a regular paper.

One paper on “Enhancing 3D Detection Through Feature Aligned Deep Fusion” is accepted by 3DV'2024.

One paper on “CoPriv: Network/Protocol Co-Optimization for Communication-Efficient Private Inference” is accepted by NeurIPs'2023.