Biography :

Zitao Mo graduated with a Master's degree from the Institute of Automation, Chinese Academy of Sciences. He Engages primarily in research involving deep learning algorithms, neural network compression and acceleration, and the co-design of software-hardware systems, with a track record of publishing papers at esteemed international conferences including ICLR and NeurIPS. Holds a firm belief in the central role of large models in achieving artificial general intelligence. Transitioning from a stance of resigned skepticism as a former nihilist, now align as a non-reductionist who agrees with scientism.

Publications :

• Xiangyu H, Zitao M, Ke C, Weixiang X, Qinghao H, Peisong W, Qingshan L, Jian C, et al. ProxyBNN: Learning Binarized Neural Networks via Proxy Matrices[C], European Conference on Computer Vision, 2020: 223-241.
• Fanrong L, Gang L, Zitao M, Xiangyu H, Jian C, et al. FSA: A Fine-Grained Systolic Accelerator for Sparse CNNs[J], IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2020, 39(11): 3589-3600.
• Xingting Y, Fanrong L, Zitao M, Jian C, et al. GLIF: A Unified Gated Leaky Integrate-and-Fire Neuron for Spiking Neural Networks[C], Conference on Neural Information Processing Systems, 2022 
• Zeyu Z, Fanrong L, Zitao M, Qinghao H, Gang L, Zejian L, Xiaoyao L, Jian C, et al. : Aggregation-Aware Quantization for Graph Neural Networks[J], ICLR 2023, 2023
• Zeyu Z, Fanrong L, Gang L, Zejian L, Zitao Mo, Qinghao H, Xiaoyao L, Jian C. MEGA: A Memory-Efficient GNN Accelerator Exploiting Degree-Aware Mixed-Precision Quantization. HPCA 2024.